paperID stringlengths 36 36 | pwc_id stringlengths 8 47 | arxiv_id stringlengths 6 16 ⌀ | nips_id float64 | url_abs stringlengths 18 329 | url_pdf stringlengths 18 742 | title stringlengths 8 325 | abstract stringlengths 1 7.27k ⌀ | authors stringlengths 2 7.06k | published stringlengths 10 10 ⌀ | conference stringlengths 12 47 ⌀ | conference_url_abs stringlengths 16 198 ⌀ | conference_url_pdf stringlengths 27 199 ⌀ | proceeding stringlengths 6 47 ⌀ | taskID stringlengths 7 1.44k | areaID stringclasses 688 values | embedding stringlengths 9.26k 12.5k | umap_embedding stringlengths 29 44 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
feb8ace8-549a-4140-915c-19d43f15f743 | graffmatch-global-matching-of-3d-lines-and | 2212.12745 | null | https://arxiv.org/abs/2212.12745v1 | https://arxiv.org/pdf/2212.12745v1.pdf | GraffMatch: Global Matching of 3D Lines and Planes for Wide Baseline LiDAR Registration | Using geometric landmarks like lines and planes can increase navigation accuracy and decrease map storage requirements compared to commonly-used LiDAR point cloud maps. However, landmark-based registration for applications like loop closure detection is challenging because a reliable initial guess is not available. Global landmark matching has been investigated in the literature, but these methods typically use ad hoc representations of 3D line and plane landmarks that are not invariant to large viewpoint changes, resulting in incorrect matches and high registration error. To address this issue, we adopt the affine Grassmannian manifold to represent 3D lines and planes and prove that the distance between two landmarks is invariant to rotation and translation if a shift operation is performed before applying the Grassmannian metric. This invariance property enables the use of our graph-based data association framework for identifying landmark matches that can subsequently be used for registration in the least-squares sense. Evaluated on a challenging landmark matching and registration task using publicly-available LiDAR datasets, our approach yields a 1.7x and 3.5x improvement in successful registrations compared to methods that use viewpoint-dependent centroid and "closest point" representations, respectively. | ['Jonathan P. How', 'Devarth Parikh', 'Parker C. Lusk'] | 2022-12-24 | null | null | null | null | ['loop-closure-detection'] | ['computer-vision'] | [ 2.24607095e-01 -3.31376344e-01 -4.30406891e-02 -4.67000574e-01
-8.18955779e-01 -7.22159982e-01 8.30837667e-01 7.41968334e-01
-6.44107521e-01 4.72609937e-01 -3.94533545e-01 -1.94247857e-01
-3.69371444e-01 -9.81282473e-01 -7.46799409e-01 -2.76394933e-01
-2.43538007e-01 7.31075108e-01 3.55665624e-01 -2.31739968e-01
6.38595700e-01 1.02105272e+00 -1.59881043e+00 -8.61936331e-01
8.89311671e-01 7.52285182e-01 -1.04704864e-01 2.01351166e-01
-9.67723727e-02 -4.09685373e-01 -2.92694420e-01 -4.08378802e-02
7.06866920e-01 5.04450966e-03 -6.24813139e-01 -1.96756274e-01
1.02495074e+00 7.24611245e-03 -7.82794412e-03 1.10047281e+00
5.11806011e-01 4.08998609e-01 4.83487368e-01 -1.26142406e+00
-4.83970195e-02 -1.08141318e-01 -7.04978943e-01 -1.94050744e-01
6.68822825e-01 -4.82991040e-01 8.66392970e-01 -1.00926661e+00
7.57015824e-01 1.05254376e+00 9.90894973e-01 -3.00168265e-02
-1.46759260e+00 -6.22242570e-01 -1.50085524e-01 4.44726013e-02
-2.13656855e+00 -4.26015973e-01 7.14548409e-01 -3.12507659e-01
8.92196655e-01 4.81969237e-01 5.03211081e-01 1.76404491e-01
1.58078089e-01 -1.94195047e-01 7.99253404e-01 -5.57596624e-01
1.33460984e-01 -2.30760604e-01 -7.28247464e-02 7.04739213e-01
6.31270289e-01 4.16177437e-02 -5.05984843e-01 -2.81640917e-01
9.00030315e-01 4.15463634e-02 -1.86253205e-01 -1.03507352e+00
-1.38207304e+00 8.03010106e-01 7.84714401e-01 1.90254763e-01
-3.02968919e-01 2.82623358e-02 -4.00219671e-02 3.24782461e-01
3.11363697e-01 4.98086244e-01 -3.67591232e-02 3.29834409e-02
-9.33694839e-01 2.53841609e-01 6.44038796e-01 1.31390977e+00
1.40416861e+00 -2.85744667e-01 5.89195311e-01 6.61064506e-01
3.04947704e-01 8.99087429e-01 1.17009163e-01 -7.96633720e-01
3.98275137e-01 9.43610013e-01 1.07104763e-01 -1.54244387e+00
-8.47544730e-01 -7.62168393e-02 -6.91769063e-01 2.79560387e-01
4.53291535e-01 4.04014856e-01 -7.38792121e-01 1.23470008e+00
5.40472627e-01 3.34764123e-01 -9.90053043e-02 7.89981544e-01
4.52464402e-01 3.67611907e-02 -3.55783224e-01 1.35621533e-01
1.16735446e+00 -2.64993042e-01 -3.06114048e-01 -2.85705000e-01
7.79342353e-01 -1.08674455e+00 6.77254617e-01 -1.37619272e-01
-6.85088694e-01 -3.75467539e-01 -1.42883730e+00 -2.85257787e-01
-4.44769949e-01 -1.52987957e-01 2.20201671e-01 5.90801835e-01
-1.07504368e+00 5.48565328e-01 -8.69489968e-01 -7.11389601e-01
-5.40410690e-02 6.72772706e-01 -6.55445576e-01 -1.78060532e-01
-6.98538065e-01 1.19130361e+00 7.64437169e-02 2.28856698e-01
1.21999115e-01 -7.19876587e-01 -1.13009405e+00 -3.38055909e-01
2.33583480e-01 -5.69814444e-01 6.67722285e-01 7.07970411e-02
-1.24709058e+00 1.06721818e+00 -4.96346474e-01 -4.81460154e-01
6.03952110e-01 -2.93286532e-01 -1.51978150e-01 -4.20861132e-02
3.73864383e-01 4.44711596e-01 4.81263220e-01 -1.10660565e+00
-4.68571901e-01 -6.97146952e-01 -1.25856623e-01 2.55799115e-01
2.98852623e-01 -2.87031114e-01 -4.64885801e-01 -1.35636762e-01
1.17101884e+00 -1.34213519e+00 -4.49462831e-01 8.12669322e-02
-2.68744111e-01 -1.09285058e-03 7.91737258e-01 -3.71183574e-01
8.12973320e-01 -2.00097799e+00 -1.34098604e-01 9.57733870e-01
1.21347755e-02 -1.21202871e-01 2.61465758e-02 4.70453173e-01
1.69657648e-01 1.05314404e-01 -1.94386467e-01 -2.19055086e-01
-1.44675910e-01 2.98133403e-01 -7.52791315e-02 1.06504512e+00
1.99010689e-02 4.90685582e-01 -8.63442600e-01 -4.45619762e-01
6.18178248e-01 6.04896367e-01 -2.83788472e-01 -2.06994653e-01
3.76381248e-01 4.08471018e-01 -2.96620280e-01 6.10195220e-01
9.55079317e-01 3.18629175e-01 1.71225367e-03 -3.92218858e-01
-4.22178984e-01 3.64426315e-01 -1.49104917e+00 1.94799161e+00
-4.73255873e-01 6.48546219e-01 -3.02112222e-01 -6.76056862e-01
1.62088692e+00 -9.55538601e-02 6.58271134e-01 -6.66580200e-01
3.45423520e-02 6.51623666e-01 -2.28522509e-01 3.13843101e-01
8.36060405e-01 1.85610741e-01 3.19772139e-02 8.13705251e-02
-3.85934502e-01 -6.34328008e-01 -7.96230212e-02 -1.82245463e-01
9.66516435e-01 2.27146357e-01 5.41555345e-01 -3.93394947e-01
7.27797985e-01 2.07307562e-01 4.30571914e-01 4.51530874e-01
2.57104188e-01 7.15816021e-01 -1.77538648e-01 -4.26174819e-01
-9.61067259e-01 -1.07564890e+00 -5.14163733e-01 2.57808447e-01
7.22405791e-01 -6.78660214e-01 -2.79372454e-01 -9.25557092e-02
4.59881306e-01 3.84561181e-01 -2.40155369e-01 -8.06378797e-02
-7.42005467e-01 -2.62085915e-01 3.54622483e-01 3.34524095e-01
4.19894487e-01 -1.55554444e-01 -6.46496296e-01 1.81473926e-01
2.02922802e-02 -1.13925207e+00 -3.75949353e-01 -8.70898291e-02
-1.09431732e+00 -1.24293947e+00 -3.35374206e-01 -5.81271529e-01
1.11787677e+00 7.58162141e-01 8.14753175e-01 3.17072988e-01
-1.79854214e-01 4.98148352e-01 -1.86784625e-01 -2.94828445e-01
1.01951197e-01 1.09720632e-01 4.51891094e-01 -1.15226932e-01
5.34506857e-01 -6.65782571e-01 -4.34211701e-01 7.86595047e-01
-4.17713404e-01 -2.63074398e-01 3.20816308e-01 6.03103280e-01
9.33115602e-01 -2.62978166e-01 2.44136620e-02 -4.02025670e-01
2.22184449e-01 -2.79828124e-02 -1.23822439e+00 8.17893744e-02
-9.21206176e-01 2.02137157e-01 3.08903549e-02 -4.33410779e-02
-2.79142976e-01 4.83772814e-01 2.47029394e-01 -2.24769413e-01
4.08993661e-02 6.43498242e-01 6.66003823e-02 -9.80384111e-01
7.11732805e-01 -9.24301967e-02 1.51617885e-01 -4.00704652e-01
3.89612675e-01 3.60946655e-01 6.61460161e-01 -5.25724113e-01
1.32702529e+00 6.41183794e-01 7.30480492e-01 -1.06064165e+00
-1.88207924e-01 -9.43613529e-01 -1.40963721e+00 1.72183644e-02
5.39899886e-01 -7.95780420e-01 -6.67488217e-01 6.68943152e-02
-1.17606449e+00 3.18973988e-01 -7.58063570e-02 6.62801325e-01
-6.46511793e-01 4.84708548e-01 2.37649694e-01 -6.04830027e-01
-2.53038615e-01 -1.04273915e+00 1.12839997e+00 1.02825873e-01
-3.29152465e-01 -9.10997093e-01 2.59374589e-01 -1.63894519e-01
2.31028184e-01 5.78890443e-01 5.80530941e-01 -4.18279678e-01
-9.06320035e-01 -7.21143484e-01 -1.56474829e-01 -3.04689080e-01
2.52888083e-01 1.12497425e-02 -5.73993504e-01 -3.84695888e-01
-3.33661914e-01 5.17483175e-01 3.91283989e-01 8.36177170e-02
2.43946612e-01 1.57087043e-01 -5.42312682e-01 7.93814898e-01
1.54253435e+00 5.74164577e-02 6.04643822e-01 8.23219895e-01
6.93953276e-01 5.88448763e-01 9.07799125e-01 1.47352040e-01
7.36454904e-01 1.16276622e+00 3.86274964e-01 7.12685473e-03
2.74062157e-01 -4.06427205e-01 -7.21481591e-02 6.63998783e-01
-3.64355236e-01 4.67202485e-01 -1.37734365e+00 4.32836831e-01
-1.87142110e+00 -4.82773036e-01 -5.19366384e-01 2.96886921e+00
4.03234690e-01 -1.24451593e-01 -1.65577605e-01 1.99370608e-01
6.48417175e-01 -2.06832469e-01 -2.51835704e-01 -4.03010026e-02
-4.43619080e-02 2.36433759e-01 1.12404621e+00 8.37848842e-01
-1.07956123e+00 9.76080775e-01 5.81355858e+00 1.08069107e-01
-1.20651627e+00 -2.67527252e-01 -3.94510180e-01 4.27264988e-01
-3.60537730e-02 4.01692212e-01 -8.67779732e-01 -9.84653682e-02
7.61945069e-01 -4.30242479e-01 1.45498559e-01 7.13132143e-01
3.13352682e-02 -2.22309142e-01 -1.04258859e+00 1.21940196e+00
1.24807917e-01 -1.16598988e+00 -1.44172058e-01 3.28238904e-01
5.14111578e-01 2.00356603e-01 -2.62848228e-01 -2.66879857e-01
-3.92481126e-02 -8.32662642e-01 3.40673536e-01 4.48034853e-01
7.19624639e-01 -7.13409722e-01 6.14598274e-01 1.57783017e-01
-1.64193654e+00 5.94167650e-01 -5.81731617e-01 -1.34330047e-02
2.36642942e-01 3.15135211e-01 -1.24088788e+00 1.01065862e+00
5.35337389e-01 6.63643599e-01 -7.04300761e-01 1.51376915e+00
-7.86695480e-02 -7.74072185e-02 -8.95282865e-01 2.35398948e-01
4.40375395e-02 -7.00380564e-01 6.59903765e-01 7.90499568e-01
8.11096191e-01 -4.19374444e-02 3.20263684e-01 6.22594774e-01
3.18849266e-01 4.80393797e-01 -8.84339869e-01 5.55999100e-01
9.00669873e-01 1.02119625e+00 -9.56500947e-01 5.50131351e-02
-3.32813919e-01 7.65558064e-01 -1.51641751e-02 1.09854080e-01
-3.67144734e-01 -5.26747048e-01 7.43889451e-01 3.42402101e-01
-6.10936582e-02 -1.08531713e+00 -4.64007020e-01 -9.69789982e-01
1.60822377e-01 -3.12769234e-01 -8.20790767e-04 -4.55270588e-01
-7.30906963e-01 4.72258061e-01 6.43424839e-02 -1.66522253e+00
-4.38621759e-01 -3.74152452e-01 -4.82615501e-01 1.06064427e+00
-1.46456730e+00 -1.14880133e+00 -6.51345968e-01 6.26918435e-01
-1.51713789e-01 1.83957279e-01 1.07017505e+00 3.30926836e-01
8.15290399e-03 4.97655153e-01 1.96837839e-02 4.11354564e-02
8.00333023e-01 -1.18955004e+00 7.76766062e-01 8.78044724e-01
4.47990566e-01 1.00285637e+00 6.64815366e-01 -7.62021482e-01
-1.69388127e+00 -9.14304078e-01 1.02254331e+00 -5.53815305e-01
5.22682726e-01 -2.47322336e-01 -9.20767009e-01 7.22265601e-01
-5.61215639e-01 1.84830859e-01 4.52609181e-01 2.16844171e-01
-4.38846380e-01 -2.06727490e-01 -1.19866025e+00 5.03158092e-01
1.17764616e+00 -5.55075169e-01 -4.00068909e-01 8.02755952e-02
3.15632284e-01 -7.43745625e-01 -1.13148797e+00 5.54651320e-01
5.94503820e-01 -5.53887784e-01 1.37065864e+00 1.07889645e-01
-6.72608137e-01 -8.07272375e-01 -3.78705055e-01 -1.10959864e+00
-3.32314074e-02 -6.26898050e-01 4.95714724e-01 1.08672178e+00
1.67701349e-01 -8.99331152e-01 7.11606026e-01 6.47675991e-01
-4.68942933e-02 -1.40416682e-01 -1.35456586e+00 -1.01120138e+00
-2.99613208e-01 -3.81921828e-01 7.90962160e-01 9.25284922e-01
-1.19639397e-01 2.05408722e-01 -1.32177174e-01 6.31851673e-01
8.48201394e-01 1.72268793e-01 1.38363874e+00 -1.85028172e+00
6.79113388e-01 -3.78941417e-01 -1.32772744e+00 -1.07018387e+00
1.27655724e-02 -1.05152035e+00 8.17102492e-02 -1.34427357e+00
-7.17755973e-01 -1.06790638e+00 5.08459732e-02 3.63793373e-01
1.68699592e-01 5.32934666e-01 1.25012159e-01 4.49523389e-01
-1.86411321e-01 3.89539063e-01 5.20397186e-01 1.03941619e-01
-5.03429353e-01 1.67475477e-01 -1.13155045e-01 6.89794600e-01
5.62056839e-01 -3.85945022e-01 -7.35296635e-04 -3.40781718e-01
1.55982867e-01 -3.02746475e-01 3.55090022e-01 -1.15563989e+00
5.77402592e-01 -7.20254853e-02 7.77501911e-02 -8.41078341e-01
5.83146095e-01 -1.22149122e+00 4.84345227e-01 3.36197674e-01
2.33762965e-01 4.15904760e-01 2.09040657e-01 5.84949613e-01
-2.31130719e-01 -1.89441353e-01 5.27121603e-01 3.34962279e-01
-7.55757987e-01 3.56105566e-01 2.67244130e-01 -5.38642585e-01
1.04276323e+00 -5.61686397e-01 -6.20208168e-03 -3.82139176e-01
-3.25400859e-01 1.39011994e-01 1.07220960e+00 4.27395195e-01
8.32874000e-01 -1.56207204e+00 -6.74073815e-01 3.91417384e-01
4.64094132e-01 3.61669391e-01 -2.96365768e-01 1.13786519e+00
-9.15210545e-01 4.53129530e-01 -1.63099036e-01 -1.28769767e+00
-1.51373780e+00 2.70835668e-01 2.94198573e-01 3.22256655e-01
-6.69037402e-01 3.60881567e-01 -4.54889864e-01 -6.62408888e-01
-6.34599254e-02 -4.29755300e-01 1.03069283e-01 4.41016965e-02
2.07037434e-01 5.56224465e-01 3.31957191e-01 -1.21791172e+00
-7.93159783e-01 1.41209948e+00 2.37490028e-01 -3.26208659e-02
1.15507269e+00 -3.56748015e-01 -1.96548462e-01 3.37225437e-01
1.14466977e+00 3.01780105e-01 -7.60573626e-01 -4.70920652e-01
5.70123196e-01 -8.25463951e-01 -1.68937311e-01 -2.02405453e-02
-6.18642330e-01 6.92111552e-01 8.02911937e-01 9.42503661e-02
5.29934168e-01 -2.17807353e-01 4.02051181e-01 6.01431608e-01
9.42689657e-01 -7.40996778e-01 -7.19548225e-01 5.33936203e-01
9.46823657e-01 -1.21907008e+00 4.22443658e-01 -7.75154829e-01
3.55697945e-02 1.28599215e+00 1.78351745e-01 -3.59661698e-01
6.33311331e-01 -9.19208005e-02 2.07150578e-01 -1.28975943e-01
2.31791779e-01 -2.88072914e-01 6.95964515e-01 7.46487558e-01
2.67342955e-01 7.08730891e-02 -4.62335825e-01 -2.85528928e-01
-7.01559663e-01 -4.27507997e-01 3.24693859e-01 1.01967871e+00
-4.99792397e-01 -1.37055898e+00 -7.03473389e-01 1.30039707e-01
2.46575117e-01 1.70519397e-01 -2.88743347e-01 1.01181579e+00
-3.12176466e-01 8.26699317e-01 3.16257834e-01 -4.62812394e-01
6.39263213e-01 -6.46075932e-03 5.73082566e-01 -4.92812634e-01
-1.07097961e-01 1.08282998e-01 -1.09691858e-01 -7.73194969e-01
-7.60299683e-01 -8.84735107e-01 -1.43289757e+00 -4.16807622e-01
-6.55083060e-01 1.15701810e-01 1.16176987e+00 7.78360426e-01
5.54751575e-01 -2.51748919e-01 4.78468150e-01 -1.09967160e+00
-3.72088611e-01 -5.63894689e-01 -4.04157311e-01 3.56801063e-01
3.48712564e-01 -1.07791269e+00 -2.03886494e-01 -2.07658127e-01] | [7.555839538574219, -2.4481711387634277] |
aab6d14f-e0af-4b2f-b704-f3488d2e31d9 | open-high-resolution-satellite-imagery-the | 2207.06418 | null | https://arxiv.org/abs/2207.06418v1 | https://arxiv.org/pdf/2207.06418v1.pdf | Open High-Resolution Satellite Imagery: The WorldStrat Dataset -- With Application to Super-Resolution | Analyzing the planet at scale with satellite imagery and machine learning is a dream that has been constantly hindered by the cost of difficult-to-access highly-representative high-resolution imagery. To remediate this, we introduce here the WorldStrat dataset. The largest and most varied such publicly available dataset, at Airbus SPOT 6/7 satellites' high resolution of up to 1.5 m/pixel, empowered by European Space Agency's Phi-Lab as part of the ESA-funded QueryPlanet project, we curate nearly 10,000 sqkm of unique locations to ensure stratified representation of all types of land-use across the world: from agriculture to ice caps, from forests to multiple urbanization densities. We also enrich those with locations typically under-represented in ML datasets: sites of humanitarian interest, illegal mining sites, and settlements of persons at risk. We temporally-match each high-resolution image with multiple low-resolution images from the freely accessible lower-resolution Sentinel-2 satellites at 10 m/pixel. We accompany this dataset with an open-source Python package to: rebuild or extend the WorldStrat dataset, train and infer baseline algorithms, and learn with abundant tutorials, all compatible with the popular EO-learn toolbox. We hereby hope to foster broad-spectrum applications of ML to satellite imagery, and possibly develop from free public low-resolution Sentinel2 imagery the same power of analysis allowed by costly private high-resolution imagery. We illustrate this specific point by training and releasing several highly compute-efficient baselines on the task of Multi-Frame Super-Resolution. High-resolution Airbus imagery is CC BY-NC, while the labels and Sentinel2 imagery are CC BY, and the source code and pre-trained models under BSD. The dataset is available at https://zenodo.org/record/6810792 and the software package at https://github.com/worldstrat/worldstrat . | ['Freddie Kalaitzis', 'Ivan Oršolić', 'Julien Cornebise'] | 2022-07-13 | null | null | null | null | ['multi-frame-super-resolution'] | ['computer-vision'] | [ 3.60293478e-01 -1.42690852e-01 -2.11668223e-01 -2.98338354e-01
-1.23749626e+00 -6.88925982e-01 8.24810565e-01 -2.35929847e-01
-6.42271578e-01 9.54418480e-01 3.24484289e-01 -2.95395344e-01
-3.14670980e-01 -1.16357148e+00 -7.17106402e-01 -8.86508048e-01
-6.88220799e-01 5.86582065e-01 -1.70018926e-01 -4.95658517e-01
-3.28666627e-01 6.74626172e-01 -1.60873520e+00 2.48448282e-01
7.72087812e-01 8.73177886e-01 3.18221092e-01 6.88037097e-01
3.75719070e-01 2.20805809e-01 7.68858045e-02 1.31953344e-01
8.39181244e-01 -8.09655711e-02 -7.18739927e-01 -2.11150050e-01
7.04527617e-01 -4.14294928e-01 -1.84271455e-01 1.00236392e+00
4.46139336e-01 -4.72386032e-02 4.10297751e-01 -8.85004103e-01
-5.63735850e-02 2.22731113e-01 -7.67933786e-01 5.37093103e-01
1.79978371e-01 2.90396839e-01 8.71738613e-01 -6.44279718e-01
8.42761815e-01 1.14367664e+00 1.12516904e+00 -9.08341184e-02
-1.49728537e+00 -7.34101832e-01 -6.40142560e-02 -2.03164265e-01
-1.81813395e+00 -6.12540305e-01 -4.98170629e-02 -7.06890881e-01
9.95612383e-01 6.97340727e-01 4.59130168e-01 1.01675272e+00
-3.73284076e-03 1.57372206e-01 1.42486978e+00 6.61601350e-02
-8.21598545e-02 -2.66654849e-01 -1.83420211e-01 4.37418252e-01
1.10770002e-01 4.29889381e-01 -1.23610280e-01 -3.95611316e-01
7.51192570e-01 2.10129052e-01 -2.85921246e-01 8.01078528e-02
-1.31535554e+00 7.95713902e-01 9.06679571e-01 1.45463392e-01
-6.69649243e-01 -1.39238387e-01 -3.13364752e-02 1.39949203e-01
9.32724774e-01 2.16973349e-01 -5.10329962e-01 7.75613263e-02
-1.25974953e+00 4.32664335e-01 2.88546771e-01 5.68489730e-01
1.13516843e+00 -1.33625343e-01 4.79167491e-01 7.18859732e-01
7.10829198e-02 1.02038789e+00 -1.54723525e-01 -1.14100075e+00
5.07795334e-01 3.61604810e-01 4.77520496e-01 -9.42727447e-01
-8.06225300e-01 -6.24825239e-01 -9.59492266e-01 1.59761935e-01
1.97351262e-01 -4.11278099e-01 -6.81751668e-01 1.46707308e+00
5.78539073e-01 2.24059541e-02 -6.25931425e-03 1.13274109e+00
4.85501677e-01 8.37812483e-01 -8.16425588e-03 -1.99468457e-03
1.47000229e+00 -3.98933917e-01 -2.80474365e-01 -4.29137200e-01
6.84092820e-01 -2.27085069e-01 6.70432031e-01 -2.87645739e-02
-3.43822688e-01 -4.21738625e-01 -6.87111557e-01 3.58879894e-01
-8.03525269e-01 3.20917293e-02 7.68575013e-01 1.89404190e-01
-1.08937669e+00 6.01828098e-01 -9.72715616e-01 -7.69550562e-01
7.15666533e-01 -2.06793286e-02 -6.49929225e-01 2.04092190e-02
-1.43826222e+00 8.84409487e-01 3.91272306e-01 3.05462122e-01
-9.15051520e-01 -8.72346401e-01 -8.85162413e-01 -3.75721246e-01
1.53206855e-01 -5.28551936e-01 6.64613485e-01 -9.65157211e-01
-5.14112175e-01 1.16646016e+00 1.43873855e-01 -6.68935359e-01
3.77308697e-01 -3.27159822e-01 -7.37846613e-01 1.62921324e-01
7.44127810e-01 8.97145689e-01 3.94924521e-01 -9.94031966e-01
-9.73743856e-01 -7.17045963e-01 -1.17030054e-01 2.38102227e-01
-9.39802602e-02 3.06723356e-01 7.57360980e-02 -4.55679297e-01
-2.27017533e-02 -1.02566075e+00 -5.21590650e-01 -5.38482331e-03
-1.11086167e-01 3.96366596e-01 5.34662187e-01 -1.12096524e+00
9.72974777e-01 -2.03851533e+00 9.39793326e-03 -7.64088407e-02
-1.33183032e-01 8.22535008e-02 -2.97960550e-01 5.45032203e-01
-2.30137736e-01 3.44506472e-01 -7.28231251e-01 5.91050796e-02
-1.90273657e-01 2.16781497e-01 -4.28741843e-01 8.79917562e-01
2.37293005e-01 6.19796515e-01 -9.91253674e-01 -1.97568953e-01
3.79898787e-01 6.12659991e-01 -1.54402241e-01 -6.94613159e-02
-1.93926632e-01 9.43151236e-01 -4.03739274e-01 7.94614136e-01
1.13423526e+00 -1.84544977e-02 1.34574488e-01 -1.01697691e-01
-7.99003065e-01 2.57901460e-01 -1.13226020e+00 1.65036380e+00
-1.57533631e-01 4.36186314e-01 6.59152210e-01 -6.37444556e-01
6.72066569e-01 -4.31042053e-02 5.11369646e-01 -7.82076299e-01
-3.51661295e-01 1.87724367e-01 -3.48706126e-01 -6.04261100e-01
5.52507281e-01 -3.05624396e-01 -9.43023339e-02 2.61251718e-01
-3.49630475e-01 -2.41840839e-01 -7.06623718e-02 -1.04327826e-02
8.47511351e-01 5.10631621e-01 2.64006525e-01 -5.20884871e-01
9.44846869e-03 6.08479679e-01 6.55192375e-01 6.78333342e-01
-1.53638190e-02 5.92208087e-01 1.78882375e-01 -8.43031287e-01
-1.14151812e+00 -1.04369509e+00 -1.01415002e+00 1.08757925e+00
-3.89658868e-01 -1.82151556e-01 -2.53002077e-01 -2.31003091e-01
2.77545840e-01 5.06515145e-01 -6.50884628e-01 6.37234807e-01
-2.07555756e-01 -1.44146252e+00 9.41244543e-01 -3.71899046e-02
7.78153956e-01 -6.08656287e-01 -5.70164084e-01 -7.39393532e-02
-4.38586742e-01 -1.03528714e+00 3.92817527e-01 1.01070702e-01
-8.24985921e-01 -9.10386503e-01 -5.60512662e-01 -6.54160306e-02
2.59217471e-01 2.86387682e-01 1.15741134e+00 -3.08729321e-01
-4.84597802e-01 2.08675474e-01 -3.06172401e-01 -5.68217896e-02
3.54876816e-02 2.63938159e-01 2.67849833e-01 -1.36508094e-02
2.02745512e-01 -7.37794161e-01 -5.61347961e-01 2.70056933e-01
-8.27277005e-01 2.68898696e-01 3.98899525e-01 3.87418181e-01
8.26139331e-01 1.92817107e-01 2.00710177e-01 -4.21756238e-01
-4.34864253e-01 -1.21865499e+00 -8.26788485e-01 -1.15009047e-01
3.87717225e-02 -3.60218406e-01 1.47405729e-01 3.33550721e-01
-8.85504663e-01 2.83425570e-01 -1.59323052e-01 -6.45593926e-02
-6.30990386e-01 7.80153573e-01 1.34922042e-01 1.76377017e-02
9.07178044e-01 1.47243693e-01 2.30075512e-02 -7.18787670e-01
3.51271778e-01 1.09182143e+00 6.79870963e-01 -4.18766826e-01
1.08151948e+00 1.07130289e+00 -8.15224051e-02 -1.17762530e+00
-1.14936936e+00 -5.27318299e-01 -7.04260886e-01 -1.62075236e-01
8.63372862e-01 -1.67174482e+00 6.44283891e-02 4.83876854e-01
-5.62609971e-01 -6.53317451e-01 -1.56780742e-02 4.36287582e-01
-2.86929697e-01 1.11376047e-01 -1.44150183e-01 -8.30134988e-01
-4.25911754e-01 -7.89433897e-01 1.28782868e+00 1.07602343e-01
7.22847471e-04 -8.03172112e-01 3.76006991e-01 4.20558125e-01
6.99832201e-01 1.09828305e+00 1.13221399e-01 -8.14550146e-02
-6.21722579e-01 -3.03749256e-02 -4.54344094e-01 1.19820036e-01
3.21171135e-01 -1.78551346e-01 -9.26461995e-01 -5.87074816e-01
-2.11720333e-01 -4.16747302e-01 1.31239808e+00 4.34243649e-01
8.67079139e-01 -3.95370036e-01 -4.88662392e-01 1.23116159e+00
1.69926310e+00 -4.31159437e-01 6.84463382e-01 8.58843923e-01
4.35554624e-01 7.92291939e-01 8.17831099e-01 6.56224370e-01
4.33801711e-01 7.42204964e-01 8.58736932e-01 -4.06937510e-01
2.45319813e-01 9.89238620e-02 2.83972144e-01 -3.03583533e-01
-5.39498925e-01 3.89111340e-01 -1.35114086e+00 6.09484494e-01
-1.72142112e+00 -1.42790997e+00 -5.96263528e-01 2.35222912e+00
6.98038042e-01 -3.39470237e-01 1.33470595e-01 -5.07448137e-01
5.41832387e-01 6.70800388e-01 -4.33053613e-01 4.34216529e-01
-6.15702152e-01 1.02040514e-01 1.22537386e+00 7.46499300e-01
-1.60260594e+00 1.01780021e+00 5.93977404e+00 6.96228683e-01
-1.07995379e+00 3.62945646e-01 5.16912818e-01 -5.11954129e-01
-3.00075233e-01 4.10929210e-02 -9.22248721e-01 3.47723573e-01
1.38696468e+00 1.67486787e-01 6.70889378e-01 7.00533926e-01
6.75092101e-01 -2.37250641e-01 -2.70198882e-01 6.15946352e-01
-3.98533762e-01 -1.52625477e+00 -2.74278790e-01 4.97945160e-01
6.39874935e-01 1.24411964e+00 -2.95903813e-02 1.21631764e-01
4.41974938e-01 -1.12953258e+00 5.33830941e-01 6.20188832e-01
1.33328998e+00 -5.72224498e-01 4.79824424e-01 3.71598333e-01
-1.33624947e+00 -2.25622743e-01 -5.07264078e-01 -3.87629539e-01
1.32066399e-01 7.40505219e-01 -3.25672746e-01 8.81576002e-01
1.28776419e+00 1.11911964e+00 -6.28660977e-01 8.27395320e-01
-4.32240777e-03 4.59428340e-01 -8.90315831e-01 8.11184227e-01
5.18328428e-01 -3.45177799e-01 7.58952260e-01 1.20930707e+00
5.57533801e-01 2.47470871e-01 3.00475836e-01 6.94986999e-01
2.24802181e-01 -2.14736596e-01 -9.43077564e-01 1.61047831e-01
4.83098924e-01 1.59929466e+00 -2.92711496e-01 -8.33794773e-02
-2.77276039e-01 6.04233027e-01 1.39647171e-01 2.95925856e-01
-8.95323634e-01 4.52000275e-02 1.20529342e+00 5.34522355e-01
5.75905778e-02 -3.22311580e-01 1.19242050e-01 -1.23050523e+00
-2.80572891e-01 -8.90100420e-01 5.32958090e-01 -1.04862404e+00
-1.01088893e+00 4.84236509e-01 3.02182674e-01 -1.45482934e+00
-1.09156542e-01 -3.82377118e-01 -2.80921847e-01 1.17588258e+00
-1.94415009e+00 -1.50066519e+00 -5.05037725e-01 3.82202774e-01
2.56714895e-02 -4.92491685e-02 9.88854349e-01 2.17108414e-01
-5.09119987e-01 -3.36318970e-01 5.34677148e-01 4.63917591e-02
5.21096826e-01 -8.44605446e-01 6.59567475e-01 7.19510555e-01
-2.54090160e-01 4.60086465e-01 6.83472216e-01 -7.46707201e-01
-1.21977961e+00 -1.60218251e+00 6.84073508e-01 -4.94068146e-01
1.11237442e+00 -2.83469886e-01 -6.95237875e-01 1.20259571e+00
-9.58290398e-02 1.75954252e-01 5.28160095e-01 2.04218209e-01
-2.43494049e-01 -4.35379773e-01 -1.27453005e+00 2.29932845e-01
8.85944724e-01 -5.23840308e-01 -4.28730577e-01 8.96919012e-01
4.69881475e-01 -2.57579654e-01 -1.15626252e+00 5.88536680e-01
5.52520573e-01 -1.24784529e+00 1.02532494e+00 -2.89978325e-01
3.04318458e-01 -5.59698999e-01 -6.26872301e-01 -1.22356677e+00
-5.73889196e-01 -2.37206623e-01 5.48572898e-01 1.06216848e+00
2.93755651e-01 -8.73553991e-01 4.09090400e-01 8.06981027e-02
-1.96992457e-01 -7.15307146e-02 -1.31620622e+00 -8.85132730e-01
3.06851715e-01 -4.41145301e-01 6.87836647e-01 1.18245220e+00
-5.23122847e-01 -2.35367268e-01 -3.91914487e-01 8.85348558e-01
9.36421394e-01 1.79547414e-01 8.91400456e-01 -1.35501635e+00
-1.32899836e-01 -1.03677385e-01 -1.66254640e-01 -4.13202345e-01
-2.49352790e-02 -8.47146273e-01 -2.78050214e-01 -1.35603786e+00
2.34958738e-01 -7.71635890e-01 4.43795323e-02 8.86288285e-01
1.15606681e-01 5.36287129e-01 -1.80299327e-01 6.40805066e-01
-3.20681393e-01 4.19933110e-01 6.31968856e-01 -7.14578992e-03
9.49086100e-02 -2.58627087e-01 -3.45571160e-01 6.15820408e-01
8.26943636e-01 -6.34869516e-01 4.19511974e-01 -6.15081131e-01
3.47378343e-01 -4.58601676e-03 9.57919478e-01 -1.29649067e+00
-3.66772920e-01 -5.02359450e-01 4.77906317e-01 -8.45254838e-01
3.78387928e-01 -8.96621883e-01 9.65692997e-01 4.28213388e-01
2.24599943e-01 -2.90659904e-01 4.68996167e-01 4.71505001e-02
7.02027306e-02 1.28461942e-01 9.78594422e-01 -4.89723831e-01
-8.30037415e-01 5.90323985e-01 -2.96607047e-01 -2.00071082e-01
7.93365657e-01 9.39641148e-02 -6.07196033e-01 -1.42367661e-01
-6.22428656e-01 4.84135836e-01 9.98073637e-01 1.65924028e-01
-1.47016034e-01 -1.05925584e+00 -1.29841852e+00 4.13197935e-01
2.60333627e-01 2.36155801e-02 5.65227091e-01 9.24420714e-01
-6.78977847e-01 4.01749283e-01 -2.97938645e-01 -6.82588875e-01
-9.42550242e-01 9.98681188e-02 7.77845740e-01 -2.30283797e-01
-7.02829301e-01 4.08084720e-01 -5.89525588e-02 -1.03434467e+00
-4.65473622e-01 4.67162533e-03 -5.31663969e-02 4.04711545e-01
9.72014725e-01 1.09160356e-01 -9.92031544e-02 -1.12986279e+00
-5.13077497e-01 5.66064537e-01 4.74616468e-01 -1.24738030e-01
1.91489768e+00 -3.49020988e-01 -3.53826255e-01 2.85529643e-01
8.51345360e-01 -2.13391900e-01 -1.52474165e+00 -1.77501231e-01
-3.40730324e-02 -4.78393018e-01 4.65401322e-01 -8.52908373e-01
-9.14665043e-01 5.34322619e-01 8.29909921e-01 1.43555924e-01
1.04575431e+00 4.86807190e-02 3.47913414e-01 2.62270689e-01
6.38662100e-01 -8.19452524e-01 -8.52388203e-01 4.55195606e-01
1.09616351e+00 -1.37362874e+00 3.49495858e-01 -1.68567132e-02
-2.98997849e-01 6.70919836e-01 1.37795731e-01 -1.64806697e-04
4.33601588e-01 2.49460250e-01 1.12909153e-02 -3.11024964e-01
-4.11154062e-01 -7.27791011e-01 -2.06186727e-01 7.88544893e-01
1.86930567e-01 4.57896858e-01 2.19566926e-01 3.58549833e-01
-2.96083927e-01 1.07544363e-01 3.34628105e-01 7.78699577e-01
-6.05300784e-01 -6.93700731e-01 -1.00421441e+00 5.59444547e-01
-2.28509068e-01 -5.30251861e-01 1.76669490e-02 7.66730666e-01
2.05176577e-01 6.27750337e-01 4.12374616e-01 -1.41216591e-01
5.32254204e-02 -1.24585629e-01 4.79129218e-02 -4.98541772e-01
-4.26715314e-01 -8.48378465e-02 4.06593591e-01 -7.67456710e-01
-7.44534135e-01 -1.17928731e+00 -8.92163932e-01 -5.77440917e-01
3.60174328e-01 1.21618778e-01 7.60588765e-01 7.09288776e-01
5.85251331e-01 -7.02851489e-02 5.93869448e-01 -1.63232005e+00
-2.11493999e-01 -1.26423371e+00 -8.91794443e-01 -6.67181157e-04
5.60517669e-01 -4.30521637e-01 -7.29059041e-01 -2.51653314e-01] | [9.429708480834961, -1.486242651939392] |
63bd6bd1-7073-4c8c-ad9f-a33f8993f8db | syntax-and-domain-aware-model-for | 2302.03908 | null | https://arxiv.org/abs/2302.03908v2 | https://arxiv.org/pdf/2302.03908v2.pdf | Syntax and Domain Aware Model for Unsupervised Program Translation | There is growing interest in software migration as the development of software and society. Manually migrating projects between languages is error-prone and expensive. In recent years, researchers have begun to explore automatic program translation using supervised deep learning techniques by learning from large-scale parallel code corpus. However, parallel resources are scarce in the programming language domain, and it is costly to collect bilingual data manually. To address this issue, several unsupervised programming translation systems are proposed. However, these systems still rely on huge monolingual source code to train, which is very expensive. Besides, these models cannot perform well for translating the languages that are not seen during the pre-training procedure. In this paper, we propose SDA-Trans, a syntax and domain-aware model for program translation, which leverages the syntax structure and domain knowledge to enhance the cross-lingual transfer ability. SDA-Trans adopts unsupervised training on a smaller-scale corpus, including Python and Java monolingual programs. The experimental results on function translation tasks between Python, Java, and C++ show that SDA-Trans outperforms many large-scale pre-trained models, especially for unseen language translation. | ['Li Zhang', 'Jia Li', 'Fang Liu'] | 2023-02-08 | null | null | null | null | ['cross-lingual-transfer'] | ['natural-language-processing'] | [-2.82330900e-01 -5.43812215e-01 -6.32288575e-01 -5.24597704e-01
-8.50063801e-01 -6.58996046e-01 9.63414162e-02 -1.34179443e-01
-3.46783519e-01 3.44277352e-01 -3.39167789e-02 -7.41435349e-01
6.37084484e-01 -7.79037058e-01 -1.01491034e+00 -7.94839114e-02
4.43440706e-01 1.84818760e-01 1.34012997e-01 -2.65941858e-01
8.33868608e-02 -2.10379034e-01 -1.00353038e+00 5.81431806e-01
1.44876480e+00 -2.62353998e-02 4.91715848e-01 1.49317667e-01
-7.51765907e-01 7.39774764e-01 -3.91694814e-01 -5.65839946e-01
1.25379845e-01 -6.93115771e-01 -1.01934302e+00 -3.68129522e-01
2.46375427e-01 -3.30792487e-01 1.06759602e-02 1.34206569e+00
3.93297583e-01 -5.78110874e-01 6.27240224e-04 -1.27385974e+00
-1.04047465e+00 1.10853052e+00 -7.10575581e-01 8.05059075e-02
2.11945698e-01 1.47052914e-01 8.63541305e-01 -1.03529906e+00
7.30169117e-01 9.55564857e-01 7.80674696e-01 6.19395912e-01
-1.18672597e+00 -8.12131822e-01 -1.36080176e-01 -7.74621144e-02
-1.39261758e+00 -2.63477474e-01 8.94933820e-01 -8.56360197e-01
1.31573510e+00 -1.93025425e-01 2.81699657e-01 1.07583749e+00
2.10187674e-01 8.31152797e-01 1.06407082e+00 -7.92714417e-01
-2.72567153e-01 5.18194377e-01 5.90560250e-02 8.29441667e-01
7.41984770e-02 -1.30230054e-01 -4.26684946e-01 -2.35627323e-01
3.91163021e-01 4.05970477e-02 -3.20149690e-01 -2.75245190e-01
-1.46490657e+00 6.53691113e-01 2.09222868e-01 6.42771006e-01
-2.53473246e-03 -1.92612648e-01 8.99745882e-01 7.84404337e-01
4.05075520e-01 3.83547246e-01 -6.26777768e-01 -4.67087120e-01
-8.65300596e-01 -8.36153328e-02 8.64949405e-01 1.46510017e+00
1.19970453e+00 5.09957187e-02 4.25221294e-01 1.01881564e+00
3.89589697e-01 5.87957382e-01 8.66914928e-01 -5.17732561e-01
1.10000265e+00 1.19719350e+00 -2.99881071e-01 -9.16090548e-01
3.87346582e-03 -1.74347863e-01 -4.80225235e-01 -1.13227606e-01
2.29256779e-01 -1.57594621e-01 -2.75004804e-01 1.65144134e+00
1.35046288e-01 -4.50756878e-01 2.40966305e-01 7.67711520e-01
6.89854741e-01 6.13592505e-01 -1.04041450e-01 8.57968405e-02
9.62689340e-01 -1.40450966e+00 -3.12316179e-01 -4.23533976e-01
1.25481892e+00 -1.16771710e+00 1.72139740e+00 -8.08678493e-02
-7.77479351e-01 -7.49737561e-01 -1.09180319e+00 -2.00299695e-01
-2.89928526e-01 6.17027104e-01 7.69969583e-01 6.62937760e-01
-9.66272414e-01 3.44258517e-01 -1.09441459e+00 -6.77223265e-01
2.44788736e-01 1.60788611e-01 -2.96246082e-01 -2.44155034e-01
-9.28914249e-01 7.02737391e-01 4.09910321e-01 1.26688266e-02
-6.87821090e-01 -5.73505342e-01 -8.25127244e-01 8.45549721e-03
-5.83840646e-02 -3.79645377e-01 1.37925446e+00 -1.57949412e+00
-1.65208042e+00 1.15156519e+00 -9.00224373e-02 1.37485042e-01
3.85549217e-01 -1.82729766e-01 -5.65413117e-01 -5.35863638e-01
5.22930861e-01 3.51959437e-01 3.04935545e-01 -9.03688610e-01
-5.51863670e-01 -2.74761438e-01 1.20047778e-01 1.75397806e-02
-7.94982493e-01 6.28171742e-01 -5.39608657e-01 -3.90238851e-01
-2.38874212e-01 -9.20084476e-01 -4.66656797e-02 -3.40441018e-01
-8.43000859e-02 -2.14662313e-01 5.36490619e-01 -9.14911449e-01
1.31988811e+00 -2.30510378e+00 1.88855991e-01 -1.46500900e-01
3.01633589e-02 1.68158889e-01 -3.96722019e-01 6.07239962e-01
1.05980881e-01 1.12485096e-01 -3.77461404e-01 9.72458124e-02
8.83477852e-02 1.15990587e-01 -3.90231133e-01 1.85496554e-01
1.28802344e-01 9.50337648e-01 -9.46617544e-01 -6.59242630e-01
-4.18909073e-01 -7.05408445e-03 -8.34069431e-01 3.72137398e-01
-2.38090873e-01 5.11856318e-01 -5.95492363e-01 8.70547295e-01
6.18993342e-01 -4.24025387e-01 5.41892111e-01 2.91123599e-01
-3.61238599e-01 6.53373301e-01 -4.95179355e-01 2.26062322e+00
-9.04750586e-01 7.49946713e-01 -2.16768652e-01 -1.00717223e+00
1.02671278e+00 3.10294896e-01 2.26464987e-01 -7.49654591e-01
-1.80676237e-01 8.32717896e-01 2.00669825e-01 -8.48958492e-01
4.02581841e-01 8.13129172e-02 -4.18546855e-01 7.23978579e-01
4.88156937e-02 1.38350666e-01 3.09088409e-01 4.88524660e-02
1.00143576e+00 6.04776740e-01 5.09170070e-03 -3.75219911e-01
7.09723532e-01 5.65095663e-01 9.43376303e-01 2.19695687e-01
-5.68864308e-02 1.97091475e-01 3.61828595e-01 -4.81769979e-01
-9.51236963e-01 -7.35040009e-01 2.18339056e-01 1.32818615e+00
-6.01761825e-02 -5.70522189e-01 -9.75809813e-01 -9.96480227e-01
-1.96927592e-01 3.45503271e-01 9.21845902e-03 -1.99164256e-01
-9.99538064e-01 -7.54115760e-01 8.19427431e-01 6.53371215e-01
4.91612077e-01 -1.10172701e+00 -1.30816400e-01 3.40140820e-01
-5.58946013e-01 -1.11186528e+00 -8.70657563e-01 -1.65078431e-01
-9.52917099e-01 -9.42642510e-01 -5.04546583e-01 -1.52468550e+00
9.88176882e-01 2.66088694e-01 1.44931555e+00 2.19283625e-01
1.39499620e-01 -5.48233055e-02 -3.35428506e-01 2.22361106e-02
-9.75310266e-01 6.08438432e-01 -1.15630843e-01 -4.10683244e-01
1.07717359e+00 -6.66583776e-01 -1.36087999e-01 3.84956837e-01
-5.72921574e-01 2.50053704e-01 8.49394321e-01 7.59191692e-01
2.34102413e-01 -2.84331858e-01 5.33592641e-01 -1.12674499e+00
5.57239532e-01 -5.94820917e-01 -9.26191807e-01 4.85364288e-01
-6.08391166e-01 -9.73284245e-02 1.13242555e+00 -5.55539727e-01
-1.15723276e+00 -3.81928869e-02 -2.78990846e-02 -9.77141783e-02
1.00496532e-02 1.07917535e+00 -2.63264596e-01 -1.30242571e-01
8.79451156e-01 5.24450123e-01 -3.60091239e-01 -6.28644645e-01
6.00803941e-02 1.15369582e+00 3.78067583e-01 -1.07207620e+00
8.70129704e-01 3.66832577e-02 -8.12807441e-01 -4.30281937e-01
-2.59156495e-01 -2.98807859e-01 -7.88484156e-01 1.35128036e-01
5.95319152e-01 -1.12933242e+00 9.03110299e-03 5.77726603e-01
-1.56314254e+00 -5.21371782e-01 2.73429304e-01 6.76855326e-01
-2.64365852e-01 3.84943277e-01 -8.73998404e-01 6.50228932e-02
-3.81973147e-01 -1.62445152e+00 7.11613953e-01 1.24656156e-01
-1.44949660e-01 -9.04820502e-01 4.82402861e-01 4.54467654e-01
6.43429518e-01 -3.55210781e-01 1.35173714e+00 -3.05536747e-01
-8.04101884e-01 -5.75182475e-02 -2.91326821e-01 5.11663437e-01
3.66936982e-01 1.47307441e-01 -4.29277748e-01 -5.13245285e-01
-4.12410758e-02 -5.00788808e-01 1.23087645e-01 -4.26020443e-01
8.83371830e-01 -1.08377464e-01 -1.93176582e-01 7.88665891e-01
1.37986493e+00 1.40470207e-01 2.19804347e-01 4.94991004e-01
1.02550733e+00 5.45225084e-01 4.58304405e-01 -9.25427303e-02
9.14100587e-01 5.04539013e-01 -2.13107303e-01 -1.19134814e-01
3.37198144e-03 -4.75428492e-01 9.22753692e-01 2.06119823e+00
2.20592737e-01 4.41161513e-01 -1.54083562e+00 8.96104634e-01
-1.65443838e+00 -4.44308907e-01 -2.66600490e-01 2.02355099e+00
1.42882311e+00 -1.41850427e-01 -1.37606651e-01 -6.34252369e-01
6.95062399e-01 -3.14744979e-01 -5.03272235e-01 -3.05761755e-01
6.36119395e-02 8.40483010e-02 2.77045190e-01 2.26803496e-01
-7.46620238e-01 1.20925081e+00 5.29884624e+00 4.50494438e-01
-1.55037558e+00 5.52520096e-01 9.04384702e-02 4.52446401e-01
-4.60189581e-01 5.28553307e-01 -7.94949651e-01 7.51857162e-01
9.70679343e-01 -6.02437496e-01 5.93447685e-01 1.28063905e+00
-5.06998189e-02 3.32645833e-01 -1.43729925e+00 8.16494644e-01
8.33243504e-02 -1.14274907e+00 -1.89775959e-01 -1.54968947e-01
1.04157472e+00 7.15651810e-01 -2.90389150e-01 8.98368001e-01
5.15242994e-01 -5.83020449e-01 5.12580276e-01 -1.83637559e-01
7.24968970e-01 -5.32899261e-01 6.63542151e-01 6.43989623e-01
-1.15822649e+00 3.19122046e-01 -6.07038260e-01 5.09054214e-02
-3.64920944e-01 3.68387491e-01 -4.93545294e-01 4.62327808e-01
8.30549896e-01 9.14966106e-01 -6.81664884e-01 6.66893601e-01
-3.72048199e-01 7.31356859e-01 -3.22505273e-02 3.67654264e-02
6.86166063e-02 -2.73723453e-01 -4.59135473e-02 1.44721472e+00
5.74456394e-01 -5.79236865e-01 6.45821154e-01 1.40375483e+00
-4.87933815e-01 5.56904912e-01 -7.78819263e-01 -4.69384313e-01
2.89837360e-01 1.08284986e+00 -4.17389870e-01 -3.33793133e-01
-1.23005223e+00 1.07282114e+00 5.64405441e-01 3.90424222e-01
-8.77287209e-01 -7.82825708e-01 4.93618190e-01 -1.53267622e-01
-3.84728163e-02 -3.77885789e-01 -1.26511529e-01 -1.72378254e+00
6.16418958e-01 -1.48519456e+00 3.38982232e-02 -3.79230827e-01
-1.27648699e+00 8.44199479e-01 -3.09033990e-01 -1.58207047e+00
-9.35556293e-02 -3.12378764e-01 -6.48305714e-01 1.10487390e+00
-1.62376904e+00 -1.31993258e+00 -1.21802077e-01 3.70998681e-01
5.58894336e-01 -5.63409209e-01 7.48504877e-01 9.20012891e-01
-7.00483322e-01 9.08280790e-01 3.28615040e-01 6.43465400e-01
1.09701788e+00 -8.85968685e-01 7.99753070e-01 1.15837455e+00
-3.85033190e-02 1.28557622e+00 6.71251565e-02 -6.37380421e-01
-1.70469582e+00 -1.28253639e+00 1.30226755e+00 -4.61602509e-01
7.89201558e-01 -4.69177455e-01 -1.25700808e+00 8.96496058e-01
4.70736414e-01 -9.84835089e-04 9.43853259e-01 1.47997662e-01
-6.44669712e-01 -1.39959395e-01 -6.43155098e-01 5.20418584e-01
8.54621530e-01 -1.14728546e+00 -5.22320867e-01 4.72446829e-01
6.27989411e-01 -4.56976712e-01 -1.11177003e+00 1.00299463e-01
3.40913534e-01 -6.10611677e-01 4.22136158e-01 -6.36917651e-01
8.69263232e-01 -4.48736697e-01 -1.77450880e-01 -1.14998698e+00
2.68071555e-02 -5.49587905e-01 3.53032857e-01 1.55194879e+00
6.55623972e-01 -6.82714045e-01 6.15300357e-01 3.73524457e-01
-2.13463187e-01 -3.35451752e-01 -4.50232774e-01 -1.04452515e+00
6.52105093e-01 -8.00790936e-02 8.57066035e-01 1.71256185e+00
4.43216711e-01 6.23049736e-01 -1.32287771e-01 1.09138504e-01
2.54534811e-01 8.47987652e-01 1.26893806e+00 -7.91856110e-01
-6.32453263e-01 -4.24694449e-01 1.36623094e-02 -1.14432526e+00
5.41881979e-01 -1.53909540e+00 2.10306510e-01 -1.12768602e+00
5.16213775e-01 -5.23634613e-01 -5.03353253e-02 7.53171742e-01
-2.56243765e-01 -2.50667781e-01 -8.21178705e-02 7.20794916e-01
-4.71561641e-01 4.46223050e-01 9.44121838e-01 -3.91519547e-01
-2.42541701e-01 -2.50091791e-01 -5.74836075e-01 7.94642448e-01
8.50859582e-01 -8.98977160e-01 -2.15245262e-01 -1.33167028e+00
3.83693874e-01 4.18284386e-02 -2.51109064e-01 -8.27327669e-01
2.09151149e-01 -3.68339986e-01 -2.87883908e-01 -1.40678957e-01
-6.22833371e-01 -6.83368027e-01 -2.85443123e-02 3.35731238e-01
-2.43773207e-01 5.28501391e-01 1.47678226e-01 9.69208106e-02
-5.34792364e-01 -2.99329013e-01 6.49734914e-01 -2.45960027e-01
-8.35005105e-01 1.43281162e-01 -2.49861985e-01 2.81260699e-01
6.73724651e-01 1.93334445e-01 -4.43268418e-01 4.00393605e-01
-2.15170309e-02 1.99569404e-01 1.03582060e+00 9.44001019e-01
1.59649402e-01 -1.54527748e+00 -6.46670580e-01 3.60994697e-01
6.19341314e-01 -5.86247966e-02 -6.57785609e-02 9.32827950e-01
-8.71007681e-01 3.14800829e-01 -4.37076509e-01 -7.15834379e-01
-1.05883670e+00 5.38536549e-01 1.87249839e-01 -2.89621770e-01
-4.28803593e-01 4.46446002e-01 1.06507160e-01 -1.41778123e+00
-1.34196490e-01 -4.71630007e-01 3.71417850e-01 -6.23377085e-01
2.64086008e-01 -1.40340105e-01 1.35954469e-01 -5.93437314e-01
-3.83504212e-01 6.28395617e-01 -3.58565032e-01 2.39031047e-01
1.25974071e+00 3.11556272e-02 -8.30145657e-01 4.50279593e-01
1.43213594e+00 2.10096806e-01 -6.77173197e-01 -7.19439566e-01
5.16695380e-01 -6.08379543e-01 -2.58369684e-01 -6.23325884e-01
-1.17988312e+00 1.19938004e+00 3.66629243e-01 -4.05760407e-01
8.64462316e-01 -2.08345801e-01 1.09118664e+00 7.36721933e-01
8.60853016e-01 -1.12572515e+00 -1.49505630e-01 6.47621691e-01
4.93374199e-01 -1.48964477e+00 -3.91031593e-01 -2.50616133e-01
-3.35521221e-01 1.11395526e+00 1.28222263e+00 2.16996208e-01
3.23398352e-01 3.56492311e-01 4.85705823e-01 1.97221339e-01
-5.05018771e-01 2.29556531e-01 -1.61262915e-01 3.59442621e-01
1.04372573e+00 1.55393016e-02 -4.70238477e-01 7.24269032e-01
2.58798804e-03 3.23040903e-01 5.56994617e-01 1.42752266e+00
2.48023793e-02 -1.83329070e+00 -1.47689521e-01 2.44198982e-02
-5.14483929e-01 -3.26489449e-01 -3.65524411e-01 6.12781882e-01
4.75435294e-02 5.89115083e-01 -3.36321235e-01 -4.62386400e-01
2.42826879e-01 1.28408253e-01 4.07503992e-01 -1.02356589e+00
-7.67831206e-01 -1.78143054e-01 -1.73767671e-01 -3.97117198e-01
-3.23784620e-01 -4.66372520e-01 -1.27660668e+00 -4.29949164e-01
-2.60030806e-01 2.80322641e-01 6.61862910e-01 8.49709809e-01
5.56593359e-01 3.64255637e-01 5.93442321e-01 -2.67271221e-01
-5.60915589e-01 -7.86846936e-01 4.68705036e-02 2.77136356e-01
-2.77433712e-02 -9.05345678e-02 4.77531366e-02 4.27527815e-01] | [7.64813756942749, 7.947559356689453] |
fe3ebe05-9e7c-4597-b879-c977dfb9049c | music-transcription-by-deep-learning-with | 1712.03228 | null | http://arxiv.org/abs/1712.03228v1 | http://arxiv.org/pdf/1712.03228v1.pdf | Music Transcription by Deep Learning with Data and "Artificial Semantic" Augmentation | In this progress paper the previous results of the single note recognition by
deep learning are presented. The several ways for data augmentation and
"artificial semantic" augmentation are proposed to enhance efficiency of deep
learning approaches for monophonic and polyphonic note recognition by increase
of dimensions of training data, their lossless and lossy transformations. | ['Sergii Stirenko', 'Vadym Ovcharenko', 'Yuri Gordienko', 'Vladyslav Sarnatskyi', 'Mariia Tkachenko', 'Anis Rojbi'] | 2017-12-08 | null | null | null | null | ['music-transcription'] | ['music'] | [ 3.23174208e-01 1.17291607e-01 1.65034592e-01 -1.79296836e-01
-8.59841287e-01 -4.69273806e-01 5.04628301e-01 -1.45842597e-01
-5.98599553e-01 1.00023437e+00 7.26513445e-01 3.79853368e-01
8.17488730e-02 -3.76579523e-01 -3.30739260e-01 -6.42165005e-01
8.27194080e-02 1.14026248e-01 -6.04383126e-02 4.89210598e-02
4.94079828e-01 8.88584733e-01 -1.56178331e+00 6.13341928e-01
3.87480902e-03 1.09897530e+00 -2.15856746e-01 1.70248806e+00
7.11472705e-02 8.82942975e-01 -1.06878817e+00 -2.39627212e-01
7.47632027e-01 -4.94172215e-01 -8.43390822e-01 -7.61750340e-02
6.54039204e-01 -2.97237068e-01 -9.01196361e-01 8.70632946e-01
1.16972768e+00 6.11619949e-01 9.57694232e-01 -1.06685841e+00
-8.73419404e-01 4.65275347e-01 8.49224180e-02 4.05370712e-01
2.64814615e-01 1.08872261e-02 8.47060442e-01 -8.30099523e-01
2.82569408e-01 9.18743372e-01 1.23553836e+00 7.15154648e-01
-3.91298056e-01 -3.37661296e-01 -5.79339087e-01 1.46314546e-01
-9.97919679e-01 -4.09221947e-01 7.88775086e-01 -8.05075318e-02
1.52051342e+00 6.70081377e-01 8.19850981e-01 7.47812271e-01
-1.62129983e-01 1.18940878e+00 7.29602933e-01 -9.56120849e-01
-1.30033299e-01 2.17320770e-01 -1.29491184e-02 7.72110701e-01
-1.56039298e-01 4.59070176e-01 -8.72767866e-01 1.57439947e-01
8.93263996e-01 1.00639509e-03 3.56395058e-02 6.08876586e-01
-9.87139106e-01 4.92296934e-01 2.20843643e-01 6.92226112e-01
-3.20168495e-01 2.94937760e-01 8.17847431e-01 5.32449961e-01
3.97601217e-01 6.79938674e-01 -5.87985694e-01 -7.11089849e-01
-1.16206086e+00 2.25524470e-01 7.58232892e-01 1.18522286e+00
1.72371194e-01 7.37259686e-01 -4.10537153e-01 1.06988776e+00
-8.48541707e-02 3.36762041e-01 1.33788764e+00 -8.35189223e-01
2.63046086e-01 3.78032058e-01 -2.29576752e-01 -6.69976294e-01
-4.97606993e-01 -4.02060986e-01 -9.13340569e-01 4.50692803e-01
-8.83330032e-03 -3.62628132e-01 -1.22029686e+00 1.10115516e+00
-5.11043847e-01 1.72431052e-01 4.96565402e-01 5.70154548e-01
8.89177501e-01 6.86279833e-01 -1.31839082e-01 -5.35666756e-02
1.04652405e+00 -1.34256208e+00 -1.33138168e+00 3.60782713e-01
5.43551207e-01 -1.44331288e+00 1.11711144e+00 5.70810139e-01
-1.46637440e+00 -1.08284998e+00 -1.21121645e+00 -4.69954818e-01
-7.35913157e-01 6.81625009e-01 6.67690456e-01 5.43095052e-01
-9.59645987e-01 9.08756077e-01 -3.39159936e-01 -5.87391816e-02
5.57724357e-01 6.40299380e-01 -4.60158974e-01 3.15442055e-01
-9.07196522e-01 8.75735343e-01 6.77468956e-01 9.15674195e-02
-7.10821211e-01 -9.19881225e-01 -5.10438263e-01 2.21058294e-01
-4.31400567e-01 -3.89633000e-01 1.27623212e+00 -6.89818084e-01
-1.66657317e+00 8.73702347e-01 3.79138201e-01 -7.30103672e-01
4.00708824e-01 -5.56865156e-01 -5.68348825e-01 -9.69513878e-03
-8.78233135e-01 8.17464471e-01 8.05596054e-01 -6.63529396e-01
-7.21648812e-01 -2.69205242e-01 -8.82555246e-02 4.38000649e-01
-5.03999293e-01 -1.05881661e-01 3.42687875e-01 -1.45621812e+00
-1.24054067e-01 -9.61709321e-01 1.65080056e-01 2.16657203e-02
-3.63221586e-01 -1.36497766e-01 1.05923188e+00 -1.00873661e+00
1.00057185e+00 -2.23430300e+00 -7.86896572e-02 -3.26753825e-01
-3.11425298e-01 8.40151191e-01 -7.53246024e-02 4.96500969e-01
6.68468997e-02 7.69053819e-03 -4.86926794e-01 -5.64153910e-01
3.08785159e-02 2.03410417e-01 -1.19370341e-01 -8.59006792e-02
-4.70473766e-02 9.23959553e-01 -3.21872324e-01 -3.77731949e-01
5.22900581e-01 6.22870624e-01 -3.86140645e-01 6.49683595e-01
2.81303197e-01 -1.08859941e-01 3.21821064e-01 8.39944661e-01
5.38549483e-01 5.49852312e-01 -5.97344875e-01 -4.92602050e-01
-4.26933728e-02 1.56125233e-01 -1.09387505e+00 2.09453464e+00
-5.98102748e-01 1.16669977e+00 -4.42527354e-01 -7.15494990e-01
1.42573202e+00 8.29443038e-01 4.15841877e-01 -1.10569388e-01
1.41666949e-01 2.11797774e-01 -2.99632937e-01 -7.83736825e-01
6.53442502e-01 -7.81061530e-01 1.13597400e-01 5.33320725e-01
5.70792496e-01 -2.93357313e-01 -3.52592438e-01 -4.03269708e-01
9.37854469e-01 -1.75367877e-01 2.69587725e-01 1.50858954e-01
5.74368238e-01 -1.33360296e-01 4.21802133e-01 4.76264864e-01
-3.52578133e-01 1.06411743e+00 -3.67070258e-01 -7.11078227e-01
-1.34256864e+00 -8.02119911e-01 2.12171853e-01 9.88889098e-01
-7.12208748e-01 8.88699740e-02 -6.29791677e-01 -7.44080603e-01
-4.20600772e-02 4.59552258e-01 -3.81026179e-01 -5.86782873e-01
-5.45357764e-01 -8.30385685e-01 1.51570261e+00 9.12834287e-01
9.41438675e-01 -1.76741350e+00 -4.16143566e-01 3.96051079e-01
1.29876420e-01 -9.70603585e-01 -6.48871481e-01 2.83623666e-01
-1.29240286e+00 -9.13518786e-01 -1.16652346e+00 -1.21962011e+00
2.84021229e-01 -2.92650640e-01 7.26230443e-01 4.85639833e-02
-9.33687150e-01 4.60278422e-01 -5.54468989e-01 -9.66072738e-01
-3.87893945e-01 -9.94871780e-02 1.60605788e-01 -1.59455404e-01
5.68494260e-01 -7.96900392e-01 -6.65214777e-01 -5.19473612e-01
-1.09444618e+00 -5.11679590e-01 1.00147104e+00 7.13009536e-01
1.48085594e-01 1.19916238e-02 3.06689948e-01 -7.12768018e-01
1.04671943e+00 3.77644658e-01 -8.16920586e-03 2.07515344e-01
-6.26356006e-01 2.91334808e-01 5.88275611e-01 -3.72101486e-01
-1.08133185e+00 4.04104710e-01 -3.90643626e-01 -5.17073810e-01
-5.32655299e-01 -2.46759206e-02 1.78725913e-01 -4.19232219e-01
5.42955160e-01 7.20606625e-01 -2.01127529e-01 -9.31943476e-01
2.17700005e-01 8.47703278e-01 8.84832799e-01 2.75496580e-02
5.88511646e-01 3.56547087e-01 7.98565224e-02 -1.00281489e+00
-3.88541579e-01 -4.86916333e-01 -1.21033633e+00 -3.63420129e-01
6.27550960e-01 -5.06784439e-01 -7.67444551e-01 9.82963204e-01
-1.20929611e+00 8.27164575e-03 -1.08639503e+00 1.05305600e+00
-8.07091892e-01 3.21083754e-01 -8.91851485e-01 -1.04698837e+00
-8.22864115e-01 -4.83871907e-01 9.49076176e-01 3.83381039e-01
-2.47723877e-01 -1.18899012e+00 5.09046972e-01 1.05845951e-01
2.44458377e-01 6.74061179e-02 8.10160398e-01 -7.80518770e-01
-1.90905303e-01 -7.75233388e-01 1.47933856e-01 1.26173544e+00
3.54123265e-01 -6.67761192e-02 -1.55578554e+00 -4.57944125e-02
2.85370052e-02 -5.15343368e-01 8.09840441e-01 3.79045993e-01
9.63436067e-01 -7.57171869e-01 4.21662390e-01 7.69590795e-01
1.57341766e+00 7.11360395e-01 7.48323858e-01 1.14280447e-01
6.44756615e-01 8.49753246e-02 3.28875422e-01 6.90003514e-01
-2.69611478e-01 3.52119267e-01 9.97575349e-04 -1.91660091e-01
-7.54970312e-01 -2.63696998e-01 1.03661574e-01 1.48247254e+00
-4.93766457e-01 -1.30684465e-01 -4.74645168e-01 5.20146608e-01
-1.29679525e+00 -9.78338420e-01 2.11808041e-01 1.87528038e+00
5.26988029e-01 -1.16348127e-02 3.81604880e-01 1.01600027e+00
6.53177977e-01 2.38337219e-01 -4.12484825e-01 -9.23587918e-01
-3.58941406e-01 4.25118685e-01 4.15828288e-01 7.07473099e-01
-9.73075509e-01 9.47661281e-01 8.19646931e+00 5.55092752e-01
-9.01827276e-01 4.45935614e-02 1.63710624e-01 -2.88917363e-01
1.94934756e-01 -5.40332317e-01 -6.84771061e-01 1.44117892e-01
9.57294405e-01 1.15231005e-02 4.09141541e-01 6.58929765e-01
-1.06095001e-01 3.95650029e-01 -8.75970423e-01 1.35243571e+00
3.49159300e-01 -1.49474776e+00 5.08015633e-01 -3.10285211e-01
7.34803617e-01 -3.04822117e-01 3.73835750e-02 2.76041687e-01
-1.87085599e-01 -7.24821806e-01 2.23099381e-01 9.30354774e-01
7.00590849e-01 -9.22366440e-01 8.14210892e-01 -2.05332711e-01
-1.22321498e+00 -2.43829474e-01 -6.63330734e-01 -3.03325981e-01
-7.21450076e-02 1.15336090e-01 -9.02088702e-01 2.90813416e-01
1.99744329e-01 6.46711648e-01 -3.94139588e-01 1.29303300e+00
1.57734036e-01 5.60181022e-01 -2.12660551e-01 -2.30432346e-01
4.01528120e-01 3.99659783e-01 6.01559937e-01 1.73989713e+00
4.74836528e-01 2.66431391e-01 -3.92540604e-01 3.19334745e-01
-2.20923260e-01 5.88650554e-02 -7.73185074e-01 -2.64573961e-01
3.01434010e-01 9.13467646e-01 -3.94077539e-01 -4.57629353e-01
5.25969155e-02 1.25399148e+00 -2.12510183e-01 5.94333187e-02
-3.82477611e-01 -1.03220356e+00 4.11015987e-01 -3.37634921e-01
1.65556923e-01 -2.44988576e-01 -5.60238063e-01 -7.85200775e-01
1.62626095e-02 -5.46233714e-01 7.40861893e-01 -7.82391906e-01
-1.11597538e+00 8.11694384e-01 -4.55501139e-01 -1.48320389e+00
-3.02894145e-01 -8.30348909e-01 -8.53221595e-01 6.15665674e-01
-1.16330647e+00 -1.00983346e+00 -2.67010987e-01 9.54734802e-01
1.04713571e+00 -1.14854705e+00 1.25396740e+00 4.62105304e-01
1.61150903e-01 1.02761662e+00 1.82544127e-01 2.37795189e-01
6.05195165e-01 -1.44390237e+00 3.34814012e-01 5.04674017e-01
8.43809783e-01 -1.74026147e-01 5.48162103e-01 -1.44863009e-01
-9.05374408e-01 -7.60103941e-01 1.09652960e+00 -3.79773825e-02
2.28608157e-02 3.21416967e-02 -6.98700309e-01 5.10771453e-01
1.92786261e-01 -1.47802934e-01 1.10922647e+00 -1.56607538e-01
-8.06600302e-02 -3.21871817e-01 -1.35536301e+00 3.25569600e-01
7.66408443e-01 -8.50898266e-01 -1.11729705e+00 6.26240492e-01
6.40469253e-01 -1.80276148e-02 -1.08130908e+00 4.24420774e-01
7.94948459e-01 -8.20661068e-01 9.91708338e-01 -1.02765942e+00
1.59993619e-01 2.93779731e-01 -2.35173836e-01 -1.17928529e+00
-3.14989924e-01 -9.42263722e-01 -5.11199474e-01 9.85734165e-01
3.09665293e-01 1.30347729e-01 1.31020904e+00 3.19961518e-01
-4.57353175e-01 -4.66088921e-01 -8.99544120e-01 -8.86288285e-01
-2.31881440e-01 -3.36360246e-01 3.02306563e-01 6.62275374e-01
5.56182861e-02 1.01825058e-01 -8.73776913e-01 -2.17059091e-01
3.73272300e-01 -4.11924273e-01 3.34172338e-01 -1.19007802e+00
-3.74616653e-01 -2.89002717e-01 -1.03550398e+00 -6.56790972e-01
-1.54656917e-01 -6.51147962e-01 1.56769112e-01 -1.26985621e+00
-2.11758479e-01 -1.44410133e-03 -9.28241313e-01 5.05380988e-01
1.02836296e-01 6.38133943e-01 6.42106712e-01 3.93359572e-01
-1.41956240e-01 6.23867452e-01 1.10589302e+00 -1.89569846e-01
-2.77722895e-01 4.80558366e-01 -1.06604770e-03 6.15812302e-01
1.08702338e+00 -5.65537095e-01 -3.46010357e-01 -3.42404693e-01
-4.91061777e-01 2.58135945e-02 2.53935009e-01 -1.64747691e+00
2.74048984e-01 5.49992383e-01 4.19585437e-01 -6.90739572e-01
8.70840490e-01 -8.15282226e-01 -4.11848277e-01 8.28749657e-01
-1.17891407e+00 3.85868400e-01 5.86134374e-01 4.76313502e-01
-6.21783674e-01 -7.80509710e-01 8.30009818e-01 -4.49952304e-01
-6.26714230e-01 3.12772274e-01 -6.58973753e-01 -1.17156073e-01
6.66913271e-01 -3.84271979e-01 2.12393671e-01 -6.17376864e-01
-1.34447765e+00 -9.85480130e-01 -4.66788441e-01 4.45846319e-01
9.26730692e-01 -1.59654212e+00 -6.54026568e-01 2.92347282e-01
-3.28296870e-01 -6.12005293e-01 2.72894233e-01 3.23074996e-01
-8.11539888e-01 7.98501313e-01 -1.20458686e+00 2.66292930e-01
-1.89569902e+00 3.81280363e-01 4.22212571e-01 -2.06765607e-01
-7.50483930e-01 1.27306712e+00 -5.49652755e-01 -3.42319578e-01
8.43542278e-01 -4.22475487e-01 -2.24760652e-01 -1.46351546e-01
5.69497406e-01 9.33631837e-01 7.58397430e-02 -2.25750759e-01
1.40678525e-01 5.12809098e-01 -1.32712051e-01 -4.71927017e-01
1.53019977e+00 3.25973988e-01 2.07497180e-01 8.80942285e-01
1.50512373e+00 -3.10129523e-01 -1.17145085e+00 1.75375734e-02
-1.30784586e-02 -4.95349079e-01 -1.87267046e-02 -1.08486366e+00
-1.05320764e+00 1.38226879e+00 9.96285975e-01 2.74452269e-01
1.46504986e+00 -6.94314420e-01 1.01739609e+00 9.60644305e-01
-4.09792542e-01 -1.48661304e+00 3.01596344e-01 3.33501160e-01
1.19203806e+00 -1.04470336e+00 -1.50641173e-01 8.72777924e-02
-6.08133912e-01 1.70707524e+00 3.80023658e-01 -6.45688176e-01
9.95425344e-01 6.74831033e-01 3.87488514e-01 8.82157013e-02
-5.98373652e-01 7.77465701e-02 4.22660142e-01 8.83573890e-01
5.54208577e-01 -1.54111922e-01 -3.32270771e-01 4.28096920e-01
-2.51552701e-01 2.69460887e-01 4.82411623e-01 1.01348877e+00
-4.29575533e-01 -1.02336049e+00 -3.67841095e-01 4.17026430e-01
-6.25624061e-01 -4.22553211e-01 -9.31364596e-01 1.07178891e+00
1.49144650e-01 4.60051537e-01 -9.60454121e-02 -7.67889023e-01
5.44621527e-01 3.73648643e-01 6.47291422e-01 -4.93930191e-01
-1.09183931e+00 -2.27157727e-01 -5.96634597e-02 1.18806414e-01
-4.42192584e-01 -3.90299946e-01 -1.03304255e+00 -4.83217612e-02
-7.46986270e-02 2.07794711e-01 9.66516733e-01 8.33296597e-01
1.40769452e-01 8.16502213e-01 4.48255658e-01 -8.06759655e-01
-7.08467603e-01 -1.31833088e+00 -9.72995222e-01 5.58996201e-01
3.73913556e-01 -1.38928413e-01 -3.99837792e-01 6.06059372e-01] | [15.684369087219238, 5.30488920211792] |
1c6bacb0-1984-4369-9a24-a9a6fb6fc7ab | salypath-a-deep-based-architecture-for-visual | 2107.00559 | null | https://arxiv.org/abs/2107.00559v1 | https://arxiv.org/pdf/2107.00559v1.pdf | SALYPATH: A Deep-Based Architecture for visual attention prediction | Human vision is naturally more attracted by some regions within their field of view than others. This intrinsic selectivity mechanism, so-called visual attention, is influenced by both high- and low-level factors; such as the global environment (illumination, background texture, etc.), stimulus characteristics (color, intensity, orientation, etc.), and some prior visual information. Visual attention is useful for many computer vision applications such as image compression, recognition, and captioning. In this paper, we propose an end-to-end deep-based method, so-called SALYPATH (SALiencY and scanPATH), that efficiently predicts the scanpath of an image through features of a saliency model. The idea is predict the scanpath by exploiting the capacity of a deep-based model to predict the saliency. The proposed method was evaluated through 2 well-known datasets. The results obtained showed the relevance of the proposed framework comparing to state-of-the-art models. | ['Rachid Harba', 'Aladine Chetouani', 'Marouane Tliba', 'Mohamed Amine Kerkouri'] | 2021-06-29 | null | null | null | null | ['scanpath-prediction'] | ['computer-vision'] | [ 6.01341128e-01 -8.55512992e-02 -2.66193300e-01 -5.56952953e-01
-1.54950231e-01 -9.65910032e-02 6.04708493e-01 2.23988578e-01
-3.17912430e-01 4.95666355e-01 4.08192426e-01 2.32829005e-01
-2.11894624e-02 -5.66004694e-01 -8.95827770e-01 -5.45652807e-01
1.16939798e-01 7.82048404e-02 7.27148712e-01 -6.93797767e-02
9.60730016e-01 4.45477575e-01 -2.10248518e+00 5.05516052e-01
6.15461171e-01 1.01236606e+00 8.18709850e-01 5.52618980e-01
4.78714146e-02 8.72439325e-01 1.68434996e-02 1.49131000e-01
-3.48786265e-03 -4.73648697e-01 -6.52686238e-01 1.24187574e-01
4.82385814e-01 3.95880900e-02 -1.51196346e-01 1.21210718e+00
3.95822376e-01 1.04309037e-01 5.46003163e-01 -7.98107564e-01
-6.75089061e-01 2.71248370e-01 -7.27400243e-01 5.66797197e-01
8.83788392e-02 3.14161986e-01 1.06612647e+00 -9.36634183e-01
7.65613377e-01 1.15081918e+00 1.14011921e-01 3.54782194e-01
-9.35586512e-01 -2.07104325e-01 1.32428765e-01 8.23275387e-01
-9.96533632e-01 -2.56094396e-01 1.01243341e+00 -4.11908537e-01
5.58919370e-01 1.73566163e-01 6.04748547e-01 7.87229419e-01
5.27844727e-01 9.98982608e-01 1.45444310e+00 -3.82293612e-01
3.21167737e-01 4.05752152e-01 2.06839549e-03 5.48753917e-01
-9.48644727e-02 2.47558832e-01 -7.29177773e-01 1.76991314e-01
7.33613431e-01 4.58043180e-02 -3.77713799e-01 -6.92541540e-01
-1.14149797e+00 5.67012131e-01 1.04069078e+00 3.22747380e-01
-7.62960553e-01 -1.45228822e-02 1.22794837e-01 -2.39140362e-01
1.88758314e-01 3.61691713e-01 -4.39012885e-01 2.12377265e-01
-1.00827301e+00 1.29270196e-01 2.47878879e-01 6.64807379e-01
9.24480855e-01 -4.25897576e-02 -5.50197423e-01 5.50385118e-01
4.85056162e-01 4.45742130e-01 7.72998154e-01 -5.03530622e-01
2.55169392e-01 4.55012619e-01 2.31833026e-01 -1.23250675e+00
-5.71000576e-01 -7.64708221e-01 -5.40455461e-01 2.28685707e-01
1.08877353e-01 2.54433155e-01 -9.32598531e-01 1.70656729e+00
1.81020334e-01 2.60872960e-01 -1.58882260e-01 1.46266901e+00
7.81132400e-01 6.39865339e-01 1.10263243e-01 4.94720368e-03
1.30021524e+00 -1.19207215e+00 -4.26157415e-01 -6.97815120e-01
5.12490608e-03 -8.54012370e-01 1.09616923e+00 1.69593811e-01
-1.18917775e+00 -9.12918866e-01 -7.96078920e-01 -6.77216128e-02
-2.60348469e-01 3.01171184e-01 3.79929125e-01 1.34638533e-01
-1.06914043e+00 3.59272003e-01 -4.84249085e-01 -6.33968532e-01
4.66899991e-01 2.66312063e-01 -6.28222525e-02 -4.80044968e-02
-9.90257025e-01 9.44328666e-01 4.49803144e-01 4.56033200e-02
-1.12253487e+00 -2.79181212e-01 -5.88513434e-01 4.22018915e-01
1.21018939e-01 -6.76730990e-01 9.53227818e-01 -1.51311481e+00
-1.20602739e+00 8.37609649e-01 -4.42583025e-01 -5.79568028e-01
2.17102051e-01 -3.12529504e-01 -6.71692193e-02 2.67002493e-01
-1.05622858e-02 8.69956434e-01 1.12221324e+00 -1.24269783e+00
-7.19858706e-01 -6.60853744e-01 -1.08530834e-01 4.90516901e-01
-1.19126543e-01 -2.15912368e-02 -4.59899664e-01 -3.92099231e-01
1.76114943e-02 -8.89274955e-01 -3.07268232e-01 -5.92087731e-02
-3.85484546e-01 3.51880863e-02 8.08605850e-01 -6.85654700e-01
7.70579696e-01 -2.09884858e+00 1.94360122e-01 1.72725946e-01
-2.83076111e-02 4.61983383e-01 -2.80492336e-01 2.20807284e-01
-1.05696358e-01 -2.93643206e-01 -1.04283266e-01 2.48372909e-02
-4.99016643e-01 -1.71849221e-01 -2.19613552e-01 5.72500587e-01
3.09594750e-01 7.42769480e-01 -8.91802251e-01 -3.96597892e-01
5.95802367e-01 5.60071826e-01 -3.64806503e-01 2.98067689e-01
-4.20089632e-01 5.94719291e-01 -5.20956039e-01 4.49619263e-01
6.24529839e-01 -2.24355698e-01 -2.01011896e-01 -3.53461325e-01
-4.63480234e-01 1.14095390e-01 -7.98186004e-01 1.69608319e+00
-3.96507144e-01 8.82282436e-01 -2.86839426e-01 -8.55809748e-01
1.14424801e+00 -1.18209109e-01 1.40469506e-01 -9.14735258e-01
4.27775800e-01 1.58284590e-01 2.19649568e-01 -5.88710308e-01
5.21328688e-01 4.36382115e-01 4.63736951e-01 1.54200569e-01
2.10443605e-02 1.06498115e-01 -8.58142525e-02 -8.32178537e-03
4.29910302e-01 3.10106180e-03 3.51737738e-01 -4.49701101e-01
8.62404406e-01 -2.05761373e-01 2.77636319e-01 6.69925749e-01
-4.16990340e-01 6.88045204e-01 3.68829966e-01 -4.31539506e-01
-1.18488777e+00 -6.99993849e-01 -4.40886654e-02 1.20197999e+00
5.60267389e-01 7.04101473e-02 -9.16714907e-01 -2.70099521e-01
-3.01499963e-01 6.99765444e-01 -8.61347258e-01 -3.21802616e-01
-3.14169616e-01 -4.23255533e-01 -1.26387924e-01 4.52150404e-01
6.61118686e-01 -1.46815300e+00 -1.30949926e+00 9.17973444e-02
-1.05784694e-02 -1.19261122e+00 -3.64119530e-01 -4.90657203e-02
-8.66498053e-01 -9.45552647e-01 -6.96214020e-01 -8.95553589e-01
6.12147093e-01 5.30735672e-01 1.01692963e+00 -1.18277827e-03
-3.90942127e-01 1.50650144e-01 -2.76356757e-01 -6.31456077e-01
7.15519935e-02 5.94105609e-02 -4.19978350e-01 6.01778030e-01
2.35219494e-01 -3.24642420e-01 -9.84127104e-01 2.62714386e-01
-9.01769757e-01 3.88628721e-01 9.85425830e-01 7.87636518e-01
6.82801247e-01 -1.01314083e-01 4.06447679e-01 -8.13362837e-01
2.19940603e-01 -3.89960319e-01 -6.58821464e-01 1.73375279e-01
-2.13175684e-01 1.59159839e-01 3.92277062e-01 -9.36925858e-02
-1.02481961e+00 3.73908758e-01 1.78322807e-01 -2.38485858e-01
-3.57351214e-01 3.61960977e-01 -2.43350610e-01 -2.12598398e-01
5.01792490e-01 5.60070992e-01 -1.98944479e-01 -3.83049160e-01
1.73626542e-01 5.50029457e-01 6.30872607e-01 5.50691225e-02
3.90079618e-01 5.72882652e-01 2.55073220e-01 -8.46148431e-01
-9.47567999e-01 -5.19079149e-01 -7.36707151e-01 -3.34660321e-01
8.58081937e-01 -7.50974715e-01 -4.11552787e-01 6.17584944e-01
-1.27751994e+00 -1.16932496e-01 1.06192790e-01 5.18429697e-01
-6.13791287e-01 5.40630929e-02 -4.71974127e-02 -8.70272994e-01
-4.16703969e-01 -1.17838609e+00 1.05945480e+00 7.23340988e-01
2.30160639e-01 -8.82686019e-01 -2.34086469e-01 3.61704975e-01
6.47964716e-01 7.46335834e-02 8.03663313e-01 -4.94550616e-01
-9.61678386e-01 1.28018394e-01 -6.89197898e-01 2.56835848e-01
-1.50256768e-01 -3.06045711e-02 -1.09880424e+00 -1.36254011e-02
-9.19776484e-02 -1.47939667e-01 9.91062582e-01 7.56377339e-01
1.38037109e+00 -4.82690632e-02 -3.49335402e-01 5.36436737e-01
1.71844077e+00 -1.41300429e-02 7.30674386e-01 3.81221950e-01
6.89537108e-01 7.57906139e-01 8.29331398e-01 5.01464248e-01
2.05922484e-01 8.96240473e-01 9.60199475e-01 -3.90393794e-01
-2.01630935e-01 -1.42459810e-01 6.27271608e-02 1.44625410e-01
-1.50433257e-01 -2.04125687e-01 -8.15253139e-01 8.09173822e-01
-1.70811999e+00 -8.65710318e-01 -2.42545933e-01 2.16661096e+00
5.61577201e-01 1.81845516e-01 -9.65195745e-02 1.86906441e-03
8.74934196e-01 1.96247563e-01 -8.17452848e-01 -6.68340027e-01
-2.67099142e-01 5.11788651e-02 6.71645939e-01 3.17023396e-01
-1.01885736e+00 1.01539946e+00 5.32447529e+00 5.94937503e-01
-1.59032643e+00 -5.72792292e-02 7.73005605e-01 1.85356259e-01
-2.58625001e-01 -7.18478635e-02 -6.85198903e-01 5.63681126e-01
6.49863362e-01 -4.07730937e-02 4.77811396e-01 8.56753469e-01
3.35691035e-01 -4.12308663e-01 -8.79787266e-01 8.56283605e-01
5.24313569e-01 -1.06355643e+00 -5.31197013e-03 -1.11795925e-01
7.91172504e-01 3.02020818e-01 5.04796028e-01 -1.52577683e-01
-3.76508236e-01 -8.36554646e-01 8.32873583e-01 5.53422868e-01
4.57026094e-01 -6.37733757e-01 8.86232018e-01 3.23889881e-01
-8.46826673e-01 -3.20748389e-01 -3.95785719e-01 -1.00840507e-02
-5.76328561e-02 6.06223524e-01 -9.18633580e-01 1.43098384e-01
6.45938694e-01 8.97670627e-01 -9.30267692e-01 1.37720621e+00
-3.33832860e-01 5.35826027e-01 2.30391193e-02 -2.20435292e-01
4.59294140e-01 2.06290781e-01 6.36412680e-01 1.07705855e+00
2.11465955e-01 -2.82907505e-02 -1.22326329e-01 9.54055488e-01
1.33959517e-01 2.44653761e-01 -4.49539870e-01 2.14676440e-01
1.67952687e-01 1.11364639e+00 -6.60677433e-01 -2.06917316e-01
-2.97559857e-01 8.28674614e-01 4.66100127e-02 3.41714799e-01
-7.15008140e-01 -2.15830386e-01 2.76250780e-01 3.20580542e-01
6.66422069e-01 8.14209282e-02 -4.51059192e-01 -8.80352736e-01
-1.06857307e-01 -5.96228182e-01 2.46833172e-02 -1.18759429e+00
-7.63888359e-01 7.70939827e-01 -2.22228542e-01 -1.12797260e+00
-2.16915719e-02 -5.83947361e-01 -6.07049048e-01 1.01022983e+00
-1.91688836e+00 -1.16491222e+00 -6.14971459e-01 7.89794862e-01
8.45048547e-01 -2.65570760e-01 3.92951488e-01 -5.90274557e-02
-1.52055308e-01 2.17255577e-01 -1.84437726e-02 -2.23873973e-01
5.03791034e-01 -1.00103199e+00 2.24760056e-01 8.98141086e-01
2.73781389e-01 2.17779696e-01 1.00508010e+00 -3.30603004e-01
-1.11627364e+00 -1.02884102e+00 1.02553260e+00 3.55380848e-02
3.55809540e-01 -1.27653807e-01 -8.16354632e-01 2.21325845e-01
2.92548120e-01 1.47477657e-01 1.36192769e-01 -1.82313904e-01
-1.07624985e-01 -3.23555768e-01 -9.84789431e-01 4.95794982e-01
6.56195283e-01 -3.59423965e-01 -3.84887338e-01 1.09433495e-01
4.26653534e-01 -3.54305148e-01 2.10973863e-02 3.08674365e-01
5.58162868e-01 -1.47072780e+00 9.21458840e-01 -3.54107022e-01
6.11875057e-01 -2.53428787e-01 -2.69042235e-02 -1.28297389e+00
-6.16694152e-01 -3.17818262e-02 1.81149334e-01 8.22779775e-01
2.20554411e-01 -3.18045765e-01 6.76175654e-01 2.92803317e-01
-8.02301764e-02 -7.84883916e-01 -5.75269401e-01 -2.16589838e-01
-6.84126019e-01 -1.38841504e-02 4.98979151e-01 4.15581226e-01
-4.60387290e-01 4.02185887e-01 -4.99696285e-01 2.67129004e-01
5.43738306e-01 3.22428346e-01 5.82422733e-01 -1.27262044e+00
-1.28339455e-01 -3.63683343e-01 -8.67746830e-01 -8.72845292e-01
-2.60795243e-02 -4.31528568e-01 1.80081919e-01 -1.40343869e+00
3.41149390e-01 -1.64093018e-01 -8.05481911e-01 2.26970986e-01
-2.55915374e-01 3.42882961e-01 4.36260581e-01 1.83826566e-01
-6.24582589e-01 5.33796906e-01 1.45258605e+00 -3.05388030e-02
-3.04619253e-01 2.49015819e-02 -6.91794991e-01 6.37693524e-01
9.61752057e-01 -2.25607559e-01 -5.15423656e-01 -6.45776451e-01
-4.98880148e-02 4.91808206e-02 6.49968684e-01 -1.34056532e+00
2.67284423e-01 -2.73244113e-01 3.62001508e-01 -6.73533142e-01
3.49709570e-01 -7.32703686e-01 -3.76381487e-01 4.46984351e-01
-7.16282248e-01 -1.57796249e-01 6.28302917e-02 5.22621095e-01
-4.76040244e-01 -1.59757003e-01 1.07744431e+00 -7.89891705e-02
-1.18089235e+00 1.75930530e-01 -5.86448684e-02 -1.52191877e-01
9.70301747e-01 -1.34441584e-01 -2.85670161e-01 -4.02319133e-01
-4.72115368e-01 -1.68112472e-01 3.23859543e-01 5.61194539e-01
8.16267669e-01 -9.83870685e-01 -7.74434328e-01 4.92897063e-01
2.09686920e-01 -2.95703888e-01 4.11363095e-01 8.76847565e-01
-3.25062156e-01 5.47906339e-01 -8.28994274e-01 -7.91665614e-01
-1.28082526e+00 7.82204688e-01 7.33740181e-02 6.22447543e-02
-2.40765229e-01 8.74316752e-01 5.91575325e-01 3.02192718e-01
1.50917009e-01 -4.04621243e-01 -6.39383554e-01 -4.18440372e-01
6.62831366e-01 2.42515542e-02 -3.78771536e-02 -9.73223448e-01
-4.19530630e-01 8.99997830e-01 -6.48155510e-02 -2.24505682e-02
1.19976819e+00 -3.43575299e-01 -5.82368635e-02 3.01515937e-01
9.80873942e-01 -2.43417948e-01 -1.23528063e+00 -4.07224089e-01
-1.36611775e-01 -7.99952328e-01 4.34459329e-01 -9.20901954e-01
-1.15552878e+00 1.22971594e+00 8.54641259e-01 -6.23157062e-02
1.42869878e+00 4.12693508e-02 6.51553810e-01 8.53812471e-02
3.77239794e-01 -1.04373682e+00 1.64648935e-01 4.03502256e-01
9.72418606e-01 -1.32280409e+00 -1.32220998e-01 -2.97636211e-01
-9.82030571e-01 8.39041352e-01 7.28328347e-01 -2.29483321e-01
5.40324152e-01 -4.06587005e-01 -7.83511028e-02 -6.20065816e-03
-7.93776870e-01 -4.55402613e-01 5.38582981e-01 6.74477577e-01
3.22142839e-01 -6.77302480e-02 -1.41291022e-01 4.08926636e-01
7.46880621e-02 2.03470111e-01 5.40795028e-01 5.28645694e-01
-8.50725710e-01 -5.75804710e-01 -4.34386641e-01 2.68353522e-01
-3.44706684e-01 -1.17045447e-01 -2.00827196e-01 2.28790209e-01
4.29806113e-01 8.10494184e-01 1.58724934e-01 -2.24230364e-01
6.89605623e-02 -5.08508563e-01 3.09239924e-01 -5.07103324e-01
-3.26911420e-01 -1.94951203e-02 -4.19732302e-01 -6.45001471e-01
-7.05895722e-01 -6.65191948e-01 -9.45990026e-01 3.13300043e-01
-7.45821446e-02 -1.80881158e-01 9.65151012e-01 8.30829501e-01
4.74059135e-01 3.82256776e-01 8.48013878e-01 -1.13105714e+00
-1.62491962e-01 -9.66050267e-01 -5.43889761e-01 5.47540247e-01
4.30467844e-01 -7.60947943e-01 -8.12347457e-02 2.58148819e-01] | [9.804020881652832, -0.3516635000705719] |
486353ef-2d2f-4e00-9c56-5ab5e66947b2 | pruning-vs-quantization-which-is-better | 2307.02973 | null | https://arxiv.org/abs/2307.02973v1 | https://arxiv.org/pdf/2307.02973v1.pdf | Pruning vs Quantization: Which is Better? | Neural network pruning and quantization techniques are almost as old as neural networks themselves. However, to date only ad-hoc comparisons between the two have been published. In this paper, we set out to answer the question on which is better: neural network quantization or pruning? By answering this question, we hope to inform design decisions made on neural network hardware going forward. We provide an extensive comparison between the two techniques for compressing deep neural networks. First, we give an analytical comparison of expected quantization and pruning error for general data distributions. Then, we provide lower bounds for the per-layer pruning and quantization error in trained networks, and compare these to empirical error after optimization. Finally, we provide an extensive experimental comparison for training 8 large-scale models on 3 tasks. Our results show that in most cases quantization outperforms pruning. Only in some scenarios with very high compression ratio, pruning might be beneficial from an accuracy standpoint. | ['Tijmen Blankevoort', 'Arash Behboodi', 'Mart van Baalen', 'Markus Nagel', 'Andrey Kuzmin'] | 2023-07-06 | null | null | null | null | ['network-pruning', 'quantization'] | ['methodology', 'methodology'] | [ 5.20056009e-01 1.80461496e-01 -1.77006140e-01 -5.91856718e-01
-5.37325859e-01 -8.18693116e-02 7.56485164e-02 3.07101578e-01
-9.11387980e-01 6.56023979e-01 -2.30154935e-02 -7.16288626e-01
-2.10825458e-01 -6.56804025e-01 -8.66210222e-01 -4.82980460e-01
-1.62505656e-01 2.15423957e-01 1.84485897e-01 2.23466858e-01
4.25887942e-01 6.16963148e-01 -1.86618781e+00 3.76482069e-01
4.69248593e-01 1.43000567e+00 2.13391766e-01 6.97584331e-01
-8.46230611e-02 9.71407175e-01 -8.79353225e-01 -5.82637668e-01
3.54857087e-01 -1.71869323e-01 -8.49822998e-01 -2.64604568e-01
7.48431623e-01 -6.64851069e-01 -4.67945486e-01 1.27139080e+00
4.72985506e-01 5.51264174e-02 5.35264850e-01 -8.87836516e-01
-4.50175762e-01 9.24695134e-01 -1.08072527e-01 7.05907643e-01
-3.79117429e-01 -5.60458265e-02 8.44403923e-01 -6.01537526e-01
2.21207991e-01 1.10679770e+00 8.94514441e-01 7.38584101e-01
-1.04424250e+00 -7.62392759e-01 9.52832401e-02 3.15231234e-01
-1.34218097e+00 -9.15147364e-01 4.79568362e-01 -2.12721862e-02
1.32507253e+00 1.47621647e-01 6.58287823e-01 6.51331306e-01
1.27253667e-01 9.89871204e-01 7.53925145e-01 -5.84046662e-01
4.59607482e-01 2.99970750e-02 3.36806118e-01 7.66119897e-01
6.31396830e-01 1.23204961e-01 -5.48389316e-01 4.91866246e-02
7.34956264e-01 -3.35698664e-01 -1.71073437e-01 9.83307883e-02
-5.07710993e-01 9.59601879e-01 3.44463736e-01 1.86756238e-01
-2.44393587e-01 6.93263769e-01 5.85055232e-01 4.99063522e-01
2.06122726e-01 3.47995579e-01 -5.21579742e-01 -1.79647297e-01
-1.38586497e+00 3.76518220e-01 9.00409758e-01 1.01852429e+00
4.11266178e-01 5.08230031e-01 2.24327847e-01 9.62788343e-01
2.63096511e-01 1.90114025e-02 6.60142779e-01 -1.34113669e+00
5.99454463e-01 -9.20580178e-02 -2.91604072e-01 -8.42970133e-01
-3.67323637e-01 -4.36830431e-01 -8.44860494e-01 1.53704971e-01
4.15236771e-01 -4.51128751e-01 -7.93733180e-01 1.52218080e+00
-2.90895462e-01 -1.60484999e-01 1.05964959e-01 5.66321850e-01
6.49686515e-01 5.24307966e-01 5.56175644e-03 -2.09138840e-01
1.13399172e+00 -6.66684628e-01 -8.61031175e-01 -3.40489298e-01
9.21255589e-01 -6.16818070e-01 7.01006591e-01 7.31164038e-01
-1.42077994e+00 -5.04263818e-01 -1.41963851e+00 -2.55464196e-01
-3.15386832e-01 2.36855343e-01 7.97705412e-01 1.06890452e+00
-1.19888484e+00 1.23899138e+00 -1.11017203e+00 -2.31006563e-01
7.99211800e-01 7.86558807e-01 2.17763945e-01 5.16048931e-02
-1.10397065e+00 9.92949843e-01 7.04655349e-01 -3.81327234e-02
-4.87259030e-01 -3.81257296e-01 -6.31606162e-01 3.44765216e-01
-3.75663824e-02 -4.66573268e-01 1.71174777e+00 -7.88756728e-01
-1.23577631e+00 5.94829082e-01 -2.13666141e-01 -1.32177985e+00
-6.63167387e-02 -2.68170029e-01 -2.85006046e-01 2.71887749e-01
-5.91462910e-01 1.23430121e+00 4.66640979e-01 -8.46296549e-01
-9.30173039e-01 -3.15239936e-01 6.60638809e-02 1.50274083e-01
-7.35940278e-01 5.27407862e-02 -2.56556541e-01 -7.71888256e-01
1.21416546e-01 -5.13645709e-01 -1.66563049e-01 4.21088964e-01
-1.38136104e-01 -2.55527735e-01 6.80204749e-01 -4.89666492e-01
1.28439748e+00 -2.13750172e+00 -3.72458547e-01 4.98065837e-02
1.66108727e-01 5.24921238e-01 -5.73758185e-02 -1.43890113e-01
1.75503343e-02 4.11959171e-01 -8.08620155e-02 -5.56949198e-01
-2.76972186e-02 4.76318270e-01 -4.34776336e-01 3.47995669e-01
2.57310569e-01 6.54729962e-01 -5.71902752e-01 -6.59280479e-01
-8.54612514e-02 5.54028630e-01 -7.57363319e-01 -2.81648844e-01
-2.36445423e-02 -4.82123166e-01 -1.37244284e-01 7.02387989e-01
4.11294550e-01 2.90969778e-02 1.27433315e-02 -1.30873308e-01
1.36954933e-01 6.39033973e-01 -8.82860661e-01 1.34973586e+00
-2.69644439e-01 1.20778942e+00 8.61324519e-02 -1.31624091e+00
9.12977993e-01 1.78882331e-01 9.79137719e-02 -7.07467914e-01
4.78542089e-01 2.50830293e-01 2.18960866e-01 -1.09548002e-01
6.43211842e-01 -2.18143404e-01 3.37941915e-01 1.91355601e-01
3.70396376e-01 1.39148505e-02 3.97478342e-01 -2.08003908e-01
1.16029215e+00 -2.98708618e-01 3.02565116e-02 -2.65207678e-01
-1.15395270e-01 -1.63981933e-02 3.30103993e-01 8.05910826e-01
-2.65939713e-01 5.37128866e-01 6.39071763e-01 -4.79026347e-01
-1.10931706e+00 -7.21481681e-01 -4.02413368e-01 9.12976384e-01
-2.74626344e-01 -5.75445533e-01 -1.06968629e+00 -2.48227835e-01
-1.93834797e-01 8.29707563e-01 -4.46840227e-01 -1.48011506e-01
-5.32236755e-01 -9.96254265e-01 1.03224289e+00 7.72774994e-01
6.18757904e-01 -8.49845529e-01 -1.14774323e+00 7.63964280e-02
1.61597848e-01 -1.11440468e+00 9.90012512e-02 1.01530015e+00
-1.80981648e+00 -7.52610981e-01 -4.94548559e-01 -8.90587747e-01
3.32364887e-01 1.26343966e-01 9.51051235e-01 3.42052013e-01
1.17979441e-02 -1.27495944e-01 -1.01108491e-01 -6.71165168e-01
-3.53180230e-01 3.62159193e-01 1.47654161e-01 -8.07163358e-01
7.48141468e-01 -7.73957312e-01 -4.12132889e-01 -3.52645628e-02
-8.47231567e-01 -1.61026955e-01 9.05802906e-01 5.90544105e-01
6.22884154e-01 4.30025846e-01 4.81813103e-01 -9.16341841e-01
8.92283678e-01 -2.62463450e-01 -6.63444459e-01 4.58539911e-02
-7.69729197e-01 3.79483700e-01 6.64725900e-01 -5.81363797e-01
-4.85005438e-01 1.03025243e-01 -5.94614863e-01 -7.51273811e-01
-2.08392575e-01 5.70834816e-01 1.52188048e-01 -2.87950095e-02
8.33395958e-01 1.15742184e-01 -2.16516346e-01 -6.94487393e-01
1.60937309e-01 5.64047635e-01 5.79534113e-01 -4.10360426e-01
3.24749678e-01 1.43456891e-01 -1.81079432e-02 -1.01713824e+00
-6.93525136e-01 -5.58893569e-02 -4.94878292e-01 2.60143369e-01
5.77879965e-01 -7.00838923e-01 -3.54383975e-01 3.65022160e-02
-1.22475755e+00 -6.45862222e-01 -5.64036608e-01 3.66064936e-01
-7.53282726e-01 1.98096111e-01 -8.59143078e-01 -8.91953170e-01
-4.90204751e-01 -1.16891205e+00 6.70433342e-01 1.25761390e-01
-2.92900592e-01 -9.27038729e-01 -4.21468139e-01 -5.33983074e-02
4.42068309e-01 -2.07037523e-01 1.03234518e+00 -8.67515624e-01
-3.47956568e-01 -1.96375832e-01 -3.73658270e-01 6.54785097e-01
-2.46712431e-01 -8.33717212e-02 -1.20192313e+00 -1.27660170e-01
2.25838453e-01 -4.88634676e-01 1.10967183e+00 5.62727869e-01
1.81771731e+00 -6.59895837e-01 -2.07232714e-01 8.47899556e-01
1.45670664e+00 2.74745822e-01 7.21188784e-01 1.70066267e-01
2.83052623e-01 4.84895200e-01 2.82347798e-01 3.37285608e-01
-1.21202648e-01 2.24517867e-01 4.30466741e-01 2.87528396e-01
-3.23061794e-01 -8.44550356e-02 1.86113074e-01 1.01212597e+00
2.88218372e-02 -3.29022884e-01 -7.86148548e-01 4.69181210e-01
-1.26156414e+00 -9.36803937e-01 2.42738500e-01 2.18349242e+00
1.00628698e+00 7.22695172e-01 -1.29625842e-01 6.69102669e-01
5.20725667e-01 8.44929218e-02 -5.75466692e-01 -8.15781713e-01
-1.12363265e-03 6.25819743e-01 9.54525530e-01 3.81415427e-01
-9.89501476e-01 7.46644080e-01 7.76057529e+00 1.08663952e+00
-1.04738569e+00 4.87037487e-02 9.56832707e-01 -6.08779848e-01
8.25923607e-02 -2.50558972e-01 -1.28295469e+00 2.13187188e-01
1.56500471e+00 1.19890109e-01 5.08921981e-01 9.76285696e-01
-2.29621559e-01 -1.40313163e-01 -1.23134494e+00 1.22197461e+00
-1.00631237e-01 -1.53245389e+00 1.73190117e-01 1.57773614e-01
5.92089832e-01 2.86490232e-01 1.18000627e-01 3.65903854e-01
2.01322526e-01 -1.18704295e+00 7.31208324e-01 1.95353568e-01
7.66986609e-01 -9.91011441e-01 7.47437179e-01 5.34761429e-01
-8.40082586e-01 -2.22549349e-01 -9.27675843e-01 -1.70646846e-01
-8.49993899e-02 6.59246027e-01 -4.76360410e-01 -1.80400699e-01
6.76537335e-01 2.56700248e-01 -5.83804905e-01 1.26450276e+00
9.85263661e-03 8.31594288e-01 -6.83496594e-01 -2.97329962e-01
2.76909143e-01 2.71119893e-01 -2.23002557e-04 1.53045738e+00
2.92059034e-01 1.36293501e-01 -5.57798862e-01 6.95145488e-01
-4.12315011e-01 -2.02370271e-01 -6.06168509e-01 -2.15812579e-01
7.54300475e-01 8.02284718e-01 -9.23879445e-01 -4.19483244e-01
-3.39944124e-01 7.65990853e-01 4.70167249e-01 1.99611127e-01
-6.38910890e-01 -7.55647242e-01 5.70845425e-01 9.91918817e-02
4.74471331e-01 -2.23226026e-01 -7.44099379e-01 -7.36622095e-01
-7.34816194e-02 -6.25487685e-01 7.36885369e-02 -5.59122384e-01
-7.06093371e-01 5.60449600e-01 1.53621554e-01 -7.11477399e-01
-3.09814960e-01 -1.03305626e+00 -3.81857872e-01 5.49344599e-01
-1.43012857e+00 -8.59744549e-02 1.15623429e-01 -4.52769967e-03
6.54969335e-01 -1.95544973e-01 8.55295777e-01 4.94170398e-01
-4.32440102e-01 1.00058758e+00 2.94346124e-01 1.63860857e-01
1.72259822e-01 -1.01977634e+00 4.27409649e-01 6.62163079e-01
3.50121588e-01 7.41688967e-01 8.09970856e-01 -2.84710199e-01
-1.20021784e+00 -1.15468264e+00 9.61168945e-01 -5.10197766e-02
3.17170352e-01 -9.69421193e-02 -1.07105696e+00 5.83159029e-01
7.13809729e-02 -8.68956894e-02 7.60119319e-01 6.34817928e-02
-8.98353755e-02 -8.32940545e-03 -1.14562523e+00 6.25053823e-01
1.07663095e+00 -2.80674011e-01 -6.27073288e-01 4.15372729e-01
7.17301548e-01 -4.53180432e-01 -9.03616071e-01 3.67347598e-01
6.36214316e-01 -1.00179052e+00 8.98653090e-01 -3.51118356e-01
6.84683859e-01 2.53947884e-01 -6.83050275e-01 -8.55816662e-01
1.08076108e-03 -3.41759294e-01 -5.24562955e-01 8.76830995e-01
4.44236189e-01 -3.14261764e-01 1.30306804e+00 4.66889083e-01
-2.29074031e-01 -1.37197399e+00 -9.44993734e-01 -1.00922370e+00
3.80201817e-01 -7.56835759e-01 4.24315572e-01 4.93273258e-01
-1.29436180e-01 2.65667796e-01 -3.25001068e-02 -1.04469359e-01
5.22350252e-01 -4.69535947e-01 1.67785868e-01 -1.09688187e+00
-2.20919013e-01 -8.94478977e-01 -5.23408413e-01 -1.61980772e+00
2.30599474e-02 -7.26236939e-01 4.11426388e-02 -1.37334704e+00
-1.98768079e-01 -4.82760221e-01 -3.21028024e-01 5.45496762e-01
3.68945986e-01 2.90489584e-01 1.69758201e-01 1.22384228e-01
-4.14355397e-01 5.00106454e-01 7.25903809e-01 -1.04564890e-01
-1.54022323e-02 5.12974570e-03 -7.26132751e-01 9.55647707e-01
1.21529543e+00 -6.01078689e-01 -5.94116509e-01 -9.61147666e-01
-3.14374603e-02 -2.03632414e-01 9.01870728e-02 -1.63561022e+00
4.77601826e-01 1.34620383e-01 5.38540959e-01 -9.01849926e-01
4.48919088e-01 -7.61655211e-01 -4.16488469e-01 6.50855005e-01
-7.24452198e-01 1.40617996e-01 5.58413088e-01 3.62711608e-01
-1.73701152e-01 -9.55346048e-01 1.02523518e+00 -6.16730414e-02
-5.38200259e-01 1.18447311e-01 -5.92486084e-01 -1.88341737e-02
4.25846905e-01 -5.72159469e-01 -1.62463784e-01 -2.57930517e-01
-7.10074246e-01 2.55577751e-02 1.50207773e-01 -6.24831840e-02
9.00440454e-01 -1.16438818e+00 -3.08361381e-01 1.36425555e-01
-3.87343794e-01 1.66590407e-01 -1.99421987e-01 4.86400962e-01
-7.91826487e-01 7.56359696e-01 -1.32397652e-01 -2.61661559e-01
-1.32866561e+00 5.38853228e-01 4.06247199e-01 -9.90989059e-02
-3.86128515e-01 1.14783013e+00 -3.71038496e-01 1.38103096e-02
1.08830607e+00 -8.54250431e-01 -1.89571660e-02 -3.20604257e-02
8.09814632e-01 4.01877791e-01 3.69160593e-01 -2.38701984e-01
4.79137972e-02 1.87620342e-01 -1.12783022e-01 -1.23062246e-01
1.19420648e+00 2.06405848e-01 3.03415179e-01 5.32781720e-01
1.46526206e+00 -7.64219224e-01 -1.23340797e+00 -2.16860965e-01
3.83806936e-02 -2.34238163e-01 4.34213519e-01 -3.92490774e-01
-1.38189292e+00 1.50746524e+00 8.07958484e-01 2.46091455e-01
1.40601480e+00 -2.64724821e-01 8.66019607e-01 1.19617724e+00
1.99288487e-01 -1.18974864e+00 -1.88031331e-01 7.51847327e-01
5.85097671e-01 -7.07790256e-01 2.98033416e-01 -2.00172156e-01
-7.32001439e-02 1.30957174e+00 4.51850176e-01 -3.20866525e-01
7.11819232e-01 5.58720171e-01 -4.08843577e-01 -4.54299450e-02
-1.08896005e+00 1.47307023e-01 -7.81686753e-02 6.89408243e-01
5.74646533e-01 -3.82438779e-01 -6.59622848e-02 4.15797889e-01
-8.23481441e-01 2.63106108e-01 2.89199412e-01 1.06614590e+00
-1.06566060e+00 -8.96094382e-01 -1.87876403e-01 9.74065483e-01
-7.88738489e-01 -4.34298843e-01 -1.49813622e-01 7.27834642e-01
1.96326956e-01 7.30865121e-01 4.55832928e-01 -6.17282331e-01
1.80706710e-01 2.15636626e-01 8.15699041e-01 -5.38065791e-01
-4.76524651e-01 -2.20931202e-01 2.91737109e-01 -3.10197324e-01
-2.09098235e-01 -4.43042845e-01 -1.24298322e+00 -6.71752036e-01
-5.46974003e-01 -6.29877225e-02 8.38375628e-01 9.22758043e-01
6.40714392e-02 5.57068944e-01 -1.26995355e-01 -8.08440924e-01
-9.75245714e-01 -8.72548223e-01 -5.81051409e-01 -2.81611800e-01
2.89375216e-01 -3.21811527e-01 -2.25593254e-01 1.57686248e-01] | [8.532161712646484, 3.14851975440979] |
33a81e94-93f4-4801-a941-f62d7a8d4209 | auto-encoding-knowledge-graph-for | 2111.04318 | null | https://arxiv.org/abs/2111.04318v2 | https://arxiv.org/pdf/2111.04318v2.pdf | Auto-Encoding Knowledge Graph for Unsupervised Medical Report Generation | Medical report generation, which aims to automatically generate a long and coherent report of a given medical image, has been receiving growing research interests. Existing approaches mainly adopt a supervised manner and heavily rely on coupled image-report pairs. However, in the medical domain, building a large-scale image-report paired dataset is both time-consuming and expensive. To relax the dependency on paired data, we propose an unsupervised model Knowledge Graph Auto-Encoder (KGAE) which accepts independent sets of images and reports in training. KGAE consists of a pre-constructed knowledge graph, a knowledge-driven encoder and a knowledge-driven decoder. The knowledge graph works as the shared latent space to bridge the visual and textual domains; The knowledge-driven encoder projects medical images and reports to the corresponding coordinates in this latent space and the knowledge-driven decoder generates a medical report given a coordinate in this space. Since the knowledge-driven encoder and decoder can be trained with independent sets of images and reports, KGAE is unsupervised. The experiments show that the unsupervised KGAE generates desirable medical reports without using any image-report training pairs. Moreover, KGAE can also work in both semi-supervised and supervised settings, and accept paired images and reports in training. By further fine-tuning with image-report pairs, KGAE consistently outperforms the current state-of-the-art models on two datasets. | ['Xu sun', 'Sheng Wang', 'Shen Ge', 'Xian Wu', 'Chenyu You', 'Fenglin Liu'] | 2021-11-08 | null | http://proceedings.neurips.cc/paper/2021/hash/876e1c59023b1a0e95808168e1a8ff89-Abstract.html | http://proceedings.neurips.cc/paper/2021/file/876e1c59023b1a0e95808168e1a8ff89-Paper.pdf | neurips-2021-12 | ['medical-report-generation'] | ['medical'] | [ 3.42265874e-01 5.42950571e-01 -3.60996425e-01 -5.32457888e-01
-1.15586090e+00 -2.42657259e-01 4.80336845e-01 1.32294983e-01
-1.11672536e-01 4.94446844e-01 2.76659638e-01 3.37869227e-02
4.31323890e-03 -9.16623294e-01 -6.19018197e-01 -5.31984866e-01
1.86122954e-01 5.93637526e-01 9.55665261e-02 2.95054197e-01
-1.69179678e-01 -2.61360228e-01 -1.18044078e+00 4.06424016e-01
7.84932256e-01 7.29472876e-01 5.20253539e-01 7.96801984e-01
-2.61889994e-01 1.14497650e+00 -4.73342836e-01 -8.78469646e-01
-1.36557072e-01 -9.37966347e-01 -8.54285598e-01 6.81072950e-01
2.02282444e-01 -2.23370448e-01 -2.61590362e-01 1.06417894e+00
4.42650497e-01 -3.64883363e-01 7.27551877e-01 -1.17318523e+00
-1.25515461e+00 7.26357639e-01 -6.23400033e-01 -2.46783104e-02
3.60123903e-01 -2.70396993e-02 8.56905401e-01 -9.55849171e-01
1.03289270e+00 8.66415799e-01 3.32962543e-01 4.51046169e-01
-1.23780751e+00 -6.64013863e-01 8.25874216e-04 4.51994874e-02
-1.77091503e+00 -2.45829552e-01 9.08060610e-01 -5.94734669e-01
7.46954143e-01 4.13660221e-02 7.13243902e-01 8.63797724e-01
2.34924927e-01 8.17178965e-01 8.37922335e-01 -2.77364314e-01
5.43520451e-02 4.25873905e-01 -2.59714931e-01 1.12411761e+00
1.35722160e-01 -1.56564668e-01 -7.37340808e-01 4.79780417e-03
8.92780125e-01 1.19905263e-01 -4.18745399e-01 -3.36713731e-01
-1.52763712e+00 9.95169878e-01 4.62852806e-01 3.11085939e-01
-6.57843590e-01 -2.16357574e-01 2.84902841e-01 2.75855571e-01
6.84085608e-01 1.17048331e-01 1.63688455e-02 3.85040790e-01
-1.21779823e+00 -2.09873705e-03 6.70464158e-01 1.27883852e+00
8.11152935e-01 -3.69196624e-01 -3.58274311e-01 7.37132311e-01
3.17806393e-01 6.04892075e-01 4.43449229e-01 -5.31799316e-01
7.57789433e-01 8.59624326e-01 -2.61311740e-01 -1.48597455e+00
-2.43137494e-01 -5.66760540e-01 -1.20037115e+00 -3.96247864e-01
-1.48175880e-01 7.41143450e-02 -9.56989110e-01 1.58564377e+00
3.22777629e-01 1.65404201e-01 3.74508411e-01 9.27856445e-01
1.19175005e+00 7.53499269e-01 -7.99827278e-02 -6.34560704e-01
1.33194888e+00 -1.13894093e+00 -9.56413984e-01 -1.17711350e-01
7.29641497e-01 -8.52446854e-01 8.78893375e-01 7.94651210e-02
-1.28890336e+00 -6.04654431e-01 -8.87943089e-01 -7.45136589e-02
-2.89600007e-02 6.16137862e-01 3.35257232e-01 1.39246687e-01
-1.11209011e+00 -2.01489300e-01 -7.53843248e-01 -3.81917715e-01
4.14410472e-01 1.52350962e-01 -7.34578192e-01 -5.50973117e-01
-7.99004614e-01 6.30963326e-01 6.45386636e-01 -4.01050374e-02
-8.56099546e-01 -5.61851144e-01 -1.15242934e+00 -1.58609822e-01
6.06869936e-01 -1.07285166e+00 1.03801346e+00 -8.00455034e-01
-1.17541039e+00 1.37515080e+00 1.95969623e-02 -2.55572110e-01
4.28811073e-01 1.91402972e-01 -6.94684803e-01 5.94343007e-01
6.49571240e-01 9.33834136e-01 1.03394938e+00 -1.36830962e+00
-4.92537469e-01 -1.86621279e-01 -2.48933479e-01 1.49724782e-01
-3.38287920e-01 -2.19859228e-01 -1.20409894e+00 -8.42213392e-01
1.63087115e-01 -1.00159240e+00 -1.89091504e-01 -2.24725693e-01
-7.31989264e-01 -3.09724882e-02 6.01137161e-01 -6.25523090e-01
1.33631372e+00 -2.16945553e+00 2.29354426e-01 2.41111040e-01
3.92154127e-01 -4.34107147e-02 -1.40826494e-01 3.30205530e-01
-2.17314899e-01 -1.79539055e-01 -4.56522495e-01 -5.73053360e-01
-2.94158757e-01 3.47632974e-01 -1.74244836e-01 2.33189255e-01
3.15020323e-01 1.11708784e+00 -1.20295990e+00 -1.18914044e+00
-1.27908394e-01 3.68361443e-01 -5.62532008e-01 5.56576312e-01
-1.04549050e-01 3.35297078e-01 -3.87838930e-01 4.43230987e-01
4.57051098e-01 -9.61210847e-01 3.78001481e-01 -5.18229187e-01
2.85563141e-01 5.41447774e-02 -7.92561293e-01 2.19741750e+00
-4.66640323e-01 2.86256075e-01 -2.38886476e-01 -9.45121467e-01
1.02812576e+00 6.50447845e-01 5.00983000e-01 -6.42318785e-01
-8.48297924e-02 1.84375539e-01 -4.57121968e-01 -7.36614287e-01
4.18933451e-01 -4.16762531e-01 -2.25400060e-01 8.05503666e-01
4.06831563e-01 -2.24669978e-01 2.57889658e-01 7.46402979e-01
1.16758859e+00 -8.95804390e-02 2.02262074e-01 3.07168633e-01
4.98740554e-01 2.16221943e-01 5.00053346e-01 4.53823596e-01
2.14600995e-01 9.21486259e-01 3.51132691e-01 -2.47251719e-01
-9.50231850e-01 -1.28507686e+00 2.55750537e-01 6.24602556e-01
1.56970888e-01 -6.84556067e-01 -6.35990918e-01 -1.25796962e+00
-3.29782069e-01 5.42093158e-01 -6.10025883e-01 -3.09620708e-01
-1.07104465e-01 -4.79846984e-01 2.72302985e-01 5.56050718e-01
5.52520871e-01 -9.86895442e-01 -1.26328796e-01 2.36360922e-01
-6.12208545e-01 -1.34516037e+00 -9.65101302e-01 -1.53499618e-01
-5.52093029e-01 -1.13888073e+00 -9.68533635e-01 -1.01511502e+00
1.37507010e+00 1.81126058e-01 1.34434342e+00 1.21625639e-01
-2.42140576e-01 6.63128793e-01 -5.06375253e-01 -2.67393053e-01
-8.40954065e-01 4.13673632e-02 -2.30251610e-01 3.13575596e-01
8.35254043e-02 -3.60011101e-01 -5.65627158e-01 1.76374346e-01
-1.29643488e+00 6.96917176e-01 9.40725982e-01 8.57340515e-01
9.39463675e-01 1.40479982e-01 5.15592039e-01 -1.12083125e+00
4.31322545e-01 -5.28277397e-01 -3.02355230e-01 4.70769882e-01
-5.34103692e-01 2.45952442e-01 3.93415391e-01 -1.27002120e-01
-1.05647135e+00 3.01562876e-01 8.98556933e-02 -7.95220256e-01
1.29589379e-01 8.28348935e-01 5.53785376e-02 5.62480807e-01
4.86801028e-01 5.36312819e-01 7.55806938e-02 -1.59127250e-01
5.11722803e-01 5.18682957e-01 8.93251359e-01 4.96692695e-02
8.67969215e-01 4.77645487e-01 -3.25742811e-01 -2.67940491e-01
-1.18707848e+00 -5.80768287e-01 -7.06312120e-01 -2.61155695e-01
1.21799946e+00 -1.13182521e+00 -2.14555666e-01 -5.41665107e-02
-1.29130781e+00 1.12592861e-01 -3.28693897e-01 6.58253789e-01
-6.91022456e-01 6.11594431e-02 -5.64867556e-01 -3.22740078e-01
-5.50110281e-01 -1.19718993e+00 1.08140635e+00 -5.88835143e-02
-1.78531885e-01 -1.07487833e+00 9.21153501e-02 5.54074764e-01
-2.27777034e-01 1.65603921e-01 7.30287433e-01 -3.41348022e-01
-6.40021622e-01 -2.09527209e-01 -4.20324981e-01 3.30165893e-01
4.52091664e-01 -2.72987485e-01 -7.23348379e-01 -1.78339824e-01
-1.00804590e-01 -5.48204064e-01 9.00084913e-01 1.21797629e-01
8.39038670e-01 -5.59529781e-01 -4.47508812e-01 3.20887566e-01
1.31424284e+00 1.01918718e-02 3.94224644e-01 -1.97516829e-01
8.91576111e-01 5.66792488e-01 6.39538229e-01 1.65463507e-01
1.10411346e+00 5.38352489e-01 1.74873486e-01 -4.51877654e-01
-1.95772767e-01 -5.95401824e-01 2.73661703e-01 1.62742829e+00
1.06742024e-01 -1.73402220e-01 -8.81480455e-01 9.26379204e-01
-2.02865410e+00 -8.92908752e-01 7.66726658e-02 1.92400658e+00
1.29706848e+00 -9.83859673e-02 3.46417949e-02 -4.77559119e-02
7.15931654e-01 -1.01768486e-01 -3.23804021e-01 9.89296511e-02
9.72232148e-02 -5.58145903e-02 2.20461324e-01 2.79398918e-01
-9.31781769e-01 6.99947059e-01 5.95567894e+00 6.26384139e-01
-1.05351508e+00 4.18834537e-01 6.49462938e-01 7.37982988e-02
-4.26064551e-01 -2.29370743e-01 -4.34104741e-01 3.25276315e-01
8.10573697e-01 -2.01284498e-01 -3.03705186e-01 8.86754036e-01
-1.17965385e-01 -5.51858693e-02 -1.14852929e+00 1.36371911e+00
5.04465580e-01 -1.80938959e+00 4.53331381e-01 1.15237094e-01
9.34999108e-01 -3.26214850e-01 1.64694339e-02 6.05305955e-02
4.96512920e-01 -7.73623466e-01 7.48150527e-01 7.45517313e-01
1.19760656e+00 -7.05107033e-01 9.50404525e-01 2.98811942e-01
-1.27601922e+00 2.90815294e-01 -1.57687902e-01 4.80740875e-01
3.84179920e-01 8.03945005e-01 -1.19228232e+00 1.09670997e+00
5.74420631e-01 9.91200209e-01 -4.84398752e-01 6.72032535e-01
-4.69667673e-01 5.66474497e-01 -3.60983820e-03 4.94624525e-01
1.63836822e-01 -7.14774206e-02 1.61545932e-01 1.28121746e+00
3.66452873e-01 1.75492898e-01 5.93938589e-01 1.12197292e+00
-1.36562973e-01 1.34866282e-01 -7.17619300e-01 -2.38260746e-01
3.62964272e-02 1.29785919e+00 -9.58711922e-01 -6.78869247e-01
-6.09085739e-01 1.31915748e+00 3.26728761e-01 2.24253431e-01
-7.73004591e-01 -1.41536474e-01 -3.09372563e-02 9.15134102e-02
2.94143885e-01 2.72306073e-02 -9.34728161e-02 -1.11868632e+00
2.44027749e-01 -4.83672231e-01 6.84898078e-01 -1.02101099e+00
-1.22566414e+00 7.90465772e-01 -3.87003906e-02 -1.56378794e+00
-4.49919492e-01 -6.96867984e-03 -2.59730518e-01 5.81674159e-01
-1.52503300e+00 -1.46919799e+00 -3.92078459e-01 8.24628234e-01
3.44810456e-01 -1.90706342e-01 1.00258493e+00 4.62331086e-01
-4.17874843e-01 5.47549248e-01 -4.12883729e-01 4.63476926e-01
7.36240387e-01 -1.15549803e+00 6.82416260e-02 7.28085637e-01
4.36760128e-01 5.17065227e-01 3.31514716e-01 -8.94367218e-01
-1.18567526e+00 -1.50510395e+00 9.76576328e-01 -3.74824107e-01
3.87552500e-01 -2.38281801e-01 -8.47883701e-01 8.29851866e-01
3.55228186e-01 2.93474525e-01 1.01214838e+00 -2.93278664e-01
-3.88987899e-01 -1.15220487e-01 -8.62724125e-01 3.98175508e-01
9.87189233e-01 -7.79879570e-01 -4.60205436e-01 5.23554742e-01
9.31271970e-01 -5.08453608e-01 -1.08625722e+00 1.55951992e-01
1.41250283e-01 -8.35820377e-01 6.86252177e-01 -6.55153990e-02
7.94101715e-01 -5.47447443e-01 1.50755033e-01 -1.32227182e+00
-2.89331079e-01 -3.58446509e-01 9.09769386e-02 1.36674607e+00
6.60893142e-01 -1.45918980e-01 5.38894653e-01 2.12410286e-01
-2.02728495e-01 -7.77347684e-01 -5.92337728e-01 -6.08857095e-01
-5.78464508e-01 -6.43223166e-01 4.07690912e-01 1.10567617e+00
1.67401195e-01 6.34995341e-01 -4.41705346e-01 3.28757316e-01
4.90564257e-01 2.71968424e-01 7.36759961e-01 -7.86900282e-01
-2.53265291e-01 -7.36066606e-03 -4.67717826e-01 -6.77257001e-01
-1.84494585e-01 -1.24757481e+00 1.88024476e-01 -2.05205727e+00
7.05676496e-01 -4.32795048e-01 -8.07484686e-02 8.49692583e-01
-4.00869310e-01 5.14970183e-01 1.72765985e-01 4.67050552e-01
-1.03983700e+00 4.84805316e-01 1.37420583e+00 -4.67133433e-01
-2.05601916e-01 -2.92110115e-01 -6.65794551e-01 6.53191805e-01
2.98831910e-01 -7.24359989e-01 -7.01887727e-01 -2.43014634e-01
4.62918192e-01 4.47115391e-01 3.31122756e-01 -8.22035015e-01
4.64912683e-01 -5.04948087e-02 3.03175360e-01 -8.29417825e-01
5.70686124e-02 -8.36177111e-01 4.20277983e-01 2.81343818e-01
-3.83163393e-01 7.22248703e-02 -1.23052552e-01 7.25162923e-01
-6.28249586e-01 1.69326991e-01 4.61653441e-01 -3.80204707e-01
-4.25160974e-01 5.95567763e-01 -2.01203744e-03 -1.17296865e-02
1.17408681e+00 -1.38852790e-01 -8.19857866e-02 -3.94947946e-01
-9.16935563e-01 3.15400183e-01 2.20596433e-01 6.16016090e-01
1.13728988e+00 -1.59982955e+00 -8.75829637e-01 2.38395691e-01
7.26340830e-01 3.95207137e-01 4.72513109e-01 1.07638919e+00
-2.15025201e-01 2.57501453e-01 2.12727979e-01 -8.82460654e-01
-1.24933898e+00 7.23071575e-01 -1.70456290e-01 -7.41272092e-01
-8.16839993e-01 7.59828389e-01 5.12811780e-01 -4.50230271e-01
-4.63144928e-02 -2.48870268e-01 -5.22410423e-02 -1.50831090e-02
7.19740927e-01 -2.36853614e-01 -5.21194488e-02 -8.27754557e-01
-2.45493338e-01 4.89392191e-01 -3.49718392e-01 -2.09442243e-01
1.33223999e+00 -2.80766398e-01 -1.51994959e-01 4.07794207e-01
1.36430979e+00 -1.68830678e-01 -8.27326417e-01 -5.46846688e-01
-1.64376751e-01 -1.85466960e-01 1.21039070e-01 -7.33890295e-01
-1.48710144e+00 4.87323344e-01 2.07508802e-01 -1.18166283e-02
1.36472857e+00 4.87448543e-01 7.61501312e-01 6.42489269e-02
3.09316248e-01 -8.86971354e-01 5.73184907e-01 7.09049823e-03
1.02872646e+00 -1.38171434e+00 -4.00892235e-02 -6.75225556e-01
-1.06444156e+00 8.59833539e-01 4.79683250e-01 2.39761412e-01
6.40286326e-01 1.94199741e-01 2.95026362e-01 -6.35495722e-01
-8.15045178e-01 -1.92387640e-01 6.04753315e-01 6.70709610e-01
4.22746867e-01 1.32639706e-01 -6.79683983e-02 6.59509897e-01
-2.75833428e-01 3.78020912e-01 3.00683498e-01 1.02024543e+00
2.57873777e-02 -9.57185566e-01 -3.98154706e-01 5.47290742e-01
-2.24942967e-01 -3.82960439e-02 -1.81629762e-01 4.87445205e-01
2.82096595e-01 8.48707497e-01 3.07106245e-02 -5.93613148e-01
1.81101352e-01 -1.07117265e-01 3.51486057e-01 -1.08529770e+00
-2.60577291e-01 6.21482022e-02 -1.06051229e-01 -4.44499075e-01
-7.51224756e-01 -2.96195298e-01 -1.46379507e+00 4.14447067e-03
-4.26873624e-01 2.54801840e-01 3.73224199e-01 1.12827075e+00
5.68480134e-01 7.47534454e-01 5.34560502e-01 -3.23700458e-01
1.04533412e-01 -8.31858695e-01 -7.10270405e-01 6.80243313e-01
3.36862713e-01 -4.17106092e-01 2.53566682e-01 8.84945095e-01] | [15.032795906066895, -1.4013023376464844] |
f41e346b-a6d6-42b6-abe0-a3afe85e2ff5 | bi-directional-joint-neural-networks-for | 2202.13079 | null | https://arxiv.org/abs/2202.13079v1 | https://arxiv.org/pdf/2202.13079v1.pdf | Bi-directional Joint Neural Networks for Intent Classification and Slot Filling | Intent classification and slot filling are two critical tasks for natural language understanding. Traditionally the two tasks proceeded independently. However, more recently joint models for intent classification and slot filling have achieved state-of-the-art performance, and have proved that there exists a strong relationship between the two tasks. In this paper, we propose a bi-directional joint model for intent classification and slot filling, which includes a multi-stage hierarchical process via BERT and bi-directional joint natural language understanding mechanisms, including intent2slot and slot2intent, to obtain mutual performance enhancement between intent classification and slot filling. The evaluations show that our model achieves state-of-the-art results on intent classification accuracy, slot filling F1, and significantly improves sentence-level semantic frame accuracy when applied to publicly available benchmark datasets, ATIS (88.6%) and SNIPS (92.8%). | ['Josiah Poon', 'Henry Weld', 'Huichun Li', 'Siqu Long', 'Soyeon Caren Han'] | 2022-02-26 | null | null | null | null | ['slot-filling'] | ['natural-language-processing'] | [ 2.73956239e-01 3.93705308e-01 -7.51557589e-01 -6.94441199e-01
-9.53869760e-01 -2.46935755e-01 6.54147267e-01 1.85922563e-01
-5.17785728e-01 7.59520590e-01 5.67641556e-01 -6.34231031e-01
2.82383859e-01 -7.85987437e-01 -2.03124747e-01 -1.34309247e-01
4.35865343e-01 7.68286467e-01 4.13404435e-01 -3.51127207e-01
4.50487405e-01 -4.94681597e-01 -1.48054576e+00 6.47920728e-01
1.15407693e+00 9.67630506e-01 2.88363248e-01 5.37297666e-01
-7.73539841e-01 1.16517544e+00 -4.26521629e-01 -2.69912034e-01
-3.13254565e-01 -2.23734215e-01 -1.35546255e+00 -1.84174374e-01
-1.18874170e-01 -2.21098393e-01 -1.72916755e-01 6.08207941e-01
1.12539809e-02 7.74108842e-02 4.58861411e-01 -1.28701627e+00
-4.69905227e-01 6.44967198e-01 -4.15709704e-01 5.05627431e-02
6.72904491e-01 -2.49730706e-01 1.49564922e+00 -6.43890023e-01
4.59955871e-01 1.21537089e+00 7.67819941e-01 4.63780463e-01
-9.39136386e-01 -5.86523354e-01 2.69320279e-01 2.49577507e-01
-9.86012697e-01 -2.80025482e-01 6.17742538e-01 -2.32115656e-01
1.35406613e+00 3.20893377e-01 2.72830933e-01 8.07016253e-01
2.71413982e-01 1.39396560e+00 1.06714153e+00 -6.84692740e-01
2.42443040e-01 1.91991925e-01 8.71539950e-01 6.77946329e-01
1.00919269e-02 -3.25086564e-01 -7.75534332e-01 -8.57539251e-02
5.08528590e-01 5.97174652e-02 1.34462044e-01 1.46067366e-01
-1.07260454e+00 1.08867228e+00 1.64888546e-01 4.92671728e-01
-2.67902285e-01 -7.81225562e-02 7.43756115e-01 6.14389069e-02
8.13232899e-01 3.74420345e-01 -6.74250662e-01 -6.03903890e-01
-8.08510602e-01 1.89052925e-01 9.56907272e-01 9.95189130e-01
8.22608173e-01 -4.88326341e-01 -5.67348719e-01 1.00850439e+00
5.90572417e-01 1.77654058e-01 6.37658536e-01 -5.74913263e-01
7.01813757e-01 1.05395103e+00 5.74369133e-02 -7.71021605e-01
-6.53229535e-01 -1.01099610e-01 -6.23902023e-01 -5.67865849e-01
9.24467109e-03 1.34719014e-01 -1.16295516e+00 1.84073174e+00
1.31007269e-01 -1.03781827e-01 3.89645964e-01 5.05482852e-01
1.04682517e+00 8.23083580e-01 9.19956684e-01 -6.83673173e-02
2.04884768e+00 -1.31342638e+00 -1.14759052e+00 -9.36987638e-01
1.14759576e+00 -7.86604822e-01 1.18717420e+00 -5.71852997e-02
-7.65838027e-01 -7.01576054e-01 -8.09854686e-01 -6.68806851e-01
-2.75139958e-01 2.33099133e-01 1.20907462e+00 7.34573364e-01
-7.82976687e-01 -1.33408695e-01 -8.13654661e-01 -4.55430537e-01
4.00351793e-01 7.50385672e-02 -2.29546919e-01 -1.56961143e-01
-1.46844578e+00 6.66391075e-01 6.68010831e-01 -4.81735259e-01
-3.06202561e-01 -8.31759095e-01 -1.15162051e+00 1.68445217e-03
4.48860586e-01 -8.88648987e-01 1.71117163e+00 -3.53172928e-01
-1.18087029e+00 1.09527314e+00 -7.34781265e-01 -6.88892901e-01
2.02583194e-01 -7.06497848e-01 -2.38584876e-01 -6.51030615e-02
7.27196634e-01 1.05283797e+00 2.19552055e-01 -8.17511916e-01
-1.07851505e+00 -5.35905182e-01 3.17562938e-01 6.99899018e-01
-2.34878033e-01 -1.68933257e-01 -5.41358650e-01 -4.54096735e-01
5.94329476e-01 -5.67442656e-01 -6.54494911e-02 -1.92615405e-01
-3.34344685e-01 -7.76316226e-01 1.00135446e+00 -8.62335205e-01
1.59086227e+00 -2.13892484e+00 -2.91875929e-01 -6.45252585e-01
7.65002565e-03 1.19919777e-01 1.75916255e-01 4.31129068e-01
6.53633103e-02 1.02299966e-01 -1.57734081e-01 -8.47350597e-01
1.03312880e-01 4.13123071e-01 -7.55502164e-01 -1.82274848e-01
-1.19411238e-01 1.15821123e+00 -7.63926268e-01 -6.54338062e-01
2.93247908e-01 3.20260739e-03 -7.60899663e-01 4.10681516e-01
-6.25289738e-01 3.24193723e-02 -6.44210875e-01 6.35494590e-01
4.03616488e-01 -4.28236872e-01 3.11102659e-01 -1.17961645e-01
3.10636554e-02 9.59736645e-01 -6.75392687e-01 2.12108612e+00
-6.25192165e-01 4.81417447e-01 -3.80211323e-01 -1.07686126e+00
9.02995527e-01 4.89055544e-01 3.91827077e-01 -1.14846802e+00
-6.44978508e-03 2.51060247e-01 -6.45373166e-01 -1.81807965e-01
1.09565258e+00 -4.68949109e-01 -7.85013735e-01 5.38458109e-01
1.35388881e-01 -2.36653805e-01 1.64151147e-01 1.74272090e-01
9.51649010e-01 6.18952848e-02 6.11963987e-01 -3.91499400e-01
5.42451859e-01 3.10417861e-01 4.71139431e-01 8.97873402e-01
-3.71249259e-01 3.97589207e-01 5.29212594e-01 -5.74609876e-01
-7.93467104e-01 -8.35294783e-01 -6.66313097e-02 1.43575966e+00
5.27623415e-01 -5.99537015e-01 -6.82982624e-01 -7.64678121e-01
-3.44436198e-01 1.23488629e+00 -5.10085523e-01 -2.72438288e-01
-3.82777780e-01 -5.94896317e-01 6.82572007e-01 7.23119855e-01
1.17696559e+00 -1.05989158e+00 -6.47327781e-01 3.10031205e-01
-9.33603287e-01 -1.55920458e+00 -1.61433280e-01 3.77859443e-01
-9.48242664e-01 -8.46337259e-01 -2.57966667e-01 -9.36807215e-01
3.71680588e-01 4.17794645e-01 1.50461173e+00 2.01004863e-01
-1.21821426e-01 3.38693053e-01 -6.51166320e-01 -3.02684188e-01
-1.00675657e-01 2.78542668e-01 -1.50526494e-01 -4.24399614e-01
6.94621146e-01 -2.41180629e-01 -4.34022665e-01 4.76471126e-01
-1.00551760e+00 7.92692721e-01 3.76867235e-01 8.63373816e-01
3.44030589e-01 1.87588826e-01 4.72975522e-01 -9.21214163e-01
5.32525957e-01 -3.28016371e-01 -5.86360842e-02 3.33791375e-01
-5.93541682e-01 1.12791494e-01 2.76013553e-01 2.07085446e-01
-1.49910426e+00 -1.94756135e-01 -5.28978646e-01 3.90300333e-01
-4.65367258e-01 5.69015026e-01 6.50978601e-03 5.24995089e-01
4.36006069e-01 4.28263992e-01 -1.89409837e-01 -3.27220231e-01
3.71740013e-01 9.15725887e-01 3.37405056e-01 -6.37674749e-01
2.15315267e-01 5.16924083e-01 -5.05330443e-01 -8.41708481e-01
-1.59551764e+00 -9.19513941e-01 -4.51242208e-01 3.06925148e-01
1.17706060e+00 -1.26759899e+00 -4.34124887e-01 6.84004962e-01
-1.26894605e+00 -2.73901433e-01 -2.70580292e-01 1.73311546e-01
-7.10950315e-01 4.68038291e-01 -6.94365025e-01 -9.83735979e-01
-6.93159342e-01 -1.06512868e+00 1.49005139e+00 4.55649614e-01
-5.35085857e-01 -1.23657131e+00 -4.26938683e-02 1.08415890e+00
2.32156560e-01 -6.29195571e-01 1.05428720e+00 -8.91288340e-01
-4.17787731e-01 -7.91871324e-02 -6.29931033e-01 -9.37287658e-02
2.28605330e-01 -8.70640159e-01 -8.89495075e-01 1.95552200e-01
7.97284171e-02 -5.62254727e-01 9.03577685e-01 4.05007422e-01
9.18163776e-01 -6.54731616e-02 -5.55040121e-01 1.37561068e-01
1.14283621e+00 5.10571837e-01 9.97257292e-01 4.59784359e-01
4.37612563e-01 3.88303995e-01 1.10136735e+00 4.38047260e-01
1.06649363e+00 8.33080769e-01 5.45640923e-02 -1.06906369e-01
-1.56438425e-02 -5.83668113e-01 8.27865973e-02 5.84092379e-01
5.14777541e-01 -2.42457464e-01 -1.03361416e+00 5.12927055e-01
-2.18893170e+00 -6.52224600e-01 -5.13202138e-02 1.74975419e+00
8.78198504e-01 4.20538187e-01 -1.13679007e-01 2.03159556e-01
4.19509917e-01 5.22170782e-01 -2.81183064e-01 -1.76998302e-01
1.93837002e-01 4.68066595e-02 8.39888379e-02 6.65567398e-01
-1.40075934e+00 1.54652321e+00 6.73988295e+00 1.05454195e+00
-5.37223995e-01 6.15687929e-02 1.16328228e+00 7.05953896e-01
-5.29012382e-01 1.81398407e-01 -1.14809668e+00 1.53253645e-01
6.36561215e-01 1.77389346e-02 -1.72048360e-01 1.07958376e+00
-2.66668975e-01 -6.60863578e-01 -8.45342457e-01 1.02813458e+00
-5.54732271e-02 -1.33970666e+00 8.77944008e-02 -2.44807154e-01
4.92258012e-01 -3.76265466e-01 -1.04917556e-01 1.00812542e+00
1.86564252e-01 -8.04761291e-01 5.80536902e-01 9.92974937e-02
6.19984329e-01 -3.02009702e-01 8.68782282e-01 5.17291129e-01
-1.48586059e+00 -2.62652617e-02 -1.00732133e-01 -5.06012082e-01
6.15437090e-01 7.75403857e-01 -7.20234215e-01 6.66978776e-01
4.11523998e-01 6.61438465e-01 -3.39212805e-01 5.30419171e-01
-2.61368573e-01 5.71881115e-01 -2.83676445e-01 -5.42046055e-02
4.09571052e-01 5.69933839e-02 1.61631048e-01 1.07416368e+00
-5.97482212e-02 3.08714390e-01 4.90883172e-01 7.43780851e-01
2.40945593e-01 -4.07979824e-02 -2.35478327e-01 -3.62085283e-01
3.48758221e-01 8.87623310e-01 -1.03973186e+00 -6.20645285e-01
-3.75466406e-01 1.01499844e+00 1.61155730e-01 3.44215035e-02
-1.04630625e+00 -9.47927311e-02 9.19409633e-01 -3.45927387e-01
-1.76293533e-02 -2.35591784e-01 -1.00780904e+00 -1.25258207e+00
4.00743335e-02 -4.76766139e-01 7.28665054e-01 -7.13151515e-01
-8.05407703e-01 4.05473590e-01 1.07019328e-01 -8.07515740e-01
-2.18368188e-01 -6.43212736e-01 -5.13993502e-01 6.20528281e-01
-1.59117949e+00 -1.19835055e+00 -3.22846830e-01 1.89546943e-01
1.16540062e+00 4.87107644e-03 1.07889009e+00 3.25974733e-01
-3.28089625e-01 3.57506037e-01 -5.54743826e-01 7.56715238e-02
1.96994320e-01 -1.09300244e+00 8.27274203e-01 5.44292212e-01
1.88367590e-01 4.15282518e-01 6.76203132e-01 -8.05431306e-01
-1.09517682e+00 -7.46922314e-01 1.53917348e+00 -4.03683931e-01
3.26202989e-01 -3.29362839e-01 -6.51375175e-01 7.76894689e-01
-1.10563161e-02 -4.72664475e-01 8.41546535e-01 6.26011133e-01
-4.68209654e-01 4.87110347e-01 -1.02840471e+00 6.33761406e-01
1.17150581e+00 -8.39165807e-01 -1.13800299e+00 4.66204047e-01
1.24754322e+00 -4.64932859e-01 -3.08246493e-01 6.40037060e-01
4.54078436e-01 -1.12089384e+00 1.04164898e+00 -4.23521936e-01
5.54872394e-01 -1.11643866e-01 -4.34832931e-01 -6.78767800e-01
-1.69842124e-01 -1.18858509e-01 -4.86990530e-03 1.05339098e+00
5.08372188e-01 -5.75678051e-01 1.02541935e+00 8.04694057e-01
-4.00616586e-01 -7.89617836e-01 -9.40114439e-01 -3.89423162e-01
-2.97618061e-01 -8.87903512e-01 3.00662309e-01 6.17705286e-01
3.38479102e-01 8.05522144e-01 -1.85217261e-01 -2.20422938e-01
1.51597604e-01 3.82263511e-01 6.33438945e-01 -1.17427027e+00
-1.43372118e-01 -2.70234734e-01 -2.03599870e-01 -1.72234130e+00
3.18333000e-01 -6.04718328e-01 1.84257045e-01 -1.97907722e+00
3.78199995e-01 -7.37908006e-01 -3.94189581e-02 9.59854305e-01
-5.33951223e-01 3.30724567e-02 -4.73112380e-03 2.11745039e-01
-1.17587161e+00 7.92589128e-01 8.46604884e-01 -5.98685667e-02
-3.15536052e-01 9.42609552e-03 -1.04317069e+00 6.52049661e-01
7.09231019e-01 -1.87013254e-01 -6.29838943e-01 -3.64871740e-01
-9.67199430e-02 2.30354086e-01 -8.42501819e-02 -1.07847536e+00
4.75889146e-02 -5.61580807e-02 -1.20321527e-01 -8.46345186e-01
6.43420219e-01 -5.92115402e-01 -1.98433295e-01 5.35661995e-01
-4.30016786e-01 -2.95093328e-01 3.15001428e-01 4.07519877e-01
-5.55303037e-01 -2.36331105e-01 4.06390041e-01 -1.83429405e-01
-1.27649665e+00 -1.34865912e-02 -3.09948057e-01 2.08868504e-01
9.16436493e-01 -2.02610940e-01 -4.64862913e-01 -4.39594567e-01
-4.36490417e-01 2.79255718e-01 -1.95533354e-02 7.05548465e-01
3.31470251e-01 -1.18650830e+00 -2.10168064e-01 3.47662747e-01
3.72148871e-01 2.77981032e-02 5.48362970e-01 5.05076051e-01
-3.59649658e-01 1.14409697e+00 3.00811566e-02 -7.68243074e-01
-1.08433270e+00 6.26902133e-02 -4.55207787e-02 -1.09092486e+00
-5.19182980e-01 1.06953144e+00 6.25011563e-01 -9.31524515e-01
2.45715275e-01 -3.55876386e-01 -2.08362326e-01 7.69589394e-02
5.59202254e-01 3.55820321e-02 -3.56732644e-02 -5.13698936e-01
-4.35511738e-01 3.34407926e-01 -3.00539672e-01 -1.56126782e-01
8.50914896e-01 -2.99211800e-01 -9.03174430e-02 4.38409626e-01
1.03303194e+00 -7.51299679e-01 -7.68299222e-01 -4.20447886e-01
1.71644107e-01 -3.68848979e-01 8.76961276e-02 -9.52565372e-01
-3.79438519e-01 6.68808937e-01 3.44661713e-01 3.51556838e-01
1.01263130e+00 1.85697362e-01 1.20329678e+00 2.86401123e-01
7.16112673e-01 -1.05320263e+00 2.97923833e-01 1.06430364e+00
4.20298904e-01 -1.39488256e+00 -1.51978984e-01 -7.37798512e-01
-8.06995451e-01 7.59400249e-01 7.78358221e-01 3.95925760e-01
5.07989585e-01 -4.24840860e-03 1.17552727e-01 -1.49306208e-01
-7.61180997e-01 -1.85597807e-01 3.41219962e-01 2.58778304e-01
8.69180381e-01 3.09883296e-01 -6.35245502e-01 7.59574294e-01
-2.95649856e-01 1.20461226e-01 -1.30918056e-01 1.20256376e+00
-7.51677394e-01 -1.23133898e+00 -1.04694158e-01 4.35813487e-01
-2.69986928e-01 -2.69971251e-01 7.82288164e-02 5.05827367e-01
-3.22394192e-01 1.06146777e+00 2.19380394e-01 -4.24720585e-01
1.22748293e-01 6.52172983e-01 -3.31815705e-02 -7.92050362e-01
-2.63378710e-01 -3.06028605e-01 4.15273607e-01 -5.72689116e-01
-2.77908206e-01 -3.66999626e-01 -1.55072188e+00 -1.66793838e-02
-3.39010745e-01 3.60545397e-01 6.18986249e-01 1.53329396e+00
5.27670205e-01 6.52110994e-01 2.12379113e-01 -5.21849275e-01
-1.74362123e-01 -8.87514055e-01 -4.18557733e-01 3.55338454e-01
7.78699061e-03 -6.60383940e-01 -1.02174431e-01 -8.38465393e-02] | [12.511347770690918, 7.347854137420654] |
2c7d54e4-954c-4a7d-b4d9-9996f3a9ee50 | bach-style-music-authoring-system-based-on | 2110.02640 | null | https://arxiv.org/abs/2110.02640v1 | https://arxiv.org/pdf/2110.02640v1.pdf | Bach Style Music Authoring System based on Deep Learning | With the continuous improvement in various aspects in the field of artificial intelligence, the momentum of artificial intelligence with deep learning capabilities into the field of music is coming. The research purpose of this paper is to design a Bach style music authoring system based on deep learning. We use a LSTM neural network to train serialized and standardized music feature data. By repeated experiments, we find the optimal LSTM model which can generate imitation of Bach music. Finally the generated music is comprehensively evaluated in the form of online audition and Turing test. The repertoires which the music generation system constructed in this article are very close to the style of Bach's original music, and it is relatively difficult for ordinary people to distinguish the musics Bach authored and AI created. | ['Lican Huang', 'Minghe Kong'] | 2021-10-06 | null | null | null | null | ['music-generation', 'music-generation'] | ['audio', 'music'] | [-9.39051658e-02 -1.60696179e-01 1.78421691e-01 6.26230091e-02
-1.07563168e-01 -5.34346521e-01 3.13234270e-01 -7.20602870e-01
-3.19783092e-01 5.06596446e-01 2.08545756e-02 6.27799556e-02
-3.29516232e-01 -8.34676981e-01 -4.35386509e-01 -5.99794745e-01
-5.74170761e-02 5.91502488e-01 -4.60067213e-01 -3.57057512e-01
4.39078182e-01 2.98340887e-01 -1.52977979e+00 4.67368543e-01
5.18513978e-01 1.02972174e+00 3.85970861e-01 8.52356553e-01
-3.41297209e-01 8.41634393e-01 -1.08830428e+00 -4.04573232e-01
4.28228408e-01 -8.23436618e-01 -9.73557532e-01 -2.52065837e-01
1.37145922e-01 -3.59334439e-01 -5.38395941e-01 1.09734631e+00
8.60791206e-01 1.15797065e-01 4.69777107e-01 -1.01933455e+00
-9.46022928e-01 1.29762387e+00 7.14266747e-02 -3.00066650e-01
3.20644528e-01 9.18853506e-02 1.13297379e+00 -4.23669189e-01
2.81081498e-01 1.04635394e+00 6.80380642e-01 7.94685960e-01
-7.31166959e-01 -1.05788434e+00 -5.04861355e-01 4.88965958e-01
-1.46934545e+00 -6.83183894e-02 1.03643084e+00 -3.01578641e-01
5.53955793e-01 5.93270622e-02 1.28326869e+00 1.23877406e+00
2.29707986e-01 1.08962607e+00 1.02023077e+00 -4.59999979e-01
1.43421620e-01 -9.74042192e-02 -5.26552498e-01 4.30786639e-01
-1.05867445e-01 3.73396784e-01 -7.86610007e-01 1.75017878e-01
1.29645956e+00 -3.47031206e-01 -6.15618937e-02 2.04861253e-01
-1.45192289e+00 5.78473508e-01 4.43950593e-01 7.71457791e-01
-3.67957711e-01 2.93816864e-01 4.48874474e-01 7.67051160e-01
-2.91969568e-01 1.21191669e+00 -4.67144519e-01 -5.67110837e-01
-1.11495543e+00 2.58954883e-01 1.02302063e+00 8.60962808e-01
6.57108575e-02 8.55874419e-01 -3.38897258e-02 8.66554558e-01
8.57065096e-02 3.32224667e-01 1.29886603e+00 -1.35201550e+00
-1.73384354e-01 6.29748404e-01 -9.11526904e-02 -7.65028656e-01
-3.86605002e-02 -9.43031013e-01 -1.10085201e+00 3.88687223e-01
3.25136095e-01 -1.66032508e-01 -5.49702942e-01 1.62128067e+00
-2.09302679e-01 9.44411457e-02 2.03929484e-01 9.01060998e-01
1.04660404e+00 6.72453761e-01 -4.47874188e-01 -5.75239733e-02
1.07297564e+00 -9.45607185e-01 -7.73696184e-01 3.08895200e-01
8.94023925e-02 -9.06585872e-01 1.30511141e+00 1.05668807e+00
-1.08916879e+00 -1.03246057e+00 -1.29263783e+00 3.94572139e-01
-8.37490559e-02 1.11083128e-01 7.78463960e-01 4.66637552e-01
-7.68052995e-01 1.19999421e+00 -1.44306585e-01 -6.88711256e-02
1.16160281e-01 5.23263514e-01 2.13851146e-02 7.33416498e-01
-1.38397658e+00 5.42502999e-01 9.58396792e-01 1.37378290e-01
-1.06519425e+00 -1.06551208e-01 -1.80136055e-01 1.15674600e-01
-1.96403980e-01 -9.29157972e-01 1.94103408e+00 -1.58046031e+00
-2.13403201e+00 7.33758807e-01 4.45209146e-01 -4.37232763e-01
3.39440495e-01 -1.55363545e-01 -5.51560998e-01 -1.31181598e-01
-1.53013632e-01 6.84029222e-01 9.37703192e-01 -1.03908837e+00
-5.84274352e-01 -1.34733170e-01 -1.57882944e-01 2.18227386e-01
-5.40156484e-01 -5.53786494e-02 -1.20105743e-01 -1.02787066e+00
1.02357508e-03 -8.90029788e-01 1.38371289e-01 -5.35247743e-01
-3.91028672e-01 -4.66004372e-01 3.60097706e-01 -4.38324511e-01
1.26960981e+00 -2.07589579e+00 2.70261258e-01 1.24510936e-01
-5.50997108e-02 3.70540529e-01 -3.84012699e-01 4.19809222e-01
8.42606276e-02 4.75811027e-02 1.56773835e-01 1.29935145e-01
2.27753967e-01 2.96944350e-01 -6.08815253e-01 -2.53032804e-01
-4.26301301e-01 9.17850435e-01 -7.77437329e-01 -4.27398324e-01
-3.29370618e-01 9.80518237e-02 -3.88402462e-01 5.46081364e-01
-3.55326742e-01 6.30329430e-01 -3.56472373e-01 6.91232860e-01
4.47313786e-02 1.66411176e-02 8.11745450e-02 1.12306930e-01
-4.59075347e-02 2.74407387e-01 -1.20178843e+00 2.03287983e+00
-2.89183617e-01 6.90647542e-01 -2.70586163e-01 -6.75374508e-01
1.19203866e+00 7.77086258e-01 3.83625209e-01 -5.27392089e-01
4.85119373e-01 3.73871356e-01 5.38567185e-01 -5.67007959e-01
4.57354158e-01 -4.27822649e-01 -6.32070452e-02 7.38831699e-01
1.00786485e-01 -4.51909840e-01 3.60022578e-03 -2.82975286e-01
7.84242988e-01 4.42275792e-01 8.55993405e-02 4.44929823e-02
3.58506590e-01 -9.88329947e-02 6.13178372e-01 7.29643404e-01
1.35158122e-01 1.87811464e-01 1.77708805e-01 -7.09003389e-01
-1.14463985e+00 -1.07078409e+00 1.66464970e-01 1.15154183e+00
-1.71854392e-01 -2.44638026e-01 -8.97388697e-01 3.97363156e-02
-2.36341208e-01 6.57616735e-01 -2.37817198e-01 -2.61153966e-01
-6.58738792e-01 -8.71403292e-02 1.28028810e+00 3.57267886e-01
9.21888053e-01 -2.10030699e+00 -5.69702864e-01 5.83334744e-01
-1.23811707e-01 -4.78480011e-01 -4.15597588e-01 -3.24318334e-02
-1.05005503e+00 -7.84359097e-01 -8.90709817e-01 -1.30270588e+00
-1.96555108e-01 -3.70739520e-01 1.10503614e+00 -1.54343769e-01
-1.31651118e-01 -2.65009813e-02 -2.97722220e-01 -8.82863700e-01
-6.92921698e-01 3.26163560e-01 4.64603782e-01 -7.30324313e-02
2.87145823e-01 -9.72284198e-01 -5.41746557e-01 1.62662461e-01
-8.89446676e-01 1.72456786e-01 8.46088171e-01 7.84755468e-01
3.47233981e-01 3.12863886e-01 7.39274979e-01 -1.71493948e-01
1.10393643e+00 8.86866301e-02 -2.81712592e-01 -7.09370673e-02
-6.81167424e-01 2.03264356e-01 8.42415929e-01 -8.73445213e-01
-7.46678948e-01 -5.64134642e-02 -2.76868809e-02 -6.13022506e-01
-2.31584549e-01 4.95926887e-01 2.17540953e-02 4.83458687e-04
5.43089330e-01 6.43224120e-01 -1.59020931e-01 -8.37692738e-01
1.20107673e-01 1.07224822e+00 9.48199987e-01 -6.59679830e-01
6.66866243e-01 -1.74228206e-01 -2.14271769e-01 -6.37534499e-01
-7.25386500e-01 1.51595995e-01 -4.75965172e-01 -5.51842272e-01
6.82561278e-01 -5.03423095e-01 -1.30526352e+00 8.55510354e-01
-1.33992302e+00 -4.74627793e-01 -5.85432231e-01 6.95498765e-01
-9.04444814e-01 2.62252182e-01 -9.14204001e-01 -7.47815788e-01
-8.02818656e-01 -7.83474982e-01 6.85150146e-01 3.58575642e-01
-5.08634984e-01 -4.37946707e-01 4.40056741e-01 1.85430080e-01
2.19227821e-01 -1.64285541e-01 1.03885794e+00 -4.47019756e-01
-5.97700119e-01 -2.43392229e-01 3.27290207e-01 7.04952598e-01
-1.28286347e-01 -2.00795252e-02 -9.86449122e-01 -3.65829393e-02
2.22048938e-01 -4.78226006e-01 5.81525445e-01 2.58944720e-01
1.19402051e+00 -4.45576757e-01 4.62354779e-01 9.08149004e-01
9.76483643e-01 7.82100022e-01 7.63032675e-01 6.55249774e-01
3.95479053e-01 2.71606803e-01 3.42181295e-01 4.19050932e-01
-1.11734115e-01 6.00255072e-01 1.97564751e-01 3.40180695e-01
-1.17476895e-01 -7.91587532e-01 5.90500414e-01 1.46201193e+00
-4.05403972e-01 1.86964869e-01 -5.95647335e-01 2.97619224e-01
-1.58159339e+00 -1.32149315e+00 2.48414367e-01 2.09874606e+00
1.05017424e+00 -5.15101813e-02 2.68726051e-01 6.35050416e-01
7.17905581e-01 -2.21854970e-01 -6.51073813e-01 -7.54401445e-01
-1.34881631e-01 6.36512280e-01 -1.67424902e-01 -4.81790490e-02
-8.54654312e-01 1.21939266e+00 7.61888695e+00 9.07366157e-01
-1.29994690e+00 -2.51837850e-01 2.82837637e-02 -3.17026824e-01
1.11700678e-02 -1.87001631e-01 -3.43057692e-01 3.81619394e-01
7.81591892e-01 -3.11052114e-01 1.10361969e+00 8.02283168e-01
1.50990009e-01 5.39144754e-01 -1.13730478e+00 1.46113610e+00
-6.45464063e-02 -1.04303157e+00 3.81335527e-01 -3.68301794e-02
7.34438241e-01 -6.22935742e-02 2.67788917e-01 5.52180111e-01
2.45170563e-01 -1.32424414e+00 9.47670102e-01 7.05065966e-01
7.60287881e-01 -7.32519567e-01 6.24259651e-01 5.34373224e-01
-9.95298684e-01 -3.27918261e-01 -3.43144745e-01 -6.31854355e-01
-1.11086227e-01 1.42052561e-01 -7.74147570e-01 2.30477586e-01
6.47573769e-01 5.32723546e-01 -2.83083290e-01 1.35666656e+00
-3.99344265e-01 6.81621909e-01 1.16832510e-01 -5.11631966e-01
2.64564931e-01 -2.94912279e-01 6.37297809e-01 6.89109206e-01
6.93423092e-01 -1.58463731e-01 -1.14284240e-01 1.04095912e+00
-2.50749707e-01 3.01240891e-01 -5.67268431e-01 -6.13517165e-01
1.79553688e-01 1.07004416e+00 -3.12622666e-01 -4.88921523e-01
4.03738499e-01 9.58557844e-01 -1.27752274e-01 2.51557171e-01
-5.53743064e-01 -6.76693916e-01 2.64122933e-01 -2.96208441e-01
7.89083913e-02 -2.68767864e-01 -5.03053367e-01 -1.06776536e+00
-2.80925900e-01 -1.32280564e+00 8.90320390e-02 -1.18878746e+00
-1.27586889e+00 8.60514462e-01 -4.77907836e-01 -1.42022681e+00
-6.03534818e-01 -6.25336945e-01 -9.72304702e-01 5.67552269e-01
-6.12640560e-01 -8.33257020e-01 -2.41334364e-01 5.38572729e-01
5.62697351e-01 -1.03992760e+00 1.21566999e+00 1.53897837e-01
-2.54236877e-01 6.67530477e-01 1.05764039e-01 4.18513507e-01
5.75833440e-01 -1.22094750e+00 1.53626874e-01 2.80321389e-02
7.82218456e-01 3.64748001e-01 6.96864605e-01 -3.12828273e-01
-1.16295075e+00 -6.10893548e-01 8.50159466e-01 1.09834611e-01
5.47722280e-01 2.26953514e-02 -4.05263394e-01 3.54406834e-01
5.54000437e-01 -1.15202618e+00 7.48145640e-01 1.10510411e-02
-2.56523967e-01 -3.56646240e-01 -7.97611713e-01 7.74000168e-01
1.03757012e+00 -4.02555913e-01 -1.05745840e+00 1.49849430e-01
5.97227693e-01 4.93175723e-02 -9.94959652e-01 4.29328114e-01
1.13020289e+00 -9.24658239e-01 5.59190869e-01 -7.54444838e-01
4.46450919e-01 -2.76837498e-01 4.56623584e-02 -1.25536132e+00
-4.70289052e-01 -8.94832730e-01 1.07704541e-02 1.01274133e+00
9.81272534e-02 -3.28182787e-01 8.68583381e-01 -1.98268771e-01
-1.58106104e-01 -4.09703553e-01 -8.59028876e-01 -1.01439726e+00
4.16865349e-02 -3.56410891e-01 8.03745389e-01 9.57298696e-01
1.30175129e-01 6.79125428e-01 -4.97398019e-01 -4.81595635e-01
4.83071566e-01 5.00231981e-01 1.02076638e+00 -1.49773538e+00
-9.14358377e-01 -9.35688853e-01 -4.30684924e-01 -8.21786642e-01
1.33521616e-01 -1.15094316e+00 -1.17110588e-01 -1.34135401e+00
7.11664781e-02 3.09434216e-02 -7.02073455e-01 4.66150105e-01
4.22002017e-01 2.60143429e-01 3.63471627e-01 3.93541157e-01
-2.73694068e-01 6.65868878e-01 1.67929614e+00 -3.79293263e-01
-4.54583466e-01 3.14299673e-01 -6.45763457e-01 9.41780865e-01
1.19893384e+00 -4.13282782e-01 -1.59591451e-01 -4.03588295e-01
5.38930714e-01 1.71402842e-01 3.65272760e-02 -1.41394079e+00
1.13713399e-01 7.46389106e-02 6.15813851e-01 -4.42505330e-01
3.42273355e-01 -5.57589650e-01 3.20411831e-01 7.72692382e-01
-6.68852568e-01 1.10046834e-01 8.27340856e-02 8.91675726e-02
-4.35436189e-01 -5.06988227e-01 7.35423386e-01 -5.38083315e-01
-5.93417585e-01 1.55026436e-01 -5.01293540e-01 -1.49909675e-01
5.47545612e-01 -2.86731154e-01 3.74022156e-01 -7.69819379e-01
-8.65769446e-01 -3.45829993e-01 2.03185990e-01 5.22376716e-01
5.94876468e-01 -1.70693696e+00 -6.34644270e-01 1.88048422e-01
-3.06927383e-01 -4.70001161e-01 7.06971437e-02 2.34996453e-01
-6.96485221e-01 3.71189922e-01 -6.31442606e-01 -2.04743475e-01
-1.22521102e+00 3.57859880e-01 4.65507060e-01 -3.84568945e-02
-5.23531914e-01 6.17662728e-01 -1.75164431e-01 -4.69062269e-01
6.30892396e-01 -7.69430101e-02 -2.15547785e-01 -2.60522068e-01
3.71705353e-01 3.30979437e-01 -5.00195086e-01 -1.97773710e-01
2.58240134e-01 4.69437242e-01 3.65385473e-01 -4.24043030e-01
1.16075337e+00 6.10464215e-01 -3.75082314e-01 8.95486832e-01
6.70113206e-01 -1.68652371e-01 -4.72885370e-01 8.04372281e-02
1.85070408e-03 -1.96961731e-01 -1.20344363e-01 -1.04606187e+00
-8.78099918e-01 9.52983916e-01 7.04947412e-01 3.35997045e-01
1.17623663e+00 -2.99299717e-01 1.26418626e+00 9.45863485e-01
4.19776708e-01 -1.39969432e+00 3.95585358e-01 7.36164212e-01
1.11096013e+00 -6.28571570e-01 -4.77177262e-01 8.09445083e-01
-6.31857753e-01 1.22864056e+00 5.80378950e-01 -4.00856018e-01
3.15754443e-01 7.50149712e-02 2.01798782e-01 -5.40838689e-02
-6.78913057e-01 -3.71428509e-03 3.71967673e-01 3.16818058e-01
6.86304808e-01 1.59842432e-01 -4.37721461e-01 1.01038182e+00
-1.22652864e+00 5.15043259e-01 4.65533167e-01 3.28260690e-01
-6.93394125e-01 -1.25324595e+00 -4.51960027e-01 2.34792039e-01
-5.11100650e-01 -1.11670680e-01 -9.38134551e-01 4.99624580e-01
3.39317322e-01 7.22326696e-01 3.62008344e-03 -9.20710087e-01
1.82318732e-01 2.70547003e-01 7.37766504e-01 -3.03076714e-01
-8.04802001e-01 1.47016272e-01 -2.30005234e-01 -1.80135384e-01
-2.16235667e-01 -6.84494525e-02 -1.19446540e+00 -5.11022747e-01
-1.33498227e-02 4.43471342e-01 7.89939404e-01 7.45191395e-01
1.06503800e-01 6.07581615e-01 6.61434889e-01 -9.68357325e-01
-8.45309973e-01 -1.26794910e+00 -1.12136805e+00 3.02387893e-01
-2.66602069e-01 -1.82263345e-01 -1.70991033e-01 -4.46844436e-02] | [16.033348083496094, 5.479020595550537] |
3aa03601-cfa4-421c-aa09-082faf133bb5 | using-objective-words-in-the-reviews-to | 1709.08521 | null | http://arxiv.org/abs/1709.08521v1 | http://arxiv.org/pdf/1709.08521v1.pdf | Using objective words in the reviews to improve the colloquial arabic sentiment analysis | One of the main difficulties in sentiment analysis of the Arabic language is
the presence of the colloquialism. In this paper, we examine the effect of
using objective words in conjunction with sentimental words on sentiment
classification for the colloquial Arabic reviews, specifically Jordanian
colloquial reviews. The reviews often include both sentimental and objective
words, however, the most existing sentiment analysis models ignore the
objective words as they are considered useless. In this work, we created two
lexicons: the first includes the colloquial sentimental words and compound
phrases, while the other contains the objective words associated with values of
sentiment tendency based on a particular estimation method. We used these
lexicons to extract sentiment features that would be training input to the
Support Vector Machines (SVM) to classify the sentiment polarity of the
reviews. The reviews dataset have been collected manually from JEERAN website.
The results of the experiments show that the proposed approach improves the
polarity classification in comparison to two baseline models, with accuracy
95.6%. | ['Omar Al-Harbi'] | 2017-09-25 | null | null | null | null | ['arabic-sentiment-analysis'] | ['natural-language-processing'] | [-2.77881980e-01 -2.35272534e-02 -8.63696709e-02 -5.97442389e-01
-1.51240364e-01 -7.10541785e-01 6.29101217e-01 4.54193443e-01
-3.91754001e-01 1.00068712e+00 3.97749752e-01 -2.03107193e-01
2.66824037e-01 -6.88525915e-01 -1.30668417e-01 -7.61350393e-01
3.83718818e-01 3.29586059e-01 -1.52615145e-01 -9.06337082e-01
1.03279793e+00 2.48411726e-02 -1.54638612e+00 4.07315820e-01
9.21613872e-01 9.30055201e-01 1.56606436e-01 5.78605115e-01
-4.47592586e-01 9.72158909e-01 -8.93361926e-01 -4.52931374e-01
-3.34798172e-02 -3.33661050e-01 -6.33593500e-01 1.87841445e-01
-4.01146263e-02 2.34557495e-01 6.52469754e-01 1.02456474e+00
2.49976590e-01 -1.40276313e-01 1.05144906e+00 -1.05007863e+00
-8.15141499e-01 4.95173186e-01 -7.43602514e-01 2.00196460e-01
3.25398386e-01 -5.97364128e-01 1.19161153e+00 -1.18743026e+00
4.71263379e-01 1.03045022e+00 5.05884826e-01 1.19306603e-02
-2.65577137e-01 -4.83882576e-01 1.95742413e-01 4.67851944e-02
-8.97498250e-01 -8.75732005e-02 8.42757523e-01 -6.32212520e-01
8.83591413e-01 7.40027279e-02 8.11344504e-01 2.79258013e-01
7.61802435e-01 5.39955020e-01 1.47569931e+00 -8.82268786e-01
1.31010875e-01 1.11577475e+00 7.64870644e-01 3.77234370e-01
4.46742356e-01 -5.93906939e-01 -3.18888396e-01 -5.23210652e-02
-3.08688462e-01 -2.95164526e-01 2.69255172e-02 5.05639911e-02
-6.17095232e-01 1.23775089e+00 -6.72738180e-02 5.03983915e-01
-4.30923879e-01 -4.87574816e-01 7.73025930e-01 3.56100291e-01
7.45051563e-01 4.98556197e-01 -9.05438781e-01 -5.44980578e-02
-5.16765177e-01 2.13557720e-01 8.95077467e-01 5.83907723e-01
7.28221714e-01 7.95475617e-02 4.60056305e-01 1.04065740e+00
7.57273972e-01 8.72636378e-01 9.15552080e-01 7.24091604e-02
3.55429083e-01 1.09748125e+00 3.40064406e-01 -1.50100100e+00
-4.65853572e-01 -1.54061660e-01 -1.42098442e-01 5.02076708e-02
1.49236142e-03 -7.54922509e-01 -8.52551579e-01 1.04789042e+00
2.32884645e-01 -9.43492234e-01 7.76035190e-01 8.15315664e-01
1.05988789e+00 1.05657828e+00 1.07520722e-01 -2.73844421e-01
1.51823294e+00 -9.63684499e-01 -1.13214850e+00 -3.11763585e-01
6.86433792e-01 -1.42688966e+00 1.10986459e+00 7.25052714e-01
-7.28373230e-01 -3.15247655e-01 -1.36682260e+00 3.20408434e-01
-1.05308676e+00 5.93019366e-01 7.80939043e-01 1.11772335e+00
-6.96810246e-01 -2.83805560e-02 -2.58710951e-01 -3.67882282e-01
-1.91017583e-01 4.07199681e-01 -3.35070193e-01 5.76521993e-01
-1.30831051e+00 1.39312053e+00 2.61417478e-01 2.60557592e-01
-5.55790551e-02 1.50169834e-01 -8.71981859e-01 -4.29056823e-01
-6.53420761e-02 4.44704592e-01 1.10313833e+00 -1.84877920e+00
-1.25243783e+00 8.56835067e-01 -2.15985343e-01 7.72145987e-02
-1.68550253e-01 -1.70389414e-01 -7.11595476e-01 6.18479699e-02
3.71773124e-01 1.33492038e-01 4.86471415e-01 -1.29945087e+00
-9.21815872e-01 -3.70925307e-01 1.05285734e-01 5.16056001e-01
-6.18990779e-01 3.76996875e-01 -1.43732682e-01 -5.25231302e-01
4.72071841e-02 -1.01999009e+00 -2.39557717e-02 -9.92272079e-01
-1.97935849e-01 -3.16646427e-01 9.93642569e-01 -6.26173913e-01
1.23310542e+00 -1.80825806e+00 -1.74790278e-01 5.32420635e-01
-5.15127122e-01 2.48310789e-01 3.48240376e-01 5.67604482e-01
-2.01111704e-01 4.94213179e-02 -7.08685368e-02 5.11095412e-02
-2.55296946e-01 8.29040185e-02 -3.50859433e-01 2.90153861e-01
2.61724621e-01 3.74016702e-01 -7.07530916e-01 -5.41623175e-01
-1.38977796e-01 2.57809490e-01 -8.40253755e-02 2.61426289e-02
-6.00300804e-02 -3.39626931e-02 -6.02331161e-01 8.80370975e-01
5.88473201e-01 1.09824039e-01 4.25085932e-01 -3.30627412e-01
-2.54857033e-01 5.53525925e-01 -1.15861452e+00 7.07630813e-01
-4.66467738e-01 5.59656680e-01 -3.53646964e-01 -8.73938620e-01
1.23334944e+00 4.43014115e-01 1.93914175e-01 -4.37809318e-01
7.45760262e-01 5.35097361e-01 4.92221266e-02 -6.81590080e-01
1.21104479e+00 -4.21867877e-01 -2.41545931e-01 5.59855342e-01
-1.45512715e-01 -5.11148334e-01 7.11084485e-01 1.52410477e-01
-2.05277670e-02 3.11756250e-03 5.71047664e-01 -5.67955434e-01
1.09735358e+00 5.16176462e-01 2.46239260e-01 -1.85129315e-01
-9.87410769e-02 3.54893059e-01 9.53002274e-01 -3.04715395e-01
-7.27380037e-01 -3.15789640e-01 -2.70056367e-01 9.48064566e-01
6.67250436e-03 -3.86652410e-01 -6.55928314e-01 -9.19241309e-01
-3.84048820e-01 5.34459949e-01 -6.62786186e-01 2.84054101e-01
-2.58469760e-01 -1.43901873e+00 -1.37911528e-01 2.67999530e-01
6.22366108e-02 -1.18669724e+00 -3.81265223e-01 3.50766592e-02
-1.00860566e-01 -7.65364707e-01 -3.60047668e-02 4.17720228e-01
-5.94780564e-01 -1.07446289e+00 -4.30096686e-01 -1.10039389e+00
1.02551258e+00 2.00564042e-01 8.82118583e-01 1.53496906e-01
3.17658514e-01 9.02022198e-02 -1.07020915e+00 -1.13966918e+00
-4.77187812e-01 2.21711606e-01 -7.95035902e-03 -5.14339507e-02
1.00880086e+00 1.17588677e-01 -2.08044067e-01 8.74598324e-02
-7.02453077e-01 -3.72580707e-01 4.82540131e-01 6.99241698e-01
1.59025580e-01 2.53352761e-01 1.29939437e+00 -1.47224665e+00
1.11128032e+00 -6.63060546e-01 -3.67878079e-01 -7.89961964e-02
-8.13966751e-01 -2.34262407e-01 7.53850400e-01 -1.51293188e-01
-1.09675789e+00 -2.49602869e-01 -3.20085853e-01 8.84168983e-01
2.75134951e-01 1.16327989e+00 6.00839779e-02 1.76422559e-02
6.09965622e-01 8.82940367e-02 2.76808385e-02 1.06583677e-01
-1.12428434e-01 1.30209672e+00 -3.60647172e-01 -3.41686219e-01
2.21039489e-01 6.66944981e-02 -4.10149366e-01 -8.46168578e-01
-1.06968904e+00 -6.39583945e-01 -4.94809985e-01 -4.38892990e-01
7.24176943e-01 -8.91801417e-01 -2.46844575e-01 6.76424980e-01
-8.78399611e-01 3.79023761e-01 1.86740056e-01 4.96232033e-01
-1.50427580e-01 3.33378583e-01 -5.22580922e-01 -1.07229102e+00
-6.95441306e-01 -1.32057488e+00 5.16636908e-01 4.74743962e-01
-6.30615950e-01 -1.17496645e+00 2.19289139e-01 3.60050231e-01
1.20417736e-01 1.20590664e-01 9.91515100e-01 -8.81161749e-01
5.82751393e-01 -5.26030838e-01 1.06427610e-01 7.93899953e-01
4.82718319e-01 3.94063413e-01 -8.02400172e-01 1.19310886e-01
2.65394181e-01 -5.41823208e-01 2.87965685e-01 1.70747772e-01
3.01602185e-01 -2.92826802e-01 1.82351276e-01 -3.29562426e-01
1.70085561e+00 6.80473924e-01 6.15437150e-01 6.68392420e-01
3.36913347e-01 8.73185217e-01 1.30765271e+00 3.46416265e-01
6.35865748e-01 -1.95940025e-02 2.59597361e-01 -4.26068343e-03
5.31200349e-01 2.01783985e-01 6.90775573e-01 1.56322765e+00
5.23395166e-02 -2.43956149e-01 -8.92793894e-01 9.70705986e-01
-1.45572162e+00 -4.39178526e-01 -7.24629402e-01 1.57240415e+00
1.12581170e+00 2.65588701e-01 -5.15722074e-02 5.98713577e-01
5.07534266e-01 1.85961127e-01 1.18372925e-01 -1.26089120e+00
-2.23373652e-01 2.51920432e-01 4.17345524e-01 6.32527888e-01
-1.31292343e+00 9.64577317e-01 5.81724310e+00 4.39521372e-01
-1.49280775e+00 -5.60443092e-04 5.97523868e-01 3.77062589e-01
-3.38416755e-01 -1.36049718e-01 -1.01199675e+00 4.13885474e-01
8.78600061e-01 1.94211584e-02 -3.36708516e-01 9.90341485e-01
2.64182985e-01 -7.34564126e-01 -2.75153935e-01 3.43406975e-01
6.56569123e-01 -8.06399465e-01 1.34093180e-01 -3.33449870e-01
1.14720416e+00 -3.22850049e-01 7.62213022e-02 1.24043554e-01
-4.86924360e-03 -8.23472917e-01 7.24832654e-01 2.14911342e-01
1.91416323e-01 -1.06751621e+00 1.64207172e+00 1.60679929e-02
-7.60400712e-01 1.54942766e-01 -4.04229701e-01 -4.60387886e-01
-3.81525159e-02 5.67734540e-01 -7.99107611e-01 3.59619200e-01
5.83157957e-01 8.61806929e-01 -6.34420156e-01 3.38728964e-01
-4.19376433e-01 7.02115834e-01 -5.12189120e-02 -9.39224362e-01
5.00979364e-01 -5.67024291e-01 6.22990392e-02 1.33546937e+00
1.50458768e-01 -3.57477181e-02 -2.70160943e-01 -1.17557712e-01
2.41638169e-01 1.18792462e+00 -5.48279345e-01 -2.92704582e-01
4.68766391e-02 1.58163989e+00 -1.09777451e+00 -3.76493067e-01
-6.52392864e-01 4.12391603e-01 -1.38496563e-01 9.70700607e-02
-4.82689530e-01 -7.95638621e-01 5.25777228e-02 -6.56171888e-02
1.89132914e-01 1.13850668e-01 -7.52304673e-01 -9.45628703e-01
1.45637631e-01 -1.20653796e+00 1.86299011e-01 -8.00736010e-01
-1.25068831e+00 9.55672622e-01 -2.44527102e-01 -1.15343499e+00
-1.02769539e-01 -1.16299462e+00 -6.22792423e-01 1.08526433e+00
-1.70396864e+00 -1.13195288e+00 2.84804087e-02 2.57339120e-01
5.78802407e-01 -5.68182170e-01 8.68143737e-01 2.71580249e-01
-3.26189280e-01 1.74266115e-01 -5.26308753e-02 6.61459491e-02
8.96327615e-01 -1.38411379e+00 -3.29265088e-01 7.73034692e-01
-5.00462949e-01 7.73272395e-01 1.13053739e+00 -8.11621308e-01
-9.16850865e-01 -6.21773005e-01 1.45771480e+00 -4.43107605e-01
8.46106946e-01 -1.79950651e-02 -4.29730743e-01 5.30357063e-01
5.94259679e-01 -8.31331134e-01 1.18972945e+00 8.00508857e-02
8.80311355e-02 -7.26821274e-02 -1.12129605e+00 4.86910701e-01
-5.45211256e-01 -1.25568256e-01 -9.65256453e-01 4.55857068e-01
2.25214064e-01 4.91975956e-02 -6.17504001e-01 1.07374266e-01
6.51458085e-01 -6.97306097e-01 3.35621208e-01 -3.46352816e-01
9.08214331e-01 -4.62250262e-01 -3.20386142e-01 -1.39975929e+00
4.25500154e-01 1.35409191e-01 4.66971099e-01 1.14733744e+00
1.05522704e+00 -8.21733594e-01 6.66441023e-01 2.30392978e-01
1.04987651e-01 -8.20974648e-01 -1.06192045e-01 1.30753070e-01
2.36324012e-01 -1.79554984e-01 1.89634264e-01 1.05377829e+00
4.72490728e-01 8.45289052e-01 -3.48220080e-01 -1.31776065e-01
-1.72791377e-01 1.45324528e-01 6.18417740e-01 -1.10176373e+00
2.81531304e-01 -1.49721012e-01 -2.60618687e-01 -3.95883501e-01
-1.10873114e-02 -5.03399849e-01 9.28626731e-02 -1.52759349e+00
-2.80103106e-02 -6.07261419e-01 -8.61402880e-03 2.48179615e-01
-3.60497326e-01 3.86881143e-01 -3.47818807e-02 -5.13416417e-02
-1.22290224e-01 2.49812573e-01 1.13195944e+00 6.09286092e-02
-3.16838026e-01 8.72422941e-03 -1.36914074e+00 1.10567367e+00
1.05112755e+00 -5.00958920e-01 -6.02966070e-01 6.99432194e-02
1.07118070e+00 -4.45696920e-01 -7.31208384e-01 -4.54435617e-01
-2.10309118e-01 -3.47931713e-01 2.62399644e-01 -9.50102925e-01
2.45157242e-01 -8.69104624e-01 -4.63066697e-01 2.57038265e-01
-6.37826920e-02 4.93575901e-01 1.51407957e-01 6.32463768e-02
-5.86572170e-01 -8.66549611e-01 7.10791111e-01 -1.63459256e-01
-5.05118370e-01 -4.24030602e-01 -8.51988018e-01 -1.45989478e-01
1.16653955e+00 -2.74693608e-01 -2.56587774e-01 -4.23232287e-01
-3.87303561e-01 1.83169991e-01 3.47098678e-01 3.92288327e-01
4.79438126e-01 -9.47214782e-01 -6.60955727e-01 9.46523342e-03
1.55730233e-01 -5.08324027e-01 -1.75158709e-01 8.35491896e-01
-1.13120866e+00 4.62302119e-01 -4.59482431e-01 -8.18774030e-02
-1.32731318e+00 2.56616294e-01 1.40573889e-01 -1.68862432e-01
3.15083325e-01 5.42485893e-01 -1.58664227e-01 -7.28259027e-01
-2.33104408e-01 -6.71085715e-02 -1.57535386e+00 9.17270541e-01
4.45082963e-01 4.49016765e-02 3.30932885e-01 -1.27210689e+00
-3.43882233e-01 8.07056069e-01 -3.30778778e-01 -2.98909456e-01
1.32993388e+00 -1.87644243e-01 -8.20350111e-01 9.26340520e-01
9.80495214e-01 8.13164592e-01 -1.02437504e-01 3.49276781e-01
1.19819939e-01 -2.30698749e-01 -5.27872518e-02 -1.12397361e+00
-9.02794182e-01 5.70783436e-01 2.97118247e-01 5.01560330e-01
1.13287926e+00 -4.95897472e-01 5.36728263e-01 3.59083146e-01
1.29055604e-01 -1.55723405e+00 -1.73172593e-01 8.70160580e-01
6.36499345e-01 -1.43418038e+00 2.24009782e-01 -5.23261309e-01
-1.24403894e+00 1.38843215e+00 6.12263560e-01 -4.20555145e-01
1.03659236e+00 1.08608283e-01 8.86317194e-01 -3.68509084e-01
-5.78339517e-01 8.80015641e-02 2.10700825e-01 4.24598992e-01
9.89336014e-01 1.25544220e-01 -1.52662253e+00 9.84925508e-01
-5.72268367e-01 -3.02725405e-01 1.28762937e+00 1.14514089e+00
-6.83142662e-01 -1.19000065e+00 -5.31830847e-01 6.36347115e-01
-1.18389094e+00 -3.04862797e-01 -5.73944807e-01 7.39817202e-01
1.69301406e-01 1.42285848e+00 -1.22655101e-01 -3.23855728e-01
2.12112024e-01 -2.31825337e-02 -1.93471938e-01 -8.06608617e-01
-1.01080465e+00 7.74226487e-02 5.69418371e-01 3.52208465e-01
-9.96479869e-01 -6.48426831e-01 -1.35488033e+00 -1.85652941e-01
-7.84746468e-01 8.62976611e-01 1.34520853e+00 1.09415352e+00
-2.62806982e-01 2.36149535e-01 9.06056404e-01 -3.78731847e-01
-1.59084603e-01 -1.33426166e+00 -6.36130929e-01 2.56837845e-01
2.51519978e-01 -5.87801218e-01 -6.02278411e-01 2.45211020e-01] | [11.0400390625, 6.894489765167236] |
a4603d17-0f45-4bf6-9ac7-9f2ab8da1c69 | unsupervised-neural-dependency-parsing | null | null | https://aclanthology.org/D16-1073 | https://aclanthology.org/D16-1073.pdf | Unsupervised Neural Dependency Parsing | null | ['Kewei Tu', 'Yong Jiang', 'Wenjuan Han'] | 2016-11-01 | null | null | null | emnlp-2016-11 | ['dependency-grammar-induction'] | ['natural-language-processing'] | [-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01
-8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01
-2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00
-3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01
-9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444e-01
7.03166842e-01 7.88774848e-01 -1.68942046e+00 7.18545914e-01
7.04417467e-01 8.52217197e-01 2.49141872e-01 1.14950800e+00
-1.95044339e-01 1.55633950e+00 -7.48382092e-01 -5.46825826e-01
3.13719302e-01 -1.23176083e-01 -7.22945035e-01 -1.01074085e-01
9.28529128e-02 -8.59008506e-02 -2.09758401e-01 9.22211111e-01
5.37373662e-01 4.49454933e-02 1.08379531e+00 -1.42548037e+00
-5.91619551e-01 6.10313773e-01 -4.01565880e-02 1.21627934e-01
1.03678203e+00 -5.39447069e-01 1.19919395e+00 -1.13026452e+00
7.20913768e-01 1.26888943e+00 8.66221786e-01 5.44149756e-01
-1.22286928e+00 -1.94712028e-01 -3.26822817e-01 -9.51717794e-02
-1.46558487e+00 -3.25250506e-01 4.25783843e-02 -2.08119690e-01
1.66093647e+00 1.26596653e+00 1.20609856e+00 1.01401424e+00
1.26658809e+00 8.34431887e-01 1.04267764e+00 -5.13792276e-01
3.35295945e-01 3.66983831e-01 1.54683650e-01 6.33519173e-01
8.40953708e-01 5.26628852e-01 -7.06372619e-01 -9.13127720e-01
9.33553874e-01 -2.94925272e-01 1.71355158e-01 -5.05680561e-01
-9.05919552e-01 6.91228509e-01 1.78732842e-01 3.83959889e-01
-1.39880210e-01 9.89067405e-02 1.26390755e-01 5.30987144e-01
-2.58292928e-02 6.47037446e-01 -9.11868811e-01 -1.33165747e-01
-8.71728659e-01 5.10332465e-01 1.25398111e+00 1.52653182e+00
1.24482810e-01 2.94908643e-01 -9.34252143e-02 3.17179203e-01
8.92314315e-01 1.01808000e+00 4.28362608e-01 -1.36146402e+00
-6.87414408e-02 1.72361732e-01 5.01781464e-01 -8.52631688e-01
-6.33224547e-01 -9.64177120e-03 -8.93263519e-01 4.49267089e-01
3.49161088e-01 4.57367361e-01 -8.02827001e-01 5.07305264e-01
4.33481112e-02 -2.34125629e-01 4.53833073e-01 5.55570945e-02
4.99930978e-01 3.76208365e-01 -1.34477139e-01 -5.73289394e-01
1.06082785e+00 -1.36716676e+00 -1.35299087e+00 2.33215362e-01
9.05734658e-01 -1.07320261e+00 4.35900748e-01 5.33875942e-01
-1.55548143e+00 -1.37560293e-01 -1.08699942e+00 1.78573877e-01
-7.27255583e-01 -3.14239264e-01 8.57801437e-01 1.43120694e+00
-1.60129595e+00 9.73287821e-01 -4.91727620e-01 6.59165755e-02
1.20568443e-02 8.24621081e-01 -2.64718989e-03 4.62812334e-01
-1.33193445e+00 1.08501506e+00 2.22979754e-01 -1.21242590e-01
-1.65216476e-01 -2.13068098e-01 -8.23704481e-01 -5.63443303e-01
-4.78693932e-01 -5.29636025e-01 1.44139910e+00 -2.59346128e-01
-1.65295815e+00 9.71794367e-01 -1.42069459e-01 -1.97814897e-01
6.14786744e-01 -1.28011424e-02 -8.31891418e-01 2.42498964e-01
-1.89849049e-01 5.76383233e-01 9.28263724e-01 -1.35132408e+00
-7.59897232e-01 -1.67359829e-01 -1.23336017e-01 2.66287565e-01
-1.25510961e-01 1.89734384e-01 2.11616129e-01 -1.12999000e-01
3.27147305e-01 -7.20919967e-01 -2.53068686e-01 -5.32041907e-01
-1.46512717e-01 -7.10518599e-01 7.70373225e-01 -4.51523662e-01
1.83705616e+00 -1.67618537e+00 -2.06720144e-01 3.98590982e-01
3.57815564e-01 -1.24705513e-03 2.40583986e-01 1.08380008e+00
-2.76906848e-01 7.32199550e-01 4.11965609e-01 -1.10722095e-01
2.24991128e-01 4.85861301e-01 -4.16602850e-01 3.05609167e-01
-1.29282743e-01 1.15307164e+00 -1.16605783e+00 -5.23096442e-01
4.11106765e-01 9.42391157e-02 -4.58732933e-01 4.79237735e-01
2.99364805e-01 1.47170946e-01 -3.56553018e-01 1.39399457e+00
1.15709066e+00 -1.31984919e-01 1.45911396e-01 5.30878425e-01
-3.79135728e-01 3.55090618e-01 -6.96863770e-01 1.05554795e+00
7.08333924e-02 5.00173986e-01 1.02364108e-01 -7.94621468e-01
3.33247900e-01 8.61540735e-01 4.37155962e-01 -9.67555881e-01
-2.26398129e-02 6.92409754e-01 1.12803578e-01 -6.07703328e-01
7.58228302e-01 3.94563079e-02 -3.64872098e-01 6.40070081e-01
-2.37588286e-01 -6.59476995e-01 9.64643434e-02 3.08100313e-01
6.22585893e-01 -5.18246442e-02 5.62923312e-01 -1.05613089e+00
7.86340594e-01 -1.82965681e-01 -1.81299388e-01 1.03415680e+00
-3.09923887e-01 3.19085121e-01 2.99841821e-01 -6.65102363e-01
-6.45341039e-01 -1.12307119e+00 -4.89381433e-01 1.30636716e+00
3.24267983e-01 -4.39044595e-01 -9.54439282e-01 -2.49762803e-01
1.77620783e-01 6.89606130e-01 -5.90509653e-01 3.84124845e-01
-5.03739953e-01 -8.32535863e-01 7.39044368e-01 3.45434904e-01
-5.07752821e-02 -1.33414865e+00 -6.58416986e-01 1.25490099e-01
-2.20292807e-01 -6.63697243e-01 -6.23428151e-02 4.48765576e-01
-1.35989368e+00 -5.18594682e-01 -6.66252747e-02 -8.20914626e-01
5.87345481e-01 2.46782884e-01 1.27047324e+00 5.39230824e-01
-2.31483161e-01 4.26904231e-01 -1.21292919e-01 -4.95818377e-01
-4.59671497e-01 -8.00336525e-02 5.28869390e-01 -5.87835789e-01
5.19427478e-01 -2.50617653e-01 -7.29350567e-01 5.37953973e-01
-6.88540697e-01 1.62748516e-01 1.79803044e-01 1.04410267e+00
1.35816500e-01 -9.34035778e-02 1.22507080e-01 -6.38007045e-01
8.72274399e-01 -1.69219792e-01 -3.78732830e-01 5.77745810e-02
-6.77108407e-01 -3.74140263e-01 3.21430594e-01 -3.25342178e-01
-1.01981449e+00 -4.87835288e-01 -9.82677937e-02 2.45538145e-01
1.11353043e-02 -1.46784872e-01 6.47139177e-02 -5.24923325e-01
8.02199244e-01 9.25758183e-02 1.99174434e-02 -6.80815242e-03
3.01039815e-01 7.09525108e-01 -6.82967342e-03 -6.68678164e-01
8.44880998e-01 4.91470337e-01 7.98524171e-02 -9.57177758e-01
-1.52186140e-01 -2.60129690e-01 -9.51962709e-01 -6.54426932e-01
6.56643391e-01 -6.78531289e-01 -9.10833478e-01 3.91110867e-01
-9.38691139e-01 -3.38627815e-01 -3.91645581e-01 4.25431967e-01
-1.01278400e+00 2.75717527e-02 -3.90154392e-01 -1.27895141e+00
-5.10977268e-01 -1.02017939e+00 9.43384409e-01 5.30070923e-02
-5.10597289e-01 -1.26927447e+00 5.87685481e-02 2.71537274e-01
1.81734428e-01 -1.73075795e-01 6.90226793e-01 -2.38256708e-01
-4.24233019e-01 -1.53791070e-01 2.34436691e-02 -1.39755070e-01
1.70832314e-02 4.95917559e-01 -9.81751978e-01 -5.31145096e-01
6.65065646e-02 -1.92070693e-01 -1.08835101e-01 6.52520418e-01
5.91872573e-01 -2.29931593e-01 -8.56000841e-01 5.40386558e-01
1.38545322e+00 3.85070026e-01 5.32770038e-01 7.28214979e-01
1.41836226e-01 5.53460240e-01 9.17806149e-01 4.63203549e-01
1.30579369e-02 3.28798652e-01 2.40537539e-01 1.49327129e-01
1.11720070e-01 -1.54819340e-01 3.77893507e-01 1.16112018e+00
-8.18235934e-01 -2.69281328e-01 -5.07867396e-01 4.42987174e-01
-1.72482407e+00 -1.40330648e+00 -4.32368398e-01 6.90478683e-01
6.25676990e-01 1.56016424e-01 -1.48347050e-01 3.35214496e-01
4.99015123e-01 -2.03574806e-01 -1.19133167e-01 -1.06291151e+00
-1.43546045e-01 3.15233678e-01 7.37729073e-01 1.00061214e+00
-7.20721722e-01 1.03317809e+00 1.29781246e+01 1.02230716e+00
2.21112028e-01 1.03134915e-01 5.16071796e-01 3.48020852e-01
-4.36954498e-01 -4.56139445e-02 -1.04416132e+00 2.72933897e-02
1.38140702e+00 -4.30666685e-01 6.85999811e-01 5.44219851e-01
3.44648361e-01 -4.23268199e-01 -1.26188684e+00 5.26221812e-01
9.73738134e-02 -1.40886843e+00 -2.83300440e-04 6.85225725e-01
7.73699820e-01 -5.08050561e-01 6.22419357e-01 3.24184299e-01
6.09259963e-01 -1.14389277e+00 8.60300779e-01 2.53660440e-01
1.03040910e+00 -6.05088234e-01 5.67372203e-01 1.68872893e-01
-1.14389896e+00 -2.20873043e-01 -8.77727985e-01 -1.00755692e+00
3.93533185e-02 -1.81779593e-01 -4.29956943e-01 3.48861217e-01
9.58353162e-01 2.99398601e-01 -3.93658698e-01 9.95779395e-01
-4.78694476e-02 1.04875881e-02 -2.71853864e-01 -4.48467314e-01
4.83122796e-01 -3.54241252e-01 4.66730654e-01 1.00164843e+00
2.48499006e-01 3.51035744e-01 -9.84472036e-02 4.01770771e-01
5.45058846e-01 3.29446048e-02 -1.19659424e+00 -1.78908288e-01
2.83276141e-01 9.16795909e-01 -4.83487815e-01 -4.22520459e-01
-2.00212970e-01 8.62069130e-01 -3.55488248e-02 5.01107454e-01
-6.11489356e-01 -4.35615242e-01 9.72222984e-01 -1.27327025e-01
-1.14700586e-01 -3.48497719e-01 -6.23769283e-01 -7.30352640e-01
-5.89872956e-01 -4.54965204e-01 5.93606755e-02 -5.53365827e-01
-1.39813089e+00 5.79277515e-01 -2.27688253e-02 -1.40553558e+00
-6.99901402e-01 -1.27676582e+00 -4.76714373e-01 4.92853165e-01
-1.11898029e+00 -1.10984349e+00 2.50124663e-01 4.52870727e-01
1.64141744e-01 -5.34416080e-01 1.39563632e+00 3.57715860e-02
1.00637585e-01 9.24474537e-01 6.69434488e-01 -7.37814724e-01
5.56605101e-01 -1.27867436e+00 5.68737745e-01 -1.37897313e-01
-4.31265175e-01 9.05828118e-01 6.28349900e-01 -5.39804697e-01
-1.41196322e+00 -3.66917729e-01 1.08350635e+00 -9.83769417e-01
6.55218959e-01 -3.86345625e-01 4.23767231e-02 7.88592756e-01
7.15902448e-01 -6.18741751e-01 8.21781039e-01 -1.83753878e-01
1.80774391e-01 5.75296998e-01 -1.39248300e+00 6.12354755e-01
1.66275799e+00 -4.63594139e-01 -6.25784039e-01 7.60327101e-01
8.13696027e-01 -6.94087505e-01 -1.30082703e+00 3.34633321e-01
8.65424156e-01 -8.75409484e-01 1.61978090e+00 -1.32660246e+00
-4.43697497e-02 2.81152606e-01 -2.61993498e-01 -9.32519078e-01
-5.96193194e-01 -1.23518765e+00 -5.33532679e-01 -5.83747849e-02
5.96577883e-01 -1.13057327e+00 3.42365682e-01 8.74560475e-01
-2.82833427e-01 -6.42737269e-01 -1.06996536e+00 -1.32016802e+00
-3.35779637e-02 -1.45572275e-01 4.90409225e-01 7.63798356e-01
6.83744550e-01 1.09839931e-01 -6.36873543e-02 -1.00294888e-01
5.33176839e-01 9.55312885e-03 4.41501856e-01 -1.34294486e+00
3.86843324e-01 -5.75816095e-01 -3.07655483e-01 -9.45992947e-01
-8.85957032e-02 -8.12076271e-01 -6.53862000e-01 -1.28511906e+00
-8.32044985e-03 -1.91056758e-01 -1.11109078e-01 -1.65725678e-01
3.67937148e-01 2.13746816e-01 1.20859891e-02 1.03788137e-01
-3.71160030e-01 6.18435517e-02 1.29639816e+00 6.91750320e-05
-1.62315920e-01 4.85058486e-01 -4.81304944e-01 7.84440815e-01
8.58408585e-02 -2.99253196e-01 -6.78878546e-01 6.11881316e-02
6.69384480e-01 4.61409837e-02 3.24159935e-02 -7.42885649e-01
5.37211418e-01 -3.75702560e-01 4.78586555e-01 -1.32223868e+00
1.30741090e-01 -9.61415648e-01 6.95283338e-02 9.40189242e-01
2.74610907e-01 1.20424610e-02 8.66204947e-02 5.39500564e-02
-1.44871444e-01 -5.70943117e-01 9.21121240e-01 -4.07591403e-01
-4.92852688e-01 -5.20386267e-03 -1.03226590e+00 8.97834301e-02
9.92593169e-01 -7.84614205e-01 -3.59281451e-01 -4.20183957e-01
-8.29068601e-01 -1.95836127e-02 6.50830388e-01 3.03609259e-02
7.30431557e-01 -1.51530886e+00 -2.30721906e-01 7.18729138e-01
-3.22939813e-01 -3.74400020e-01 -1.70157343e-01 6.58265352e-01
-1.32361674e+00 1.02442718e+00 -5.41665435e-01 -4.55340147e-01
-1.14228773e+00 4.78126436e-01 4.28307921e-01 -2.41845414e-01
-2.15481281e-01 1.11954463e+00 2.71224789e-02 -8.23025763e-01
1.85185194e-01 -9.21545625e-02 -7.54407048e-01 3.55081353e-03
6.88606799e-01 1.05194807e+00 -2.91290224e-01 -6.04341030e-01
-4.56784427e-01 6.65885091e-01 2.32151806e-01 -2.87484169e-01
9.17833567e-01 -2.19243199e-01 -9.89108324e-01 4.28274393e-01
8.38715494e-01 -1.36269778e-01 -3.69319022e-02 4.16855574e-01
1.25943512e-01 -8.02164078e-01 -4.32406247e-01 -3.55811834e-01
-1.85641110e-01 5.54822803e-01 5.34874737e-01 8.99602413e-01
8.57008278e-01 -3.02566767e-01 8.18335712e-01 9.66778398e-01
5.72402716e-01 -1.68019545e+00 -2.13140488e-01 6.89524531e-01
9.12339568e-01 -9.28350806e-01 5.44190466e-01 -7.27165341e-01
-4.14997995e-01 1.32979155e+00 4.68304873e-01 -1.55325383e-01
1.27306652e+00 5.74917436e-01 1.14069022e-02 -3.70670199e-01
-9.44949508e-01 1.12705544e-01 3.75366658e-01 1.11147714e+00
5.06513238e-01 5.07374525e-01 -9.66500878e-01 3.21953118e-01
-7.28706717e-01 -2.34555230e-01 4.90474731e-01 1.41972518e+00
-6.43810987e-01 -1.20391917e+00 -7.23931909e-01 4.61561680e-01
-5.49773455e-01 -1.16372630e-01 -5.28106689e-01 8.46754074e-01
-5.76629937e-02 1.49448860e+00 -1.97535474e-03 -5.03491640e-01
4.11356747e-01 1.41089618e-01 7.21762300e-01 -1.23501487e-01
-9.04846430e-01 4.15413082e-01 3.84890139e-01 -1.23056793e+00
-8.58632207e-01 -1.05834293e+00 -1.40667629e+00 -1.19437599e+00
-5.12782812e-01 1.89310342e-01 3.83317530e-01 3.90289724e-01
-2.06836104e-01 2.85260603e-02 9.80917513e-01 -1.07949340e+00
-3.90341938e-01 -9.39418614e-01 -1.00026262e+00 -7.84516707e-02
2.89751232e-01 -8.00943017e-01 -7.83523321e-01 2.83909619e-01] | [-7.221755027770996, 3.7373082637786865] |
9c165da6-eeb6-4016-a1cc-316335ebcf3c | a-flexible-frame-rate-vision-aided-inertial | 2210.12476 | null | https://arxiv.org/abs/2210.12476v1 | https://arxiv.org/pdf/2210.12476v1.pdf | A Flexible-Frame-Rate Vision-Aided Inertial Object Tracking System for Mobile Devices | Real-time object pose estimation and tracking is challenging but essential for emerging augmented reality (AR) applications. In general, state-of-the-art methods address this problem using deep neural networks which indeed yield satisfactory results. Nevertheless, the high computational cost of these methods makes them unsuitable for mobile devices where real-world applications usually take place. In addition, head-mounted displays such as AR glasses require at least 90~FPS to avoid motion sickness, which further complicates the problem. We propose a flexible-frame-rate object pose estimation and tracking system for mobile devices. It is a monocular visual-inertial-based system with a client-server architecture. Inertial measurement unit (IMU) pose propagation is performed on the client side for high speed tracking, and RGB image-based 3D pose estimation is performed on the server side to obtain accurate poses, after which the pose is sent to the client side for visual-inertial fusion, where we propose a bias self-correction mechanism to reduce drift. We also propose a pose inspection algorithm to detect tracking failures and incorrect pose estimation. Connected by high-speed networking, our system supports flexible frame rates up to 120 FPS and guarantees high precision and real-time tracking on low-end devices. Both simulations and real world experiments show that our method achieves accurate and robust object tracking. | ['Yi-Ping Hung', 'Hsiao-Ching Tseng', 'I-Ju Hsieh', 'Kuan-Wei Tseng', 'Yo-Chung Lau'] | 2022-10-22 | null | null | null | null | ['3d-pose-estimation'] | ['computer-vision'] | [-2.17345566e-01 -5.48048854e-01 -1.26317739e-01 -2.36005589e-01
-4.94103909e-01 -2.82832682e-01 -4.73709852e-02 -4.37518805e-01
-3.92997384e-01 5.43864131e-01 -2.73479283e-01 -3.47310781e-01
1.86340079e-01 -5.92489660e-01 -8.93943310e-01 -5.69400549e-01
1.79503679e-01 3.83615136e-01 4.53037590e-01 1.09313335e-02
-1.69159502e-01 8.51414204e-01 -1.71503901e+00 -4.43113089e-01
5.68445742e-01 1.47594810e+00 2.83467144e-01 7.12694347e-01
2.88089156e-01 2.11211935e-01 -6.14785731e-01 2.44188886e-02
1.59909829e-01 3.62740844e-01 4.09825481e-02 1.16548494e-01
7.09614336e-01 -9.04809535e-01 -4.76388961e-01 1.03707743e+00
1.00412452e+00 -2.80910879e-01 -1.27088547e-01 -1.29847229e+00
-1.75100997e-01 -3.31570446e-01 -6.54260159e-01 2.93228105e-02
7.57857442e-01 1.67418897e-01 2.55278081e-01 -9.11501050e-01
4.16824520e-01 9.68562603e-01 1.07264805e+00 6.91260219e-01
-8.15138638e-01 -7.23142743e-01 2.04654112e-01 -4.42309491e-02
-1.52874732e+00 -5.52170098e-01 6.67127192e-01 -2.25699022e-01
7.89655566e-01 4.16252881e-01 9.23658490e-01 8.10216784e-01
5.69000542e-01 5.93454778e-01 3.76361012e-01 -5.74829467e-02
2.06643134e-01 5.30342646e-02 -6.85859174e-02 4.94803011e-01
6.13872349e-01 -1.58042144e-02 -4.81508136e-01 -1.36886552e-01
1.22823012e+00 5.94402552e-01 -6.70398474e-01 -6.43060803e-01
-1.37487698e+00 1.16362892e-01 7.73117006e-01 -1.82120893e-02
-5.69246471e-01 3.46551925e-01 2.26016104e-01 5.36599793e-02
3.74975830e-01 -3.73253554e-01 -5.61900795e-01 -3.03182393e-01
-6.24191582e-01 -1.82399189e-03 3.87079120e-01 1.35705185e+00
1.83617085e-01 2.90095180e-01 2.84676403e-01 3.06821257e-01
9.62216675e-01 1.05435157e+00 4.03482825e-01 -7.56228983e-01
4.30720419e-01 4.68658507e-01 4.24128860e-01 -1.01914799e+00
-6.39206886e-01 -5.05050123e-01 -8.17686379e-01 3.36107761e-01
-6.27949834e-03 -1.33453429e-01 -6.27722681e-01 1.41850293e+00
8.74885261e-01 4.59046543e-01 -3.77148002e-01 1.39770710e+00
7.88253427e-01 2.88371056e-01 -2.03323990e-01 -4.87934589e-01
1.41291809e+00 -4.06742901e-01 -8.99536014e-01 -9.29381326e-02
4.23267871e-01 -7.74658442e-01 1.01688814e+00 4.71391678e-01
-9.34767962e-01 -7.20637798e-01 -1.29185688e+00 1.02033213e-01
2.52887249e-01 3.16702068e-01 5.49726427e-01 9.52354729e-01
-1.00833392e+00 3.13361883e-01 -1.40957952e+00 -3.51801902e-01
-1.40070498e-01 1.02819288e+00 -1.13512643e-01 5.03580123e-02
-7.87061036e-01 6.74296141e-01 -5.73068112e-02 6.39717817e-01
-2.91761279e-01 -4.59241152e-01 -7.93027639e-01 -2.73509860e-01
3.00309718e-01 -1.07048261e+00 1.48144484e+00 -5.00372052e-01
-1.92834926e+00 4.95654464e-01 -3.79285306e-01 -7.31086135e-02
4.84367907e-01 -7.42770433e-01 -6.20413780e-01 -9.81952026e-02
-2.20200986e-01 2.61899590e-01 7.89818108e-01 -1.19452870e+00
-5.30308545e-01 -7.64382720e-01 -1.65049195e-01 3.97123963e-01
-4.10687089e-01 -3.62348258e-02 -8.70950878e-01 -7.96658546e-02
7.99023390e-01 -1.12968647e+00 3.42268310e-02 4.99183238e-01
-9.22902524e-02 2.27054641e-01 1.23734295e+00 -4.36112732e-01
1.05796361e+00 -2.12236476e+00 -2.77515024e-01 4.20669131e-02
3.02364081e-01 2.57760137e-01 4.39451814e-01 -3.00517529e-01
1.55680045e-01 -6.95030630e-01 6.37728453e-01 -6.18200243e-01
-2.07185000e-01 1.34153962e-01 -2.49540523e-01 8.75050783e-01
-4.68094558e-01 7.09355295e-01 -4.92044359e-01 -2.36353680e-01
5.94506979e-01 9.94570553e-01 -3.82037848e-01 1.27902731e-01
1.20788634e-01 6.62652373e-01 -4.58579659e-01 9.60020959e-01
7.39711106e-01 -4.38734889e-01 1.15516096e-01 -4.43683118e-01
-1.29280537e-01 3.53747845e-01 -1.69128108e+00 1.57631528e+00
-5.31981528e-01 5.24344683e-01 2.83396304e-01 -1.33128956e-01
8.32917213e-01 3.69555324e-01 5.82417130e-01 -6.60123944e-01
5.35129309e-01 3.54382217e-01 -4.14086521e-01 -5.64098842e-02
7.85818517e-01 2.13172466e-01 1.36893317e-01 7.24263042e-02
-5.30869961e-01 1.50867164e-01 -6.42108440e-01 -1.23601202e-02
9.42779422e-01 5.04046381e-01 1.89923346e-01 3.72851938e-01
4.75542247e-01 -3.16335618e-01 7.53392935e-01 3.03405374e-01
-2.60942101e-01 6.83231533e-01 -6.13855898e-01 -5.65621316e-01
-8.22612822e-01 -1.21660101e+00 -2.53421403e-02 6.46749675e-01
7.06672549e-01 -2.36641124e-01 -3.24104279e-01 -2.13232547e-01
2.27971599e-01 6.10879287e-02 -3.78001668e-02 -1.42870128e-01
-7.39610970e-01 -4.37307566e-01 -3.16683501e-02 9.39717472e-01
4.35297221e-01 -4.82736737e-01 -8.92015636e-01 3.83462787e-01
-2.28212681e-02 -1.24055851e+00 -4.65200394e-01 -3.17622602e-01
-1.32983160e+00 -7.81955600e-01 -6.75903976e-01 -4.22918797e-01
6.90678537e-01 7.72760510e-01 8.34832728e-01 1.44862384e-01
-8.07086751e-02 5.89135528e-01 8.97132009e-02 -3.75140697e-01
1.87166184e-01 -3.13097894e-01 7.88458645e-01 -1.53646320e-01
2.38743693e-01 -4.00388837e-01 -8.33440244e-01 6.78190053e-01
-3.36169928e-01 -1.07907645e-01 3.05670470e-01 3.04875493e-01
8.07262003e-01 -4.03203666e-01 1.59304529e-01 6.32647127e-02
1.28081292e-01 1.21675894e-01 -1.25828815e+00 -4.89492863e-02
-5.54317236e-01 -5.61269164e-01 2.01942757e-01 -7.10636854e-01
-8.13337505e-01 4.49369580e-01 -3.08153749e-01 -7.73507655e-01
4.42997143e-02 2.19193235e-01 -3.61143947e-01 -4.06648874e-01
4.61978704e-01 9.04478952e-02 5.80437928e-02 -4.23110247e-01
1.19531907e-01 1.03734720e+00 7.66112626e-01 3.37996287e-03
7.91533291e-01 7.16631770e-01 -1.33644283e-01 -8.72315586e-01
-3.84495884e-01 -5.02652705e-01 -3.97769660e-01 -5.85872412e-01
5.68243265e-01 -1.55514228e+00 -1.56161225e+00 4.63332951e-01
-1.20896435e+00 9.58137661e-02 -1.26188025e-02 9.43004966e-01
-4.80240107e-01 3.30379993e-01 -6.10727966e-01 -1.02859187e+00
-6.47189379e-01 -1.26628268e+00 1.49899971e+00 5.14454722e-01
1.88819487e-02 -3.83758098e-01 -6.98539615e-02 2.33690381e-01
3.46896797e-01 -1.86645426e-02 -2.17513397e-01 2.28688434e-01
-1.07729554e+00 -5.20563304e-01 -2.45047927e-01 -3.06271642e-01
6.99161664e-02 -2.09544957e-01 -8.21570814e-01 -7.32587397e-01
2.43446097e-01 1.18760072e-01 -3.73063485e-05 7.09496319e-01
6.63174272e-01 1.30912289e-01 -7.58340061e-01 7.47980833e-01
1.30449116e+00 3.84286165e-01 6.00105524e-01 5.28831601e-01
1.06521046e+00 -3.52104038e-01 8.89018238e-01 3.74006510e-01
6.76072598e-01 1.26415908e+00 6.98369920e-01 6.95577264e-02
-9.48763732e-03 1.53230399e-01 4.73782629e-01 1.07616305e+00
-6.48757368e-02 -3.82823609e-02 -6.65100873e-01 8.20287690e-02
-1.90144062e+00 -4.18418407e-01 -4.31825191e-01 2.68821573e+00
5.12703419e-01 3.30559760e-01 1.64609402e-01 2.53698051e-01
6.91772521e-01 -2.98006296e-01 -7.02571690e-01 3.14644963e-01
4.38922375e-01 -2.71545678e-01 5.75069726e-01 3.29311460e-01
-9.42938447e-01 5.79526901e-01 5.48400640e+00 1.45602077e-01
-1.62189209e+00 2.43943661e-01 -1.42736256e-01 -4.25487339e-01
1.82331115e-01 -5.47816873e-01 -1.23634052e+00 4.25662577e-01
9.58194613e-01 2.97515869e-01 -1.09696627e-01 1.43880916e+00
1.21642336e-01 -1.86798036e-01 -8.43709588e-01 1.42483401e+00
-1.26944324e-02 -1.09844255e+00 -6.95227146e-01 2.26603553e-01
2.83531278e-01 3.71244758e-01 -1.01741761e-01 6.17347881e-02
-1.39148533e-01 -1.79987431e-01 7.30764925e-01 3.83368522e-01
1.09144366e+00 -7.34704852e-01 7.51143575e-01 3.77612412e-01
-1.59882903e+00 8.71079490e-02 -3.53844702e-01 -1.80279672e-01
5.85876465e-01 7.14910388e-01 -6.91428423e-01 5.03838360e-01
8.43093574e-01 3.61883074e-01 -3.06166708e-01 1.28178108e+00
-4.98177148e-02 3.77246588e-02 -7.61639714e-01 -1.83596492e-01
-4.68057036e-01 1.47724628e-01 6.20741725e-01 3.82908672e-01
6.01159155e-01 9.11767930e-02 3.24816048e-01 2.65021592e-01
1.02453031e-01 -3.52398753e-01 -5.53677917e-01 5.66697478e-01
4.25138116e-01 1.07980156e+00 -6.91818953e-01 -2.55435377e-01
-3.83827925e-01 1.06356573e+00 -2.27105364e-01 4.49837521e-02
-1.07998466e+00 -2.52561212e-01 9.07388926e-01 2.72442341e-01
2.03116059e-01 -8.53599548e-01 -3.72802258e-01 -1.45132279e+00
4.37806755e-01 -5.44558406e-01 -1.30144313e-01 -1.04578888e+00
-6.99983597e-01 6.31575346e-01 -5.45488894e-01 -1.90941644e+00
-3.51507872e-01 -5.51091075e-01 -6.76577538e-02 5.91284335e-01
-1.15377605e+00 -1.02045667e+00 -7.69890845e-01 7.13498473e-01
3.21181297e-01 6.04179576e-02 6.25239313e-01 8.13431680e-01
-5.30764401e-01 6.16484344e-01 6.54625595e-02 -7.29070678e-02
5.59385717e-01 -7.02389061e-01 6.69086516e-01 8.33033800e-01
7.19351545e-02 8.22876513e-01 7.45376050e-01 -1.01648510e+00
-2.18638206e+00 -8.73667836e-01 6.04876101e-01 -5.65172911e-01
2.27008432e-01 -3.37296844e-01 -7.21817553e-01 7.44188607e-01
-4.25148875e-01 5.32630980e-01 2.77938277e-01 -6.40394241e-02
-1.82890408e-02 -3.00713688e-01 -1.01992822e+00 6.31007910e-01
1.18321288e+00 -4.87092644e-01 -2.43013561e-01 2.93130964e-01
8.76019895e-01 -1.29084516e+00 -6.62209451e-01 5.04345417e-01
9.69059348e-01 -7.87627280e-01 1.10048366e+00 1.38354003e-01
-5.68515480e-01 -8.86995673e-01 -2.24849343e-01 -7.15549946e-01
-2.01784819e-01 -6.47664607e-01 -6.90706611e-01 7.36163139e-01
-1.54327959e-01 -7.14892566e-01 1.14782727e+00 6.84198618e-01
-3.56897675e-02 -5.42940855e-01 -1.25313711e+00 -9.40141201e-01
-1.15283966e+00 -8.18330646e-01 4.93043780e-01 4.91874665e-01
-3.77446949e-01 3.57159436e-01 -4.05413985e-01 7.66994894e-01
7.38197386e-01 5.22027127e-02 1.21213078e+00 -1.27895880e+00
-9.22764316e-02 2.35847384e-01 -7.97487378e-01 -1.53726709e+00
-3.60923618e-01 -9.68034342e-02 1.14645630e-01 -1.29952204e+00
-2.99736321e-01 -4.63790327e-01 1.52229205e-01 -7.12850839e-02
5.45567721e-02 6.54780865e-01 1.99891657e-01 3.09844524e-01
-6.52544081e-01 6.03363395e-01 7.43125796e-01 3.37786466e-01
-4.94614422e-01 5.03696442e-01 -1.45450488e-01 9.00847375e-01
4.24017817e-01 -3.46957386e-01 -1.37386173e-01 -7.34850347e-01
3.09718519e-01 3.09049875e-01 5.46267986e-01 -1.49223638e+00
4.66329694e-01 2.31061906e-01 8.58241796e-01 -1.06088173e+00
7.54744709e-01 -1.47402287e+00 5.10870397e-01 6.96541786e-01
6.44795120e-01 4.08612967e-01 2.68650979e-01 4.59657282e-01
1.40178353e-01 3.66018713e-01 4.89027083e-01 3.45200658e-01
-3.42123598e-01 5.93924046e-01 -8.88511240e-02 -7.07842827e-01
9.58219945e-01 -6.93394899e-01 -1.04230762e-01 -6.24201894e-01
-6.36786580e-01 -7.32531250e-02 6.79058015e-01 5.43254912e-01
8.26360583e-01 -1.75491595e+00 3.58809382e-02 3.31778675e-01
-5.89115173e-02 2.07252279e-01 2.07404777e-01 1.11815965e+00
-4.84935015e-01 5.20358741e-01 1.02214515e-01 -1.40340924e+00
-1.64855433e+00 3.97791833e-01 2.08066180e-01 2.36438543e-01
-7.16497302e-01 6.56895816e-01 -2.30718613e-01 -3.87013972e-01
5.10198534e-01 -3.66499901e-01 2.29790196e-01 -3.29019338e-01
8.27902794e-01 3.42202753e-01 3.62412184e-01 -8.06885958e-01
-7.62145162e-01 9.75540102e-01 1.07355297e-01 -1.05654672e-01
1.04702497e+00 -6.36382043e-01 3.20361823e-01 2.88618356e-01
1.02137506e+00 -4.55458425e-02 -1.31961703e+00 -2.42944360e-01
-3.00915450e-01 -6.94616199e-01 3.62365037e-01 -2.52892911e-01
-1.10151458e+00 6.89704478e-01 1.30998635e+00 -1.63449094e-01
1.09294617e+00 -2.71255672e-01 9.74523425e-01 4.58469033e-01
1.04799342e+00 -8.50163758e-01 -3.46119478e-02 5.31438589e-01
7.15448141e-01 -1.29101634e+00 3.39462727e-01 -4.62927788e-01
-2.38541663e-01 9.23240244e-01 8.81211817e-01 1.47623673e-01
4.93740529e-01 5.51451743e-01 3.06704760e-01 2.49433462e-02
-3.07795614e-01 -5.00078127e-02 2.44882137e-01 6.68757200e-01
2.58210629e-01 -1.36021569e-01 3.90086174e-02 2.35661745e-01
-1.83254965e-02 3.32144260e-01 2.21325755e-01 1.25925517e+00
-4.06720579e-01 -6.69062316e-01 -9.85992551e-01 2.50476543e-02
-3.24225247e-01 4.21717346e-01 2.59963721e-01 6.86290979e-01
-1.55030206e-01 6.49420738e-01 2.61802018e-01 -6.10142410e-01
5.89658141e-01 -2.65993327e-01 6.10042930e-01 -3.71559978e-01
-6.02140069e-01 6.16697729e-01 -1.69192880e-01 -1.03496039e+00
-1.73789799e-01 -5.37629724e-01 -1.39848566e+00 -3.90397638e-01
-7.70114124e-01 -3.93647939e-01 1.23653686e+00 7.12874532e-01
7.62563229e-01 5.68075478e-01 5.17820597e-01 -1.44952178e+00
-3.87197971e-01 -6.68066025e-01 -2.61641115e-01 -1.52978659e-01
5.77374220e-01 -8.37995350e-01 7.65520781e-02 -1.30703896e-01] | [7.119724273681641, -1.7976369857788086] |
cb33f052-ed45-445c-8e2d-b51e605dddd5 | open-set-adversarial-defense-with-clean | 2202.05953 | null | https://arxiv.org/abs/2202.05953v1 | https://arxiv.org/pdf/2202.05953v1.pdf | Open-set Adversarial Defense with Clean-Adversarial Mutual Learning | Open-set recognition and adversarial defense study two key aspects of deep learning that are vital for real-world deployment. The objective of open-set recognition is to identify samples from open-set classes during testing, while adversarial defense aims to robustify the network against images perturbed by imperceptible adversarial noise. This paper demonstrates that open-set recognition systems are vulnerable to adversarial samples. Furthermore, this paper shows that adversarial defense mechanisms trained on known classes are unable to generalize well to open-set samples. Motivated by these observations, we emphasize the necessity of an Open-Set Adversarial Defense (OSAD) mechanism. This paper proposes an Open-Set Defense Network with Clean-Adversarial Mutual Learning (OSDN-CAML) as a solution to the OSAD problem. The proposed network designs an encoder with dual-attentive feature-denoising layers coupled with a classifier to learn a noise-free latent feature representation, which adaptively removes adversarial noise guided by channel and spatial-wise attentive filters. Several techniques are exploited to learn a noise-free and informative latent feature space with the aim of improving the performance of adversarial defense and open-set recognition. First, we incorporate a decoder to ensure that clean images can be well reconstructed from the obtained latent features. Then, self-supervision is used to ensure that the latent features are informative enough to carry out an auxiliary task. Finally, to exploit more complementary knowledge from clean image classification to facilitate feature denoising and search for a more generalized local minimum for open-set recognition, we further propose clean-adversarial mutual learning, where a peer network (classifying clean images) is further introduced to mutually learn with the classifier (classifying adversarial images). | ['Vishal M. Patel', 'Pong C. Yuen', 'Pramuditha Perera', 'Rui Shao'] | 2022-02-12 | null | null | null | null | ['adversarial-defense', 'open-set-learning'] | ['adversarial', 'miscellaneous'] | [ 4.64485049e-01 -3.05712037e-02 1.85142502e-01 -2.54450798e-01
-1.22586775e+00 -1.08428442e+00 4.13927376e-01 -5.28554499e-01
4.43414412e-02 5.00712633e-01 -7.84698725e-02 -2.62182266e-01
-1.33599669e-01 -9.48075414e-01 -1.08922529e+00 -1.06280959e+00
-2.73836702e-01 -2.24401385e-01 -2.96737641e-01 -3.87108654e-01
1.06167994e-01 6.88913167e-01 -1.44287932e+00 3.89529854e-01
1.05634797e+00 1.12082398e+00 -1.29435420e-01 7.70916224e-01
7.45948404e-02 6.63856804e-01 -8.61389935e-01 -3.67643207e-01
9.40888405e-01 -4.97112006e-01 -6.09214127e-01 8.93998370e-02
6.63538814e-01 -3.13452363e-01 -6.57703161e-01 1.42859471e+00
4.36596006e-01 9.78346691e-02 7.41107702e-01 -1.20838189e+00
-1.17301404e+00 3.17659944e-01 -1.60160929e-01 4.08104919e-02
2.82077670e-01 5.69061816e-01 6.82155192e-01 -6.13863826e-01
3.46655101e-01 1.12845230e+00 4.88859981e-01 9.38951373e-01
-1.10973239e+00 -9.56476152e-01 2.46005803e-01 9.06613395e-02
-1.26799619e+00 -5.28701305e-01 8.66670370e-01 -5.46908639e-02
4.87158775e-01 8.03501606e-01 6.07247241e-02 1.66685152e+00
3.63970280e-01 4.19254988e-01 1.28133976e+00 -3.80007297e-01
2.49580011e-01 2.83530146e-01 -8.33983868e-02 6.25067532e-01
-5.02758361e-02 7.57744431e-01 -1.06632113e-01 -2.40902677e-01
7.31289864e-01 2.68834859e-01 -6.68823957e-01 -2.16924772e-01
-7.21425951e-01 7.72768915e-01 9.31957603e-01 2.80268371e-01
-1.95095748e-01 -4.55866866e-02 2.03554228e-01 9.70399976e-01
2.19794407e-01 6.19599879e-01 -2.05256835e-01 3.26644778e-01
-4.86138940e-01 -2.96139807e-01 1.01472938e+00 7.81064808e-01
6.94562256e-01 4.20696437e-01 -4.52855416e-02 7.07964540e-01
1.73486352e-01 9.18252468e-01 3.49352211e-01 -7.79706538e-01
3.50901574e-01 1.43347636e-01 -2.97106683e-01 -9.73837316e-01
3.20867062e-01 -7.02898026e-01 -1.17422521e+00 7.83533454e-01
1.84403658e-01 -1.57783523e-01 -1.00928879e+00 1.80243921e+00
1.77969381e-01 4.69590515e-01 6.63105130e-01 8.06012511e-01
5.41162908e-01 7.49956906e-01 -4.38990742e-01 -1.03624001e-01
9.51682687e-01 -7.69870639e-01 -3.93501997e-01 -2.75804847e-01
2.17303500e-01 -4.90419418e-01 7.98784316e-01 4.27777916e-01
-5.32182395e-01 -6.81996286e-01 -1.25777602e+00 4.29179341e-01
-5.83801448e-01 -5.41962743e-01 2.34640464e-01 9.34418261e-01
-6.09924734e-01 5.87717831e-01 -4.86723006e-01 5.86195067e-02
7.43641973e-01 4.52101648e-01 -8.57357144e-01 -2.77861089e-01
-1.26089382e+00 6.68778300e-01 6.47687092e-02 2.62472272e-01
-1.62722433e+00 -3.08731765e-01 -1.09079313e+00 8.17770958e-02
3.24102402e-01 -5.99154174e-01 5.02396703e-01 -1.47351372e+00
-1.33731246e+00 6.86179399e-01 3.40249479e-01 -6.84652686e-01
4.75027442e-01 -1.19750313e-01 -7.02213645e-01 3.22511464e-01
-3.46731395e-02 2.71494955e-01 1.64040244e+00 -1.60952806e+00
-1.82736173e-01 -3.96889448e-01 2.82032847e-01 -1.20066982e-02
-7.12964654e-01 -8.88926834e-02 5.53286821e-02 -9.89846170e-01
1.09517276e-01 -7.97673583e-01 -3.00246805e-01 -4.73519899e-02
-5.24289727e-01 4.17124450e-01 1.06477189e+00 -6.13714159e-01
8.53091002e-01 -2.39007759e+00 8.61684233e-02 5.26603580e-01
2.94634521e-01 4.14035290e-01 -5.89828789e-01 2.64857411e-01
-5.93577743e-01 1.89731538e-01 -3.86300027e-01 -5.45260459e-02
-8.36098120e-02 4.95063394e-01 -8.24652314e-01 6.49999857e-01
3.67138326e-01 7.31947422e-01 -7.16691971e-01 6.99969903e-02
1.90775976e-01 4.07783180e-01 -4.74568397e-01 4.32555199e-01
2.48147044e-02 7.28692949e-01 -3.60491604e-01 7.14678347e-01
9.52178240e-01 2.78530151e-01 -2.58828431e-01 -3.58385295e-02
2.04507783e-01 -3.57038528e-01 -1.25285125e+00 1.23828900e+00
-5.07477999e-01 3.31840634e-01 5.13387203e-01 -1.07887065e+00
1.01167226e+00 2.95872569e-01 1.02953039e-01 -5.29245973e-01
3.57293874e-01 2.78261155e-01 -1.28743181e-03 -4.22185749e-01
-2.23904520e-01 6.37408942e-02 -2.20890209e-01 2.80733526e-01
2.24359825e-01 -4.37864736e-02 -5.18591821e-01 1.01767480e-01
1.46206307e+00 -2.48500109e-01 -4.38673422e-02 -1.24543443e-01
8.60569835e-01 -5.12004912e-01 6.51948988e-01 1.10643280e+00
-2.41951585e-01 6.64425850e-01 1.45056903e-01 -4.49473888e-01
-6.62993014e-01 -1.33372641e+00 -1.73294023e-01 6.95032001e-01
2.66090840e-01 8.20330530e-02 -9.20373380e-01 -1.29373705e+00
-6.13961294e-02 3.87740999e-01 -8.23459089e-01 -6.52436852e-01
-3.08315873e-01 -1.75198719e-01 7.32044101e-01 3.70308042e-01
6.72974050e-01 -9.37463403e-01 5.65507449e-02 -2.07357571e-01
7.60402381e-02 -7.84082770e-01 -5.08218706e-01 6.00847423e-01
-4.76396888e-01 -1.24568272e+00 -3.66527081e-01 -9.01011467e-01
8.68935704e-01 3.63684297e-01 4.98373777e-01 6.08237088e-02
-1.32449567e-01 4.47119445e-01 -5.72026908e-01 -3.12596530e-01
-8.41608942e-01 -3.74193579e-01 4.24809784e-01 4.43847179e-01
-2.30982341e-02 -7.35083699e-01 -2.12003067e-01 4.72927034e-01
-1.22899544e+00 -6.19554698e-01 5.92653930e-01 1.28478360e+00
4.87037778e-01 2.99970388e-01 6.62786484e-01 -6.95817113e-01
3.45479310e-01 -4.95102108e-01 -5.44844151e-01 2.62180030e-01
-1.79132253e-01 6.49145767e-02 1.28524899e+00 -6.08072698e-01
-8.31176758e-01 -1.70105368e-01 -4.63417172e-01 -1.10509193e+00
-4.77436602e-01 -5.50497659e-02 -7.71407962e-01 -6.88739896e-01
1.05480337e+00 4.98800397e-01 1.75503343e-01 -1.87264889e-01
5.05756915e-01 9.15438533e-01 8.95016730e-01 -4.35302198e-01
1.68567681e+00 6.61674142e-01 -1.92937359e-01 -6.54057205e-01
-1.08979690e+00 -2.35799387e-01 -5.22048175e-01 1.23461951e-02
7.22046912e-01 -8.85370076e-01 -3.83197784e-01 6.66944265e-01
-7.86145210e-01 -1.82379618e-01 -5.47949731e-01 9.47558060e-02
-5.53523004e-01 5.76570272e-01 -4.56672966e-01 -7.40478337e-01
-5.41545153e-01 -1.17823565e+00 8.12755048e-01 1.56073764e-01
3.18305433e-01 -9.74008024e-01 -1.89290687e-01 3.96678925e-01
2.98571378e-01 5.17414570e-01 3.82364929e-01 -1.08315241e+00
-6.15647376e-01 -5.84369779e-01 2.04598680e-01 1.28385246e+00
3.22735041e-01 -3.26726884e-01 -1.37082839e+00 -6.05805337e-01
6.96895480e-01 -5.79054534e-01 9.12787795e-01 3.43112126e-02
1.41940057e+00 -6.86954796e-01 -3.26529108e-02 1.30282629e+00
1.33777153e+00 1.54252231e-01 9.08244789e-01 2.66554743e-01
7.85706699e-01 3.57617557e-01 4.29471463e-01 1.40140906e-01
-3.45543444e-01 4.06902507e-02 8.23247313e-01 -2.34412596e-01
-3.70116835e-03 -1.84502810e-01 7.25088418e-01 4.90138769e-01
4.64737236e-01 -3.11786085e-01 -4.10944551e-01 2.42892683e-01
-1.36274147e+00 -1.30185390e+00 2.39889234e-01 2.29612923e+00
5.69500685e-01 2.62183428e-01 -4.60402608e-01 3.07436854e-01
7.25566447e-01 2.76837558e-01 -5.86066723e-01 -4.19575572e-01
-4.19030011e-01 3.20601940e-01 3.72492015e-01 4.35040206e-01
-1.48082542e+00 7.20596910e-01 5.38951015e+00 1.01215518e+00
-1.14749157e+00 8.89341757e-02 6.29774630e-01 2.09997699e-01
-3.01037937e-01 2.11298093e-02 -4.24906582e-01 4.83792037e-01
7.57187963e-01 3.06849092e-01 6.99650645e-01 8.40947688e-01
-2.59284317e-01 5.09446561e-01 -1.19875646e+00 8.70864987e-01
3.72953683e-01 -1.23918080e+00 4.48321737e-02 2.51927346e-01
9.70611930e-01 5.75574674e-03 4.12144065e-01 4.76874709e-01
3.62284720e-01 -9.72255290e-01 5.28698206e-01 6.18938029e-01
9.57336664e-01 -8.26007605e-01 6.79373443e-01 5.96756399e-01
-8.51266623e-01 -6.00460768e-01 -3.99022430e-01 1.59424707e-01
-3.78259063e-01 2.35520467e-01 -3.23073059e-01 7.21150100e-01
6.58250988e-01 5.16659319e-01 -4.00190324e-01 6.02150559e-01
-4.12687719e-01 5.87285459e-01 -2.29619935e-01 5.76852858e-01
3.17027450e-01 -1.34943083e-01 9.80261922e-01 8.27449560e-01
1.57437064e-02 6.38495088e-02 3.94298971e-01 9.86947536e-01
-3.29446644e-01 -3.84239227e-01 -1.12373924e+00 1.16133064e-01
4.85026330e-01 1.12362754e+00 -1.88560024e-01 -5.69418557e-02
-2.45794028e-01 1.45728385e+00 2.19707131e-01 4.28601891e-01
-8.53658974e-01 -6.12714231e-01 9.28988516e-01 -3.47965926e-01
3.63102555e-01 6.53422326e-02 -9.91743878e-02 -1.30957747e+00
5.02891913e-02 -1.25579262e+00 4.38239276e-01 -4.11265641e-01
-1.87451839e+00 7.40621865e-01 -5.35214901e-01 -1.57693493e+00
-3.44338827e-02 -5.32961071e-01 -1.06974649e+00 9.96471465e-01
-1.52240920e+00 -1.28861904e+00 -2.31581315e-01 9.64015007e-01
4.97441798e-01 -5.04364014e-01 1.15027583e+00 -4.77346918e-03
-6.97595179e-01 1.01936519e+00 4.70052093e-01 5.52834690e-01
7.03869224e-01 -1.18774605e+00 3.82934690e-01 1.34696615e+00
3.11646104e-01 7.33883798e-01 3.51663709e-01 -5.62161446e-01
-1.60748994e+00 -1.49887133e+00 -2.95251627e-02 -6.08483970e-01
6.16612732e-01 -5.70778608e-01 -1.02138412e+00 6.04380250e-01
9.01746377e-02 6.72584295e-01 7.29730070e-01 -3.00227284e-01
-8.43782187e-01 -4.01901543e-01 -1.48310888e+00 3.56982350e-01
9.67648983e-01 -8.99025440e-01 -6.76752210e-01 4.54099715e-01
9.11820531e-01 -3.38413417e-01 -8.15056503e-01 2.07993671e-01
2.16958165e-01 -7.64627337e-01 1.17011225e+00 -5.32289088e-01
2.52262712e-01 -2.24391952e-01 -4.28137422e-01 -1.42724907e+00
-3.93653274e-01 -1.04341412e+00 -3.18526328e-01 1.42104793e+00
2.39351895e-02 -9.94671047e-01 4.44759756e-01 -1.74229816e-02
-3.77992302e-01 -5.51336884e-01 -9.61093664e-01 -1.21566188e+00
1.79483831e-01 -4.16732579e-01 5.03478408e-01 9.92015064e-01
-3.63709837e-01 -6.55506179e-02 -4.85000163e-01 8.28731239e-01
8.83240998e-01 2.17840020e-02 7.97185302e-01 -7.92959154e-01
-3.77368093e-01 -1.19149648e-02 -5.28567195e-01 -7.86531389e-01
5.22871494e-01 -9.40527141e-01 6.89473376e-02 -8.73064101e-01
-1.53563589e-01 -5.57249308e-01 -5.80038726e-01 4.42482889e-01
-1.69846907e-01 5.50885856e-01 1.32080764e-01 1.59023911e-01
-4.61072683e-01 5.83923519e-01 1.17960536e+00 -4.28452462e-01
3.29059660e-02 1.54080287e-01 -1.00022876e+00 5.91128111e-01
5.85559428e-01 -5.47162294e-01 -3.39981288e-01 -3.92787188e-01
-4.24726725e-01 2.76187193e-02 7.37276196e-01 -1.25311840e+00
1.53376162e-01 -7.95763806e-02 5.37065506e-01 -3.34451757e-02
2.27924570e-01 -1.21225345e+00 -1.68717042e-01 5.83407521e-01
-1.93749860e-01 -5.41794837e-01 4.09571193e-02 9.08279836e-01
-2.79767841e-01 -3.36536348e-01 1.01049685e+00 -9.41109136e-02
-5.79569340e-01 5.14665782e-01 -1.33488059e-01 -1.17893936e-03
1.35038650e+00 -4.40958917e-01 -2.98942149e-01 -3.24723184e-01
-8.74310613e-01 2.38888457e-01 4.83911753e-01 5.35278559e-01
9.71115232e-01 -1.27018452e+00 -6.94775701e-01 8.46932471e-01
1.80705473e-01 -1.42400235e-01 4.22047406e-01 2.39058658e-01
-2.34655812e-01 -1.30602047e-01 -1.26001000e-01 -4.60326135e-01
-1.11385202e+00 9.24817860e-01 5.91707230e-01 -1.19007990e-01
-5.51106155e-01 1.15856969e+00 3.79207432e-01 -6.77031398e-01
4.09602106e-01 2.09144741e-01 4.79901955e-02 -3.65395129e-01
7.09581196e-01 1.37078998e-04 1.91959888e-02 -6.54641628e-01
-1.31749466e-01 3.42286199e-01 -8.59173015e-02 3.48710030e-01
1.33815277e+00 -1.12834126e-01 -6.81650266e-02 -1.04081213e-01
1.54694998e+00 2.01568455e-01 -1.35421431e+00 -2.67229140e-01
-6.87547922e-01 -6.56268001e-01 1.39785847e-02 -7.19575644e-01
-1.01377189e+00 7.24760950e-01 9.73903537e-01 2.63116986e-01
1.44674599e+00 -3.49956930e-01 8.95310819e-01 4.55110610e-01
3.27288330e-01 -7.24698663e-01 2.37109572e-01 6.12702012e-01
1.09886730e+00 -1.41930997e+00 -5.50629616e-01 -5.01657486e-01
-3.66333008e-01 1.08482885e+00 6.63767457e-01 -4.88985807e-01
6.96119666e-01 3.51365685e-01 1.86107576e-01 -6.00474365e-02
-5.24785936e-01 -2.82129571e-02 2.86888421e-01 1.11554646e+00
-5.23384929e-01 -5.40070282e-03 5.51071405e-01 6.92294776e-01
-1.34406865e-01 -6.39055848e-01 2.95083314e-01 8.12636137e-01
-3.47427100e-01 -1.07667923e+00 -9.94721949e-01 3.95966291e-01
-3.43844086e-01 -3.81663553e-02 -3.52070838e-01 2.13465869e-01
4.71949428e-01 1.12500429e+00 -1.18017137e-01 -7.46681750e-01
1.79491043e-01 -1.93494037e-01 3.88019383e-01 -7.71086872e-01
-8.52624834e-01 -1.86244905e-01 -2.60094851e-01 -6.82262957e-01
1.41912565e-01 -3.96412611e-01 -7.90100098e-01 -1.35057285e-01
-6.48064673e-01 2.27687642e-01 4.54707503e-01 8.43735933e-01
1.46118522e-01 6.95930660e-01 1.57135856e+00 -6.93233848e-01
-1.21977091e+00 -6.39643073e-01 -5.21609366e-01 6.05127573e-01
7.07996786e-01 -2.83017904e-01 -7.79505491e-01 -2.48742830e-02] | [5.549810886383057, 7.979093551635742] |
dc909f77-2ed7-459e-a121-740fa9b4d4f9 | finite-element-model-updating-using-fish | 1308.2307 | null | http://arxiv.org/abs/1308.2307v1 | http://arxiv.org/pdf/1308.2307v1.pdf | Finite Element Model Updating Using Fish School Search Optimization Method | A recent nature inspired optimization algorithm, Fish School Search (FSS) is
applied to the finite element model (FEM) updating problem. This method is
tested on a GARTEUR SM-AG19 aeroplane structure. The results of this algorithm
are compared with two other metaheuristic algorithms; Genetic Algorithm (GA)
and Particle Swarm Optimization (PSO). It is observed that on average, the FSS
and PSO algorithms give more accurate results than the GA. A minor modification
to the FSS is proposed. This modification improves the performance of FSS on
the FEM updating problem which has a constrained search space. | ['T. Marwala', 'F. De Lima Neto', 'L. Mthembu', 'I. Boulkabeit'] | 2013-08-10 | null | null | null | null | ['nature-inspired-optimization-algorithm'] | ['computer-code'] | [ 1.39468715e-01 -5.44331789e-01 3.04336131e-01 3.93048048e-01
5.03213346e-01 -3.15684766e-01 2.85564572e-01 3.42723615e-02
-5.67066789e-01 1.15087366e+00 -1.18887231e-01 -1.16501085e-01
-8.94488752e-01 -1.10809624e+00 -1.52424663e-01 -1.00869083e+00
-2.52790675e-02 5.08978009e-01 5.87856054e-01 -9.16184783e-01
9.28346694e-01 7.82192528e-01 -2.04227877e+00 -6.17291868e-01
8.64481926e-01 6.44819796e-01 5.85364640e-01 9.36927855e-01
2.95110434e-01 3.90252173e-01 -9.06368315e-01 3.74075264e-01
1.55240819e-01 -2.62898028e-01 -6.70126438e-01 -3.55888307e-01
-5.58369100e-01 5.47597528e-01 2.16626182e-01 9.80009019e-01
7.38203228e-01 1.16628861e+00 6.73622012e-01 -1.12264884e+00
-1.59091428e-01 1.28444627e-01 -4.99966443e-01 6.47548854e-01
4.10011768e-01 -6.82585016e-02 2.84274399e-01 -4.92912650e-01
4.62764472e-01 1.37094235e+00 8.66997659e-01 2.87716299e-01
-4.22690183e-01 -2.76156873e-01 -4.50160652e-01 1.24448255e-01
-1.24208760e+00 2.58558005e-01 6.27604842e-01 9.47929844e-02
1.34534276e+00 8.79783571e-01 1.18488991e+00 -7.95965269e-02
8.80816400e-01 3.02800566e-01 8.93686652e-01 -5.97713888e-01
6.73022807e-01 -4.67110246e-01 7.87984952e-02 5.65937400e-01
9.15040731e-01 6.67076051e-01 -2.73833185e-01 -6.06982052e-01
4.84925628e-01 -3.68558526e-01 -2.03332365e-01 1.42274693e-01
-6.95893347e-01 1.06101441e+00 1.28923237e-01 7.52630353e-01
-5.51283956e-01 1.43473968e-01 2.08932031e-02 1.52099222e-01
1.53121442e-01 1.16815281e+00 -5.91119647e-01 -3.53005797e-01
-4.79552925e-01 6.79857194e-01 1.20905936e+00 2.27824926e-01
1.83299229e-01 6.44413412e-01 3.23482662e-01 5.09595931e-01
7.44724095e-01 8.59862447e-01 6.54693246e-01 -9.46892142e-01
-3.70084047e-01 5.35676003e-01 4.48464692e-01 -1.40078104e+00
-6.61245465e-01 -5.88144660e-01 -2.85951197e-01 6.35733545e-01
-7.16775060e-02 -5.74168146e-01 -5.49918115e-01 9.46328640e-01
8.49787951e-01 2.05269933e-01 4.32100773e-01 6.33224010e-01
8.53603601e-01 1.05711591e+00 -2.01374546e-01 -3.38989049e-01
1.04443741e+00 -1.36152673e+00 -7.53832698e-01 8.24212953e-02
2.77884215e-01 -1.05035210e+00 3.25081140e-01 3.42006117e-01
-1.00588524e+00 -3.40067059e-01 -1.27622247e+00 9.80193377e-01
-4.15675044e-01 -4.34804648e-01 5.97784162e-01 1.07137048e+00
-8.58999610e-01 1.08614767e+00 -8.85326266e-01 -5.27125597e-01
-2.72803098e-01 6.61989629e-01 1.54480889e-01 5.95761836e-01
-1.05023623e+00 1.22021997e+00 6.70985043e-01 1.19619541e-01
-1.02590695e-01 -5.67271948e-01 -7.98614025e-01 5.09499945e-02
3.54995847e-01 -7.06509471e-01 7.91464388e-01 -6.28370106e-01
-2.22374845e+00 1.15009740e-01 -1.16489626e-01 -3.62581909e-01
1.01873271e-01 3.78827959e-01 -3.56203139e-01 2.20999330e-01
-2.89222360e-01 -7.46723786e-02 5.59809446e-01 -1.35009181e+00
-8.30337167e-01 2.14833170e-01 -1.48323923e-01 4.50882405e-01
3.97430360e-02 3.19483995e-01 5.63868463e-01 -9.56833541e-01
1.17208295e-01 -1.00707781e+00 -4.99933660e-01 -5.23129404e-01
3.15951109e-01 -5.28885722e-01 8.77238154e-01 -2.84609675e-01
1.35210490e+00 -1.64995611e+00 3.95071238e-01 8.36122036e-01
-7.26746142e-01 5.03741086e-01 9.27487165e-02 8.54423463e-01
7.82611668e-02 -1.28571749e-01 -2.48724654e-01 2.25526646e-01
-1.17115431e-01 6.26583338e-01 3.54692847e-01 5.44350207e-01
-5.55549979e-01 4.78025764e-01 -8.76371920e-01 -1.86264902e-01
3.22789848e-01 -9.37207788e-03 -6.72196209e-01 -8.39055926e-02
7.53389671e-02 3.29923511e-01 -6.47851169e-01 9.82737124e-01
8.64018917e-01 2.59949584e-02 -8.68322849e-02 -7.48054683e-02
-7.62958288e-01 -5.13566136e-01 -1.67617834e+00 1.18393219e+00
-2.08998495e-03 8.34889561e-02 3.00668269e-01 -1.01199126e+00
1.07071710e+00 2.79668987e-01 7.86906421e-01 -2.78315216e-01
5.78825533e-01 3.83393735e-01 2.97321051e-01 -7.33855128e-01
5.41275501e-01 -9.61684585e-02 4.29379582e-01 5.02751648e-01
-5.37421405e-02 -6.47164524e-01 6.05772078e-01 -4.06066418e-01
7.45441258e-01 2.12141834e-02 4.60533887e-01 -1.27236068e+00
1.04593027e+00 4.57027942e-01 5.75727046e-01 5.96148431e-01
1.55894645e-02 8.00703093e-02 -2.75274038e-01 -6.14496112e-01
-7.86414444e-01 -4.46451038e-01 -3.32799047e-01 7.06993043e-01
7.25545526e-01 -3.26248705e-01 -5.89187503e-01 3.84018794e-02
1.12785250e-01 8.24808955e-01 -5.28563261e-01 -1.45302236e-01
-9.31737900e-01 -1.40404975e+00 1.37125164e-01 -3.11605185e-02
5.16518176e-01 -1.23691261e+00 -1.26846480e+00 6.28787160e-01
3.26727360e-01 -2.15871274e-01 1.63649440e-01 -1.90830484e-01
-9.96545017e-01 -1.30531251e+00 -4.51412022e-01 -8.07114601e-01
6.11792326e-01 6.84739053e-02 9.22374904e-01 1.01353705e+00
-3.45714480e-01 4.70121205e-01 -7.55513191e-01 -8.15152466e-01
-6.39913797e-01 -3.06463271e-01 1.97568208e-01 -4.27341044e-01
-9.91711244e-02 -6.33031130e-01 -4.54039931e-01 4.98905838e-01
-7.30221331e-01 -5.37384272e-01 6.84724674e-02 1.08924091e+00
4.68024760e-01 1.16639090e+00 3.43541354e-01 -2.30911553e-01
5.64085245e-01 -3.25769663e-01 -8.52103055e-01 1.55291528e-01
-6.20155990e-01 -6.02574460e-02 3.87556463e-01 -4.52923626e-01
-9.12992537e-01 -3.40762317e-01 -3.77156824e-01 -2.96958461e-02
9.84024815e-03 5.83963037e-01 4.01224703e-01 -1.22333312e+00
4.64070141e-01 2.01739505e-01 4.34966266e-01 -6.23947859e-01
-9.29031253e-01 4.93511230e-01 3.17615002e-01 -5.27432978e-01
8.07438254e-01 3.69556755e-01 6.44066155e-01 -1.05618453e+00
-2.05160946e-01 -3.50501776e-01 8.08840767e-02 -4.74080205e-01
6.29424334e-01 1.45197343e-02 -1.00307381e+00 8.38847637e-01
-6.90446556e-01 6.25195652e-02 -2.74942577e-01 6.64725780e-01
-3.41368675e-01 3.52567643e-01 -1.61589354e-01 -1.12291753e+00
-6.62984848e-01 -1.04927337e+00 5.25660634e-01 8.97853792e-01
-1.23754196e-01 -1.29396462e+00 4.06602949e-01 9.30103585e-02
7.52910674e-01 9.13930535e-01 4.96108681e-01 -3.95942599e-01
2.41500258e-01 -1.85705021e-01 8.46774399e-01 -1.47550836e-01
2.13051021e-01 5.41162670e-01 -7.05463439e-02 -7.40508080e-01
7.44053483e-01 4.79711115e-01 1.06686667e-01 6.31153882e-01
4.67160702e-01 -1.23870149e-01 -5.05170882e-01 5.83590627e-01
1.85995519e+00 1.10524833e+00 4.91982609e-01 1.40262270e+00
-2.22150579e-01 2.87460089e-01 1.08274329e+00 6.88002706e-01
-1.48978204e-01 3.15399528e-01 8.32112312e-01 2.20276490e-01
2.29507580e-01 3.43583941e-01 -1.04902141e-01 9.10510957e-01
-6.00320041e-01 -7.39702523e-01 -9.02961969e-01 1.74167380e-01
-1.65506053e+00 -8.44733179e-01 -2.85669148e-01 1.58785880e+00
2.02547148e-01 -1.73944935e-01 -1.54858053e-01 5.92019022e-01
9.26643789e-01 -8.65504369e-02 -1.82798266e-01 -6.30395591e-01
-3.80121410e-01 6.38677716e-01 5.33085823e-01 5.43817878e-01
-9.34581876e-01 3.91169727e-01 6.75199938e+00 9.98975456e-01
-1.03947949e+00 -7.92273059e-02 -3.10223132e-01 2.97736764e-01
-7.80737922e-02 -1.62839025e-01 -5.24222791e-01 8.47384155e-01
6.97816670e-01 -7.95690179e-01 7.33035326e-01 3.56534839e-01
4.17031907e-02 -7.05194950e-01 1.11200012e-01 6.74123645e-01
1.64227903e-01 -1.30664384e+00 -1.91227160e-02 -3.70844156e-01
1.04034615e+00 -1.45136312e-01 -2.34793410e-01 -3.38451684e-01
8.29237923e-02 -8.19299638e-01 6.51960373e-01 5.99453747e-01
-2.51881570e-01 -1.37390351e+00 1.32550704e+00 4.57070082e-01
-1.25844407e+00 -5.09345949e-01 -6.02308810e-01 -3.32179308e-01
5.23971796e-01 -1.21188194e-01 -1.12650685e-01 1.01203740e+00
1.19975972e+00 3.44808638e-01 -2.76535898e-01 1.82430124e+00
3.17396969e-01 3.61751527e-01 -9.27379787e-01 -7.77240336e-01
5.13004899e-01 -5.78733623e-01 1.59974635e+00 2.33925968e-01
7.46845126e-01 5.33385575e-01 -1.40881851e-01 3.08970690e-01
9.36042964e-01 3.21522743e-01 3.94880176e-02 3.62534449e-02
6.01519287e-01 8.74214530e-01 -1.08342659e+00 -1.29150838e-01
1.68184966e-01 -2.02363059e-02 -7.09384024e-01 -8.25241059e-02
-8.49693775e-01 -7.73726523e-01 4.79930699e-01 -1.12310201e-01
3.77950132e-01 3.99847999e-02 -4.89997026e-03 -4.51733232e-01
-9.29397464e-01 -7.30212152e-01 5.26191354e-01 -8.20603371e-01
-1.05275989e+00 6.69596672e-01 3.35881740e-01 -1.20931625e+00
-4.12624404e-02 -8.10090959e-01 -9.56827879e-01 5.28598547e-01
-1.03152144e+00 -5.49167752e-01 -1.20943449e-01 2.24844426e-01
5.16792595e-01 -7.31856525e-01 6.61234021e-01 4.60025147e-02
-3.92450362e-01 1.70089453e-02 5.98781765e-01 -7.61991322e-01
-1.58390164e-01 -8.73503149e-01 -7.56843388e-02 7.25801408e-01
-6.74790263e-01 3.40177357e-01 1.40038526e+00 -1.01641142e+00
-1.70155275e+00 -4.57611322e-01 3.92529905e-01 1.67280942e-01
5.84823668e-01 7.29484499e-01 -5.54280639e-01 7.09798634e-02
5.52665055e-01 -5.02071738e-01 4.29259360e-01 -7.52097905e-01
1.32085681e+00 3.32700074e-01 -1.71570325e+00 3.35348338e-01
6.87980473e-01 6.94577992e-01 -8.26574087e-01 3.28033417e-01
3.45560670e-01 -8.74958754e-01 -1.19419467e+00 8.56485903e-01
5.19055367e-01 -6.39527440e-01 1.15223193e+00 -2.84947485e-01
-3.59247863e-01 -7.68804014e-01 1.15741774e-01 -1.24483907e+00
-5.31496763e-01 -9.21566784e-01 5.26189245e-03 7.51236022e-01
8.31229612e-02 -1.27359331e+00 6.39909446e-01 1.30409181e-01
-2.72667944e-01 -8.54162574e-01 -1.21541977e+00 -1.01596940e+00
-5.76473102e-02 5.18911123e-01 1.25101781e+00 9.24278319e-01
-4.09557849e-01 -3.24397862e-01 -7.41117969e-02 3.74141365e-01
6.68610454e-01 2.08410010e-01 3.13708931e-01 -1.57025325e+00
-3.67756546e-01 -1.52537420e-01 -4.25334215e-01 1.98996916e-01
2.52591781e-02 -1.88983396e-01 -3.20990533e-01 -1.44265819e+00
-5.23972154e-01 -3.70626032e-01 -2.42573142e-01 -3.75972204e-02
-1.98655337e-01 2.36123845e-01 1.55927584e-01 1.69230811e-02
1.16110869e-01 6.11891985e-01 1.48644805e+00 9.98474434e-02
-2.31378943e-01 2.96625048e-01 -5.69529943e-02 9.01230216e-01
1.02440047e+00 -7.56995857e-01 -8.95858929e-02 -1.56111300e-01
3.73836994e-01 4.02292944e-02 -1.96125388e-01 -1.50008595e+00
4.91037041e-01 -5.04122138e-01 6.85684085e-02 -9.02795315e-01
8.46079439e-02 -1.15278697e+00 1.09227955e+00 1.49145699e+00
5.33370733e-01 7.07406759e-01 4.47677910e-01 4.71399844e-01
-2.78036207e-01 -1.18925190e+00 1.13045514e+00 -2.22872809e-01
-6.47972107e-01 -3.21347892e-01 -8.27147365e-01 -4.58091378e-01
1.27375436e+00 -8.21664095e-01 -2.35149011e-01 7.84796625e-02
-5.87185085e-01 2.64118373e-01 7.58582413e-01 4.98628765e-02
3.05896252e-01 -1.13159788e+00 -5.39637327e-01 2.83723712e-01
-7.91880369e-01 -2.43224949e-01 1.63258493e-01 8.13903332e-01
-1.78197312e+00 1.18850522e-01 -6.68233991e-01 -3.04137886e-01
-1.46980834e+00 2.03990698e-01 6.85893297e-01 -1.09726794e-01
-2.15977743e-01 9.71059144e-01 -8.55088711e-01 -5.54579973e-01
-4.56293643e-01 8.22010934e-02 -9.60104823e-01 -2.28104979e-01
2.50113219e-01 1.40627062e+00 -8.68182257e-02 -9.78933334e-01
-4.85131264e-01 1.08721089e+00 9.82930303e-01 1.72624171e-01
1.88067770e+00 -3.47669399e-03 -8.23984802e-01 -2.56603509e-01
5.08859873e-01 2.00869694e-01 -6.03456736e-01 5.77981472e-01
-1.98457792e-01 -7.25984693e-01 1.55079469e-01 -6.97255969e-01
-9.87507403e-01 -1.54685333e-01 6.91241562e-01 4.08757657e-01
1.32641256e+00 -7.61909246e-01 7.27772832e-01 6.05616450e-01
4.46246445e-01 -1.42033327e+00 -3.19501460e-01 8.70570421e-01
8.46800685e-01 -6.02602899e-01 5.58828235e-01 -3.63185346e-01
-9.53435898e-02 1.58533359e+00 6.16266012e-01 -7.21231461e-01
9.38043833e-01 7.16843843e-01 -3.52198422e-01 -2.70571530e-01
-5.87665737e-01 2.01926246e-01 2.35297918e-01 3.92855138e-01
-2.61842102e-01 -4.89045829e-01 -1.07124877e+00 3.19530815e-01
-4.51162487e-01 -1.47231489e-01 3.70836914e-01 1.73810351e+00
-9.74789143e-01 -1.06044316e+00 -9.55578864e-01 2.58920282e-01
-7.24036634e-01 2.39366367e-01 3.93889844e-02 1.03214908e+00
7.86966383e-02 1.07371283e+00 8.36881921e-02 -6.01973347e-02
4.02775943e-01 -6.19163692e-01 4.15258884e-01 -6.55637905e-02
-1.25844312e+00 3.83697152e-02 8.06413591e-02 -3.31490546e-01
-9.37114179e-01 -6.65066063e-01 -1.37406826e+00 -7.25662112e-01
-9.37609673e-01 1.32471871e+00 7.60753393e-01 8.49632263e-01
-9.03902873e-02 5.89839756e-01 7.24150121e-01 -1.12052965e+00
-4.08253759e-01 -6.77829862e-01 -9.01375353e-01 -3.33432734e-01
-1.67882368e-01 -1.31279910e+00 -8.02876353e-01 -5.61456680e-01] | [5.635908126831055, 3.4574685096740723] |
e0591688-29b1-45fb-89aa-de481babac3a | stacked-convolutional-and-recurrent-neural-1 | 1706.02047 | null | http://arxiv.org/abs/1706.02047v1 | http://arxiv.org/pdf/1706.02047v1.pdf | Stacked Convolutional and Recurrent Neural Networks for Bird Audio Detection | This paper studies the detection of bird calls in audio segments using
stacked convolutional and recurrent neural networks. Data augmentation by
blocks mixing and domain adaptation using a novel method of test mixing are
proposed and evaluated in regard to making the method robust to unseen data.
The contributions of two kinds of acoustic features (dominant frequency and log
mel-band energy) and their combinations are studied in the context of bird
audio detection. Our best achieved AUC measure on five cross-validations of the
development data is 95.5% and 88.1% on the unseen evaluation data. | ['Emre Çakır', 'Sharath Adavanne', 'Tuomas Virtanen', 'Konstantinos Drossos'] | 2017-06-07 | null | null | null | null | ['bird-audio-detection'] | ['audio'] | [ 3.41443568e-01 -2.84515768e-01 5.56282759e-01 -4.60766643e-01
-6.99443936e-01 -7.45044291e-01 2.66417086e-01 1.41014203e-01
-7.15712130e-01 5.10336876e-01 1.90270960e-01 -9.93227512e-02
-2.34793186e-01 -4.56385344e-01 -4.19780999e-01 -5.45679748e-01
-8.57756793e-01 -1.33878917e-01 2.36226708e-01 -2.22138360e-01
-1.33914396e-01 4.54904616e-01 -1.95010984e+00 6.24706924e-01
2.58667946e-01 1.10569906e+00 -2.07353875e-01 1.35174644e+00
4.74480838e-01 4.68979686e-01 -1.32978952e+00 -1.96638465e-01
2.33393490e-01 -4.62309211e-01 -7.31016874e-01 -5.01714200e-02
4.01243448e-01 -1.90649718e-01 6.55992404e-02 5.87920070e-01
9.25233543e-01 4.58720982e-01 3.85066211e-01 -1.00188589e+00
-1.99480534e-01 8.92524123e-01 -1.17352627e-01 7.37116039e-01
2.36074254e-01 -1.73377588e-01 1.02049541e+00 -6.97224081e-01
-1.40431330e-01 7.49229670e-01 1.00869429e+00 5.07734656e-01
-1.35534334e+00 -8.74201357e-01 -2.30760291e-01 -1.07975468e-01
-1.28799093e+00 -4.82559562e-01 8.08097064e-01 -4.57325071e-01
1.06659901e+00 6.35214329e-01 5.13574243e-01 8.60655487e-01
-3.27847868e-01 4.31684405e-01 1.00823939e+00 -6.36964500e-01
1.35548338e-01 3.48140538e-01 2.24036604e-01 2.55839348e-01
-2.74024844e-01 6.52965665e-01 -6.25485778e-01 -3.57244134e-01
4.28673893e-01 -5.54018676e-01 -1.17347322e-01 2.52869338e-01
-9.07008410e-01 7.95910537e-01 3.94098639e-01 5.60953975e-01
-2.18028858e-01 -7.20267594e-02 6.59271657e-01 4.72855747e-01
7.75942147e-01 6.51922166e-01 -5.97656786e-01 -6.47513568e-02
-1.04310548e+00 8.13933685e-02 9.44835126e-01 5.56289017e-01
2.13070050e-01 6.13512933e-01 1.17087327e-02 1.27349150e+00
2.65385330e-01 2.64257073e-01 8.67194951e-01 -3.87713313e-01
1.39049754e-01 2.05249920e-01 2.48457864e-03 -5.79030931e-01
-6.56768203e-01 -9.84561443e-01 -4.12760973e-01 -4.70438153e-02
4.49046604e-02 -5.92852056e-01 -9.80880260e-01 1.79132402e+00
2.10751995e-01 4.73456830e-01 1.56324774e-01 6.47209525e-01
1.38052368e+00 7.90798843e-01 2.40989719e-02 -1.38021737e-01
1.06477702e+00 -7.08586097e-01 -6.17689312e-01 -5.99745847e-02
1.19604677e-01 -1.01399648e+00 6.02085412e-01 5.02708972e-01
-9.73968506e-01 -1.12224674e+00 -1.35188377e+00 5.43221056e-01
-5.95233023e-01 3.80505443e-01 1.03281833e-01 9.76471901e-01
-1.06706250e+00 4.38362241e-01 -3.88370544e-01 -7.33044446e-02
-1.95472121e-01 8.19490790e-01 -2.52028227e-01 7.98127651e-01
-1.32191658e+00 3.91658992e-01 3.27324837e-01 3.85317713e-01
-1.23011100e+00 -5.57979405e-01 -6.47011161e-01 -5.79090267e-02
-2.20283717e-01 -1.38567880e-01 1.62662590e+00 -8.48667026e-01
-1.39419115e+00 8.96467865e-01 4.66272116e-01 -1.09508646e+00
2.97260374e-01 -4.58068728e-01 -1.09498620e+00 5.78331351e-02
-7.08877146e-02 4.95262176e-01 9.80315208e-01 -7.19178677e-01
-9.24551725e-01 -1.19573332e-01 -3.44068736e-01 -5.78327999e-02
-4.54665065e-01 2.36959010e-01 3.47770810e-01 -9.43669617e-01
-8.83249566e-02 -8.11068714e-01 -6.35712817e-02 -9.03848529e-01
-6.40004203e-02 -4.34183441e-02 8.59158099e-01 -7.82669842e-01
1.29628038e+00 -2.42004299e+00 -3.06393564e-01 1.69366196e-01
-8.22735876e-02 6.28842115e-01 -2.72207677e-01 4.19837892e-01
-3.29933643e-01 -5.83792990e-03 -1.13626309e-01 -2.43990093e-01
-9.92044508e-02 -1.20442219e-01 -5.97240090e-01 2.35685781e-01
3.49991947e-01 2.99186468e-01 -6.79244339e-01 -2.40546428e-02
2.25733086e-01 8.09994161e-01 -3.88340533e-01 6.24353707e-01
2.38088369e-01 1.34784058e-01 1.27322614e-01 3.87072295e-01
4.23750699e-01 6.07311130e-01 -2.23861128e-01 -6.19818680e-02
-3.44044089e-01 6.23393297e-01 -1.18042731e+00 1.17631090e+00
-4.93959308e-01 7.12493956e-01 1.41525358e-01 -6.82797909e-01
1.23306537e+00 6.48349702e-01 -4.77354303e-02 -1.54094100e-01
1.69847667e-01 3.66805106e-01 3.67904127e-01 -2.84976125e-01
4.08401072e-01 -2.32560232e-01 -1.91947982e-01 1.98878706e-01
6.11713052e-01 -4.93903495e-02 7.55095407e-02 -2.19641522e-01
9.47611094e-01 -1.89710930e-01 5.79448417e-02 -2.12701485e-01
4.72095579e-01 -3.24187666e-01 4.44404006e-01 8.08547258e-01
-2.41257444e-01 4.64124262e-01 -2.05564667e-02 -5.28510034e-01
-7.00522006e-01 -8.16476643e-01 -3.66873056e-01 1.67916775e+00
-5.61989009e-01 -3.48411560e-01 -8.62289429e-01 -3.56067747e-01
-1.37100533e-01 5.62606931e-01 -8.33898365e-01 -2.65720367e-01
-4.78291273e-01 -1.05319321e+00 1.23361874e+00 6.55364573e-01
4.29773360e-01 -1.31571949e+00 -8.57742429e-01 5.42400122e-01
1.49044514e-01 -9.42020297e-01 -1.69571236e-01 8.86867642e-01
-6.71162546e-01 -7.29350209e-01 -4.92543399e-01 -1.13555801e+00
-7.06319287e-02 6.97258860e-02 1.08701825e+00 -2.47232690e-01
-4.57861632e-01 4.55327630e-01 -6.15329802e-01 -6.64398432e-01
-5.29266655e-01 1.72324955e-01 2.90339917e-01 2.61303037e-01
3.52767229e-01 -6.92598522e-01 -3.76753807e-01 4.81876016e-01
-8.11178505e-01 -6.35369360e-01 3.85384470e-01 1.04147363e+00
3.25975239e-01 -3.75605375e-02 7.67657757e-01 -3.86051983e-01
8.52607667e-01 -1.53726131e-01 -5.73020756e-01 7.16691390e-02
-8.86818245e-02 -4.23838943e-01 3.21780264e-01 -8.06397855e-01
-7.26830184e-01 1.41346842e-01 -3.34270060e-01 -2.40598410e-01
-6.65836215e-01 2.12222800e-01 3.15067172e-01 -8.41708109e-02
9.20862675e-01 2.73971986e-02 -3.74012619e-01 -8.03470433e-01
-1.24848532e-02 6.93119228e-01 6.24256074e-01 -7.66543150e-02
4.85266656e-01 -7.56567568e-02 -3.53853256e-01 -1.28486669e+00
-5.66379964e-01 -6.67262793e-01 -7.52950668e-01 -3.92611861e-01
6.99350715e-01 -7.68043280e-01 -4.77516830e-01 5.07761180e-01
-8.82960320e-01 -1.30610485e-02 -4.83026654e-01 8.29937339e-01
-1.40269518e-01 -1.86084896e-01 -6.33526385e-01 -1.34328139e+00
-6.29893720e-01 -8.24945450e-01 8.19320560e-01 1.61456734e-01
-3.27160001e-01 -5.58947623e-01 5.92718899e-01 3.78645845e-02
4.71816957e-01 3.06166142e-01 3.80683184e-01 -1.33738196e+00
3.08626354e-01 -6.11948311e-01 4.51816499e-01 8.61193717e-01
4.77163643e-02 2.36474350e-01 -1.79132462e+00 -2.95919538e-01
1.97956041e-01 -3.18478256e-01 1.01314735e+00 5.54102540e-01
7.90085375e-01 -2.21906841e-01 1.27593577e-01 3.62597853e-01
7.49028623e-01 7.03384161e-01 1.84122808e-02 9.89241377e-02
1.20243646e-01 6.77456677e-01 3.54762435e-01 3.24822038e-01
-4.94183421e-01 4.30697322e-01 3.21341902e-01 -2.30277196e-01
-8.89155641e-02 -2.80966848e-01 3.01071137e-01 9.46378350e-01
7.28766993e-02 -1.95023134e-01 -9.27806675e-01 7.37918198e-01
-1.24073374e+00 -9.25561786e-01 1.99921487e-04 2.19787621e+00
4.39778477e-01 2.43600667e-01 7.56713390e-01 9.76972222e-01
1.04770839e+00 -6.54613459e-03 -1.44439295e-01 -7.06888258e-01
-1.36984482e-01 5.52077830e-01 2.76542664e-01 3.90823781e-01
-1.71187317e+00 4.81578737e-01 7.51071262e+00 5.41145146e-01
-9.83182609e-01 1.06326468e-01 5.46963036e-01 -3.75010848e-01
5.10281444e-01 -5.51224768e-01 -7.80499637e-01 1.70792490e-01
1.51167202e+00 3.45452309e-01 4.00149941e-01 8.42941940e-01
-1.87188670e-01 2.08938062e-01 -8.07978809e-01 7.51258492e-01
1.12782612e-01 -7.61203647e-01 -3.39724779e-01 -1.38711646e-01
5.69374681e-01 3.20514500e-01 5.36796451e-01 4.84226733e-01
2.99512818e-02 -7.92235136e-01 7.41277933e-01 4.03241068e-02
5.75494111e-01 -1.08162653e+00 1.02775204e+00 1.52407244e-01
-1.23644459e+00 -2.29476526e-01 -5.84383830e-02 -1.76448211e-01
-2.68457323e-01 2.01674178e-01 -1.56039560e+00 1.07662722e-01
1.12856400e+00 2.42964238e-01 -8.28740656e-01 1.40101123e+00
1.76193774e-01 1.20826018e+00 -5.16816318e-01 -2.96782315e-01
4.18096781e-01 4.03526604e-01 6.45382285e-01 1.72722208e+00
1.25034958e-01 -2.69538105e-01 -1.21850461e-01 4.76622105e-01
1.30718481e-02 2.40164384e-01 -5.33243537e-01 -1.84071362e-01
3.02003771e-01 1.26830637e+00 -7.59092033e-01 5.28568886e-02
2.30258573e-02 3.69131744e-01 -1.25621751e-01 1.00428186e-01
-6.94181144e-01 -8.27093780e-01 3.58434647e-01 -1.88257203e-01
4.15985733e-01 2.74415970e-01 1.46510243e-01 -4.38652873e-01
-1.82274699e-01 -7.31857896e-01 7.01236486e-01 -2.68780679e-01
-1.34424877e+00 1.29300404e+00 1.23864330e-01 -1.47232866e+00
-5.28528988e-01 -4.18422997e-01 -7.20674813e-01 7.15440691e-01
-9.81767595e-01 -1.02066302e+00 -8.90178084e-02 2.55749404e-01
4.44785982e-01 -7.04424202e-01 1.10317194e+00 5.08375525e-01
-4.18274969e-01 7.96295404e-01 4.73356731e-02 3.61585349e-01
4.86126512e-01 -1.18500793e+00 6.93091869e-01 6.89784944e-01
7.70471454e-01 3.39769840e-01 7.70414472e-01 -2.24371731e-01
-4.82199162e-01 -1.06590855e+00 5.72638988e-01 -2.63154507e-01
4.87922579e-01 -5.94647408e-01 -9.49637234e-01 3.51236761e-01
1.62911966e-01 -3.88294794e-02 1.39654887e+00 2.98984617e-01
-2.45770037e-01 -3.01973552e-01 -1.01751459e+00 -5.76739758e-02
3.62584591e-01 -7.58167505e-01 -8.52508485e-01 -4.02743742e-02
6.67184651e-01 -2.49188557e-01 -6.85025692e-01 9.32779729e-01
7.64041424e-01 -9.61877227e-01 8.81066561e-01 -7.02471375e-01
-3.69628578e-01 -3.89383405e-01 -2.76939332e-01 -1.25577605e+00
-1.83728337e-01 -8.23801696e-01 1.48603186e-01 1.48506999e+00
7.16283619e-01 -3.18926066e-01 3.31463546e-01 -1.28631502e-01
-1.73056886e-01 -2.76247263e-01 -1.14375234e+00 -6.29711628e-01
-3.89455527e-01 -7.88214266e-01 5.63706517e-01 5.55853724e-01
-2.21679941e-01 5.57138801e-01 -6.21106505e-01 2.45770827e-01
1.12936080e-01 -8.10680464e-02 5.92395008e-01 -1.28355181e+00
-3.51222306e-01 -2.82714128e-01 -7.70713449e-01 -3.07774156e-01
-1.12121604e-01 -5.28651595e-01 2.20229179e-01 -7.56980062e-01
-5.36661744e-01 1.35688037e-02 -9.71686482e-01 1.52227953e-01
2.38910973e-01 8.47092092e-01 -1.01143904e-01 -2.38669619e-01
-3.31018925e-01 2.83016384e-01 2.44358256e-01 -5.78882210e-02
-5.58991134e-01 7.11801171e-01 -9.40762386e-02 6.43012881e-01
1.04097128e+00 -5.78813255e-01 -2.78078794e-01 -2.01116353e-01
-1.30709186e-01 2.47649737e-02 5.84587038e-01 -1.49184096e+00
-1.12313114e-01 7.92167723e-01 6.67471111e-01 -7.79491782e-01
6.32395446e-01 -5.51300406e-01 1.40573919e-01 5.75381219e-01
-9.10463572e-01 1.43881381e-01 7.00168490e-01 5.77198565e-01
-5.29923677e-01 -4.09537077e-01 9.21231747e-01 1.29785776e-01
-4.72608149e-01 -3.71582568e-01 -5.24688721e-01 -9.79949236e-02
7.76900828e-01 -7.39962608e-02 1.34822726e-01 -4.38881248e-01
-1.20056474e+00 -2.53479093e-01 -4.60315943e-01 5.92933714e-01
5.65203488e-01 -1.11771262e+00 -8.77439082e-01 8.10441077e-01
-7.52816573e-02 -7.85165250e-01 3.61015052e-01 4.41194952e-01
-3.45618308e-01 3.54754448e-01 -2.97410458e-01 -5.90362787e-01
-1.86291075e+00 2.94197857e-01 6.96103454e-01 8.21749344e-02
2.83823516e-02 1.48619354e+00 4.89517599e-02 -3.55296731e-01
7.34359205e-01 -5.22953808e-01 -4.90981072e-01 4.25018281e-01
6.60593629e-01 5.27180612e-01 4.47334021e-01 -8.31870556e-01
-4.12785679e-01 1.65304631e-01 -5.85540682e-02 -4.10574347e-01
1.42646348e+00 3.34065765e-01 9.07853171e-02 8.51770580e-01
1.20112145e+00 1.24766612e-02 -6.49713814e-01 -1.69914499e-01
4.46036272e-03 -1.55372858e-01 1.52063414e-01 -1.05231571e+00
-6.75551414e-01 1.29080510e+00 1.37809086e+00 1.19033766e+00
1.22014403e+00 -2.51721293e-01 4.55325156e-01 1.74076244e-01
-2.65593529e-01 -1.08573604e+00 -2.30002269e-01 3.19125295e-01
9.37407613e-01 -9.36273098e-01 -3.26724082e-01 1.05533428e-01
-5.72751999e-01 1.08588195e+00 5.53368330e-01 -2.46026203e-01
8.66813600e-01 2.42793605e-01 3.50751787e-01 4.83661070e-02
-9.04642761e-01 -3.64121139e-01 7.00209320e-01 6.74899995e-01
7.29703724e-01 1.13632180e-01 -4.26643202e-03 6.44401610e-01
-6.43801749e-01 -5.45694411e-01 -2.18568984e-02 8.25621903e-01
-3.48156929e-01 -6.12272739e-01 -6.20752513e-01 4.78451401e-01
-9.91789699e-01 7.85853863e-02 -8.48688006e-01 7.43017554e-01
3.54097784e-01 1.34629309e+00 9.19754580e-02 -9.43889558e-01
4.63402659e-01 3.14041346e-01 -4.26106006e-02 -6.14735365e-01
-1.50118303e+00 7.01599121e-01 2.33876705e-01 2.48889089e-01
-6.81735933e-01 -4.45192009e-01 -6.76017106e-01 3.64131868e-01
-8.05110514e-01 8.62833321e-01 7.02404797e-01 4.32206064e-01
1.73702482e-02 8.53006005e-01 8.48216176e-01 -7.91340590e-01
-3.85303319e-01 -1.41330326e+00 -6.97636008e-01 -6.14259951e-03
8.95836294e-01 -4.33881342e-01 -6.30445540e-01 -9.18341894e-03] | [15.226616859436035, 5.311604022979736] |
e63b0780-e6e0-420e-aa23-ab016b6888f8 | localized-trajectories-for-2d-and-3d-action | 1904.05244 | null | http://arxiv.org/abs/1904.05244v1 | http://arxiv.org/pdf/1904.05244v1.pdf | Localized Trajectories for 2D and 3D Action Recognition | The Dense Trajectories concept is one of the most successful approaches in
action recognition, suitable for scenarios involving a significant amount of
motion. However, due to noise and background motion, many generated
trajectories are irrelevant to the actual human activity and can potentially
lead to performance degradation. In this paper, we propose Localized
Trajectories as an improved version of Dense Trajectories where motion
trajectories are clustered around human body joints provided by RGB-D cameras
and then encoded by local Bag-of-Words. As a result, the Localized Trajectories
concept provides a more discriminative representation of actions as compared to
Dense Trajectories. Moreover, we generalize Localized Trajectories to 3D by
using the modalities offered by RGB-D cameras. One of the main advantages of
using RGB-D data to generate trajectories is that they include radial
displacements that are perpendicular to the image plane. Extensive experiments
and analysis are carried out on five different datasets. | ['Björn Ottersten', 'Enjie Ghorbel', 'Konstantinos Papadopoulos', 'Girum Demisse', 'Djamila Aouada', 'Michel Antunes'] | 2019-04-10 | null | null | null | null | ['3d-human-action-recognition'] | ['computer-vision'] | [ 4.15687114e-02 -4.03208464e-01 -3.08000356e-01 -8.79229456e-02
-3.85270596e-01 -4.88008022e-01 7.91349947e-01 -8.02949145e-02
-3.10426205e-01 6.28464222e-01 6.34450853e-01 3.51133704e-01
-2.07114115e-01 -6.45951033e-01 -4.91755515e-01 -8.65160406e-01
2.40532476e-02 1.90079466e-01 5.37233174e-01 -7.30184019e-02
3.01369950e-02 8.34176779e-01 -1.65218854e+00 3.17733735e-01
4.08495873e-01 6.96137607e-01 -3.44587048e-03 3.99625778e-01
-7.76899382e-02 9.15276587e-01 -4.84190851e-01 -1.67814165e-01
3.19163501e-01 -6.13656342e-01 -4.71178651e-01 5.96167386e-01
1.79221615e-01 -4.87078875e-01 -7.46303141e-01 7.98264265e-01
2.36901060e-01 5.00184715e-01 4.45217252e-01 -1.24882960e+00
-2.20174447e-01 7.29102790e-02 -4.65449840e-01 7.89510086e-02
8.91594291e-01 2.27204710e-01 5.95306337e-01 -8.28084588e-01
8.08886468e-01 1.24531734e+00 5.61998665e-01 6.20428562e-01
-8.93215954e-01 -1.40855327e-01 2.06359997e-01 4.46697593e-01
-1.43293667e+00 -3.67763638e-01 8.02287161e-01 -3.21673989e-01
6.92302227e-01 3.25645238e-01 9.26525056e-01 1.42170537e+00
1.32711321e-01 1.09062874e+00 7.47322977e-01 -2.35281810e-01
3.33418250e-01 -2.69234926e-01 -3.31217766e-01 4.15497094e-01
6.88033625e-02 1.51966931e-02 -4.14398789e-01 -1.72651559e-01
9.11835253e-01 7.39489496e-01 -4.83158022e-01 -8.76469612e-01
-1.48647535e+00 5.97201467e-01 4.34987158e-01 4.45564568e-01
-5.60672700e-01 3.21971625e-01 2.52218217e-01 -1.56972870e-01
2.54806846e-01 -3.21710221e-02 4.98249792e-02 -5.99431753e-01
-8.24656248e-01 4.90474433e-01 2.23084733e-01 9.44179893e-01
5.68706155e-01 -8.41066837e-02 -2.58108526e-01 3.72966468e-01
2.73747176e-01 5.55487096e-01 6.65094912e-01 -1.08234179e+00
5.86519301e-01 9.04620171e-01 4.41134959e-01 -1.46155965e+00
-3.17422420e-01 -9.43767354e-02 -8.56085002e-01 -1.06979609e-01
5.88002026e-01 1.45504817e-01 -6.81265593e-01 1.29542208e+00
4.79967117e-01 4.12769765e-01 -5.18906750e-02 1.16084051e+00
3.08251113e-01 6.64152205e-01 -6.52811080e-02 -9.25960541e-02
9.53138173e-01 -9.31139410e-01 -8.62071455e-01 -3.73423696e-02
6.94235265e-01 -4.87581849e-01 8.79267752e-01 2.45066643e-01
-7.86203206e-01 -6.69320822e-01 -5.51964223e-01 1.07527189e-01
-2.94453114e-01 2.60944009e-01 4.25123662e-01 4.89207566e-01
-6.68528616e-01 5.53837895e-01 -1.21717381e+00 -4.72401112e-01
2.24098772e-01 2.66550258e-02 -6.78385675e-01 -2.92224377e-01
-8.28001797e-01 5.51136971e-01 3.62295598e-01 1.49969324e-01
-6.71799302e-01 -1.70090934e-03 -8.81712198e-01 -3.06007862e-01
4.82615381e-01 -3.23732734e-01 1.06008828e+00 -7.04956293e-01
-1.35259485e+00 4.01007056e-01 -3.08855027e-01 -2.74204791e-01
6.82438076e-01 -4.62963969e-01 -2.88537741e-01 6.02140248e-01
-1.04008175e-01 3.42390031e-01 8.58397841e-01 -9.71110404e-01
-6.21577084e-01 -5.48629165e-01 3.24204285e-03 7.89172277e-02
-3.32398891e-01 -2.03199923e-01 -5.90117812e-01 -7.01456547e-01
2.02010676e-01 -1.22457707e+00 -4.28375751e-01 -3.01100742e-02
-2.02174425e-01 -1.04133971e-01 1.04981840e+00 -6.09061003e-01
1.35477245e+00 -2.19523597e+00 4.87788707e-01 2.29536667e-01
-6.24554045e-02 3.49441975e-01 5.86542906e-03 6.84689641e-01
2.39580929e-01 -1.73071921e-01 -1.05141930e-01 -4.36888725e-01
-2.02678114e-01 5.25198162e-01 -1.24404222e-01 7.13862360e-01
9.54167396e-02 8.54088843e-01 -1.20061982e+00 -5.51753819e-01
5.30584276e-01 4.22367960e-01 -2.50955850e-01 2.26348154e-02
-1.40816912e-01 9.19603109e-01 -7.70073295e-01 6.75447226e-01
3.59369785e-01 1.47348549e-02 1.29902497e-01 -1.29405081e-01
-5.84190972e-02 -8.70187581e-02 -1.23833954e+00 2.08377600e+00
-4.44230586e-02 3.07009727e-01 -3.11700851e-01 -9.08364236e-01
7.57950187e-01 4.59468663e-01 1.02319789e+00 -5.15809298e-01
6.86890557e-02 -2.04607807e-02 -4.49855030e-01 -5.30986845e-01
5.07412970e-01 -5.06095309e-03 -2.35342786e-01 2.95299321e-01
-2.23531306e-01 2.07174718e-01 1.09329194e-01 3.04390024e-02
1.36735094e+00 5.15068114e-01 3.19768161e-01 4.75077540e-01
5.74501932e-01 1.43078282e-01 7.29602873e-01 4.95696843e-01
-3.39772612e-01 8.45513105e-01 1.54853016e-01 -5.37561536e-01
-1.00803173e+00 -1.06146348e+00 3.00642610e-01 2.74218529e-01
3.07427794e-01 -7.21336305e-01 -4.29028958e-01 -7.74635494e-01
-7.47329891e-02 3.89943600e-01 -4.29688066e-01 -1.13612965e-01
-8.33490014e-01 -2.54106373e-01 4.80505347e-01 7.79396772e-01
5.77098310e-01 -8.15504611e-01 -1.04713404e+00 2.29405791e-01
-4.15500194e-01 -1.39118779e+00 -3.98726672e-01 -5.14839172e-01
-1.25920212e+00 -1.10617638e+00 -1.07922506e+00 -2.12001026e-01
6.81091309e-01 6.97702348e-01 7.21168995e-01 -1.74185336e-01
-3.08123767e-01 8.18787813e-01 -9.43220139e-01 1.63673103e-01
-1.64263666e-01 -3.89970303e-01 7.41276070e-02 4.13124532e-01
4.99529004e-01 -3.75214845e-01 -7.71988928e-01 4.76992518e-01
-9.14641619e-01 -1.74704656e-01 5.32103717e-01 5.70497334e-01
7.08053827e-01 1.67877004e-01 -1.48755638e-02 -4.21109796e-01
1.69275209e-01 -5.50267339e-01 -1.78760111e-01 3.68932113e-02
1.38621584e-01 -2.09904283e-01 5.41899085e-01 -4.75594521e-01
-1.09963417e+00 5.83617568e-01 7.17933103e-02 -9.10748065e-01
-5.98273695e-01 1.28886446e-01 -1.91794619e-01 1.67120978e-01
4.01392639e-01 3.35191309e-01 4.40584384e-02 -6.79968774e-01
5.38865745e-01 4.34777826e-01 1.55453220e-01 -2.65834808e-01
6.62509680e-01 8.45681787e-01 2.03034312e-01 -1.06170547e+00
-4.25586790e-01 -7.85036147e-01 -9.09802496e-01 -6.25065267e-01
1.00351906e+00 -7.98201501e-01 -5.95562160e-01 5.28391421e-01
-1.01060140e+00 -2.97656953e-02 -3.95760506e-01 8.98662627e-01
-7.60311127e-01 8.00299883e-01 -5.69851160e-01 -9.82704401e-01
2.82408327e-01 -1.03762233e+00 1.23543513e+00 1.50525570e-01
-2.76551574e-01 -8.91325772e-01 1.35090187e-01 2.19014987e-01
-7.23857135e-02 6.86506391e-01 4.91707623e-01 -1.49651811e-01
-5.55132389e-01 -6.76231384e-01 2.05591977e-01 2.48372301e-01
4.13855106e-01 -8.35989490e-02 -5.31680703e-01 -1.40225977e-01
-2.05528922e-02 -1.29683122e-01 5.60484350e-01 2.20144704e-01
9.87236619e-01 -1.94672629e-01 -5.70329368e-01 2.61562735e-01
1.23548019e+00 2.17730656e-01 8.76886606e-01 1.40364543e-01
9.30351794e-01 5.64122438e-01 9.89038289e-01 5.88809907e-01
1.67503968e-01 9.85657930e-01 3.47019076e-01 2.93503940e-01
-6.17824160e-02 -4.87761766e-01 5.61015189e-01 6.26000226e-01
-6.15835249e-01 -1.68488041e-01 -8.21436942e-01 6.38419986e-01
-2.19390464e+00 -1.27516544e+00 -4.05992389e-01 2.40486884e+00
3.63313556e-01 -7.25258961e-02 6.41378284e-01 4.13886517e-01
5.68527579e-01 2.44320914e-01 -2.97688246e-01 9.68450829e-02
-4.36690040e-02 -3.68327014e-02 4.53868270e-01 1.71865243e-02
-1.01869631e+00 7.41665065e-01 5.79644203e+00 7.41596043e-01
-8.66137564e-01 5.96043579e-02 -1.04894200e-02 -3.02133232e-01
-4.34700996e-02 -6.17196821e-02 -6.32213235e-01 5.93397200e-01
7.26128638e-01 3.41221750e-01 -1.97255626e-01 8.20353091e-01
4.31459337e-01 -2.63655782e-01 -9.89520073e-01 1.13284159e+00
7.72769190e-03 -9.81563807e-01 2.72425245e-02 1.82788029e-01
6.98774338e-01 -3.48503083e-01 -1.74464211e-01 -2.27227174e-02
1.04244472e-02 -7.25832224e-01 7.94053972e-01 8.04895222e-01
3.91122252e-01 -7.20183790e-01 5.99307120e-01 6.17376685e-01
-1.21810484e+00 -1.58736378e-01 -5.37528574e-01 -1.11898609e-01
4.49114114e-01 3.79456431e-01 -5.30793250e-01 7.01974332e-01
6.83547735e-01 1.10698879e+00 -3.78147095e-01 1.14671528e+00
-3.03977817e-01 3.49005848e-01 -2.62057096e-01 -1.79408461e-01
4.52830315e-01 -3.91539216e-01 7.30060577e-01 1.00030529e+00
6.01288080e-01 2.72238374e-01 2.14094207e-01 5.47041118e-01
4.74184394e-01 -9.68097672e-02 -1.03795934e+00 -1.88048497e-01
2.12167174e-01 9.12025452e-01 -6.88138723e-01 -1.94109261e-01
-6.54899538e-01 1.31701732e+00 -9.55935642e-02 3.87876004e-01
-7.15451598e-01 -1.19525619e-01 6.45647585e-01 2.22764686e-01
3.53293806e-01 -8.43093693e-01 3.68576229e-01 -1.24051273e+00
1.69319957e-01 -6.71566904e-01 9.40019414e-02 -5.75953782e-01
-1.00253868e+00 2.21565902e-01 2.92475998e-01 -1.91348696e+00
-4.57143098e-01 -5.85648060e-01 -2.15920970e-01 3.37311864e-01
-9.93667543e-01 -1.16743064e+00 -5.13861418e-01 9.58568633e-01
6.72261178e-01 9.69608054e-02 8.29227448e-01 2.61534870e-01
-4.08471465e-01 1.05789736e-01 2.51546472e-01 2.46875510e-01
3.96673381e-01 -9.50429916e-01 2.37245545e-01 7.98999071e-01
5.11675239e-01 4.27758813e-01 5.46632171e-01 -7.05468357e-01
-1.61126566e+00 -1.05226696e+00 6.18634760e-01 -5.62852919e-01
4.40427035e-01 -1.83957487e-01 -6.29189312e-01 6.72314703e-01
-2.98366070e-01 1.20890543e-01 7.77028322e-01 -3.80422294e-01
7.82499015e-02 -2.86494810e-02 -9.16381299e-01 6.51739657e-01
1.32549620e+00 -4.09333676e-01 -6.16051376e-01 1.95591673e-01
2.48911038e-01 -4.40016717e-01 -9.03474271e-01 1.88512608e-01
5.20129442e-01 -1.22680128e+00 1.15113056e+00 -5.31000733e-01
2.94128299e-01 -4.64418769e-01 -2.19134375e-01 -1.18821394e+00
-1.05620973e-01 -3.30365211e-01 -4.91968125e-01 8.60613048e-01
-2.58866519e-01 -2.12671429e-01 1.03286028e+00 3.56392503e-01
-4.87557985e-02 -5.46282232e-01 -9.47180927e-01 -1.02001643e+00
-2.81836718e-01 -5.16019464e-01 4.29769248e-01 5.05852103e-01
1.47182355e-03 -2.19830975e-01 -6.41090810e-01 -1.00918531e-01
6.79360032e-01 2.08353043e-01 1.03965235e+00 -8.74044955e-01
-1.78359225e-01 -3.14819068e-02 -1.02811992e+00 -1.53953481e+00
3.90689149e-02 -3.29242527e-01 1.81898683e-01 -1.35126090e+00
-1.55331552e-01 -3.37218881e-01 -1.04945460e-02 2.95285434e-01
6.96267337e-02 2.92818040e-01 3.51580590e-01 5.17968118e-01
-6.30427182e-01 8.89605284e-01 1.29065251e+00 -2.11345088e-02
-1.54892206e-01 3.04146379e-01 1.24388091e-01 9.14738476e-01
7.07333565e-01 -3.22112739e-01 -5.82239568e-01 -3.45006734e-01
-5.85339293e-02 3.40236396e-01 6.23342097e-01 -1.18616307e+00
2.24101797e-01 -3.24993908e-01 4.57730114e-01 -7.67670035e-01
8.68861735e-01 -1.09397078e+00 3.02580118e-01 4.03058797e-01
-6.96154088e-02 1.74729638e-02 -5.90365753e-02 1.05794597e+00
-5.25247395e-01 -5.27163036e-02 3.65941823e-01 -5.01977921e-01
-9.25438643e-01 3.31787080e-01 -6.36009336e-01 -3.21908861e-01
1.48924518e+00 -6.80034578e-01 1.96098328e-01 -5.35844624e-01
-7.87049353e-01 -1.28210932e-01 7.02011108e-01 5.49179554e-01
9.71986949e-01 -1.68897545e+00 -3.27934057e-01 1.04480930e-01
2.34521165e-01 -9.30660665e-02 3.86149853e-01 1.03282809e+00
-4.91551578e-01 7.59198964e-01 -3.09600562e-01 -7.99873710e-01
-1.30396044e+00 6.64713204e-01 1.05541144e-02 2.13085301e-02
-1.13034689e+00 5.05195856e-01 -1.49203092e-01 -9.37812626e-02
2.37586588e-01 -4.68215436e-01 -1.81372970e-01 -5.78170642e-02
6.38937891e-01 7.31806934e-01 -1.40132815e-01 -1.16275072e+00
-5.14290273e-01 7.91973531e-01 4.41840470e-01 -1.50267869e-01
1.14492548e+00 -2.67086506e-01 2.89895326e-01 6.77091658e-01
1.01150656e+00 1.02576539e-01 -1.39190209e+00 -2.24014580e-01
1.17419355e-01 -1.00790036e+00 -3.03308785e-01 -1.71626911e-01
-1.04632664e+00 1.00009739e+00 4.16092962e-01 7.03564957e-02
1.12115157e+00 -1.25144005e-01 1.01546717e+00 2.50140607e-01
8.16086531e-01 -9.87746000e-01 4.22006637e-01 1.72797665e-01
7.16431677e-01 -1.06161296e+00 -1.83797423e-02 -3.54691863e-01
-8.13361764e-01 1.11032021e+00 3.83975297e-01 -1.42678276e-01
3.98195952e-01 -2.06072077e-01 -1.79786682e-01 -5.83936088e-02
-3.10559630e-01 -3.67144078e-01 2.04506963e-01 7.56399393e-01
1.47208989e-01 9.18026362e-03 -2.71741748e-01 2.08730370e-01
3.79009932e-01 1.37672722e-01 4.20238376e-01 1.30770612e+00
-2.30401993e-01 -1.19456875e+00 -5.84846556e-01 7.15956613e-02
-1.90242544e-01 6.81878984e-01 -3.79789084e-01 7.94957578e-01
1.85368106e-01 1.07238531e+00 2.64309458e-02 -4.74815935e-01
5.70185423e-01 6.18619360e-02 7.61243761e-01 -4.72391039e-01
-2.16001287e-01 1.48977444e-01 -1.44283608e-01 -1.10238993e+00
-7.89928973e-01 -1.03365898e+00 -1.37386799e+00 -2.13858172e-01
-1.01598307e-01 6.21153153e-02 4.38513666e-01 9.42707062e-01
4.73493129e-01 2.52139926e-01 4.55331981e-01 -1.00606930e+00
-5.03633857e-01 -8.21920753e-01 -7.63807893e-01 8.50423098e-01
2.23171413e-01 -9.95455563e-01 -2.07917124e-01 2.47364610e-01] | [7.9792094230651855, 0.36264243721961975] |
dcc35a6d-615d-44d5-a096-8afd949ca32d | leveraging-hidden-positives-for-unsupervised | 2303.15014 | null | https://arxiv.org/abs/2303.15014v1 | https://arxiv.org/pdf/2303.15014v1.pdf | Leveraging Hidden Positives for Unsupervised Semantic Segmentation | Dramatic demand for manpower to label pixel-level annotations triggered the advent of unsupervised semantic segmentation. Although the recent work employing the vision transformer (ViT) backbone shows exceptional performance, there is still a lack of consideration for task-specific training guidance and local semantic consistency. To tackle these issues, we leverage contrastive learning by excavating hidden positives to learn rich semantic relationships and ensure semantic consistency in local regions. Specifically, we first discover two types of global hidden positives, task-agnostic and task-specific ones for each anchor based on the feature similarities defined by a fixed pre-trained backbone and a segmentation head-in-training, respectively. A gradual increase in the contribution of the latter induces the model to capture task-specific semantic features. In addition, we introduce a gradient propagation strategy to learn semantic consistency between adjacent patches, under the inherent premise that nearby patches are highly likely to possess the same semantics. Specifically, we add the loss propagating to local hidden positives, semantically similar nearby patches, in proportion to the predefined similarity scores. With these training schemes, our proposed method achieves new state-of-the-art (SOTA) results in COCO-stuff, Cityscapes, and Potsdam-3 datasets. Our code is available at: https://github.com/hynnsk/HP. | ['Jae-Pil Heo', 'SuBeen Lee', 'WonJun Moon', 'Hyun Seok Seong'] | 2023-03-27 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Seong_Leveraging_Hidden_Positives_for_Unsupervised_Semantic_Segmentation_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Seong_Leveraging_Hidden_Positives_for_Unsupervised_Semantic_Segmentation_CVPR_2023_paper.pdf | cvpr-2023-1 | ['unsupervised-semantic-segmentation'] | ['computer-vision'] | [ 3.32379520e-01 3.83034229e-01 -3.07974279e-01 -5.84763348e-01
-8.84442985e-01 -2.72025555e-01 4.67311680e-01 2.03893200e-01
-3.09097618e-01 5.01531243e-01 2.26512626e-02 2.28576630e-01
-9.22552943e-02 -6.82753861e-01 -8.26618433e-01 -7.49936819e-01
1.11455575e-01 3.91855597e-01 5.85217535e-01 -3.20226103e-02
4.63484228e-02 4.34446670e-02 -1.47036493e+00 2.78418481e-01
9.73159313e-01 1.16850209e+00 5.78407943e-01 6.89129010e-02
-3.52741517e-02 6.23644471e-01 1.13549577e-02 -4.57579345e-01
2.98453391e-01 -1.84463769e-01 -9.56978321e-01 4.32058573e-01
6.55928552e-01 2.18276173e-01 6.19651563e-02 1.15569842e+00
1.90672562e-01 2.18630508e-02 4.12853241e-01 -1.13053107e+00
-5.10363281e-01 4.99058902e-01 -7.04164147e-01 4.10153717e-02
-1.02490269e-01 1.56279758e-01 1.44706893e+00 -7.32904017e-01
7.39119828e-01 7.80799866e-01 8.58198225e-01 4.01566625e-01
-1.12459719e+00 -2.69716263e-01 5.39155841e-01 1.18421853e-01
-1.27033758e+00 -2.15062335e-01 9.73439693e-01 -3.52174670e-01
6.17085218e-01 2.10867487e-02 7.01558828e-01 9.04189467e-01
-7.10498020e-02 9.81789172e-01 1.13538206e+00 -3.77299130e-01
3.69074106e-01 3.39258075e-01 1.04014210e-01 8.06558967e-01
-3.15034576e-02 -2.83895046e-01 -6.00504398e-01 1.72868386e-01
6.75929666e-01 1.89731419e-02 -2.53432721e-01 -7.08515823e-01
-9.67101395e-01 7.16091812e-01 8.23766708e-01 4.77696538e-01
-4.22253370e-01 8.73590931e-02 3.28747123e-01 -8.53804201e-02
5.37779212e-01 2.60776252e-01 -6.37642801e-01 3.42670977e-01
-1.15742159e+00 -2.00446546e-02 4.09300894e-01 7.52426267e-01
1.16534436e+00 -2.65255541e-01 -6.50335103e-02 9.26099122e-01
3.93630922e-01 7.05642849e-02 3.99095595e-01 -1.03875196e+00
3.55394214e-01 7.06009269e-01 -3.03605199e-01 -1.11115897e+00
-2.17695609e-01 -9.23939347e-01 -6.76648140e-01 -1.03036486e-01
3.36178958e-01 2.98143387e-01 -1.18248236e+00 1.77485394e+00
5.00349164e-01 3.02301526e-01 -5.46801202e-02 8.78729761e-01
5.08557916e-01 2.22503692e-01 2.69955188e-01 1.86668083e-01
1.27510881e+00 -1.30970895e+00 -2.95644343e-01 -4.49389845e-01
5.23631155e-01 -4.92212862e-01 1.24729776e+00 3.13547775e-02
-9.56643283e-01 -5.15900314e-01 -7.24271536e-01 -8.07911307e-02
-3.59546602e-01 1.16232283e-01 6.79385185e-01 3.44797462e-01
-1.12741709e+00 4.42027748e-01 -7.64261007e-01 -4.73206699e-01
7.75794089e-01 3.39566767e-02 -1.41091913e-01 -6.89512417e-02
-1.04056001e+00 5.11332452e-01 2.67235994e-01 1.93950072e-01
-8.42811167e-01 -8.22767019e-01 -8.14671457e-01 -5.77842481e-02
2.62143910e-01 -7.66148150e-01 9.08763409e-01 -1.43498659e+00
-1.05801475e+00 1.22428763e+00 -5.44791222e-02 -4.93531406e-01
6.74047470e-01 -2.78784689e-02 -1.33659036e-04 2.48873040e-01
6.72003925e-01 1.08770025e+00 7.46140063e-01 -1.48391926e+00
-9.21825409e-01 -5.09291470e-01 5.03029786e-02 4.26386505e-01
-3.92058372e-01 -3.34188789e-01 -8.01847637e-01 -7.24350631e-01
3.80364031e-01 -7.79020905e-01 -3.44958067e-01 1.63813904e-01
-4.49929357e-01 -2.04717994e-01 5.80810785e-01 -6.02845550e-01
8.48229825e-01 -2.23393607e+00 9.42947045e-02 4.39952284e-01
1.37968585e-01 -6.83581978e-02 -7.30837360e-02 -7.21653178e-02
2.97523677e-01 -2.81274199e-01 -7.21404552e-01 -5.20947397e-01
-1.45241290e-01 3.16955954e-01 -2.11624354e-02 2.73851544e-01
3.28431964e-01 8.59712481e-01 -1.08288860e+00 -5.81943393e-01
2.54284680e-01 3.47237796e-01 -5.75495958e-01 -7.36467689e-02
-3.10240299e-01 4.62301701e-01 -6.03125513e-01 8.69897544e-01
4.60501015e-01 -5.66614687e-01 7.33707771e-02 -2.47990727e-01
-1.55197596e-02 1.14306621e-01 -9.56064761e-01 1.90580857e+00
-3.78875792e-01 3.52357119e-01 9.47170705e-02 -1.38132894e+00
8.10447991e-01 7.97742978e-02 5.50151825e-01 -8.85322511e-01
-3.05201411e-02 3.96433264e-01 -4.86951321e-01 -3.36274415e-01
3.04186821e-01 -1.47281587e-03 6.06563240e-02 -4.99367109e-03
1.90356866e-01 -5.88788763e-02 7.66236410e-02 1.11693762e-01
8.53916883e-01 2.91900009e-01 3.27447578e-02 -2.64126539e-01
4.23365533e-01 2.82823205e-01 6.36683345e-01 6.01807654e-01
-3.46038133e-01 9.93025303e-01 1.39168173e-01 -2.09983453e-01
-8.66243362e-01 -1.02536476e+00 -2.47417733e-01 1.07705867e+00
3.91548604e-01 -6.60641566e-02 -8.45655024e-01 -9.63845730e-01
-1.21701159e-01 4.44608569e-01 -7.68061697e-01 -3.96631658e-02
-3.39531630e-01 -5.98394752e-01 2.91011423e-01 5.09865701e-01
8.34021091e-01 -1.16915870e+00 -5.01718760e-01 1.43169150e-01
-3.61995578e-01 -1.17204595e+00 -3.35396260e-01 3.37428272e-01
-9.12846506e-01 -8.85945261e-01 -9.43170428e-01 -1.09163642e+00
9.67373669e-01 3.74845445e-01 1.11732864e+00 -6.19714567e-03
-3.01708460e-01 3.31425905e-01 -2.62455910e-01 -5.65663679e-03
8.10567737e-02 2.97696620e-01 -5.35161614e-01 2.24525735e-01
1.12759411e-01 -5.33223093e-01 -7.90428817e-01 3.01342607e-01
-6.42234325e-01 2.82538205e-01 5.50719440e-01 7.80577600e-01
1.09145260e+00 2.90350597e-02 3.90432268e-01 -9.99831498e-01
-4.54831608e-02 -5.44907212e-01 -3.42092723e-01 3.89107436e-01
-5.87091684e-01 -6.02143034e-02 3.76558125e-01 -1.83767915e-01
-1.03770554e+00 3.81976634e-01 -3.26830707e-02 -3.86697084e-01
-2.88092703e-01 3.44720036e-01 -2.19329640e-01 6.32054731e-02
4.63022798e-01 3.47852439e-01 -3.68480623e-01 -3.53583425e-01
4.40698147e-01 3.51310968e-01 6.04350686e-01 -6.24922097e-01
5.74611843e-01 7.92726040e-01 -3.78436446e-01 -5.28270602e-01
-1.25742841e+00 -8.27183902e-01 -5.11704266e-01 -2.13550493e-01
9.47468817e-01 -1.04365265e+00 1.09900996e-01 5.44590533e-01
-6.89184666e-01 -5.69633186e-01 -6.22781575e-01 1.83913141e-01
-6.68204963e-01 2.07539201e-01 -2.70416439e-01 -3.53064477e-01
-3.59829336e-01 -9.51067865e-01 1.26011050e+00 3.71054083e-01
-1.27304718e-01 -1.17865002e+00 -3.79406358e-03 7.45080471e-01
2.25425601e-01 2.07665578e-01 7.22389042e-01 -3.91483486e-01
-6.04206502e-01 -8.63900185e-02 -4.78680670e-01 5.74575067e-01
2.36629367e-01 -2.21021920e-01 -1.09973192e+00 -5.91024719e-02
-1.12629779e-01 -3.18277538e-01 1.03968334e+00 6.22027814e-01
1.16345286e+00 -1.76843107e-01 -3.87985408e-01 6.29458904e-01
1.49136662e+00 -4.38937098e-01 4.09719408e-01 4.83128786e-01
8.91864419e-01 7.79968798e-01 5.54175198e-01 2.08555698e-01
7.85534441e-01 6.19005620e-01 4.15323615e-01 -4.46166992e-01
-3.64011258e-01 -4.18041229e-01 1.26246542e-01 5.54755151e-01
3.11500151e-02 4.04861271e-02 -8.69595051e-01 1.01238453e+00
-1.95781934e+00 -6.61589563e-01 -6.60983101e-02 2.07969809e+00
1.01589251e+00 3.44015598e-01 2.40445919e-02 -1.67767759e-02
7.48263776e-01 2.28647724e-01 -4.53127623e-01 1.67460963e-01
-1.87477931e-01 7.56547153e-02 4.88720179e-01 3.87258530e-01
-1.12011993e+00 1.14261210e+00 4.66497660e+00 1.05208659e+00
-1.18346691e+00 3.62219065e-01 9.81983542e-01 9.86812189e-02
-4.96616483e-01 1.19707763e-01 -6.98732555e-01 5.45028329e-01
3.35498124e-01 3.53402376e-01 6.34222701e-02 8.21086705e-01
2.09588885e-01 -2.40309775e-01 -7.25630581e-01 5.04923224e-01
-8.14793725e-03 -1.27228105e+00 8.33748505e-02 -1.57576919e-01
1.01905346e+00 2.45679259e-01 2.64398724e-01 5.57388775e-02
1.91178799e-01 -6.20556116e-01 1.01126957e+00 2.82856613e-01
4.44119334e-01 -3.77710670e-01 6.89073682e-01 1.05223551e-01
-1.20635676e+00 -8.11090767e-02 -2.39560038e-01 3.53722721e-01
-5.06217917e-03 8.67005825e-01 -6.90961003e-01 5.81982553e-01
9.57749963e-01 9.02475238e-01 -5.65221786e-01 1.02201152e+00
-3.69881183e-01 6.76038742e-01 -3.11326414e-01 3.65492642e-01
6.19050801e-01 -1.60077214e-02 2.50745714e-01 1.08357525e+00
3.61231267e-02 -2.57982939e-01 2.43391931e-01 8.77341390e-01
-7.86303356e-02 4.11766209e-02 -1.50287524e-01 3.50821614e-01
2.52591968e-01 1.38322937e+00 -1.11640394e+00 -2.30023235e-01
-3.81732881e-01 1.12589133e+00 5.04748166e-01 4.95763332e-01
-8.36896598e-01 6.12038709e-02 3.63762081e-01 3.59556913e-01
4.25398111e-01 1.14307646e-02 -6.53902531e-01 -9.63663459e-01
3.14302146e-01 -4.54914540e-01 5.16404212e-01 -6.56323910e-01
-1.31734383e+00 4.25179660e-01 -2.25758687e-01 -1.01429820e+00
2.42314458e-01 -1.36311755e-01 -6.48118019e-01 5.52392602e-01
-1.96299291e+00 -1.50227702e+00 -4.70368505e-01 6.46132231e-01
7.77526736e-01 1.47970960e-01 4.75844055e-01 2.95697927e-01
-4.80699927e-01 5.75421989e-01 2.60516740e-02 2.97422763e-02
5.04218876e-01 -1.26016045e+00 1.60773784e-01 6.85881555e-01
3.37103248e-01 2.81554878e-01 4.44987625e-01 -6.09930277e-01
-5.44281363e-01 -1.38940883e+00 1.04223466e+00 -3.27839077e-01
6.20721102e-01 -2.43601680e-01 -8.94056261e-01 4.72270131e-01
-1.21457018e-01 2.80562252e-01 3.25487554e-01 5.33355884e-02
-4.92103249e-01 -7.82855228e-02 -1.17871797e+00 4.00615335e-01
1.09565330e+00 -4.89297420e-01 -4.53155905e-01 3.94591093e-01
6.73711598e-01 -1.17935374e-01 -6.34916961e-01 4.45936143e-01
3.39641333e-01 -9.06064510e-01 8.67985010e-01 -7.20423907e-02
5.40451884e-01 -4.36442196e-01 -1.84376121e-01 -9.93151069e-01
-2.64011264e-01 -1.26838148e-01 3.27417701e-01 1.26573336e+00
6.10009074e-01 -4.53711390e-01 1.05161846e+00 3.97826582e-01
-3.64110529e-01 -9.78922665e-01 -8.17579925e-01 -6.52260303e-01
-1.14160098e-01 -3.07472497e-01 2.40568548e-01 1.15903950e+00
-2.92544991e-01 1.33951619e-01 4.74125221e-02 3.20074797e-01
7.36601114e-01 6.42814636e-02 3.55677873e-01 -1.10055256e+00
-2.72954047e-01 -4.71273869e-01 -3.91758263e-01 -9.04097140e-01
7.49405771e-02 -1.07223737e+00 1.79648235e-01 -1.65981805e+00
3.38600069e-01 -9.01308000e-01 -6.38805151e-01 7.26355076e-01
-3.46580148e-01 6.34595096e-01 1.76760759e-02 3.68871063e-01
-1.04897094e+00 5.80573499e-01 1.17584181e+00 -2.32244551e-01
-1.59068391e-01 1.30613208e-01 -6.41132116e-01 6.37091279e-01
8.59391809e-01 -5.23558795e-01 -4.93168652e-01 -4.67674315e-01
1.26311094e-01 -5.06245852e-01 8.02118063e-01 -1.04622447e+00
1.69203416e-01 3.88408974e-02 2.26250947e-01 -2.73538351e-01
7.59494677e-02 -8.44316602e-01 1.16064271e-04 2.56818265e-01
-5.56456149e-01 -4.61539894e-01 -5.35320528e-02 6.66345537e-01
-2.83545554e-01 -1.64109513e-01 8.59717846e-01 -2.53256559e-01
-9.86572027e-01 3.41095209e-01 2.70953588e-03 2.80800968e-01
9.17015553e-01 -5.55720747e-01 -1.09598534e-02 6.35579228e-02
-6.85760677e-01 4.59189475e-01 5.60079932e-01 3.59683990e-01
3.65108997e-01 -9.29033160e-01 -4.78309810e-01 1.46490052e-01
4.22221422e-01 4.13900793e-01 4.47651684e-01 9.76595700e-01
-2.36534685e-01 1.32920995e-01 -2.16790393e-01 -9.11462545e-01
-9.50395226e-01 6.17486089e-02 4.27423149e-01 -2.37827972e-01
-6.99218094e-01 1.17241895e+00 4.51258957e-01 -4.78367537e-01
4.00822520e-01 -3.11572909e-01 -8.43551978e-02 3.64504345e-02
-3.85890901e-02 -4.31995094e-02 2.32227355e-01 -5.62703907e-01
-3.65729451e-01 7.36970663e-01 -1.62299186e-01 1.73964165e-02
1.25022507e+00 -3.68345171e-01 5.30321486e-02 3.34217340e-01
1.05785549e+00 -3.43115985e-01 -1.59136009e+00 -5.23104668e-01
1.88307405e-01 -3.63381475e-01 2.17652634e-01 -8.37692261e-01
-1.43677270e+00 6.50595963e-01 7.33363986e-01 1.18745510e-02
1.14505005e+00 5.41827381e-01 7.92632043e-01 -1.43670723e-01
3.60698193e-01 -1.23251402e+00 1.25567064e-01 2.93412924e-01
3.92929494e-01 -1.33037853e+00 -2.65458941e-01 -6.45471513e-01
-7.67752588e-01 5.01272678e-01 6.35453165e-01 -7.19655231e-02
5.30324161e-01 -5.97972311e-02 8.43481198e-02 -3.55018556e-01
-3.98811311e-01 -4.91639197e-01 2.53238410e-01 7.21173286e-01
3.10480982e-01 1.77206144e-01 -6.76179826e-02 4.58588749e-01
1.45555437e-02 -7.47485012e-02 -6.28351793e-03 7.69130290e-01
-4.94115055e-01 -6.89801693e-01 4.70526852e-02 3.27954173e-01
-3.11520010e-01 -1.90329298e-01 -1.47473201e-01 5.28892279e-01
3.69270682e-01 6.21903360e-01 2.16492131e-01 -3.46640445e-05
2.07625747e-01 -1.09783135e-01 3.01703572e-01 -6.79272473e-01
-5.22618890e-01 5.79803213e-02 -1.53762832e-01 -6.17758036e-01
-6.12285852e-01 -6.50893033e-01 -1.48738933e+00 2.16878116e-01
-2.41294742e-01 -3.24310618e-03 5.63856781e-01 9.53751981e-01
2.79359311e-01 5.05916953e-01 6.09344482e-01 -6.99721396e-01
-2.70786136e-01 -5.40825903e-01 -4.99178499e-01 5.10537744e-01
6.48745298e-02 -5.12602329e-01 -3.54766905e-01 2.53899306e-01] | [9.561334609985352, 0.591437041759491] |
940b9260-6089-4b4c-ae50-1816e675c22b | dual-reweighted-lp-norm-minimization-for-salt | 1811.09173 | null | https://arxiv.org/abs/1811.09173v3 | https://arxiv.org/pdf/1811.09173v3.pdf | Dual Reweighted Lp-Norm Minimization for Salt-and-pepper Noise Removal | The robust principal component analysis (RPCA), which aims to estimate underlying low-rank and sparse structures from the degraded observation data, has found wide applications in computer vision. It is usually replaced by the principal component pursuit (PCP) model in order to pursue the convex property, leading to the undesirable overshrink problem. In this paper, we propose a dual weighted lp-norm (DWLP) model with a more reasonable weighting rule and weaker powers, which greatly generalizes the previous work and provides a better approximation to the rank minimization problem for original matrix as well as the l0-norm minimization problem for sparse data. Moreover, an approximate closed-form solution is introduced to solve the lp-norm minimization, which has more stability in the nonconvex optimization and provides a more accurate estimation for the low-rank and sparse matrix recovery problem. We then apply the DWLP model to remove salt-and-pepper noise by exploiting the image nonlocal self-similarity. Both qualitative and quantitative experiments demonstrate that the proposed method outperforms other state-of-the-art methods. In terms of PSNR evaluation, our DWLP achieves about 7.188dB, 5.078dB, 3.854dB, 2.536dB and 0.158dB improvements over the current WSNM-RPCA under 10\% to 50\% salt-and-pepper noise with an interval 10\% respectively. | ['Huiwen Dong', 'Chuangbai Xiao', 'Jing Yu'] | 2018-11-22 | null | null | null | null | ['salt-and-pepper-noise-removal'] | ['computer-vision'] | [ 2.49679431e-01 -3.68250817e-01 4.44510542e-02 6.11543581e-02
-8.84261847e-01 -2.40060002e-01 1.67989433e-01 -2.24741861e-01
-2.03800574e-01 5.63490987e-01 6.60767019e-01 3.49194445e-02
-3.40983272e-01 -4.41621304e-01 -5.49017549e-01 -1.27488792e+00
-8.03985372e-02 -2.74828523e-01 6.82947412e-02 -8.61743242e-02
1.04647979e-01 3.09561789e-01 -1.25836587e+00 9.72119346e-03
9.49611902e-01 1.12741196e+00 4.73244667e-01 1.93582162e-01
3.10219437e-01 5.89675963e-01 -2.66186744e-01 -5.62042631e-02
4.06899869e-01 -2.38993257e-01 -7.85332769e-02 4.27930385e-01
1.99687347e-01 -2.23681197e-01 -6.55527592e-01 1.54752994e+00
6.13552332e-01 6.41677082e-02 3.96311104e-01 -9.95328724e-01
-5.60256302e-01 2.59149313e-01 -1.37286878e+00 1.48788139e-01
3.01998496e-01 -9.08252224e-03 6.94477141e-01 -1.03328931e+00
3.46847504e-01 1.40768123e+00 6.38076663e-01 4.02623899e-02
-1.06889832e+00 -6.94547355e-01 -5.17269969e-02 5.13307035e-01
-1.60586381e+00 -5.09203672e-01 1.02798712e+00 -1.91015214e-01
2.28402287e-01 5.18639088e-01 2.82844722e-01 6.36808276e-01
1.00653194e-01 8.30267727e-01 1.38995767e+00 -3.10258597e-01
-7.23145977e-02 -2.38619015e-01 1.01878541e-02 5.77437401e-01
5.50999939e-01 -6.57451600e-02 -3.32952499e-01 -3.98638010e-01
7.13334322e-01 2.33006448e-01 -7.85918951e-01 -2.50844061e-01
-1.42199838e+00 5.40015042e-01 4.58837956e-01 1.46182150e-01
-6.57534957e-01 -9.42180678e-02 2.61464179e-01 -8.88476893e-03
2.21059442e-01 -1.77823409e-01 -6.94200546e-02 1.77187815e-01
-8.52864921e-01 1.58203855e-01 5.86351931e-01 9.95842576e-01
4.77807194e-01 4.74093109e-01 -2.24639162e-01 1.16647673e+00
8.67538035e-01 1.06038535e+00 4.11458284e-01 -1.21546388e+00
6.27832055e-01 2.52377301e-01 1.78943738e-01 -1.67407012e+00
-1.46084845e-01 -8.14416528e-01 -1.37463760e+00 1.70468111e-02
1.50530323e-01 1.11871183e-01 -5.93567133e-01 1.42574763e+00
2.55509555e-01 5.21657169e-01 2.78050452e-01 1.20826960e+00
7.11781263e-01 9.64491844e-01 -2.70051301e-01 -8.82156491e-01
1.37461185e+00 -8.01792145e-01 -1.03535140e+00 -2.28980839e-01
-5.26492670e-02 -1.10187101e+00 8.68814111e-01 7.09579051e-01
-9.27315950e-01 -4.44034070e-01 -1.18644094e+00 2.46201143e-01
4.74047542e-01 2.33692199e-01 1.79991558e-01 4.14885640e-01
-6.60935700e-01 2.65455633e-01 -7.68387735e-01 -1.80602401e-01
2.48903573e-01 4.94422205e-02 -4.40905541e-01 -7.29216337e-01
-8.72418404e-01 3.77948374e-01 -1.12533569e-01 5.85496128e-01
-6.75255835e-01 -4.40003812e-01 -6.58859253e-01 -1.33482143e-01
4.68928903e-01 -2.49363184e-01 5.36379457e-01 -5.77752590e-01
-1.30590081e+00 5.23294806e-01 -4.44530159e-01 -1.14721157e-01
2.15710476e-01 -3.18070590e-01 -6.00828946e-01 3.19226384e-01
1.59791112e-01 -1.05418310e-01 9.49398160e-01 -1.38069689e+00
-3.48095149e-01 -5.73785603e-01 -4.67157781e-01 3.27977628e-01
-2.86904395e-01 7.80328959e-02 -7.50099778e-01 -1.13256812e+00
9.67433453e-01 -9.46535826e-01 -4.08325732e-01 4.22166884e-02
-2.80647069e-01 2.97158450e-01 9.36147749e-01 -9.61031556e-01
1.25426638e+00 -2.48160481e+00 2.97653526e-01 2.88699985e-01
4.96794619e-02 2.83475637e-01 -1.86197266e-01 4.37701792e-01
-4.41100150e-02 -3.85550976e-01 -5.18609762e-01 -1.36752725e-01
-4.07104760e-01 2.13728413e-01 -1.57574683e-01 1.01361990e+00
-3.55433404e-01 3.28876525e-01 -8.67528975e-01 -3.95204961e-01
1.40241921e-01 6.65556610e-01 -3.62288743e-01 1.12079486e-01
4.91261989e-01 3.82919520e-01 -4.61698860e-01 8.76390457e-01
1.14854121e+00 7.14425743e-02 -5.01143709e-02 -7.68373907e-01
-2.32534081e-01 -5.25859892e-01 -1.89520264e+00 1.76145458e+00
-1.62062138e-01 3.51469755e-01 9.41824555e-01 -1.13809431e+00
9.96348619e-01 2.89317518e-01 6.75903618e-01 -4.98730719e-01
7.69121125e-02 4.02236044e-01 -2.07345396e-01 -7.90613532e-01
1.70652047e-01 -2.48514846e-01 1.69453591e-01 -1.34949043e-01
-4.78369892e-01 2.51382589e-01 6.22066632e-02 1.79872811e-01
1.04367936e+00 -9.68999639e-02 3.41891468e-01 -7.46988893e-01
1.01327026e+00 -3.44553024e-01 1.19983459e+00 4.08053666e-01
-2.91933537e-01 6.81002259e-01 1.79677859e-01 -1.16037093e-01
-6.15916789e-01 -9.27038133e-01 -2.13221177e-01 5.88361561e-01
4.86434042e-01 -2.25942377e-02 -3.86037380e-01 -1.34572372e-01
-1.27607390e-01 2.78531879e-01 -4.34236564e-02 1.72764696e-02
-5.79116404e-01 -1.14728272e+00 2.45583087e-01 1.82224154e-01
9.20989156e-01 -6.00839376e-01 1.64595470e-02 1.82269320e-01
-4.51571703e-01 -1.13543761e+00 -5.17583609e-01 -3.28913182e-01
-9.29416120e-01 -9.69602823e-01 -1.02157068e+00 -8.98572505e-01
9.08469975e-01 8.30448329e-01 5.42860806e-01 -1.09519541e-01
4.59984802e-02 2.64224470e-01 -5.59294105e-01 -6.10699877e-02
1.05911687e-01 -7.06007540e-01 3.04481030e-01 5.41190326e-01
1.45923167e-01 -8.35100234e-01 -7.28614151e-01 5.33621967e-01
-1.00122011e+00 -1.80151314e-01 9.62091506e-01 9.76663947e-01
8.73161793e-01 4.18594807e-01 2.71696776e-01 -5.20954967e-01
6.02820158e-01 -4.55882639e-01 -6.33572757e-01 -6.77654818e-02
-7.42625177e-01 -8.32071975e-02 6.37636483e-01 -3.25530589e-01
-1.09912932e+00 3.29131395e-01 1.21153882e-02 -7.16271043e-01
1.33469194e-01 6.50411963e-01 -6.20806813e-01 -2.86175787e-01
3.99111092e-01 6.85922742e-01 1.72656015e-01 -8.62352669e-01
2.49461204e-01 7.41826534e-01 7.93142676e-01 -5.85404634e-01
1.28576577e+00 8.53157103e-01 2.38093942e-01 -1.10215175e+00
-6.74166024e-01 -8.59564185e-01 -1.65729061e-01 -1.18761025e-02
4.70844358e-01 -1.34577942e+00 -6.46705449e-01 6.90650582e-01
-8.13757896e-01 2.11369857e-01 5.10600731e-02 7.59868920e-01
-2.30178729e-01 1.02356958e+00 -6.58423603e-01 -8.87060523e-01
-4.34749633e-01 -1.20286131e+00 6.96243048e-01 1.86212972e-01
4.80508953e-01 -3.91808391e-01 -2.76608288e-01 5.29406011e-01
5.44445932e-01 3.57884645e-01 4.11978662e-01 2.18865033e-02
-3.98639411e-01 -1.97589666e-01 -4.90225792e-01 5.79817176e-01
5.46316896e-03 -4.89092171e-01 -6.35138452e-01 -5.61962485e-01
5.80782890e-01 2.04446778e-01 6.93781674e-01 3.84594560e-01
9.65839446e-01 -7.23390520e-01 -1.51722118e-01 8.96881342e-01
1.67640841e+00 1.54323876e-01 8.67783487e-01 3.71182144e-01
8.03081155e-01 5.67316830e-01 8.34690154e-01 6.04499161e-01
3.01332057e-01 5.78692377e-01 5.92398345e-01 -8.48772898e-02
-3.93786170e-02 -1.95081485e-03 4.46935713e-01 1.21548653e+00
-9.54927038e-03 3.06087174e-02 -6.45354092e-01 5.26485980e-01
-1.83814108e+00 -7.65837133e-01 -6.32743061e-01 2.17009926e+00
7.11073399e-01 -1.21287115e-01 -2.31506720e-01 4.54800576e-01
7.80246079e-01 4.85598743e-01 -3.76629949e-01 2.61260509e-01
-3.72038215e-01 -3.15871418e-01 8.01795602e-01 5.34685552e-01
-1.00232160e+00 3.41818869e-01 5.04997396e+00 1.05418575e+00
-9.18286860e-01 7.71578327e-02 3.34317923e-01 1.77159816e-01
-1.20387450e-01 2.86659505e-02 -4.22012448e-01 6.62512183e-01
4.48817194e-01 -1.13126049e-02 5.68958998e-01 7.40611792e-01
7.10069299e-01 -7.45686740e-02 -3.13307136e-01 1.44513750e+00
3.03499460e-01 -7.55037904e-01 -1.24132738e-01 1.98490798e-01
6.88545465e-01 -1.14594556e-01 -2.36217622e-02 -1.16395883e-01
-1.41428739e-01 -5.65286815e-01 5.26345849e-01 4.57619280e-01
6.01059198e-01 -6.70488358e-01 9.67740178e-01 3.36064041e-01
-1.32931209e+00 -2.02752739e-01 -5.80081940e-01 2.30128411e-02
2.65841037e-01 1.15616310e+00 -2.95448466e-03 5.98532081e-01
9.34495270e-01 9.29899096e-01 -7.05509633e-02 1.21593177e+00
-2.08925888e-01 6.97394192e-01 -4.10907805e-01 3.68122607e-01
1.32500648e-01 -6.99047565e-01 1.07557452e+00 1.16819239e+00
5.78146338e-01 4.85434920e-01 3.33404303e-01 4.22923893e-01
1.10312425e-01 3.01836729e-01 -1.33108333e-01 3.02623302e-01
5.06090045e-01 1.29370987e+00 -3.25756192e-01 -5.88635467e-02
-4.78957504e-01 1.06816745e+00 -3.83759558e-01 7.32152820e-01
-5.08399963e-01 -5.72123170e-01 5.64283490e-01 1.68254614e-01
3.54169130e-01 -3.82601231e-01 -8.44400376e-02 -1.17569935e+00
4.60117519e-01 -1.28106534e+00 2.87170738e-01 -6.52471662e-01
-1.30152464e+00 2.88781047e-01 -1.61810696e-01 -1.60917866e+00
3.55428159e-01 -3.67302895e-01 -6.49326622e-01 8.97734761e-01
-1.76329136e+00 -1.06421030e+00 -5.80409825e-01 7.50410080e-01
3.53434443e-01 -2.23859847e-01 3.56039852e-01 6.02099061e-01
-7.53318608e-01 2.87619382e-01 7.09531724e-01 -5.33122150e-03
6.03393376e-01 -7.88557589e-01 -4.09764022e-01 1.24882996e+00
-3.35928470e-01 7.01960683e-01 1.06800878e+00 -5.05104959e-01
-2.08975649e+00 -1.00168562e+00 4.52760369e-01 3.34778130e-01
6.07356846e-01 1.26545012e-01 -9.61108804e-01 2.29757056e-01
2.77561061e-02 3.89247775e-01 4.96621817e-01 -3.94099206e-01
-3.62194985e-01 -6.79251850e-01 -1.25625062e+00 3.64958227e-01
7.91147709e-01 -2.38906443e-01 -2.70598203e-01 2.69055963e-01
5.59172630e-01 -2.75347888e-01 -1.02425933e+00 5.49707115e-01
4.26435918e-01 -8.71380031e-01 1.11416817e+00 1.44272089e-01
5.52315861e-02 -9.64743972e-01 -6.72553480e-01 -1.03162992e+00
-7.05301642e-01 -8.00265789e-01 -2.68631458e-01 1.38666904e+00
8.06462113e-03 -5.47412634e-01 6.37189209e-01 1.89630404e-01
-1.88092992e-01 -6.82717443e-01 -1.04889500e+00 -8.75539839e-01
-5.64346552e-01 -2.51623094e-01 1.99164838e-01 8.69667709e-01
-1.52271464e-01 1.31735623e-01 -7.70791233e-01 6.12456441e-01
1.38303030e+00 -9.31733660e-03 5.66563189e-01 -8.56475174e-01
-3.88225526e-01 -7.60832801e-02 -4.27731782e-01 -1.42390645e+00
-1.80627897e-01 -5.92873275e-01 3.24832737e-01 -1.50838506e+00
2.18526855e-01 -3.05767864e-01 -5.79605222e-01 9.58781093e-02
-2.71767437e-01 3.41771215e-01 1.75762519e-01 6.51528180e-01
-3.78910720e-01 7.00756967e-01 1.18226230e+00 -2.46966764e-01
-1.55079812e-01 -6.90919757e-02 -9.05163348e-01 6.70030057e-01
5.04219472e-01 -3.77198368e-01 -3.41467947e-01 -6.90016568e-01
-4.02832025e-04 2.66465276e-01 7.60545656e-02 -1.09422004e+00
2.16353759e-01 -1.56647697e-01 1.83804691e-01 -4.97961909e-01
4.20994014e-01 -1.05284154e+00 2.13960245e-01 5.18749297e-01
2.44154632e-02 -1.46119043e-01 -3.90177704e-02 1.03592420e+00
-4.57581311e-01 -1.33453771e-01 8.81010175e-01 -1.11289032e-01
-6.86886430e-01 2.85034746e-01 -1.84339702e-01 -2.42207080e-01
8.21620584e-01 -2.85298765e-01 -1.95932850e-01 -5.21719933e-01
-4.61621404e-01 2.95118302e-01 2.31097832e-01 9.29267853e-02
8.90151680e-01 -1.51699579e+00 -1.11338127e+00 2.14828521e-01
4.73414510e-02 -4.48487177e-02 4.11934674e-01 1.15996695e+00
-5.72507501e-01 9.40125138e-02 8.20890535e-03 -5.16624689e-01
-1.25848281e+00 5.61999559e-01 -9.60331038e-02 -1.36042699e-01
-6.19085312e-01 7.49841809e-01 4.97996956e-02 -2.60033435e-03
2.41386145e-01 1.16947651e-01 -2.93543994e-01 -1.35132834e-01
7.05674291e-01 8.63735557e-01 -2.30486751e-01 -1.15354085e+00
-3.25657636e-01 8.15681577e-01 2.86936253e-01 -1.14306612e-02
1.45843577e+00 -4.97681141e-01 -5.36611080e-01 6.87997341e-02
1.40795374e+00 3.83327544e-01 -1.16368496e+00 -6.50444627e-01
-1.99316725e-01 -7.44195759e-01 4.47107345e-01 -3.37797523e-01
-1.30056894e+00 5.95133305e-01 7.57020473e-01 -2.44053658e-02
1.40471911e+00 -4.19176489e-01 9.82968032e-01 2.43074715e-01
4.18642431e-01 -8.04685295e-01 -8.49090293e-02 2.23709419e-01
1.23128998e+00 -1.09878182e+00 5.96948326e-01 -6.56738997e-01
-4.79796320e-01 9.15682256e-01 2.67932504e-01 -2.89077431e-01
6.74247563e-01 1.50903761e-01 5.03205918e-02 -1.21056717e-02
-1.86783686e-01 4.10725474e-02 2.64397919e-01 5.61184585e-01
1.23864628e-01 3.75344865e-02 -6.22678638e-01 5.71753561e-01
2.17532948e-01 -1.79291874e-01 4.10496354e-01 7.03112662e-01
-6.51923180e-01 -8.23674619e-01 -1.07699442e+00 4.20198858e-01
-6.26506388e-01 -6.00170530e-02 2.92928427e-01 1.87172741e-01
4.30726074e-02 1.24339521e+00 -5.43989539e-01 -3.71428818e-01
4.79605615e-01 -5.63772917e-01 2.21590437e-02 -2.00698227e-01
1.25649393e-01 7.98885345e-01 -1.81752935e-01 -7.85594940e-01
-6.73123360e-01 -8.16543877e-01 -1.09243119e+00 -9.08564106e-02
-3.23039889e-01 3.05095971e-01 5.60555279e-01 5.96924007e-01
2.95424849e-01 1.85943097e-01 8.06940794e-01 -7.30386257e-01
-5.73006570e-01 -9.19017494e-01 -1.01726449e+00 4.51481640e-01
4.07298654e-01 -3.73475760e-01 -7.26758242e-01 1.93187088e-01] | [11.311866760253906, -2.4016928672790527] |
d3e0e706-5359-433d-b52f-db607790e5e0 | end-to-end-sensor-modeling-for-lidar-point | 1907.07748 | null | https://arxiv.org/abs/1907.07748v1 | https://arxiv.org/pdf/1907.07748v1.pdf | End-to-end sensor modeling for LiDAR Point Cloud | Advanced sensors are a key to enable self-driving cars technology. Laser scanner sensors (LiDAR, Light Detection And Ranging) became a fundamental choice due to its long-range and robustness to low light driving conditions. The problem of designing a control software for self-driving cars is a complex task to explicitly formulate in rule-based systems, thus recent approaches rely on machine learning that can learn those rules from data. The major problem with such approaches is that the amount of training data required for generalizing a machine learning model is big, and on the other hand LiDAR data annotation is very costly compared to other car sensors. An accurate LiDAR sensor model can cope with such problem. Moreover, its value goes beyond this because existing LiDAR development, validation, and evaluation platforms and processes are very costly, and virtual testing and development environments are still immature in terms of physical properties representation. In this work we propose a novel Deep Learning-based LiDAR sensor model. This method models the sensor echos, using a Deep Neural Network to model echo pulse widths learned from real data using Polar Grid Maps (PGM). We benchmark our model performance against comprehensive real sensor data and very promising results are achieved that sets a baseline for future works. | ['Moemen Abdel-Razek', 'Mohamed Elsobky', 'Khaled Elmadawi', 'Mohamed Zahran', 'Hesham M. Eraqi'] | 2019-07-17 | null | null | null | null | ['sensor-modeling'] | ['computer-vision'] | [-8.57052859e-03 -3.78203452e-01 -1.24545678e-01 -7.62521923e-01
-3.65198255e-01 -1.74334630e-01 5.90018928e-01 7.09642768e-02
-3.45605642e-01 7.25713193e-01 -6.09699249e-01 -4.67559963e-01
-1.63850054e-01 -1.17174029e+00 -7.52991199e-01 -6.09554470e-01
-9.37140882e-02 9.07110095e-01 7.74437249e-01 -5.31771839e-01
2.06826001e-01 9.47859168e-01 -2.32353663e+00 -3.95046413e-01
9.88830984e-01 1.34155023e+00 2.64158726e-01 5.42234421e-01
-4.03036684e-01 2.95120358e-01 -1.87182993e-01 7.36506581e-02
2.81863600e-01 2.03927115e-01 1.03559799e-01 -4.42496657e-01
5.28881967e-01 -2.10914373e-01 -1.27408609e-01 8.50035071e-01
6.61329985e-01 -6.38811588e-02 6.98843956e-01 -1.46674502e+00
-2.11480990e-01 2.45986268e-01 -2.28122711e-01 1.25942891e-02
-3.80771250e-01 4.45575655e-01 4.28821772e-01 -7.83839762e-01
1.22678190e-01 9.11252201e-01 6.89031601e-01 4.80887949e-01
-1.09549689e+00 -9.97356653e-01 -4.35560048e-01 5.11936665e-01
-1.45272589e+00 -6.59296513e-01 9.08646286e-01 -6.44477665e-01
8.37471366e-01 1.73157901e-02 6.70300543e-01 8.13413799e-01
4.82367635e-01 -2.35342644e-02 1.37188816e+00 -1.24454826e-01
5.09022176e-01 4.22177792e-01 2.63905108e-01 6.21304750e-01
4.75776017e-01 8.59964430e-01 -4.44137037e-01 4.10629809e-01
1.78081587e-01 -2.58091658e-01 2.58777648e-01 -4.65072811e-01
-4.78941262e-01 8.90551925e-01 2.72412002e-01 1.70730874e-01
-4.37627167e-01 3.29346150e-01 2.10551023e-01 2.81585127e-01
1.65879071e-01 5.41346632e-02 -3.72623831e-01 -1.38137877e-01
-9.96682703e-01 2.84598619e-01 7.91043043e-01 9.04939711e-01
1.42152464e+00 4.04542834e-01 3.26788902e-01 6.82878256e-01
6.28850937e-01 1.15462983e+00 2.70077378e-01 -9.58549917e-01
-6.87130392e-02 4.78579283e-01 8.88044015e-03 -7.18239307e-01
-4.83553231e-01 -1.85057044e-01 -8.26887369e-01 6.65094852e-01
-1.30043477e-01 -6.06014542e-02 -9.28305268e-01 1.29393232e+00
2.67885178e-01 3.28587502e-01 9.50401649e-02 6.21686816e-01
8.49044383e-01 6.04416490e-01 9.58795473e-02 -1.77099600e-01
9.99832332e-01 -2.54077494e-01 -6.16664827e-01 -4.30903196e-01
3.00019920e-01 -3.08947057e-01 8.81829083e-01 4.04415548e-01
-5.42631328e-01 -9.78217006e-01 -1.34940112e+00 1.49305627e-01
-8.44292343e-01 -2.03029364e-01 7.33455956e-01 9.50674117e-01
-9.09632206e-01 5.42170048e-01 -6.69375420e-01 -3.35311502e-01
2.61740714e-01 4.85605836e-01 7.00932322e-03 -4.00328897e-02
-1.35524797e+00 1.20724213e+00 4.31155145e-01 1.69675872e-01
-8.27298403e-01 -8.28040302e-01 -8.22111547e-01 -1.71742246e-01
2.46261030e-01 -1.84748903e-01 1.14445055e+00 1.07829250e-01
-1.75582576e+00 7.28484333e-01 2.33836900e-02 -8.35818470e-01
2.58895904e-01 2.63261329e-02 -7.89799392e-01 -2.27898151e-01
-7.45992139e-02 8.62132251e-01 1.04605663e+00 -1.39153504e+00
-6.23954952e-01 -2.24706635e-01 -4.17565554e-01 -5.25837779e-01
-1.61577910e-01 -3.00024480e-01 -8.28046724e-02 4.73695695e-01
-2.51475453e-01 -9.51652288e-01 -2.78084397e-01 -5.20504303e-02
2.41521411e-02 -3.43838453e-01 1.22495246e+00 -3.52745913e-02
9.00951385e-01 -1.90468860e+00 -7.79971421e-01 2.22501859e-01
-8.20493251e-02 6.06851578e-01 1.75693375e-03 2.21836880e-01
2.29871765e-01 -6.73574880e-02 -2.91567892e-01 -5.54142818e-02
1.97033733e-01 6.11540139e-01 -5.09648442e-01 9.11104754e-02
4.14607108e-01 8.25182021e-01 -5.05960166e-01 -6.64792180e-01
8.29342365e-01 6.42042041e-01 -1.82526723e-01 1.65351436e-01
-4.50726748e-01 3.90082538e-01 -2.30419084e-01 5.61473370e-01
1.16514325e+00 3.38966399e-01 -3.00868571e-01 -2.69261837e-01
-7.09062696e-01 1.29186079e-01 -1.25028503e+00 1.24844301e+00
-8.91460836e-01 7.41276681e-01 1.16625600e-01 -1.19406486e+00
1.58898950e+00 -1.02030024e-01 6.32422864e-01 -1.12664294e+00
1.22819491e-01 4.72525746e-01 3.26032117e-02 -8.27496827e-01
5.74545741e-01 -3.14366728e-01 -1.42202573e-02 3.16285230e-02
-2.35712811e-01 -9.19785500e-01 2.39647344e-01 -3.14707845e-01
7.18982100e-01 2.25414485e-01 -2.14458942e-01 -2.06398025e-01
7.25379109e-01 1.93030596e-01 6.17710829e-01 5.18355250e-01
-6.70091361e-02 1.79143131e-01 -7.04433024e-02 -5.65100312e-01
-8.61424923e-01 -8.45851600e-01 -5.37546754e-01 6.90298557e-01
1.96634054e-01 -2.97942851e-02 -3.74953598e-01 -2.36215703e-02
4.98584777e-01 9.13280249e-01 -1.52038768e-01 -2.33114347e-01
-6.33364916e-01 -5.84808052e-01 4.86507297e-01 4.65200871e-01
6.15390301e-01 -9.55647230e-01 -1.01423275e+00 4.62830186e-01
4.18310553e-01 -1.41283989e+00 4.60409313e-01 4.00140196e-01
-7.51206875e-01 -1.00436664e+00 1.92200840e-01 -2.06971377e-01
-1.04533337e-01 2.36254409e-01 1.03805101e+00 5.19642420e-02
-4.91967887e-01 2.90455282e-01 5.19733280e-02 -1.23112941e+00
-6.32950425e-01 -1.28933221e-01 3.66125494e-01 -4.01697755e-02
8.20327938e-01 -7.11773932e-01 -3.73779982e-01 2.71903545e-01
-8.21936905e-01 -2.46821627e-01 8.47581327e-01 3.58848780e-01
6.91562891e-01 2.93223858e-01 6.09344721e-01 -5.53893149e-01
3.64651561e-01 -3.06990147e-01 -1.31511688e+00 -9.59359333e-02
-1.12577128e+00 2.05421254e-01 4.37405348e-01 -1.61519110e-01
-8.31280231e-01 2.59309739e-01 -4.27139312e-01 -3.57536286e-01
-6.43657148e-01 3.90164167e-01 -2.56722897e-01 -2.81749427e-01
6.45863891e-01 1.13276243e-01 2.15199351e-01 -2.69505501e-01
2.04098195e-01 9.33165133e-01 5.66994309e-01 -6.33080542e-01
1.08226073e+00 4.38959926e-01 6.21476054e-01 -1.42482412e+00
-3.99496466e-01 -2.88426101e-01 -7.68230319e-01 -5.29401839e-01
6.49808049e-01 -8.13289404e-01 -1.00729060e+00 4.03193206e-01
-1.06058621e+00 -5.09204328e-01 -4.06634897e-01 5.33918142e-01
-4.84032840e-01 1.09465376e-01 1.88514918e-01 -1.12757194e+00
-2.20545366e-01 -1.10922897e+00 1.03838086e+00 3.57260317e-01
2.13491395e-01 -8.04403603e-01 2.48285070e-01 3.03832144e-01
1.06114793e+00 3.40128005e-01 7.37345338e-01 -1.52140036e-01
-9.72561479e-01 -4.22539234e-01 -2.49404669e-01 5.35401165e-01
-2.27398127e-01 3.37516069e-01 -1.23101401e+00 1.23347729e-01
7.47686550e-02 -3.82614285e-01 8.83062065e-01 2.91866243e-01
1.07967150e+00 3.83176178e-01 -4.77584660e-01 4.22274232e-01
1.59651303e+00 1.27703905e-01 6.17048919e-01 7.62350634e-02
4.23733532e-01 7.34701812e-01 9.31334257e-01 1.35996804e-01
4.46775258e-01 6.99837744e-01 7.25831091e-01 1.31290227e-01
-1.15641825e-01 -8.30732584e-02 3.62556309e-01 5.97382724e-01
-7.94263259e-02 3.05892248e-02 -1.21559262e+00 3.16269219e-01
-1.58756268e+00 -8.84973645e-01 -5.85384846e-01 2.26866341e+00
4.92817491e-01 4.33752149e-01 -1.54431656e-01 3.12427014e-01
4.32066649e-01 -1.02318212e-01 -7.01322675e-01 -5.48831701e-01
1.43807188e-01 6.24973297e-01 7.52193272e-01 4.71013278e-01
-8.38122368e-01 9.02361989e-01 5.82439470e+00 5.32416880e-01
-1.76147938e+00 1.74306691e-01 -1.25641346e-01 3.33064258e-01
-1.48285910e-01 -4.82682213e-02 -1.32620025e+00 4.65395182e-01
1.72546446e+00 1.06326446e-01 2.31375128e-01 7.20913529e-01
5.80511451e-01 -3.38595778e-01 -9.50108767e-01 1.08500290e+00
-2.01495275e-01 -1.19785774e+00 -3.49870712e-01 2.33219013e-01
2.15688556e-01 5.95226228e-01 -7.49306083e-02 6.48685932e-01
2.57628798e-01 -9.50671017e-01 5.69492042e-01 8.18096817e-01
1.02919805e+00 -6.03006661e-01 6.83626592e-01 5.91383159e-01
-1.25168693e+00 -6.64944723e-02 -7.82384872e-01 -6.96355402e-02
1.76906541e-01 9.17423964e-01 -9.17369008e-01 4.79343295e-01
7.03461707e-01 5.03698826e-01 -5.52268505e-01 9.54874992e-01
-1.53389439e-01 6.53814197e-01 -6.90254450e-01 -3.21820676e-01
-1.05676286e-01 -2.30297491e-01 2.58216083e-01 1.29318941e+00
7.17904866e-01 -2.57783711e-01 1.43829018e-01 1.09506905e+00
2.52156198e-01 -2.80077398e-01 -1.16776550e+00 -7.49005154e-02
6.02226794e-01 1.48730254e+00 -2.91044205e-01 -7.76446164e-02
-3.75459075e-01 1.28352921e-02 -2.30279431e-01 7.29628578e-02
-8.58973086e-01 -3.40771675e-01 8.75730991e-01 5.75947225e-01
1.94119692e-01 -5.58673859e-01 -2.32503116e-01 -5.01092792e-01
-2.17517868e-01 -2.22000927e-01 -2.31279016e-01 -7.38013148e-01
-1.15390515e+00 2.84160465e-01 2.06358120e-01 -1.22940731e+00
-4.01861727e-01 -9.96484816e-01 -6.03182495e-01 7.84830928e-01
-2.29121351e+00 -1.18705094e+00 -8.34398806e-01 4.29692924e-01
1.87332734e-01 -2.16888189e-01 8.33396018e-01 5.33165038e-01
-3.12213451e-01 1.86392561e-01 -1.72477379e-01 -1.98766440e-01
6.34220958e-01 -1.12611079e+00 1.95123628e-01 5.45530081e-01
-1.36207655e-01 1.34357437e-01 8.08700323e-01 -4.50856119e-01
-1.94044113e+00 -1.27982795e+00 6.45589590e-01 -4.01120365e-01
7.50656128e-01 -5.44112504e-01 -1.02262974e+00 9.60743055e-02
1.63439348e-01 3.99511904e-01 3.90011907e-01 -2.31991529e-01
-9.07870829e-02 -8.63073111e-01 -1.06884992e+00 -3.90036516e-02
8.08603823e-01 -5.04623473e-01 -3.60240608e-01 4.88896444e-02
4.12798792e-01 -2.02274472e-01 -7.29131043e-01 7.46329248e-01
5.38462818e-01 -1.23879516e+00 7.56164670e-01 -1.00197211e-01
-5.40967798e-03 -5.95436931e-01 -1.82541043e-01 -1.08832538e+00
9.54177380e-02 -3.93245071e-01 -2.04277754e-01 1.31679487e+00
2.55348146e-01 -7.93847799e-01 8.21089864e-01 5.69391251e-01
-4.65767175e-01 -4.01892990e-01 -1.05555511e+00 -1.13430107e+00
1.85618758e-01 -1.08584929e+00 8.57766807e-01 5.45695662e-01
-7.04575896e-01 5.91951847e-01 -3.85803990e-02 3.89159709e-01
8.59690249e-01 1.26010463e-01 1.12646985e+00 -2.04462194e+00
2.25461051e-01 -2.40582705e-01 -6.05561197e-01 -7.14151800e-01
2.98862845e-01 -6.78018332e-01 4.74848688e-01 -1.49857926e+00
-4.99779582e-01 -1.08002043e+00 -1.41170472e-01 1.78826928e-01
3.68906230e-01 1.98401317e-01 -1.18755780e-01 -5.42755313e-02
-9.93211865e-02 5.60080528e-01 7.41706431e-01 -3.66278678e-01
-1.86787754e-01 1.60855085e-01 -2.07659230e-01 4.44734812e-01
9.17809546e-01 -4.83124286e-01 -4.46949869e-01 -2.52599269e-01
4.39687312e-01 -1.94133654e-01 5.83538592e-01 -1.65006304e+00
5.62065482e-01 -2.45520413e-01 -2.15285365e-02 -1.05461812e+00
6.20519876e-01 -1.23714602e+00 1.40630886e-01 5.70125878e-01
3.15571725e-01 -2.06630602e-01 4.37502682e-01 4.26617235e-01
-2.29351580e-01 -2.39725575e-01 9.90531087e-01 1.51933953e-01
-1.19308436e+00 6.30329251e-01 -3.81435305e-01 -1.62006125e-01
1.04653597e+00 -3.70526731e-01 -1.85315251e-01 -1.02809094e-01
-2.17503548e-01 2.23123506e-01 2.98215419e-01 4.24989194e-01
7.10165083e-01 -1.21251380e+00 -6.72371268e-01 5.63282013e-01
3.47319007e-01 2.60263622e-01 1.18664722e-03 5.93592286e-01
-3.62452477e-01 6.53294265e-01 -2.71889567e-01 -1.10793972e+00
-8.63774896e-01 5.40021479e-01 4.76896316e-01 2.01485291e-01
-4.04169768e-01 3.20899427e-01 -5.67539990e-01 -6.13734782e-01
1.64413713e-02 -4.43973273e-01 -3.23199898e-01 -2.30639521e-02
3.76746863e-01 3.06884646e-01 5.09416759e-01 -6.00790858e-01
-3.57901037e-01 9.53796387e-01 4.67344195e-01 1.71987282e-03
1.56860423e+00 3.82102728e-02 4.53788713e-02 5.87926269e-01
8.18541586e-01 -1.34396434e-01 -1.13094056e+00 -1.92491338e-01
8.92270803e-02 -1.68343723e-01 4.37395096e-01 -4.28204238e-01
-1.00871587e+00 1.49102235e+00 1.18945789e+00 3.03360492e-01
8.45827162e-01 -1.86850935e-01 7.92184055e-01 7.87444472e-01
5.77869534e-01 -1.39059913e+00 -1.74302459e-01 7.22529590e-01
7.17832983e-01 -1.60698366e+00 -1.32528573e-01 -3.15919191e-01
-3.37545455e-01 1.26389110e+00 6.66863203e-01 7.41057545e-02
1.14904857e+00 7.99743176e-01 2.20851153e-01 -3.74510109e-01
-6.18816972e-01 -5.83729088e-01 -7.68940076e-02 9.77616966e-01
1.74781650e-01 8.92328918e-02 -9.60376393e-03 2.15676859e-01
-2.11301595e-01 3.37457329e-01 3.49039197e-01 8.30695093e-01
-9.13753331e-01 -1.23772156e+00 -3.21605384e-01 4.01976764e-01
1.83992416e-01 2.40044981e-01 -1.20107951e-02 8.07103574e-01
6.44773543e-01 1.08385646e+00 1.58884510e-01 -6.86349928e-01
6.54288054e-01 3.97254854e-01 2.12662488e-01 -5.22582352e-01
7.81098977e-02 -4.64221328e-01 -7.75067657e-02 -4.52417493e-01
-5.13742566e-01 -5.92328370e-01 -1.47729290e+00 -3.65646303e-01
-3.22430670e-01 -9.37325880e-03 1.41650629e+00 8.35148156e-01
4.65686768e-01 4.25781876e-01 7.84909785e-01 -8.93700540e-01
-6.41843557e-01 -9.04058158e-01 -5.41989267e-01 -5.23375459e-02
3.30194056e-01 -1.06870246e+00 -1.26712143e-01 -5.92866950e-02] | [7.9723801612854, -1.940273404121399] |
625a9acf-8bdb-4edc-aaf5-5180add06319 | alphaevolve-a-learning-framework-to-discover | 2103.16196 | null | https://arxiv.org/abs/2103.16196v2 | https://arxiv.org/pdf/2103.16196v2.pdf | AlphaEvolve: A Learning Framework to Discover Novel Alphas in Quantitative Investment | Alphas are stock prediction models capturing trading signals in a stock market. A set of effective alphas can generate weakly correlated high returns to diversify the risk. Existing alphas can be categorized into two classes: Formulaic alphas are simple algebraic expressions of scalar features, and thus can generalize well and be mined into a weakly correlated set. Machine learning alphas are data-driven models over vector and matrix features. They are more predictive than formulaic alphas, but are too complex to mine into a weakly correlated set. In this paper, we introduce a new class of alphas to model scalar, vector, and matrix features which possess the strengths of these two existing classes. The new alphas predict returns with high accuracy and can be mined into a weakly correlated set. In addition, we propose a novel alpha mining framework based on AutoML, called AlphaEvolve, to generate the new alphas. To this end, we first propose operators for generating the new alphas and selectively injecting relational domain knowledge to model the relations between stocks. We then accelerate the alpha mining by proposing a pruning technique for redundant alphas. Experiments show that AlphaEvolve can evolve initial alphas into the new alphas with high returns and weak correlations. | ['Beng Chin Ooi', 'Zhaojing Luo', 'Gang Chen', 'Meihui Zhang', 'Wei Wang', 'Can Cui'] | 2021-03-30 | null | null | null | null | ['stock-prediction'] | ['time-series'] | [-5.96686602e-01 1.83337498e-02 -4.30965692e-01 -3.95407438e-01
-1.42151460e-01 -5.74421763e-01 4.11508262e-01 1.28252253e-01
-2.84011029e-02 1.01182079e+00 -1.54872477e-01 -3.93857867e-01
-3.69847924e-01 -1.65918946e+00 -5.46684027e-01 -4.28679764e-01
-7.32759118e-01 6.07413292e-01 6.59075260e-01 -7.26338267e-01
4.59530771e-01 4.18069124e-01 -1.77031052e+00 6.32723451e-01
1.01422787e+00 1.58687341e+00 -2.34981269e-01 1.71997353e-01
-3.18799227e-01 1.13585746e+00 -9.32739258e-01 -8.50724578e-01
8.75639677e-01 -3.08250129e-01 -2.06993282e-01 -3.96870613e-01
-4.85943884e-01 -9.98580530e-02 5.89279346e-02 1.09907937e+00
-2.67454963e-02 -3.23912174e-01 5.67375898e-01 -1.67041063e+00
-4.56673831e-01 1.33566189e+00 -7.98345208e-01 4.27174062e-01
-1.55604146e-02 4.31944504e-02 1.24350595e+00 -8.46199036e-01
2.39894256e-01 1.05129135e+00 5.83827496e-01 -2.71809269e-02
-9.36345994e-01 -1.31332660e+00 2.01965809e-01 9.00518298e-02
-1.22687638e+00 2.04203054e-01 7.50295877e-01 -4.29429710e-01
8.77301753e-01 4.52077001e-01 1.18780982e+00 6.84051096e-01
3.47935706e-01 9.91420150e-01 1.24570489e+00 -8.41135755e-02
5.42513967e-01 4.53189939e-01 2.55953074e-01 4.65084344e-01
9.83155906e-01 6.21823549e-01 -8.10481191e-01 -5.61922073e-01
8.36801112e-01 5.64741753e-02 1.38721066e-02 -1.93382293e-01
-1.14698875e+00 1.36249518e+00 1.77352801e-01 2.69128770e-01
-4.33820188e-01 -3.73549312e-01 2.96798408e-01 9.04198468e-01
3.38370770e-01 1.11623514e+00 -9.06941175e-01 -3.07761841e-02
-6.41611755e-01 6.03418648e-01 9.66453016e-01 1.10044718e+00
8.97162497e-01 4.69027668e-01 -6.61525056e-02 4.87406641e-01
1.30657315e-01 4.72860068e-01 7.62788057e-01 -4.47486699e-01
5.30946195e-01 1.24705565e+00 -2.04301402e-01 -8.78931701e-01
-6.00654721e-01 -9.36356783e-01 -8.82946372e-01 3.84483814e-01
-1.46832198e-01 -3.49771976e-01 -6.77541137e-01 1.40625560e+00
-6.65018857e-02 3.98418233e-02 3.05963546e-01 3.72817993e-01
3.29062879e-01 7.37499177e-01 -4.97685909e-01 -7.40368366e-01
1.00479698e+00 -5.51138282e-01 -5.84411025e-01 5.02508543e-02
5.04844844e-01 -4.02609289e-01 7.84403741e-01 8.52973163e-01
-1.38271463e+00 -3.36692303e-01 -1.41875422e+00 8.08104515e-01
-4.29861516e-01 -4.23193157e-01 1.04898763e+00 7.04124868e-01
-6.39757156e-01 7.22856283e-01 -4.31411088e-01 8.47655416e-01
3.70448589e-01 5.81426203e-01 2.94510007e-01 8.61788452e-01
-1.71010458e+00 6.99680507e-01 8.15577745e-01 -3.25270772e-01
-3.47617656e-01 -1.16677523e+00 -6.50677979e-01 9.63206887e-02
5.42092860e-01 -5.74063420e-01 1.14342988e+00 -9.18756247e-01
-1.29056859e+00 2.29903743e-01 6.65365458e-01 -1.11368358e+00
4.64419633e-01 -5.39611988e-02 -9.30907488e-01 -1.89256832e-01
3.80917862e-02 1.15844823e-01 8.34582508e-01 -1.07635093e+00
-1.14012754e+00 -2.38156065e-01 -5.58921471e-02 -2.24209249e-01
-4.77285534e-01 -5.82934767e-02 2.49576002e-01 -1.25342691e+00
-6.82664290e-02 -6.19708955e-01 -5.79458952e-01 -7.91283965e-01
-3.46571386e-01 -1.94593966e-02 4.30919051e-01 5.25473766e-02
1.75787044e+00 -1.78981507e+00 -1.71623021e-01 1.16756725e+00
2.60745525e-01 2.35086933e-01 3.94368738e-01 5.40194213e-02
-1.65595561e-01 2.67157316e-01 -2.42624968e-01 5.94114542e-01
-1.94085136e-01 2.86464542e-01 -8.27964365e-01 -2.10877344e-01
3.95733029e-01 8.23498845e-01 -6.77785337e-01 -3.99352074e-01
-2.39570290e-01 -4.01218563e-01 -6.66009903e-01 4.44724768e-01
-6.45909131e-01 -2.54125744e-01 -4.88522083e-01 8.57940078e-01
6.51745677e-01 -3.12957242e-02 -3.28983851e-02 -2.30869856e-02
-3.05266559e-01 3.02591384e-03 -1.44443595e+00 5.67601621e-01
-1.38522059e-01 -2.96133101e-01 -3.82860243e-01 -8.09919000e-01
1.68702078e+00 -2.45932043e-01 6.11353874e-01 -6.24958277e-01
3.01526338e-01 5.03797174e-01 3.41978788e-01 3.03536579e-02
4.46511805e-01 -4.13861006e-01 -3.11392158e-01 6.44187272e-01
-1.70310006e-01 6.85341954e-02 5.75246990e-01 1.45813972e-01
9.69456792e-01 -4.25250679e-01 5.51139772e-01 -5.17230511e-01
4.47224498e-01 1.56852648e-01 1.07639670e+00 6.39962614e-01
5.23051322e-01 1.13082342e-01 9.58671749e-01 -8.46061647e-01
-8.80616903e-01 -1.15334070e+00 -3.29713941e-01 9.35086966e-01
6.95288032e-02 -7.63169289e-01 -6.95280060e-02 -7.02545226e-01
5.43668509e-01 7.06525564e-01 -7.25219727e-01 -3.58490378e-01
-4.59479958e-01 -1.36127484e+00 4.07147169e-01 7.73695290e-01
7.08991766e-01 -9.15676236e-01 -7.16754973e-01 3.10110599e-01
4.73265409e-01 -3.74590188e-01 -6.90238178e-02 5.81946790e-01
-8.34314048e-01 -1.10634446e+00 -3.50881189e-01 -4.61196363e-01
2.72350639e-01 -2.36546427e-01 1.33113182e+00 -7.17920437e-02
6.00294769e-02 -3.21490973e-01 -5.59711218e-01 -1.14660120e+00
-6.20142519e-01 6.44731447e-02 4.16314185e-01 3.30626070e-01
4.64514434e-01 -9.42899346e-01 -3.39126110e-01 4.63964432e-01
-8.70413482e-01 -1.89622134e-01 9.99860048e-01 9.58060801e-01
6.13273859e-01 3.78458589e-01 1.00406623e+00 -1.19331098e+00
9.93221879e-01 -7.55887270e-01 -1.03316450e+00 3.61412466e-01
-1.24404073e+00 6.53823614e-01 9.68792558e-01 -6.30019188e-01
-1.07320499e+00 -1.47602350e-01 2.50324368e-01 -6.12378776e-01
6.79657936e-01 8.71150255e-01 -5.07095158e-02 1.69422761e-01
8.24306309e-01 5.49920611e-02 6.19221255e-02 -4.33910280e-01
1.77668884e-01 4.10540789e-01 5.12609422e-01 -6.59614503e-01
1.03064322e+00 1.53622359e-01 2.79929727e-01 -3.19943964e-01
-6.01509750e-01 -5.28277233e-02 -3.81378531e-01 1.04998723e-01
8.50699991e-02 -6.75413609e-01 -7.27578759e-01 2.31810048e-01
-5.70001185e-01 8.72335136e-02 -6.62964880e-01 4.16418046e-01
-4.77443457e-01 2.87543237e-02 -6.17587030e-01 -8.38785648e-01
-4.39708769e-01 -6.06408060e-01 3.87176126e-01 1.12561747e-01
-3.33247274e-01 -5.21179199e-01 2.20107198e-01 -4.22206908e-01
3.58230531e-01 3.05733144e-01 1.11942923e+00 -1.31460476e+00
-6.12157583e-01 -1.27157956e-01 2.26485491e-01 3.12966317e-01
-1.24001317e-01 1.07697450e-01 -3.33590925e-01 -7.49043748e-02
1.95311159e-01 -2.48406693e-01 9.11919296e-01 -6.86151683e-02
1.10059464e+00 -6.70729756e-01 -4.87303920e-02 9.05324876e-01
9.12173927e-01 8.62059593e-01 5.47887802e-01 7.76009977e-01
-5.11126406e-02 3.61187279e-01 9.93104219e-01 9.89401758e-01
-7.46453255e-02 4.43585008e-01 1.28259614e-01 1.63002014e-01
7.26371944e-01 -4.16184157e-01 3.61827821e-01 1.03938568e+00
-3.62520874e-01 5.49339235e-01 -6.07598007e-01 2.51439273e-01
-1.89892697e+00 -1.12285638e+00 7.96304643e-02 2.06383085e+00
1.35518301e+00 7.60614336e-01 5.63050389e-01 4.57262278e-01
4.59076762e-01 -1.95541292e-01 -6.37898266e-01 -5.16569018e-01
-6.21733844e-01 5.79706907e-01 5.93818843e-01 1.01670697e-01
-1.05049920e+00 5.90295494e-01 6.09741068e+00 6.11988425e-01
-9.41617191e-01 -5.52765250e-01 6.49804413e-01 -2.47446805e-01
-9.04125333e-01 -6.50411332e-03 -1.27964485e+00 6.68844283e-01
9.50234711e-01 -9.31938946e-01 -1.54258594e-01 1.42841411e+00
-5.75460136e-01 4.78899419e-01 -9.51217413e-01 8.33946526e-01
-3.06269795e-01 -1.68017650e+00 5.23580849e-01 7.11351931e-02
8.97305191e-01 -4.08031613e-01 3.11354816e-01 6.83978856e-01
8.31793547e-01 -9.59866226e-01 6.37585104e-01 4.36285287e-01
4.19149816e-01 -1.23045123e+00 8.78969848e-01 2.20420361e-01
-1.16720939e+00 -6.38988793e-01 -6.79491580e-01 -1.15671210e-01
-4.63773124e-02 8.00409257e-01 -6.80010796e-01 6.36169076e-01
8.35891962e-01 6.51221097e-01 -7.24118769e-01 8.85030270e-01
9.65441242e-02 4.03778255e-01 -3.47576648e-01 -3.99516314e-01
1.47804379e-01 -3.80101711e-01 2.83435911e-01 1.03790092e+00
5.93617618e-01 2.13885397e-01 1.46741077e-01 7.98572600e-01
2.30817467e-01 2.29225427e-01 -5.79489172e-01 3.50087017e-01
5.96829355e-01 8.47864866e-01 -3.97412568e-01 -5.36019325e-01
-4.08603281e-01 3.47658247e-02 6.53801253e-03 -5.98993413e-02
-7.74257600e-01 -6.81886613e-01 6.81305051e-01 3.23082030e-01
3.41391623e-01 3.40514034e-01 -7.17933059e-01 -1.38938344e+00
5.52364066e-02 -1.29872561e+00 9.49331582e-01 -2.52151072e-01
-1.48399508e+00 9.80675519e-01 2.96066642e-01 -1.61263323e+00
-6.77007556e-01 -7.30285645e-01 -4.44688320e-01 5.61986029e-01
-1.05282211e+00 -5.22456229e-01 1.33939579e-01 9.30741787e-01
2.31356069e-01 -1.20791876e+00 6.12654805e-01 -2.53557891e-01
-2.54926234e-01 7.01053917e-01 -1.52572453e-01 1.06665105e-01
2.15058997e-01 -1.28239751e+00 4.23128724e-01 5.97330511e-01
2.51683474e-01 8.85710239e-01 5.27230203e-01 -1.04567730e+00
-1.01808798e+00 -1.16396761e+00 4.46324587e-01 -1.48094416e-01
9.81232524e-01 -2.44160742e-01 -1.19721258e+00 6.37532115e-01
-7.28591383e-02 -2.00894371e-01 7.76996315e-01 2.09053293e-01
-7.41352558e-01 -6.98828995e-01 -9.52452660e-01 6.00269735e-01
8.83639812e-01 1.19241841e-01 -1.01122904e+00 2.47415639e-02
7.97718167e-01 7.23564401e-02 -1.20828843e+00 1.04396284e+00
4.38653171e-01 -1.34532249e+00 9.74176705e-01 -9.03848112e-01
2.89565414e-01 -3.59954417e-01 1.38580009e-01 -1.41319811e+00
-4.73274738e-01 -9.06215966e-01 -6.64252102e-01 9.22010541e-01
9.21565056e-01 -8.74056518e-01 8.43386173e-01 3.72853726e-01
-1.76400840e-02 -1.00784326e+00 -6.67955041e-01 -1.14365077e+00
1.21782832e-01 -2.78415978e-01 1.37819099e+00 1.03026319e+00
5.78833044e-01 4.32314843e-01 -3.12573493e-01 -2.34547079e-01
6.55297577e-01 7.80942976e-01 7.34278381e-01 -1.60240877e+00
-7.59361565e-01 -8.28236997e-01 -5.51165819e-01 -5.36819339e-01
-1.73810855e-01 -8.35581481e-01 -5.85904241e-01 -4.29636449e-01
-4.54738215e-02 -8.47353816e-01 -6.21750116e-01 4.49489087e-01
1.57318443e-01 -1.87644318e-01 -6.77543785e-03 4.54596311e-01
2.61458308e-02 3.91309857e-01 8.94438446e-01 -1.16406642e-01
-6.84690595e-01 7.21147895e-01 -1.22497749e+00 8.10602248e-01
1.17210090e+00 -1.45188823e-01 -6.53773427e-01 5.04813969e-01
8.93290460e-01 1.24469846e-01 -3.39607328e-01 -6.32393837e-01
1.16897762e-01 -4.35269892e-01 4.64423001e-01 -8.97601247e-01
2.44761501e-02 -4.45003361e-01 2.89937258e-01 7.32727289e-01
-4.28585261e-01 6.37988150e-01 -1.70664340e-02 3.63702059e-01
-6.00914001e-01 -1.75724372e-01 7.43976235e-01 -3.36282551e-01
-4.93175000e-01 3.43586564e-01 2.43871417e-02 1.30208120e-01
1.42250335e+00 2.03852765e-02 -4.14464563e-01 -2.98359483e-01
-4.79273885e-01 3.40725839e-01 -2.56286692e-02 5.19917071e-01
8.16338241e-01 -1.46264744e+00 -7.83079982e-01 7.78561294e-01
8.85534286e-02 1.73763521e-02 -4.94352072e-01 3.50894600e-01
-4.58281040e-01 3.04149181e-01 -6.36473954e-01 -2.23236591e-01
-8.14257562e-01 9.83774483e-01 1.65136501e-01 -5.99763632e-01
-4.13313299e-01 6.59334898e-01 3.42974544e-01 -4.25183494e-03
2.39282206e-01 -6.03385866e-01 -4.53422248e-01 2.97642410e-01
9.30756509e-01 2.76085734e-01 6.54200763e-02 -1.61063741e-03
-1.83232740e-01 2.35957742e-01 -3.28136414e-01 -2.83733048e-02
1.63504517e+00 5.39955556e-01 -3.53520066e-01 5.00054002e-01
7.97465503e-01 2.60462046e-01 -8.42770100e-01 -2.00325936e-01
5.69132149e-01 -6.00833654e-01 -5.62301278e-01 -6.90616906e-01
-1.35341763e+00 4.03640777e-01 -4.88323607e-02 6.80590510e-01
1.17038226e+00 1.86351817e-02 6.61459088e-01 5.79499602e-01
5.92882335e-01 -1.24346006e+00 1.51674718e-01 3.63906652e-01
1.04888177e+00 -5.65636575e-01 1.33423388e-01 -3.84159654e-01
-1.20273542e+00 1.23763573e+00 9.11317170e-01 -3.95413667e-01
1.03998542e+00 1.08598387e+00 -8.84466693e-02 -1.19720958e-01
-1.24422312e+00 -1.01638228e-01 -3.66846509e-02 5.24682343e-01
-4.86048311e-02 6.01956360e-02 -4.56648290e-01 1.56968272e+00
-9.94901240e-01 -9.90030020e-02 6.41084731e-01 8.04829478e-01
-7.36766636e-01 -1.21261215e+00 -3.53253216e-01 1.08316076e+00
-3.12241971e-01 -1.55517057e-01 -5.19234478e-01 8.83237481e-01
1.08666196e-01 3.76075089e-01 1.67406186e-01 -1.14337289e+00
6.58943951e-01 -5.65877669e-02 -2.23943423e-02 -4.81147110e-01
-1.00104427e+00 -1.11412615e-01 1.85188860e-01 -3.40222657e-01
2.14763060e-02 -6.79550827e-01 -1.02603698e+00 -3.79384786e-01
-2.10457057e-01 5.50889611e-01 -3.31893191e-02 4.16170239e-01
1.71764776e-01 2.04438627e-01 1.14856172e+00 1.86669648e-01
-1.09749746e+00 -7.11155653e-01 -9.65936124e-01 2.85074025e-01
-1.69254482e-01 -8.31757069e-01 -3.05402726e-01 -3.76458794e-01] | [4.543420791625977, 4.175474166870117] |
ac2af3aa-c0c0-48fe-bbd7-117793b6922b | recursive-neural-structural-correspondence | null | null | https://aclanthology.org/P18-1202 | https://aclanthology.org/P18-1202.pdf | Recursive Neural Structural Correspondence Network for Cross-domain Aspect and Opinion Co-Extraction | Fine-grained opinion analysis aims to extract aspect and opinion terms from each sentence for opinion summarization. Supervised learning methods have proven to be effective for this task. However, in many domains, the lack of labeled data hinders the learning of a precise extraction model. In this case, unsupervised domain adaptation methods are desired to transfer knowledge from the source domain to any unlabeled target domain. In this paper, we develop a novel recursive neural network that could reduce domain shift effectively in word level through syntactic relations. We treat these relations as invariant {``}pivot information{''} across domains to build structural correspondences and generate an auxiliary task to predict the relation between any two adjacent words in the dependency tree. In the end, we demonstrate state-of-the-art results on three benchmark datasets. | ['Sinno Jialin Pan', 'Wenya Wang'] | 2018-07-01 | null | null | null | acl-2018-7 | ['extract-aspect', 'fine-grained-opinion-analysis'] | ['natural-language-processing', 'natural-language-processing'] | [ 5.09082735e-01 2.86673278e-01 -3.70691568e-01 -7.03112006e-01
-7.82831490e-01 -8.04280996e-01 4.84325886e-01 4.92402136e-01
-2.25745127e-01 1.11163759e+00 5.17185926e-01 -3.42568427e-01
2.05727324e-01 -7.77933657e-01 -5.17726064e-01 -5.41073799e-01
3.65765303e-01 6.03935122e-01 1.80715218e-01 -5.81451237e-01
4.30363357e-01 1.17955670e-01 -1.07994986e+00 6.45260870e-01
1.35876024e+00 7.85298884e-01 -1.46109775e-01 1.21099539e-01
-6.80450261e-01 8.70047867e-01 -1.02263916e+00 -8.18482161e-01
2.25484706e-02 -5.68416178e-01 -1.06831431e+00 1.62743837e-01
3.31962407e-01 -1.05395548e-01 5.74770421e-02 1.18470907e+00
4.58761066e-01 6.88155070e-02 1.06492198e+00 -7.58275688e-01
-7.67160952e-01 7.61923254e-01 -6.89666271e-01 1.58777654e-01
2.06275403e-01 -3.65490675e-01 1.24762917e+00 -8.32858026e-01
8.81915748e-01 1.09714365e+00 2.70249665e-01 4.79604155e-01
-9.75264192e-01 -4.69213903e-01 6.14568114e-01 2.62168258e-01
-8.62966299e-01 -4.27927256e-01 1.20436680e+00 -2.36716554e-01
1.22186542e+00 -1.88179672e-01 4.77982044e-01 1.12756073e+00
2.01980099e-01 9.28878605e-01 1.00201654e+00 -6.35078907e-01
3.13052475e-01 1.31079257e-01 5.27979493e-01 4.87265706e-01
4.59754884e-01 -5.54533541e-01 -6.90522432e-01 -5.05548455e-02
1.94279611e-01 -4.69619602e-01 -3.16614330e-01 -4.38447118e-01
-8.05600703e-01 9.80410337e-01 3.09902251e-01 4.54261780e-01
-4.57932621e-01 -4.39925343e-01 6.74158812e-01 4.44845855e-01
1.10098755e+00 7.75506675e-01 -9.86329734e-01 1.06764711e-01
-6.72291458e-01 1.05820321e-01 1.07429779e+00 9.34086502e-01
7.57239282e-01 -3.80978622e-02 -1.94659948e-01 1.15475833e+00
-1.25419691e-01 4.33928013e-01 5.11740923e-01 -7.55603611e-01
9.05591667e-01 9.41497266e-01 9.27878097e-02 -9.78674710e-01
-2.92626768e-01 -4.98495519e-01 -1.04285789e+00 -3.12413186e-01
3.00998300e-01 -5.17041385e-01 -9.38685000e-01 1.62874079e+00
3.36035848e-01 -2.63680309e-01 4.61168349e-01 6.45427227e-01
9.52393949e-01 7.17577040e-01 -6.27282560e-02 -5.13002813e-01
1.35608876e+00 -1.07787240e+00 -8.56230974e-01 -6.88554466e-01
7.07017362e-01 -6.16753280e-01 9.33988273e-01 2.51126975e-01
-8.64714444e-01 -3.79936814e-01 -9.92966652e-01 -1.82162806e-01
-5.26324749e-01 1.37026519e-01 4.91268992e-01 2.48015285e-01
-6.58117175e-01 3.98100495e-01 -5.53466380e-01 -3.18484247e-01
5.30094028e-01 3.64089817e-01 -4.36373353e-01 -1.52744696e-01
-1.42969382e+00 1.04437685e+00 5.79342067e-01 -1.03385597e-02
-1.86656922e-01 -4.51456964e-01 -1.00061834e+00 5.39254118e-03
4.15835828e-01 -9.80191112e-01 1.31751990e+00 -1.39089930e+00
-1.48806691e+00 1.16030097e+00 -5.06606579e-01 -4.99287069e-01
-2.97873691e-02 -4.61147040e-01 -2.92438745e-01 1.35638162e-01
3.34838212e-01 4.95711535e-01 1.00373018e+00 -1.25541854e+00
-8.01161528e-01 -5.81067979e-01 3.52665901e-01 4.82373625e-01
-5.66692472e-01 9.59883779e-02 -3.69117558e-01 -8.42360914e-01
4.26640399e-02 -6.52437925e-01 -2.66260982e-01 -5.65191388e-01
-4.22563791e-01 -6.03068650e-01 5.69117486e-01 -7.81032443e-01
1.25963330e+00 -1.81640184e+00 5.16394198e-01 -1.05176136e-01
1.18293114e-01 5.26258349e-01 -3.50928366e-01 4.13686991e-01
-3.32832448e-02 -4.95819338e-02 -5.30108988e-01 -4.56553958e-02
-6.32138923e-02 2.69179761e-01 -3.99108529e-01 7.10454881e-02
6.39033318e-01 8.95868421e-01 -9.16790664e-01 -5.20910203e-01
-1.11238420e-01 1.23165861e-01 -5.25213957e-01 2.92198032e-01
-4.46736693e-01 4.16275680e-01 -6.98276818e-01 4.14942473e-01
6.66603744e-01 -8.90345797e-02 3.28117311e-01 -2.83614457e-01
2.06632823e-01 7.19924271e-01 -7.31404603e-01 1.75348723e+00
-5.65240264e-01 3.82774889e-01 -4.29012068e-02 -1.43701696e+00
1.14401829e+00 1.31783992e-01 2.07099214e-01 -6.75499260e-01
2.16128349e-01 1.52161315e-01 -5.93269318e-02 -5.57331264e-01
6.57054365e-01 -4.59501743e-01 -3.55625808e-01 3.51733267e-01
2.35872030e-01 -4.44868237e-01 4.87732857e-01 1.65737495e-02
8.84334266e-01 2.09097546e-02 7.14835167e-01 -2.62230873e-01
8.67089450e-01 3.44759554e-01 8.00104678e-01 4.16466773e-01
-2.69258134e-02 5.50942421e-01 7.10691571e-01 -3.68500769e-01
-7.57804573e-01 -8.65248263e-01 1.08166695e-01 1.23949993e+00
1.95693836e-01 -3.80477637e-01 -8.00529599e-01 -1.18359709e+00
-3.07176411e-01 9.30713534e-01 -4.86696571e-01 -1.99524626e-01
-7.69346058e-01 -6.72148883e-01 2.60634452e-01 5.77707291e-01
7.57250011e-01 -1.10317719e+00 -1.92671075e-01 2.78114915e-01
-6.94630504e-01 -1.15701330e+00 -3.26041967e-01 4.36540842e-01
-9.51049626e-01 -9.12656367e-01 -6.60413206e-01 -1.17313755e+00
8.52019250e-01 7.25038499e-02 1.51405334e+00 -3.27188134e-01
4.32902604e-01 -2.54895270e-01 -6.98092282e-01 -6.48215771e-01
-2.72978008e-01 6.33676708e-01 -1.48292869e-01 -2.50669450e-01
8.18605959e-01 -4.56497252e-01 -4.53898907e-01 -1.22382985e-02
-8.66023839e-01 -1.31088331e-01 6.58615232e-01 1.01325500e+00
5.60597718e-01 5.13136610e-02 8.22949946e-01 -1.32091010e+00
1.11987364e+00 -1.74615920e-01 -2.90898860e-01 4.11204368e-01
-2.52973795e-01 2.99788088e-01 8.42255950e-01 -2.47072592e-01
-1.41973233e+00 -8.36384967e-02 -3.22661132e-01 2.23723501e-01
-3.77931118e-01 9.76416051e-01 -4.47518498e-01 5.92153013e-01
7.39846647e-01 1.24611422e-01 -4.09585118e-01 -3.12376916e-01
5.04513025e-01 9.15529847e-01 3.49962503e-01 -6.70880497e-01
7.14247227e-01 2.40836397e-01 -3.21709543e-01 -8.20363045e-01
-1.66101074e+00 -4.30017412e-01 -1.08403885e+00 3.47830653e-01
8.21157873e-01 -9.02630508e-01 1.47808909e-01 3.13510627e-01
-1.50457001e+00 6.17038980e-02 -4.41891938e-01 1.67950869e-01
-3.14426184e-01 4.24291790e-01 -3.61453652e-01 -3.40250373e-01
-7.92588115e-01 -7.63021171e-01 1.10666978e+00 3.48202169e-01
-5.25945902e-01 -1.09186912e+00 2.32799932e-01 5.42211652e-01
1.51421696e-01 -7.91672170e-02 1.14379060e+00 -9.88664508e-01
3.10854241e-02 -1.98478937e-01 -1.63310483e-01 7.39255190e-01
4.68835235e-01 -2.20335618e-01 -7.68334925e-01 -9.78567079e-02
3.91719550e-01 -4.04434294e-01 9.93727505e-01 3.27608466e-01
7.60905743e-01 -4.20619309e-01 -2.28291631e-01 2.26131022e-01
9.52349126e-01 9.55426842e-02 4.10235137e-01 3.17515999e-01
6.22181237e-01 1.03819799e+00 8.19443047e-01 1.27132282e-01
4.75930750e-01 3.64110708e-01 -1.09911293e-01 -1.29118171e-02
-6.92143589e-02 -1.96939871e-01 4.97661948e-01 1.13016474e+00
3.18180285e-02 -5.21233797e-01 -7.47621834e-01 8.04854393e-01
-1.90234232e+00 -6.19133532e-01 -7.06871450e-02 1.69303095e+00
1.13640094e+00 3.62832695e-01 -2.51630723e-01 -1.57252729e-01
7.11772621e-01 4.59108949e-01 -6.09655917e-01 -6.97193265e-01
-3.44923526e-01 4.47243840e-01 8.34676251e-02 4.14576828e-01
-1.31123924e+00 1.47021985e+00 5.22472763e+00 7.91671038e-01
-9.32230592e-01 -1.80818155e-01 5.80702662e-01 2.81223297e-01
-3.59889448e-01 -3.17689851e-02 -7.74770379e-01 1.14035960e-02
6.77125037e-01 -2.13610694e-01 -1.99305505e-01 6.75835848e-01
1.87193945e-01 -1.10447900e-02 -9.61319923e-01 6.04240775e-01
3.46970171e-01 -1.13811624e+00 4.13014114e-01 -3.15610170e-01
1.03293741e+00 -1.51486903e-01 -2.87282795e-01 3.30311716e-01
4.12465811e-01 -7.87255466e-01 2.17873439e-01 1.26665369e-01
5.72879255e-01 -9.49643195e-01 1.20588660e+00 4.06949937e-01
-9.09897208e-01 3.33858013e-01 -5.26621401e-01 -2.30678856e-01
2.61412024e-01 8.16327810e-01 -7.08705902e-01 9.01616156e-01
3.34815681e-01 8.74427438e-01 -4.99841213e-01 5.15370548e-01
-9.39908683e-01 6.29238129e-01 -1.69986654e-02 -1.41637772e-01
1.83946908e-01 -4.03493404e-01 4.38702732e-01 1.17024779e+00
2.49862000e-02 1.69220701e-01 6.32729903e-02 3.63961130e-01
-4.41062182e-01 3.83327842e-01 -7.82735944e-01 -7.85990581e-02
1.07043222e-01 1.18728936e+00 -6.22225344e-01 -3.88481408e-01
-5.65863192e-01 1.17115235e+00 6.36761785e-01 4.51345801e-01
-4.21722084e-01 -5.56935072e-01 7.01779187e-01 -1.97363272e-01
5.28624475e-01 -9.12005156e-02 -4.50911164e-01 -1.61262977e+00
2.27743194e-01 -1.11850023e+00 4.04599547e-01 -4.56224203e-01
-1.71047342e+00 6.64555609e-01 -8.96786153e-02 -1.11554587e+00
-4.58377331e-01 -6.81999028e-01 -5.65982282e-01 8.10402274e-01
-1.77659214e+00 -1.21327996e+00 1.67084321e-01 3.73256594e-01
9.89645898e-01 -1.94329992e-01 1.04633307e+00 -1.20037727e-01
-5.39809346e-01 4.24205869e-01 3.56328599e-02 3.76484096e-01
1.05590034e+00 -1.37476945e+00 4.86712754e-01 9.03863132e-01
1.92396566e-01 7.89930403e-01 8.70831072e-01 -6.69723392e-01
-9.42725778e-01 -1.05941892e+00 1.28489828e+00 -3.58311683e-01
6.96148276e-01 -2.30751529e-01 -9.70051467e-01 5.23518801e-01
6.74528062e-01 -3.86761606e-01 9.04612303e-01 5.14211118e-01
-4.46017623e-01 -1.92876548e-01 -9.15995061e-01 5.95892370e-01
9.34614778e-01 -4.96906102e-01 -1.38118076e+00 1.41009569e-01
7.52860308e-01 -2.62293041e-01 -7.02119589e-01 5.67939281e-01
2.57364839e-01 -7.06719577e-01 6.36725366e-01 -9.00598407e-01
8.95846486e-01 -2.99776018e-01 1.07082073e-03 -1.86018574e+00
-5.96040748e-02 -2.05714568e-01 -1.40887231e-01 1.61163068e+00
8.21508467e-01 -5.13993800e-01 8.17371249e-01 3.51412922e-01
-1.52681723e-01 -6.86873138e-01 -6.56698763e-01 -4.21553910e-01
6.64844334e-01 -7.73497820e-02 3.62823963e-01 9.14486766e-01
2.31070548e-01 1.26702356e+00 -1.22870237e-01 9.91197899e-02
3.16203713e-01 5.49381793e-01 5.96588790e-01 -1.26944375e+00
-6.15967661e-02 -2.60663629e-01 -4.02514078e-02 -1.46372795e+00
8.24862778e-01 -7.42465079e-01 9.84101892e-02 -1.98090231e+00
2.46231079e-01 6.27442971e-02 -2.25923374e-01 3.19045037e-01
-5.54689646e-01 -1.05056748e-01 -2.08130553e-01 -7.66065195e-02
-7.23622024e-01 8.68924320e-01 1.27021992e+00 -5.04164934e-01
-2.50917852e-01 1.39616102e-01 -1.24066126e+00 9.76775587e-01
1.19306445e+00 -4.51880187e-01 -5.64485490e-01 -7.34647751e-01
2.09087834e-01 -1.30798459e-01 -4.04750496e-01 -4.99959886e-01
1.50990173e-01 -2.39422381e-01 2.65545368e-01 -7.78209150e-01
6.05565123e-02 -5.92853904e-01 -8.36508751e-01 1.20108165e-01
-4.48608309e-01 5.14107943e-03 1.16102919e-01 4.26356167e-01
-7.78673887e-01 -4.49018538e-01 5.81928372e-01 -1.87667266e-01
-7.78876543e-01 -5.04567400e-02 -2.77146459e-01 5.89569271e-01
7.08376110e-01 5.97031973e-02 -5.04516721e-01 -3.78089845e-01
-5.07948756e-01 3.16191316e-01 4.67219204e-01 5.21399260e-01
5.22976696e-01 -9.42224085e-01 -8.63156974e-01 -1.43213561e-02
3.26915592e-01 5.05772471e-01 6.75494298e-02 3.54456782e-01
-3.00799310e-01 5.11254191e-01 -9.79724750e-02 -2.50876039e-01
-1.35564661e+00 4.72393870e-01 1.63317367e-01 -8.06828022e-01
-2.44627699e-01 8.32829356e-01 4.01873142e-01 -7.43596315e-01
-4.71220212e-03 -3.08387965e-01 -6.95658743e-01 5.01749575e-01
3.77053708e-01 -1.04038984e-01 3.19029480e-01 -6.00420356e-01
-3.20681930e-01 6.42231941e-01 -5.01966774e-01 1.35699064e-02
1.51983845e+00 -4.12172824e-01 -3.99634212e-01 3.54378879e-01
1.07922304e+00 2.23344774e-03 -9.33301151e-01 -3.86195451e-01
3.65666687e-01 1.66101661e-02 -2.74789751e-01 -7.24656165e-01
-7.76656210e-01 8.69705558e-01 -1.19085517e-02 2.25492686e-01
1.27168357e+00 2.46202469e-01 8.48659635e-01 7.44072080e-01
6.81724250e-02 -1.32758653e+00 7.22450987e-02 1.12362254e+00
8.14987361e-01 -1.33006644e+00 3.20266634e-02 -6.92209125e-01
-9.12721395e-01 1.06568122e+00 7.13762462e-01 -1.18261695e-01
3.96537364e-01 7.05307499e-02 2.95641929e-01 -1.81134060e-01
-7.72168517e-01 -3.63479555e-01 2.95672178e-01 8.10274184e-01
7.13153541e-01 -1.57490984e-01 -6.87664509e-01 6.92703903e-01
-3.38747799e-01 -1.21648788e-01 3.97157013e-01 1.00119448e+00
-6.36653483e-01 -1.50398278e+00 -3.86154689e-02 5.48544645e-01
-6.92930579e-01 -2.28160322e-01 -9.50984836e-01 3.85533541e-01
1.36053517e-01 1.28190231e+00 -1.17107667e-01 3.02265044e-02
6.92539930e-01 2.44294986e-01 4.52324569e-01 -1.07972467e+00
-5.77545285e-01 7.58149549e-02 6.17316127e-01 2.19651517e-02
-8.14359725e-01 -5.46934903e-01 -1.13737452e+00 7.27834925e-02
-2.65447944e-01 4.08438206e-01 1.99978858e-01 1.26586485e+00
3.11753750e-01 6.41658902e-01 5.34964740e-01 -4.58936185e-01
-5.52611411e-01 -1.15231991e+00 -3.99558961e-01 6.14772558e-01
2.15578958e-01 -3.66435081e-01 -7.65242279e-02 2.53377259e-01] | [11.376290321350098, 6.802160739898682] |
af07af70-e86c-43e0-87d1-81bbdf6c877f | a-data-driven-latent-semantic-analysis-for | 2207.14687 | null | https://arxiv.org/abs/2207.14687v7 | https://arxiv.org/pdf/2207.14687v7.pdf | A Data-driven Latent Semantic Analysis for Automatic Text Summarization using LDA Topic Modelling | With the advent and popularity of big data mining and huge text analysis in modern times, automated text summarization became prominent for extracting and retrieving important information from documents. This research investigates aspects of automatic text summarization from the perspectives of single and multiple documents. Summarization is a task of condensing huge text articles into short, summarized versions. The text is reduced in size for summarization purpose but preserving key vital information and retaining the meaning of the original document. This study presents the Latent Dirichlet Allocation (LDA) approach used to perform topic modelling from summarised medical science journal articles with topics related to genes and diseases. In this study, PyLDAvis web-based interactive visualization tool was used to visualise the selected topics. The visualisation provides an overarching view of the main topics while allowing and attributing deep meaning to the prevalence individual topic. This study presents a novel approach to summarization of single and multiple documents. The results suggest the terms ranked purely by considering their probability of the topic prevalence within the processed document using extractive summarization technique. PyLDAvis visualization describes the flexibility of exploring the terms of the topics' association to the fitted LDA model. The topic modelling result shows prevalence within topics 1 and 2. This association reveals that there is similarity between the terms in topic 1 and 2 in this study. The efficacy of the LDA and the extractive summarization methods were measured using Latent Semantic Analysis (LSA) and Recall-Oriented Understudy for Gisting Evaluation (ROUGE) metrics to evaluate the reliability and validity of the model. | ['Mahmoud El-Haj', 'Elaine L. L. Pang', 'Daniel F. O. Onah'] | 2022-07-23 | null | null | null | null | ['extractive-summarization'] | ['natural-language-processing'] | [-1.94652304e-02 5.06472826e-01 -1.77289113e-01 1.22287542e-01
-6.05162382e-01 -3.92077029e-01 8.10323954e-01 1.03749549e+00
-9.74274352e-02 7.57470191e-01 1.25948572e+00 -3.47566158e-02
-5.97342491e-01 -5.68320096e-01 1.38958484e-01 -8.28243077e-01
-1.85476840e-01 6.30475581e-01 6.40191808e-02 -1.05310082e-02
8.18010569e-01 1.67586893e-01 -1.69934392e+00 5.82378983e-01
1.11445928e+00 1.59621194e-01 3.97249907e-01 7.70966411e-01
-7.21518278e-01 2.51578271e-01 -1.12902200e+00 1.82810888e-01
-2.53796101e-01 -5.22686422e-01 -6.72939479e-01 2.39383131e-01
4.82859723e-02 -8.33762717e-03 3.74816895e-01 7.20974445e-01
6.90002680e-01 1.81139435e-03 9.84773517e-01 -9.20236051e-01
-2.38242418e-01 4.93753970e-01 -8.45160961e-01 3.52444172e-01
7.44241953e-01 -3.23804080e-01 7.55350173e-01 -5.57238698e-01
1.01448429e+00 1.57527709e+00 3.53580773e-01 -1.12410940e-01
-1.01083815e+00 -1.03801437e-01 -1.07268050e-01 -9.23929736e-02
-1.06122279e+00 -1.73798739e-03 4.74978417e-01 -7.47715592e-01
1.22535205e+00 6.42056346e-01 8.96403611e-01 3.54203016e-01
6.54527247e-01 5.48151970e-01 9.64842677e-01 -5.48574328e-01
3.99497658e-01 5.36325336e-01 7.86565542e-01 1.94338754e-01
8.04741323e-01 -8.79827619e-01 -5.86691141e-01 -6.41792119e-01
-5.60390279e-02 1.26216814e-01 2.72775982e-02 -1.27788156e-01
-1.16906250e+00 9.51172471e-01 -3.83507490e-01 5.64538240e-01
-8.57483089e-01 -4.39829618e-01 9.21674609e-01 -3.46803591e-02
1.08901978e+00 3.91960680e-01 -3.15497592e-02 -2.71610647e-01
-1.54084563e+00 2.90383160e-01 9.08542752e-01 5.68634510e-01
4.15921867e-01 -1.12681046e-01 -5.02678096e-01 7.40101457e-01
4.38983440e-01 2.54858494e-01 9.86254156e-01 -5.36231995e-01
1.15350649e-01 1.10868442e+00 -1.27839357e-01 -1.28401601e+00
-4.92192984e-01 -2.79545128e-01 -5.98369777e-01 -1.22324772e-01
-2.93783963e-01 -1.56613544e-01 -8.50466490e-01 9.82210100e-01
4.05786395e-01 -5.94915628e-01 3.54084224e-01 2.65180975e-01
1.26542032e+00 1.24400377e+00 3.73910248e-01 -6.88872874e-01
1.93406093e+00 -4.18350816e-01 -1.28337181e+00 3.74654174e-01
7.01741755e-01 -8.90718102e-01 7.06583738e-01 5.34779727e-01
-1.09161568e+00 -1.87108159e-01 -1.09706819e+00 -1.62542582e-01
-6.52736545e-01 1.12765498e-01 3.51931959e-01 6.44194663e-01
-1.15212917e+00 2.83616781e-01 -6.20786369e-01 -1.14789987e+00
3.20772856e-01 6.03903607e-02 -2.67166138e-01 2.87133962e-01
-8.64261925e-01 7.91855931e-01 8.92328143e-01 -5.94672978e-01
-3.29349995e-01 -7.33880043e-01 -4.21483785e-01 3.51846009e-01
-8.00529197e-02 -8.77060235e-01 6.11567855e-01 -2.70598054e-01
-1.02780545e+00 8.07413280e-01 -4.21707064e-01 -4.63839412e-01
3.29521120e-01 -2.61174887e-01 -1.14981569e-01 6.47626877e-01
4.60593492e-01 5.30302703e-01 4.93214875e-01 -1.13472962e+00
-8.14463675e-01 -5.73006451e-01 -6.75634980e-01 4.36239481e-01
-4.70022261e-01 1.97195098e-01 -1.68375522e-01 -6.45586073e-01
3.59259814e-01 -3.45723659e-01 6.00959212e-02 -8.14856350e-01
-5.61965525e-01 -6.60669386e-01 1.14588773e+00 -1.01366794e+00
1.75303185e+00 -1.88969064e+00 3.63270147e-03 1.24579921e-01
5.39623916e-01 2.46644299e-02 4.85088021e-01 1.25303018e+00
-5.92551306e-02 3.74445915e-01 -1.31226733e-01 -1.32160634e-01
-3.43582146e-02 -8.60020593e-02 -3.82977486e-01 3.27301055e-01
-3.71728569e-01 4.64517653e-01 -7.12589860e-01 -9.34848309e-01
-6.80022454e-03 5.11289597e-01 -2.14128494e-01 -6.33328781e-02
-2.26034839e-02 -7.00300038e-02 -4.36714858e-01 3.19348484e-01
5.42122006e-01 1.22232147e-01 1.47806704e-01 -1.36077523e-01
-5.33165753e-01 2.00627953e-01 -9.23180521e-01 1.32445550e+00
1.83162272e-01 9.29948270e-01 -3.25826436e-01 -8.80040467e-01
1.14093506e+00 5.92878282e-01 6.59024239e-01 -2.32993543e-01
-1.36812329e-01 -6.30048439e-02 -3.96806270e-01 -6.99578583e-01
9.95129228e-01 -1.62530482e-01 -5.38542010e-02 8.25177729e-01
-8.95224418e-03 -1.49785683e-01 4.06839579e-01 7.86962271e-01
6.42295003e-01 -1.36426136e-01 1.04538178e+00 -7.33040631e-01
8.13182369e-02 3.82711649e-01 -6.87935427e-02 4.56331581e-01
2.24712133e-01 4.03787225e-01 1.10757232e+00 -1.89934969e-01
-1.16641724e+00 -6.85429096e-01 -2.97781199e-01 8.45502019e-01
-3.55292588e-01 -8.34452510e-01 -8.16421866e-01 -1.96943015e-01
-1.65749520e-01 1.19218063e+00 -6.35997891e-01 7.84692019e-02
-3.52766030e-02 -1.12014127e+00 2.38174975e-01 -2.98287809e-01
3.01672250e-01 -8.32711041e-01 -1.19311464e+00 6.25958294e-02
-1.33557424e-01 -1.30667299e-01 2.93567359e-01 -1.31501317e-01
-1.31434178e+00 -6.10444963e-01 -1.14148462e+00 -4.56671685e-01
7.22982049e-01 1.80064440e-01 7.21516967e-01 -3.65470707e-01
-3.95247102e-01 4.98886883e-01 -3.82758886e-01 -8.32333863e-01
-5.43843150e-01 2.63340741e-01 -1.22637555e-01 -6.29140794e-01
4.93826240e-01 -4.24293995e-01 -6.36516273e-01 -2.96090305e-01
-1.11626828e+00 1.62434608e-01 4.46838856e-01 3.13434362e-01
2.30885923e-01 3.54442030e-01 7.35595822e-01 -9.29046810e-01
1.51966941e+00 -8.86480689e-01 8.69071186e-02 2.08673358e-01
-8.86327982e-01 4.70359847e-02 -2.66362935e-01 1.14525505e-03
-1.16107261e+00 -6.24580503e-01 4.03535813e-01 2.52961338e-01
-2.45201603e-01 7.89379120e-01 1.34620577e-01 9.77980316e-01
6.89097047e-01 3.29267859e-01 1.39101848e-01 -5.73387444e-01
3.73496085e-01 1.02041161e+00 1.09717220e-01 -8.88659731e-02
-1.44572869e-01 5.85500002e-01 -2.21999049e-01 -1.39739966e+00
-3.91095281e-01 -1.09453762e+00 -5.64488947e-01 -4.34136510e-01
8.53464007e-01 -6.15287662e-01 -5.55659056e-01 -2.96682119e-03
-1.20797312e+00 5.90689838e-01 -6.19396627e-01 4.14042413e-01
-4.57821965e-01 6.87196314e-01 3.81469876e-02 -9.38955247e-01
-8.91882658e-01 -6.73630774e-01 1.06289625e+00 1.54576302e-01
-9.66459036e-01 -1.02921045e+00 4.80482042e-01 1.31112203e-01
2.74840176e-01 5.33583045e-01 1.27906394e+00 -1.13737357e+00
2.61158466e-01 -5.46268582e-01 -3.33831497e-02 -2.48183325e-01
4.45424020e-01 3.40605751e-02 -7.80101120e-01 -7.48474300e-02
2.72759765e-01 3.64085644e-01 9.36304748e-01 9.03188765e-01
2.43576497e-01 -7.41594434e-01 -7.03848839e-01 -1.94218248e-01
1.30278432e+00 3.30466062e-01 7.18406320e-01 5.85760474e-01
3.61560494e-01 1.18573892e+00 4.38828409e-01 7.03654110e-01
1.06953010e-01 2.32321367e-01 -2.52255380e-01 1.11791320e-01
-8.64834785e-02 -5.88625036e-02 2.76371479e-01 9.95588422e-01
1.48196205e-01 -2.79941767e-01 -1.10120142e+00 6.60672605e-01
-1.74138510e+00 -9.99700189e-01 -3.22624952e-01 2.05966115e+00
6.22400939e-01 1.01685882e-01 3.27429563e-01 1.24781437e-01
7.32948422e-01 1.70434564e-01 -6.87644929e-02 -9.53471482e-01
-1.04052506e-01 -4.81879175e-01 1.52914628e-01 4.23251212e-01
-6.53189659e-01 6.22746587e-01 6.26363468e+00 1.01977456e+00
-8.21873307e-01 -1.37966229e-02 5.22237122e-01 -2.07490340e-01
-4.43770766e-01 1.59504279e-01 -9.05755043e-01 4.73301709e-01
1.20075524e+00 -9.40239608e-01 -7.39101589e-01 6.50100112e-01
9.04096544e-01 -1.05221307e+00 -4.42229450e-01 4.41824645e-01
3.74325991e-01 -1.23880124e+00 7.22291291e-01 3.29413176e-01
7.59008288e-01 -4.48058188e-01 -1.81978837e-01 -1.13301180e-01
-1.46455437e-01 -6.02430761e-01 3.61735851e-01 8.86498272e-01
4.80579674e-01 -6.85968041e-01 9.59933758e-01 2.67500371e-01
-5.13062358e-01 2.18278691e-01 -5.57775915e-01 -2.37667374e-02
2.89047837e-01 7.50984371e-01 -1.38373113e+00 6.78215683e-01
4.89242107e-01 6.87889278e-01 -5.54531038e-01 1.20584834e+00
2.26704359e-01 7.68945634e-01 -2.56987751e-01 -2.11046755e-01
2.06963301e-01 -2.86930442e-01 1.19244921e+00 1.51482642e+00
5.39956093e-01 -8.08738098e-02 -3.86733919e-01 4.62829202e-01
5.71013629e-01 7.08494604e-01 -6.82149351e-01 -3.95242363e-01
3.43014479e-01 1.03115785e+00 -1.50126803e+00 -7.96414673e-01
1.77933425e-01 5.34408629e-01 -4.65221196e-01 1.28267661e-01
9.82000902e-02 -4.46018308e-01 -1.21648043e-01 4.12093371e-01
8.09702650e-02 1.63391791e-02 -6.55533791e-01 -5.48672795e-01
-3.49586934e-01 -5.49999535e-01 6.49475396e-01 -8.05675447e-01
-6.32408023e-01 4.84784245e-01 8.47442448e-01 -1.02481246e+00
-2.44025081e-01 1.72849417e-01 -7.66028106e-01 8.25162053e-01
-5.47080934e-01 -8.83136570e-01 -1.20895337e-02 -1.45649105e-01
9.02297020e-01 -3.40750337e-01 9.56473529e-01 -4.28380966e-01
-3.92181993e-01 -2.88099974e-01 7.26192713e-01 -7.65535831e-01
6.93800986e-01 -1.45188844e+00 -8.36814269e-02 4.46379453e-01
-4.55686718e-01 8.25251639e-01 1.39766765e+00 -1.38415027e+00
-5.37508965e-01 -3.42543721e-01 1.27661872e+00 -1.49737984e-01
4.26060468e-01 -1.46227879e-02 -9.12011087e-01 1.16948716e-01
9.00242329e-01 -1.48795223e+00 1.19076419e+00 1.23369411e-01
3.93862247e-01 3.35740566e-01 -1.05723023e+00 4.88410473e-01
-8.59969333e-02 1.42327100e-01 -1.16376972e+00 4.78959203e-01
6.59264445e-01 2.59156287e-01 -7.50761867e-01 -2.24447012e-01
5.47888637e-01 -8.73785555e-01 7.07089424e-01 -5.07316470e-01
6.31961882e-01 1.02278426e-01 2.35342473e-01 -1.24637139e+00
-9.18474272e-02 -7.09635794e-01 6.21804520e-02 1.45519722e+00
2.20987171e-01 -4.50369626e-01 6.35949552e-01 4.35932398e-01
-4.91688587e-02 -5.81202209e-01 -7.95057416e-01 -6.78480566e-02
-1.74769796e-02 1.78619213e-02 1.70549586e-01 7.55352974e-01
7.11728752e-01 4.51541692e-01 -8.78718197e-02 -3.35712373e-01
5.37796497e-01 -1.29966065e-01 6.18114352e-01 -1.67151511e+00
6.54526591e-01 -5.31527102e-01 -6.11028790e-01 -8.03105161e-02
-4.21218872e-01 -7.37544537e-01 -5.24131060e-01 -2.46208119e+00
6.67254806e-01 4.53983396e-01 1.94836691e-01 -2.68465374e-02
-9.39558670e-02 -3.92594725e-01 5.07319793e-02 8.57100666e-01
-5.78741193e-01 2.43054122e-01 8.46231699e-01 1.15607396e-01
-7.05732465e-01 -4.60807867e-02 -9.48896945e-01 7.56579161e-01
8.77987981e-01 -7.15476692e-01 -6.47166073e-01 3.59974891e-01
6.59219384e-01 -1.57603458e-01 -5.92819899e-02 -7.02054560e-01
3.76099408e-01 2.27019146e-01 1.53928459e-01 -1.34256613e+00
-7.54013658e-02 -3.91054869e-01 1.46845669e-01 5.98439336e-01
-5.69516122e-01 8.15032423e-02 4.92942512e-01 6.00906730e-01
-1.98114499e-01 -4.24211949e-01 3.35133106e-01 -1.56939745e-01
-3.69248956e-01 -6.36882007e-01 -1.01667631e+00 -2.38778770e-01
1.07021201e+00 -7.89402902e-01 -2.36704022e-01 -6.32609606e-01
-9.89454389e-01 2.76095957e-01 2.04509050e-01 4.78818491e-02
5.73541224e-01 -8.39857817e-01 -9.73329127e-01 -2.45597497e-01
-7.38714263e-02 -9.11209583e-02 6.54603362e-01 9.29687619e-01
-8.27762485e-01 1.04092383e+00 -4.38922584e-01 -4.54947740e-01
-1.76370811e+00 3.83778393e-01 -4.45464611e-01 -4.54079360e-01
-9.26532626e-01 2.80496120e-01 3.39547664e-01 2.98616320e-01
1.40373588e-01 -1.47468328e-01 -1.15399981e+00 1.16840231e+00
6.77685857e-01 1.15907776e+00 -8.68038535e-02 -6.26858592e-01
-1.04520947e-01 4.45131004e-01 -3.75957876e-01 -3.14559162e-01
1.51531804e+00 -7.10564911e-01 -7.17423737e-01 8.28806758e-01
1.09081495e+00 8.10760260e-02 -3.85188162e-01 1.53831035e-01
5.82936406e-01 6.77766427e-02 2.02457264e-01 -8.53626370e-01
-1.10431753e-01 7.55069911e-01 6.01255059e-01 8.00335348e-01
8.98334205e-01 2.17568368e-01 2.07003593e-01 2.56777495e-01
-6.37413204e-01 -1.23647547e+00 -2.76115656e-01 3.25542055e-02
1.14662874e+00 -7.62492657e-01 7.68645346e-01 -1.96271524e-01
-8.56150389e-01 1.19122636e+00 -2.19019383e-01 2.02002496e-01
6.87270224e-01 -4.33531590e-02 -1.68606624e-01 -9.80955482e-01
-7.41295815e-01 2.53770113e-01 4.75657105e-01 3.95773172e-01
7.05991685e-01 8.57274011e-02 -1.36380982e+00 6.63174331e-01
-5.33544660e-01 -1.62158355e-01 6.38745546e-01 8.79599154e-01
-1.12823188e+00 -4.92382258e-01 -7.01371193e-01 7.26503253e-01
-8.43231618e-01 3.41214128e-02 -6.32652521e-01 7.31298387e-01
-2.39660800e-01 8.87373328e-01 2.36016467e-01 2.77326494e-01
-1.14566390e-03 3.46692264e-01 -3.09330463e-01 -8.35329175e-01
-4.76549208e-01 6.96732640e-01 1.34478942e-01 -5.46889864e-02
-5.73671997e-01 -9.48361933e-01 -1.20883858e+00 -4.00127433e-02
-1.60609275e-01 9.00149703e-01 1.18324482e+00 7.28257895e-01
5.40576756e-01 6.93719447e-01 4.58702855e-02 -6.22360945e-01
2.44950112e-02 -1.47869408e+00 -7.95250058e-01 -2.49934681e-02
1.59488872e-01 -2.79144913e-01 -2.90686667e-01 2.93437272e-01] | [12.407842636108398, 9.571134567260742] |
19099f8a-1dce-4a70-aacd-64c6fca7bff4 | evaluating-and-modelling-hanabi-playing | 1704.07069 | null | http://arxiv.org/abs/1704.07069v1 | http://arxiv.org/pdf/1704.07069v1.pdf | Evaluating and Modelling Hanabi-Playing Agents | Agent modelling involves considering how other agents will behave, in order
to influence your own actions. In this paper, we explore the use of agent
modelling in the hidden-information, collaborative card game Hanabi. We
implement a number of rule-based agents, both from the literature and of our
own devising, in addition to an Information Set Monte Carlo Tree Search
(IS-MCTS) agent. We observe poor results from IS-MCTS, so construct a new,
predictor version that uses a model of the agents with which it is paired. We
observe a significant improvement in game-playing strength from this agent in
comparison to IS-MCTS, resulting from its consideration of what the other
agents in a game would do. In addition, we create a flawed rule-based agent to
highlight the predictor's capabilities with such an agent. | ['Diego Perez-Liebana', 'Joseph Walton-Rivers', 'Piers R. Williams', 'Simon M. Lucas', 'Richard Bartle'] | 2017-04-24 | null | null | null | null | ['game-of-hanabi'] | ['playing-games'] | [-1.45912662e-01 5.46880305e-01 2.39962429e-01 5.51391095e-02
-4.20955628e-01 -4.29801971e-01 1.08914292e+00 -2.56759167e-01
-8.28571379e-01 1.09844983e+00 5.16586721e-01 -5.06023347e-01
-3.68294179e-01 -1.02443123e+00 -2.69063145e-01 -6.26900554e-01
-3.08515906e-01 1.28657186e+00 5.66650033e-01 -6.55414999e-01
1.88442960e-01 -2.26995453e-01 -1.39820504e+00 1.94981560e-01
3.71589780e-01 2.91867733e-01 1.57756448e-01 1.04676938e+00
5.04222155e-01 1.85728860e+00 -8.45521092e-01 -6.07735038e-01
4.95926797e-01 -8.14713418e-01 -7.32154250e-01 -1.70677397e-02
-6.64572299e-01 -6.35273039e-01 -1.65716082e-01 5.91153860e-01
6.25722826e-01 2.77670205e-01 5.64579964e-01 -1.17590654e+00
2.08943874e-01 1.17897856e+00 -2.26378426e-01 5.48291244e-02
4.78608966e-01 7.55176961e-01 1.02489185e+00 1.12220243e-01
9.42050576e-01 1.21966088e+00 5.00403047e-01 6.71619236e-01
-1.10508525e+00 -5.56475818e-01 -1.83727648e-02 2.97909617e-01
-1.11161816e+00 -2.61540681e-01 4.15537328e-01 -3.57616127e-01
1.03373945e+00 2.96359748e-01 1.01834512e+00 9.77295458e-01
3.63028646e-02 8.62705290e-01 1.57376206e+00 -6.95619881e-01
8.51313591e-01 -3.11577376e-02 -4.49269801e-01 3.59358341e-01
1.20734237e-01 1.00507057e+00 -4.44994509e-01 -5.85378885e-01
8.67745519e-01 -4.68842089e-01 2.90955216e-01 -3.95890623e-01
-9.64603186e-01 1.17977953e+00 4.32191091e-03 2.57355690e-01
-7.25110054e-01 4.52914029e-01 4.77596670e-02 3.46186519e-01
1.66076496e-01 7.18014896e-01 -3.58548522e-01 -7.31110930e-01
-7.05938578e-01 9.95599627e-01 1.33154106e+00 6.01925910e-01
3.65667313e-01 1.52186211e-02 -3.30633335e-02 2.06520259e-01
6.68512464e-01 -8.96558762e-02 2.58819282e-01 -1.68962908e+00
1.32096335e-01 2.78452009e-01 3.98856997e-01 -4.41814423e-01
-6.70485556e-01 -4.20252949e-01 -2.30042502e-01 1.19791472e+00
5.01453400e-01 -6.25916183e-01 -4.91275042e-01 1.66915119e+00
3.02269787e-01 1.17442332e-01 3.92440706e-01 7.24188328e-01
1.17768243e-01 4.15911466e-01 1.52285993e-01 -3.51068974e-01
1.28958237e+00 -8.62259805e-01 -4.55973327e-01 -1.39878392e-01
9.86740172e-01 -4.11246449e-01 3.25236052e-01 6.79307163e-01
-1.45388258e+00 4.25233431e-02 -9.88071859e-01 7.28110313e-01
1.35446154e-02 -8.23824883e-01 1.04431915e+00 8.96087110e-01
-1.19979763e+00 5.14225066e-01 -8.25020015e-01 -2.70021349e-01
4.75123487e-02 5.69553375e-01 2.97910661e-01 4.40312624e-01
-1.15710366e+00 1.36912251e+00 3.61160040e-01 -4.85786617e-01
-1.02092302e+00 -1.76827937e-01 -4.20849383e-01 4.48570512e-02
1.01197195e+00 -8.99239182e-01 1.91897666e+00 -8.86760473e-01
-1.99125779e+00 4.55078125e-01 4.53351885e-01 -7.50915706e-01
9.39795732e-01 5.41097045e-01 2.53846142e-02 -3.80902618e-01
1.80691674e-01 3.14912826e-01 2.16025263e-01 -1.49437833e+00
-1.15262032e+00 -1.95789844e-01 5.90535462e-01 7.71076739e-01
6.74916625e-01 3.14525068e-01 1.04766235e-01 -2.35025004e-01
-4.83354926e-01 -1.04023695e+00 -9.35905814e-01 -8.40375304e-01
-2.69791454e-01 -2.05184132e-01 7.52488822e-02 -5.22458926e-02
1.04704440e+00 -1.46992362e+00 -1.31584406e-01 3.95142436e-01
4.54568893e-01 4.85066548e-02 1.64759949e-01 7.85953462e-01
3.80898863e-01 2.22384468e-01 2.47768804e-01 -3.03932607e-01
4.68203783e-01 3.89470786e-01 3.10347706e-01 6.71395585e-02
-5.53022504e-01 6.55996919e-01 -8.63828242e-01 -3.43517035e-01
1.46558002e-01 -1.96173400e-01 -8.85524154e-01 1.16977647e-01
-6.22415006e-01 2.83836842e-01 -6.62247837e-01 -1.27601251e-01
2.27021977e-01 3.96894477e-02 7.15342700e-01 1.01019371e+00
-4.18285012e-01 5.20860553e-01 -1.34355116e+00 1.05048978e+00
-1.92429602e-01 2.69897580e-01 2.72525609e-01 -3.34789157e-01
4.43808615e-01 5.99020422e-01 2.45044976e-01 -7.05339015e-01
2.84445733e-01 1.04456946e-01 1.01596224e+00 -7.67648816e-02
5.09973526e-01 -4.66103613e-01 -6.64186180e-02 1.14919746e+00
-2.47738212e-01 1.10422801e-02 3.69446367e-01 4.25209135e-01
1.41910541e+00 2.88978487e-01 6.75164402e-01 -1.23472288e-01
4.58134040e-02 4.78877038e-01 4.12744910e-01 1.69596350e+00
-1.18091404e-01 1.93191811e-01 6.33313835e-01 -3.36889595e-01
-1.03750277e+00 -4.53588217e-01 3.65318716e-01 1.24283230e+00
-7.89081082e-02 -6.10421717e-01 -7.74197400e-01 -3.82211059e-01
-3.29466403e-01 1.19592977e+00 -7.14392841e-01 1.04254194e-01
-4.86790776e-01 -8.30470145e-01 5.08795500e-01 1.59775749e-01
4.79352832e-01 -1.31895459e+00 -1.18507469e+00 8.67137134e-01
2.23294765e-01 -4.56366152e-01 -2.00974792e-01 3.45068038e-01
-2.21630678e-01 -9.83014822e-01 -6.67456612e-02 -1.08719602e-01
-9.47868004e-02 -2.18204945e-01 1.07886684e+00 2.79308379e-01
3.36474597e-01 3.12587708e-01 -5.88973284e-01 -8.08878899e-01
-1.03845525e+00 4.69257077e-03 7.34557360e-02 -4.50382888e-01
3.81055951e-01 -5.86286485e-01 -4.66996789e-01 3.45281839e-01
-4.80257004e-01 4.01902974e-01 4.43960309e-01 7.70784318e-01
-2.83030570e-01 6.53976381e-01 1.91758350e-02 -1.01855659e+00
9.48954463e-01 -5.94186127e-01 -8.06949675e-01 -2.52019584e-01
-6.89778149e-01 6.70438632e-02 5.44543266e-01 -2.94274300e-01
-1.26411736e+00 -2.00088814e-01 4.52737212e-02 3.74288529e-01
-1.66940928e-01 5.63993454e-01 2.45597609e-03 8.22895095e-02
7.34447300e-01 7.66923204e-02 3.63794684e-01 -2.66837865e-01
1.95625678e-01 5.44440687e-01 1.22083329e-01 -7.05935180e-01
6.69500411e-01 1.54505715e-01 -2.12026149e-01 -1.70088768e-01
-1.23667076e-01 4.45936024e-02 -9.22865048e-03 -4.25695062e-01
5.81552446e-01 -7.71969080e-01 -1.48093891e+00 6.16826713e-01
-1.09645832e+00 -9.58826482e-01 -5.96453369e-01 5.84588647e-01
-1.08993232e+00 -2.03511745e-01 -5.28171480e-01 -1.36789167e+00
3.85058612e-01 -1.11581528e+00 2.01237485e-01 2.87070006e-01
-4.44528848e-01 -1.08611000e+00 5.98277271e-01 4.89434451e-01
4.81283724e-01 -6.14671223e-02 5.05766988e-01 -1.21686542e+00
-9.17052090e-01 3.62411030e-02 4.66336906e-01 -4.24508154e-01
-3.63639712e-01 -2.71139950e-01 -5.43425918e-01 -9.40317214e-02
2.20781088e-01 -1.49161518e-01 1.72620207e-01 4.68270421e-01
-1.12231418e-01 -5.89554250e-01 -2.96092868e-01 -2.23517176e-02
1.18534851e+00 9.87649262e-01 7.17176676e-01 1.06883585e+00
3.82196084e-02 7.15457141e-01 5.47754228e-01 7.77311802e-01
8.96113575e-01 1.20507836e+00 3.52883548e-01 5.85972480e-02
1.27246886e-01 -3.37216258e-01 4.10689682e-01 3.57924521e-01
-7.18104362e-01 -3.66194338e-01 -7.71628916e-01 1.10843889e-01
-2.36715126e+00 -1.52550030e+00 -1.15051515e-01 1.94338739e+00
8.34741652e-01 5.26345789e-01 7.17872262e-01 -7.11075515e-02
4.28628445e-01 1.96124669e-02 -3.38482648e-01 -5.39200306e-01
9.72412601e-02 2.28461161e-01 7.03574717e-01 9.45493579e-01
-5.30605912e-01 9.74354863e-01 7.12915182e+00 9.32554901e-01
-9.61804464e-02 4.60891992e-01 4.90341187e-01 -3.57102722e-01
-1.91287726e-01 5.46917737e-01 -5.41425645e-01 5.88256240e-01
1.03320158e+00 -5.79378068e-01 9.37917650e-01 5.21483600e-01
6.82420671e-01 -7.76187956e-01 -9.29720521e-01 2.64698356e-01
-2.44201794e-01 -1.54147816e+00 -4.11834151e-01 8.36066604e-01
6.15715802e-01 -2.74718702e-02 -3.05485576e-01 6.15421653e-01
1.90921605e+00 -1.09853137e+00 1.16286469e+00 3.47315431e-01
-9.11136419e-02 -7.76299596e-01 1.01491499e+00 9.58944976e-01
-7.35557199e-01 -1.62087888e-01 1.85791537e-01 -1.22356129e+00
1.41582578e-01 -3.27324718e-01 -1.17554212e+00 4.74306494e-01
2.98281193e-01 -2.01196656e-01 -9.11748260e-02 9.46678221e-01
-2.37325892e-01 7.54969835e-01 -4.03915584e-01 -5.45560360e-01
5.91535509e-01 -4.35833186e-01 6.39943182e-01 4.80338305e-01
-9.00807753e-02 7.71788418e-01 4.01886255e-01 7.82491446e-01
4.87638682e-01 -2.47544661e-01 -4.79253769e-01 4.13195312e-01
5.82571864e-01 8.71899545e-01 -6.64484859e-01 -4.44972098e-01
-2.17945635e-01 4.11539823e-01 -4.30752411e-02 1.84044257e-01
-5.57448208e-01 2.80342579e-01 5.96296370e-01 1.01512931e-01
2.95231700e-01 1.57018434e-02 -2.54992545e-01 -6.60042167e-01
-4.28159446e-01 -1.38724017e+00 2.83727467e-01 -8.59286547e-01
-9.43418801e-01 4.24365282e-01 9.65831876e-02 -9.37396765e-01
-1.15004945e+00 -1.71609789e-01 -8.38962495e-01 8.05008531e-01
-8.01839530e-01 -8.56815755e-01 3.78561854e-01 3.46062452e-01
3.38075250e-01 -5.43921649e-01 7.29135394e-01 -3.18959743e-01
-1.09210394e-01 1.48236051e-01 2.04332709e-01 -1.06180936e-01
-1.47760004e-01 -1.36631525e+00 5.37565351e-01 5.67570090e-01
1.09937556e-01 3.17504942e-01 1.14622378e+00 -5.99594176e-01
-9.95416522e-01 -3.46289545e-01 4.31422472e-01 -7.04427600e-01
7.26946056e-01 -1.95656970e-01 -1.30807787e-01 9.16840553e-01
5.24325192e-01 -7.92122960e-01 5.91603994e-01 2.48646140e-01
2.35514313e-01 4.93167251e-01 -1.15481615e+00 1.05698621e+00
1.16621625e+00 -7.42641613e-02 -5.83600044e-01 7.61658177e-02
3.09072465e-01 -3.13481301e-01 -4.23752725e-01 -2.13907376e-01
4.53788221e-01 -1.54502141e+00 5.74785650e-01 -5.85058451e-01
1.54115245e-01 -4.60178167e-01 6.57045245e-02 -1.77608490e+00
-6.04026556e-01 -1.14502764e+00 2.78862000e-01 8.47868800e-01
4.19983745e-01 -9.55875039e-01 1.40032601e+00 9.68847573e-01
2.89593101e-01 -2.84949958e-01 -1.05680382e+00 -7.97922373e-01
2.41326466e-01 -5.47898769e-01 6.08768702e-01 6.79482937e-01
5.06772578e-01 3.95549655e-01 -6.61634386e-01 -1.48840815e-01
9.67484057e-01 -3.75502706e-01 1.01788616e+00 -1.05111289e+00
-1.09480441e+00 -5.80633581e-01 -4.00797904e-01 -8.40950727e-01
-3.72811615e-01 -5.15484810e-01 7.48654902e-02 -1.42498767e+00
4.67929751e-01 -6.48977697e-01 2.82090604e-01 4.35381949e-01
6.78429753e-02 -6.43567517e-02 7.50067711e-01 1.85509101e-01
-8.80973756e-01 2.24291652e-01 1.10783017e+00 2.16987476e-01
-2.74605244e-01 2.68876910e-01 -9.57950056e-01 1.02632368e+00
8.52152467e-01 -7.00899124e-01 -2.42842123e-01 1.31310895e-01
3.40950131e-01 5.90979576e-01 3.38966995e-01 -1.02396929e+00
6.98172271e-01 -3.89374524e-01 1.61770638e-02 -1.17295355e-01
4.35191482e-01 -5.46307683e-01 9.13993418e-01 1.00088656e+00
-4.16094571e-01 -7.77637139e-02 -3.20529610e-01 2.55650401e-01
1.89797223e-01 -7.09194183e-01 2.45250240e-01 -8.48299623e-01
-5.93349159e-01 -1.43592283e-01 -1.33314288e+00 -3.94402146e-02
1.20187819e+00 -5.35347939e-01 -3.79635572e-01 -1.21509886e+00
-8.03970516e-01 3.88882667e-01 7.04685748e-01 -3.46584707e-01
1.10281743e-01 -8.97921920e-01 -1.00834513e+00 -2.38599464e-01
-2.85444885e-01 -3.28278035e-01 -6.35992065e-02 5.32823861e-01
-4.82408226e-01 1.56523615e-01 -3.21603388e-01 1.53061926e-01
-1.23867953e+00 3.10049266e-01 5.59384823e-01 -8.48822832e-01
-3.75668049e-01 4.19046313e-01 3.22265252e-02 -5.11996210e-01
-5.95901581e-03 3.85835856e-01 -2.94965029e-01 -1.47858784e-01
4.25344169e-01 3.73179942e-01 -4.32154715e-01 -3.28721851e-01
-1.74870819e-01 -2.08172470e-01 -9.33304802e-02 -1.06298983e+00
1.15523851e+00 -1.17843419e-01 1.12576351e-01 2.90500939e-01
1.62829176e-01 9.56188217e-02 -1.47238779e+00 -1.65283144e-01
1.42482489e-01 -4.18046385e-01 6.03013970e-02 -1.20595002e+00
-5.24082422e-01 -2.81551797e-02 2.41779238e-01 8.79578769e-01
6.20841444e-01 -1.09946735e-01 6.51525185e-02 2.33672127e-01
1.30001128e+00 -1.11383522e+00 -3.35356683e-01 6.70624852e-01
3.24974209e-01 -9.25599396e-01 2.23047659e-01 9.19554159e-02
-1.16037071e+00 6.80075169e-01 4.93784934e-01 -8.96797255e-02
2.64339060e-01 6.53011382e-01 2.19226927e-01 -4.35212970e-01
-1.59481633e+00 -7.74390459e-01 -8.65456164e-01 1.04150927e+00
-1.27940282e-01 3.46156895e-01 -3.57944965e-01 5.66303432e-01
-5.99231482e-01 2.71689087e-01 1.15175402e+00 1.15678656e+00
-4.61199254e-01 -1.42625105e+00 -5.09290576e-01 4.20442492e-01
-2.36174911e-01 -1.04282692e-01 -5.24580359e-01 1.08682632e+00
2.89070040e-01 1.30619180e+00 1.34456098e-01 -4.42003995e-01
1.47618786e-01 -2.72246212e-01 3.76590371e-01 -7.01196671e-01
-1.11220980e+00 5.66453673e-02 7.51697481e-01 -4.47460353e-01
-4.24243808e-01 -1.09643161e+00 -1.03550518e+00 -8.42225254e-01
-2.10531369e-01 5.33753455e-01 3.17720741e-01 9.95526254e-01
-9.76775959e-02 5.15008688e-01 6.32451773e-01 -1.02473676e+00
-7.08204925e-01 -9.52624679e-01 -9.10473108e-01 3.52030285e-02
-1.58057317e-01 -7.67399549e-01 -3.69271785e-01 -6.05742157e-01] | [3.4856793880462646, 1.5815770626068115] |
8590d353-4e3b-4578-85a0-c284d9479bbf | light-sampling-field-and-brdf-representation-1 | 2304.05472 | null | https://arxiv.org/abs/2304.05472v1 | https://arxiv.org/pdf/2304.05472v1.pdf | Light Sampling Field and BRDF Representation for Physically-based Neural Rendering | Physically-based rendering (PBR) is key for immersive rendering effects used widely in the industry to showcase detailed realistic scenes from computer graphics assets. A well-known caveat is that producing the same is computationally heavy and relies on complex capture devices. Inspired by the success in quality and efficiency of recent volumetric neural rendering, we want to develop a physically-based neural shader to eliminate device dependency and significantly boost performance. However, no existing lighting and material models in the current neural rendering approaches can accurately represent the comprehensive lighting models and BRDFs properties required by the PBR process. Thus, this paper proposes a novel lighting representation that models direct and indirect light locally through a light sampling strategy in a learned light sampling field. We also propose BRDF models to separately represent surface/subsurface scattering details to enable complex objects such as translucent material (i.e., skin, jade). We then implement our proposed representations with an end-to-end physically-based neural face skin shader, which takes a standard face asset (i.e., geometry, albedo map, and normal map) and an HDRI for illumination as inputs and generates a photo-realistic rendering as output. Extensive experiments showcase the quality and efficiency of our PBR face skin shader, indicating the effectiveness of our proposed lighting and material representations. | ['Yajie Zhao', 'Yunxuan Cai', 'Wenbin Teng', 'Hanyuan Xiao', 'Jing Yang'] | 2023-04-11 | light-sampling-field-and-brdf-representation | https://openreview.net/forum?id=yYEb8v65X8 | https://openreview.net/pdf?id=yYEb8v65X8 | iclr-2023-3 | ['neural-rendering'] | ['computer-vision'] | [ 3.60383630e-01 -9.84586701e-02 6.95291400e-01 -4.33031648e-01
-3.32339883e-01 -2.25766554e-01 5.46305239e-01 -4.48770851e-01
1.31878063e-01 6.45872712e-01 -1.01899758e-01 -4.75125968e-01
2.12799147e-01 -1.22197056e+00 -7.76679993e-01 -5.29622674e-01
2.04286858e-01 6.87103346e-02 -2.54533757e-02 -3.40589195e-01
-1.43888239e-02 9.93283689e-01 -1.99071360e+00 1.31587908e-01
1.07537246e+00 1.18548691e+00 2.13166028e-01 7.79364049e-01
-5.25370836e-01 6.23759866e-01 -6.56577408e-01 -2.50989169e-01
4.11957294e-01 -3.72484356e-01 -1.13424383e-01 -1.18202291e-01
8.97194207e-01 -6.71851695e-01 -1.16596140e-01 5.30585885e-01
6.07441366e-01 2.93377250e-01 7.63378859e-01 -8.81515145e-01
-7.23525286e-01 -3.54846150e-01 -7.91865170e-01 -3.91179055e-01
2.67006963e-01 2.35483363e-01 4.18094039e-01 -9.18082535e-01
4.39053506e-01 1.35699892e+00 5.71134031e-01 8.91354382e-01
-1.01738048e+00 -8.47713828e-01 3.04977238e-01 -4.68525767e-01
-1.23588228e+00 -3.75763893e-01 1.08110666e+00 -1.31091386e-01
8.10577035e-01 8.43229651e-01 1.03755689e+00 9.38526511e-01
2.15409949e-01 3.63425225e-01 1.53190362e+00 -2.45355278e-01
3.86468470e-01 1.84668809e-01 -1.72745690e-01 9.32818651e-01
5.86144626e-03 3.47222209e-01 -5.31831920e-01 -1.65358081e-01
1.39200497e+00 7.29117543e-02 -4.81178463e-01 -2.92340457e-01
-4.23671126e-01 4.84026045e-01 7.97815263e-01 -3.01982701e-01
-3.07712168e-01 5.12350321e-01 -6.19756384e-03 -6.42771497e-02
8.70611310e-01 -3.10492590e-02 -1.08547941e-01 4.86012809e-02
-8.81460845e-01 1.69041783e-01 6.91308975e-01 7.26899147e-01
8.17151725e-01 4.69816685e-01 8.36686511e-03 1.01388133e+00
6.25900626e-01 9.23615396e-01 -2.59116620e-01 -1.00477922e+00
1.10614309e-02 4.08034444e-01 1.72149986e-01 -9.13581312e-01
-1.30874529e-01 -2.06356645e-01 -8.10359120e-01 1.00949538e+00
-1.64585605e-01 5.91505580e-02 -1.20761836e+00 1.30212724e+00
6.51522398e-01 5.26264131e-01 -2.75876056e-02 1.12131906e+00
1.40857220e+00 8.38186979e-01 -3.46629508e-02 1.14702873e-01
1.25678372e+00 -7.80598640e-01 -5.66891074e-01 2.76582241e-02
1.55370846e-01 -7.25193143e-01 1.54529452e+00 6.01120532e-01
-1.22749507e+00 -4.14435744e-01 -9.44181025e-01 -4.32033747e-01
-1.79495424e-01 -5.82438335e-02 1.01909232e+00 9.56261694e-01
-1.14803874e+00 4.52761829e-01 -7.24293590e-01 5.29018603e-02
4.70919728e-01 -3.30997072e-03 2.06547305e-02 -1.44981652e-01
-8.80468965e-01 5.29510260e-01 -4.45584327e-01 4.21093434e-01
-1.00727201e+00 -1.06737304e+00 -6.83137298e-01 6.63005188e-02
-9.70179662e-02 -9.14166272e-01 9.47778225e-01 -9.70232308e-01
-2.14642739e+00 7.10206985e-01 -1.31932244e-01 1.69066936e-01
5.78071952e-01 -4.25951988e-01 -2.31594473e-01 7.83972368e-02
-6.20335877e-01 4.40694660e-01 8.05219412e-01 -2.03629827e+00
-1.17690794e-01 -2.78623760e-01 3.20604831e-01 3.61598760e-01
-3.00234016e-02 -1.30465120e-01 -3.04257601e-01 -4.92175877e-01
-5.13709001e-02 -4.56300169e-01 -1.00033171e-01 8.67156208e-01
-3.15887928e-01 3.41455281e-01 8.72293234e-01 -8.44206095e-01
7.34400988e-01 -2.01848602e+00 -3.79326612e-01 3.06048691e-01
2.15240642e-01 2.12578848e-01 -1.08732074e-01 6.26823455e-02
-1.13919772e-01 5.14959246e-02 -2.72555858e-01 -6.87080920e-01
-1.06411003e-01 2.04634488e-01 -4.66086835e-01 3.97379458e-01
1.26883760e-01 5.66368341e-01 -6.24487042e-01 -1.08492009e-01
6.24328017e-01 1.66787469e+00 -5.09512067e-01 3.62536371e-01
-3.03845614e-01 6.75909400e-01 -1.63575664e-01 6.72733247e-01
1.29457629e+00 1.18059739e-01 -1.71536162e-01 -2.52433658e-01
-3.26803565e-01 1.73066825e-01 -1.12885940e+00 1.54850316e+00
-1.17786491e+00 5.79005361e-01 4.23567057e-01 -4.21440750e-02
1.22729003e+00 3.80966961e-02 2.93561481e-02 -1.07931721e+00
2.38758668e-01 1.02477789e-01 -5.33239007e-01 -1.35781989e-01
5.77832222e-01 -2.96844929e-01 6.00433111e-01 1.86338052e-01
-5.47970295e-01 -5.37532330e-01 -7.14717984e-01 -9.71445963e-02
6.92941666e-01 6.63543344e-01 -2.72461712e-01 -2.05333993e-01
2.70104617e-01 -4.50102150e-01 2.58036375e-01 2.97357827e-01
4.17571813e-01 9.44256544e-01 -7.07392171e-02 -5.70901096e-01
-7.21966028e-01 -1.38835728e+00 -2.61172980e-01 1.02429307e+00
2.33085737e-01 -5.99497929e-02 -7.36330748e-01 -6.15586676e-02
-1.35191858e-01 8.34446549e-01 -6.01043820e-01 1.15247995e-01
-6.67825758e-01 -4.40363437e-01 2.25508451e-01 3.20425600e-01
5.82651854e-01 -9.29422438e-01 -1.01394880e+00 -1.57902896e-01
3.48508298e-01 -6.30045354e-01 1.18949726e-01 -1.23826332e-01
-8.65473270e-01 -8.58544946e-01 -8.34723473e-01 -2.68419415e-01
6.56914830e-01 3.92463863e-01 1.37740469e+00 3.00325274e-01
-5.01020372e-01 4.11205947e-01 -7.46077076e-02 -5.31334281e-01
-2.13979974e-01 -5.70895016e-01 -3.69184792e-01 3.50806080e-02
-4.04703379e-01 -8.03028405e-01 -1.16819119e+00 2.23483160e-01
-1.03035700e+00 6.68540180e-01 7.05237165e-02 2.08030045e-01
7.60551095e-01 -3.38855952e-01 -3.58474767e-03 -9.56435621e-01
5.67349613e-01 -1.30309597e-01 -7.90470064e-01 2.33826950e-01
-1.94433540e-01 -2.58096159e-01 5.92695892e-01 -3.30870152e-01
-1.66014183e+00 -3.66632938e-01 -3.13288540e-01 -6.12353206e-01
-1.41222566e-01 1.01799637e-01 -2.88534701e-01 -3.43291283e-01
6.20348990e-01 3.01789958e-02 -3.36436868e-01 -5.43838084e-01
3.66183579e-01 5.70351064e-01 3.39120299e-01 -8.67438734e-01
7.19802320e-01 9.03943121e-01 3.01675618e-01 -1.13993883e+00
-3.49307299e-01 1.05062388e-01 -6.87978491e-02 -5.51095605e-01
5.56169033e-01 -8.88339937e-01 -1.08246922e+00 4.97079551e-01
-1.25207710e+00 -6.61747277e-01 -3.88559967e-01 7.57362470e-02
-4.53866512e-01 4.24737856e-02 -4.59634960e-01 -1.34842098e+00
-6.08906925e-01 -1.11806500e+00 1.37666667e+00 5.74265420e-01
2.45756537e-01 -8.88305962e-01 1.45019159e-01 1.53179348e-01
7.14482725e-01 8.71138871e-01 7.46771038e-01 8.39236140e-01
-8.65638256e-01 2.69067675e-01 -6.34508312e-01 3.36001515e-01
2.35301137e-01 4.78559077e-01 -1.71569204e+00 -2.34556906e-02
-6.98174015e-02 -4.73421179e-02 7.40988672e-01 4.10899252e-01
1.45712745e+00 -1.65548921e-02 -1.44750401e-01 1.13096952e+00
1.77862632e+00 2.06443235e-01 9.58269656e-01 -1.73960254e-01
1.09064984e+00 6.93326592e-01 2.58080721e-01 5.35962284e-01
3.06172580e-01 5.32741725e-01 7.59458959e-01 -8.10404301e-01
-5.08188367e-01 -1.91164359e-01 1.27442986e-01 6.06266379e-01
-6.55685425e-01 -3.28091085e-01 -6.58340812e-01 -1.04014829e-01
-1.24992251e+00 -4.91911411e-01 -3.17439377e-01 2.38512564e+00
6.57598615e-01 -2.24271446e-01 -2.72350580e-01 -1.66315883e-01
2.41061285e-01 1.85762182e-01 -6.54687345e-01 -9.56438243e-01
-7.00906292e-02 8.30048561e-01 3.96459818e-01 6.66738868e-01
-4.42645460e-01 7.24557757e-01 5.59524298e+00 5.54665625e-01
-1.55500674e+00 5.22244759e-02 6.73039973e-01 -3.22798789e-01
-1.03022778e+00 -3.42508078e-01 -5.12758374e-01 1.83990166e-01
8.29284966e-01 4.73200798e-01 7.77958035e-01 7.97863483e-01
5.11482000e-01 -1.78802207e-01 -7.70580649e-01 1.19897330e+00
6.33705556e-02 -1.24904561e+00 2.93428063e-01 1.40812382e-01
6.41837537e-01 -2.15437382e-01 3.16793710e-01 5.17108180e-02
3.91148657e-01 -1.29404676e+00 7.91251183e-01 8.55731606e-01
1.33354759e+00 -7.51951575e-01 7.39608258e-02 1.03572004e-01
-1.16805553e+00 4.08722043e-01 -3.75546724e-01 2.48375144e-02
3.38859886e-01 6.77358031e-01 -4.20570076e-01 5.55093586e-01
6.92399561e-01 1.19919628e-01 -3.06461692e-01 9.43365097e-01
-1.80327460e-01 4.41695601e-01 -4.76216137e-01 1.11676678e-01
1.03191438e-03 -5.00296295e-01 1.37586102e-01 1.09005880e+00
4.23052639e-01 1.18475378e-01 -9.90545452e-02 1.29424679e+00
-8.14926401e-02 8.77438709e-02 -5.45539558e-01 4.26715255e-01
1.56656712e-01 1.34377074e+00 -6.69051886e-01 -3.09630055e-02
-3.33492100e-01 1.04802668e+00 2.59882599e-01 5.91016531e-01
-1.06936407e+00 -1.55298233e-01 8.82214725e-01 4.16345447e-01
-1.43616542e-01 -2.02741817e-01 -3.54487717e-01 -9.63283837e-01
-7.61785954e-02 -4.54304665e-01 -4.02389467e-01 -1.30673361e+00
-1.24853384e+00 7.54235804e-01 -1.35300413e-01 -9.64167297e-01
3.51563841e-01 -7.41110563e-01 -7.37760305e-01 1.52024734e+00
-2.07558227e+00 -1.46389079e+00 -9.66436863e-01 5.94345272e-01
2.79278576e-01 4.89354014e-01 9.90806341e-01 3.03236812e-01
-3.34328800e-01 3.02965254e-01 1.52555499e-02 -2.52149493e-01
2.84641117e-01 -1.13836586e+00 5.72926104e-01 4.58299637e-01
-2.41778493e-01 7.45200217e-01 5.62061369e-01 -5.40618122e-01
-1.60122395e+00 -9.42853510e-01 -2.14579791e-01 -8.79271254e-02
-1.78206805e-02 -7.43263602e-01 -9.82959569e-01 1.77217886e-01
1.24641642e-01 2.11792082e-01 5.93009174e-01 -3.74724008e-02
-4.90074873e-01 -1.72191903e-01 -1.49112844e+00 7.56701589e-01
1.26263797e+00 -4.94690657e-01 7.78801367e-02 1.60212189e-01
5.81582427e-01 -7.96680510e-01 -4.98219490e-01 3.19150686e-01
8.39160323e-01 -1.34034789e+00 1.14513361e+00 1.70222670e-02
5.69346607e-01 -4.53661799e-01 -2.09095031e-01 -1.31828749e+00
2.79534105e-02 -5.01859069e-01 -2.69512177e-01 1.15855563e+00
-1.72691345e-01 -7.22982645e-01 6.64320409e-01 8.54241967e-01
-3.31824988e-01 -8.72037053e-01 -6.98501885e-01 -4.14415598e-01
7.74493068e-02 -5.49464405e-01 9.52989638e-01 7.07524717e-01
-1.06126153e+00 -1.59580946e-01 -2.99048305e-01 3.51870477e-01
7.48279333e-01 2.98668802e-01 8.78637791e-01 -1.49774373e+00
-3.36396039e-01 -2.19927549e-01 1.52711466e-01 -1.03366315e+00
7.34742556e-04 -5.27907312e-01 9.73471701e-02 -1.82293892e+00
-2.00374648e-01 -1.05816281e+00 2.37063393e-02 1.16936378e-01
6.23629615e-02 5.41487396e-01 7.34983236e-02 -1.34899423e-01
1.66079119e-01 6.74381733e-01 1.63227642e+00 1.28824726e-01
-4.21423376e-01 -2.48803347e-01 -5.66815913e-01 8.66950989e-01
6.15696430e-01 9.27636027e-03 -5.95625877e-01 -8.21619272e-01
4.83202726e-01 -9.22447070e-03 7.15595067e-01 -9.04826403e-01
-3.99248987e-01 -2.09183976e-01 5.78528345e-01 -4.81758118e-01
8.86396110e-01 -1.06103170e+00 5.05074501e-01 1.17749974e-01
5.05802184e-02 -3.05147737e-01 3.20676476e-01 2.75506288e-01
3.69331568e-01 1.56369448e-01 9.02598202e-01 -1.86101004e-01
-3.60444754e-01 4.24838692e-01 1.61678895e-01 -3.12054336e-01
6.52845144e-01 -5.96910655e-01 -4.57291484e-01 -2.91504085e-01
-3.22932065e-01 -2.45694980e-01 7.94462562e-01 1.84353516e-01
9.46760654e-01 -1.16117477e+00 -5.26816785e-01 4.56901729e-01
-4.68928777e-02 4.83534485e-01 5.48898578e-01 1.29814819e-01
-1.23339427e+00 -2.34990105e-01 -2.45268449e-01 -4.91247267e-01
-1.30445588e+00 8.43382999e-02 6.56769574e-01 1.66450873e-01
-9.09666002e-01 9.63992476e-01 7.89134264e-01 -6.05504394e-01
9.47414041e-02 -6.79678142e-01 1.08570434e-01 -3.91179651e-01
7.46873260e-01 4.09739166e-01 2.15295896e-01 -4.02356535e-01
-9.48019698e-02 7.57162929e-01 4.64737594e-01 -4.30390313e-02
1.36499977e+00 1.13934316e-01 -6.32035285e-02 5.06752729e-01
9.12133574e-01 2.63752371e-01 -1.63424170e+00 2.39396363e-01
-1.09662092e+00 -8.93497109e-01 4.60961342e-01 -1.02396011e+00
-1.33884263e+00 1.24801934e+00 8.59263122e-01 -5.18166423e-02
1.25879490e+00 -4.04602319e-01 8.62455189e-01 -8.41313750e-02
6.28660440e-01 -6.61941409e-01 -2.74083167e-01 2.48389870e-01
1.14323485e+00 -7.85686433e-01 -6.40851557e-02 -9.15117145e-01
-2.16869935e-01 1.00757349e+00 6.26908481e-01 -2.31575266e-01
6.98121130e-01 7.14547098e-01 2.98705488e-01 -3.82864296e-01
-2.32415661e-01 1.28996179e-01 2.76691318e-01 8.47792268e-01
5.82671344e-01 2.01652884e-01 1.94461882e-01 7.71375149e-02
-2.29125485e-01 6.97302595e-02 3.28231692e-01 7.13095725e-01
-2.21095741e-01 -6.81201279e-01 -4.98372555e-01 3.70643288e-01
-1.42281679e-02 -1.63334981e-01 -1.11587919e-01 4.36172158e-01
2.63938338e-01 5.88969886e-01 3.22775424e-01 -2.31115390e-02
4.17238146e-01 -3.78380209e-01 6.51755929e-01 -5.35510123e-01
-7.53789008e-01 8.65692645e-02 -8.97136107e-02 -6.74759626e-01
-3.89651477e-01 2.26448420e-02 -1.29064262e+00 -5.32337964e-01
-2.06060827e-01 -4.02289897e-01 1.19392431e+00 3.00236642e-01
3.44361812e-01 8.66498291e-01 6.19546115e-01 -1.36035466e+00
1.78478733e-01 -6.40018582e-01 -7.19795346e-01 2.83433288e-01
3.58349651e-01 -8.60063076e-01 -2.53655463e-01 -2.45229691e-01] | [9.58371353149414, -3.1157357692718506] |
90cf0b25-67e8-4614-9d2e-4e6fd9664259 | is-a-green-screen-really-necessary-for-real | 2011.11961 | null | https://arxiv.org/abs/2011.11961v4 | https://arxiv.org/pdf/2011.11961v4.pdf | MODNet: Real-Time Trimap-Free Portrait Matting via Objective Decomposition | Existing portrait matting methods either require auxiliary inputs that are costly to obtain or involve multiple stages that are computationally expensive, making them less suitable for real-time applications. In this work, we present a light-weight matting objective decomposition network (MODNet) for portrait matting in real-time with a single input image. The key idea behind our efficient design is by optimizing a series of sub-objectives simultaneously via explicit constraints. In addition, MODNet includes two novel techniques for improving model efficiency and robustness. First, an Efficient Atrous Spatial Pyramid Pooling (e-ASPP) module is introduced to fuse multi-scale features for semantic estimation. Second, a self-supervised sub-objectives consistency (SOC) strategy is proposed to adapt MODNet to real-world data to address the domain shift problem common to trimap-free methods. MODNet is easy to be trained in an end-to-end manner. It is much faster than contemporaneous methods and runs at 67 frames per second on a 1080Ti GPU. Experiments show that MODNet outperforms prior trimap-free methods by a large margin on both Adobe Matting Dataset and a carefully designed photographic portrait matting (PPM-100) benchmark proposed by us. Further, MODNet achieves remarkable results on daily photos and videos. Our code and models are available at https://github.com/ZHKKKe/MODNet, and the PPM-100 benchmark is released at https://github.com/ZHKKKe/PPM. | ['Qiong Yan', 'Kaican Li', 'Jiayu Sun', 'Rynson W. H. Lau', 'Zhanghan Ke'] | 2020-11-24 | null | null | null | null | ['video-matting'] | ['computer-vision'] | [ 2.06253529e-01 -4.27821785e-01 -2.93540120e-01 -2.86842853e-01
-7.35663354e-01 -2.80051947e-01 2.86900192e-01 -5.49791694e-01
-2.44905084e-01 5.15475392e-01 -3.85132432e-02 -4.63734865e-02
1.67563409e-01 -6.82424426e-01 -7.40679562e-01 -6.41816795e-01
3.96996439e-01 2.33826116e-01 1.45514637e-01 -2.26865858e-01
-1.74640298e-01 2.44841009e-01 -1.25992107e+00 5.19348323e-01
1.04086554e+00 1.12922740e+00 4.33064550e-01 5.38515866e-01
2.46989056e-01 5.11818111e-01 -2.34215170e-01 -7.19411373e-01
7.29295075e-01 -2.39753768e-01 -6.76311672e-01 7.06208348e-01
1.02620566e+00 -5.46558082e-01 -6.74819350e-01 1.02109504e+00
3.83263409e-01 -3.62381325e-05 3.82181615e-01 -1.45959210e+00
-7.48132885e-01 2.21079588e-01 -8.33729148e-01 -3.08777448e-02
-8.04543495e-03 4.98085678e-01 1.03339934e+00 -1.01527381e+00
5.91500700e-01 1.27142680e+00 8.80764961e-01 3.78031313e-01
-1.43646502e+00 -7.84105778e-01 1.82915181e-01 2.43490413e-01
-1.53177595e+00 -4.52720076e-01 7.86097825e-01 -5.72191104e-02
6.54724956e-01 3.90646666e-01 8.20879161e-01 1.12817979e+00
2.03922123e-01 1.07300496e+00 1.20040452e+00 1.14748336e-01
4.48380075e-02 -4.89863865e-02 -2.26951510e-01 8.52753282e-01
1.86257154e-01 -2.62216389e-01 -7.29541361e-01 4.34228592e-02
1.02230954e+00 1.55686125e-01 -2.51993030e-01 -1.00648001e-01
-1.23989224e+00 5.74889481e-01 7.40201473e-01 1.63788170e-01
-3.79366636e-01 4.37177271e-01 1.60332471e-01 4.07957792e-01
7.34707057e-01 5.94212525e-02 -2.56875932e-01 2.35892199e-02
-1.22176135e+00 2.10067794e-01 4.75191653e-01 9.44510281e-01
7.81053841e-01 2.60667235e-01 -2.45378725e-02 9.46081161e-01
1.28625482e-01 6.74287438e-01 2.10628837e-01 -1.02350605e+00
5.86589158e-01 5.55127621e-01 -1.52167410e-01 -1.20128500e+00
-8.24038237e-02 -4.39744562e-01 -1.22226143e+00 1.54982820e-01
3.85741532e-01 1.38027340e-01 -1.09681439e+00 1.53839552e+00
4.50213969e-01 2.59574026e-01 -2.19979167e-01 1.10453868e+00
8.62884343e-01 8.65935326e-01 -1.50675267e-01 1.73624620e-01
1.35588646e+00 -1.38440943e+00 -6.20740354e-01 -4.29496586e-01
3.22044432e-01 -1.01220870e+00 1.20735490e+00 4.29287493e-01
-1.16750968e+00 -5.23910284e-01 -1.00573325e+00 -3.74923885e-01
-1.89177945e-01 4.04742628e-01 7.15154290e-01 4.07860339e-01
-1.09345794e+00 7.42257953e-01 -9.16828513e-01 -3.68881762e-01
7.74904668e-01 4.28560555e-01 -4.19782639e-01 -2.25696117e-01
-7.73583114e-01 5.40064394e-01 2.76747137e-01 2.23388478e-01
-9.79440093e-01 -7.44536042e-01 -8.36542428e-01 -9.82073396e-02
3.91703695e-01 -8.93571079e-01 1.02549386e+00 -1.50089335e+00
-1.57219243e+00 9.24838662e-01 -2.13821024e-01 -3.75975549e-01
6.50095224e-01 -1.51667446e-01 -3.18770438e-01 3.28104138e-01
9.98670682e-02 9.49806094e-01 1.28956854e+00 -1.01404023e+00
-3.22663009e-01 -1.86913293e-02 -3.57483141e-02 2.87702739e-01
-6.15141392e-01 -2.52154410e-01 -8.75490546e-01 -1.09898293e+00
-6.18540123e-03 -9.95708287e-01 -9.90179777e-02 6.18785143e-01
-5.33604980e-01 1.34197086e-01 1.03908348e+00 -8.73894513e-01
1.03707027e+00 -2.10970736e+00 2.76579380e-01 2.44682725e-03
3.58651996e-01 4.12910849e-01 -5.28667092e-01 3.25588614e-01
2.52489410e-02 -1.02659687e-01 -4.67578650e-01 -8.12726974e-01
-5.42346537e-02 3.05977523e-01 -3.19506496e-01 5.39528370e-01
1.63061112e-01 1.09843302e+00 -6.20810986e-01 -4.96586263e-01
2.25789815e-01 5.49733102e-01 -5.46375573e-01 -3.32396440e-02
-2.65904427e-01 1.58421904e-01 -1.99901849e-01 1.18167543e+00
1.02694201e+00 -4.60593045e-01 -1.49999978e-02 -5.22183061e-01
6.82500973e-02 1.42902464e-01 -1.10258532e+00 1.98859859e+00
-3.72537315e-01 7.33282864e-01 3.17482769e-01 -7.40619123e-01
7.05664814e-01 -1.32157803e-02 4.56220716e-01 -4.95407403e-01
1.55730888e-01 1.42141730e-01 -4.37625408e-01 -1.64382488e-01
5.28587699e-01 8.80336761e-02 1.12172320e-01 4.16572869e-01
1.96269065e-01 -3.32030445e-01 3.03640306e-01 2.71177709e-01
9.97585893e-01 1.09465621e-01 -2.45416407e-02 -3.21923077e-01
2.46121556e-01 1.40076518e-01 8.56101990e-01 2.79250622e-01
-1.19456768e-01 8.85730147e-01 1.92042068e-01 -7.46210635e-01
-1.02120662e+00 -1.13086128e+00 -2.70394236e-02 7.87319660e-01
3.65279585e-01 -7.28116333e-01 -6.28861308e-01 -5.26158929e-01
3.52664068e-02 2.99924612e-01 -6.27724648e-01 1.30593017e-01
-4.77388382e-01 -8.36305797e-01 3.97897422e-01 5.31573474e-01
1.21941149e+00 -6.97123766e-01 -2.19237834e-01 6.48938268e-02
-3.57728064e-01 -1.22339535e+00 -9.80451465e-01 -3.34425688e-01
-1.00312495e+00 -9.36212063e-01 -8.29452813e-01 -7.11133122e-01
7.81115830e-01 6.76957667e-01 1.02931333e+00 2.13718176e-01
-3.60072792e-01 2.01123625e-01 -8.38756561e-02 -1.74606130e-01
9.98488665e-02 2.67539918e-01 -2.68724803e-02 2.37796098e-01
-5.95247038e-02 -6.84030890e-01 -7.38532841e-01 6.17965817e-01
-1.09169161e+00 6.94143355e-01 7.72167802e-01 9.49468613e-01
7.55605221e-01 -1.08925179e-01 1.31808311e-01 -4.65889156e-01
2.62876272e-01 -1.68458045e-01 -5.72037876e-01 2.91966110e-01
-3.79421651e-01 -3.60104680e-01 6.20893776e-01 -6.49576187e-01
-7.70711958e-01 1.30572870e-01 -2.36909747e-01 -7.48391986e-01
1.90242857e-01 2.68443227e-01 -2.34198958e-01 -3.65531951e-01
3.80724162e-01 3.82011473e-01 2.02995390e-01 -4.31402981e-01
1.52304426e-01 2.64666140e-01 3.72333705e-01 -4.37532991e-01
1.24684608e+00 7.93928027e-01 -6.93872124e-02 -8.51322174e-01
-9.61990714e-01 -1.39469936e-01 -5.30963600e-01 -2.43369728e-01
6.37782514e-01 -1.12038589e+00 -5.78427434e-01 8.16329002e-01
-7.83967853e-01 -8.66026282e-01 -1.06553175e-01 1.95287630e-01
-4.67585802e-01 5.41615367e-01 -7.75777459e-01 -2.42230222e-01
-6.71122253e-01 -1.14128113e+00 1.17777514e+00 2.41591603e-01
-3.04403845e-02 -9.53595519e-01 -3.27415705e-01 8.02427411e-01
5.44138849e-01 2.73363531e-01 3.04024577e-01 1.52723342e-01
-9.27191138e-01 -5.98837622e-02 -4.81109798e-01 4.24462706e-01
7.55473822e-02 -4.07540724e-02 -8.55642676e-01 -6.03901684e-01
-1.66660205e-01 -4.43229020e-01 1.09097850e+00 3.22791785e-01
1.44824600e+00 -5.54054558e-01 -7.78779835e-02 1.22252059e+00
1.38402951e+00 -3.05305719e-01 7.89239943e-01 3.44324499e-01
1.00575483e+00 9.17131305e-02 6.24930620e-01 2.40123436e-01
5.53339064e-01 8.37519467e-01 4.60468531e-01 -4.88746643e-01
-4.80951160e-01 -2.53782868e-01 5.62168062e-01 6.59282088e-01
-1.94986492e-01 -2.63550788e-01 -6.09733224e-01 4.79173183e-01
-2.00060415e+00 -9.40292358e-01 -5.80357313e-02 1.96433711e+00
8.21635723e-01 -9.08249766e-02 2.19460174e-01 -5.82022890e-02
5.27340353e-01 4.26195860e-01 -6.90490842e-01 -4.81538177e-02
-4.06380177e-01 1.60708949e-01 5.50794899e-01 4.87475693e-01
-1.16257918e+00 1.17552257e+00 5.28539801e+00 1.27210033e+00
-1.27525091e+00 2.55991846e-01 9.73538220e-01 -2.92287767e-01
-2.35824540e-01 -4.44135908e-03 -5.45983195e-01 5.60107052e-01
3.45893532e-01 -1.83125749e-01 6.56714618e-01 8.23758781e-01
1.19070187e-01 -8.70955437e-02 -8.28370273e-01 1.34444678e+00
2.20517009e-01 -1.66496921e+00 6.25585616e-02 1.30822212e-01
8.12406421e-01 -1.86295100e-02 3.66377831e-01 2.10174304e-02
2.24814899e-02 -9.77040112e-01 7.35573590e-01 1.99695915e-01
9.79426563e-01 -6.81389630e-01 5.28580368e-01 -4.49997559e-02
-1.32594895e+00 2.36520007e-01 -5.33499777e-01 3.90653089e-02
1.45573765e-01 6.30556405e-01 -5.19832551e-01 5.36134303e-01
7.93565929e-01 1.11337531e+00 -6.34207606e-01 9.39279199e-01
-2.85943329e-01 6.49791837e-01 -5.11620045e-01 2.46130079e-01
4.44421619e-01 -2.47221231e-01 6.51445985e-01 1.01293826e+00
3.09576750e-01 -9.70387924e-03 1.31912276e-01 8.86843860e-01
-4.35132027e-01 -1.21133089e-01 -3.09017360e-01 5.95293418e-02
9.03876051e-02 1.60635507e+00 -7.71659434e-01 -3.49718630e-01
-1.34956792e-01 1.42540896e+00 1.66888833e-01 3.76805842e-01
-1.12261319e+00 -8.26491937e-02 9.62876678e-01 3.77836883e-01
4.27611589e-01 -3.87523800e-01 -4.23304558e-01 -1.57755315e+00
2.16113329e-01 -8.98920417e-01 2.97177732e-01 -8.66142392e-01
-1.43002772e+00 6.53054595e-01 -7.35415071e-02 -1.40850306e+00
3.26448679e-01 -6.06371045e-01 -8.07902217e-01 5.64299941e-01
-1.54157424e+00 -1.64492643e+00 -7.75475979e-01 8.79597485e-01
8.45831513e-01 3.44184786e-02 5.57743073e-01 3.83272409e-01
-8.56990993e-01 8.50600898e-01 -4.31449600e-02 1.98629737e-01
8.96055877e-01 -9.80006337e-01 5.87824106e-01 1.07656288e+00
1.60591781e-01 1.97830603e-01 4.49341774e-01 -6.10158086e-01
-1.69036603e+00 -1.47996998e+00 6.76688313e-01 -1.96015984e-01
6.86660886e-01 -5.21737635e-01 -8.14998925e-01 6.37617946e-01
5.65870762e-01 9.28369239e-02 4.88913536e-01 -2.00961351e-01
-5.24707854e-01 -5.86197376e-01 -1.19879591e+00 8.87027264e-01
1.16492128e+00 -2.52053380e-01 -1.69437930e-01 6.24842048e-01
4.90677446e-01 -6.35148644e-01 -7.55474091e-01 1.26375630e-01
5.48995852e-01 -7.96892881e-01 1.07292295e+00 3.68734822e-02
5.38230598e-01 -5.27365565e-01 -1.00546576e-01 -1.20915115e+00
-4.02544677e-01 -9.61694896e-01 -1.27790689e-01 1.05703044e+00
1.73147231e-01 -7.47257292e-01 7.80131817e-01 5.33179164e-01
-9.08651054e-02 -1.00992560e+00 -1.04241407e+00 -9.31382000e-01
-1.96619362e-01 -3.17080498e-01 4.25526798e-01 9.19366479e-01
-5.07670164e-01 1.66769296e-01 -7.44312167e-01 3.95570789e-03
9.56805408e-01 3.33386183e-01 9.47315037e-01 -6.84606731e-01
-2.99052179e-01 -3.35651398e-01 -4.13256586e-01 -1.16211772e+00
-6.41749278e-02 -8.51312816e-01 -3.07097584e-01 -1.43168998e+00
2.55651474e-01 -4.22880471e-01 2.93698851e-02 8.33307862e-01
-8.02413002e-02 9.68420029e-01 5.65300167e-01 4.44520950e-01
-5.76743066e-01 7.99647331e-01 1.38395596e+00 -4.43721861e-01
-1.12656794e-01 -2.22316787e-01 -6.05919540e-01 7.67436087e-01
1.01901913e+00 -3.87048215e-01 -2.50946432e-01 -7.93003201e-01
6.53651636e-03 -3.84109735e-01 7.03526676e-01 -1.06967258e+00
2.64788568e-01 -2.33578160e-01 5.68581879e-01 -6.44881606e-01
7.43013620e-01 -7.18371332e-01 2.84609824e-01 4.15832609e-01
8.83478895e-02 1.99933738e-01 4.13684040e-01 3.61398131e-01
-1.94450304e-01 2.02179670e-01 1.00886607e+00 5.60789406e-02
-7.71254361e-01 7.61635184e-01 -1.46259278e-01 -1.24162987e-01
1.13181281e+00 -3.16877365e-01 -3.86132061e-01 -4.86808449e-01
-4.08370882e-01 2.79013455e-01 7.91672111e-01 3.13037306e-01
8.66764247e-01 -1.52900684e+00 -8.06572616e-01 4.94314432e-02
-1.26357436e-01 8.43437016e-02 6.01933599e-01 1.12923503e+00
-7.83615410e-01 -8.11580867e-02 -3.61894190e-01 -6.18087709e-01
-1.52908528e+00 1.98875770e-01 1.59697101e-01 -2.91103005e-01
-7.60332882e-01 9.64336574e-01 2.88466275e-01 -1.55085534e-01
1.81865722e-01 -2.44670823e-01 6.12876534e-01 -1.19559005e-01
6.06963158e-01 1.81625813e-01 -1.15479723e-01 -5.81095755e-01
-2.96270907e-01 6.68941677e-01 -2.57378846e-01 1.09880283e-01
1.60042262e+00 -4.14039269e-02 -1.41566008e-01 -4.02497239e-02
1.08234406e+00 -3.25001627e-01 -1.52598596e+00 -2.65229523e-01
-6.62694633e-01 -9.63096917e-01 2.06913561e-01 -6.50774658e-01
-1.43491781e+00 6.73948705e-01 4.41654742e-01 -3.25167447e-01
1.62138116e+00 -2.33472779e-01 1.29377818e+00 3.72527480e-01
1.52604371e-01 -1.14156365e+00 4.40210193e-01 2.10727125e-01
1.19371271e+00 -1.20856690e+00 3.64576042e-01 -6.09552622e-01
-7.80337334e-01 1.10812771e+00 7.74506211e-01 -4.09883022e-01
3.93984973e-01 2.57001460e-01 -4.27120999e-02 -1.05075017e-02
-7.12225497e-01 -8.83212313e-02 4.33968723e-01 2.72406012e-01
1.79569311e-02 -4.27582338e-02 -4.64800820e-02 1.31785706e-01
-6.02200851e-02 8.03725123e-02 2.61410177e-01 8.81720364e-01
-1.44381849e-02 -1.11097264e+00 -4.31206286e-01 3.63919199e-01
-2.68309593e-01 -1.82393149e-01 -3.99390042e-01 8.04729342e-01
1.92504376e-01 7.23111987e-01 4.31147031e-02 -5.47443509e-01
2.92688748e-03 -4.56969976e-01 5.37257195e-01 -3.30600142e-01
-5.90009391e-01 1.21764898e-01 -3.49770524e-02 -7.81514287e-01
-4.80486959e-01 -4.66461003e-01 -6.55733943e-01 -8.94376755e-01
-4.58188578e-02 -4.07177866e-01 4.18190658e-01 6.07875049e-01
5.95444500e-01 2.41203859e-01 6.13664150e-01 -1.09918535e+00
-1.68362662e-01 -8.09977233e-01 -3.03531557e-01 3.76894861e-01
5.16616702e-02 -4.63118255e-01 -1.83362260e-01 2.38620952e-01] | [10.632401466369629, -0.8946421146392822] |
84f2e4b9-146e-41c4-85aa-226e694297b6 | netwalk-a-flexible-deep-embedding-approach | null | null | https://www.kdd.org/kdd2018/accepted-papers/view/netwalk-a-flexible-deep-embedding-approach-for-anomaly-detection-in-dynamic | https://dl.acm.org/doi/pdf/10.1145/3219819.3220024 | NetWalk: A Flexible Deep Embedding Approach for Anomaly Detection in Dynamic Networks | Massive and dynamic networks arise in many practical applications such as social media, security and public health. Given an evolutionary network, it is crucial to detect structural anomalies, such as vertices and edges whose “behaviors’’ deviate from underlying majority of the network, in a real-time fashion. Recently, network embedding has proven a powerful tool in learning the low-dimensional representations of vertices in networks that can capture and preserve the network structure. However, most existing network embedding approaches are designed for static networks, and thus may not be perfectly suited for a dynamic environment in which the network representation has to be constantly updated. In this paper, we propose a novel approach, NetWalk, for anomaly detection in dynamic networks by learning network representations which can be updated dynamically as the network evolves. We first encode the vertices of the dynamic network to vector representations by clique embedding, which jointly minimizes the pairwise distance of vertex representations of each walk derived from the dynamic networks, and the deep autoencoder reconstruction error serving as a global regularization. The vector representations can be computed with constant space requirements using reservoir sampling. On the basis of the learned low-dimensional vertex representations, a clustering-based technique is employed to incrementally and dynamically detect network anomalies. Compared with existing approaches, NetWalk has several advantages: 1) the network embedding can be updated dynamically, 2) streaming network nodes and edges can be encoded efficiently with constant memory space usage, 3). flexible to be applied on different types of networks, and 4) network anomalies can be detected in real-time. Extensive experiments on four real datasets demonstrate the effectiveness of NetWalk. | ['Wenchao Yu; Wei Cheng; Charu Aggarwal; Kai Zhang; Haifeng Chen; Wei Wang'] | 2018-07-19 | null | null | null | acm-sigkdd-international-conference-on | ['learning-network-representations'] | ['methodology'] | [-3.67342904e-02 3.36773545e-02 4.17145677e-02 4.09239829e-02
3.49732012e-01 -5.26863694e-01 3.42650115e-01 5.54977536e-01
1.35086998e-02 3.37837130e-01 -2.30451658e-01 -1.62029326e-01
-5.30598879e-01 -1.25172675e+00 -6.14406288e-01 -7.57065058e-01
-7.90695369e-01 6.49878681e-01 4.08650577e-01 -2.22343326e-01
1.98754575e-02 6.99197710e-01 -1.23660648e+00 -1.95507601e-01
5.67191482e-01 9.37685728e-01 -1.15889154e-01 6.97838068e-01
-2.53981799e-01 5.82119226e-01 -3.97779465e-01 -9.62679610e-02
2.42015719e-01 -3.74822646e-01 -5.47319293e-01 2.24299893e-01
-7.92323723e-02 -7.52495527e-02 -1.02033317e+00 1.21848345e+00
2.27001056e-01 1.54662073e-01 4.21606570e-01 -1.45507669e+00
-6.34567916e-01 6.69559121e-01 -3.74831021e-01 4.05471891e-01
6.45532683e-02 1.67640626e-01 1.08212543e+00 -4.36946183e-01
6.33051574e-01 1.18554378e+00 8.27979982e-01 4.74643737e-01
-1.54139555e+00 -5.28727353e-01 5.03015816e-01 2.95043856e-01
-1.19498575e+00 -1.48560479e-01 1.15755999e+00 -4.88388300e-01
7.42629886e-01 3.29829454e-01 1.08432209e+00 1.06884396e+00
1.39857918e-01 3.97368491e-01 2.78711140e-01 2.91852597e-02
3.12304527e-01 -5.06024733e-02 -1.08592592e-01 8.70344460e-01
4.59359586e-01 -3.00869122e-02 -1.41273305e-01 -5.55723488e-01
6.53499007e-01 5.87821662e-01 -3.09655160e-01 -5.88226259e-01
-1.16346800e+00 1.02839875e+00 7.19580889e-01 4.44859684e-01
-4.87323493e-01 4.65363353e-01 7.46004403e-01 7.29711831e-01
4.96850967e-01 3.33129406e-01 -2.96115249e-01 -6.70980215e-02
-4.64148253e-01 -2.40647905e-02 9.36452687e-01 5.47023535e-01
8.15817595e-01 2.91591704e-01 3.64691883e-01 6.12362862e-01
3.37995976e-01 2.99590051e-01 6.34605527e-01 -6.11546338e-01
1.32780895e-01 1.01220024e+00 -3.59532416e-01 -1.91612303e+00
-3.33289206e-01 -1.39107734e-01 -1.36533248e+00 -1.73820108e-01
7.26492703e-02 -1.33212507e-01 -5.23727119e-01 1.57063186e+00
5.54490149e-01 8.41143250e-01 -9.12906677e-02 3.47419798e-01
4.82038945e-01 7.96167314e-01 -5.15909970e-01 -1.45781040e-01
7.32510269e-01 -5.25313795e-01 -5.10664463e-01 -1.67832822e-02
5.22286773e-01 -9.06763971e-02 5.19808412e-01 -1.15056023e-01
-7.09388137e-01 -1.37264699e-01 -1.10201609e+00 5.73522270e-01
-4.95860755e-01 -7.11743712e-01 6.54506445e-01 3.98844987e-01
-1.16908145e+00 9.15215254e-01 -1.16551721e+00 -3.87796640e-01
2.89951921e-01 5.27185917e-01 -2.91854948e-01 2.67113447e-02
-1.15654171e+00 2.32385948e-01 5.81161499e-01 3.49109143e-01
-8.28139067e-01 -4.26873803e-01 -1.01878798e+00 2.79094398e-01
4.04253244e-01 -2.16070116e-01 5.52069306e-01 -1.06722474e+00
-1.37620628e+00 2.84021944e-01 6.60396134e-03 -4.18746859e-01
3.07735324e-01 4.38659281e-01 -8.14622581e-01 2.27720276e-01
-2.66984761e-01 -2.02725574e-01 9.41818535e-01 -9.25241649e-01
-1.91880301e-01 -3.24514240e-01 -7.82119110e-03 -1.71685323e-01
-7.15721667e-01 -4.34696734e-01 -2.73072153e-01 -4.97675151e-01
4.78945017e-01 -1.12682903e+00 -4.31957871e-01 2.50412136e-01
-4.18038785e-01 -1.42782062e-01 1.33840585e+00 -2.98923790e-01
1.24003887e+00 -2.06833243e+00 4.51625437e-01 8.88166249e-01
7.03300834e-01 3.68946761e-01 -2.52538174e-01 7.27697909e-01
-7.34275728e-02 2.35917777e-01 -3.17981511e-01 -8.53002369e-02
-3.38861346e-01 4.37918484e-01 -1.16389453e-01 6.54175639e-01
1.77666038e-01 7.98793018e-01 -1.24227536e+00 -1.56732053e-01
2.14709789e-01 5.78731060e-01 -6.12130165e-01 9.71881300e-02
-1.84316978e-01 3.45395833e-01 -4.92002189e-01 4.72707361e-01
3.27542722e-01 -6.84391916e-01 5.62297046e-01 1.45993575e-01
3.75462055e-01 -3.20343345e-01 -1.27715111e+00 1.17946279e+00
-2.51853466e-01 8.15375030e-01 7.75206089e-02 -1.46716368e+00
1.02918994e+00 3.98692936e-01 7.66269207e-01 -3.52594018e-01
-1.66159440e-02 6.37601092e-02 1.67820543e-01 -3.87905627e-01
1.74873725e-01 3.63685787e-01 3.70677337e-02 6.54459536e-01
-3.47333290e-02 4.16968137e-01 1.22075789e-01 4.38737214e-01
1.47177434e+00 -8.34101379e-01 1.36006951e-01 9.52661932e-02
6.14933074e-01 -3.86049390e-01 6.06831491e-01 4.55995142e-01
-1.76018491e-01 1.79332830e-02 9.77019727e-01 -7.37962782e-01
-1.06716156e+00 -9.75598276e-01 1.54294685e-01 5.57569146e-01
1.75007448e-01 -4.87741023e-01 -3.69954586e-01 -8.19368720e-01
3.45672846e-01 -8.05136934e-03 -7.52461553e-01 -5.25472939e-01
-5.59699476e-01 -6.55875623e-01 3.35740387e-01 1.75222695e-01
2.14224815e-01 -1.08452475e+00 -1.35610074e-01 5.81751883e-01
1.60663888e-01 -7.98863769e-01 -3.18931252e-01 -2.15727314e-01
-1.10691464e+00 -1.38665020e+00 -8.59806538e-02 -7.97086477e-01
9.17427361e-01 2.19758525e-01 8.12315583e-01 6.28774047e-01
-4.26406771e-01 5.46578646e-01 -2.19899893e-01 1.65329725e-01
-5.76208472e-01 -2.54001450e-02 4.23193246e-01 5.66167593e-01
1.83049634e-01 -1.14897394e+00 -6.00026309e-01 1.96102545e-01
-1.05307508e+00 -5.84669650e-01 3.54646832e-01 1.02604413e+00
5.93620360e-01 5.46801448e-01 7.26502061e-01 -1.20657325e+00
6.35007322e-01 -9.30367708e-01 -7.24103689e-01 1.26958251e-01
-8.68584931e-01 -5.51502369e-02 9.57757592e-01 -7.06518650e-01
-1.63704976e-01 -2.80965000e-01 1.52963564e-01 -6.92077458e-01
2.73492932e-01 6.05391860e-01 3.90846469e-03 -3.22210014e-01
4.88441706e-01 4.44978684e-01 3.64349842e-01 -2.95459211e-01
2.42260024e-01 4.54599530e-01 7.68245533e-02 -2.41937011e-01
9.91195798e-01 5.57995856e-01 2.12715998e-01 -1.01001084e+00
-1.22641966e-01 -3.88733834e-01 -5.68355560e-01 -2.62230724e-01
1.72564834e-01 -5.28738558e-01 -9.29005742e-01 4.20779645e-01
-7.32074082e-01 -1.49331033e-01 -3.09346199e-01 2.19243571e-01
-1.41382441e-01 7.46232271e-01 -7.13657320e-01 -5.86230516e-01
-3.81063193e-01 -6.92584097e-01 2.78587013e-01 8.25006291e-02
-8.25170651e-02 -1.56701267e+00 2.35249475e-01 -4.40560609e-01
5.37640274e-01 8.32921624e-01 1.18994212e+00 -8.22143555e-01
-6.03129923e-01 -7.97158480e-01 -2.70168036e-02 2.60480911e-01
3.76740098e-01 2.95131326e-01 -3.98550034e-01 -8.53852808e-01
-2.83308387e-01 1.25284255e-01 4.87923831e-01 1.03161089e-01
1.43440092e+00 -7.49095440e-01 -5.62392533e-01 7.33880281e-01
1.43694973e+00 2.24070437e-02 2.92459786e-01 4.86652926e-03
9.13201809e-01 4.23416704e-01 -1.43355459e-01 5.27446508e-01
1.67177424e-01 2.26317957e-01 7.18054533e-01 2.72913784e-01
3.85419905e-01 -3.98710012e-01 3.41876030e-01 1.51452112e+00
1.13083802e-01 -1.46889463e-01 -8.97502601e-01 7.65910029e-01
-2.02422905e+00 -1.12798357e+00 9.13195759e-02 2.13811111e+00
2.83844292e-01 3.23897004e-01 1.70820251e-01 2.88802922e-01
9.32465792e-01 4.47094887e-01 -1.01668239e+00 -3.50060433e-01
2.00520426e-01 -1.12720586e-01 2.90637702e-01 2.76887953e-01
-7.27720380e-01 5.53943574e-01 5.62132215e+00 2.38822803e-01
-1.28804171e+00 1.44776022e-02 3.04018915e-01 -1.01272561e-01
-4.23480719e-01 7.04021081e-02 -1.60646811e-01 8.16534519e-01
1.22163105e+00 -2.87701786e-01 7.59578884e-01 8.75151813e-01
-1.40182404e-02 5.99628508e-01 -1.11111033e+00 9.93617892e-01
-1.81479126e-01 -1.43418515e+00 6.05792068e-02 1.61644846e-01
6.14557266e-01 2.63880640e-01 5.35575785e-02 1.75113156e-01
2.87263960e-01 -9.80113983e-01 -3.57438461e-03 4.37035114e-01
6.82824135e-01 -9.53916013e-01 5.49643517e-01 2.40995213e-01
-1.45684183e+00 -3.44489038e-01 -5.13550341e-01 1.08444154e-01
-2.12558284e-02 7.60254264e-01 -7.83677578e-01 3.46011966e-01
7.14027166e-01 1.10880899e+00 -3.46165806e-01 8.51813495e-01
3.72349098e-02 6.34780526e-01 -3.50636065e-01 -2.24347264e-01
4.24188256e-01 -4.51579005e-01 9.41249847e-01 7.34221876e-01
2.82994717e-01 -2.11618602e-01 2.80303240e-01 6.94831610e-01
-1.76900759e-01 -1.57310516e-02 -1.03514242e+00 -5.36709249e-01
7.96389520e-01 1.14645839e+00 -8.09569657e-01 -1.09290645e-01
-2.87875533e-01 1.06333482e+00 5.54454982e-01 4.24855977e-01
-5.55413842e-01 -6.92719042e-01 9.38593864e-01 1.95228219e-01
4.28037673e-01 -2.58845270e-01 4.25729871e-01 -1.24479330e+00
1.38207003e-01 -6.93307221e-01 4.24318373e-01 -2.05962416e-02
-1.40637815e+00 7.45575726e-01 -4.74731714e-01 -1.22284770e+00
-2.87401497e-01 -3.73909384e-01 -8.64372849e-01 3.44893664e-01
-1.24733067e+00 -6.45555675e-01 -3.54530275e-01 8.66173506e-01
6.65113479e-02 -2.98953146e-01 1.11250627e+00 2.08224341e-01
-8.58608305e-01 5.40571690e-01 7.72981226e-01 5.17334461e-01
-4.33226563e-02 -1.20736563e+00 5.86525202e-01 7.38909662e-01
3.78654093e-01 4.57185537e-01 4.09264416e-01 -7.40578532e-01
-1.63483942e+00 -1.37123787e+00 4.47880566e-01 7.09177181e-02
1.04466975e+00 -4.85377878e-01 -1.19979095e+00 9.17006612e-01
-3.50175917e-01 5.92207551e-01 6.49258316e-01 2.41706118e-01
-3.88467282e-01 -3.97225082e-01 -1.25371206e+00 6.40273333e-01
1.05197906e+00 -7.27147222e-01 -6.57059392e-03 5.96853197e-01
8.25991035e-01 -1.90500334e-01 -9.94488597e-01 7.07407668e-03
2.52413630e-01 -6.06564224e-01 9.46899951e-01 -8.80024910e-01
4.83382940e-02 -2.56160796e-01 1.38254628e-01 -1.49064219e+00
-4.59187031e-01 -7.85376072e-01 -1.03357184e+00 9.55793202e-01
2.48910680e-01 -1.13999534e+00 8.77186596e-01 3.06789041e-01
4.59431499e-01 -9.42768872e-01 -1.00944138e+00 -7.14607358e-01
-4.22519237e-01 -3.64297442e-02 8.94464433e-01 1.18788660e+00
4.48659016e-03 1.00199223e-01 -4.40920830e-01 3.99769455e-01
6.34744167e-01 9.53821316e-02 5.23713827e-01 -1.73652625e+00
-4.35894281e-01 -4.00367409e-01 -1.19644523e+00 -5.70003510e-01
2.34444514e-01 -1.13815999e+00 -3.83783877e-01 -1.17217016e+00
-1.54134229e-01 -6.32327974e-01 -5.46094656e-01 2.64944583e-01
1.25047609e-01 -1.27666667e-01 -2.19380945e-01 3.08064222e-01
-3.94586205e-01 6.89622581e-01 7.95242846e-01 -3.26713175e-01
-3.32480013e-01 1.09373666e-01 -4.21901762e-01 6.86163783e-01
7.39626050e-01 -5.74354887e-01 -6.14141047e-01 -2.76538044e-01
3.15073490e-01 2.13048197e-02 1.83330789e-01 -9.62546527e-01
3.20458055e-01 8.28683972e-02 3.13446283e-01 -2.84311533e-01
2.63368338e-01 -1.16493726e+00 3.43290150e-01 8.06444407e-01
-7.04201087e-02 4.31536198e-01 -1.19044147e-01 1.50410604e+00
-1.65975168e-01 -4.40155230e-02 7.13343501e-01 -8.44817832e-02
-6.12657964e-01 1.00706542e+00 -2.98441410e-01 5.96063882e-02
1.34522808e+00 -1.10484049e-01 -1.16981573e-01 -4.69321132e-01
-7.88272500e-01 4.04888958e-01 5.34588277e-01 4.62992489e-01
9.64266539e-01 -1.47231960e+00 -4.94627774e-01 3.86465639e-01
7.33748600e-02 7.01876432e-02 4.25604075e-01 6.41389191e-01
-6.81058645e-01 -1.34192361e-02 -9.27036181e-02 -6.91791296e-01
-1.00556803e+00 7.72233188e-01 4.40263569e-01 -2.56680995e-01
-1.13690293e+00 5.88662267e-01 -5.50415754e-01 -4.93144214e-01
1.10413976e-01 -3.78651097e-02 -1.41725525e-01 8.31068158e-02
5.72608411e-01 3.44219804e-01 -1.23604111e-01 -5.89305103e-01
-3.36142004e-01 3.84286761e-01 -1.67720795e-01 3.24794233e-01
1.62419188e+00 -7.67033473e-02 -4.66154963e-01 7.34153807e-01
1.44113350e+00 -2.77360767e-01 -1.00272596e+00 -7.17735231e-01
1.71406478e-01 -6.26151085e-01 1.34354795e-03 6.45240694e-02
-1.59280860e+00 6.17511332e-01 5.04050493e-01 7.64090776e-01
9.14720118e-01 -1.80254474e-01 9.62755620e-01 6.68778956e-01
1.39555782e-01 -8.98785055e-01 2.57666528e-01 3.76580268e-01
5.84184229e-01 -1.00315297e+00 -2.38162428e-01 -5.25426231e-02
-2.40504101e-01 1.25983489e+00 4.79279131e-01 -4.47263479e-01
1.13269281e+00 -1.67252064e-01 -3.87035996e-01 -5.34576118e-01
-8.12749028e-01 2.76257247e-01 -5.33807203e-02 5.01508951e-01
-2.07570687e-01 1.20280869e-01 3.69466901e-01 4.06216569e-02
2.42622957e-01 -5.99710405e-01 6.15699530e-01 7.69163251e-01
-3.51925701e-01 -8.51953447e-01 1.39450476e-01 9.53881502e-01
-2.03751549e-01 2.22732589e-01 -3.21029305e-01 5.08195400e-01
-2.93197900e-01 6.87797666e-01 3.18333358e-01 -5.03464341e-01
2.30430439e-01 -1.10957408e-02 -1.28238767e-01 -5.57174861e-01
-3.49487454e-01 -5.24698615e-01 -3.30345064e-01 -7.36116111e-01
-3.82522233e-02 -5.41894972e-01 -1.01233470e+00 -7.96542943e-01
-3.24449986e-01 3.11531335e-01 5.25649905e-01 6.08607113e-01
7.58824646e-01 5.45679927e-01 1.12582338e+00 -6.53997481e-01
-4.37749952e-01 -6.78943098e-01 -8.31166446e-01 4.86876726e-01
6.15237176e-01 -5.21882415e-01 -7.37340093e-01 -5.31318605e-01] | [7.1270904541015625, 6.124075889587402] |
cc5bab27-f0ee-4b71-bbac-e59642b0aa12 | frankmocap-fast-monocular-3d-hand-and-body | 2008.08324 | null | https://arxiv.org/abs/2008.08324v1 | https://arxiv.org/pdf/2008.08324v1.pdf | FrankMocap: Fast Monocular 3D Hand and Body Motion Capture by Regression and Integration | Although the essential nuance of human motion is often conveyed as a combination of body movements and hand gestures, the existing monocular motion capture approaches mostly focus on either body motion capture only ignoring hand parts or hand motion capture only without considering body motion. In this paper, we present FrankMocap, a motion capture system that can estimate both 3D hand and body motion from in-the-wild monocular inputs with faster speed (9.5 fps) and better accuracy than previous work. Our method works in near real-time (9.5 fps) and produces 3D body and hand motion capture outputs as a unified parametric model structure. Our method aims to capture 3D body and hand motion simultaneously from challenging in-the-wild monocular videos. To construct FrankMocap, we build the state-of-the-art monocular 3D "hand" motion capture method by taking the hand part of the whole body parametric model (SMPL-X). Our 3D hand motion capture output can be efficiently integrated to monocular body motion capture output, producing whole body motion results in a unified parrametric model structure. We demonstrate the state-of-the-art performance of our hand motion capture system in public benchmarks, and show the high quality of our whole body motion capture result in various challenging real-world scenes, including a live demo scenario. | ['Hanbyul Joo', 'Yu Rong', 'Takaaki Shiratori'] | 2020-08-19 | null | null | null | null | ['3d-human-reconstruction'] | ['computer-vision'] | [-5.66740096e-01 -7.41388738e-01 -5.63070178e-01 2.78065473e-01
-4.00975466e-01 -6.93767846e-01 4.65242773e-01 -1.26177394e+00
-2.85147697e-01 4.94904816e-01 5.62212288e-01 2.13152096e-01
4.57417786e-01 -3.88767302e-01 -4.87171888e-01 -5.83451211e-01
2.77780414e-01 5.85024238e-01 3.24031293e-01 -8.72600377e-02
-4.19724315e-01 7.11341202e-01 -1.31790209e+00 -2.61641163e-02
4.06923108e-02 5.43609202e-01 3.23664993e-02 1.56460512e+00
5.14994502e-01 8.05867910e-01 -4.12319332e-01 -1.71203479e-01
5.48608065e-01 -4.21674043e-01 -9.10964191e-01 3.96104530e-02
7.36197650e-01 -1.16386354e+00 -1.05638576e+00 3.81639868e-01
1.22135770e+00 -7.12333918e-02 4.87500876e-01 -1.26756656e+00
-1.20462485e-01 -2.11457945e-02 -8.14635932e-01 1.69212557e-03
9.51275408e-01 8.02385390e-01 4.68919754e-01 -9.94779766e-01
1.33035362e+00 1.48097479e+00 5.64135551e-01 1.03611076e+00
-9.26245332e-01 -6.43600047e-01 -1.60265997e-01 -9.22351405e-02
-1.57242632e+00 -4.40155983e-01 4.67241764e-01 -8.15443516e-01
1.01100767e+00 2.41417557e-01 1.29100609e+00 1.30585742e+00
4.10175025e-01 1.38813770e+00 1.71814859e-01 -1.04442216e-01
-1.97488129e-01 -7.89548635e-01 -1.58261821e-01 7.09720612e-01
1.66036382e-01 3.39513928e-01 -7.11836874e-01 -6.41907081e-02
1.50263870e+00 1.83132127e-01 -5.73579371e-01 -5.93647480e-01
-1.67085028e+00 2.32389852e-01 2.42953375e-01 -1.77982971e-02
-4.81302470e-01 9.34082448e-01 2.46769607e-01 -1.00969508e-01
7.39658102e-02 -5.48696876e-01 -3.97034883e-01 -6.67570174e-01
-1.38592029e+00 9.18672621e-01 6.07708037e-01 1.29767799e+00
1.50209755e-01 1.35710254e-01 -4.43677157e-01 3.20652783e-01
3.71430337e-01 1.22876036e+00 4.34273362e-01 -1.31981218e+00
6.23594940e-01 2.63104409e-01 5.27939439e-01 -6.27215922e-01
-5.77574074e-01 6.99349567e-02 -8.44707608e-01 2.30328366e-01
5.51312685e-01 -5.02132416e-01 -7.20552444e-01 1.50832689e+00
7.36815751e-01 -1.79469213e-01 -2.85189331e-01 1.57610357e+00
1.05168700e+00 5.10088861e-01 1.14045046e-01 2.03268960e-01
1.26848030e+00 -1.21439171e+00 -8.85137260e-01 -2.17332438e-01
5.91772676e-01 -6.41205072e-01 9.52310205e-01 2.66313761e-01
-1.24603975e+00 -7.23171115e-01 -5.59023201e-01 -3.19447309e-01
4.04094100e-01 4.02358204e-01 3.61507803e-01 6.16705716e-01
-7.26930320e-01 2.43931845e-01 -1.29963768e+00 -4.76412594e-01
6.34367093e-02 4.23784256e-01 -6.79176569e-01 -1.45812124e-01
-8.87040257e-01 5.31142294e-01 1.40429690e-01 4.69805859e-02
-8.25680315e-01 -6.07785881e-01 -1.03013396e+00 -2.74140388e-01
2.40603879e-01 -1.55489302e+00 1.55944884e+00 -2.70790666e-01
-1.78892314e+00 1.00264478e+00 -3.07957351e-01 7.32228085e-02
1.13451445e+00 -9.26946580e-01 7.82019049e-02 4.26049769e-01
-1.33015737e-01 1.17069268e+00 6.58829868e-01 -9.74906802e-01
-5.17676771e-01 -4.09659863e-01 -2.48216689e-01 4.04651612e-01
3.85293633e-01 1.18521966e-01 -1.11577642e+00 -8.74150038e-01
-1.52383465e-02 -1.37021458e+00 1.91718429e-01 5.96989393e-01
-3.30801070e-01 2.09816650e-01 1.01483905e+00 -8.30828667e-01
1.31330085e+00 -1.67879200e+00 6.05898082e-01 -5.42261004e-01
2.02424482e-01 3.54963899e-01 1.77439630e-01 2.80964822e-01
3.74095649e-01 -2.60937870e-01 -2.02973802e-02 -5.99560380e-01
1.88861787e-02 1.51017159e-01 -1.52592376e-01 6.12600803e-01
-4.28849131e-01 1.50672722e+00 -1.07998335e+00 -7.88654566e-01
5.78175366e-01 6.76850736e-01 -6.43623590e-01 3.13668340e-01
1.36440191e-02 7.64738619e-01 -1.67988181e-01 1.21024692e+00
5.16839802e-01 -6.98163062e-02 -3.00936829e-02 -5.78707159e-02
1.37523413e-01 -3.87238801e-01 -1.28248549e+00 2.16141462e+00
1.36403507e-02 7.04846382e-01 2.26612329e-01 3.10214341e-01
3.92456591e-01 5.90192080e-01 8.08675230e-01 -7.57635981e-02
2.23280728e-01 -2.86180489e-02 -2.45219424e-01 -5.46674669e-01
6.06660366e-01 -1.17904529e-01 -6.60282522e-02 4.69837397e-01
8.47921818e-02 -4.42186743e-01 -1.98198542e-01 -1.37145177e-01
8.54420781e-01 9.04548168e-01 4.98880655e-01 2.03691855e-01
2.73769140e-01 2.75562495e-01 2.52858967e-01 4.99362499e-01
-5.54709911e-01 1.13927901e+00 -1.45200655e-01 -4.49381292e-01
-1.00301707e+00 -1.21877861e+00 4.87761915e-01 8.68287325e-01
2.67518997e-01 -4.35837388e-01 -7.13648677e-01 -3.07854831e-01
2.22596839e-01 -2.88842678e-01 -4.16052192e-01 2.00496674e-01
-1.05333734e+00 -2.69250065e-01 1.00569618e+00 1.16013658e+00
7.99034297e-01 -1.19058168e+00 -1.31879544e+00 1.65298522e-01
-5.35515368e-01 -1.25287187e+00 -1.13395679e+00 -6.88277245e-01
-8.55929375e-01 -1.10816956e+00 -1.69589126e+00 -4.80426013e-01
-8.30102935e-02 4.82965410e-01 1.00455189e+00 -1.58140883e-01
-4.55315351e-01 6.57150745e-01 -1.74065486e-01 4.97635338e-04
6.28223643e-02 1.03083856e-01 3.27480882e-01 -2.74213523e-01
1.56151906e-01 -2.66089469e-01 -1.05395925e+00 6.21031165e-01
-5.57640135e-01 2.37268776e-01 2.00239211e-01 7.29577243e-01
4.57266122e-01 -9.11924720e-01 -2.71839857e-01 4.84195203e-02
8.91991854e-02 -2.23161638e-01 -3.62872362e-01 -1.45662844e-01
3.03104997e-01 -2.82248318e-01 9.04828906e-02 -7.62130499e-01
-1.03227901e+00 5.84304214e-01 -9.11626071e-02 -1.03731620e+00
-1.02168687e-01 5.89460097e-02 -2.96455920e-01 5.90020604e-02
5.59713423e-01 4.16267067e-01 -4.15637009e-02 -5.23528755e-01
6.29418135e-01 5.80744863e-01 1.22914934e+00 -5.16875565e-01
5.87861001e-01 1.03868186e+00 1.24509364e-01 -9.16945338e-01
-9.29686651e-02 -8.76106739e-01 -1.32561338e+00 -5.66481590e-01
1.08107972e+00 -1.19403696e+00 -1.02603257e+00 1.03073752e+00
-1.33649886e+00 -7.34178483e-01 -6.35651350e-02 8.04587901e-01
-1.01483727e+00 5.89377165e-01 -9.14908826e-01 -8.30148637e-01
-7.20103979e-01 -9.28147912e-01 1.78520513e+00 -9.97905359e-02
-8.57954025e-01 -6.44000947e-01 6.11953616e-01 1.13548242e-01
-1.29047945e-01 5.61632991e-01 1.56238014e-02 5.94264925e-01
-5.04817843e-01 -4.33781803e-01 3.22586298e-02 -2.92525202e-01
3.35766412e-02 1.17949866e-01 -7.56741047e-01 -3.95894170e-01
-5.70392370e-01 -1.57546431e-01 5.08102000e-01 8.74623954e-01
3.21739614e-01 -7.27079511e-02 -5.52742362e-01 9.98346806e-01
1.03314042e+00 -6.44442588e-02 7.08544910e-01 1.18521117e-01
1.26682043e+00 4.30218488e-01 6.51284635e-01 6.93969846e-01
4.59164649e-01 1.33863282e+00 1.48287877e-01 1.73392624e-01
-5.89004099e-01 -4.62981969e-01 5.20601571e-01 6.06386483e-01
-6.20044708e-01 -2.79981762e-01 -9.66651797e-01 6.96679711e-01
-1.96431792e+00 -1.27975357e+00 -3.65248710e-01 1.87456775e+00
7.20578790e-01 -6.06223404e-01 9.24770236e-01 5.99413104e-02
4.79052067e-01 1.32207930e-01 -4.71982896e-01 2.32938379e-01
-4.91809696e-02 3.69542353e-02 4.60737675e-01 5.70859611e-01
-1.17162848e+00 1.26358545e+00 6.70035410e+00 3.34119260e-01
-1.20467961e+00 6.95446879e-02 -3.88084710e-01 -8.92592132e-01
3.83608907e-01 -3.52927506e-01 -9.34966743e-01 3.58384818e-01
3.18868846e-01 1.64713651e-01 6.37684762e-02 8.82255018e-01
3.67622077e-01 -1.40256584e-01 -1.15826440e+00 1.69333720e+00
-7.20812082e-02 -9.44917202e-01 9.73417237e-03 2.84935087e-01
8.25537086e-01 -9.96289402e-02 -3.57693970e-01 -3.46382372e-02
2.53961831e-01 -1.06703103e+00 1.26301384e+00 6.58332527e-01
1.22772801e+00 -3.39535892e-01 5.00728428e-01 8.30311060e-01
-1.73518491e+00 2.40920424e-01 -3.42746340e-02 -3.22291940e-01
8.32721531e-01 -3.58330943e-02 -3.52975398e-01 4.90033954e-01
7.83783615e-01 7.92063475e-01 -5.73055334e-02 9.54498053e-01
-2.35848993e-01 2.73871928e-01 -4.19515997e-01 1.68192506e-01
-2.29236260e-02 3.36200356e-01 7.45762527e-01 1.46246028e+00
3.12860340e-01 5.00748873e-01 -6.70580044e-02 7.06485152e-01
1.56402349e-01 -2.67181844e-01 -6.04140341e-01 2.41577938e-01
2.93101341e-01 7.09923208e-01 -2.90645450e-01 -5.55275619e-01
-1.79935336e-01 1.58346379e+00 -2.23219976e-01 2.18864068e-01
-8.66921663e-01 -1.19437180e-01 1.10474503e+00 2.91802466e-01
2.97748774e-01 -7.01844573e-01 -1.56429544e-01 -1.64800918e+00
1.23884149e-01 -5.74503779e-01 1.95475772e-01 -1.15990436e+00
-6.59093857e-01 2.00797468e-01 2.63865501e-01 -1.64335334e+00
-8.87710989e-01 -5.91198862e-01 -2.31275335e-01 9.58359182e-01
-7.43495107e-01 -1.46663320e+00 -1.05435479e+00 1.03917682e+00
6.96322143e-01 2.43046209e-01 6.31919622e-01 3.31115872e-01
-1.16324872e-01 5.34463346e-01 -3.08264852e-01 4.57899362e-01
9.42185879e-01 -8.38366687e-01 1.00103784e+00 6.63168907e-01
-5.52108139e-02 4.23620343e-01 5.20342290e-01 -1.01361823e+00
-1.78960764e+00 -8.53108406e-01 6.34835184e-01 -1.32089603e+00
-6.26280671e-03 -1.54029682e-01 -3.66338462e-01 9.44597304e-01
-2.59982735e-01 3.68783504e-01 1.85314730e-01 -7.59777069e-01
-1.97576974e-02 4.86084700e-01 -1.03681719e+00 8.04030836e-01
1.79967976e+00 -4.17403311e-01 -7.17160225e-01 -3.60281676e-01
3.80630493e-01 -1.02018189e+00 -7.12073982e-01 2.33776525e-01
1.64225876e+00 -6.92794502e-01 1.46216536e+00 -5.91272712e-01
5.19171476e-01 -5.38186789e-01 -2.81757563e-01 -8.71145606e-01
-3.51687253e-01 -8.87746751e-01 -9.72066939e-01 6.68899477e-01
-6.94690764e-01 1.05816789e-01 1.19274485e+00 5.73200345e-01
4.27913964e-01 -4.12706167e-01 -8.77091110e-01 -9.78534997e-01
4.77528647e-02 -4.87309486e-01 5.08317709e-01 5.20462692e-01
1.78720970e-02 -3.81610319e-02 -1.05244660e+00 -1.29223034e-01
6.66381598e-01 6.66062757e-02 1.79948032e+00 -9.44957316e-01
-3.43413383e-01 -2.97747403e-01 -5.63531697e-01 -1.92690599e+00
-1.39677450e-01 -3.53898168e-01 5.56253782e-03 -1.39395237e+00
3.52474988e-01 4.72918034e-01 7.28707314e-01 2.73445517e-01
-2.18696386e-01 6.08359098e-01 7.72948503e-01 7.44554222e-01
-1.51310936e-01 4.93875235e-01 1.65382159e+00 -4.05252427e-02
-3.76586378e-01 -1.44925550e-01 1.80339172e-01 8.05654585e-01
1.65349856e-01 -3.23337406e-01 -5.05916812e-02 -5.58311105e-01
-3.23232919e-01 7.77571738e-01 9.47969079e-01 -1.17215049e+00
3.16788852e-01 -2.67078251e-01 6.53006256e-01 -1.09559965e+00
6.31543219e-01 -6.77893221e-01 8.81389201e-01 8.87159944e-01
3.93037081e-01 5.72886467e-02 1.90267056e-01 2.65349239e-01
3.59653868e-02 4.39349204e-01 5.15910089e-01 -2.73069024e-01
-8.19731772e-01 5.98589122e-01 -2.89076567e-01 6.72979653e-02
7.90503681e-01 -6.21322513e-01 -2.01196119e-01 -6.62693322e-01
-8.97326231e-01 1.28334090e-01 7.61767030e-01 6.15627110e-01
5.75495780e-01 -1.68959999e+00 -8.18908811e-01 1.07598186e-01
1.12317521e-02 4.42587100e-02 3.73701155e-01 8.10856402e-01
-1.16371107e+00 8.48226488e-01 -4.68493879e-01 -1.05282187e+00
-1.68988848e+00 3.63636553e-01 5.08265734e-01 -1.18082292e-01
-9.87160802e-01 4.66636211e-01 4.09451514e-01 -4.55492586e-01
3.86341326e-02 -4.47148085e-01 2.95052111e-01 -3.50088716e-01
5.75957060e-01 8.62578750e-01 -4.33111191e-01 -1.12393665e+00
-5.70746422e-01 1.45558143e+00 8.99291873e-01 -7.09739804e-01
6.70688689e-01 -2.82314181e-01 5.21997571e-01 2.19425485e-01
9.93613124e-01 6.33011684e-02 -1.69442248e+00 1.98069401e-02
-7.99065769e-01 -8.96476865e-01 -3.84248883e-01 -7.02975810e-01
-9.77965355e-01 1.16574073e+00 4.81566072e-01 -8.23326170e-01
8.92959535e-01 7.59434626e-02 1.15992022e+00 1.77354530e-01
9.35192764e-01 -6.98053181e-01 -2.99616111e-03 5.81621885e-01
1.08433211e+00 -9.47208166e-01 3.84088121e-02 -5.27690232e-01
-6.73230648e-01 1.07739937e+00 6.21755004e-01 -8.06007758e-02
5.59183896e-01 3.66293013e-01 1.00382388e-01 -7.32723475e-02
-3.87325883e-01 -1.89458102e-01 4.99221653e-01 8.21035087e-01
3.02335948e-01 3.09278131e-01 -1.41547918e-01 8.27367306e-01
-3.43270183e-01 6.85660362e-01 1.67404771e-01 1.27104437e+00
8.32149610e-02 -7.12518752e-01 -8.26507866e-01 -3.24561536e-01
-3.92826527e-01 3.07624549e-01 -5.98678648e-01 1.23842514e+00
-3.92326489e-02 5.56408226e-01 -9.33875516e-02 -5.31111062e-01
5.28324485e-01 3.73222269e-02 9.86483455e-01 -3.40952903e-01
-5.17452240e-01 3.73399436e-01 -2.27921501e-01 -9.97579396e-01
-3.06030303e-01 -5.53938866e-01 -1.17141461e+00 -8.92107069e-01
8.68968368e-02 -6.81925118e-01 1.87278196e-01 7.24960566e-01
2.89299160e-01 9.88038629e-02 -2.04923049e-01 -1.91807234e+00
-3.80225331e-01 -1.00675774e+00 -6.60775781e-01 4.17545587e-01
5.72661281e-01 -8.80138934e-01 9.66416821e-02 5.73487282e-01] | [7.158892631530762, -0.6883317232131958] |
e3936dbf-261d-4268-827e-fdaeab4d1a87 | employing-linear-prediction-coding-in-feature | null | null | https://aclanthology.org/O13-5008 | https://aclanthology.org/O13-5008.pdf | 雜訊環境下應用線性估測編碼於特徵時序列之強健性語音辨識 (Employing Linear Prediction Coding in Feature Time Sequences for Robust Speech Recognition in Noisy Environments) [In Chinese] | null | ['Jeih-weih Hung', 'Wen-yu Tseng', 'Hao-teng Fan'] | 2013-12-01 | employing-linear-prediction-coding-in-feature-3 | https://aclanthology.org/O13-1015 | https://aclanthology.org/O13-1015.pdf | roclingijclclp-2013-10 | ['robust-speech-recognition'] | ['speech'] | [-8.63703638e-02 1.71006292e-01 -6.22772932e-01 -4.08054382e-01
-8.41685571e-03 -9.08429027e-01 6.55310392e-01 -6.53472245e-01
-2.85945535e-01 1.06888819e+00 -4.63127941e-02 -1.01159286e+00
-3.91567826e-01 -9.63214397e-01 -4.95059669e-01 -6.31337762e-01
-9.79754329e-01 7.25764990e-01 3.30370307e-01 -6.93831444e-01
7.03166842e-01 7.88774848e-01 -1.68942046e+00 7.18545914e-01
7.04417467e-01 8.52217197e-01 2.49141872e-01 1.14950800e+00
-1.95044339e-01 1.55633950e+00 -7.48382092e-01 -5.46825826e-01
3.13719302e-01 -1.23176083e-01 -7.22945035e-01 -1.01074085e-01
9.28529128e-02 -8.59008506e-02 -2.09758401e-01 9.22211111e-01
5.37373662e-01 4.49454933e-02 1.08379531e+00 -1.42548037e+00
-5.91619551e-01 6.10313773e-01 -4.01565880e-02 1.21627934e-01
1.03678203e+00 -5.39447069e-01 1.19919395e+00 -1.13026452e+00
7.20913768e-01 1.26888943e+00 8.66221786e-01 5.44149756e-01
-1.22286928e+00 -1.94712028e-01 -3.26822817e-01 -9.51717794e-02
-1.46558487e+00 -3.25250506e-01 4.25783843e-02 -2.08119690e-01
1.66093647e+00 1.26596653e+00 1.20609856e+00 1.01401424e+00
1.26658809e+00 8.34431887e-01 1.04267764e+00 -5.13792276e-01
3.35295945e-01 3.66983831e-01 1.54683650e-01 6.33519173e-01
8.40953708e-01 5.26628852e-01 -7.06372619e-01 -9.13127720e-01
9.33553874e-01 -2.94925272e-01 1.71355158e-01 -5.05680561e-01
-9.05919552e-01 6.91228509e-01 1.78732842e-01 3.83959889e-01
-1.39880210e-01 9.89067405e-02 1.26390755e-01 5.30987144e-01
-2.58292928e-02 6.47037446e-01 -9.11868811e-01 -1.33165747e-01
-8.71728659e-01 5.10332465e-01 1.25398111e+00 1.52653182e+00
1.24482810e-01 2.94908643e-01 -9.34252143e-02 3.17179203e-01
8.92314315e-01 1.01808000e+00 4.28362608e-01 -1.36146402e+00
-6.87414408e-02 1.72361732e-01 5.01781464e-01 -8.52631688e-01
-6.33224547e-01 -9.64177120e-03 -8.93263519e-01 4.49267089e-01
3.49161088e-01 4.57367361e-01 -8.02827001e-01 5.07305264e-01
4.33481112e-02 -2.34125629e-01 4.53833073e-01 5.55570945e-02
4.99930978e-01 3.76208365e-01 -1.34477139e-01 -5.73289394e-01
1.06082785e+00 -1.36716676e+00 -1.35299087e+00 2.33215362e-01
9.05734658e-01 -1.07320261e+00 4.35900748e-01 5.33875942e-01
-1.55548143e+00 -1.37560293e-01 -1.08699942e+00 1.78573877e-01
-7.27255583e-01 -3.14239264e-01 8.57801437e-01 1.43120694e+00
-1.60129595e+00 9.73287821e-01 -4.91727620e-01 6.59165755e-02
1.20568443e-02 8.24621081e-01 -2.64718989e-03 4.62812334e-01
-1.33193445e+00 1.08501506e+00 2.22979754e-01 -1.21242590e-01
-1.65216476e-01 -2.13068098e-01 -8.23704481e-01 -5.63443303e-01
-4.78693932e-01 -5.29636025e-01 1.44139910e+00 -2.59346128e-01
-1.65295815e+00 9.71794367e-01 -1.42069459e-01 -1.97814897e-01
6.14786744e-01 -1.28011424e-02 -8.31891418e-01 2.42498964e-01
-1.89849049e-01 5.76383233e-01 9.28263724e-01 -1.35132408e+00
-7.59897232e-01 -1.67359829e-01 -1.23336017e-01 2.66287565e-01
-1.25510961e-01 1.89734384e-01 2.11616129e-01 -1.12999000e-01
3.27147305e-01 -7.20919967e-01 -2.53068686e-01 -5.32041907e-01
-1.46512717e-01 -7.10518599e-01 7.70373225e-01 -4.51523662e-01
1.83705616e+00 -1.67618537e+00 -2.06720144e-01 3.98590982e-01
3.57815564e-01 -1.24705513e-03 2.40583986e-01 1.08380008e+00
-2.76906848e-01 7.32199550e-01 4.11965609e-01 -1.10722095e-01
2.24991128e-01 4.85861301e-01 -4.16602850e-01 3.05609167e-01
-1.29282743e-01 1.15307164e+00 -1.16605783e+00 -5.23096442e-01
4.11106765e-01 9.42391157e-02 -4.58732933e-01 4.79237735e-01
2.99364805e-01 1.47170946e-01 -3.56553018e-01 1.39399457e+00
1.15709066e+00 -1.31984919e-01 1.45911396e-01 5.30878425e-01
-3.79135728e-01 3.55090618e-01 -6.96863770e-01 1.05554795e+00
7.08333924e-02 5.00173986e-01 1.02364108e-01 -7.94621468e-01
3.33247900e-01 8.61540735e-01 4.37155962e-01 -9.67555881e-01
-2.26398129e-02 6.92409754e-01 1.12803578e-01 -6.07703328e-01
7.58228302e-01 3.94563079e-02 -3.64872098e-01 6.40070081e-01
-2.37588286e-01 -6.59476995e-01 9.64643434e-02 3.08100313e-01
6.22585893e-01 -5.18246442e-02 5.62923312e-01 -1.05613089e+00
7.86340594e-01 -1.82965681e-01 -1.81299388e-01 1.03415680e+00
-3.09923887e-01 3.19085121e-01 2.99841821e-01 -6.65102363e-01
-6.45341039e-01 -1.12307119e+00 -4.89381433e-01 1.30636716e+00
3.24267983e-01 -4.39044595e-01 -9.54439282e-01 -2.49762803e-01
1.77620783e-01 6.89606130e-01 -5.90509653e-01 3.84124845e-01
-5.03739953e-01 -8.32535863e-01 7.39044368e-01 3.45434904e-01
-5.07752821e-02 -1.33414865e+00 -6.58416986e-01 1.25490099e-01
-2.20292807e-01 -6.63697243e-01 -6.23428151e-02 4.48765576e-01
-1.35989368e+00 -5.18594682e-01 -6.66252747e-02 -8.20914626e-01
5.87345481e-01 2.46782884e-01 1.27047324e+00 5.39230824e-01
-2.31483161e-01 4.26904231e-01 -1.21292919e-01 -4.95818377e-01
-4.59671497e-01 -8.00336525e-02 5.28869390e-01 -5.87835789e-01
5.19427478e-01 -2.50617653e-01 -7.29350567e-01 5.37953973e-01
-6.88540697e-01 1.62748516e-01 1.79803044e-01 1.04410267e+00
1.35816500e-01 -9.34035778e-02 1.22507080e-01 -6.38007045e-01
8.72274399e-01 -1.69219792e-01 -3.78732830e-01 5.77745810e-02
-6.77108407e-01 -3.74140263e-01 3.21430594e-01 -3.25342178e-01
-1.01981449e+00 -4.87835288e-01 -9.82677937e-02 2.45538145e-01
1.11353043e-02 -1.46784872e-01 6.47139177e-02 -5.24923325e-01
8.02199244e-01 9.25758183e-02 1.99174434e-02 -6.80815242e-03
3.01039815e-01 7.09525108e-01 -6.82967342e-03 -6.68678164e-01
8.44880998e-01 4.91470337e-01 7.98524171e-02 -9.57177758e-01
-1.52186140e-01 -2.60129690e-01 -9.51962709e-01 -6.54426932e-01
6.56643391e-01 -6.78531289e-01 -9.10833478e-01 3.91110867e-01
-9.38691139e-01 -3.38627815e-01 -3.91645581e-01 4.25431967e-01
-1.01278400e+00 2.75717527e-02 -3.90154392e-01 -1.27895141e+00
-5.10977268e-01 -1.02017939e+00 9.43384409e-01 5.30070923e-02
-5.10597289e-01 -1.26927447e+00 5.87685481e-02 2.71537274e-01
1.81734428e-01 -1.73075795e-01 6.90226793e-01 -2.38256708e-01
-4.24233019e-01 -1.53791070e-01 2.34436691e-02 -1.39755070e-01
1.70832314e-02 4.95917559e-01 -9.81751978e-01 -5.31145096e-01
6.65065646e-02 -1.92070693e-01 -1.08835101e-01 6.52520418e-01
5.91872573e-01 -2.29931593e-01 -8.56000841e-01 5.40386558e-01
1.38545322e+00 3.85070026e-01 5.32770038e-01 7.28214979e-01
1.41836226e-01 5.53460240e-01 9.17806149e-01 4.63203549e-01
1.30579369e-02 3.28798652e-01 2.40537539e-01 1.49327129e-01
1.11720070e-01 -1.54819340e-01 3.77893507e-01 1.16112018e+00
-8.18235934e-01 -2.69281328e-01 -5.07867396e-01 4.42987174e-01
-1.72482407e+00 -1.40330648e+00 -4.32368398e-01 6.90478683e-01
6.25676990e-01 1.56016424e-01 -1.48347050e-01 3.35214496e-01
4.99015123e-01 -2.03574806e-01 -1.19133167e-01 -1.06291151e+00
-1.43546045e-01 3.15233678e-01 7.37729073e-01 1.00061214e+00
-7.20721722e-01 1.03317809e+00 1.29781246e+01 1.02230716e+00
2.21112028e-01 1.03134915e-01 5.16071796e-01 3.48020852e-01
-4.36954498e-01 -4.56139445e-02 -1.04416132e+00 2.72933897e-02
1.38140702e+00 -4.30666685e-01 6.85999811e-01 5.44219851e-01
3.44648361e-01 -4.23268199e-01 -1.26188684e+00 5.26221812e-01
9.73738134e-02 -1.40886843e+00 -2.83300440e-04 6.85225725e-01
7.73699820e-01 -5.08050561e-01 6.22419357e-01 3.24184299e-01
6.09259963e-01 -1.14389277e+00 8.60300779e-01 2.53660440e-01
1.03040910e+00 -6.05088234e-01 5.67372203e-01 1.68872893e-01
-1.14389896e+00 -2.20873043e-01 -8.77727985e-01 -1.00755692e+00
3.93533185e-02 -1.81779593e-01 -4.29956943e-01 3.48861217e-01
9.58353162e-01 2.99398601e-01 -3.93658698e-01 9.95779395e-01
-4.78694476e-02 1.04875881e-02 -2.71853864e-01 -4.48467314e-01
4.83122796e-01 -3.54241252e-01 4.66730654e-01 1.00164843e+00
2.48499006e-01 3.51035744e-01 -9.84472036e-02 4.01770771e-01
5.45058846e-01 3.29446048e-02 -1.19659424e+00 -1.78908288e-01
2.83276141e-01 9.16795909e-01 -4.83487815e-01 -4.22520459e-01
-2.00212970e-01 8.62069130e-01 -3.55488248e-02 5.01107454e-01
-6.11489356e-01 -4.35615242e-01 9.72222984e-01 -1.27327025e-01
-1.14700586e-01 -3.48497719e-01 -6.23769283e-01 -7.30352640e-01
-5.89872956e-01 -4.54965204e-01 5.93606755e-02 -5.53365827e-01
-1.39813089e+00 5.79277515e-01 -2.27688253e-02 -1.40553558e+00
-6.99901402e-01 -1.27676582e+00 -4.76714373e-01 4.92853165e-01
-1.11898029e+00 -1.10984349e+00 2.50124663e-01 4.52870727e-01
1.64141744e-01 -5.34416080e-01 1.39563632e+00 3.57715860e-02
1.00637585e-01 9.24474537e-01 6.69434488e-01 -7.37814724e-01
5.56605101e-01 -1.27867436e+00 5.68737745e-01 -1.37897313e-01
-4.31265175e-01 9.05828118e-01 6.28349900e-01 -5.39804697e-01
-1.41196322e+00 -3.66917729e-01 1.08350635e+00 -9.83769417e-01
6.55218959e-01 -3.86345625e-01 4.23767231e-02 7.88592756e-01
7.15902448e-01 -6.18741751e-01 8.21781039e-01 -1.83753878e-01
1.80774391e-01 5.75296998e-01 -1.39248300e+00 6.12354755e-01
1.66275799e+00 -4.63594139e-01 -6.25784039e-01 7.60327101e-01
8.13696027e-01 -6.94087505e-01 -1.30082703e+00 3.34633321e-01
8.65424156e-01 -8.75409484e-01 1.61978090e+00 -1.32660246e+00
-4.43697497e-02 2.81152606e-01 -2.61993498e-01 -9.32519078e-01
-5.96193194e-01 -1.23518765e+00 -5.33532679e-01 -5.83747849e-02
5.96577883e-01 -1.13057327e+00 3.42365682e-01 8.74560475e-01
-2.82833427e-01 -6.42737269e-01 -1.06996536e+00 -1.32016802e+00
-3.35779637e-02 -1.45572275e-01 4.90409225e-01 7.63798356e-01
6.83744550e-01 1.09839931e-01 -6.36873543e-02 -1.00294888e-01
5.33176839e-01 9.55312885e-03 4.41501856e-01 -1.34294486e+00
3.86843324e-01 -5.75816095e-01 -3.07655483e-01 -9.45992947e-01
-8.85957032e-02 -8.12076271e-01 -6.53862000e-01 -1.28511906e+00
-8.32044985e-03 -1.91056758e-01 -1.11109078e-01 -1.65725678e-01
3.67937148e-01 2.13746816e-01 1.20859891e-02 1.03788137e-01
-3.71160030e-01 6.18435517e-02 1.29639816e+00 6.91750320e-05
-1.62315920e-01 4.85058486e-01 -4.81304944e-01 7.84440815e-01
8.58408585e-02 -2.99253196e-01 -6.78878546e-01 6.11881316e-02
6.69384480e-01 4.61409837e-02 3.24159935e-02 -7.42885649e-01
5.37211418e-01 -3.75702560e-01 4.78586555e-01 -1.32223868e+00
1.30741090e-01 -9.61415648e-01 6.95283338e-02 9.40189242e-01
2.74610907e-01 1.20424610e-02 8.66204947e-02 5.39500564e-02
-1.44871444e-01 -5.70943117e-01 9.21121240e-01 -4.07591403e-01
-4.92852688e-01 -5.20386267e-03 -1.03226590e+00 8.97834301e-02
9.92593169e-01 -7.84614205e-01 -3.59281451e-01 -4.20183957e-01
-8.29068601e-01 -1.95836127e-02 6.50830388e-01 3.03609259e-02
7.30431557e-01 -1.51530886e+00 -2.30721906e-01 7.18729138e-01
-3.22939813e-01 -3.74400020e-01 -1.70157343e-01 6.58265352e-01
-1.32361674e+00 1.02442718e+00 -5.41665435e-01 -4.55340147e-01
-1.14228773e+00 4.78126436e-01 4.28307921e-01 -2.41845414e-01
-2.15481281e-01 1.11954463e+00 2.71224789e-02 -8.23025763e-01
1.85185194e-01 -9.21545625e-02 -7.54407048e-01 3.55081353e-03
6.88606799e-01 1.05194807e+00 -2.91290224e-01 -6.04341030e-01
-4.56784427e-01 6.65885091e-01 2.32151806e-01 -2.87484169e-01
9.17833567e-01 -2.19243199e-01 -9.89108324e-01 4.28274393e-01
8.38715494e-01 -1.36269778e-01 -3.69319022e-02 4.16855574e-01
1.25943512e-01 -8.02164078e-01 -4.32406247e-01 -3.55811834e-01
-1.85641110e-01 5.54822803e-01 5.34874737e-01 8.99602413e-01
8.57008278e-01 -3.02566767e-01 8.18335712e-01 9.66778398e-01
5.72402716e-01 -1.68019545e+00 -2.13140488e-01 6.89524531e-01
9.12339568e-01 -9.28350806e-01 5.44190466e-01 -7.27165341e-01
-4.14997995e-01 1.32979155e+00 4.68304873e-01 -1.55325383e-01
1.27306652e+00 5.74917436e-01 1.14069022e-02 -3.70670199e-01
-9.44949508e-01 1.12705544e-01 3.75366658e-01 1.11147714e+00
5.06513238e-01 5.07374525e-01 -9.66500878e-01 3.21953118e-01
-7.28706717e-01 -2.34555230e-01 4.90474731e-01 1.41972518e+00
-6.43810987e-01 -1.20391917e+00 -7.23931909e-01 4.61561680e-01
-5.49773455e-01 -1.16372630e-01 -5.28106689e-01 8.46754074e-01
-5.76629937e-02 1.49448860e+00 -1.97535474e-03 -5.03491640e-01
4.11356747e-01 1.41089618e-01 7.21762300e-01 -1.23501487e-01
-9.04846430e-01 4.15413082e-01 3.84890139e-01 -1.23056793e+00
-8.58632207e-01 -1.05834293e+00 -1.40667629e+00 -1.19437599e+00
-5.12782812e-01 1.89310342e-01 3.83317530e-01 3.90289724e-01
-2.06836104e-01 2.85260603e-02 9.80917513e-01 -1.07949340e+00
-3.90341938e-01 -9.39418614e-01 -1.00026262e+00 -7.84516707e-02
2.89751232e-01 -8.00943017e-01 -7.83523321e-01 2.83909619e-01] | [-7.289156913757324, 3.623215675354004] |
166f6b61-83b9-4737-8ee0-7656794864ea | as-projective-as-possible-image-stitching | null | null | http://openaccess.thecvf.com/content_cvpr_2013/html/Zaragoza_As-Projective-As-Possible_Image_Stitching_2013_CVPR_paper.html | http://openaccess.thecvf.com/content_cvpr_2013/papers/Zaragoza_As-Projective-As-Possible_Image_Stitching_2013_CVPR_paper.pdf | As-Projective-As-Possible Image Stitching with Moving DLT | We investigate projective estimation under model inadequacies, i.e., when the underpinning assumptions of the projective model are not fully satisfied by the data. We focus on the task of image stitching which is customarily solved by estimating a projective warp -a model that is justified when the scene is planar or when the views differ purely by rotation. Such conditions are easily violated in practice, and this yields stitching results with ghosting artefacts that necessitate the usage of deghosting algorithms. To this end we propose as-projective-as-possible warps, i.e., warps that aim to be globally projective, yet allow local non-projective deviations to account for violations to the assumed imaging conditions. Based on a novel estimation technique called Moving Direct Linear Transformation (Moving DLT), our method seamlessly bridges image regions that are inconsistent with the projective model. The result is highly accurate image stitching, with significantly reduced ghosting effects, thus lowering the dependency on post hoc deghosting. | ['David Suter', 'Tat-Jun Chin', 'Michael S. Brown', 'Julio Zaragoza'] | 2013-06-01 | null | null | null | cvpr-2013-6 | ['image-stitching'] | ['computer-vision'] | [ 7.11079895e-01 4.54163626e-02 2.21476942e-01 2.44535264e-02
-3.83221358e-01 -7.17847288e-01 8.29070985e-01 -3.62190038e-01
-2.99557865e-01 5.49279451e-01 -2.43720170e-02 -1.63086653e-01
-1.00397021e-01 -2.97949404e-01 -7.34724164e-01 -8.86115611e-01
9.23851728e-02 2.37866536e-01 2.06443787e-01 -1.59055263e-01
5.14388561e-01 5.58924735e-01 -1.35932100e+00 -1.27092740e-02
7.08124220e-01 4.43924308e-01 -1.70301303e-01 9.02931750e-01
4.00475562e-01 4.25496697e-01 -5.41620076e-01 -5.35634339e-01
6.68394387e-01 -4.53558654e-01 -6.17979944e-01 4.41765606e-01
8.66787434e-01 -6.27001345e-01 -6.26925007e-02 1.13418078e+00
1.33965373e-01 -1.31543338e-01 5.14899671e-01 -1.26212800e+00
1.84051711e-02 1.17091872e-01 -8.45574200e-01 -1.60329819e-01
4.44405615e-01 1.60077453e-01 5.69378018e-01 -5.92029810e-01
8.08918238e-01 1.03261483e+00 8.27116489e-01 1.11260116e-01
-1.73693407e+00 -3.75159055e-01 -3.09436768e-01 -8.27924386e-02
-1.60763073e+00 -7.50459194e-01 9.09603536e-01 -4.50616151e-01
4.02314603e-01 5.83253980e-01 4.37457711e-01 1.15056980e+00
5.64559996e-01 -5.93389496e-02 1.24951196e+00 -6.29206061e-01
6.59604147e-02 1.28783867e-01 -3.13987851e-01 4.61921185e-01
4.46069121e-01 3.44686806e-01 -4.59028304e-01 -2.31148705e-01
1.13697708e+00 -4.15979743e-01 -4.44227904e-01 -7.90522814e-01
-1.64396799e+00 5.03412902e-01 4.86526042e-02 4.01474267e-01
-1.92117423e-01 1.80969033e-02 2.45388225e-01 4.84166473e-01
3.71829510e-01 6.81140482e-01 -3.41913328e-02 7.85340071e-02
-1.36283457e+00 2.76127607e-01 6.37115777e-01 9.14141357e-01
6.99454606e-01 1.93103179e-01 3.33612055e-01 5.52176654e-01
2.72809509e-02 4.36213255e-01 1.73497975e-01 -1.07947385e+00
3.41319382e-01 2.49531381e-02 2.63052255e-01 -1.29032874e+00
-1.91185892e-01 -2.40342706e-01 -8.33466947e-01 4.59793538e-01
5.79992294e-01 1.24077789e-01 -6.22806966e-01 1.63124144e+00
5.70687950e-01 4.99611169e-01 1.04894400e-01 7.82066405e-01
7.24026784e-02 3.40608358e-01 -4.61046189e-01 -5.50542831e-01
1.22956979e+00 -4.05979127e-01 -7.83849835e-01 -1.23228222e-01
7.23182619e-01 -1.23367918e+00 8.75753939e-01 5.87297261e-01
-1.03278029e+00 -4.67968881e-01 -1.37977350e+00 3.37379649e-02
1.01712987e-01 -3.01865309e-01 1.77226543e-01 9.79440749e-01
-1.09177244e+00 7.21446514e-01 -7.33073294e-01 -1.84909299e-01
-2.59651005e-01 3.59822124e-01 -7.69684732e-01 4.44824584e-02
-6.90009832e-01 1.12911630e+00 5.02705216e-01 2.71524042e-01
-2.92678922e-01 -6.29022062e-01 -8.27673316e-01 -1.71856642e-01
4.73284036e-01 -6.30923331e-01 9.27667379e-01 -1.34238839e+00
-1.46779656e+00 9.22014296e-01 -4.42366228e-02 -3.73856395e-01
1.15970564e+00 -2.06012323e-01 -2.08621249e-01 2.45146766e-01
-1.75654173e-01 6.37933493e-01 1.50864720e+00 -1.77183175e+00
1.03508011e-01 -1.20896518e-01 -2.44037807e-01 2.31546968e-01
2.75762677e-02 -1.24853641e-01 -5.22220790e-01 -6.22516811e-01
3.62105876e-01 -1.29231071e+00 -1.57438636e-01 -7.65456632e-02
-7.15742290e-01 8.46740723e-01 1.06065083e+00 -8.93003345e-01
9.42413628e-01 -2.27452135e+00 1.84410453e-01 4.69186395e-01
2.16289118e-01 1.41441852e-01 -1.36406153e-01 6.14399195e-01
-4.49883521e-01 -1.22538261e-01 -3.05513293e-01 -5.02463698e-01
-2.97453076e-01 3.57994735e-01 -4.17900801e-01 1.05196929e+00
5.97250648e-02 5.17923713e-01 -6.46173120e-01 -6.15957618e-01
5.49014390e-01 4.98918891e-01 -6.25474334e-01 1.77235663e-01
5.11234514e-02 8.70251656e-01 5.04754335e-02 1.35796905e-01
1.09923804e+00 2.10762739e-01 3.60219479e-01 -5.08536637e-01
-5.74287057e-01 -5.45086861e-02 -1.36152339e+00 1.50750470e+00
-2.30473340e-01 7.15700686e-01 1.28091583e-02 -6.25977159e-01
9.01223958e-01 2.51837134e-01 6.75501645e-01 -4.11478668e-01
1.92137435e-01 4.59691793e-01 7.88382366e-02 -3.71801168e-01
7.50999570e-01 -2.15037644e-01 2.84074813e-01 1.67765677e-01
-7.87338242e-02 -6.26728594e-01 9.12961550e-03 1.00705899e-01
8.57677579e-01 4.07892585e-01 5.77424288e-01 -5.95235288e-01
4.50055122e-01 -9.02500078e-02 4.36719269e-01 5.92020750e-01
8.65345672e-02 1.01314533e+00 6.72791839e-01 -2.16743037e-01
-1.69991994e+00 -7.75800586e-01 -1.85499683e-01 1.61885582e-02
4.04856682e-01 -3.25139731e-01 -1.06595111e+00 -2.75784135e-01
-3.73711228e-01 6.03855371e-01 -6.17413461e-01 -1.85753107e-01
-7.49130845e-01 -3.66368532e-01 5.80160081e-01 -1.03040732e-01
4.96636391e-01 -4.44798946e-01 -8.05722117e-01 -1.03144872e-03
-1.88713759e-01 -1.46736860e+00 -5.98320782e-01 -1.63378179e-01
-9.73346412e-01 -1.31569886e+00 -8.55809093e-01 -3.39910269e-01
9.45356369e-01 5.53753972e-01 8.56168211e-01 1.53299153e-01
-8.86674300e-02 3.82070988e-01 -2.95819342e-01 3.00664127e-01
-8.05455983e-01 -3.17985326e-01 1.50213301e-01 2.78643668e-01
-3.38368177e-01 -6.52469456e-01 -4.23530370e-01 6.24455273e-01
-1.37380862e+00 6.21875286e-01 3.58417362e-01 7.84165502e-01
2.30427369e-01 1.76870123e-01 -2.56278634e-01 -8.05441260e-01
2.12412149e-01 1.22584105e-01 -7.99984515e-01 9.32530016e-02
-4.73831892e-01 1.97378561e-01 5.61433494e-01 -7.92924404e-01
-1.08465958e+00 -5.52673824e-02 8.45648441e-03 -6.76935554e-01
-1.08777039e-01 1.93838939e-01 -1.33925244e-01 -6.24838114e-01
7.10062385e-01 1.92145824e-01 1.79016203e-01 -2.09508047e-01
2.47055605e-01 2.58997262e-01 6.73400223e-01 -4.44780052e-01
1.32040823e+00 7.11016238e-01 3.30882847e-01 -1.41419470e+00
-1.76322535e-01 -2.85722375e-01 -1.01119351e+00 -4.05463576e-01
7.07531512e-01 -5.14595985e-01 -4.76042598e-01 5.87511539e-01
-1.25447965e+00 -2.15608463e-01 -1.58950895e-01 5.64179182e-01
-7.38992512e-01 1.14917552e+00 -2.83801436e-01 -6.37339413e-01
9.43453312e-02 -1.32083285e+00 9.46381390e-01 -2.23751545e-01
-4.12449181e-01 -9.25362289e-01 2.87393987e-01 1.45218000e-01
2.47534439e-01 6.72723114e-01 6.57817006e-01 -2.71230936e-01
-5.47436953e-01 9.35515575e-03 -7.97263160e-02 3.49190086e-01
5.79777211e-02 3.98258895e-01 -9.58099306e-01 -5.08198559e-01
3.39647502e-01 1.72383741e-01 3.54606628e-01 3.03207248e-01
4.82078403e-01 -5.50312281e-01 4.31650765e-02 7.92339504e-01
1.52716839e+00 6.95040897e-02 9.02295649e-01 3.79331470e-01
6.60584509e-01 9.11658287e-01 4.35404062e-01 2.91088581e-01
-1.87159508e-01 1.23943365e+00 4.97200996e-01 -1.63215607e-01
-7.23378211e-02 -1.90418214e-01 3.52506846e-01 6.96254015e-01
-1.42357215e-01 -1.10498652e-01 -6.88141286e-01 2.85536408e-01
-1.62875128e+00 -8.71020973e-01 -5.85804820e-01 2.89695978e+00
6.58800900e-01 1.86029881e-01 -1.52637407e-01 3.28354329e-01
7.11596966e-01 4.56186712e-01 -1.89240307e-01 -4.44797873e-01
1.86989121e-02 -2.04662643e-02 8.43490064e-01 8.46447170e-01
-9.23438489e-01 6.87099993e-01 5.87381554e+00 4.75423962e-01
-1.20104873e+00 -7.51545206e-02 2.51539022e-01 2.54637957e-01
-3.08201224e-01 4.31288511e-01 -2.85071254e-01 3.90010864e-01
5.97922027e-01 1.96225092e-01 4.33705837e-01 3.55621845e-01
3.03054035e-01 -2.62029231e-01 -1.05367637e+00 1.01116276e+00
2.69655466e-01 -9.16648090e-01 -2.01101080e-01 3.92790079e-01
6.44181013e-01 -3.61321628e-01 2.10693330e-01 -3.44293982e-01
-1.39982834e-01 -5.77754200e-01 8.79570544e-01 2.40970284e-01
9.23629105e-01 -4.88032669e-01 6.03525877e-01 2.04539344e-01
-8.30940843e-01 3.59893292e-01 -1.80609584e-01 3.33265424e-01
4.12528813e-01 6.13644898e-01 -6.37185574e-01 8.02842617e-01
1.12975650e-01 4.96472836e-01 -3.99907649e-01 9.33655798e-01
-8.33285004e-02 1.55585438e-01 -4.73634422e-01 7.64605999e-01
1.85466707e-01 -5.81420958e-01 8.25494647e-01 9.78035927e-01
6.48805022e-01 -1.09801047e-01 -1.95071697e-01 7.27790415e-01
4.79972810e-01 6.39588237e-02 -9.12105858e-01 2.89263129e-01
2.17589736e-01 1.04212165e+00 -9.32399273e-01 -1.50857732e-01
-2.58720547e-01 1.31443655e+00 -3.11046869e-01 3.44425887e-01
-8.18555117e-01 1.30619213e-01 5.31849563e-01 1.57401457e-01
5.78325912e-02 -5.59162796e-01 -2.00164288e-01 -1.53763103e+00
1.54124662e-01 -1.12996626e+00 -9.08788592e-02 -8.17406952e-01
-6.94283962e-01 4.28226203e-01 2.93771476e-01 -1.63772690e+00
-4.29935008e-01 -3.96784097e-01 -6.30729318e-01 5.45285046e-01
-1.31928909e+00 -1.27347529e+00 -3.13813567e-01 6.31835222e-01
2.68066555e-01 4.81166869e-01 5.80988348e-01 1.62745416e-01
-2.41537988e-01 3.67362946e-01 2.54175365e-02 -3.73790562e-01
1.03671026e+00 -8.31154406e-01 3.03030699e-01 1.24520791e+00
3.27848732e-01 7.33433008e-01 1.36034691e+00 -6.40310466e-01
-1.34905684e+00 -5.94272196e-01 7.46607780e-01 -2.08225429e-01
6.22945607e-01 -4.01192039e-01 -1.13792312e+00 7.15867341e-01
2.18808249e-01 -2.26338938e-01 2.73186773e-01 -3.34200174e-01
-6.10453308e-01 1.27668738e-01 -1.08368242e+00 8.28175068e-01
7.38140285e-01 -3.77424926e-01 -3.45021576e-01 1.33382291e-01
4.96439338e-01 -6.07850015e-01 -6.89082980e-01 1.84072644e-01
7.42915571e-01 -1.36117041e+00 1.15545058e+00 7.18099102e-02
3.19685727e-01 -4.18895572e-01 -2.81804260e-02 -1.00877857e+00
-4.92132641e-02 -1.12946427e+00 2.35496044e-01 1.09441340e+00
3.34065291e-04 -8.28724205e-01 5.72159529e-01 5.72416306e-01
-4.01852205e-02 1.67359948e-01 -1.23662806e+00 -1.02648103e+00
-2.96839148e-01 -1.66127741e-01 3.42245221e-01 1.41049170e+00
-1.59495205e-01 -9.67375711e-02 -8.96720886e-01 4.23256904e-01
7.73630440e-01 -3.03292602e-01 1.18488705e+00 -9.11444187e-01
-4.71736878e-01 -2.95852035e-01 -6.02915227e-01 -8.24312687e-01
-1.58867836e-02 -1.04304895e-01 5.54529019e-02 -5.42842567e-01
-9.70405564e-02 -3.62066180e-01 3.95656914e-01 1.30108282e-01
1.12187453e-01 4.58650321e-01 2.63159037e-01 5.85303962e-01
1.93916336e-01 3.28134120e-01 1.16497147e+00 2.73656100e-01
-2.45190933e-01 -1.83503136e-01 -2.34411731e-01 6.95799053e-01
6.48008704e-01 -3.89766574e-01 -4.38840240e-01 -3.38532001e-01
4.00224656e-01 3.56053203e-01 5.59620976e-01 -1.02305293e+00
6.34048209e-02 -1.45021603e-01 -3.87203135e-03 -2.49540180e-01
5.51362991e-01 -1.29248297e+00 8.56143177e-01 6.30686283e-01
-2.06295192e-01 1.30639359e-01 1.55013308e-01 1.79154903e-01
-1.73672289e-01 -4.58262116e-01 9.90752518e-01 2.69739538e-01
-3.09368849e-01 -5.71620837e-02 9.48402565e-03 -5.20634949e-01
1.16534054e+00 -6.58662856e-01 -2.95643032e-01 -4.39020425e-01
-3.93014044e-01 -5.40297389e-01 1.15446162e+00 1.33022651e-01
4.35348690e-01 -1.15112162e+00 -7.08183944e-01 5.87232530e-01
4.05892953e-02 -2.04391703e-01 3.82128060e-01 1.22084618e+00
-9.99317229e-01 4.77780625e-02 -1.84870362e-01 -6.91577196e-01
-1.72664356e+00 7.20965445e-01 2.38442779e-01 -3.01261306e-01
-9.00598645e-01 2.12346435e-01 3.39637756e-01 -1.59395665e-01
-3.35356712e-01 -2.25634426e-01 1.41928926e-01 -2.22327963e-01
3.59876007e-01 1.46457568e-01 2.29852363e-01 -9.54414308e-01
-8.56274813e-02 7.75144815e-01 -6.92492500e-02 -5.21953881e-01
9.14630055e-01 -2.07286760e-01 -4.19575162e-02 1.21643923e-01
1.15368068e+00 4.48407382e-01 -1.52723336e+00 -7.91853517e-02
-2.35905662e-01 -9.52831030e-01 8.40222687e-02 -2.56740361e-01
-8.53937149e-01 6.88239396e-01 3.91935736e-01 3.16507936e-01
9.83234346e-01 -4.92133051e-01 4.38504636e-01 -1.67342331e-02
3.04353356e-01 -7.23899901e-01 -1.53324887e-01 1.31387487e-01
9.34286237e-01 -8.83399367e-01 4.03825879e-01 -6.13359630e-01
-4.53562677e-01 1.25829875e+00 8.14329237e-02 -2.67773718e-01
7.56910592e-02 3.51862431e-01 -1.48955639e-03 4.75313365e-02
-2.61162221e-01 2.46553659e-01 2.69075632e-01 7.16283917e-01
8.66039246e-02 -1.30435959e-01 -3.33980739e-01 -5.40399790e-01
-2.47100085e-01 -2.80177474e-01 8.37193787e-01 8.72969866e-01
-5.85269891e-02 -1.23106837e+00 -1.04919159e+00 -1.51837572e-01
-1.02904439e-01 -4.89740930e-02 -4.67378020e-01 1.16791093e+00
-2.27209106e-01 7.47704148e-01 -2.48374287e-02 -4.08811688e-01
1.09094240e-01 -1.61179602e-01 7.83031404e-01 -9.99044478e-02
-4.93161529e-01 5.32119036e-01 1.58779234e-01 -6.32125854e-01
-5.86831510e-01 -6.92281842e-01 -5.71940541e-01 -7.81992257e-01
-5.49074888e-01 -1.32657737e-01 7.64421880e-01 9.33796406e-01
1.18137017e-01 1.49443895e-01 6.66464329e-01 -1.26132941e+00
-4.93216187e-01 -6.46501601e-01 -5.65641820e-01 4.79147702e-01
4.90757525e-01 -5.94894171e-01 -5.70517302e-01 4.42881733e-01] | [9.367898941040039, -2.3596506118774414] |
89c36a5a-bc0b-4683-b868-199d76897a76 | learning-high-precision-bounding-box-for | 2106.01883 | null | https://arxiv.org/abs/2106.01883v5 | https://arxiv.org/pdf/2106.01883v5.pdf | Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence | Existing rotated object detectors are mostly inherited from the horizontal detection paradigm, as the latter has evolved into a well-developed area. However, these detectors are difficult to perform prominently in high-precision detection due to the limitation of current regression loss design, especially for objects with large aspect ratios. Taking the perspective that horizontal detection is a special case for rotated object detection, in this paper, we are motivated to change the design of rotation regression loss from induction paradigm to deduction methodology, in terms of the relation between rotation and horizontal detection. We show that one essential challenge is how to modulate the coupled parameters in the rotation regression loss, as such the estimated parameters can influence to each other during the dynamic joint optimization, in an adaptive and synergetic way. Specifically, we first convert the rotated bounding box into a 2-D Gaussian distribution, and then calculate the Kullback-Leibler Divergence (KLD) between the Gaussian distributions as the regression loss. By analyzing the gradient of each parameter, we show that KLD (and its derivatives) can dynamically adjust the parameter gradients according to the characteristics of the object. It will adjust the importance (gradient weight) of the angle parameter according to the aspect ratio. This mechanism can be vital for high-precision detection as a slight angle error would cause a serious accuracy drop for large aspect ratios objects. More importantly, we have proved that KLD is scale invariant. We further show that the KLD loss can be degenerated into the popular $l_{n}$-norm loss for horizontal detection. Experimental results on seven datasets using different detectors show its consistent superiority, and codes are available at https://github.com/yangxue0827/RotationDetection and https://github.com/open-mmlab/mmrotate. | ['Junchi Yan', 'Qi Tian', 'Wentao Wang', 'Qi Ming', 'Jirui Yang', 'Xiaojiang Yang', 'Xue Yang'] | 2021-06-03 | null | http://proceedings.neurips.cc/paper/2021/hash/98f13708210194c475687be6106a3b84-Abstract.html | http://proceedings.neurips.cc/paper/2021/file/98f13708210194c475687be6106a3b84-Paper.pdf | neurips-2021-12 | ['object-detection-in-aerial-images'] | ['computer-vision'] | [-2.51981854e-01 -2.31526002e-01 -1.65300593e-01 -1.03071921e-01
-5.66802204e-01 -4.24196571e-01 2.69664854e-01 -2.45799109e-01
-5.96281469e-01 2.55766034e-01 -1.21343926e-01 -2.44059607e-01
-1.43663213e-01 -6.12004101e-01 -5.74613690e-01 -1.01570117e+00
1.44249320e-01 1.45336539e-01 5.51955283e-01 -2.45414376e-01
2.62552291e-01 6.50283217e-01 -1.37086797e+00 -2.55682856e-01
6.51350081e-01 9.77830946e-01 2.93418527e-01 5.12461841e-01
4.33814138e-01 1.03605002e-01 -5.99731207e-01 -3.28570902e-01
4.89433616e-01 -1.19569920e-01 -2.31947586e-01 -2.09817290e-01
1.29780337e-01 -2.25342050e-01 -1.73081592e-01 1.28700030e+00
6.99258208e-01 -1.06749766e-01 7.20294118e-01 -1.02391016e+00
-2.93089777e-01 4.68539625e-01 -1.09796751e+00 1.34160504e-01
1.13577262e-01 1.47828117e-01 9.93284643e-01 -1.20235622e+00
4.36663091e-01 1.28009224e+00 6.28057420e-01 3.34837526e-01
-1.03911877e+00 -7.77233958e-01 3.50979686e-01 1.38937701e-02
-1.59709120e+00 -4.72481884e-02 7.57096648e-01 -6.32674277e-01
4.82885033e-01 1.89333960e-01 6.07097447e-01 9.22621250e-01
1.05925038e-01 7.83389151e-01 7.28373349e-01 -4.69541371e-01
-1.79209337e-01 2.76279807e-01 2.51322407e-02 6.23609066e-01
7.08134830e-01 8.00706074e-02 -8.51566866e-02 1.20292135e-01
8.67916048e-01 -1.59063146e-01 -4.71237898e-01 -7.22505569e-01
-1.15352213e+00 8.49745214e-01 6.87489033e-01 2.32909903e-01
1.29223838e-01 3.76507156e-02 2.33620837e-01 6.79394826e-02
1.77444369e-01 5.07479250e-01 -2.82242328e-01 7.93992430e-02
-3.45059484e-01 3.26906592e-01 4.28020567e-01 7.44033158e-01
4.32394981e-01 -6.79046959e-02 -2.49323353e-01 7.68223584e-01
6.15240574e-01 6.58575118e-01 4.10594523e-01 -3.70900333e-01
6.20064616e-01 4.31871116e-01 2.65959322e-01 -1.21480918e+00
-6.70570791e-01 -7.90105224e-01 -6.63611531e-01 2.71676362e-01
6.80493534e-01 -1.70960709e-01 -4.07444000e-01 1.69503784e+00
6.21642113e-01 -4.50319380e-01 -3.92743737e-01 1.24635530e+00
4.39079314e-01 2.61582643e-01 -2.30068207e-01 -1.89365551e-01
1.71579456e+00 -6.67208076e-01 -5.32229781e-01 -1.49286568e-01
4.95154709e-01 -1.00695193e+00 1.17824602e+00 3.65185708e-01
-8.68534088e-01 -5.82128465e-01 -1.43579531e+00 1.26142412e-01
-1.73477799e-01 5.70995867e-01 4.44458842e-01 5.48311234e-01
-4.99400347e-01 2.33518884e-01 -7.59569764e-01 -2.52277762e-01
1.97882771e-01 2.91477978e-01 1.05982363e-01 4.25857246e-01
-1.11923230e+00 8.89629543e-01 3.14250231e-01 3.39270502e-01
-3.49403381e-01 -4.81793970e-01 -5.73130071e-01 -1.32010549e-01
5.41093886e-01 -6.51769161e-01 1.10560417e+00 -3.88424456e-01
-1.41130328e+00 7.44968057e-01 2.55336255e-01 -2.43645191e-01
1.02461505e+00 -4.68437940e-01 -2.43286997e-01 -6.77167997e-02
2.54572500e-02 4.87045109e-01 1.06006551e+00 -1.19382882e+00
-6.73222899e-01 -6.55499518e-01 -2.38304455e-02 3.20967406e-01
-3.98874938e-01 3.21776830e-02 -7.03763723e-01 -6.84766769e-01
3.78047705e-01 -1.11876881e+00 -1.21455863e-01 1.57774255e-01
-6.38076782e-01 -6.51728585e-02 8.09587121e-01 -2.54768610e-01
1.39331985e+00 -2.40209961e+00 -1.99277420e-02 9.70062688e-02
8.28254968e-02 2.53916889e-01 2.80039579e-01 4.51472066e-02
5.61660901e-02 -2.39856895e-02 -6.24543689e-02 8.87608388e-04
-9.88921672e-02 -4.80551004e-01 -3.51028472e-01 8.73235464e-01
1.73291638e-01 5.96059442e-01 -8.49865317e-01 -4.15286511e-01
2.14530051e-01 6.74981892e-01 -4.79182214e-01 2.22757068e-02
1.93460673e-01 3.37100416e-01 -7.91884243e-01 5.44592559e-01
8.34612131e-01 -1.80177167e-01 1.20851202e-02 -5.82696438e-01
-3.88163388e-01 -5.36343530e-02 -1.47250295e+00 1.15690076e+00
-1.60666406e-01 5.26751578e-01 1.19225234e-01 -8.78638864e-01
1.01811659e+00 -1.53900608e-01 3.67916107e-01 -4.41934764e-01
3.72815877e-01 2.17105284e-01 9.59060267e-02 -2.58085489e-01
5.57558060e-01 1.20332297e-02 -1.41156197e-01 -9.71587822e-02
-4.51004028e-01 -1.96482450e-01 1.38739452e-01 -8.15057456e-02
4.90374863e-01 1.46154091e-01 5.24572730e-01 -7.58291855e-02
4.76997793e-01 -1.85127959e-01 4.91938204e-01 6.48392916e-01
-1.70316502e-01 7.07610488e-01 6.65444016e-01 -1.02887280e-01
-8.26981604e-01 -1.03535557e+00 -8.20840359e-01 7.62351632e-01
5.76318800e-01 -1.74566805e-01 -5.30033112e-01 -5.44678748e-01
2.12271735e-01 3.69882762e-01 -4.20459092e-01 -3.46888721e-01
-7.99586892e-01 -1.19558668e+00 4.17092562e-01 5.25453329e-01
5.00612617e-01 -6.83017075e-01 -8.84487510e-01 -3.32894884e-02
4.46091928e-02 -1.03925407e+00 -5.28895199e-01 9.27537233e-02
-6.08631313e-01 -1.10381472e+00 -6.55048013e-01 -5.25867701e-01
7.25397170e-01 4.64082748e-01 5.92406631e-01 -2.00543195e-01
-5.50078452e-01 2.16020629e-01 -3.84343684e-01 -5.08148789e-01
1.24501679e-02 1.57454967e-01 1.78325564e-01 -1.62137337e-02
9.19401944e-02 -3.78956258e-01 -8.84290457e-01 8.80053222e-01
-6.75525248e-01 -1.32219633e-02 7.35032260e-01 6.37303054e-01
6.37275875e-01 -5.16202934e-02 2.94701517e-01 -6.03785694e-01
4.90377426e-01 -1.42161608e-01 -1.23067641e+00 1.17953241e-01
-6.19706750e-01 7.75641948e-02 3.41101229e-01 -5.72469234e-01
-8.53801727e-01 4.71433513e-02 7.94781372e-02 -3.97389382e-01
2.97282338e-01 1.64834663e-01 -2.45218053e-01 -2.93392595e-03
6.74889982e-01 1.91471688e-02 -1.77397311e-01 -4.28956896e-01
4.23750013e-01 5.69635153e-01 2.82430321e-01 -4.52654392e-01
1.08958066e+00 7.20312953e-01 2.33847290e-01 -7.69666314e-01
-7.21050084e-01 -5.70945799e-01 -4.70512211e-01 -2.96460450e-01
8.61316204e-01 -7.64807820e-01 -9.20953810e-01 5.09245992e-01
-9.47109401e-01 3.54962945e-02 -1.40706450e-01 7.54473090e-01
-1.92291215e-01 4.39643204e-01 -4.01256144e-01 -7.79095352e-01
-2.56209880e-01 -1.38034010e+00 1.05870581e+00 3.17422003e-01
1.72450468e-01 -5.34745097e-01 -5.78011647e-02 1.19728915e-01
1.98220506e-01 9.16460603e-02 5.64825118e-01 -3.75689894e-01
-4.79055315e-01 -5.53446829e-01 -4.39336747e-01 2.34401003e-01
-1.01118796e-01 2.88609087e-01 -7.87160635e-01 -4.52014357e-01
2.03135967e-01 -2.91361474e-02 9.39938843e-01 5.50825775e-01
1.09099782e+00 -3.11574005e-02 -5.38779557e-01 7.70252228e-01
1.27940738e+00 5.79066575e-02 4.69725043e-01 5.06748438e-01
7.64993608e-01 3.90439779e-01 9.51401174e-01 5.52591085e-01
1.02262311e-01 1.22252154e+00 5.71986139e-01 6.62288442e-02
-1.05029374e-01 -1.97647423e-01 4.78200495e-01 6.72288895e-01
-1.45178750e-01 3.58602628e-02 -6.74980521e-01 -1.78623979e-03
-1.78967738e+00 -7.47142315e-01 -2.17656180e-01 2.67522669e+00
6.39305711e-01 4.38642532e-01 2.35948995e-01 -1.33780660e-02
8.31477463e-01 1.55779213e-01 -5.67448676e-01 9.80415940e-02
-1.40695348e-01 -4.30843920e-01 7.36155212e-01 4.57649797e-01
-1.27084112e+00 7.93115377e-01 5.00771189e+00 9.86726820e-01
-1.42135024e+00 -1.31581664e-01 2.55027056e-01 -1.12259492e-01
5.85230850e-02 2.97434088e-02 -1.42352223e+00 4.93304074e-01
5.50824553e-02 4.06194068e-02 1.01880003e-02 1.10357893e+00
2.60775864e-01 -5.07958280e-03 -7.73855150e-01 8.30662251e-01
-2.06378847e-01 -6.76092625e-01 -1.06603995e-01 1.24842227e-01
3.82386178e-01 -2.10685022e-02 3.52755338e-01 2.54066378e-01
-2.33248249e-01 -5.52093029e-01 8.27545762e-01 2.66840845e-01
6.52650774e-01 -5.65917790e-01 5.69415510e-01 2.61166245e-01
-1.40433240e+00 -1.23937443e-01 -4.88342404e-01 2.50549495e-01
8.20266679e-02 7.81477630e-01 -9.31000113e-01 4.24626261e-01
7.49547362e-01 3.84345174e-01 -5.56546032e-01 1.15139401e+00
-5.61147332e-01 1.84232518e-01 -3.91146123e-01 -1.69441760e-01
-2.13611066e-01 -5.23720205e-01 1.01840544e+00 1.22039557e+00
3.33839744e-01 -3.04977328e-01 1.29517436e-01 7.83980668e-01
1.07722074e-01 2.52335221e-01 -3.44552487e-01 4.53672498e-01
5.69431365e-01 1.43833375e+00 -7.72798121e-01 6.71785884e-03
-3.48892510e-01 4.09102052e-01 2.56757379e-01 3.73891532e-01
-1.28059399e+00 -5.02890944e-01 6.16713345e-01 3.80661249e-01
6.17999434e-01 -2.04082102e-01 -2.19829306e-01 -1.31590271e+00
4.32588249e-01 -5.80326676e-01 3.15089047e-01 -4.82858539e-01
-1.06889737e+00 3.97210985e-01 1.61542490e-01 -1.70484936e+00
9.59809572e-02 -7.74369597e-01 -4.27593887e-01 5.07763684e-01
-1.32495570e+00 -8.46147597e-01 -3.86108637e-01 3.23749930e-01
3.24374616e-01 1.46156669e-01 1.84113130e-01 4.56602126e-01
-9.24896240e-01 8.76126170e-01 4.28000204e-02 2.47809157e-01
8.18592906e-01 -1.08833802e+00 -3.39104310e-02 8.36939454e-01
-2.31254417e-02 7.77899384e-01 9.26455021e-01 -3.67739230e-01
-1.47397530e+00 -9.02189016e-01 2.81946480e-01 -2.55092651e-01
8.10610592e-01 -4.96435821e-01 -7.02564716e-01 3.97238761e-01
-3.94865423e-01 2.07154274e-01 1.54237524e-01 1.67171732e-02
-4.30358380e-01 -3.43028754e-01 -9.04419601e-01 7.81089723e-01
9.91179347e-01 -1.17726885e-01 -2.07789838e-01 2.95633823e-01
7.21416056e-01 -4.62247282e-01 -6.31040752e-01 8.56992662e-01
8.62405837e-01 -9.22522008e-01 1.08366978e+00 -2.62536228e-01
-7.11313933e-02 -7.48480439e-01 -7.54648894e-02 -9.51989949e-01
-2.46267155e-01 -4.22568291e-01 -1.94045492e-02 1.00878382e+00
4.72001195e-01 -9.68432724e-01 5.52096725e-01 6.98204860e-02
-3.65452133e-02 -1.01664472e+00 -9.44815040e-01 -9.26453471e-01
-5.41947111e-02 -4.48283285e-01 1.18561447e-01 6.29168808e-01
-1.07631810e-01 3.51393312e-01 -2.33809680e-01 3.97547007e-01
5.75992942e-01 1.75883710e-01 9.84388769e-01 -9.61475968e-01
-5.77329993e-01 -9.19481456e-01 -5.47222316e-01 -1.46706855e+00
-4.66679573e-01 -5.52320182e-01 1.55603383e-02 -1.01058793e+00
1.95657820e-01 -4.84049827e-01 -2.73282975e-01 2.41431780e-02
-3.31952006e-01 9.22342986e-02 1.99500203e-01 4.05496627e-01
-4.20784742e-01 6.24360025e-01 1.40186536e+00 2.81902682e-02
-3.56000245e-01 4.04781371e-01 -6.95640206e-01 8.94952416e-01
7.23667502e-01 -4.76429522e-01 -1.66288778e-01 -2.32643247e-01
4.65916246e-01 -2.05953717e-01 3.12523901e-01 -9.96682882e-01
2.01392010e-01 3.16763557e-02 2.84784287e-01 -5.53116918e-01
2.92739421e-01 -6.68970466e-01 -2.39995837e-01 6.33367062e-01
-2.82140691e-02 -1.05652653e-01 2.74757873e-02 4.90212888e-01
-1.08965978e-01 -4.45770584e-02 1.08732235e+00 2.80321151e-01
-2.48031408e-01 1.59647211e-01 1.55587457e-02 7.30677396e-02
1.05076373e+00 -2.84133911e-01 -3.66972059e-01 -9.70803052e-02
-5.31007707e-01 1.97179049e-01 4.05858427e-01 4.48393285e-01
4.27958697e-01 -1.17158163e+00 -6.14837289e-01 1.25448316e-01
2.24416941e-01 1.58046618e-01 1.24991737e-01 1.35232425e+00
-3.96234185e-01 3.83126467e-01 1.70389608e-01 -7.77416885e-01
-1.21171188e+00 5.35332859e-01 4.30579275e-01 -2.83222616e-01
-5.93589783e-01 8.82652521e-01 5.80476046e-01 -1.24289349e-01
3.86287510e-01 -5.02640903e-01 -2.66437024e-01 2.99697548e-01
4.32758242e-01 3.92345309e-01 -2.75533404e-02 -5.01395464e-01
-5.07596314e-01 1.09351039e+00 -1.62003905e-01 6.16751239e-02
9.39240992e-01 5.09710200e-02 2.64354169e-01 2.39828110e-01
1.17515790e+00 3.26703191e-01 -1.34465694e+00 -5.07813431e-02
-2.17507482e-01 -5.27584016e-01 -5.02341837e-02 -1.77937984e-01
-1.07407129e+00 6.64293885e-01 7.46559799e-01 3.21854174e-01
9.60764766e-01 2.16508985e-01 3.91971290e-01 2.81338125e-01
2.88218319e-01 -9.77119446e-01 2.52737999e-01 3.54587346e-01
1.02311242e+00 -1.16177750e+00 4.32480514e-01 -6.46502554e-01
-5.13480723e-01 1.11895061e+00 7.94611633e-01 -1.42837524e-01
7.40546346e-01 2.42831081e-01 6.99727014e-02 -2.23983869e-01
-1.77333787e-01 -2.06254274e-01 4.59560007e-01 2.34575897e-01
4.20066863e-01 1.97835281e-01 -6.37660503e-01 5.32843232e-01
-3.24437737e-01 -5.68684101e-01 1.57233551e-01 6.47593319e-01
-6.07578337e-01 -9.27732825e-01 -6.04520857e-01 1.21012315e-01
-3.64244998e-01 2.17697233e-01 -1.09498478e-01 1.12010646e+00
1.65458933e-01 5.18816233e-01 -1.58194855e-01 -3.06634128e-01
6.09719753e-01 -4.38342541e-01 5.73843360e-01 -3.76427621e-01
-1.00579612e-01 2.72651643e-01 -2.42906287e-01 -4.34723318e-01
-8.61968771e-02 -7.67505646e-01 -1.23386562e+00 1.03853561e-01
-7.82722831e-01 -5.99512234e-02 8.18296432e-01 5.52896082e-01
3.35003100e-02 3.32158059e-01 7.33149171e-01 -8.22377980e-01
-1.06487072e+00 -8.91481161e-01 -5.12611508e-01 2.42731079e-01
1.62102386e-01 -1.04102814e+00 -7.06340790e-01 -5.06247520e-01] | [8.620624542236328, -0.8266986012458801] |
e34053b1-347c-4e46-a4e3-b3225b8550d5 | on-the-construction-of-distribution-free | 2203.03150 | null | https://arxiv.org/abs/2203.03150v1 | https://arxiv.org/pdf/2203.03150v1.pdf | On the Construction of Distribution-Free Prediction Intervals for an Image Regression Problem in Semiconductor Manufacturing | The high-volume manufacturing of the next generation of semiconductor devices requires advances in measurement signal analysis. Many in the semiconductor manufacturing community have reservations about the adoption of deep learning; they instead prefer other model-based approaches for some image regression problems, and according to the 2021 IEEE International Roadmap for Devices and Systems (IRDS) report on Metrology a SEMI standardization committee may endorse this philosophy. The semiconductor manufacturing community does, however, communicate a need for state-of-the-art statistical analyses to reduce measurement uncertainty. Prediction intervals which characterize the reliability of the predictive performance of regression models can impact decisions, build trust in machine learning, and be applied to other regression models. However, we are not aware of effective and sufficiently simple distribution-free approaches that offer valid coverage for important classes of image data, so we consider the distribution-free conformal prediction and conformalized quantile regression framework.The image regression problem that is the focus of this paper pertains to line edge roughness (LER) estimation from noisy scanning electron microscopy images. LER affects semiconductor device performance and reliability as well as the yield of the manufacturing process; the 2021 IRDS emphasizes the crucial importance of LER by devoting a white paper to it in addition to mentioning or discussing it in the reports of multiple international focus teams. It is not immediately apparent how to effectively use normalized conformal prediction and quantile regression for LER estimation. The modeling techniques we apply appear to be novel for finding distribution-free prediction intervals for image data and will be presented at the 2022 SEMI Advanced Semiconductor Manufacturing Conference. | ['Serap A. Savari', 'Inimfon I. Akpabio'] | 2022-03-07 | null | null | null | null | ['prediction-intervals'] | ['miscellaneous'] | [ 2.70255595e-01 -6.81722760e-02 3.12322471e-02 -5.51645458e-01
-1.23007238e+00 -4.98488247e-02 2.91313440e-01 2.56354451e-01
-1.44459397e-01 8.10722768e-01 -4.01227832e-01 -6.75440967e-01
-7.10744262e-01 -6.93041205e-01 -4.47963655e-01 -8.28121483e-01
3.12539518e-01 7.87239015e-01 -1.77303150e-01 1.81576852e-02
7.72907197e-01 8.63025129e-01 -1.20635378e+00 2.45221227e-01
5.86181521e-01 1.40680218e+00 1.07348293e-01 4.01418298e-01
9.32226554e-02 5.44413567e-01 -6.66625381e-01 -1.96293578e-01
-6.28175586e-02 -2.32214913e-01 -3.37078542e-01 1.66632456e-03
8.17092583e-02 -7.87978917e-02 3.29283118e-01 8.22287381e-01
3.59753788e-01 -3.59595239e-01 1.05612755e+00 -9.41754937e-01
-6.41169727e-01 3.43820184e-01 -6.98020816e-01 2.33263120e-01
-1.89189643e-01 -7.49171153e-02 7.13235140e-01 -7.55385578e-01
4.98419195e-01 7.77678490e-01 7.50245035e-01 2.07707793e-01
-1.12218523e+00 -5.05263090e-01 -2.22599790e-01 1.02006830e-02
-1.24771333e+00 -3.79059881e-01 5.83982706e-01 -6.48257196e-01
1.11164391e+00 1.78223342e-01 1.80693164e-01 7.31874824e-01
1.34679389e+00 1.18867680e-01 1.24875486e+00 -4.89678890e-01
2.65657127e-01 2.58039623e-01 -2.49817278e-02 4.76019800e-01
3.66884381e-01 3.99192214e-01 3.60576026e-02 1.50351062e-01
8.10745001e-01 -1.43905804e-01 2.09587529e-01 -4.16944563e-01
-6.51744604e-01 8.82423937e-01 8.89172330e-02 5.04321456e-01
-1.89682424e-01 3.31051201e-01 4.73960042e-01 3.73867542e-01
5.41367352e-01 5.50684750e-01 -5.06127834e-01 -1.77452609e-01
-1.04781151e+00 1.44357800e-01 5.38942873e-01 8.74727666e-01
4.26153392e-01 9.77009609e-02 1.18166178e-01 6.46329522e-01
3.05945843e-01 4.48807210e-01 -2.36203089e-01 -1.11479235e+00
-7.16011152e-02 1.14268661e-01 2.06673011e-01 -6.79439008e-01
-6.92723274e-01 -3.16881150e-01 -6.38603926e-01 6.77317798e-01
1.21898189e-01 -2.54474849e-01 -8.28567982e-01 7.85907984e-01
-3.03588599e-01 -6.27285838e-01 -3.94340634e-01 5.46189427e-01
3.77109736e-01 3.69642556e-01 -3.11009079e-01 -4.20964181e-01
7.48347998e-01 -1.56558648e-01 -8.18840086e-01 1.62852585e-01
4.23410803e-01 -9.62728202e-01 5.98647654e-01 8.77165198e-01
-1.17667580e+00 -4.28947508e-01 -1.66985583e+00 2.17121586e-01
-3.35380226e-01 1.18110605e-01 7.54395247e-01 8.13778222e-01
-6.63446844e-01 9.93282855e-01 -1.04433000e+00 -1.33896142e-01
6.49720192e-01 6.48663402e-01 -5.90039454e-02 -3.60870034e-01
-6.26028001e-01 1.22476947e+00 -3.89637470e-01 1.64166808e-01
-5.12901723e-01 -7.84012437e-01 -6.86494887e-01 -2.78762668e-01
1.56799421e-01 -3.55621248e-01 1.06432581e+00 -5.89262366e-01
-1.32610428e+00 6.45184338e-01 8.16605464e-02 -3.48814666e-01
3.23861301e-01 7.31955692e-02 -6.87375426e-01 -1.86171770e-01
7.94778839e-02 2.20382586e-01 9.01015759e-01 -1.32882905e+00
-6.79911315e-01 -4.27949399e-01 -3.35032970e-01 -4.38758612e-01
4.28507566e-01 8.26216266e-02 -1.72006134e-02 1.08596705e-01
2.38389477e-01 -6.21450126e-01 -3.37628752e-01 -5.62384844e-01
-3.82479429e-01 -1.61265180e-01 7.47159839e-01 -4.61726248e-01
8.86839271e-01 -2.00431395e+00 -4.17677820e-01 4.07238960e-01
-6.46778122e-02 -5.42294741e-01 3.43608081e-01 4.68922317e-01
-2.71930750e-02 1.88750982e-01 -2.42547348e-01 -1.82071105e-01
1.31826848e-01 -9.99398381e-02 -9.17633921e-02 9.21857715e-01
5.55759668e-01 6.82224512e-01 -3.75778109e-01 7.06512388e-03
7.11842716e-01 3.91905040e-01 -6.30251467e-02 -4.12293077e-01
3.36997509e-02 3.96849155e-01 -3.24994177e-01 9.78969514e-01
7.54350305e-01 -2.10094422e-01 4.21542749e-02 -4.09049362e-01
-3.34102750e-01 4.61277785e-03 -6.51045024e-01 1.16245115e+00
-5.66865265e-01 6.44780517e-01 2.12506428e-02 -1.13290060e+00
1.30268860e+00 -2.21640826e-03 7.93439686e-01 -9.05553520e-01
1.94238991e-01 5.16840458e-01 2.44815364e-01 -2.33651817e-01
4.25131291e-01 -6.94007695e-01 -2.29952872e-01 -2.69314107e-02
-2.08920866e-01 -9.64120448e-01 -2.35465959e-01 -2.28982329e-01
1.19776964e+00 1.54340997e-01 -1.30331174e-01 -7.18443930e-01
2.41405830e-01 -6.00642897e-02 3.82577091e-01 5.52875519e-01
-1.01472102e-01 8.35824549e-01 6.05053544e-01 -2.67808288e-01
-1.44673276e+00 -1.27200806e+00 -9.10922885e-01 1.52878970e-01
-3.32366563e-02 -3.05334739e-02 -5.75004160e-01 -3.73476028e-01
2.58913159e-01 9.12933946e-01 -4.05772388e-01 -2.91502886e-02
-1.14927620e-01 -8.58947456e-01 -1.79689124e-01 6.52119756e-01
-2.49442756e-01 -8.16015720e-01 -4.28820997e-01 2.12222278e-01
6.73331320e-01 -1.08470547e+00 2.28540972e-01 7.48464525e-01
-6.98938847e-01 -1.00990570e+00 -2.68291533e-01 -2.90015578e-01
5.96364141e-01 -4.01330113e-01 1.13065684e+00 -9.17243138e-02
-5.90952933e-01 4.74428803e-01 -7.25901648e-02 -8.40934813e-01
-7.11687803e-01 -1.62142366e-01 -9.33972560e-03 -6.10803127e-01
6.20788515e-01 -2.65216291e-01 -4.70338553e-01 3.93954903e-01
-4.55150723e-01 -3.75480711e-01 7.45668054e-01 8.58764529e-01
8.36453199e-01 5.11572897e-01 1.17726398e+00 -1.24539948e+00
6.12159848e-01 -2.25588903e-01 -9.03521955e-01 -8.30616653e-02
-1.20523310e+00 -3.66758704e-01 4.09936696e-01 4.21169996e-01
-9.02618229e-01 -3.35133851e-01 -2.69199580e-01 -2.45083019e-01
-1.02776766e-01 6.34616077e-01 -4.81778830e-02 -1.54911250e-01
4.68531996e-01 -4.69441950e-01 1.31053567e-01 1.56494312e-03
-2.32686475e-01 7.78317273e-01 2.66828924e-01 -4.40368652e-01
5.37642539e-01 5.33821881e-01 4.37927693e-01 -1.06637287e+00
-4.99112546e-01 -3.06679904e-01 -4.46947128e-01 -2.29835555e-01
9.33742046e-01 -6.00729823e-01 -6.40983641e-01 1.75298020e-01
-7.52758026e-01 -1.74828112e-01 -3.52439821e-01 4.08798754e-01
-1.00004601e+00 5.84748900e-03 -6.30319238e-01 -1.05139530e+00
-3.55542600e-02 -1.45175195e+00 1.07620621e+00 -1.05354384e-01
-4.35499489e-01 -1.03272247e+00 -4.82633173e-01 5.53541541e-01
5.71950078e-01 4.54595953e-01 1.44253707e+00 -4.11793649e-01
-6.46086991e-01 -7.84025848e-01 -3.42434525e-01 8.15170527e-01
3.82813603e-01 2.40348130e-01 -9.89067793e-01 -2.75165498e-01
3.85807246e-01 -1.20783411e-01 7.37390876e-01 1.30017900e+00
1.42916811e+00 8.18098783e-01 -3.52786601e-01 2.05055133e-01
1.65338719e+00 5.60589552e-01 9.30372894e-01 1.41241193e-01
3.70481372e-01 7.44731367e-01 9.28388655e-01 3.35415095e-01
-1.08305454e-01 3.93175006e-01 3.37517381e-01 -2.83552986e-02
2.73283780e-01 3.54159951e-01 1.79149628e-01 8.09837878e-01
4.68869647e-03 -7.56695345e-02 -7.78882742e-01 3.22550625e-01
-1.36432338e+00 -4.82001543e-01 -3.54226589e-01 2.38045144e+00
8.75679553e-02 7.86585689e-01 -2.98128009e-01 8.09673145e-02
4.58074838e-01 -2.57667154e-01 -6.33954167e-01 -8.51197660e-01
1.16059348e-01 5.16668320e-01 8.95712316e-01 4.94510621e-01
-8.22118282e-01 4.13861901e-01 6.59576607e+00 9.54519451e-01
-9.79061842e-01 -9.27159488e-02 1.25554633e+00 -9.30090025e-02
-4.77320880e-01 -1.01726480e-01 -7.77814031e-01 3.00072640e-01
1.31275427e+00 3.40493545e-02 1.70861796e-01 6.04161859e-01
2.65524596e-01 -6.18583143e-01 -1.26379359e+00 8.56182873e-01
-1.37301043e-01 -1.59138858e+00 -7.19813764e-01 3.96950632e-01
9.41971421e-01 1.46229908e-01 4.03652310e-01 1.65840119e-01
-1.65165123e-02 -1.62511241e+00 2.82250851e-01 9.10506725e-01
9.98423576e-01 -1.28712821e+00 1.19545162e+00 -1.31969646e-01
-5.21761835e-01 -1.83450863e-01 -6.44114733e-01 1.35094970e-01
3.04668844e-01 1.12194407e+00 -6.84441626e-01 6.92765951e-01
7.15473533e-01 3.46234649e-01 -1.73939645e-01 6.73145175e-01
3.93033534e-01 3.21613908e-01 -2.03412533e-01 -3.44545506e-02
-6.04536347e-02 -4.27501112e-01 1.28510982e-01 8.66624773e-01
5.77727377e-01 -4.19979125e-01 -2.34784171e-01 1.03897619e+00
1.66668981e-01 -1.33514673e-01 -6.53161287e-01 -3.07920009e-01
2.48337954e-01 1.25352705e+00 -9.34548855e-01 3.99057716e-01
-7.41557539e-01 4.52032000e-01 -4.44812655e-01 3.09843775e-02
-5.81878781e-01 -2.22785875e-01 5.03867805e-01 5.52408695e-01
3.42773974e-01 -1.98916361e-01 -1.03548968e+00 -1.80522382e-01
-1.92738846e-01 -7.64795840e-01 -2.34534636e-01 -7.55782962e-01
-1.82750726e+00 2.87474126e-01 1.60852596e-02 -8.43784392e-01
-1.82879850e-01 -1.36254203e+00 -4.63350147e-01 1.01739585e+00
-1.28537035e+00 -9.00652170e-01 1.45990595e-01 -1.79667860e-01
4.42173660e-01 -3.30857575e-01 7.00210571e-01 -3.39251235e-02
-1.34075701e-01 9.51020569e-02 5.31348825e-01 -4.51384246e-01
6.67209744e-01 -1.46400166e+00 2.46327057e-01 1.65904745e-01
-2.72041619e-01 3.44101667e-01 1.02110541e+00 -7.29307115e-01
-1.39096093e+00 -8.10007036e-01 4.29517716e-01 -8.94190490e-01
7.73821294e-01 -5.68346232e-02 -5.68029404e-01 4.72099394e-01
1.71490848e-01 -2.98581988e-01 6.41791642e-01 2.64627427e-01
7.10731328e-01 -3.85959566e-01 -1.31615376e+00 2.15518638e-01
2.65414804e-01 -5.65489829e-01 2.07994934e-02 3.31340671e-01
3.32009107e-01 -2.49684274e-01 -1.29111183e+00 8.77083182e-01
5.00342309e-01 -1.23895860e+00 6.02753341e-01 -2.50983890e-02
7.08604753e-01 1.19193271e-02 -4.69440520e-01 -1.01795435e+00
-1.32396311e-01 -1.44856393e-01 3.90508264e-01 1.11651695e+00
6.27682984e-01 -5.51136255e-01 1.11555767e+00 6.01410687e-01
-5.49798071e-01 -1.32470429e+00 -9.95531619e-01 -8.17821741e-01
6.13764763e-01 -8.83051991e-01 3.22613895e-01 4.89173442e-01
-2.00243995e-01 9.16908309e-02 -4.32626903e-02 1.59103215e-01
6.36075437e-01 2.45876163e-02 4.01353300e-01 -1.41528034e+00
-1.19916864e-01 -3.54293346e-01 -6.57052696e-01 -2.66307324e-01
-1.21121950e-01 -4.15099412e-01 2.15102062e-01 -1.76947725e+00
6.64755180e-02 -5.69934607e-01 -4.86654401e-01 -5.22468090e-01
5.04570246e-01 3.80601026e-02 -2.01088458e-01 -1.76224768e-01
-1.10372692e-01 2.47482002e-01 1.19444788e+00 -2.37058237e-01
3.12605679e-01 3.25064331e-01 -7.46119201e-01 6.41487896e-01
7.43964732e-01 -1.19577080e-01 -4.52735841e-01 8.73031467e-02
4.76047695e-01 2.47760609e-01 2.31333718e-01 -1.20754826e+00
9.27632973e-02 -8.49183723e-02 1.09122074e+00 -8.29731524e-01
2.93168902e-01 -9.75431025e-01 1.89574569e-01 1.36542946e-01
8.09579939e-02 4.89738844e-02 4.16489422e-01 3.13685477e-01
-1.15617462e-01 -5.93571544e-01 9.40916240e-01 -1.10270180e-01
-3.49107563e-01 2.55218506e-01 -4.21730906e-01 -5.01177371e-01
9.75017667e-01 -4.38500375e-01 -2.08281428e-01 -4.87634063e-01
-7.50912488e-01 -9.76515934e-03 7.08824813e-01 1.82540327e-01
7.09508479e-01 -1.03977406e+00 -5.92503250e-01 1.67772546e-01
1.54718244e-02 -1.27787232e-01 5.69091976e-01 1.00005770e+00
-3.90506178e-01 5.18482983e-01 -1.24245368e-01 -7.25085318e-01
-7.32662976e-01 5.32788634e-01 5.10648906e-01 -2.89048761e-01
-3.12513947e-01 8.05366635e-01 6.04742728e-02 -2.53587991e-01
-3.43867451e-01 -2.98329145e-01 2.32003570e-01 -2.16222927e-01
5.90904392e-02 3.66327792e-01 7.06021845e-01 -8.22358951e-02
-2.50127167e-01 3.82395834e-01 -3.10281217e-01 2.65201088e-02
1.44261551e+00 -1.37214512e-01 -1.56151131e-01 1.06440091e+00
1.13678539e+00 8.78212005e-02 -1.31355739e+00 6.24261677e-01
3.58126640e-01 -1.01070419e-01 4.70918745e-01 -1.05711257e+00
-7.63298154e-01 9.54812586e-01 8.99058819e-01 3.60362560e-01
1.08523595e+00 1.44910696e-03 3.78875405e-01 -1.45587197e-03
4.66684043e-01 -1.61839414e+00 -1.19721919e-01 3.05516541e-01
9.50566947e-01 -1.22483087e+00 5.54658711e-01 -4.84846860e-01
-4.38957840e-01 1.31357992e+00 4.80569482e-01 -1.75910220e-01
7.86675453e-01 9.42712784e-01 9.12482217e-02 -5.21371245e-01
-6.34002507e-01 3.61644924e-01 7.24404603e-02 9.70260024e-01
6.73970759e-01 7.77727962e-02 -1.59031540e-01 5.53135455e-01
-3.94973755e-02 2.43039001e-02 4.89795357e-01 7.72285402e-01
-4.98566270e-01 -1.23903561e+00 -2.59449303e-01 1.08420539e+00
-5.42494535e-01 4.51960117e-02 -1.84155267e-03 1.05237472e+00
6.34232908e-02 8.93597960e-01 4.55518782e-01 -2.85203010e-01
2.73479015e-01 -2.00559688e-03 8.77370059e-01 -6.15707099e-01
-7.03220963e-02 1.47656381e-01 2.53452659e-01 -3.45272005e-01
4.01116721e-03 -8.75834882e-01 -1.26975095e+00 -3.48460823e-01
-4.95595843e-01 -1.40394270e-02 1.13486195e+00 1.11477458e+00
-9.01351422e-02 1.04631984e+00 6.12724662e-01 -7.56297886e-01
-6.87974036e-01 -9.17828202e-01 -1.21885729e+00 -2.81491041e-01
1.89011976e-01 -7.28841424e-01 -2.35100180e-01 -4.90493298e-01] | [7.047554969787598, 2.1998209953308105] |
5ec652c4-0cd5-416c-9d10-c5ba5222c936 | applications-of-artificial-intelligence-1 | 2105.15103 | null | https://arxiv.org/abs/2105.15103v1 | https://arxiv.org/pdf/2105.15103v1.pdf | Applications of Artificial Intelligence, Machine Learning and related techniques for Computer Networking Systems | This article presents a primer/overview of applications of Artificial Intelligence and Machine Learning (AI/ML) techniques to address problems in the domain of computer networking. In particular, the techniques have been used to support efficient and accurate traffic prediction, traffic classification, anomaly detection, network management, network security, network resource allocation and optimization, network scheduling algorithms, fault diagnosis and many more such applications. The article first summarizes some of the key networking concepts and a few representative machine learning techniques and algorithms. The article then presents details regarding the availability of data sets for networking applications and machine learning software and toolkits for processing these data sets. Highlights of some of the standards activities, pursued by ITU-T and ETSI, which are related to AI/ML for networking, are also presented. Finally, the article discusses a small set of representative networking problems where AI/ML techniques have been successfully applied. | ['Krishna M. Sivalingam'] | 2021-04-21 | null | null | null | null | ['traffic-classification'] | ['miscellaneous'] | [ 1.55456156e-01 -3.54129463e-01 -6.31741643e-01 -4.81007665e-01
1.20210379e-01 -1.27036765e-01 1.56419277e-01 1.25932798e-01
-1.32643104e-01 9.83106792e-01 -8.53792250e-01 -9.81617332e-01
-7.83039987e-01 -7.30762303e-01 9.54748541e-02 -4.48017299e-01
-7.71107435e-01 9.95507956e-01 3.46708000e-01 2.15782430e-02
5.25977194e-01 1.31625724e+00 -1.63365340e+00 2.66025454e-01
2.68312365e-01 1.73270655e+00 -3.89094234e-01 9.49359059e-01
-5.04418552e-01 9.77618515e-01 -8.97911906e-01 -2.54102260e-01
4.16246802e-01 -4.12453376e-02 -9.09743607e-01 2.10878596e-01
1.79753944e-01 8.10080487e-03 -4.99036223e-01 6.10716522e-01
1.59339994e-01 1.80364177e-01 6.52694583e-01 -2.07721877e+00
3.95464599e-01 2.10809946e-01 -2.15546951e-01 1.38145733e+00
9.34497565e-02 2.31605321e-02 7.44461417e-01 -2.70239115e-01
1.97496250e-01 1.29724228e+00 4.79297221e-01 4.16068822e-01
-1.08265638e+00 -8.31597328e-01 -5.80986124e-03 8.42349172e-01
-1.15262330e+00 -7.53086030e-01 8.53489995e-01 -3.28822970e-01
1.20842588e+00 5.62881529e-01 4.54685748e-01 4.71834272e-01
7.47905821e-02 6.04887843e-01 3.81099641e-01 -5.99848330e-01
2.43005335e-01 4.73417372e-01 2.31390581e-01 6.10535979e-01
2.20751479e-01 1.01840653e-01 3.26423228e-01 -4.62852955e-01
8.52931023e-01 -3.10323119e-01 4.52493519e-01 -2.39758834e-01
-7.97301233e-01 7.49523580e-01 -8.26576725e-02 3.94124448e-01
-7.41858125e-01 -2.09990010e-01 8.92857194e-01 9.73981798e-01
2.71503001e-01 4.46901619e-01 -7.33434737e-01 -3.70199144e-01
-9.51866031e-01 -5.93783520e-02 1.25179696e+00 1.07350242e+00
7.94197977e-01 8.79431725e-01 3.89977902e-01 9.66811121e-01
4.28285360e-01 5.06731123e-02 -1.19812889e-02 -1.43776464e+00
3.43141317e-01 9.65678692e-02 -4.73890126e-01 -1.08784342e+00
-6.27569258e-01 -3.58098783e-02 -6.72244012e-01 5.64813554e-01
2.46706828e-01 -3.76679242e-01 -4.97129142e-01 1.03021646e+00
-7.42610171e-02 5.90599716e-01 1.23285510e-01 4.74662669e-02
4.33236271e-01 7.28853285e-01 1.40958950e-01 -7.28323758e-01
8.21693122e-01 -9.09886777e-01 -8.08480561e-01 -8.97907242e-02
6.23009086e-01 -8.75846326e-01 -5.87202646e-02 4.00780350e-01
-1.10280573e+00 -5.43469429e-01 -5.60226381e-01 6.56337559e-01
-5.49230635e-01 -6.22851372e-01 7.78656840e-01 1.13475931e+00
-8.54608476e-01 5.79747438e-01 -5.67938626e-01 -9.57593977e-01
6.51570141e-01 8.71609151e-01 9.13521722e-02 6.36806712e-02
-9.79812860e-01 6.84329510e-01 5.70568979e-01 -4.17157263e-01
-2.84090877e-01 -7.63204753e-01 -6.08762801e-01 -8.01196024e-02
2.79464126e-01 -1.24516621e-01 1.22828066e+00 -7.35369265e-01
-1.27036524e+00 5.72899997e-01 1.04199946e-01 -7.18480229e-01
1.43331336e-02 3.21915269e-01 -1.62154317e+00 2.37619191e-01
5.39142340e-02 2.39243403e-01 6.44214392e-01 -8.30704391e-01
-1.11734676e+00 2.18247734e-02 -1.68586075e-01 -5.04979670e-01
-8.06805268e-02 6.05175793e-01 -2.50320286e-01 -6.50824845e-01
-1.38084311e-03 -4.66311783e-01 -3.32233459e-01 1.05226718e-01
-4.64044571e-01 -3.48530352e-01 1.68302691e+00 -3.20256412e-01
1.42854822e+00 -1.74707079e+00 -5.20228267e-01 1.29668605e+00
4.69476134e-02 7.62557387e-01 -8.03747252e-02 6.31761312e-01
-5.68932235e-01 3.92914504e-01 3.36504132e-01 2.47137725e-01
-3.79367590e-01 7.10425794e-01 -5.35651147e-02 2.54625175e-03
9.96716768e-02 2.52948552e-01 -6.38099730e-01 -5.22395909e-01
8.04152429e-01 -9.14144069e-02 -4.94691133e-01 -4.15847786e-02
-1.21167846e-01 5.26708901e-01 -5.72740436e-01 1.07119679e+00
4.05221522e-01 5.26301973e-02 9.31612551e-02 -3.04960549e-01
-1.84388310e-01 2.04387128e-01 -1.20350397e+00 4.50502098e-01
-4.38279301e-01 9.89865601e-01 8.36556479e-02 -1.89769351e+00
9.10711706e-01 7.14573503e-01 1.26925170e+00 -6.40155911e-01
1.43239081e-01 3.64864916e-01 2.74896771e-01 -7.27913260e-01
-2.60235608e-01 8.52854848e-02 5.84340334e-01 7.44999707e-01
6.80140108e-02 3.22839886e-01 8.72562587e-01 -9.14270058e-02
1.37998748e+00 -9.40885246e-01 4.96746480e-01 5.95727526e-02
1.25057840e+00 -1.56312615e-01 7.08491862e-01 9.21382129e-01
-7.78500199e-01 -2.35495523e-01 6.04257226e-01 -8.86085272e-01
-1.12117350e+00 -9.04680789e-01 -1.26887694e-01 1.00218487e+00
-2.41559625e-01 -8.06501210e-02 -2.90691108e-01 -5.60234308e-01
1.75738446e-02 6.78258359e-01 1.13215417e-01 9.70439985e-02
-6.92473114e-01 -6.72655523e-01 5.06147265e-01 2.64696598e-01
4.22439009e-01 -1.30508232e+00 1.98193818e-01 6.36822224e-01
1.76901355e-01 -1.80137086e+00 4.22381788e-01 2.01807898e-02
-1.24390709e+00 -1.24190474e+00 3.09738100e-01 -7.23508358e-01
1.83233589e-01 -2.41065603e-02 1.27963960e+00 5.39898515e-01
-7.79047847e-01 3.98040354e-01 -2.13135958e-01 -4.49234486e-01
-7.59170592e-01 -1.53166996e-02 5.09153545e-01 -4.96649332e-02
5.85601747e-01 -1.15370476e+00 4.22516577e-02 6.73096180e-01
-4.46235865e-01 -8.34160149e-01 6.61171615e-01 3.04288417e-01
2.77740747e-01 7.97109604e-01 7.86000192e-01 -1.08316624e+00
3.53598446e-01 -8.21400642e-01 -5.45386195e-01 1.23955114e-02
-8.24090719e-01 -4.20512885e-01 5.76319814e-01 -1.32865697e-01
-5.80716789e-01 -6.72749460e-01 -3.86687726e-01 -4.24650818e-01
-6.54844165e-01 -1.56205166e-02 -3.43665838e-01 -6.72084451e-01
5.92964709e-01 -1.09728150e-01 4.41007733e-01 -5.57619631e-01
-3.77349734e-01 9.45800841e-01 1.20276958e-01 -5.43627441e-01
8.91458750e-01 3.91170561e-01 2.01175854e-01 -1.42992246e+00
-4.89877075e-01 -6.59471869e-01 -5.16710341e-01 -3.93744409e-01
2.59073615e-01 -5.31020463e-02 -5.25783777e-01 2.80527025e-01
-8.15106809e-01 1.30064219e-01 -4.70454723e-01 5.76938987e-01
-8.44222963e-01 3.80388826e-01 -8.12167525e-01 -8.94296587e-01
-1.26696214e-01 -1.11261737e+00 -3.77720185e-02 1.56807050e-01
-3.09038460e-01 -1.41106153e+00 -2.09410623e-01 6.13461912e-01
7.69918799e-01 4.23319787e-02 1.54220510e+00 -1.32424879e+00
-4.44741875e-01 -5.08723140e-01 -3.93481374e-01 6.07658148e-01
2.74390757e-01 5.66725314e-01 -5.96363783e-01 -1.49031341e-01
-2.82210916e-01 4.36232463e-02 2.23474175e-01 3.62782687e-01
1.56769621e+00 -5.74633300e-01 -5.77902794e-01 4.54726696e-01
1.25778925e+00 1.04792178e+00 6.45415485e-01 5.07167280e-01
2.57134110e-01 5.41458726e-01 5.33567190e-01 5.25750458e-01
-3.04408461e-01 4.99598205e-01 5.98709345e-01 1.02548078e-02
8.79693404e-02 7.74253607e-01 -9.43484828e-02 5.28959155e-01
-2.26125881e-01 -5.29425681e-01 -8.82652819e-01 9.82053019e-03
-1.53377140e+00 -1.55915570e+00 -1.78866014e-01 1.67338312e+00
-6.77071437e-02 7.25710988e-01 6.83346093e-01 1.07830167e+00
1.00423384e+00 5.44217080e-02 -5.20997822e-01 -1.00504541e+00
2.90834278e-01 3.72031957e-01 5.29743314e-01 3.92110676e-01
-1.24594820e+00 8.22082102e-01 7.75486660e+00 9.42745864e-01
-8.55069101e-01 -3.08739305e-01 6.13097608e-01 3.43116492e-01
5.45224249e-01 -2.68681139e-01 -4.29601103e-01 4.82718945e-01
1.51912665e+00 -2.66707867e-01 5.34361780e-01 9.77209926e-01
6.30187154e-01 1.76677853e-01 -8.26067805e-01 8.91542375e-01
-4.45560038e-01 -1.47356260e+00 4.61191088e-01 6.31964505e-02
3.71855497e-01 2.15338379e-01 -1.22071929e-01 3.07587504e-01
-1.77623361e-01 -5.19479394e-01 -2.82067895e-01 2.67073512e-01
4.07668382e-01 -1.12557662e+00 8.46057296e-01 -2.61774600e-01
-9.77318943e-01 -6.24236107e-01 -9.77222696e-02 -2.21761093e-01
3.20349872e-01 7.02575505e-01 -8.32285047e-01 5.01014888e-01
6.30663693e-01 8.33790362e-01 -2.72773981e-01 1.65021157e+00
6.39113247e-01 5.65876663e-01 -5.91970563e-01 2.77961463e-01
1.99373811e-01 -2.21329436e-01 9.16936636e-01 1.23200226e+00
-8.71792734e-02 5.32326400e-02 4.63478655e-01 2.41443530e-01
1.90226063e-01 -5.72383357e-03 -3.46173793e-01 -9.91853997e-02
8.46590281e-01 1.13851082e+00 -6.94871008e-01 -3.94876659e-01
-8.27902615e-01 4.31437731e-01 -5.06825089e-01 5.71901441e-01
-6.00107908e-01 -9.42100823e-01 1.45276105e+00 5.06340683e-01
-3.58973676e-03 -1.17767014e-01 -1.37047037e-01 -5.09482026e-01
-4.15903181e-01 -9.03611958e-01 8.63240421e-01 -3.22396845e-01
-1.47580755e+00 7.35233903e-01 1.64927423e-01 -1.10389376e+00
-5.12135148e-01 -9.92054343e-01 -1.12597704e+00 2.02360839e-01
-1.49723721e+00 -3.52496028e-01 -4.78074187e-03 7.89931118e-01
6.64598286e-01 -1.24631655e+00 9.19167995e-01 9.28011656e-01
-1.22361743e+00 5.02988219e-01 1.76213562e-01 3.55592817e-01
1.27854481e-01 -8.97185922e-01 3.17836851e-01 6.09053016e-01
3.45238782e-02 -6.14480115e-02 6.73297524e-01 -2.59514630e-01
-7.15000331e-01 -1.03355396e+00 6.52008295e-01 2.79004604e-01
9.36454892e-01 3.90818328e-01 -7.22198129e-01 8.42372358e-01
-3.03807318e-01 2.69899905e-01 1.05351472e+00 -1.33495033e-01
1.07475102e-03 -6.89742863e-01 -1.71395671e+00 3.22014272e-01
6.84531212e-01 -1.77805021e-01 2.02333137e-01 8.05889308e-01
1.24002688e-01 1.69310406e-01 -9.84473109e-01 3.72462898e-01
2.22703621e-01 -1.08283937e+00 1.18316972e+00 -1.14278650e+00
-6.44963324e-01 -6.34333640e-02 -1.13474615e-01 -4.79249358e-01
-4.40076739e-01 -9.82135296e-01 -6.70227587e-01 1.18810821e+00
3.50018620e-01 -7.48851955e-01 1.21536684e+00 3.89925659e-01
-6.94304751e-03 -4.90148783e-01 -1.11457837e+00 -1.02802896e+00
-5.35235941e-01 -1.07239592e+00 4.25550193e-01 8.81711245e-01
-1.89890668e-01 2.21432015e-01 -1.29883394e-01 3.97033207e-02
9.25163209e-01 -4.54137564e-01 6.13200247e-01 -1.72304153e+00
2.58663803e-01 -7.49828041e-01 -1.51760674e+00 -5.58474004e-01
3.20937902e-01 -3.73316586e-01 -8.95680368e-01 -1.08614922e+00
-7.01949120e-01 -1.08070862e+00 -5.49668789e-01 3.99367273e-01
8.00518334e-01 1.53794035e-01 -3.49917322e-01 3.32967371e-01
-5.19356906e-01 -2.96622455e-01 5.36676586e-01 1.86719880e-01
-2.22243704e-02 1.10731280e+00 -3.73056859e-01 7.83607543e-01
1.46493530e+00 -2.07413003e-01 -4.96393442e-01 -4.68843477e-03
-5.24211407e-01 -1.37733415e-01 1.13439485e-01 -1.50657034e+00
4.00161564e-01 -3.16494703e-01 2.93989509e-01 -6.36802018e-01
2.45504886e-01 -1.56591010e+00 -1.39571980e-01 5.01412332e-01
5.94562329e-02 2.05768898e-01 2.32856318e-01 5.94224751e-01
-3.03793967e-01 -2.52938122e-01 1.21367919e+00 1.74714178e-02
-1.38737857e+00 7.38757193e-01 -1.28263712e+00 8.08745250e-02
1.51384699e+00 -6.33210778e-01 2.44358078e-01 -5.01980364e-01
-1.08799195e+00 2.53703117e-01 -1.24325715e-01 3.31768960e-01
4.77313042e-01 -1.17601311e+00 -3.34703684e-01 5.64898312e-01
-1.08132167e-02 -7.75968313e-01 -1.08796783e-01 8.29388678e-01
-8.97364259e-01 6.80336952e-01 -7.14655697e-01 -4.24530506e-01
-1.51124167e+00 7.12531924e-01 5.87598264e-01 -2.22650290e-01
-1.81017652e-01 4.61174220e-01 -8.45789135e-01 -1.02106459e-01
5.96221149e-01 3.07880580e-01 -3.34101111e-01 -3.86641443e-01
6.41741633e-01 1.14500535e+00 2.78670132e-01 -4.59241480e-01
-5.60249150e-01 3.37695748e-01 -4.20867801e-01 3.25631052e-01
1.14691246e+00 -4.02797163e-01 -2.68537074e-01 2.61307687e-01
1.23518479e+00 -7.05155969e-01 -2.11195707e-01 -5.03675938e-01
7.13921368e-01 -3.44385147e-01 7.42113739e-02 -3.94714266e-01
-1.65558028e+00 9.16228652e-01 7.25527465e-01 8.08825612e-01
1.37751198e+00 -2.77730793e-01 7.46664226e-01 8.45689654e-01
1.98026687e-01 -1.13830197e+00 1.63288694e-02 7.56247818e-01
2.01609917e-02 -1.27468109e+00 -1.20435089e-01 -8.63800406e-01
-5.93382940e-02 1.66421020e+00 7.64501333e-01 8.53107572e-02
1.38619208e+00 4.05949593e-01 1.02210552e-01 -1.40738159e-01
-9.05940235e-01 -2.26666406e-01 -1.71611428e-01 1.04001391e+00
1.61710307e-01 -4.03027207e-01 5.07840872e-01 -2.88069904e-01
2.14480713e-01 -2.30294272e-01 4.02371973e-01 1.16982365e+00
-7.48300195e-01 -1.50927126e+00 -3.67175788e-01 1.05946958e+00
-6.93051577e-01 -4.19016927e-02 -1.64425850e-01 9.91786242e-01
-9.69916061e-02 1.17523599e+00 6.48575842e-01 -5.86834252e-01
4.31521118e-01 1.32435396e-01 -1.05435304e-01 -4.18666422e-01
-1.32694393e-01 -4.49240953e-01 5.46333551e-01 -7.86089599e-01
-2.95340985e-01 -4.96095717e-01 -1.06226075e+00 -1.11588728e+00
-2.18507141e-01 6.18234158e-01 5.35670042e-01 1.08418453e+00
1.31112099e-01 6.66354716e-01 9.91812944e-01 -6.05577588e-01
1.83744013e-01 -6.00652754e-01 -8.66495848e-01 -4.77728657e-02
3.26740861e-01 -7.07882226e-01 -5.22589982e-01 -1.87040195e-01] | [5.080911636352539, 7.200492858886719] |
98107cbe-b51e-4511-8598-ed490f20f334 | on-recognizing-transparent-objects-in | 1606.01001 | null | http://arxiv.org/abs/1606.01001v1 | http://arxiv.org/pdf/1606.01001v1.pdf | On Recognizing Transparent Objects in Domestic Environments Using Fusion of Multiple Sensor Modalities | Current object recognition methods fail on object sets that include both
diffuse, reflective and transparent materials, although they are very common in
domestic scenarios. We show that a combination of cues from multiple sensor
modalities, including specular reflectance and unavailable depth information,
allows us to capture a larger subset of household objects by extending a state
of the art object recognition method. This leads to a significant increase in
robustness of recognition over a larger set of commonly used objects. | ['Paul Plöger', 'Frederik Hegger', 'Alexander Hagg'] | 2016-06-03 | null | null | null | null | ['transparent-objects'] | ['computer-vision'] | [ 7.68185735e-01 -4.56105977e-01 7.19570294e-02 -4.80960339e-01
-3.78552616e-01 -5.03559530e-01 4.14232790e-01 -5.83630562e-01
4.77881804e-02 5.72777569e-01 1.67643860e-01 3.86333048e-01
-1.41220748e-01 -8.86665940e-01 -2.44798571e-01 -7.77971208e-01
1.46785483e-01 1.66086644e-01 3.61937046e-01 -2.93372601e-01
-1.09209843e-01 1.04061460e+00 -2.09159851e+00 6.00885570e-01
2.94509679e-01 1.29574656e+00 1.92366511e-01 6.44399703e-01
1.94442540e-01 3.12738270e-01 -3.86277974e-01 -4.06415969e-01
5.77808917e-01 2.07614988e-01 -4.17595923e-01 4.25747395e-01
8.85280371e-01 -8.17300439e-01 -4.74058449e-01 8.17001522e-01
5.36906958e-01 1.66721955e-01 6.61329925e-01 -7.69044638e-01
-8.89932156e-01 1.44219786e-01 -4.87977177e-01 -1.57642230e-01
1.02266264e+00 3.54153588e-02 8.81340027e-01 -9.50708508e-01
4.20917630e-01 1.29595315e+00 6.63237274e-01 6.95589304e-01
-1.04411113e+00 -4.56475675e-01 1.37626594e-02 1.43208191e-01
-1.11668527e+00 -7.80575395e-01 1.15026689e+00 -6.86915368e-02
1.10017264e+00 6.67102277e-01 6.34163678e-01 1.30212688e+00
-3.76642048e-01 9.43884432e-01 1.36189783e+00 -5.66559076e-01
-8.64845365e-02 2.11317599e-01 3.07732552e-01 4.92378414e-01
7.39062667e-01 4.07166809e-01 -5.29510438e-01 -1.27015620e-01
7.97631979e-01 4.60825861e-01 -5.68801224e-01 -5.78556776e-01
-1.04863012e+00 2.24679530e-01 3.38171989e-01 6.51549101e-02
-4.06889647e-01 -1.02332570e-01 -1.70875818e-01 1.75746650e-01
2.50360578e-01 3.56800407e-01 -2.53819525e-01 3.61042209e-02
-4.70464736e-01 -1.74981430e-02 1.18215370e+00 9.54748690e-01
7.83591151e-01 1.54192656e-01 1.78913265e-01 1.04277515e+00
5.08499384e-01 1.23192286e+00 -1.14524968e-01 -6.86928213e-01
3.80332172e-01 6.52525246e-01 4.76696849e-01 -1.05501354e+00
-3.07623446e-01 1.09562807e-01 -4.18922096e-01 2.99003720e-01
3.16867560e-01 2.08496332e-01 -1.27120161e+00 1.18691254e+00
4.79906797e-01 -8.84398967e-02 1.94499910e-01 1.25184584e+00
1.24088454e+00 2.30906337e-01 -5.20476520e-01 1.34016022e-01
1.35565305e+00 -6.20862246e-01 -4.95380670e-01 -3.00138235e-01
1.25643164e-02 -9.01652217e-01 7.04341531e-01 7.34601140e-01
-8.55833948e-01 -3.91205817e-01 -1.22363174e+00 9.17815119e-02
-4.90499973e-01 -1.90171115e-02 1.10513532e+00 1.36373425e+00
-6.45787656e-01 7.28185102e-02 -6.92734838e-01 -6.80238962e-01
4.82123345e-01 5.92497528e-01 -7.05368400e-01 -7.72942781e-01
-5.74860573e-01 9.10328448e-01 -9.15847067e-03 3.90700102e-01
-4.56867754e-01 -4.02666777e-01 -8.34531844e-01 -4.05604988e-01
2.84897536e-01 -4.56962764e-01 8.91511679e-01 -9.26806152e-01
-1.82811213e+00 1.16979885e+00 -7.99513515e-03 2.07903847e-01
3.48508984e-01 -5.93131781e-01 -5.91133952e-01 5.05972803e-01
-6.46271586e-01 2.28652000e-01 9.35402513e-01 -1.63609385e+00
3.95067707e-02 -6.77168131e-01 1.44675598e-01 4.89674993e-02
-4.49743479e-01 2.58372962e-01 -3.57368439e-01 -3.92229557e-01
5.37596881e-01 -8.69317889e-01 1.90995172e-01 2.46962860e-01
-4.75727975e-01 3.45717013e-01 1.05761838e+00 -4.32158738e-01
4.88484710e-01 -1.92702079e+00 1.26467748e-02 3.58323425e-01
6.00026324e-02 3.48820150e-01 -4.04344946e-01 3.86817425e-01
7.00444728e-02 -3.06983948e-01 4.82706651e-02 -3.59602571e-01
-2.49907393e-02 2.63287604e-01 -4.44953084e-01 7.11632609e-01
1.46749228e-01 8.15038145e-01 -6.35388792e-01 3.82031836e-02
5.32053113e-01 7.20387578e-01 -2.76522785e-01 2.10414633e-01
7.89425448e-02 1.99237615e-01 -5.65560937e-01 1.57176292e+00
9.76679802e-01 6.15267865e-02 -1.39439162e-02 -2.15800777e-01
2.02328876e-01 1.60300180e-01 -1.52005076e+00 1.34425354e+00
4.07340797e-03 4.39139813e-01 3.59279811e-01 -7.54788995e-01
1.14646029e+00 2.48209998e-01 7.01984763e-01 -6.48819268e-01
7.03716204e-02 1.61333203e-01 -2.40955412e-01 -6.26278758e-01
5.19885182e-01 -1.25797734e-01 2.80013353e-01 3.65707278e-01
-1.84226736e-01 -3.30315143e-01 -3.72344643e-01 -4.48002875e-01
1.16460931e+00 1.90654710e-01 1.75140381e-01 1.63954094e-01
6.01558462e-02 -3.53615135e-01 2.76585490e-01 8.97941232e-01
-9.25528407e-02 1.08212209e+00 -5.63106716e-01 -4.63962376e-01
-7.11327434e-01 -1.33654535e+00 -4.64161456e-01 7.88483024e-01
4.14934784e-01 1.78782463e-01 -1.50049195e-01 -2.76886314e-01
3.63540053e-01 1.31694928e-01 -4.85505432e-01 -2.57874548e-01
-4.35146421e-01 -7.75297523e-01 3.61024827e-01 6.99223518e-01
5.82285225e-01 -9.67185438e-01 -6.61794305e-01 -3.02683562e-02
1.89258590e-01 -1.47003412e+00 1.10663585e-01 -1.90052707e-02
-9.09719110e-01 -1.02790821e+00 -7.43918180e-01 -2.60847926e-01
5.91637611e-01 9.16447759e-01 1.17366159e+00 1.09682418e-02
-6.91450715e-01 1.21497118e+00 -5.25349915e-01 -5.53297281e-01
7.85631090e-02 -3.90875846e-01 2.83549905e-01 3.03513944e-01
7.23797083e-01 -4.80861932e-01 -4.06960249e-01 6.00233197e-01
-7.10821807e-01 -3.43737705e-03 5.05146265e-01 4.71481442e-01
1.82292223e-01 -3.65964085e-01 -7.32045039e-04 -6.76651955e-01
-1.49975829e-02 -3.70539039e-01 -6.05823338e-01 3.22094828e-01
-1.10522337e-01 -3.15729976e-01 4.26826514e-02 -9.25441921e-01
-1.25357091e+00 1.04896925e-01 1.67913422e-01 -4.64054495e-01
-5.07329941e-01 2.97210701e-02 -3.77017885e-01 -5.54117918e-01
6.29823744e-01 5.96006773e-02 -1.98581502e-01 -7.02522516e-01
7.44715780e-02 6.55411363e-01 3.62530082e-01 -6.51202500e-01
8.41839075e-01 9.06915605e-01 8.94995406e-02 -1.16981292e+00
-5.37905276e-01 -7.18603611e-01 -7.99661934e-01 -2.91235387e-01
3.10723424e-01 -6.36288881e-01 -8.07156026e-01 7.81773984e-01
-1.04779613e+00 -9.66010913e-02 -2.98580676e-01 8.42725813e-01
-3.92224282e-01 2.99960583e-01 -3.90765518e-01 -1.31900477e+00
-1.72062125e-02 -7.18457341e-01 1.29255962e+00 2.62074441e-01
2.74296813e-02 -6.63899124e-01 -7.14887772e-03 7.09133327e-01
2.59879142e-01 2.70944208e-01 3.37341577e-01 -3.26526284e-01
-1.08237660e+00 -3.93993109e-01 -4.66734052e-01 2.91592747e-01
6.08391404e-01 4.71021011e-02 -1.36729467e+00 -2.62777954e-01
2.04217248e-02 -4.40954238e-01 1.25948703e+00 2.03060612e-01
6.47948980e-01 -6.11051247e-02 -5.15056968e-01 6.88028753e-01
1.43996513e+00 2.25659236e-01 9.46100175e-01 2.96069514e-02
8.95835280e-01 7.36762345e-01 3.74997258e-01 4.02171105e-01
1.87810808e-01 8.00668657e-01 5.57520628e-01 -1.85977727e-01
-3.16279382e-01 3.40968758e-01 4.37767118e-01 4.00088966e-01
-5.43578267e-01 -3.34007800e-01 -9.77024317e-01 4.17058319e-01
-1.50464821e+00 -1.15510046e+00 9.08183604e-02 2.28790617e+00
5.76720774e-01 -2.54760742e-01 6.29631877e-02 2.85664529e-01
5.32325506e-01 8.40804279e-02 -5.57227910e-01 -7.69527778e-02
-5.38332045e-01 2.65117675e-01 4.45718735e-01 1.75075114e-01
-1.01442897e+00 5.19121826e-01 8.83228683e+00 1.83112711e-01
-1.15106928e+00 -3.30083191e-01 -1.81963563e-01 -2.44277999e-01
-4.53261733e-01 -4.98986512e-01 -9.69492018e-01 6.38245493e-02
4.78653044e-01 6.51914895e-01 5.28286576e-01 7.81466424e-01
-2.76656628e-01 -3.53353918e-01 -1.48428988e+00 1.27530348e+00
6.75746977e-01 -8.91185224e-01 -7.16030151e-02 1.46486968e-01
7.20603585e-01 3.14203906e-03 1.44800276e-01 -1.41296059e-01
3.47309738e-01 -1.06763256e+00 5.17621636e-01 8.12820613e-01
5.56416988e-01 8.11565965e-02 4.53372657e-01 -9.18259695e-02
-9.87296879e-01 -1.19423494e-01 -4.34376270e-01 -1.82578132e-01
-1.36741981e-01 7.64546871e-01 -5.96299231e-01 4.78927135e-01
8.65458846e-01 6.73461735e-01 -3.18599492e-01 1.19477367e+00
9.39157829e-02 4.25154656e-01 -9.38513339e-01 -1.38319835e-01
-3.75057727e-01 -1.42668426e-01 7.42904246e-01 1.18968141e+00
1.30303830e-01 4.45199668e-01 1.59196302e-01 8.58249724e-01
-6.32116497e-02 -3.21114182e-01 -1.02850831e+00 1.18343838e-01
2.85523593e-01 1.15203893e+00 -5.49565375e-01 2.19978672e-02
-7.06706047e-01 9.32811558e-01 1.66392967e-01 6.39829516e-01
-3.84611666e-01 -1.23375043e-01 8.34529102e-01 -9.60602015e-02
5.51062524e-01 -5.59559762e-01 -2.96241969e-01 -1.43832088e+00
3.96230072e-01 -7.79881299e-01 7.18209743e-02 -1.04256713e+00
-1.51801693e+00 4.12839614e-02 5.27082868e-02 -1.17339599e+00
1.79288402e-01 -1.10490346e+00 -3.16271007e-01 6.83282316e-01
-1.77828503e+00 -1.45992601e+00 -6.73475981e-01 5.84344625e-01
1.78976357e-01 -1.45727426e-01 1.03242218e+00 2.31234610e-01
-3.14192653e-01 2.52250254e-01 2.12224066e-01 1.46808792e-02
5.44776261e-01 -7.98008978e-01 -1.22643299e-01 5.73850334e-01
2.30388954e-01 5.68151116e-01 6.44820929e-01 -4.60054785e-01
-2.17672324e+00 -5.25882363e-01 -6.26052245e-02 -8.04660022e-01
1.72612101e-01 -5.10192871e-01 -7.81530619e-01 5.41590810e-01
-3.95743638e-01 2.39588842e-01 7.42694020e-01 4.11018580e-01
-9.43569720e-01 -3.90750796e-01 -1.34145343e+00 3.97906572e-01
1.23285985e+00 -7.39107430e-01 -7.52884924e-01 1.00001357e-01
1.40398636e-01 -3.88722867e-01 -8.45766902e-01 7.71079779e-01
1.17765236e+00 -1.00025415e+00 1.53532076e+00 -3.96731108e-01
1.27764657e-01 -2.56138723e-02 -8.52310538e-01 -8.38729441e-01
-2.59196311e-01 -4.42855507e-01 -5.49909949e-01 1.19944775e+00
-1.07747182e-01 -9.06415641e-01 7.67888546e-01 1.06122983e+00
6.21189550e-02 -4.06923264e-01 -8.32371831e-01 -8.88011694e-01
-5.30530393e-01 -5.83178222e-01 4.31434155e-01 6.41781390e-01
-2.81602263e-01 -1.32573336e-01 -4.09171671e-01 3.14385235e-01
7.79919624e-01 5.30343056e-01 6.66093349e-01 -1.57230175e+00
-1.15769520e-01 -9.36274454e-02 -6.70924485e-01 -1.20283723e+00
-5.22965975e-02 -4.25038576e-01 1.24908425e-01 -1.46323526e+00
3.71617973e-01 -7.40404248e-01 -4.17257905e-01 4.80553389e-01
9.81096700e-02 8.90028358e-01 8.78023803e-02 2.78615326e-01
-3.86163831e-01 4.55752432e-01 1.22746384e+00 -4.03110921e-01
-8.26150253e-02 -2.86053028e-02 -6.65615797e-01 7.97065914e-01
2.73831099e-01 -1.54612273e-01 -3.73482183e-02 -6.68815613e-01
5.08344211e-02 -4.62142140e-01 6.06041849e-01 -1.02763915e+00
-1.38356969e-01 -4.71587032e-01 7.42724776e-01 -4.43420023e-01
1.16602111e+00 -1.20847321e+00 1.14537105e-01 3.41563299e-02
1.80056304e-01 -6.07351542e-01 2.78888166e-01 5.82436979e-01
1.35583818e-01 4.43388103e-03 7.01391697e-01 -1.30036086e-01
-7.76332676e-01 1.80859730e-01 -2.32172012e-02 -5.35088062e-01
8.47134590e-01 -1.13615167e+00 -5.05027533e-01 -1.66073203e-01
-4.09780473e-01 -4.33492541e-01 6.17397964e-01 6.82044744e-01
9.09596324e-01 -1.41611612e+00 -5.14777780e-01 4.62471336e-01
2.22827777e-01 -3.22746456e-01 -1.10085510e-01 6.60639763e-01
-1.38805389e-01 3.30189854e-01 -4.51210618e-01 -8.05193603e-01
-1.45083582e+00 2.19022945e-01 2.28328839e-01 3.91498178e-01
-6.73169255e-01 8.04896772e-01 1.76032156e-01 -2.93671012e-01
8.96358266e-02 -4.11575735e-01 -1.19828016e-01 -5.72964661e-02
5.28841972e-01 5.27903438e-01 1.45299599e-01 -7.39183128e-01
-6.75738752e-01 9.30287361e-01 1.60998031e-01 1.43723652e-01
1.51904345e+00 -1.07545450e-01 3.07837594e-02 6.61388755e-01
8.40401530e-01 2.33287856e-01 -1.25655615e+00 -5.37533283e-01
-5.02218187e-01 -9.69696224e-01 -3.39810513e-02 -8.34443390e-01
-1.02770948e+00 5.96978366e-01 7.59017348e-01 2.75681734e-01
1.22431183e+00 6.23762570e-02 6.03912711e-01 9.73292291e-01
8.48159850e-01 -9.42272961e-01 3.13699245e-01 3.90833706e-01
1.04091442e+00 -1.62376451e+00 4.05399531e-01 -9.23744023e-01
-1.64394081e-01 1.25844288e+00 4.32547331e-01 1.51529051e-02
6.77489042e-01 5.77260673e-01 -2.11990681e-02 -2.22665027e-01
-3.93493235e-01 -5.02807140e-01 6.13445938e-01 1.09041381e+00
1.68046847e-01 1.03893831e-01 4.40962970e-01 4.14688796e-01
1.39751971e-01 -1.55134633e-01 2.63905704e-01 1.18377805e+00
-4.46077943e-01 -8.01140070e-01 -9.08444524e-01 4.22632635e-01
-2.66001403e-01 1.46898374e-01 -5.45368314e-01 5.88008881e-01
-1.90331101e-01 9.83421266e-01 2.72582285e-02 -2.43631154e-01
4.15107191e-01 -1.36989430e-01 1.26251101e+00 -8.17436576e-01
-2.25547343e-01 -2.06603870e-01 2.61825234e-01 -7.49565065e-01
-9.00314987e-01 -7.15650260e-01 -7.27258801e-01 -2.24518970e-01
-7.53721178e-01 -8.08849573e-01 6.80887163e-01 9.60805416e-01
1.16590023e-01 3.66336554e-02 5.23714960e-01 -1.35241985e+00
-1.88313752e-01 -8.07354927e-01 -7.71227717e-01 7.73692310e-01
5.49155533e-01 -1.01211536e+00 -2.02358499e-01 1.56847939e-01] | [9.696157455444336, -2.790952682495117] |
a14eee39-9744-4de3-80a6-53170534a826 | learning-illuminant-estimation-from-object | 1805.09264 | null | http://arxiv.org/abs/1805.09264v1 | http://arxiv.org/pdf/1805.09264v1.pdf | Learning Illuminant Estimation from Object Recognition | In this paper we present a deep learning method to estimate the illuminant of
an image. Our model is not trained with illuminant annotations, but with the
objective of improving performance on an auxiliary task such as object
recognition. To the best of our knowledge, this is the first example of a deep
learning architecture for illuminant estimation that is trained without ground
truth illuminants. We evaluate our solution on standard datasets for color
constancy, and compare it with state of the art methods. Our proposal is shown
to outperform most deep learning methods in a cross-dataset evaluation setup,
and to present competitive results in a comparison with parametric solutions. | ['Joost Van de Weijer', 'Raimondo Schettini', 'Marco Buzzelli'] | 2018-05-23 | null | null | null | null | ['color-constancy'] | ['computer-vision'] | [ 1.75243855e-01 -3.11901689e-01 2.79449880e-01 -5.69839597e-01
-6.36730433e-01 -6.53343797e-01 6.60546780e-01 -2.42917687e-01
-5.16819239e-01 7.02914596e-01 -1.36860758e-01 -1.68079048e-01
2.61765450e-01 -4.37364072e-01 -9.29874778e-01 -8.18469286e-01
1.03782669e-01 4.81610119e-01 -1.80930406e-01 1.14852466e-01
3.22902113e-01 7.01412916e-01 -1.38740730e+00 2.30683595e-01
5.16640365e-01 1.30280089e+00 -8.15595388e-02 9.01705325e-01
-8.44376758e-02 6.82463348e-01 -6.60273194e-01 -4.57941860e-01
7.03590512e-01 -2.48607963e-01 -7.13129342e-01 9.01866257e-02
1.07861257e+00 -5.50612807e-01 -1.65450171e-01 6.28193974e-01
7.43346095e-01 -7.34595209e-02 7.79165149e-01 -1.19981956e+00
-6.66192889e-01 2.68937908e-02 -6.02556407e-01 2.37008221e-02
2.13179290e-02 2.19342366e-01 9.45010722e-01 -7.75798142e-01
4.18042570e-01 8.76544237e-01 7.68110752e-01 2.96359658e-01
-1.22077131e+00 -1.43544406e-01 2.97914110e-02 2.02063203e-01
-1.13121068e+00 -4.45203006e-01 6.94073260e-01 -4.39332008e-01
1.05894756e+00 1.89622492e-01 2.76443928e-01 1.21520519e+00
-1.34524286e-01 8.96716654e-01 1.45459759e+00 -7.36904740e-01
2.52814233e-01 9.70240086e-02 -1.29415661e-01 6.67749643e-01
1.21280029e-01 2.75704205e-01 -4.57366526e-01 3.63341808e-01
6.59762621e-01 -6.70194983e-01 -3.52375925e-01 -4.65156972e-01
-1.08900952e+00 5.00909746e-01 8.09316516e-01 -1.51406601e-02
-1.72348946e-01 6.00643039e-01 2.77680695e-01 1.06953792e-01
8.47581506e-01 6.04393780e-01 -8.30723643e-01 -6.29270896e-02
-6.68456316e-01 -2.82911919e-02 7.71832347e-01 7.59832203e-01
9.33386564e-01 1.43194452e-01 -1.84260905e-01 7.20092893e-01
4.59765941e-02 6.77231908e-01 1.45793810e-01 -1.07977223e+00
1.29493818e-01 1.25898182e-01 4.52365547e-01 -3.34921360e-01
-6.61104381e-01 -6.45328701e-01 -4.36964631e-01 7.03357160e-01
7.59982109e-01 -3.54065001e-01 -1.03732276e+00 1.34685349e+00
-7.07278326e-02 3.93131196e-01 -2.25443423e-01 1.27077949e+00
9.78482008e-01 6.02216840e-01 -2.85135746e-01 2.52277702e-01
8.91994119e-01 -1.16536736e+00 -2.88078129e-01 -2.64335632e-01
1.06879726e-01 -1.22318101e+00 1.15671444e+00 8.94316316e-01
-7.50114918e-01 -5.51442266e-01 -1.00690794e+00 -5.05799890e-01
-4.58496183e-01 6.04996681e-01 1.08973801e+00 1.10944915e+00
-1.15119267e+00 4.52940851e-01 -4.65146035e-01 -2.46615738e-01
3.54910105e-01 -7.08196908e-02 -1.13522023e-01 -1.31964102e-01
-7.47216761e-01 1.06881142e+00 1.04913682e-01 2.79951781e-01
-9.15026426e-01 -8.06903243e-01 -7.66516268e-01 -2.29424223e-01
1.57346189e-01 -6.12404704e-01 1.49582350e+00 -1.31633329e+00
-1.88117123e+00 9.58465159e-01 -1.64299935e-01 -3.93851578e-01
6.71417415e-01 -4.96622652e-01 -6.31052535e-03 -1.02119192e-01
-3.84141684e-01 7.80620098e-01 7.95097113e-01 -1.71488035e+00
-2.39293069e-01 -6.47411570e-02 3.36948514e-01 9.86684859e-02
-2.46602111e-02 -1.85113937e-01 -6.34417355e-01 -1.08165443e-01
-2.58441687e-01 -8.29906940e-01 -9.65079889e-02 3.33952487e-01
-5.38902998e-01 -2.20084339e-02 3.97970021e-01 -4.80664223e-01
2.76514530e-01 -1.82715845e+00 -2.35316724e-01 -1.18739530e-02
-5.20369299e-02 2.07317755e-01 -4.67042089e-01 1.25522792e-01
-8.52454752e-02 -1.85109541e-01 -2.09913641e-01 -8.22351038e-01
2.34753624e-01 9.66126919e-02 -3.16524595e-01 8.47542644e-01
3.38045806e-01 9.61887002e-01 -8.63795280e-01 -3.50657776e-02
6.68626785e-01 6.89753532e-01 -2.65565068e-01 3.29763889e-01
-4.29269820e-01 3.63927782e-01 3.76058608e-01 5.78761399e-01
1.02260327e+00 -5.59775755e-02 -9.36739892e-02 -6.40749931e-01
-4.49819595e-01 2.17897803e-01 -9.15898740e-01 1.92248225e+00
-9.05202389e-01 1.35383284e+00 -2.77244568e-01 -7.63769150e-01
9.06229496e-01 4.19281460e-02 4.53399628e-01 -9.32545662e-01
2.41720051e-01 2.28910238e-01 -4.83522192e-02 -3.91440123e-01
3.78629923e-01 8.35721642e-02 4.78153914e-01 3.91521394e-01
1.07408762e-01 -5.56005478e-01 6.03337958e-03 -1.08998805e-01
5.80559969e-01 7.57014751e-01 1.53909266e-01 -1.17761478e-01
1.99211240e-01 -2.85966694e-01 -7.72598162e-02 7.15658665e-01
-9.09002945e-02 1.03310859e+00 3.77942920e-01 -7.48888016e-01
-1.27346468e+00 -1.19615388e+00 -5.34143090e-01 1.03889000e+00
1.96084052e-01 7.48632327e-02 -6.29513144e-01 -3.98897171e-01
-3.58186550e-02 7.47034550e-01 -8.04498434e-01 5.28983831e-01
-2.70192981e-01 -8.47290516e-01 3.95101190e-01 5.70233345e-01
5.86280286e-01 -8.73433530e-01 -7.56671369e-01 -2.25601077e-01
-7.84747824e-02 -1.48657203e+00 -9.64349732e-02 4.09000307e-01
-3.86793345e-01 -1.31312704e+00 -7.14482069e-01 -4.32391554e-01
4.36569989e-01 1.92902327e-01 1.59104800e+00 -2.93985844e-01
-7.24019289e-01 5.54538012e-01 -1.36763722e-01 -9.29305732e-01
7.45594501e-02 -6.85768947e-02 -4.67157066e-01 -4.25627567e-02
4.20284957e-01 -1.08693421e-01 -1.05110884e+00 2.34520808e-01
-7.88407922e-01 7.31113926e-02 6.72837913e-01 6.58351183e-01
3.91352475e-01 -2.47414246e-01 -4.03635763e-02 -7.83206463e-01
1.30261660e-01 -8.12197998e-02 -1.00809407e+00 5.85228465e-02
-4.42214787e-01 1.76520810e-01 5.39346159e-01 6.63328171e-02
-1.35329866e+00 4.00980890e-01 -1.85844660e-01 -2.53537476e-01
-5.54981291e-01 3.70267220e-02 8.19276944e-02 -4.39794779e-01
1.13669586e+00 -1.68814331e-01 -3.51144612e-01 -5.85669696e-01
6.82288826e-01 3.48888904e-01 6.49459422e-01 -4.06856298e-01
6.59356177e-01 7.66771317e-01 6.22089326e-01 -6.68836892e-01
-1.01409149e+00 -7.65899897e-01 -8.72563720e-01 -2.95817018e-01
5.59523702e-01 -8.96667242e-01 -7.50264287e-01 7.30458736e-01
-1.31559050e+00 -8.61904562e-01 8.21683481e-02 4.87549692e-01
-6.48190260e-01 2.64961988e-01 -4.63234276e-01 -7.81119525e-01
-1.06588960e-01 -8.97908092e-01 1.38363147e+00 1.01281144e-01
3.34522814e-01 -1.12932241e+00 3.07040632e-01 1.86548576e-01
5.01305580e-01 4.73247111e-01 1.86543226e-01 9.43923816e-02
-8.02273691e-01 -1.77522600e-01 -8.65866482e-01 6.18884921e-01
2.95673192e-01 1.58797681e-01 -1.87299931e+00 -4.56165820e-02
-3.04484427e-01 -6.14985943e-01 1.32890248e+00 6.99175358e-01
1.47021377e+00 1.63490534e-01 1.08857319e-01 9.48538065e-01
1.83900225e+00 -3.17449033e-01 9.61178720e-01 5.49870193e-01
8.69848490e-01 3.75950068e-01 3.46646488e-01 4.63050455e-01
1.11947618e-01 6.88626885e-01 1.07617807e+00 -6.54774785e-01
-5.06373227e-01 3.48260403e-01 -1.33337468e-01 -2.18618661e-01
-3.41437399e-01 -4.17490065e-01 -7.81538904e-01 3.94493341e-01
-1.52328205e+00 -6.77837670e-01 -3.81029308e-01 2.20142865e+00
5.38710535e-01 -1.54824004e-01 1.41887426e-01 7.76621103e-02
2.54554212e-01 1.92610502e-01 -5.63831210e-01 -5.94569683e-01
-3.22330713e-01 5.40989399e-01 8.96648645e-01 7.45233893e-01
-1.30517256e+00 7.03952610e-01 7.26291704e+00 7.70386979e-02
-1.44658971e+00 8.43107700e-02 7.99746692e-01 -2.40722775e-01
-2.76003331e-01 -2.42027700e-01 -2.37053856e-01 7.11251646e-02
5.00373840e-01 6.04466796e-01 7.34912276e-01 7.22823739e-01
7.17117861e-02 -3.61960232e-01 -1.33265424e+00 1.25332558e+00
5.64286649e-01 -9.64567602e-01 -5.68756580e-01 -1.50587380e-01
9.86638784e-01 4.84311819e-01 2.98757374e-01 -6.64737523e-02
1.95301354e-01 -1.15512192e+00 7.55592585e-01 6.12327695e-01
8.26663256e-01 -5.35558820e-01 7.76968122e-01 -2.26302594e-01
-6.75083518e-01 2.95507848e-01 -6.25356793e-01 -1.35120898e-01
-2.09481046e-02 7.64811814e-01 -1.10898685e+00 5.12943089e-01
5.28206110e-01 5.98125637e-01 -6.74624920e-01 1.64266288e+00
-4.40846264e-01 5.46258271e-01 -2.32956707e-01 -9.16339606e-02
3.12194139e-01 4.25500646e-02 1.20735124e-01 1.52263951e+00
2.17378348e-01 -4.93649334e-01 -9.26498175e-02 9.04928327e-01
-1.74487919e-01 3.48362923e-02 -5.07394910e-01 3.32479417e-01
-1.32482439e-01 1.48386526e+00 -5.48732877e-01 -2.14020237e-02
-5.17941117e-01 1.21456933e+00 4.01789188e-01 6.99641705e-01
-8.78490329e-01 -4.35882688e-01 5.64230561e-01 -1.67895406e-01
2.37829387e-01 -3.27671766e-01 -6.50789857e-01 -1.02493131e+00
1.08981235e-02 -3.08580965e-01 1.90205220e-02 -1.34504044e+00
-1.37776518e+00 5.29113054e-01 -1.49860427e-01 -1.03649950e+00
-8.78708437e-02 -1.60144198e+00 -5.77284515e-01 1.22663093e+00
-2.25270128e+00 -1.16340590e+00 -8.24558258e-01 4.62203681e-01
3.58460546e-01 5.27836941e-03 7.77140915e-01 1.02110028e-01
-2.24803358e-01 3.91176701e-01 2.56028414e-01 7.40145594e-02
1.05289817e+00 -1.82685649e+00 6.08375609e-01 9.90835786e-01
4.54415947e-01 2.24784881e-01 9.69901741e-01 6.72216341e-02
-1.33695972e+00 -1.03910542e+00 3.30549955e-01 -5.52804828e-01
4.84166592e-01 -5.69284916e-01 -4.03095812e-01 5.54638565e-01
5.82470894e-01 3.75736117e-01 6.45245433e-01 4.13050294e-01
-8.77179205e-01 -4.50427890e-01 -9.10713851e-01 2.88330883e-01
7.37505555e-01 -4.64011610e-01 8.52912962e-02 6.33035660e-01
3.96808743e-01 -7.94198036e-01 -2.76529849e-01 2.29188263e-01
7.68363595e-01 -1.47370601e+00 1.14171016e+00 -2.69384921e-01
6.26283646e-01 -3.21370423e-01 -1.61685228e-01 -1.58713901e+00
-2.71411479e-01 -3.78131926e-01 1.52939484e-01 8.53515267e-01
4.32418048e-01 -4.55691516e-01 7.74710894e-01 4.20137405e-01
-4.66243237e-01 -3.90056670e-01 -7.07615256e-01 -6.87737703e-01
3.89558300e-02 -6.13530278e-01 4.26049531e-01 4.98149306e-01
-6.89420342e-01 1.21374980e-01 -6.51348352e-01 3.60593140e-01
8.76359403e-01 3.95490557e-01 1.07658219e+00 -1.01591361e+00
-4.24831152e-01 -6.74492419e-01 -3.04589033e-01 -8.80029619e-01
1.11107573e-01 -4.51901495e-01 4.64401960e-01 -1.81995595e+00
9.46978778e-02 -2.98742175e-01 -4.19604659e-01 3.59351367e-01
-7.49019757e-02 8.25137854e-01 -1.16621941e-01 -2.64434218e-01
-5.25559545e-01 3.07652235e-01 1.14601064e+00 -2.51484692e-01
1.96707398e-02 -1.24779150e-01 -4.55700487e-01 3.61490399e-01
1.29344153e+00 1.49498284e-01 -4.47687387e-01 -9.07864213e-01
3.61046314e-01 -7.71099806e-01 6.62010908e-01 -9.46308434e-01
-2.17237592e-01 -1.48326635e-01 8.27959538e-01 -4.39215660e-01
4.96657610e-01 -6.78110421e-01 -3.69775951e-01 4.56102714e-02
-1.57999903e-01 -5.19722223e-01 5.05176425e-01 7.43898451e-02
1.81550667e-01 -1.68389916e-01 7.46563554e-01 -6.58977479e-02
-9.64985192e-01 2.83325195e-01 3.10817733e-02 -2.33264461e-01
5.74541509e-01 1.39297470e-02 -8.22623432e-01 -2.99746871e-01
-1.19438134e-01 -3.52577627e-01 3.68282139e-01 2.26453304e-01
3.45946729e-01 -1.14821827e+00 -7.18939185e-01 -1.11621976e-01
4.28423703e-01 -1.12090372e-01 -7.09377229e-02 4.83048320e-01
-9.44508493e-01 5.70863843e-01 -3.39034826e-01 -7.91870952e-01
-9.69680905e-01 5.69560409e-01 6.92979455e-01 2.09535956e-01
-2.83792078e-01 1.01270032e+00 4.64346766e-01 -2.59749889e-01
3.39995384e-01 -5.12352109e-01 1.97496563e-01 -4.38386858e-01
4.95939076e-01 2.06225514e-01 5.10158539e-01 -2.37616092e-01
-1.94120437e-01 5.32846451e-01 3.87667179e-01 -8.53703469e-02
1.27668977e+00 -9.63250399e-02 -1.52022302e-01 6.22611403e-01
1.31502855e+00 1.75488051e-02 -1.68671346e+00 -1.76610962e-01
-5.17248154e-01 -8.84207129e-01 4.13356781e-01 -1.40075362e+00
-1.15346479e+00 1.06485105e+00 8.22511792e-01 -3.12927738e-02
1.32715273e+00 -1.90443903e-01 3.27351391e-01 7.30674803e-01
1.60583168e-01 -9.50028419e-01 5.04465222e-01 4.25314605e-01
1.04932117e+00 -1.71028447e+00 3.53436805e-02 -3.03569973e-01
-4.68364745e-01 1.39472985e+00 7.39527941e-01 -1.87283009e-01
3.63728940e-01 4.95148569e-01 5.60957432e-01 4.66647111e-02
-4.66359824e-01 -6.11887813e-01 6.78656578e-01 9.50878382e-01
9.03057694e-01 -4.76379842e-02 1.37762398e-01 -2.73253202e-01
-1.04853183e-01 -1.03894979e-01 4.60210770e-01 4.12425756e-01
-4.16806877e-01 -9.00732696e-01 -3.85375023e-01 3.92640233e-02
-2.07366019e-01 -3.38144571e-01 -5.69073081e-01 7.18954563e-01
2.25381896e-01 9.02022958e-01 2.44498909e-01 1.54146671e-01
1.37245208e-01 -1.25313073e-01 1.07912517e+00 -2.67594814e-01
-1.30985454e-01 1.42983392e-01 2.13830918e-01 -6.09392464e-01
-6.82713926e-01 -4.00730520e-01 -5.43041646e-01 -1.66033372e-01
-1.28978148e-01 -4.77992982e-01 1.19552374e+00 6.33504629e-01
-2.67897427e-01 5.81281483e-01 7.68866301e-01 -1.20195711e+00
1.20295517e-01 -9.36654806e-01 -4.69459802e-01 5.40487766e-01
6.00643098e-01 -5.17134249e-01 -2.98298955e-01 -2.23012380e-02] | [10.270445823669434, -2.560140609741211] |
93268256-8a0c-42a5-98e3-4900bfbfc04c | task-adaptive-few-shot-node-classification | 2206.11972 | null | https://arxiv.org/abs/2206.11972v1 | https://arxiv.org/pdf/2206.11972v1.pdf | Task-Adaptive Few-shot Node Classification | Node classification is of great importance among various graph mining tasks. In practice, real-world graphs generally follow the long-tail distribution, where a large number of classes only consist of limited labeled nodes. Although Graph Neural Networks (GNNs) have achieved significant improvements in node classification, their performance decreases substantially in such a few-shot scenario. The main reason can be attributed to the vast generalization gap between meta-training and meta-test due to the task variance caused by different node/class distributions in meta-tasks (i.e., node-level and class-level variance). Therefore, to effectively alleviate the impact of task variance, we propose a task-adaptive node classification framework under the few-shot learning setting. Specifically, we first accumulate meta-knowledge across classes with abundant labeled nodes. Then we transfer such knowledge to the classes with limited labeled nodes via our proposed task-adaptive modules. In particular, to accommodate the different node/class distributions among meta-tasks, we propose three essential modules to perform \emph{node-level}, \emph{class-level}, and \emph{task-level} adaptations in each meta-task, respectively. In this way, our framework can conduct adaptations to different meta-tasks and thus advance the model generalization performance on meta-test tasks. Extensive experiments on four prevalent node classification datasets demonstrate the superiority of our framework over the state-of-the-art baselines. Our code is provided at https://github.com/SongW-SW/TENT. | ['Jundong Li', 'Chen Chen', 'Chuxu Zhang', 'Kaize Ding', 'Song Wang'] | 2022-06-23 | null | null | null | null | ['graph-mining'] | ['graphs'] | [ 2.35375583e-01 8.71917233e-02 -5.25269568e-01 -2.94079095e-01
-2.90032327e-01 -1.36269927e-01 5.39291501e-01 3.17863792e-01
-2.89639562e-01 5.84436655e-01 -8.14812034e-02 -2.20265895e-01
-2.21841335e-01 -1.20088184e+00 -3.79998982e-01 -7.54131913e-01
7.90832713e-02 2.78959543e-01 4.29556519e-01 -2.55334824e-01
-6.91227764e-02 -1.76805809e-01 -1.47316885e+00 3.69141363e-02
1.23178804e+00 7.41247952e-01 5.05009174e-01 4.31976527e-01
-4.15694654e-01 6.11859977e-01 -5.72231174e-01 -5.90317130e-01
-9.39144492e-02 -3.57382357e-01 -6.40142858e-01 8.74990076e-02
1.95386544e-01 -3.86610292e-02 -6.73271358e-01 1.46186781e+00
5.31747878e-01 3.59396845e-01 7.27348328e-01 -1.72689652e+00
-7.47165382e-01 8.66202593e-01 -8.81248832e-01 3.80419254e-01
-2.94591755e-01 1.50343195e-01 1.10512245e+00 -6.89036608e-01
5.32370210e-01 1.03913260e+00 5.31809032e-01 6.56256199e-01
-1.11791039e+00 -8.96504700e-01 4.29217249e-01 4.50664520e-01
-1.41136980e+00 -3.10969979e-01 1.11474109e+00 -2.06947103e-01
5.82092762e-01 -1.87550280e-02 4.28740978e-01 1.43189251e+00
2.23345414e-01 7.63070345e-01 6.12343848e-01 -1.55525625e-01
2.51012057e-01 2.19022520e-02 3.65708083e-01 5.99105179e-01
4.75136638e-01 -4.85657662e-01 -1.70848712e-01 -1.35645717e-01
5.11120260e-01 5.52472055e-01 -2.71885991e-01 -3.46421361e-01
-9.95005071e-01 6.92091644e-01 5.81491232e-01 4.35073376e-01
-2.54091352e-01 1.82957530e-01 9.02014434e-01 2.91384637e-01
8.02241325e-01 -1.30301356e-01 -5.00142515e-01 1.79171711e-01
-6.40235722e-01 -1.05021790e-01 6.20842874e-01 1.16993058e+00
9.50110257e-01 8.08586478e-02 -5.82198858e-01 1.26101613e+00
2.50465125e-01 1.94186673e-01 6.72275722e-01 -4.35186774e-01
7.45356977e-01 7.58533180e-01 -5.34609854e-01 -1.07566059e+00
-5.39675534e-01 -1.14713228e+00 -1.36120772e+00 -2.08853319e-01
1.33232072e-01 -1.85483038e-01 -1.14421034e+00 2.12022185e+00
4.77651596e-01 6.21841192e-01 -2.14367673e-01 4.98160183e-01
1.19046772e+00 5.08107007e-01 5.41802585e-01 -3.80224228e-01
1.46935916e+00 -1.16065073e+00 -6.77275538e-01 -4.64280397e-01
8.85672808e-01 -2.05160111e-01 1.18945551e+00 -2.32459977e-01
-5.73924422e-01 -6.52528703e-01 -9.66451764e-01 1.68749303e-01
-3.83164227e-01 -2.11541504e-01 5.82407594e-01 5.54289401e-01
-6.29923403e-01 5.29165864e-01 -4.88022178e-01 -3.23472530e-01
6.92913353e-01 -7.30060861e-02 -2.19031870e-01 -1.63418218e-01
-1.35966754e+00 2.95167804e-01 6.09101653e-01 -4.39628735e-02
-9.33541417e-01 -9.19388473e-01 -8.70868027e-01 4.14601982e-01
8.51014555e-01 -7.92760909e-01 1.10218465e+00 -5.33875763e-01
-8.16766262e-01 6.83171153e-01 -1.00492746e-01 4.54366095e-02
3.37643385e-01 4.63476330e-01 -5.33801556e-01 -2.12811381e-01
2.64213204e-01 3.42829347e-01 9.47271526e-01 -1.14952087e+00
-7.65603423e-01 -4.51895684e-01 1.12602524e-01 1.35164782e-01
-9.61151719e-01 -6.76630676e-01 -7.70477951e-01 -8.26489449e-01
5.20875342e-02 -7.86378205e-01 -1.74605399e-01 -2.72945523e-01
-4.99452531e-01 -6.57990277e-01 5.98028421e-01 -1.40780032e-01
1.57740724e+00 -2.08353209e+00 5.19541316e-02 9.99091938e-03
7.39221573e-01 2.83284128e-01 -5.52365601e-01 4.37650561e-01
-6.60634274e-03 3.50516886e-01 -1.62727281e-01 -4.21934634e-01
-1.65010974e-01 1.65854752e-01 1.14031740e-01 1.75652623e-01
-2.37624392e-01 1.02999520e+00 -1.20948076e+00 -6.74234271e-01
-1.34587390e-02 1.92578241e-01 -1.15113586e-01 2.02594567e-02
-2.60284513e-01 3.27726528e-02 -7.16766477e-01 7.73000181e-01
8.41010988e-01 -6.10063910e-01 3.13537329e-01 -3.19056958e-01
5.23876965e-01 3.80747765e-02 -9.49905157e-01 1.80621207e+00
-5.74915886e-01 2.63010889e-01 1.06561579e-01 -1.20773399e+00
8.55464876e-01 1.45337880e-01 3.77661586e-01 -6.05987728e-01
1.65740401e-01 7.61172101e-02 3.39025497e-01 -4.65249211e-01
2.68885255e-01 -5.25192544e-03 8.51202160e-02 4.28729057e-01
3.25292200e-01 3.63493145e-01 4.28967297e-01 4.78591293e-01
1.41218066e+00 -2.40081519e-01 3.10665518e-01 -2.59575099e-01
4.52234954e-01 -2.27460295e-01 8.56166482e-01 8.14085245e-01
-5.40195823e-01 2.46600345e-01 6.10639989e-01 -2.73748249e-01
-7.38053262e-01 -7.96456754e-01 4.74141687e-02 1.64490426e+00
3.33412439e-01 -6.26812398e-01 -5.83657384e-01 -9.78624225e-01
5.37598133e-02 7.07595885e-01 -9.29950237e-01 -6.60321355e-01
-2.95031190e-01 -1.02001369e+00 4.71156716e-01 5.56909084e-01
5.96635699e-01 -1.22766554e+00 -8.49624127e-02 3.17979008e-01
2.87040956e-02 -9.77784097e-01 -4.86079663e-01 1.71146572e-01
-1.00043726e+00 -1.19632292e+00 -7.42963493e-01 -9.09694612e-01
7.32234001e-01 9.70481932e-01 1.25296533e+00 4.68920171e-01
-2.48041153e-01 1.16590992e-01 -6.03513181e-01 -2.40482584e-01
-1.16703831e-01 6.11395478e-01 -1.33187279e-01 -7.90922269e-02
3.23814154e-01 -1.01203573e+00 -7.06354082e-01 3.20954680e-01
-9.21375096e-01 6.23615570e-02 7.13390529e-01 9.63475227e-01
3.72692287e-01 3.56352001e-01 9.12202358e-01 -1.47837448e+00
8.63130987e-01 -9.76150393e-01 1.66614354e-02 4.94061023e-01
-7.97329068e-01 -2.80280650e-01 7.74364829e-01 -6.23283148e-01
-9.38440025e-01 -5.44343472e-01 1.13944776e-01 -6.27939641e-01
3.89992632e-03 8.14979374e-01 -3.63357097e-01 2.34930351e-01
5.65106332e-01 2.51428515e-01 -9.14403126e-02 -3.36873531e-01
1.68803617e-01 4.74704236e-01 2.47132435e-01 -5.00803053e-01
1.05358624e+00 2.19945103e-01 1.38522387e-01 -6.04008436e-01
-9.52514470e-01 -3.64765942e-01 -3.73384118e-01 -2.79763788e-01
4.94838357e-01 -9.25730407e-01 -4.63131219e-01 6.03650868e-01
-8.56123090e-01 -6.05495155e-01 -8.82357433e-02 9.73898843e-02
2.04365440e-02 3.50487173e-01 -5.68353832e-01 -6.06544554e-01
-6.09371901e-01 -1.04739892e+00 7.44273603e-01 4.71255451e-01
2.56744742e-01 -1.35538220e+00 -4.59938832e-02 1.52048841e-01
4.14537221e-01 4.80322354e-02 1.23590994e+00 -8.05620611e-01
-1.19189404e-01 -1.20581247e-01 -4.56991047e-01 7.68458052e-03
3.69245261e-01 -9.19933245e-02 -8.49695027e-01 -5.90405881e-01
-3.98359716e-01 -4.26779717e-01 1.13749623e+00 2.39417806e-01
1.56038833e+00 1.48409158e-02 -7.85922110e-01 5.31474352e-01
1.19214249e+00 -1.62379026e-01 2.71858931e-01 1.86897695e-01
1.05254996e+00 5.02476275e-01 5.96067488e-01 4.14212167e-01
6.90865695e-01 4.73205954e-01 5.80873549e-01 2.01096937e-01
-3.17700297e-01 -4.70791370e-01 -5.62807247e-02 8.57424378e-01
1.45077720e-01 -5.15558124e-01 -1.00091648e+00 6.01383150e-01
-2.12250805e+00 -9.04971004e-01 -1.02454871e-01 1.94009888e+00
6.14897549e-01 3.51874411e-01 5.53202182e-02 6.52203783e-02
1.16480005e+00 6.96574867e-01 -8.59130263e-01 1.82078376e-01
2.75610328e-01 -2.21681729e-01 1.95466474e-01 -6.71371669e-02
-1.04109311e+00 8.12228143e-01 4.29512501e+00 1.45110595e+00
-7.71707237e-01 5.16508162e-01 7.13279009e-01 1.02404445e-01
-5.44700503e-01 -4.97821644e-02 -6.94498658e-01 6.81583881e-01
6.08206689e-01 -4.91108328e-01 1.33534864e-01 1.05034482e+00
-1.53003350e-01 3.02231520e-01 -9.57983196e-01 8.97045135e-01
1.10053189e-01 -1.12572205e+00 2.80619729e-02 1.59143344e-01
8.46988618e-01 3.70226614e-02 -2.46343147e-02 1.10312057e+00
1.88254371e-01 -6.10020936e-01 3.20413649e-01 -1.81607381e-02
7.90323019e-01 -6.76118255e-01 6.62113428e-01 7.38139570e-01
-1.70346403e+00 -1.98125273e-01 -7.89552927e-01 1.69150114e-01
-1.30899221e-01 8.22254360e-01 -4.67588693e-01 6.75871372e-01
6.85843766e-01 8.39667737e-01 -8.30648959e-01 9.10282791e-01
-1.41028658e-01 5.73357701e-01 3.76252010e-02 -2.44263038e-01
1.02193318e-01 -1.32287607e-01 4.59600478e-01 9.26253617e-01
2.65815914e-01 9.03239623e-02 4.52142149e-01 5.68944693e-01
-6.45079434e-01 1.10700121e-02 -4.89199787e-01 6.13126121e-02
1.04564750e+00 1.80894256e+00 -8.60082805e-01 -3.42535496e-01
-5.69405794e-01 7.48399794e-01 7.93385029e-01 5.19483507e-01
-7.96113133e-01 -5.97867668e-01 4.67198402e-01 3.49041633e-02
1.20886497e-01 1.36290878e-01 -1.59365818e-01 -1.21679819e+00
9.82079059e-02 -5.51917017e-01 6.97612941e-01 -3.45189422e-01
-1.62905705e+00 4.88380462e-01 7.92241544e-02 -1.28642869e+00
1.74873456e-01 -3.36583614e-01 -1.27412057e+00 6.14071488e-01
-1.39893639e+00 -1.37387216e+00 -8.73327255e-01 4.38944459e-01
7.81866968e-01 -2.92274505e-01 4.88792479e-01 4.07539785e-01
-9.87955153e-01 9.30916846e-01 1.32720883e-03 3.23840022e-01
7.47018218e-01 -1.09746635e+00 5.37876487e-01 6.91989243e-01
3.84531282e-02 5.78444779e-01 4.18321937e-01 -9.31131601e-01
-1.35667646e+00 -1.38810229e+00 3.64840508e-01 -8.90714228e-02
7.87378669e-01 -4.37980294e-01 -1.15470922e+00 6.22560441e-01
-4.98214252e-02 4.13349986e-01 6.53888822e-01 5.57508945e-01
-5.31150758e-01 -3.88720632e-01 -9.01976287e-01 6.29310548e-01
1.52727127e+00 -4.99703020e-01 -2.76325822e-01 4.27009225e-01
9.21428382e-01 -1.15138367e-01 -8.36419106e-01 7.31227696e-01
2.65500665e-01 -7.09895253e-01 6.73501492e-01 -6.58272505e-01
4.93814468e-01 -1.76645517e-02 1.57130659e-01 -1.70785749e+00
-7.11275935e-01 -2.21286863e-01 -5.07130921e-01 1.54022467e+00
4.60498631e-01 -6.95283175e-01 1.03810084e+00 3.73614222e-01
-2.63354689e-01 -9.09748971e-01 -8.56192231e-01 -7.03651488e-01
-5.53864911e-02 -2.78371811e-01 5.54647744e-01 1.11818147e+00
-1.30390242e-01 6.41337633e-01 -3.10366094e-01 -1.10373810e-01
9.26465452e-01 1.66416034e-01 7.89049804e-01 -1.53183067e+00
-2.97019362e-01 -4.90666091e-01 -2.63771892e-01 -7.11523056e-01
5.33879042e-01 -1.26651764e+00 -6.08877391e-02 -1.62223911e+00
7.18928635e-01 -7.77463317e-01 -7.73624361e-01 6.68498755e-01
-8.91913712e-01 3.34945470e-02 5.56503348e-02 3.71807516e-01
-8.95442903e-01 7.22749770e-01 1.36630547e+00 -3.69371712e-01
-1.57034114e-01 1.01412021e-01 -7.67187953e-01 5.34058809e-01
8.68508458e-01 -6.92062438e-01 -8.74144912e-01 -4.65309858e-01
1.96644962e-01 -1.93338394e-01 1.99449494e-01 -9.41876531e-01
4.74503398e-01 -1.26583889e-01 3.22307825e-01 -3.79919231e-01
9.54217389e-02 -6.34939015e-01 -4.71090339e-02 6.03451848e-01
-2.78396457e-01 -1.14252411e-01 -1.10606968e-01 1.10549450e+00
2.99564358e-02 -3.93405408e-01 5.51091552e-01 -1.75733358e-01
-9.54837859e-01 9.82483685e-01 1.07011221e-01 3.61418575e-01
1.16312599e+00 3.37610282e-02 -8.62040579e-01 -1.55723676e-01
-5.21392822e-01 7.14451671e-01 3.68089080e-01 6.45787954e-01
4.28194940e-01 -1.41105878e+00 -6.93416297e-01 -7.51378536e-02
5.95703900e-01 1.26155570e-01 8.82132173e-01 8.58974695e-01
7.99838379e-02 -1.69549614e-01 -1.51236787e-01 -4.43945497e-01
-1.06857169e+00 6.56499445e-01 1.23843610e-01 -4.54296887e-01
-5.19355714e-01 8.70555103e-01 2.58517593e-01 -6.10543311e-01
3.08323205e-01 3.04492384e-01 -2.98728496e-01 3.23221445e-01
4.58911628e-01 5.95302999e-01 -4.52217646e-02 -2.95974582e-01
-3.54690164e-01 2.84409732e-01 -3.97795975e-01 5.05942941e-01
1.19088912e+00 -5.05024008e-02 -5.36754131e-02 4.94953454e-01
1.12630117e+00 -3.11985940e-01 -1.00071061e+00 -5.34621000e-01
-1.78960696e-01 -2.40310431e-01 3.84280086e-02 -3.91966909e-01
-1.36656392e+00 9.61546421e-01 2.55763918e-01 3.50691915e-01
1.00831878e+00 2.95280479e-02 7.74943709e-01 3.87859195e-01
3.91223729e-01 -1.12243736e+00 2.37691686e-01 2.80501723e-01
2.78309643e-01 -1.42415690e+00 1.11932024e-01 -6.86980963e-01
-4.72485840e-01 8.58495057e-01 1.15550232e+00 -1.53847574e-03
8.84298861e-01 -1.22614302e-01 -2.14628160e-01 -2.26258248e-01
-9.26502645e-01 -3.06650758e-01 1.77945301e-01 5.80994844e-01
4.16453332e-01 1.91680640e-01 -2.56325334e-01 7.41201162e-01
4.20827508e-01 -1.81455508e-01 2.71865696e-01 7.75172174e-01
-5.27629256e-01 -1.07506883e+00 2.38257706e-01 9.82110202e-01
-1.21922009e-01 -1.57369420e-01 -2.06634775e-02 7.08878517e-01
1.06213391e-01 9.75813210e-01 -4.15953510e-02 -7.15351164e-01
2.92169064e-01 6.84558004e-02 1.59833327e-01 -9.27650392e-01
-5.52735329e-01 -1.93163022e-01 -6.29414916e-02 -3.06254089e-01
-1.63393021e-01 -2.16374710e-01 -1.11214423e+00 -4.63573188e-01
-3.73460203e-01 1.90688834e-01 1.59270078e-01 7.72630453e-01
3.99915665e-01 1.02946079e+00 6.70037866e-01 -6.92758977e-01
-8.02266300e-01 -1.35235929e+00 -7.72470415e-01 5.42729259e-01
-1.74205601e-01 -8.04627240e-01 -3.87718797e-01 -5.45895994e-01] | [7.427776336669922, 6.159577369689941] |
bf73e1c8-d168-4c79-bfaa-f3a1968ea423 | image-retargeting-by-content-aware-synthesis | 1403.6566 | null | http://arxiv.org/abs/1403.6566v2 | http://arxiv.org/pdf/1403.6566v2.pdf | Image Retargeting by Content-Aware Synthesis | Real-world images usually contain vivid contents and rich textural details,
which will complicate the manipulation on them. In this paper, we design a new
framework based on content-aware synthesis to enhance content-aware image
retargeting. By detecting the textural regions in an image, the textural image
content can be synthesized rather than simply distorted or cropped. This method
enables the manipulation of textural & non-textural regions with different
strategy since they have different natures. We propose to retarget the textural
regions by content-aware synthesis and non-textural regions by fast
multi-operators. To achieve practical retargeting applications for general
images, we develop an automatic and fast texture detection method that can
detect multiple disjoint textural regions. We adjust the saliency of the image
according to the features of the textural regions. To validate the proposed
method, comparisons with state-of-the-art image targeting techniques and a user
study were conducted. Convincing visual results are shown to demonstrate the
effectiveness of the proposed method. | ['Tong-Yee Lee', 'Xiaopeng Zhang', 'Weiming Dong', 'Fuzhang Wu', 'Yan Kong', 'Xing Mei'] | 2014-03-26 | null | null | null | null | ['image-retargeting'] | ['computer-vision'] | [ 7.67141640e-01 -8.74032974e-02 -3.61844003e-02 -5.11043929e-02
-5.62865473e-02 -5.00416398e-01 4.80538577e-01 -8.97037834e-02
-1.28318176e-01 4.38720822e-01 3.56177002e-01 6.18290976e-02
1.60383299e-01 -8.60160112e-01 -5.38076699e-01 -6.72075689e-01
3.30543160e-01 -2.56471813e-01 9.12561297e-01 -4.93960947e-01
8.91647637e-01 4.64484066e-01 -1.78392732e+00 4.27143335e-01
9.67438579e-01 6.23453617e-01 7.25590169e-01 5.44573486e-01
-2.53711075e-01 4.60065216e-01 -6.76585793e-01 6.92345155e-03
3.01414222e-01 -5.25276899e-01 -4.44183111e-01 5.48909068e-01
3.29292417e-01 -4.69194353e-01 -1.87318698e-01 1.38146746e+00
5.24542570e-01 1.32004142e-01 6.53191209e-01 -9.15887415e-01
-9.92918670e-01 4.18132961e-01 -1.25829065e+00 2.66542822e-01
2.67558038e-01 -9.14469063e-02 3.79825234e-01 -8.40260684e-01
6.22932315e-01 1.39253378e+00 8.55099037e-02 4.45409656e-01
-1.01630294e+00 -5.94213128e-01 2.77669698e-01 2.89722711e-01
-1.36771488e+00 -4.88875270e-01 1.21895504e+00 -2.45958447e-01
2.16854021e-01 7.61981308e-01 3.54001760e-01 5.52016616e-01
5.13963997e-01 6.91911638e-01 1.20481515e+00 -7.41187751e-01
-4.12680246e-02 3.72540146e-01 -3.24943542e-01 8.10604513e-01
1.36147767e-01 2.80018169e-02 -3.03955555e-01 1.64174497e-01
1.23163748e+00 1.13276832e-01 -5.59722304e-01 -6.40081286e-01
-1.46392798e+00 3.41414779e-01 1.95025861e-01 3.64378452e-01
-5.81366718e-02 -5.23894988e-02 3.61445904e-01 9.92279202e-02
4.75995630e-01 2.17504844e-01 -5.47686331e-02 3.03386718e-01
-9.59466755e-01 1.21563733e-01 -3.18783484e-02 8.95475864e-01
9.03253496e-01 1.46601617e-01 -4.59231943e-01 1.01191378e+00
7.51526952e-02 6.52978122e-01 6.51288033e-01 -5.87640524e-01
3.53248894e-01 5.94652116e-01 2.32844442e-01 -1.54055655e+00
-1.93613261e-01 -1.26469806e-02 -8.75368416e-01 5.23099065e-01
-7.93675333e-02 -3.01019568e-02 -1.06121385e+00 1.21334469e+00
5.63793004e-01 -2.22783864e-01 -1.88735411e-01 1.23653138e+00
8.41309309e-01 9.12855923e-01 -1.07318282e-01 -2.44124204e-01
1.20660472e+00 -1.19201386e+00 -9.71670628e-01 -9.20632631e-02
2.36396283e-01 -1.16167533e+00 1.35533607e+00 2.44784981e-01
-1.08489931e+00 -6.75353587e-01 -1.22956276e+00 -6.29194034e-03
-2.27728680e-01 4.65219170e-01 9.96586084e-02 6.65868640e-01
-8.70007813e-01 6.03477322e-02 -2.80052155e-01 -2.94647515e-01
-2.31429897e-02 4.74427454e-02 -2.81848788e-01 1.00844994e-01
-9.99973059e-01 8.43497872e-01 7.00721979e-01 -6.04108945e-02
-7.19064951e-01 -3.03921908e-01 -7.30888784e-01 1.50714412e-01
4.99553680e-01 -1.67633131e-01 8.87974739e-01 -1.55111265e+00
-1.61804342e+00 8.42185616e-01 -1.62860855e-01 1.54037312e-01
4.19546068e-01 6.74817264e-02 -6.89810395e-01 4.84648287e-01
1.78227112e-01 6.09983623e-01 1.49387610e+00 -1.47937751e+00
-8.29164267e-01 3.55311893e-02 -1.04460590e-01 5.91548324e-01
-4.36380059e-01 2.00301096e-01 -6.59444988e-01 -1.35697794e+00
1.85964659e-01 -4.20573115e-01 -1.07164858e-02 2.54651725e-01
-6.50365591e-01 4.00894016e-01 1.54920554e+00 -6.59385979e-01
1.29008055e+00 -2.37315536e+00 1.64233804e-01 2.28417575e-01
4.18183744e-01 2.30241269e-01 -2.81167150e-01 2.77290672e-01
-6.64340407e-02 1.78931221e-01 -2.40454838e-01 3.64586055e-01
-4.54224259e-01 -3.18987876e-01 -1.83215737e-01 3.32604378e-01
1.02731191e-01 6.47620440e-01 -6.58983707e-01 -7.18916237e-01
6.42868340e-01 1.34636283e-01 -3.09295267e-01 4.46646065e-02
-1.52794421e-01 4.09720272e-01 -5.94746232e-01 7.74033546e-01
1.06930244e+00 1.91124395e-01 -2.07581207e-01 -5.20468891e-01
-3.32472742e-01 -5.21141887e-01 -1.17317033e+00 1.13069248e+00
-3.71667445e-01 7.25449920e-01 1.51277795e-01 -6.90437198e-01
1.19321930e+00 -4.70250323e-02 2.23348722e-01 -9.12214041e-01
3.05505395e-01 1.11147866e-01 -3.01645845e-01 -8.64876211e-01
9.98066187e-01 -5.91527484e-02 9.26155299e-02 4.91174281e-01
-6.42377436e-01 -2.31922701e-01 5.51930517e-02 5.55733107e-02
3.24682981e-01 7.93487281e-02 2.42984518e-01 -4.90564674e-01
8.83350015e-01 -1.02116361e-01 3.41134340e-01 5.05764425e-01
-1.37127474e-01 8.15536797e-01 1.49716049e-01 -3.82800132e-01
-1.19046211e+00 -8.68812621e-01 -1.09277494e-01 1.10913551e+00
1.08923018e+00 -1.53173372e-01 -8.96955729e-01 -5.54236293e-01
-3.01341772e-01 6.20903015e-01 -7.26008058e-01 -2.19239473e-01
-5.09836555e-01 -6.29980147e-01 -3.94430794e-02 -2.64990740e-02
1.03723538e+00 -1.09293318e+00 -8.46017361e-01 1.06909826e-01
-5.04542232e-01 -8.31007421e-01 -8.41651678e-01 -3.77646863e-01
-6.64782345e-01 -8.40804458e-01 -1.31761742e+00 -1.13468087e+00
9.11014497e-01 9.76890266e-01 5.69456756e-01 1.50371939e-01
-2.88979739e-01 1.81983873e-01 -6.31481409e-01 -2.35554844e-01
-4.74731386e-01 -4.01836596e-02 -3.12942654e-01 4.57465947e-01
-2.18256503e-01 -3.07320327e-01 -7.68120825e-01 8.16637576e-01
-1.39402747e+00 7.06581235e-01 6.41073048e-01 6.30447388e-01
5.37800968e-01 5.44835746e-01 4.72012103e-01 -7.50560105e-01
5.37323296e-01 -6.21179529e-02 -4.70143974e-01 4.56262380e-01
-3.52492988e-01 2.89717270e-03 5.19826770e-01 -6.95095778e-01
-1.38055992e+00 -1.40013620e-01 4.41718549e-01 -1.74376860e-01
-1.68342870e-02 1.07366070e-02 -2.45240837e-01 -4.45148677e-01
6.94420695e-01 6.08849049e-01 -3.93441133e-02 -3.93034518e-01
5.54905057e-01 1.01957786e+00 5.33691168e-01 -3.58485997e-01
8.95060837e-01 4.76172179e-01 -2.95051754e-01 -8.12229514e-01
-2.69373059e-01 -7.73873776e-02 -4.10674632e-01 -5.35030007e-01
8.36660564e-01 -6.53657734e-01 -2.25230291e-01 5.51527500e-01
-9.50387836e-01 -2.37970456e-01 4.33264673e-02 4.75022495e-02
-4.29736465e-01 8.70476842e-01 -2.48837665e-01 -5.34776747e-01
-4.04082328e-01 -1.26907158e+00 1.10104203e+00 3.97614449e-01
1.63056731e-01 -6.15971267e-01 -3.41288835e-01 8.54053795e-02
7.22306907e-01 3.25151563e-01 1.08183181e+00 2.14460775e-01
-6.36610091e-01 1.27880111e-01 -6.68319821e-01 5.65701351e-03
6.22039795e-01 1.27219439e-01 -6.87067270e-01 -1.81712225e-01
-1.75904691e-01 2.69875646e-01 6.94644213e-01 3.26181948e-01
1.45501649e+00 -5.40586770e-01 -4.68157113e-01 4.00089443e-01
1.45555627e+00 3.86172235e-01 9.16037917e-01 7.11623192e-01
8.47291052e-01 6.82664394e-01 9.69557881e-01 3.67619127e-01
-6.50613680e-02 8.34412456e-01 4.79033172e-01 -5.70702851e-01
-3.00120562e-01 -4.78463992e-02 2.18400285e-01 4.32375997e-01
1.51164606e-01 -2.87888706e-01 -4.41794872e-01 6.44373119e-01
-1.66967261e+00 -8.47788334e-01 -1.46074176e-01 2.04271078e+00
8.91803086e-01 6.80722296e-02 -1.01438917e-01 2.21687570e-01
1.35848069e+00 2.42334902e-01 -4.37833101e-01 -5.37827909e-01
-2.55868345e-01 -2.60555327e-01 6.48327589e-01 4.58828449e-01
-1.08933353e+00 1.17554665e+00 6.47700930e+00 1.34297931e+00
-1.58218038e+00 -1.95264071e-02 8.22678328e-01 7.34436437e-02
-5.95642686e-01 -1.94139242e-01 -4.12079275e-01 5.82355618e-01
1.90490838e-02 -3.89730632e-01 4.25985098e-01 5.71318209e-01
4.90601003e-01 -6.53766274e-01 -3.04890186e-01 1.17285728e+00
1.89515278e-01 -1.21287799e+00 5.08477747e-01 -5.76738119e-01
9.90716696e-01 -8.37301433e-01 3.98863703e-01 -2.40685582e-01
-3.55131090e-01 -5.64153254e-01 1.02228177e+00 3.80628109e-01
1.20055628e+00 -7.71397293e-01 3.47391695e-01 -2.01260522e-02
-1.24420369e+00 -3.38287950e-02 -5.95229089e-01 2.53420174e-01
5.65695614e-02 7.04833329e-01 -4.89220709e-01 4.71968591e-01
7.95912862e-01 6.89431846e-01 -8.70619714e-01 1.01059389e+00
-3.50382552e-02 1.52571008e-01 1.48233771e-01 -3.06415744e-02
-5.84186502e-02 -1.45013988e-01 7.02917695e-01 1.15136397e+00
3.64108980e-01 1.14360690e-01 -1.60536095e-01 9.19951558e-01
3.43900681e-01 6.25642359e-01 -5.94887257e-01 1.29680470e-01
3.48917127e-01 1.25681579e+00 -1.05359459e+00 -5.50808549e-01
-3.32431734e-01 1.41630113e+00 -2.97993422e-01 4.78572845e-01
-8.69579971e-01 -8.67106736e-01 -6.26297742e-02 4.07981694e-01
2.67759353e-01 -6.98529556e-02 -5.01138628e-01 -1.05185497e+00
-6.45988584e-02 -1.05609572e+00 -8.05949345e-02 -1.23498809e+00
-8.28101158e-01 5.96990526e-01 1.12135485e-01 -1.37627804e+00
3.91799957e-01 -1.68473154e-01 -6.64815784e-01 9.05214489e-01
-1.34442127e+00 -1.04369140e+00 -6.04213834e-01 6.01302862e-01
7.54746258e-01 -1.69534713e-01 1.81162655e-01 9.78157297e-02
-4.66883034e-01 5.61269104e-01 1.28684521e-01 -3.13380361e-01
9.06449199e-01 -6.68951213e-01 2.70485997e-01 1.18866503e+00
-4.85085696e-01 1.42570049e-01 8.35282683e-01 -8.64596963e-01
-9.83522713e-01 -1.05958652e+00 3.74835610e-01 8.16339552e-02
4.20973934e-02 -2.52098441e-01 -7.83140838e-01 2.99444586e-01
2.66087621e-01 -5.25431633e-01 4.72996663e-03 -6.40974581e-01
7.23628476e-02 -1.33901179e-01 -1.30889904e+00 1.22653830e+00
7.46072888e-01 -9.44759101e-02 -4.66400504e-01 1.02392752e-02
7.77892053e-01 -3.29227507e-01 -2.79780328e-01 6.16497636e-01
5.46136975e-01 -1.10211122e+00 9.17362332e-01 3.64977926e-01
5.11875451e-01 -7.92998672e-01 -2.94385385e-02 -9.49074507e-01
-5.26669919e-01 -6.15184546e-01 2.32076630e-01 1.21958137e+00
6.85051605e-02 -4.85782325e-01 6.23313844e-01 2.67413080e-01
1.68868735e-01 -2.24994823e-01 -4.24693227e-01 -4.02258992e-01
-5.18997729e-01 2.86366433e-01 6.37577713e-01 9.55374241e-01
-2.58685332e-02 8.60480070e-02 -6.90914929e-01 6.27131760e-02
4.35060024e-01 1.99433446e-01 7.11450815e-01 -7.17906177e-01
-2.81873923e-02 -6.17352247e-01 -5.38353562e-01 -9.66532648e-01
-3.92565578e-01 -4.07839000e-01 3.04771820e-03 -1.34207296e+00
3.95295978e-01 -4.31587964e-01 -9.64393467e-02 5.42585671e-01
-4.72973436e-01 5.23272395e-01 8.96331668e-02 1.44507468e-01
-2.66146213e-01 7.44020760e-01 1.87783098e+00 -3.21054339e-01
-5.03534198e-01 -3.96045029e-01 -7.70273626e-01 3.92690867e-01
1.03631556e+00 -1.43788323e-01 -5.90475500e-01 -3.61072242e-01
-1.23417132e-01 -7.61343390e-02 1.52890041e-01 -9.36915874e-01
-2.22751364e-01 -4.84496713e-01 5.09698570e-01 -8.13257873e-01
-2.65508201e-02 -8.77240002e-01 -3.62639241e-02 4.60573733e-01
-3.31171185e-01 7.82615542e-02 4.38844204e-01 5.17444551e-01
-1.08335190e-01 -2.45936409e-01 1.00365114e+00 3.30936052e-02
-9.55075264e-01 -3.54605168e-02 -7.99821556e-01 -3.57204378e-01
1.37450135e+00 -5.41712105e-01 -3.70070785e-01 -5.57625651e-01
-3.18756908e-01 -1.92136168e-01 7.53449500e-01 6.03865325e-01
9.97535348e-01 -1.29320598e+00 -6.91425800e-01 4.98043776e-01
1.70795858e-01 -3.62103581e-01 6.60461843e-01 6.06321573e-01
-9.36335802e-01 -3.06930505e-02 -7.78067827e-01 -4.84814048e-01
-1.51653910e+00 1.04943180e+00 3.12800169e-01 3.63675773e-01
-7.04802334e-01 3.29083413e-01 9.65362251e-01 1.81569923e-02
-2.53700111e-02 -1.86751872e-01 -5.12050569e-01 -3.50040913e-01
7.12974191e-01 4.42607492e-01 -2.67528832e-01 -6.43897951e-01
1.19614322e-02 1.04276872e+00 -3.07256579e-01 -2.41233110e-01
6.43872261e-01 -7.75040269e-01 -2.64108390e-01 1.66378021e-01
9.31420386e-01 3.47351283e-01 -1.14003098e+00 -2.62899607e-01
-4.57320750e-01 -9.48557436e-01 2.30391443e-01 -8.48191619e-01
-1.23191810e+00 6.97407007e-01 9.66094673e-01 1.22107953e-01
1.62597799e+00 -4.14994091e-01 7.45727181e-01 -2.10658953e-01
2.38765597e-01 -1.15568507e+00 3.17902267e-01 -4.52614069e-04
1.15934801e+00 -1.02370703e+00 1.14011042e-01 -8.65100801e-01
-7.00164676e-01 1.16807961e+00 6.96781814e-01 -3.67040932e-02
3.69044185e-01 -1.76497195e-02 2.43662342e-01 -5.70919737e-02
-2.43244857e-01 -1.48038790e-01 4.78673816e-01 6.94289863e-01
1.16799578e-01 -5.95642477e-02 -6.14442945e-01 1.06559917e-01
1.20999999e-01 -3.06523085e-01 8.99368227e-01 7.98502505e-01
-9.28068459e-01 -9.33493435e-01 -9.50232506e-01 2.53154218e-01
-4.00728345e-01 -1.53543711e-01 -3.51590902e-01 5.81050575e-01
3.79733257e-02 1.13274550e+00 -1.09824970e-01 -5.18317997e-01
1.38769031e-01 -4.84970957e-01 4.89620924e-01 -3.45523328e-01
-1.45569265e-01 4.18244421e-01 -2.16040850e-01 -3.66691083e-01
-4.14492756e-01 -1.87600076e-01 -9.88660574e-01 -3.13218415e-01
-3.83322507e-01 -1.29789710e-01 4.86422747e-01 5.11031449e-01
3.11832249e-01 6.50199413e-01 9.48748589e-01 -1.08936286e+00
1.81880802e-01 -8.24255049e-01 -7.06499755e-01 4.12723184e-01
3.61752987e-01 -6.73937082e-01 -1.62129059e-01 3.21276039e-01] | [11.127891540527344, -1.1675915718078613] |
aa2f7984-f5ac-499b-b27f-59e87517c4bf | individualized-rank-aggregation-using-nuclear | 1410.0860 | null | http://arxiv.org/abs/1410.0860v1 | http://arxiv.org/pdf/1410.0860v1.pdf | Individualized Rank Aggregation using Nuclear Norm Regularization | In recent years rank aggregation has received significant attention from the
machine learning community. The goal of such a problem is to combine the
(partially revealed) preferences over objects of a large population into a
single, relatively consistent ordering of those objects. However, in many
cases, we might not want a single ranking and instead opt for individual
rankings. We study a version of the problem known as collaborative ranking. In
this problem we assume that individual users provide us with pairwise
preferences (for example purchasing one item over another). From those
preferences we wish to obtain rankings on items that the users have not had an
opportunity to explore. The results here have a very interesting connection to
the standard matrix completion problem. We provide a theoretical justification
for a nuclear norm regularized optimization procedure, and provide
high-dimensional scaling results that show how the error in estimating user
preferences behaves as the number of observations increase. | ['Yu Lu', 'Sahand N. Negahban'] | 2014-10-03 | null | null | null | null | ['collaborative-ranking'] | ['graphs'] | [ 2.34295517e-01 1.46007817e-02 -8.06473121e-02 -8.17740977e-01
-1.05773914e+00 -7.10393190e-01 2.79429048e-01 2.47364223e-01
-7.05772042e-01 9.36504602e-01 8.05124283e-01 2.26770472e-02
-7.58071899e-01 -5.82901239e-01 -5.19819200e-01 -8.14956605e-01
-5.00820994e-01 1.02256298e+00 -3.55671644e-01 -1.61699846e-01
3.10159355e-01 1.43143103e-01 -1.54287922e+00 6.64311767e-01
1.02350080e+00 6.83277130e-01 -1.39446009e-03 4.84062016e-01
2.17941955e-01 2.98237205e-01 -1.69372201e-01 -4.81242418e-01
7.06737638e-01 -9.36909765e-02 -9.08988714e-01 4.53911692e-01
5.76543510e-01 -2.14470461e-01 -1.36847496e-02 1.23822987e+00
4.90177631e-01 6.22756124e-01 7.96028912e-01 -1.14992535e+00
-5.67023516e-01 9.72134531e-01 -5.59980750e-01 -2.19334975e-01
6.28204465e-01 -6.86816156e-01 1.74646926e+00 -1.02054155e+00
5.27075887e-01 1.17430353e+00 2.79644817e-01 1.90781802e-01
-1.75572622e+00 -7.12589547e-02 2.90672451e-01 1.42365536e-02
-1.13227463e+00 -2.65677452e-01 5.54054260e-01 -5.13571382e-01
2.06014346e-02 8.40198398e-01 3.25866610e-01 6.52051449e-01
-2.54653960e-01 8.64310384e-01 1.41759348e+00 -3.03274959e-01
4.28150356e-01 3.77836406e-01 4.58678097e-01 2.01073125e-01
7.85770178e-01 -2.73600131e-01 -6.33406043e-01 -8.05294693e-01
3.24152589e-01 2.19177380e-01 -1.72959611e-01 -1.02006733e+00
-1.40029073e+00 1.01254439e+00 2.63144493e-01 2.23999768e-02
-5.09025693e-01 -2.07791194e-01 -2.28440575e-02 6.70710802e-01
4.58308101e-01 7.04975665e-01 -5.20017385e-01 3.16014051e-01
-7.47996032e-01 4.76895750e-01 9.19236183e-01 8.80934298e-01
9.66951370e-01 -5.62228620e-01 1.50883328e-02 8.97779167e-01
2.41532952e-01 3.37219596e-01 1.57325156e-02 -1.33781147e+00
6.69861257e-01 2.99965918e-01 7.27625966e-01 -1.01063788e+00
-3.57609868e-01 -2.69247234e-01 -8.34328651e-01 1.23829670e-01
7.52391517e-01 -3.92334461e-01 -2.29663640e-01 1.86074483e+00
4.31197956e-02 -2.55357891e-01 -1.71161350e-02 1.05046916e+00
7.22192749e-02 5.38822532e-01 -2.77813226e-01 -4.87895399e-01
1.10272062e+00 -4.70784694e-01 -4.96977478e-01 -3.39702405e-02
5.61841488e-01 -7.33848810e-01 8.18210244e-01 7.37725019e-01
-1.19737637e+00 -2.22881228e-01 -7.27415860e-01 1.48780391e-01
-7.42054284e-02 -3.52712013e-02 1.18297374e+00 7.19202936e-01
-1.18363452e+00 8.26217473e-01 -2.51663506e-01 -3.91593128e-01
6.23086207e-02 9.45601225e-01 -5.36508381e-01 -1.84536621e-01
-9.38254535e-01 6.02304041e-01 -1.82540044e-02 1.06653631e-01
-3.63166392e-01 -3.65617901e-01 -3.77536088e-01 1.14869536e-03
5.51793575e-01 -7.58070350e-01 1.15465260e+00 -9.16543782e-01
-8.97240639e-01 6.04995251e-01 -1.29883841e-01 1.33096352e-02
4.31804687e-01 -5.03106356e-01 -2.74439603e-01 -4.95255083e-01
2.51070410e-01 2.06229478e-01 3.58736843e-01 -1.40785503e+00
-9.14561152e-01 -6.15250051e-01 3.10913682e-01 5.59253752e-01
-5.34283698e-01 6.73567876e-02 -2.11380437e-01 -3.06973755e-01
4.57091242e-01 -1.28083885e+00 -6.36316538e-01 -4.38055396e-01
-3.29942733e-01 -8.84637684e-02 -1.88113600e-02 -4.00468111e-01
1.04379070e+00 -1.83586991e+00 6.23373628e-01 8.19545805e-01
2.34822303e-01 -4.78348732e-01 -2.81520367e-01 6.48925245e-01
-9.34288129e-02 2.03793511e-01 -1.10764064e-01 -3.20367128e-01
3.25657099e-01 1.91342101e-01 -3.69199634e-01 7.12261975e-01
-5.19208014e-01 3.74504745e-01 -1.06532204e+00 -3.44412506e-01
-2.66703159e-01 -5.71791716e-02 -1.09834957e+00 6.60714954e-02
1.11116260e-01 2.80918539e-01 -7.59599149e-01 3.87861311e-01
8.19506347e-01 -3.05356413e-01 8.20201218e-01 -2.07077160e-01
-2.29496527e-02 1.32419005e-01 -1.80271411e+00 1.61141253e+00
-1.71476945e-01 2.92895228e-01 2.92340726e-01 -9.38548148e-01
3.67666364e-01 2.49656036e-01 1.00871110e+00 -1.07381262e-01
-4.30587120e-03 3.28896821e-01 -9.49975699e-02 -3.09116602e-01
7.18692422e-01 -2.08796322e-01 -3.49424809e-01 7.78151453e-01
-2.78173327e-01 4.03253883e-01 4.62110281e-01 4.08977479e-01
9.57461953e-01 -3.66010040e-01 2.85014123e-01 -5.71633399e-01
2.37413004e-01 -1.52420057e-02 5.87568760e-01 1.01021278e+00
5.53902388e-01 6.91757083e-01 4.89017904e-01 -3.42197806e-01
-1.06175983e+00 -1.16346109e+00 -1.90056324e-01 1.37784445e+00
1.27633989e-01 -3.30881953e-01 -3.48957866e-01 -5.86891115e-01
3.35359454e-01 3.38338822e-01 -5.72599232e-01 4.40691561e-01
-3.24321777e-01 -1.23899519e+00 -3.66334885e-01 2.75493860e-01
1.40756005e-02 -6.12033844e-01 6.65794015e-02 1.83205813e-01
-4.46827322e-01 -5.03988802e-01 -8.10427547e-01 1.94647521e-01
-1.11617112e+00 -9.21598256e-01 -7.41588056e-01 -4.81962234e-01
1.03477001e+00 4.43356544e-01 1.26910686e+00 -1.78519726e-01
3.16804320e-01 5.44030011e-01 -3.01985592e-01 1.91057753e-02
1.47646546e-01 -7.70734623e-02 7.69841254e-01 5.39207101e-01
3.87415260e-01 -6.57877743e-01 -4.66920972e-01 5.25116086e-01
-8.69950712e-01 -3.63381982e-01 8.02416265e-01 6.80632889e-01
3.76265883e-01 1.17369890e-01 4.67318296e-01 -1.29514134e+00
9.94457603e-01 -6.67218149e-01 -7.12656081e-01 3.23973954e-01
-6.72905087e-01 4.73113894e-01 3.10475588e-01 -4.54281747e-01
-8.06993127e-01 2.44626030e-01 4.85731989e-01 1.88777879e-01
1.22174501e-01 7.86174119e-01 -4.77863699e-01 2.49528155e-01
5.77094018e-01 -3.17044854e-01 -1.84040755e-01 -8.73026490e-01
6.14998817e-01 6.65783644e-01 1.70711294e-01 -9.36644793e-01
7.77711332e-01 4.61162955e-01 8.29197541e-02 -6.81130946e-01
-9.56111431e-01 -6.91213131e-01 -6.89369857e-01 7.12069198e-02
5.27719975e-01 -8.16321552e-01 -8.70353460e-01 -2.87503541e-01
-8.65276694e-01 6.57647029e-02 -3.59447330e-01 8.57639194e-01
-6.25119925e-01 4.32652503e-01 -2.11229518e-01 -9.84309673e-01
3.18411142e-01 -8.71122897e-01 6.31462932e-01 -1.18826210e-01
-4.70160156e-01 -9.23280537e-01 4.02381301e-01 4.66496915e-01
1.94150344e-01 -1.00058943e-01 7.84868300e-01 -9.27056670e-01
-6.71713531e-01 -3.13974887e-01 -1.87967569e-02 1.80464029e-01
2.55549163e-01 -5.38030624e-01 -6.38380587e-01 -4.85840201e-01
-8.55506286e-02 -2.16290548e-01 7.86236167e-01 4.42341566e-01
8.57227802e-01 -6.18533254e-01 -2.54582107e-01 1.49549946e-01
1.36563456e+00 -8.00075531e-02 2.91592747e-01 3.41043882e-02
5.84119499e-01 9.36290801e-01 6.92232668e-01 6.23015642e-01
2.24089563e-01 8.30714881e-01 2.09299251e-02 1.00424811e-01
7.83566535e-01 8.44834968e-02 2.42422462e-01 6.68982089e-01
-4.85992789e-01 -2.42658511e-01 -3.10210228e-01 2.88124979e-01
-2.23845267e+00 -1.15465713e+00 -9.11133140e-02 2.95948982e+00
6.66853309e-01 -2.56018728e-01 4.64756727e-01 -7.77215287e-02
7.70253539e-01 -1.04728848e-01 -3.12481880e-01 -2.62546718e-01
-1.07602149e-01 -3.55048716e-01 7.12275028e-01 7.70515263e-01
-1.03813362e+00 9.14002657e-02 7.01579523e+00 3.83708835e-01
-3.49744469e-01 -1.74342170e-01 7.34675527e-01 -3.99506390e-01
-7.82832086e-01 1.84043333e-01 -6.31660223e-01 3.04105163e-01
6.06816232e-01 -3.41223955e-01 8.21110249e-01 6.86830699e-01
1.91532761e-01 -1.29538715e-01 -1.71201217e+00 9.49388206e-01
-2.83070523e-02 -8.63396227e-01 -2.91201994e-02 7.75140822e-01
1.10304379e+00 -1.76055059e-01 2.77264416e-01 -5.24851307e-03
9.16570783e-01 -9.23573852e-01 3.58650893e-01 7.64994085e-01
5.25939584e-01 -7.76513577e-01 6.12677217e-01 3.22239339e-01
-8.34723651e-01 -4.41330761e-01 -7.63151705e-01 -2.55192339e-01
2.88397402e-01 9.68456388e-01 -3.44553471e-01 4.64592397e-01
4.10668284e-01 4.13066655e-01 -3.23138177e-01 1.29706609e+00
3.26217324e-01 2.18468741e-01 -4.74381357e-01 -4.06755656e-02
-3.65122920e-03 -9.35128450e-01 4.55800593e-01 7.28349686e-01
6.14590526e-01 3.81241202e-01 3.36634785e-01 3.13817471e-01
-3.89261067e-01 3.57023448e-01 -6.21670127e-01 -2.30602492e-02
2.02639565e-01 1.27133107e+00 -4.35187042e-01 -3.11482728e-01
-6.97457612e-01 9.07901704e-01 3.32163721e-01 4.11779255e-01
-3.76312464e-01 1.50002390e-01 8.96649063e-01 1.17502920e-01
-9.30285174e-03 -1.29441753e-01 -2.26510897e-01 -1.43534410e+00
1.46095425e-01 -9.58290517e-01 6.66448653e-01 -4.51761663e-01
-1.49469733e+00 1.77293807e-01 2.47704074e-01 -1.31034470e+00
-1.65447041e-01 -5.34587979e-01 -2.92397588e-01 8.19970787e-01
-7.78075159e-01 -3.66850734e-01 2.29153231e-01 4.33324069e-01
1.80213332e-01 -2.00791195e-01 7.51947403e-01 3.62960726e-01
-1.51267648e-01 1.40791968e-01 7.30968356e-01 -3.72449845e-01
8.51312518e-01 -1.69257855e+00 -1.38493612e-01 5.11007905e-01
4.49856788e-01 1.02699900e+00 1.01679611e+00 -5.86605310e-01
-1.62396634e+00 -6.93189979e-01 1.24265385e+00 -9.26178753e-01
5.46265423e-01 -3.26678842e-01 -4.59409177e-01 7.62544453e-01
6.51887506e-02 -2.89043427e-01 1.09053636e+00 9.00910258e-01
-1.60685897e-01 -3.37137043e-01 -9.77150917e-01 7.51915395e-01
1.06234813e+00 -3.23252082e-01 -5.17393172e-01 5.79535723e-01
1.99913234e-01 2.69623995e-01 -7.50941515e-01 8.61242041e-02
8.74024868e-01 -1.01357484e+00 1.21776104e+00 -9.26810622e-01
8.70522559e-02 -2.96870291e-01 -7.07050622e-01 -1.49199843e+00
-7.60362804e-01 -6.12492323e-01 2.93871492e-01 9.31421518e-01
7.29266286e-01 -4.95468229e-01 1.06205285e+00 1.38111079e+00
2.85612285e-01 -5.10390401e-01 -6.63239300e-01 -6.67501271e-01
-1.31245747e-01 -6.65704757e-02 5.38606822e-01 8.52221370e-01
4.29481298e-01 5.87220430e-01 -8.98647189e-01 2.54888386e-01
1.04409945e+00 4.40414697e-01 7.25597322e-01 -1.77947986e+00
-5.55028617e-01 -1.89907804e-01 1.36030205e-02 -1.20714903e+00
-2.79949486e-01 -9.76387620e-01 -1.61905840e-01 -1.50333393e+00
7.35031843e-01 -7.72773027e-01 -6.76905155e-01 1.15439922e-01
-1.17461711e-01 2.09485173e-01 6.67677000e-02 3.20691228e-01
-1.07588470e+00 1.48511931e-01 9.13103938e-01 -1.41628355e-01
-4.21328783e-01 4.65471089e-01 -1.44665062e+00 6.25411034e-01
6.44678593e-01 -5.46919405e-01 -5.22885859e-01 -4.19431806e-01
8.45956028e-01 2.66227156e-01 -1.57336146e-01 -6.23966277e-01
4.66161877e-01 -5.52243114e-01 4.25291210e-01 -4.18017179e-01
4.03933138e-01 -1.04257393e+00 4.98433501e-01 -5.07946052e-02
-7.99092948e-01 1.08605094e-01 -6.28776133e-01 6.35376811e-01
-4.03089561e-02 -3.59664559e-01 3.61230195e-01 -1.27267852e-01
-3.38487983e-01 3.97055119e-01 -3.03311735e-01 -2.02686518e-01
5.84305704e-01 7.17605883e-03 4.60069105e-02 -7.08655000e-01
-1.22591901e+00 4.28650856e-01 5.89967966e-01 1.24264613e-01
4.91871774e-01 -1.64360321e+00 -9.79131818e-01 -1.06249340e-01
1.31316051e-01 -4.17181581e-01 -2.55138008e-03 8.33561838e-01
1.94040492e-01 4.63354468e-01 -6.38784096e-02 -9.16829482e-02
-1.37282383e+00 8.02881479e-01 -2.36278534e-01 -2.50213027e-01
-1.02602668e-01 7.03095496e-01 3.70687813e-01 -4.61894810e-01
1.55568108e-01 7.77045563e-02 -3.51230443e-01 5.80425262e-01
5.35816014e-01 6.24131978e-01 -9.48695466e-02 -5.17655790e-01
-3.10790479e-01 3.64139199e-01 -2.44665816e-01 -5.80595970e-01
1.63427269e+00 -3.68537128e-01 -4.23211247e-01 4.80107754e-01
1.14487469e+00 6.62800491e-01 -9.53949988e-01 -4.82145935e-01
2.63790756e-01 -8.76104295e-01 -3.13843787e-01 -5.09873211e-01
-8.35652113e-01 2.09942311e-01 3.11701387e-01 5.84959209e-01
9.37278509e-01 6.96146488e-02 8.96281824e-02 9.61835802e-01
6.90169573e-01 -1.22073805e+00 -1.67479571e-02 3.03946137e-01
9.28670049e-01 -1.23819709e+00 3.56732965e-01 -2.46657595e-01
-8.45390916e-01 7.71023154e-01 -3.42578925e-02 -2.67873585e-01
6.43010795e-01 -3.32473576e-01 -1.66785255e-01 -1.55966163e-01
-7.61316538e-01 -1.84018254e-01 6.03290796e-01 3.62831533e-01
5.79524755e-01 3.93193543e-01 -4.80250806e-01 5.48224866e-01
-1.30178005e-01 -3.86084110e-01 7.51231790e-01 6.42235637e-01
-6.76578403e-01 -1.64023614e+00 -5.06060839e-01 1.03232431e+00
-3.52481842e-01 -8.87492225e-02 -5.46361446e-01 1.36352152e-01
-1.19985215e-01 1.20126271e+00 -2.75083512e-01 -4.01623815e-01
3.42880428e-01 6.42057359e-02 4.17724043e-01 -8.45007539e-01
-9.28909555e-02 3.20633978e-01 3.54246289e-01 -5.59432626e-01
-5.12101233e-01 -1.18992960e+00 -5.61021507e-01 -4.63936657e-01
-3.16498578e-01 8.05118680e-01 4.64901537e-01 6.88024879e-01
4.22157422e-02 -9.08913761e-02 9.66890812e-01 -8.09672296e-01
-8.54905009e-01 -6.21184707e-01 -1.19177127e+00 6.52177811e-01
4.21648204e-01 -5.58280587e-01 -5.58157146e-01 -6.09563291e-02] | [9.523265838623047, 5.506854057312012] |
0fe8c421-1d29-4be7-bdc1-df571d22e15b | enhancement-on-model-interpretability-and | 2204.03173 | null | https://arxiv.org/abs/2204.03173v3 | https://arxiv.org/pdf/2204.03173v3.pdf | Automated Sleep Staging via Parallel Frequency-Cut Attention | This paper proposes a novel framework for automatically capturing the time-frequency nature of electroencephalogram (EEG) signals of human sleep based on the authoritative sleep medicine guidance. The framework consists of two parts: the first part extracts informative features by partitioning the input EEG spectrograms into a sequence of time-frequency patches. The second part is constituted by an attention-based architecture to efficiently search for the correlation between partitioned time-frequency patches and defining factors of sleep stages in parallel. The proposed pipeline is validated on the Sleep Heart Health Study dataset with new state-of-the-art results for the stages wake, N2, and N3, obtaining respective F1 scores of 0.93, 0.88, and 0.87, with only EEG signals used. The proposed method also has a high inter-rater reliability of 0.80 kappa. We also visualize the correspondence between sleep staging decisions and features extracted by the proposed method, providing strong interpretability for our model. | ['Shigehiko Kanaya', 'Wei Chen', 'Lingwei Zhu', 'MD Altaf-Ul-Amin', 'Naoaki Ono', 'Toshiyo Tamura', 'Ming Huang', 'Ziwei Yang', 'Zheng Chen'] | 2022-04-07 | null | null | null | null | ['sleep-staging'] | ['medical'] | [ 1.02617100e-01 1.32008240e-01 1.15406156e-01 -4.09378141e-01
-5.19736230e-01 -2.80539989e-01 2.20316634e-01 2.73903191e-01
-4.98946071e-01 5.87448955e-01 3.50697637e-01 -5.41694053e-02
-5.35867572e-01 -1.56005025e-01 1.03172027e-01 -6.55836463e-01
-5.26369512e-01 1.27317056e-01 -9.26389396e-02 1.67254061e-02
5.18777847e-01 1.16943754e-01 -1.75364423e+00 2.44009659e-01
1.11532092e+00 1.23543835e+00 1.20634638e-01 6.83709800e-01
3.21664512e-01 3.49047720e-01 -7.63603032e-01 1.16451107e-01
-2.01740652e-01 -5.45871675e-01 -7.45810747e-01 3.39100063e-02
-2.12093905e-01 2.23210037e-01 4.70453221e-03 9.33767140e-01
5.04060864e-01 1.36185884e-01 4.12006110e-01 -1.05257750e+00
-2.58949131e-01 3.62958431e-01 -3.15572262e-01 1.06006813e+00
5.54423332e-01 2.19400853e-01 1.06586385e+00 -5.51506042e-01
1.11864999e-01 4.58899677e-01 5.35453320e-01 4.19530839e-01
-1.10608745e+00 -6.81548297e-01 -2.21285239e-01 7.45053709e-01
-1.37615383e+00 -3.60402703e-01 6.58851266e-01 -5.60014725e-01
1.31289232e+00 5.07242918e-01 1.35377622e+00 8.55576515e-01
7.54366696e-01 2.73569316e-01 1.44111741e+00 -2.34681413e-01
5.10216415e-01 8.94115493e-02 7.07137287e-01 4.46652949e-01
2.44108792e-02 -1.29220784e-01 -8.51767659e-01 -1.31912038e-01
2.15695634e-01 1.87660173e-01 -3.75286460e-01 3.31617713e-01
-1.00450647e+00 6.00948334e-01 3.33133519e-01 6.28767610e-01
-6.84404373e-01 -3.53224963e-01 3.49199772e-01 7.20000714e-02
5.27358413e-01 5.86847782e-01 -3.65289778e-01 -4.48745817e-01
-1.61564004e+00 -8.73461440e-02 5.29874325e-01 5.62275052e-01
6.14365637e-01 -2.64121652e-01 -3.26041907e-01 7.00439215e-01
3.67559195e-01 2.61564165e-01 9.79704142e-01 -5.58803856e-01
-9.53131691e-02 7.79940844e-01 -8.00118446e-02 -6.61980987e-01
-1.14559197e+00 -6.47698224e-01 -8.51794600e-01 -3.71896833e-01
-2.91699499e-01 8.12778398e-02 -5.84738076e-01 1.37322462e+00
-8.52383524e-02 1.02789350e-01 -1.98631138e-01 8.18578839e-01
8.21432114e-01 3.98330063e-01 1.28836915e-01 -4.95641500e-01
2.02090216e+00 -6.84059381e-01 -1.03600454e+00 -3.57019067e-01
-4.08868976e-02 -2.68537134e-01 1.05912423e+00 6.22079790e-01
-1.15342832e+00 -7.17977345e-01 -1.22384441e+00 -4.31780629e-02
-1.69208050e-01 4.42015529e-01 3.15447748e-01 6.18419945e-01
-1.39060354e+00 6.14934802e-01 -1.19181156e+00 -4.57356304e-01
6.07548892e-01 7.43110895e-01 -1.81665435e-01 6.42203867e-01
-1.04807794e+00 8.83436561e-01 7.04809874e-02 1.81384757e-01
-7.19453990e-01 -7.29728520e-01 -4.51560289e-01 4.43258047e-01
-3.68687421e-01 -7.44596362e-01 8.40917885e-01 -6.91752791e-01
-1.26201892e+00 8.49608123e-01 -5.16528130e-01 -5.60614169e-01
-8.77327099e-02 -2.85333186e-01 -6.95612729e-01 5.01554370e-01
1.57722786e-01 3.68343085e-01 7.97328353e-01 -5.14404833e-01
-5.81468761e-01 -6.65073872e-01 -2.92910993e-01 -1.00252125e-02
-4.73937452e-01 1.92130804e-01 -1.74924195e-01 -4.12753433e-01
-5.37317432e-02 -7.45150506e-01 5.76244220e-02 -5.91269016e-01
-3.07432234e-01 -3.90120447e-01 1.82543084e-01 -7.73596406e-01
1.67816007e+00 -2.26815748e+00 1.74045801e-01 2.56863862e-01
6.62795424e-01 -7.19197169e-02 2.64665127e-01 2.59502113e-01
-2.83626616e-01 -1.23822249e-01 -3.23266238e-01 -4.40177917e-01
-2.19719572e-04 -2.04713076e-01 -4.22346480e-02 5.85878015e-01
2.92148888e-01 7.44013906e-01 -8.36137950e-01 -1.39783278e-01
2.36632183e-01 5.02435386e-01 -4.56649244e-01 4.26285177e-01
4.93431240e-01 4.64911640e-01 -8.14199075e-02 3.95592272e-01
3.09742600e-01 -3.59820217e-01 9.91721228e-02 -2.59905219e-01
-3.15262914e-01 6.32327080e-01 -5.18192232e-01 1.76073277e+00
-2.37185404e-01 8.25056076e-01 -2.50415325e-01 -7.97572613e-01
8.91333401e-01 3.82750988e-01 5.93508840e-01 -7.02636659e-01
3.43523085e-01 -9.81552973e-02 3.32453728e-01 -8.72409999e-01
2.14584768e-01 -2.64957070e-01 1.55932114e-01 5.24775803e-01
5.36517203e-01 1.48412973e-01 3.23323786e-01 -7.02912807e-02
1.30121076e+00 -2.38289684e-01 7.10977435e-01 -8.40880394e-01
5.68715334e-01 -4.64759439e-01 3.22866023e-01 3.00899833e-01
-3.70523930e-01 4.12851810e-01 5.91749847e-01 -4.27593410e-01
-6.02807939e-01 -8.10230613e-01 -3.51685494e-01 6.41493678e-01
-1.19751662e-01 -1.08226907e+00 -1.00648272e+00 -4.66081649e-01
-4.82675195e-01 8.59930277e-01 -1.18502331e+00 -5.80668807e-01
7.40220919e-02 -1.08747745e+00 2.85403818e-01 4.67007786e-01
2.14944094e-01 -1.25116277e+00 -1.43891060e+00 -8.98943543e-02
-3.32231849e-01 -8.76627266e-01 -1.89442754e-01 6.19662404e-01
-8.90263021e-01 -1.16548777e+00 -3.37797850e-01 -3.51041824e-01
6.34197235e-01 -2.09980756e-02 1.23734593e+00 5.25115803e-02
-5.20264208e-01 2.86077648e-01 -3.74295384e-01 -4.41587538e-01
2.30918556e-01 1.70821905e-01 1.52494192e-01 1.30528271e-01
7.16732979e-01 -9.72070456e-01 -1.12662327e+00 2.21977547e-01
-5.77476978e-01 -6.80818483e-02 7.48926044e-01 5.23391783e-01
5.80758214e-01 4.03996706e-02 5.38652599e-01 -4.22837973e-01
8.49916160e-01 -8.20450425e-01 -2.40344077e-01 -2.36803256e-02
-9.81263518e-01 -1.74715787e-01 6.80615067e-01 -4.95892055e-02
-4.51583445e-01 -1.29767001e-01 -6.58898056e-02 -1.22506805e-01
-4.80466515e-01 1.25591069e-01 1.45630732e-01 4.52766418e-01
6.29748404e-01 5.00035584e-01 -2.43099868e-01 -3.88406605e-01
-2.32553959e-01 8.51139426e-01 4.86764044e-01 -6.08220436e-02
2.31136709e-01 3.52224112e-01 -1.98034957e-01 -1.15114546e+00
-7.79845834e-01 -8.09658766e-01 -7.74268389e-01 -7.60863498e-02
1.08335888e+00 -8.59564126e-01 -7.60028541e-01 1.72573235e-02
-7.46202409e-01 -9.11742225e-02 -3.77226442e-01 6.02685630e-01
-4.48481739e-01 2.32763544e-01 -1.80133045e-01 -8.70613456e-01
-8.95494759e-01 -1.17217231e+00 1.06345677e+00 3.69286478e-01
-9.22893047e-01 -7.64707386e-01 4.51298028e-01 1.89220622e-01
2.75025308e-01 -1.43831134e-01 8.65119696e-01 -8.50240707e-01
9.65078697e-02 5.22687845e-02 1.38052270e-01 1.22778110e-01
2.61283040e-01 -1.57772109e-01 -1.39045393e+00 -1.54871523e-01
4.74931210e-01 -5.40642487e-03 6.09872401e-01 5.64853489e-01
1.08496082e+00 -1.59463167e-01 -3.67180035e-02 6.78762496e-01
1.07903421e+00 3.14270377e-01 6.45339131e-01 3.29942256e-01
1.23165302e-01 4.56414968e-01 1.00552864e-01 5.87426364e-01
3.90364259e-01 4.61264163e-01 1.90935344e-01 -8.46898332e-02
3.14338446e-01 3.49092990e-01 4.16304797e-01 7.63525844e-01
-2.27674171e-01 2.43145049e-01 -9.44925845e-01 5.70506155e-01
-1.55381513e+00 -7.55773246e-01 -1.17881782e-01 2.00465775e+00
5.31537950e-01 2.14416355e-01 4.46532011e-01 4.72565383e-01
2.72121549e-01 -1.10388026e-01 -4.35890883e-01 -4.48758900e-01
2.92093962e-01 7.89202631e-01 -1.10994823e-01 7.52174705e-02
-8.81140411e-01 3.26987058e-01 6.62312317e+00 3.24015617e-01
-9.53467965e-01 2.31494755e-01 4.84535456e-01 -6.33587897e-01
1.48096964e-01 -2.39767656e-01 -3.60237002e-01 8.91240418e-01
1.72816241e+00 -2.22283155e-01 6.81562185e-01 4.52412218e-01
7.27958441e-01 -2.06888527e-01 -1.02869606e+00 1.22035217e+00
1.45975560e-01 -1.06214428e+00 -3.70939463e-01 -5.30103445e-02
2.90098786e-01 8.72154012e-02 -8.93830284e-02 1.01468720e-01
-5.06761491e-01 -1.07833564e+00 7.40673184e-01 7.93571413e-01
8.02069485e-01 -7.40111530e-01 8.80913913e-01 9.40102786e-02
-1.12431848e+00 -2.23340452e-01 -1.56651527e-01 -9.96354967e-02
-2.41114467e-01 8.45046818e-01 -7.74771035e-01 5.18367469e-01
1.14861739e+00 1.04279923e+00 -1.04534233e+00 1.21327138e+00
-4.24994886e-01 9.74432647e-01 -1.05489112e-01 -2.66141705e-02
-1.71338946e-01 -9.58935693e-02 3.11267912e-01 1.32067966e+00
3.75097871e-01 2.67429858e-01 -4.52241689e-01 1.06798232e+00
3.15750271e-01 2.52856757e-03 -2.78474241e-01 1.20775089e-01
2.73919165e-01 1.75918424e+00 -1.15068316e+00 -1.11421362e-01
-2.08745956e-01 8.12213719e-01 1.04032472e-01 3.47878896e-02
-6.84120417e-01 -4.43937540e-01 5.09436846e-01 -5.96399009e-02
2.69562285e-02 9.80004370e-02 -4.19498324e-01 -1.09112227e+00
-5.97163029e-02 -7.81073570e-01 4.01584595e-01 -8.66960168e-01
-1.12746453e+00 1.09039247e+00 -1.90866347e-02 -1.11732602e+00
-1.01794444e-01 -2.08737418e-01 -8.28028977e-01 8.99414837e-01
-1.25059187e+00 -6.13463998e-01 -4.78313565e-01 6.47556484e-01
5.83056092e-01 2.41549071e-02 1.15321672e+00 2.59469897e-01
-8.04578543e-01 3.29582840e-01 -3.10697615e-01 -2.97680140e-01
3.37473392e-01 -1.49536538e+00 2.98034281e-01 7.15882659e-01
2.01041520e-01 1.03796577e+00 5.54834545e-01 -2.44781703e-01
-1.03589642e+00 -7.98423290e-01 1.01988316e+00 -3.93147558e-01
5.99126279e-01 -4.15649056e-01 -6.86397851e-01 3.31410736e-01
3.78791898e-01 -3.00821871e-01 1.36468971e+00 2.95087487e-01
1.91760942e-01 -3.43615741e-01 -9.91032302e-01 3.10067646e-02
6.72914922e-01 -5.66846550e-01 -9.82541323e-01 2.28903651e-01
-3.33802812e-02 -7.84835741e-02 -9.25602019e-01 -4.98920679e-02
6.29419327e-01 -1.33997691e+00 6.07832372e-01 -2.50535995e-01
3.97703320e-01 -1.92160174e-01 2.02150524e-01 -1.40367115e+00
-6.58848763e-01 -8.40155065e-01 -1.47207782e-01 8.15629363e-01
3.72236341e-01 -4.12809014e-01 2.50139058e-01 3.63936961e-01
-3.96050781e-01 -1.04139173e+00 -9.20660138e-01 -2.06825495e-01
-5.78436792e-01 -4.34567571e-01 5.50159395e-01 4.96995091e-01
5.60908318e-01 6.70302808e-01 -7.58829862e-02 1.17594883e-01
2.34126210e-01 2.27519140e-01 8.19939673e-02 -1.39852154e+00
-3.65104973e-02 -4.57981795e-01 -5.63519657e-01 -3.14990819e-01
-7.19674936e-05 -1.01434755e+00 -5.67456111e-02 -1.55059326e+00
5.78087091e-01 1.36298984e-01 -9.82652903e-01 6.35109007e-01
-2.71406502e-01 4.00354236e-01 -1.60631627e-01 2.55458921e-01
-6.91936135e-01 5.18691719e-01 4.99395669e-01 1.69170320e-01
-4.23157275e-01 3.47455256e-02 -9.34902370e-01 8.38252664e-01
9.62243676e-01 -5.06834984e-01 -6.72312737e-01 -3.06306574e-02
9.67625678e-02 -1.47937253e-01 2.82858521e-01 -1.41107666e+00
1.22875106e-02 3.34287316e-01 6.82262242e-01 -4.54022944e-01
2.22966433e-01 -7.19388843e-01 2.48582900e-01 4.82519507e-01
-3.59950274e-01 4.44116145e-01 3.69317651e-01 3.37961763e-01
-6.45319745e-02 1.39250636e-01 8.01618934e-01 5.36462627e-02
-2.81246841e-01 -3.26557681e-02 -5.70346951e-01 1.70064233e-02
6.26302183e-01 -2.09790006e-01 -1.64106086e-01 -6.47250935e-02
-9.93565559e-01 6.34232536e-02 1.19537607e-01 3.13461930e-01
6.18090749e-01 -9.40222442e-01 -4.89847690e-01 7.60836542e-01
1.20366141e-01 -5.20923197e-01 4.29774851e-01 1.44666326e+00
-1.55216098e-01 6.42574668e-01 -6.41844511e-01 -7.76876509e-01
-1.16185534e+00 3.10669959e-01 2.50094682e-01 -3.51242721e-01
-8.18986833e-01 6.87356412e-01 -1.51881687e-02 4.27888811e-01
-6.43999726e-02 -8.21313083e-01 -6.75804675e-01 3.63914639e-01
8.18117321e-01 5.23063958e-01 6.65235579e-01 -5.80597281e-01
-7.67619252e-01 4.93428588e-01 1.87654734e-01 3.08267176e-02
1.65363479e+00 -1.14841737e-01 -4.04273689e-01 7.02332318e-01
9.08889592e-01 -7.64038116e-02 -8.67377102e-01 5.97688556e-01
9.34630334e-02 -4.94968928e-02 8.98323059e-02 -9.32261169e-01
-1.00529361e+00 9.43794012e-01 1.03095102e+00 5.28480172e-01
1.58944273e+00 -1.64974391e-01 5.84601283e-01 -2.64827684e-02
1.67907298e-01 -7.95386970e-01 -3.46878529e-01 6.82949945e-02
6.37153745e-01 -7.52472401e-01 2.71224529e-01 2.39202708e-01
-5.79828560e-01 1.19630158e+00 4.92017716e-02 -2.01497853e-01
7.71412849e-01 1.31161530e-02 -1.17937766e-01 -7.01665640e-01
-8.86136115e-01 -2.90999897e-02 8.49813759e-01 4.37300056e-01
5.04352868e-01 7.62823373e-02 -6.58610165e-01 1.30278289e+00
-7.39519536e-01 1.06793471e-01 2.33153135e-01 5.05734980e-01
-3.83470833e-01 -7.46117771e-01 -5.58552220e-02 6.93423331e-01
-7.13554442e-01 -1.38446599e-01 -3.58474493e-01 4.61191416e-01
2.81573445e-01 1.39183116e+00 2.14865461e-01 -6.91019356e-01
2.86264211e-01 3.99907082e-01 3.78558457e-01 -7.46090353e-01
-1.08392406e+00 1.98814943e-01 -2.43764088e-01 -8.33627343e-01
-5.67060828e-01 -6.23967350e-01 -1.32937038e+00 3.32176715e-01
3.79514359e-02 6.75244212e-01 4.86805320e-01 9.06183243e-01
7.72338569e-01 8.80119383e-01 5.51195800e-01 -9.11056697e-01
3.32958214e-02 -1.20751810e+00 -6.75581038e-01 3.44936252e-01
6.98887169e-01 -5.65422535e-01 -5.18159866e-01 2.54535943e-01] | [13.49147891998291, 3.5084829330444336] |
f3e57e61-d886-4e74-92de-44ae69d772b5 | supervised-knowledge-may-hurt-novel-class | 2306.03648 | null | https://arxiv.org/abs/2306.03648v1 | https://arxiv.org/pdf/2306.03648v1.pdf | Supervised Knowledge May Hurt Novel Class Discovery Performance | Novel class discovery (NCD) aims to infer novel categories in an unlabeled dataset by leveraging prior knowledge of a labeled set comprising disjoint but related classes. Given that most existing literature focuses primarily on utilizing supervised knowledge from a labeled set at the methodology level, this paper considers the question: Is supervised knowledge always helpful at different levels of semantic relevance? To proceed, we first establish a novel metric, so-called transfer flow, to measure the semantic similarity between labeled/unlabeled datasets. To show the validity of the proposed metric, we build up a large-scale benchmark with various degrees of semantic similarities between labeled/unlabeled datasets on ImageNet by leveraging its hierarchical class structure. The results based on the proposed benchmark show that the proposed transfer flow is in line with the hierarchical class structure; and that NCD performance is consistent with the semantic similarities (measured by the proposed metric). Next, by using the proposed transfer flow, we conduct various empirical experiments with different levels of semantic similarity, yielding that supervised knowledge may hurt NCD performance. Specifically, using supervised information from a low-similarity labeled set may lead to a suboptimal result as compared to using pure self-supervised knowledge. These results reveal the inadequacy of the existing NCD literature which usually assumes that supervised knowledge is beneficial. Finally, we develop a pseudo-version of the transfer flow as a practical reference to decide if supervised knowledge should be used in NCD. Its effectiveness is supported by our empirical studies, which show that the pseudo transfer flow (with or without supervised knowledge) is consistent with the corresponding accuracy based on various datasets. Code is released at https://github.com/J-L-O/SK-Hurt-NCD | ['Haojin Yang', 'Christoph Meinel', 'Di Hu', 'Ben Dai', 'Jona Otholt', 'Ziyun Li'] | 2023-06-06 | null | null | null | null | ['novel-class-discovery', 'novel-class-discovery', 'semantic-textual-similarity', 'semantic-similarity'] | ['computer-vision', 'methodology', 'natural-language-processing', 'natural-language-processing'] | [ 2.28091583e-01 1.48852915e-01 -4.00151044e-01 -6.97544873e-01
-3.62858951e-01 -6.36695445e-01 6.19578123e-01 1.98679954e-01
-3.25318009e-01 7.14402318e-01 2.80080196e-02 -1.91253617e-01
-4.12304699e-01 -7.64801741e-01 -5.63413143e-01 -5.26867151e-01
1.99698746e-01 1.53347760e-01 3.28744441e-01 -3.53604904e-03
4.45467502e-01 3.95236403e-01 -1.95125115e+00 1.89376742e-01
8.59617233e-01 1.01590383e+00 8.17864463e-02 -1.74274072e-02
-2.90808320e-01 6.78547740e-01 -4.04913247e-01 -4.36956912e-01
5.43052614e-01 -6.31220460e-01 -1.15771031e+00 1.44192472e-01
3.80240440e-01 -3.03785745e-02 5.48017547e-02 1.24782944e+00
2.17219889e-01 -3.51990154e-03 6.90195918e-01 -1.41063225e+00
-5.79138815e-01 5.83890438e-01 -3.29407960e-01 2.15900525e-01
2.75826693e-01 -3.22407894e-02 1.08774233e+00 -8.83553863e-01
7.15127468e-01 1.06225765e+00 7.18023181e-01 3.59976113e-01
-1.06136632e+00 -8.17243934e-01 3.34134907e-01 3.82356584e-01
-1.60911012e+00 -2.14485601e-01 8.62886369e-01 -2.94383168e-01
4.03271765e-01 1.31847009e-01 3.85061562e-01 8.76763880e-01
-3.05697620e-01 7.50951648e-01 1.51797378e+00 -6.02858901e-01
3.79840642e-01 5.87792218e-01 5.90004563e-01 5.75804114e-01
5.44317722e-01 1.60874590e-01 -4.13647354e-01 7.90829137e-02
3.29339385e-01 -1.07453940e-02 -2.04557523e-01 -6.07467294e-01
-1.02401853e+00 7.76871622e-01 4.36113745e-01 6.55083895e-01
-1.24851383e-01 -4.65674311e-01 4.69461143e-01 4.36577410e-01
4.54069942e-01 5.93431473e-01 -5.37485003e-01 1.03185259e-01
-8.56551051e-01 -1.95350591e-02 8.73259604e-01 1.04659128e+00
1.15263319e+00 -2.27393582e-01 -6.08861297e-02 7.23666310e-01
2.37449050e-01 2.63421476e-01 6.67521000e-01 -7.92916417e-01
1.36197297e-04 8.70011926e-01 -8.56601745e-02 -1.05163300e+00
-3.40578675e-01 -5.25779545e-01 -5.34963429e-01 4.07981351e-02
4.05580580e-01 6.25093356e-02 -7.98583925e-01 1.85497522e+00
4.73165393e-01 4.23022479e-01 3.21555495e-01 8.27943444e-01
9.03364360e-01 1.29532382e-01 -3.53465937e-02 -2.94190168e-01
1.16708755e+00 -8.84810686e-01 -6.81010067e-01 7.95615166e-02
8.63180041e-01 -6.42210424e-01 1.28254175e+00 1.90254495e-01
-3.85327697e-01 -6.71229303e-01 -1.11233342e+00 2.56258190e-01
-6.92565322e-01 3.25167961e-02 6.09194815e-01 8.28610659e-01
-9.34966505e-01 4.91423160e-01 -4.64599878e-01 -8.47290933e-01
3.65892082e-01 1.38644278e-01 -3.19747508e-01 -2.73340732e-01
-1.21842217e+00 8.82326126e-01 9.11331356e-01 -9.43161771e-02
-7.38624334e-01 -6.57864988e-01 -6.23768449e-01 -5.68072274e-02
5.88856399e-01 -5.14967859e-01 1.02252877e+00 -1.34982014e+00
-1.18157959e+00 8.97348344e-01 -7.23453388e-02 -4.04744029e-01
3.64953130e-01 -3.44715826e-02 -5.87422609e-01 2.93868393e-01
3.49357098e-01 7.59942234e-01 5.57549477e-01 -1.60550940e+00
-7.88158476e-01 -4.07707036e-01 3.62672031e-01 3.21626008e-01
-6.20165527e-01 -2.73322433e-01 -1.89355612e-01 -5.86320162e-01
2.28393033e-01 -9.37219203e-01 8.23850036e-02 -2.15897411e-01
-4.55251157e-01 -3.16415399e-01 1.08361268e+00 -1.63871169e-01
1.12980950e+00 -2.14098763e+00 -4.84883279e-01 4.17216986e-01
1.87869102e-01 3.26828957e-01 -1.62972342e-02 3.00234735e-01
-2.92110175e-01 1.74516171e-01 -4.25940514e-01 1.23137221e-01
-8.44295025e-02 2.70625442e-01 -1.01220533e-01 1.81690082e-01
5.98317944e-02 7.51525104e-01 -1.11014950e+00 -5.43540180e-01
1.30227119e-01 1.14368517e-02 -2.17366114e-01 1.29451767e-01
1.48870125e-02 3.70588809e-01 -4.19096529e-01 7.85524905e-01
7.64197171e-01 -3.99774104e-01 3.06947410e-01 -4.64567035e-01
2.03101356e-02 -6.22082651e-02 -1.19701803e+00 1.53454185e+00
-8.47980008e-02 2.86425799e-01 -4.35793638e-01 -1.44231462e+00
1.22650349e+00 1.81926623e-01 4.09893334e-01 -5.17937303e-01
1.73417464e-01 3.58409494e-01 2.36762520e-02 -4.98119384e-01
2.52404034e-01 -2.64004022e-01 2.11682200e-01 4.98318285e-01
2.45044887e-01 4.54765558e-02 2.79332340e-01 1.90223515e-01
1.10430622e+00 8.78153592e-02 4.43138957e-01 -6.30466580e-01
6.68367505e-01 2.55609274e-01 5.14502704e-01 7.99596608e-01
-4.76084083e-01 4.18224216e-01 3.01132917e-01 -1.54405192e-01
-7.50678062e-01 -9.21707332e-01 -2.05984503e-01 7.75435150e-01
5.67691088e-01 -2.91083902e-01 -7.22733021e-01 -1.25617373e+00
-5.40948063e-02 8.67533207e-01 -7.26295352e-01 -2.38585502e-01
-3.03739179e-02 -5.35408437e-01 6.10420883e-01 4.25956160e-01
9.38078284e-01 -9.10270691e-01 -4.10595685e-01 -2.02113539e-01
-1.03825428e-01 -1.21476889e+00 -9.04113427e-02 1.64056689e-01
-9.47833121e-01 -1.57681239e+00 -4.35462385e-01 -8.81053150e-01
9.54945982e-01 6.57174587e-01 9.65281367e-01 1.27416655e-01
5.51550202e-02 5.92742980e-01 -8.79956484e-01 -3.53801340e-01
-3.65541875e-01 2.11666018e-01 9.49389115e-02 1.64479300e-01
7.75105119e-01 -6.05858505e-01 -4.18507725e-01 6.62901461e-01
-1.22080314e+00 8.71967617e-03 5.61794996e-01 7.91685760e-01
4.99655843e-01 3.31025451e-01 9.44665670e-01 -1.20226133e+00
4.47047353e-01 -7.55718827e-01 -3.78791064e-01 4.08700734e-01
-1.06372082e+00 2.09669843e-02 5.92152059e-01 -3.17086399e-01
-1.27771473e+00 -3.92217301e-02 9.72675234e-02 -5.10333180e-01
-4.92418140e-01 6.18492365e-01 -3.25148225e-01 5.36825471e-02
7.03149259e-01 1.65441513e-01 1.14198253e-02 -4.76954222e-01
4.41760093e-01 7.29393065e-01 3.47027183e-01 -7.17329443e-01
8.45016718e-01 6.88460708e-01 -1.21938884e-01 -3.92393559e-01
-1.29549098e+00 -5.35602093e-01 -9.75897074e-01 -1.46077767e-01
5.88619113e-01 -7.33092308e-01 -3.91367108e-01 2.19625980e-01
-6.30437016e-01 -3.96179408e-02 -5.08015215e-01 7.27216780e-01
-4.11286265e-01 5.14468551e-01 -2.40251392e-01 -5.41224062e-01
-1.34731293e-01 -9.38079834e-01 4.93950069e-01 2.12827623e-01
-8.68921205e-02 -1.20485222e+00 -1.89865202e-01 4.69467163e-01
1.36115074e-01 1.39544234e-01 1.03845417e+00 -1.16717386e+00
-2.40377486e-01 -4.41449210e-02 -4.33987767e-01 7.14713573e-01
5.18357873e-01 -1.77882761e-01 -1.06776810e+00 -1.91383466e-01
1.70823723e-01 -3.18322688e-01 7.38112092e-01 -2.74192113e-02
1.12508512e+00 -9.30647328e-02 -2.34138444e-01 2.03355640e-01
1.58510077e+00 2.20441580e-01 3.40864629e-01 4.78611827e-01
6.20964766e-01 7.88087368e-01 1.09629714e+00 2.97878981e-01
3.64236653e-01 4.82903749e-01 3.54153305e-01 -6.13970086e-02
-1.76420346e-01 -3.48396301e-01 1.93242833e-01 8.84548604e-01
7.60052577e-02 -3.99148539e-02 -9.03873146e-01 5.91443717e-01
-1.86612821e+00 -7.58716822e-01 1.90402493e-02 2.28170109e+00
8.56634617e-01 2.61083156e-01 -2.65312335e-03 3.48149866e-01
1.01237535e+00 -2.01328292e-01 -6.35338426e-01 -2.59157717e-01
-1.60544544e-01 1.44260257e-01 4.15350258e-01 1.77059069e-01
-1.00792849e+00 9.74202394e-01 5.96321678e+00 1.05957508e+00
-1.11262846e+00 1.41043678e-01 4.75241989e-01 4.02209610e-01
-3.90909404e-01 3.02751958e-01 -7.53026605e-01 3.19915414e-01
7.29976714e-01 -4.69363332e-01 -8.22459310e-02 8.93176019e-01
1.85101360e-01 -2.06842005e-01 -1.27994907e+00 7.57553279e-01
2.34601647e-01 -1.12479520e+00 2.27469802e-01 -1.24982841e-01
8.39672565e-01 -3.05346042e-01 -4.78351861e-02 3.42208356e-01
3.37614834e-01 -6.10813856e-01 4.53456998e-01 2.23465651e-01
5.97187817e-01 -6.08370364e-01 9.63579714e-01 4.24304992e-01
-1.09792352e+00 -5.79595231e-02 -2.42695048e-01 -5.22152595e-02
-3.33604127e-01 6.88044786e-01 -1.22218478e+00 9.89657342e-01
7.17574954e-01 1.00577927e+00 -8.02537918e-01 8.82468939e-01
-4.26457196e-01 8.40050817e-01 -7.17474818e-02 2.19810009e-01
2.84974426e-01 -1.32075414e-01 3.45899284e-01 1.05568206e+00
2.44254708e-01 2.02257503e-02 2.10608274e-01 8.22267354e-01
-1.86882913e-01 3.52265745e-01 -7.45838225e-01 1.49687544e-01
6.39832437e-01 1.09674275e+00 -1.08398533e+00 -6.61804378e-01
-5.81560791e-01 6.97320282e-01 9.15989950e-02 3.68589371e-01
-7.44497538e-01 -2.51087904e-01 2.66943455e-01 7.12387264e-02
2.63161242e-01 1.81781307e-01 -4.80409294e-01 -1.06230748e+00
-4.64084335e-02 -7.01699197e-01 6.61641717e-01 -7.20048010e-01
-1.58597469e+00 5.06489336e-01 3.30630988e-01 -1.53832161e+00
1.08727448e-01 -3.78554255e-01 -2.50342876e-01 4.43224162e-01
-1.70933831e+00 -1.01523745e+00 -5.94141662e-01 7.81206846e-01
5.53896964e-01 -1.33888468e-01 8.15589726e-01 3.07970136e-01
-3.86880368e-01 7.36032605e-01 2.02466697e-02 1.04715906e-01
9.25561845e-01 -1.13052201e+00 -1.92264125e-01 8.97264123e-01
4.85257208e-02 6.92811489e-01 5.32767653e-01 -7.06367493e-01
-9.42005396e-01 -1.20656598e+00 6.40854657e-01 -2.93921202e-01
5.50536692e-01 -6.41705515e-03 -1.08425426e+00 5.46854496e-01
-5.19426353e-02 9.40713584e-02 8.94064426e-01 7.21492469e-02
-6.28849030e-01 -2.36199573e-01 -1.50053847e+00 3.21335554e-01
1.21427512e+00 -4.00564790e-01 -9.09146369e-01 2.30140746e-01
8.56494069e-01 1.39621228e-01 -9.00721252e-01 8.35067213e-01
4.20718849e-01 -1.04270947e+00 5.88272095e-01 -4.33168173e-01
3.59786481e-01 -5.59619784e-01 -3.81333053e-01 -1.11054707e+00
-8.96489620e-02 1.35415018e-01 1.45584092e-01 1.35234511e+00
3.88238579e-01 -7.62579024e-01 7.92689085e-01 4.40176427e-01
-1.51050419e-01 -6.45662189e-01 -5.60994148e-01 -1.24040937e+00
-4.29491401e-02 -4.08933580e-01 5.74831307e-01 1.60512328e+00
-1.68067560e-01 3.11671585e-01 -7.94575177e-03 1.89705685e-01
5.61546981e-01 2.32851624e-01 7.21044719e-01 -1.29399335e+00
6.80851191e-02 -2.67587245e-01 -5.71957111e-01 -7.10386217e-01
3.46673012e-01 -1.12096798e+00 -1.32463813e-01 -1.33775556e+00
4.04396713e-01 -7.15958238e-01 -7.13636458e-01 7.87720799e-01
-2.18771920e-01 3.35193634e-01 2.09342495e-01 5.34908950e-01
-7.19726741e-01 4.94080335e-01 1.15025616e+00 -8.54385495e-02
-1.50399238e-01 -4.17273305e-02 -1.01322687e+00 9.09171820e-01
8.45135391e-01 -5.12299001e-01 -7.06214130e-01 -9.60198417e-02
-1.08119935e-01 -4.46828127e-01 2.29468644e-01 -1.06844246e+00
1.65678397e-01 -1.39898404e-01 2.83476394e-02 -4.00971502e-01
-1.56942941e-02 -1.04379618e+00 8.57278258e-02 4.54846829e-01
-4.60860580e-01 -3.39420736e-01 -4.04498503e-02 5.01564205e-01
-5.43853521e-01 -3.99680465e-01 8.19627166e-01 -1.94279090e-01
-1.16215038e+00 1.06832348e-01 1.07495099e-01 8.91951397e-02
1.21623182e+00 -6.19189918e-01 -2.31954888e-01 -1.66580424e-01
-6.69525206e-01 9.91130695e-02 5.29291987e-01 5.46177208e-01
4.43126321e-01 -1.37761378e+00 -6.06853247e-01 -1.53552899e-02
6.06553793e-01 -2.34904096e-01 6.30008727e-02 8.04290950e-01
-2.28622913e-01 3.77334774e-01 -2.11612269e-01 -6.96555316e-01
-1.26644444e+00 6.88940883e-01 1.44992948e-01 -9.10612419e-02
-3.83045852e-01 5.37916362e-01 2.23321542e-01 -5.61474681e-01
9.92758051e-02 -1.63102925e-01 -3.74425679e-01 1.57100946e-01
2.13272378e-01 3.51233453e-01 2.47439325e-01 -5.24111271e-01
-3.91511500e-01 4.26656276e-01 -1.03513993e-01 1.98630646e-01
1.08144104e+00 -3.11487079e-01 2.20991857e-02 5.87412298e-01
1.22212422e+00 -1.38143763e-01 -9.12481546e-01 -5.98272324e-01
3.53737921e-01 -6.30343497e-01 -1.59199908e-01 -8.53514552e-01
-1.07808959e+00 5.00041664e-01 8.54000390e-01 9.05046985e-02
1.32279980e+00 -2.98839319e-03 5.10205448e-01 5.04657984e-01
5.24844766e-01 -1.09120631e+00 1.25211105e-01 4.16648507e-01
4.88092095e-01 -1.48821437e+00 7.11029619e-02 -8.12071204e-01
-7.51933277e-01 9.68226314e-01 8.80338550e-01 1.35903165e-01
8.73567939e-01 -1.44667685e-01 2.53141284e-01 -2.59022862e-01
-4.05981839e-01 -5.38216293e-01 1.87406301e-01 5.39270937e-01
3.07137996e-01 2.90478710e-02 -7.38865137e-01 5.75134456e-01
-1.45460516e-01 1.82190046e-01 4.67849016e-01 1.06899893e+00
-5.18808722e-01 -1.25980878e+00 -2.97634900e-01 5.10301530e-01
-1.66886732e-01 -4.27864976e-02 -4.73503023e-01 8.47803473e-01
4.46001440e-01 1.15074706e+00 -1.96028784e-01 -5.39280713e-01
2.51224190e-01 9.84323472e-02 2.17845768e-01 -8.61928463e-01
-4.27905828e-01 -2.79967546e-01 -5.02523892e-02 -4.04414833e-01
-1.08767521e+00 -3.58261347e-01 -1.26576960e+00 -1.16456263e-01
-4.79860097e-01 4.73958582e-01 5.59659660e-01 1.11257136e+00
3.93360615e-01 2.11304590e-01 8.75714362e-01 -2.22904056e-01
-4.40561622e-01 -9.18797433e-01 -6.39038861e-01 7.86021590e-01
-1.82355657e-01 -9.42255676e-01 -6.26123309e-01 1.96660534e-01] | [9.657169342041016, 3.0069901943206787] |
db06ccb1-fe5e-4a45-afd2-aad89842ca30 | performance-comparison-of-3d-correspondence | 1909.00866 | null | https://arxiv.org/abs/1909.00866v1 | https://arxiv.org/pdf/1909.00866v1.pdf | Performance comparison of 3D correspondence grouping algorithm for 3D plant point clouds | Plant Phenomics can be used to monitor the health and the growth of plants. Computer vision applications like stereo reconstruction, image retrieval, object tracking, and object recognition play an important role in imaging based plant phenotyping. This paper offers a comparative evaluation of some popular 3D correspondence grouping algorithms, motivated by the important role that they can play in tasks such as model creation, plant recognition and identifying plant parts. Another contribution of this paper is the extension of 2D maximum likelihood matching to 3D Maximum Likelihood Estimation Sample Consensus (MLEASAC). MLESAC is efficient and is computationally less intense than 3D random sample consensus (RANSAC). We test these algorithms on 3D point clouds of plants along with two standard benchmarks addressing shape retrieval and point cloud registration scenarios. The performance is evaluated in terms of precision and recall. | ['Shiva Azimi', 'Tapan K. Gandhi'] | 2019-09-02 | null | null | null | null | ['plant-phenotyping'] | ['computer-vision'] | [ 3.06379586e-01 -5.85673034e-01 -2.08166465e-01 -3.49004716e-02
-4.24966753e-01 -1.08591449e+00 5.14437735e-01 6.60463572e-01
1.99924991e-01 1.27712395e-02 -5.05882084e-01 -4.78023648e-01
-4.31160480e-01 -8.11384201e-01 -2.23509416e-01 -7.62239099e-01
1.16207730e-02 1.02301049e+00 2.95223176e-01 -2.73338184e-02
4.36183602e-01 1.57421410e+00 -1.45908749e+00 -7.05459639e-02
5.95381677e-01 9.28127944e-01 7.77275681e-01 7.01946259e-01
-1.36361673e-01 -2.66219139e-01 -2.81846493e-01 9.70540345e-02
3.59483540e-01 8.38334039e-02 -6.70914173e-01 7.21200883e-01
5.75373888e-01 -5.58047332e-02 6.48817003e-01 9.90219831e-01
5.15269399e-01 -2.05302283e-01 6.95272267e-01 -1.51289010e+00
-5.97186089e-01 -1.69882700e-02 -9.40247118e-01 -4.71550316e-01
4.77122188e-01 2.14007753e-03 6.56070709e-01 -1.03334522e+00
6.73805356e-01 1.27915072e+00 8.94056678e-01 -1.35445401e-01
-1.81803977e+00 -4.21014726e-01 -1.73708335e-01 1.50482193e-01
-1.56845868e+00 -1.23090565e-01 4.97432381e-01 -6.65473342e-01
4.62369204e-01 6.91736341e-01 5.84313095e-01 8.73953551e-02
-5.01246043e-02 6.02857828e-01 8.71551692e-01 -4.32755679e-01
3.47731292e-01 -4.23971951e-01 -4.55779731e-02 5.89771986e-01
2.12026343e-01 2.96818942e-01 -3.40996981e-02 -7.10603654e-01
9.77134466e-01 -3.34442481e-02 -3.33147168e-01 -9.10056233e-01
-1.24737275e+00 6.39312387e-01 5.43833792e-01 1.97419077e-01
-4.60695177e-01 -2.18842596e-01 1.93360776e-01 5.43473773e-02
3.74043107e-01 4.60859478e-01 -5.94636261e-01 4.27380800e-01
-9.69043791e-01 1.62500203e-01 7.27383614e-01 1.22133911e+00
5.01900733e-01 -1.48332238e-01 1.13875598e-01 8.36165547e-01
4.44309831e-01 7.84438252e-01 2.50490662e-02 -1.28995848e+00
-4.90475804e-01 7.91372240e-01 8.44688416e-02 -1.19982350e+00
-1.05818756e-01 1.25507899e-02 -9.18556869e-01 6.57168388e-01
1.75953478e-01 6.69069886e-01 -7.13878810e-01 9.40095067e-01
5.67666292e-01 3.60225707e-01 -2.84185857e-01 5.90816855e-01
9.86942470e-01 4.43479180e-01 -6.16685636e-02 -2.69185096e-01
1.33602798e+00 -2.93810070e-01 -2.34345883e-01 1.24123164e-01
5.27118862e-01 -1.46573293e+00 3.23182583e-01 3.64983320e-01
-8.19801450e-01 -5.05576134e-01 -5.08452594e-01 1.51524812e-01
-3.51714581e-01 4.64991182e-01 5.63033223e-01 6.16365135e-01
-9.85005856e-01 7.11102664e-01 -6.43453658e-01 -6.84050679e-01
5.45348763e-01 3.86734635e-01 -6.12902641e-01 2.07102168e-02
2.28298843e-01 7.82532811e-01 4.42653149e-01 4.86666486e-02
-5.31917036e-01 -8.31612825e-01 -4.82788086e-01 4.95599909e-03
2.01395705e-01 -3.78144652e-01 9.31425452e-01 -2.40025476e-01
-1.51377809e+00 1.52101457e+00 -3.05111259e-01 -1.45662025e-01
-1.29904412e-03 1.31969079e-01 2.74208225e-02 -1.25244960e-01
2.16589466e-01 7.63439476e-01 4.45229471e-01 -1.38437510e+00
-4.07620221e-01 -9.01901543e-01 -7.69054592e-01 -5.50790466e-02
3.71219963e-01 2.91189462e-01 -1.81928694e-01 -3.45785737e-01
1.06920266e+00 -1.28736639e+00 -2.38924980e-01 8.65346849e-01
-4.25880760e-01 1.54446706e-01 1.43269324e+00 -6.80411339e-01
1.57733947e-01 -1.76224887e+00 1.61417842e-01 2.74548173e-01
-7.06162900e-02 4.97524142e-01 -2.91482866e-01 3.53909194e-01
-3.61481398e-01 1.22314617e-02 -3.82761836e-01 -6.68619573e-02
-3.95008773e-01 2.79896200e-01 -3.45886856e-01 6.78948164e-01
1.00635685e-01 8.72782230e-01 -6.82803154e-01 -3.32638234e-01
7.89110363e-01 3.31513047e-01 6.83284625e-02 -7.69520178e-02
-4.19435687e-02 5.51560283e-01 -2.59632707e-01 1.26685405e+00
1.30000579e+00 -2.02864721e-01 9.05636698e-02 -3.37182432e-01
-5.65910220e-01 -3.55912805e-01 -1.24889517e+00 1.32813108e+00
-2.91421004e-02 4.40444946e-01 4.53571469e-01 -9.10243809e-01
1.31971943e+00 4.04059231e-01 9.21887100e-01 1.64295897e-01
-1.61404625e-01 2.48985156e-01 -1.76650703e-01 3.28710407e-01
1.09872922e-01 6.09276295e-01 5.85432708e-01 6.98865131e-02
-1.30042449e-01 -9.96782124e-01 -2.59767342e-02 -6.45290613e-02
8.53054047e-01 2.00706571e-01 8.69895458e-01 -6.13415480e-01
6.33875370e-01 4.09464061e-01 3.81578118e-01 3.47805440e-01
-2.57751316e-01 6.90888405e-01 8.73658583e-02 -1.72101602e-01
-1.03194201e+00 -1.14040661e+00 -4.53159720e-01 5.29131353e-01
3.23308855e-01 -3.56244773e-01 9.12195130e-04 -2.39440531e-01
4.94197041e-01 4.75963473e-01 -7.91504383e-02 1.27741858e-01
-6.08587451e-02 -5.86693764e-01 2.73703754e-01 3.34609836e-01
3.78370762e-01 -9.15018559e-01 -6.81894720e-01 6.87513649e-02
1.68673292e-01 -1.09431338e+00 1.50791565e-02 8.29498563e-03
-1.32179463e+00 -1.16488266e+00 -6.26972198e-01 -7.89557457e-01
6.14688873e-01 9.60651815e-01 1.22600877e+00 7.62085989e-02
-7.26483703e-01 4.21524763e-01 -4.81635511e-01 -6.35622203e-01
-4.84080255e-01 -1.76394805e-01 -2.22848251e-01 -4.86567646e-01
4.24945533e-01 -7.74830461e-01 -1.69364452e-01 7.72730350e-01
-5.78004062e-01 -8.56487155e-02 2.74558395e-01 7.27973521e-01
1.11058235e+00 6.46021888e-02 -2.66473025e-01 -5.85567296e-01
1.45657942e-01 1.12232342e-02 -1.39691520e+00 6.56485558e-01
-3.28169376e-01 -4.66382533e-01 1.07527390e-01 -2.94143230e-01
-5.32347739e-01 8.23603153e-01 2.67132431e-01 -4.51711178e-01
-7.46913433e-01 4.78384703e-01 -2.99280882e-01 -6.57364190e-01
7.43083000e-01 6.37377575e-02 2.88964748e-01 -6.28473639e-01
2.09243700e-01 4.63151336e-01 7.62762010e-01 -4.81718719e-01
1.12264323e+00 7.52499282e-01 7.34641016e-01 -1.29954648e+00
-3.96172017e-01 -1.10944664e+00 -1.13065350e+00 -2.04007521e-01
5.06121099e-01 -5.76995492e-01 -1.18912184e+00 6.81773186e-01
-1.32321000e+00 3.93834151e-02 -2.97755718e-01 4.45254296e-01
-6.61165595e-01 5.53260028e-01 -2.59538651e-01 -7.02411652e-01
-4.79645312e-01 -1.11353660e+00 1.70145500e+00 1.51200965e-01
-2.26151850e-02 -7.68939853e-01 -6.10845909e-02 9.94425733e-03
1.25463426e-01 5.19640267e-01 7.66538918e-01 -2.28038400e-01
-5.21998405e-01 -5.52117705e-01 -5.49034238e-01 -2.04727337e-01
4.16567028e-01 9.20273542e-01 -8.92975569e-01 -3.05558652e-01
-2.42036089e-01 3.58873457e-02 1.93127200e-01 8.84876490e-01
1.02592707e+00 5.05557299e-01 -7.20230103e-01 4.76839155e-01
1.58731997e+00 2.36770585e-01 5.57214856e-01 -5.70370594e-06
6.59497082e-01 8.39522123e-01 8.28982294e-01 4.44820881e-01
-2.32052803e-01 8.80058050e-01 8.68558109e-01 -8.64992887e-02
5.97759336e-02 2.52785385e-01 -3.37775558e-01 5.16301870e-01
-2.19744623e-01 5.85382506e-02 -1.24565852e+00 2.44946837e-01
-1.79176688e+00 -1.13959277e+00 -8.91091406e-01 2.32756233e+00
2.64305055e-01 -5.53542793e-01 -1.53929859e-01 2.04995513e-01
1.01164329e+00 -3.65147173e-01 -5.08123517e-01 -6.22677989e-02
-4.34040666e-01 2.61490643e-01 9.10619676e-01 2.48233318e-01
-1.13536894e+00 9.77813184e-01 6.42478895e+00 5.87159455e-01
-9.00139093e-01 -2.53122419e-01 3.07546854e-01 8.21454465e-01
2.60423392e-01 5.47568142e-01 -7.24694490e-01 -1.43253384e-02
3.40109229e-01 8.53285380e-03 2.16912761e-01 1.03360522e+00
2.93004721e-01 -4.12615091e-01 -8.61679792e-01 1.12210917e+00
-2.80829787e-01 -1.38717377e+00 -2.23626271e-01 4.25686538e-01
5.96896589e-01 4.71641421e-02 -2.29221910e-01 -4.66239154e-01
5.99543214e-01 -7.78821766e-01 3.60722274e-01 1.90333411e-01
7.30509043e-01 -3.85500312e-01 3.58162880e-01 5.38999021e-01
-1.64702606e+00 2.36197934e-01 -8.53487611e-01 4.14352357e-01
2.20032781e-01 8.78303826e-01 -1.12019134e+00 8.12572956e-01
7.02359200e-01 8.27829063e-01 -7.49831557e-01 1.64567542e+00
1.06010593e-01 1.56012744e-01 -5.92861891e-01 5.50126553e-01
-1.75636455e-01 -6.38005912e-01 7.70665824e-01 6.13556564e-01
7.21897542e-01 4.83470559e-02 5.94534338e-01 9.84369576e-01
3.97718072e-01 1.46117255e-01 -6.97035193e-01 2.23918771e-03
8.21036339e-01 1.39459229e+00 -1.26377785e+00 5.34278154e-02
9.97345150e-02 9.38431501e-01 -3.42126429e-01 -4.26440269e-01
-1.63210720e-01 4.94182557e-02 5.37537992e-01 -7.08704069e-02
4.60875213e-01 -5.65522075e-01 -2.74058878e-01 -5.75668991e-01
-4.71361339e-01 -6.41016603e-01 1.01950683e-01 -1.27568328e+00
-1.30782497e+00 3.68431099e-02 2.42899135e-01 -1.22754788e+00
-8.09047818e-02 -1.04495263e+00 -7.02275276e-01 1.15353537e+00
-9.74233150e-01 -1.24566519e+00 -8.06746781e-01 1.67515159e-01
4.53688502e-01 -1.64071530e-01 1.54349887e+00 -1.19796619e-01
-3.00529152e-01 -3.56562138e-01 4.78006274e-01 -3.83617967e-01
4.95255142e-01 -1.17498493e+00 6.10684335e-01 6.65955067e-01
2.70453334e-01 1.22296654e-01 5.29736638e-01 -9.05225694e-01
-1.36668658e+00 -8.73089314e-01 7.05267489e-01 -3.31790745e-01
3.06751311e-01 1.49161905e-01 -8.26181829e-01 3.46131533e-01
-3.03953886e-01 2.39167169e-01 5.79249382e-01 -1.41310334e-01
-7.69655704e-02 9.73115861e-02 -1.44976115e+00 3.97757143e-01
8.30163538e-01 -4.20777440e-01 2.37917930e-01 7.01041400e-01
1.17888622e-01 -4.32390392e-01 -1.14925241e+00 8.14572036e-01
6.32926166e-01 -6.32731199e-01 1.39782572e+00 -1.80069819e-01
-2.08721325e-01 -7.69106507e-01 -5.14203608e-01 -1.14050829e+00
-7.09507883e-01 -5.16949236e-01 3.26376438e-01 1.27691758e+00
-2.32623473e-01 -2.19489932e-01 9.89070177e-01 1.78467467e-01
-1.63237620e-02 1.50968842e-02 -5.21489263e-01 -9.62596357e-01
-2.49596417e-01 -1.16625428e-01 7.32680261e-01 1.10762560e+00
-6.89994574e-01 -1.75225005e-01 2.91431904e-01 6.41704381e-01
8.08564663e-01 6.92910075e-01 1.05295932e+00 -2.27541471e+00
-1.33614400e-02 -4.43397760e-01 -8.52873862e-01 -6.88273966e-01
2.03311220e-01 -9.90856647e-01 -1.18841901e-01 -1.35008895e+00
-1.10326232e-02 -5.00067532e-01 7.43097842e-01 2.48869628e-01
2.45789096e-01 8.18232819e-02 2.19739154e-01 5.73450863e-01
1.90995887e-01 1.06773831e-01 9.72047687e-01 -1.13482662e-01
-1.86463043e-01 5.07811308e-01 1.88009162e-02 7.94556141e-01
1.10185492e+00 -3.49763334e-01 -1.34123191e-01 -2.92661905e-01
-4.66966152e-01 1.68122873e-02 8.13037872e-01 -8.02273750e-01
9.02940705e-02 -4.12334740e-01 4.85098839e-01 -1.32366717e+00
5.75279355e-01 -1.31294525e+00 6.98815405e-01 6.59723699e-01
7.15348870e-02 3.05799663e-01 2.29884312e-01 2.65599847e-01
1.05556756e-01 -7.60514140e-01 9.98784900e-01 -2.88555354e-01
-7.95749843e-01 3.53212267e-01 -2.02952772e-01 -7.89489269e-01
1.21872520e+00 -6.69007957e-01 -2.11321637e-01 -3.67887244e-02
-6.32230163e-01 -4.60421667e-02 7.84982562e-01 3.79607379e-02
6.42863810e-01 -1.29704404e+00 -8.77966821e-01 1.80568039e-01
3.21141541e-01 -9.79375094e-03 -3.01394075e-01 7.54855156e-01
-1.14604783e+00 5.11531472e-01 -4.93240088e-01 -1.38990879e+00
-1.94791710e+00 3.92145365e-01 1.05324708e-01 8.44167452e-03
-2.91503787e-01 6.30303204e-01 -8.61956477e-02 -9.37787771e-01
1.01388972e-02 -1.36242211e-01 -1.28711805e-01 -1.67623073e-01
4.17635813e-02 6.11962318e-01 3.49224925e-01 -9.71596599e-01
-4.09093916e-01 7.91214824e-01 3.41156065e-01 2.15120688e-01
1.30551338e+00 1.83365464e-01 -4.97839659e-01 3.05512995e-01
7.39492714e-01 -3.15814525e-01 -7.07061827e-01 -1.70501098e-01
4.90462184e-01 -9.53313172e-01 1.79712594e-01 -4.32981461e-01
-8.95853639e-01 6.25566721e-01 1.01718044e+00 5.64235866e-01
9.97907639e-01 -1.48295565e-02 1.13683060e-01 3.81059736e-01
4.84935641e-01 -5.51446974e-01 -5.41153789e-01 3.68590713e-01
9.98801172e-01 -1.24909830e+00 3.64837766e-01 -1.09967160e+00
-7.60772973e-02 1.10516500e+00 3.39773417e-01 -1.69201985e-01
9.71947134e-01 3.31777483e-01 -2.72939503e-01 -2.29470760e-01
-3.85196120e-01 -4.46492493e-01 3.12108785e-01 1.19167554e+00
5.63056827e-01 2.06463724e-01 -2.55099498e-02 -5.98048747e-01
1.56339098e-04 -2.46351913e-01 1.12891205e-01 1.16516256e+00
-5.01444876e-01 -1.41274190e+00 -9.67228293e-01 3.69089544e-01
2.25398764e-01 2.31549531e-01 -8.51204932e-01 6.57815158e-01
-1.31418779e-01 6.03881955e-01 1.15192123e-01 1.31218493e-01
3.00545901e-01 -2.31493339e-01 5.79462945e-01 -7.83793509e-01
-3.89860421e-01 2.70420969e-01 -5.15767932e-01 -4.34572250e-01
-6.30716383e-01 -1.01685131e+00 -6.38245940e-01 -5.12656927e-01
-1.03205085e+00 -2.76439697e-01 1.16516685e+00 4.86549467e-01
3.93184870e-01 -5.54550625e-03 5.96195281e-01 -1.13211346e+00
-4.68423784e-01 -5.25806963e-01 -7.71084011e-01 1.61330447e-01
-3.41126263e-01 -5.22065103e-01 6.39316142e-02 3.08461905e-01] | [8.952740669250488, -1.7834653854370117] |
aeb57cbb-436a-449e-9f5a-aec420c6a87f | enriching-relation-extraction-with-openie | 2212.09376 | null | https://arxiv.org/abs/2212.09376v1 | https://arxiv.org/pdf/2212.09376v1.pdf | Enriching Relation Extraction with OpenIE | Relation extraction (RE) is a sub-discipline of information extraction (IE) which focuses on the prediction of a relational predicate from a natural-language input unit (such as a sentence, a clause, or even a short paragraph consisting of multiple sentences and/or clauses). Together with named-entity recognition (NER) and disambiguation (NED), RE forms the basis for many advanced IE tasks such as knowledge-base (KB) population and verification. In this work, we explore how recent approaches for open information extraction (OpenIE) may help to improve the task of RE by encoding structured information about the sentences' principal units, such as subjects, objects, verbal phrases, and adverbials, into various forms of vectorized (and hence unstructured) representations of the sentences. Our main conjecture is that the decomposition of long and possibly convoluted sentences into multiple smaller clauses via OpenIE even helps to fine-tune context-sensitive language models such as BERT (and its plethora of variants) for RE. Our experiments over two annotated corpora, KnowledgeNet and FewRel, demonstrate the improved accuracy of our enriched models compared to existing RE approaches. Our best results reach 92% and 71% of F1 score for KnowledgeNet and FewRel, respectively, proving the effectiveness of our approach on competitive benchmarks. | ['Martin Theobald', 'Maria Biryukov', 'Alessandro Temperoni'] | 2022-12-19 | null | null | null | null | ['open-information-extraction'] | ['natural-language-processing'] | [ 2.93233931e-01 6.75821185e-01 -4.98258203e-01 -3.00599456e-01
-7.89332092e-01 -7.26612508e-01 6.50716066e-01 6.18663609e-01
-3.71064514e-01 1.11135030e+00 5.27857363e-01 -4.68022555e-01
-8.24614763e-02 -1.12988555e+00 -8.50045025e-01 -2.61667401e-01
-2.16874480e-01 6.56206727e-01 2.25423113e-01 -3.78956169e-01
-2.02974845e-02 4.03664947e-01 -1.43429303e+00 5.32406092e-01
5.94584584e-01 1.11697900e+00 -7.67324045e-02 5.28046906e-01
-5.31528473e-01 1.16511226e+00 -5.50544977e-01 -8.23772371e-01
-2.72764176e-01 1.94904536e-01 -1.63847339e+00 -2.91191846e-01
-5.69213070e-02 2.05069929e-01 -7.04951882e-02 8.38214040e-01
-3.09872627e-03 1.75584048e-01 7.03983068e-01 -7.90873885e-01
-5.59117913e-01 1.27974427e+00 -3.42630595e-01 1.83982179e-01
6.71720445e-01 -5.02205670e-01 1.77442420e+00 -8.33016276e-01
1.09548223e+00 1.03326035e+00 5.68178654e-01 3.84128273e-01
-1.28900051e+00 -3.30185801e-01 4.10445519e-02 4.48906660e-01
-1.43859005e+00 -4.54750925e-01 5.42786598e-01 -3.87475550e-01
1.60529208e+00 4.83762592e-01 1.37686998e-01 1.06100321e+00
2.89920513e-02 1.00504279e+00 6.35865629e-01 -6.85079098e-01
6.27901778e-02 4.39570963e-01 7.40791142e-01 4.26660627e-01
5.14885247e-01 -1.66727901e-01 -3.41017544e-01 -2.33070686e-01
3.31466317e-01 -4.08585221e-01 -3.67403984e-01 -3.52592841e-02
-1.17107308e+00 8.43696058e-01 3.79319847e-01 5.48056364e-01
-3.34899276e-01 -3.35908383e-01 6.54788136e-01 1.99607626e-01
3.99523884e-01 8.49317193e-01 -1.14811611e+00 -2.65252478e-02
-4.59096879e-01 3.01284969e-01 1.41764915e+00 1.07530034e+00
8.54577661e-01 -4.65191305e-01 -3.98880690e-01 8.59105945e-01
-2.89907847e-02 1.42235920e-01 3.50031048e-01 -3.45383078e-01
9.63676631e-01 9.20801580e-01 3.61632109e-02 -9.21701849e-01
-5.22016168e-01 -3.51251692e-01 -7.82694995e-01 -8.23038399e-01
1.30640075e-01 -2.68599361e-01 -4.79781330e-01 1.52720654e+00
3.89994681e-01 -7.62149971e-03 5.56864440e-01 4.40202415e-01
1.47371781e+00 8.43947232e-01 2.15196744e-01 -4.96664643e-01
1.69354796e+00 -5.79712987e-01 -7.19556093e-01 -4.93575186e-01
6.81628823e-01 -3.37925613e-01 3.54737550e-01 1.39500335e-01
-8.08971763e-01 -3.09817046e-01 -8.78677964e-01 -4.01406646e-01
-6.49745643e-01 -6.27502576e-02 7.85293937e-01 1.63164482e-01
-4.43716019e-01 5.65266848e-01 -4.31342900e-01 -7.38159120e-02
3.66101325e-01 4.76794988e-01 -8.24950457e-01 -3.82949337e-02
-1.59663999e+00 1.06435025e+00 9.23285306e-01 1.60461664e-01
-9.39006135e-02 -5.46070218e-01 -1.38023818e+00 3.83217871e-01
9.32283342e-01 -5.28926373e-01 1.01231050e+00 -4.11181301e-01
-1.00910485e+00 9.29163694e-01 -3.95007491e-01 -7.57078171e-01
-2.82963872e-01 -3.72524738e-01 -4.30022180e-01 -8.72916728e-02
2.62616873e-02 3.47886831e-01 3.30386639e-01 -1.19280863e+00
-5.84256768e-01 -4.49644893e-01 3.19876105e-01 -5.70362695e-02
-7.72712156e-02 3.69212449e-01 -2.18839273e-01 -3.96231174e-01
6.14099950e-03 -5.70412457e-01 -1.83709189e-01 -1.00939250e+00
-1.02379060e+00 -7.89656997e-01 2.38193244e-01 -8.66610765e-01
1.57915497e+00 -2.07086349e+00 3.91350061e-01 2.70087391e-01
4.95560437e-01 3.07708055e-01 9.10630599e-02 5.64834535e-01
-3.79183650e-01 3.87210995e-01 -1.39027983e-01 2.67597474e-02
1.73289374e-01 4.57944989e-01 -5.63487470e-01 -7.01538939e-03
6.30792379e-01 1.36809540e+00 -7.45935559e-01 -7.16988921e-01
-1.51421353e-01 2.67589718e-01 -2.89420962e-01 2.10847259e-01
-5.21078587e-01 5.68234660e-02 -5.27434647e-01 5.11923194e-01
3.34818155e-01 -2.33611867e-01 4.20141220e-01 -2.55700171e-01
8.03748667e-02 1.08922267e+00 -1.28368425e+00 1.03350377e+00
-3.85219634e-01 5.40427029e-01 -2.99890518e-01 -1.12555420e+00
8.49640250e-01 4.05689985e-01 2.58803010e-01 -3.97183806e-01
2.16586500e-01 -2.81986175e-03 -1.10128947e-01 -6.28335476e-01
6.75841510e-01 -4.96642739e-02 -4.02756065e-01 1.95606276e-01
3.14632893e-01 1.28042772e-01 5.17293572e-01 3.36901993e-01
1.37486756e+00 -1.27942730e-02 1.12820542e+00 -4.40123416e-02
8.39159966e-01 -2.16451799e-03 7.12172270e-01 5.24271548e-01
2.33619928e-01 1.13430060e-01 8.11293364e-01 -3.90866756e-01
-7.84592152e-01 -7.18919396e-01 -4.83748078e-01 1.02311897e+00
-2.43276954e-01 -9.81004417e-01 -4.43002224e-01 -9.08242106e-01
3.48249674e-02 7.37292171e-01 -5.13963997e-01 8.37841332e-02
-8.68885159e-01 -6.59654677e-01 5.50199866e-01 7.33493447e-01
1.26120755e-02 -1.33596587e+00 -1.16755009e-01 3.86274636e-01
-5.37694037e-01 -1.53058684e+00 1.01982400e-01 8.37528944e-01
-4.52074587e-01 -1.24237835e+00 7.51163065e-02 -9.68413174e-01
3.06790799e-01 -4.26756382e-01 1.46934187e+00 -2.56924838e-01
9.65455398e-02 -6.29521012e-02 -7.51459122e-01 -3.95805687e-01
-4.87691671e-01 2.30462283e-01 -5.57271689e-02 -1.01478603e-02
6.86530888e-01 -4.04370964e-01 1.49739727e-01 -9.12246704e-02
-9.04980540e-01 -3.90079319e-02 6.85928643e-01 9.45459127e-01
6.62028253e-01 1.92422539e-01 2.29496211e-01 -1.45482099e+00
5.42173028e-01 -5.17531097e-01 -2.91091204e-01 5.74154258e-01
-4.99577701e-01 5.44275701e-01 4.64853376e-01 -4.40267533e-01
-9.43200767e-01 -1.60969153e-01 -2.46012494e-01 1.55101165e-01
-3.87174755e-01 1.00214064e+00 -4.04095173e-01 3.72632653e-01
5.85773468e-01 2.16601238e-01 -6.70728981e-01 -4.52262700e-01
4.50676203e-01 8.49382758e-01 5.64436853e-01 -7.98218787e-01
7.89894462e-01 -8.67303554e-03 -1.33725241e-01 -7.95327067e-01
-1.29373491e+00 -7.04768956e-01 -7.58696377e-01 4.72422779e-01
6.83293462e-01 -8.01947594e-01 -7.87615418e-01 -2.85160959e-01
-1.34247005e+00 9.75697860e-02 -4.25339103e-01 1.73955753e-01
-2.33746082e-01 1.82712883e-01 -8.52619052e-01 -7.83448875e-01
-6.01116836e-01 -6.78988159e-01 1.05680537e+00 1.92300528e-01
-6.39117539e-01 -8.78109932e-01 1.65958315e-01 5.45255482e-01
-1.38665304e-01 8.38279128e-02 1.24611890e+00 -1.39809167e+00
-3.76253873e-01 -2.62934357e-01 -3.04344207e-01 4.11057621e-01
-1.37853056e-01 -2.01109961e-01 -9.61080670e-01 1.75532997e-01
-2.44996861e-01 -3.79171997e-01 8.47446859e-01 -1.18518069e-01
8.21813822e-01 -7.34097719e-01 -5.41129589e-01 3.50813031e-01
1.50072944e+00 -2.78614629e-02 7.81863332e-01 3.37985516e-01
6.22223496e-01 6.92846060e-01 5.51783323e-01 3.17563802e-01
6.02918923e-01 5.23304701e-01 1.22571416e-01 3.05612236e-01
2.13365689e-01 -2.87496805e-01 2.75096029e-01 7.81095743e-01
-2.84138530e-01 -9.94293094e-02 -9.40082729e-01 5.76852560e-01
-1.60730970e+00 -1.13573027e+00 -2.83974349e-01 1.83377218e+00
1.53680134e+00 3.29913706e-01 -2.69241720e-01 4.86992270e-01
4.19997066e-01 1.04398191e-01 -8.32846090e-02 -6.00200355e-01
-1.82362407e-01 7.09294021e-01 3.54589760e-01 4.00400788e-01
-1.41108632e+00 1.10327995e+00 5.56609249e+00 8.17059040e-01
-6.57249749e-01 -1.38805985e-01 3.77276570e-01 4.47607756e-01
-3.16603631e-01 1.01854123e-01 -1.55975223e+00 -8.56434330e-02
1.17550349e+00 -1.93332747e-01 2.30363995e-01 9.47151661e-01
-3.94672781e-01 -1.06841959e-01 -1.60333514e+00 6.73498929e-01
7.80429840e-02 -1.63505042e+00 2.35194266e-02 5.84501354e-03
6.29189968e-01 4.91886511e-02 -6.93390131e-01 8.22406054e-01
3.41393411e-01 -9.55350280e-01 5.23213625e-01 3.37518275e-01
7.73533821e-01 -6.94807708e-01 1.26716924e+00 4.76933867e-01
-1.25745988e+00 -1.55399203e-01 -2.99973696e-01 -1.69869334e-01
-1.55733213e-01 6.16911471e-01 -9.99281824e-01 8.69954884e-01
5.16533256e-01 6.07915401e-01 -6.01496160e-01 5.46741545e-01
-5.78124523e-01 6.66613817e-01 -2.24398673e-01 -2.88004696e-01
-9.47623104e-02 1.80204868e-01 5.39386034e-01 1.47800243e+00
-5.01250207e-01 3.68200600e-01 9.87073556e-02 8.48758817e-01
-3.46392751e-01 2.60615170e-01 -4.89855081e-01 -1.57777146e-01
3.08125734e-01 1.22146404e+00 -2.89336115e-01 -3.91082495e-01
-5.13111889e-01 4.46971834e-01 7.74225950e-01 3.36557776e-01
-5.12216508e-01 -7.12104917e-01 5.76995850e-01 -1.05843358e-01
7.90328205e-01 7.79185221e-02 -2.99772501e-01 -1.39382911e+00
3.20632607e-01 -7.26581931e-01 6.29392266e-01 -4.56024855e-01
-1.35621858e+00 8.22746456e-01 5.27400970e-02 -7.09843338e-01
-5.52533746e-01 -6.46152794e-01 -2.25293532e-01 6.00258529e-01
-1.45786488e+00 -1.06752777e+00 3.59104484e-01 4.86730248e-01
2.96868861e-01 1.47398025e-01 1.15093017e+00 7.37089291e-02
-7.65987754e-01 5.77658772e-01 -7.79946819e-02 9.04552698e-01
2.85626352e-01 -1.41231823e+00 2.30718747e-01 6.64468288e-01
6.91910803e-01 8.05094182e-01 5.88724375e-01 -6.35187387e-01
-1.34170985e+00 -1.01136434e+00 1.89402139e+00 -7.51540542e-01
1.00457990e+00 -5.54123163e-01 -9.94731724e-01 1.07680535e+00
-1.45580456e-01 -1.31418899e-01 7.74158359e-01 8.66693735e-01
-5.68401396e-01 -1.36535391e-01 -7.95458019e-01 3.69250298e-01
7.80574977e-01 -6.57331288e-01 -1.29709661e+00 2.17881680e-01
1.15271389e+00 -4.80411381e-01 -1.16335499e+00 6.41901314e-01
2.39571407e-01 -4.80979562e-01 1.03775692e+00 -1.19952202e+00
6.71975613e-01 1.03852682e-01 -3.57141584e-01 -8.80137622e-01
-2.32752979e-01 -4.24172312e-01 -6.55520797e-01 1.57662940e+00
9.48928714e-01 -3.22996378e-01 4.70584005e-01 7.30349541e-01
1.33689865e-01 -9.48434651e-01 -8.20600867e-01 -4.32358027e-01
-8.37375000e-02 -8.85384142e-01 5.04188180e-01 8.20683539e-01
5.74945092e-01 1.12176049e+00 4.84873280e-02 2.39132434e-01
1.56105757e-01 4.65428174e-01 4.82957542e-01 -1.50452089e+00
-2.68622696e-01 -2.13785335e-01 -5.50991654e-01 -8.43535483e-01
5.71744323e-01 -1.06829429e+00 7.63806403e-02 -1.58100140e+00
3.09294462e-01 -4.33311582e-01 -2.09910855e-01 8.22205842e-01
-1.76211670e-01 -1.47632033e-01 -9.42567289e-02 -1.08460255e-01
-7.69432902e-01 3.39371413e-01 9.12494898e-01 -4.14998561e-01
-2.68849492e-01 1.31879151e-01 -9.86206174e-01 6.31195068e-01
2.95036554e-01 -4.36469913e-01 1.52324647e-01 -1.35933654e-02
7.36020327e-01 2.26243839e-01 9.41306874e-02 -5.10370374e-01
1.66998476e-01 -2.02909172e-01 1.02748111e-01 -4.15950388e-01
1.79257262e-02 -6.70821309e-01 -2.35926256e-01 -1.23482430e-02
-4.31192845e-01 -4.19005632e-01 8.32613558e-02 4.59866941e-01
-5.62602341e-01 -3.04079175e-01 1.26522869e-01 -9.13214833e-02
-7.33347356e-01 6.05543256e-02 -6.45486042e-02 3.22092056e-01
7.96434939e-01 2.28772938e-01 -4.16127771e-01 8.27419013e-02
-9.58258808e-01 2.24546686e-01 -3.03828597e-01 3.96937162e-01
6.66429818e-01 -9.41852629e-01 -8.17499995e-01 2.17704251e-01
2.79884577e-01 4.72018003e-01 -2.45930642e-01 6.65345252e-01
-1.06069654e-01 9.79382873e-01 3.71453047e-01 -1.64435416e-01
-1.36782122e+00 7.61361539e-01 -9.01063457e-02 -1.00024903e+00
-6.47790492e-01 1.02077770e+00 3.27101834e-02 -4.14728612e-01
3.62745255e-01 -7.32490599e-01 -8.33567023e-01 1.38042569e-01
5.99002719e-01 -4.33639027e-02 3.31007004e-01 -7.15754092e-01
-5.97644627e-01 7.52950534e-02 -2.79648751e-01 3.05411935e-01
1.72791374e+00 1.07306093e-01 -5.43427289e-01 5.43520153e-01
1.13488197e+00 2.68249124e-01 -3.17176700e-01 -6.50313377e-01
5.79705417e-01 2.33025342e-01 -2.21955985e-01 -6.97969913e-01
-5.61062276e-01 4.43394065e-01 -4.97124165e-01 3.29760373e-01
8.54458094e-01 8.70393217e-01 8.29400837e-01 8.67323339e-01
5.04033148e-01 -7.25068271e-01 -5.72303712e-01 8.35479677e-01
8.46112072e-01 -1.21651459e+00 9.00817961e-02 -7.58563399e-01
-6.24843180e-01 9.54300404e-01 3.81167948e-01 1.22964583e-01
6.75644815e-01 5.26313603e-01 -3.49007308e-01 -1.88294604e-01
-1.08622038e+00 -5.57411969e-01 6.62689865e-01 3.93000633e-01
4.91517156e-01 2.94189155e-01 -2.67208934e-01 1.35841501e+00
-5.20629406e-01 -2.96912581e-01 3.11197728e-01 8.26998234e-01
-3.51781070e-01 -1.19942415e+00 -1.87304407e-01 6.32152677e-01
-5.74216187e-01 -4.39517915e-01 -5.80943406e-01 7.81832039e-01
3.02444756e-01 9.44174528e-01 -1.13062464e-01 -4.53833938e-01
3.66682053e-01 4.29999858e-01 2.95262724e-01 -1.01396048e+00
-8.07061136e-01 -2.98508495e-01 7.97458053e-01 -5.51675141e-01
-4.62242484e-01 -6.49796247e-01 -1.35741246e+00 -1.42208859e-01
-6.80092812e-01 4.19089496e-01 3.22248012e-01 1.58663273e+00
1.44408951e-02 4.73509133e-01 3.71221960e-01 -2.17835650e-01
-4.35669512e-01 -1.08118927e+00 -5.21739483e-01 4.49352980e-01
3.45227867e-01 -4.67773438e-01 -9.88710001e-02 8.28761011e-02] | [9.397538185119629, 8.621182441711426] |
c1196b97-ce6f-4f3f-b97b-5e57a58d7922 | shrec-2021-classification-in-cryo-electron | 2203.10035 | null | https://arxiv.org/abs/2203.10035v1 | https://arxiv.org/pdf/2203.10035v1.pdf | SHREC 2021: Classification in cryo-electron tomograms | Cryo-electron tomography (cryo-ET) is an imaging technique that allows three-dimensional visualization of macro-molecular assemblies under near-native conditions. Cryo-ET comes with a number of challenges, mainly low signal-to-noise and inability to obtain images from all angles. Computational methods are key to analyze cryo-electron tomograms. To promote innovation in computational methods, we generate a novel simulated dataset to benchmark different methods of localization and classification of biological macromolecules in tomograms. Our publicly available dataset contains ten tomographic reconstructions of simulated cell-like volumes. Each volume contains twelve different types of complexes, varying in size, function and structure. In this paper, we have evaluated seven different methods of finding and classifying proteins. Seven research groups present results obtained with learning-based methods and trained on the simulated dataset, as well as a baseline template matching (TM), a traditional method widely used in cryo-ET research. We show that learning-based approaches can achieve notably better localization and classification performance than TM. We also experimentally confirm that there is a negative relationship between particle size and performance for all methods. | ['Fa Zhang', 'Xuefeng Cui', 'Cheng Chen', 'Yaoyu Wang', 'Min Xu', 'Sinuo Liu', 'Xiangrui Zeng', 'Konstantinos Moustakas', 'Evangelia I. Zacharaki', 'Stavros Gerolymatos', 'Giorgos Papoulias', 'Filiz Bunyak', 'Tommi White', 'Nguyen P. Nguyen', 'Emmanuel Moebel', 'Daisuke Kihara', 'Xiao Wang', 'Friedrich Förster', 'Remco C. Veltkamp', 'M. Cristina Trueba', 'Gijs van der Schot', 'Marten L. Chaillet', 'Ilja Gubins'] | 2022-03-18 | null | null | null | null | ['template-matching', 'electron-tomography', 'tomographic-reconstructions'] | ['computer-vision', 'medical', 'medical'] | [ 2.08657250e-01 -6.93538785e-01 2.79471487e-01 -2.63172150e-01
-9.77107882e-01 -6.57881737e-01 4.82033461e-01 2.21385345e-01
-5.22681415e-01 9.64095771e-01 -3.52952629e-01 -3.01498890e-01
8.07783678e-02 -4.08844709e-01 -7.39009917e-01 -1.09143710e+00
-2.34928846e-01 1.06079221e+00 3.87577981e-01 8.00118819e-02
5.41898310e-01 8.62350047e-01 -1.34710479e+00 8.45672607e-01
3.25871795e-01 8.41128051e-01 8.71187806e-01 8.10047925e-01
1.08708456e-01 4.76838499e-01 -4.07088608e-01 -2.08594650e-02
-8.42770785e-02 -2.50953704e-01 -8.44685018e-01 -5.98978885e-02
6.24453127e-01 4.18416977e-01 2.02059478e-01 6.68169856e-01
8.05446625e-01 -2.63830930e-01 9.44058299e-01 -6.56057060e-01
-3.22597563e-01 -2.48369381e-01 -2.40472466e-01 5.25186181e-01
6.63345218e-01 1.29309341e-01 5.94435096e-01 -1.20866394e+00
1.33115041e+00 1.07703185e+00 8.73490453e-01 4.99303132e-01
-1.86020637e+00 -3.39942008e-01 -3.83565396e-01 3.09057564e-01
-9.68067467e-01 -4.99588013e-01 3.42812359e-01 -5.57684004e-01
1.26739252e+00 4.26031053e-01 6.83295190e-01 9.13965881e-01
1.00297129e+00 1.44961283e-01 1.73402035e+00 -5.18442452e-01
3.79116029e-01 4.17942228e-03 -1.44039020e-01 7.04239845e-01
-2.70616217e-03 -1.52975276e-01 -3.64489466e-01 -5.81373274e-01
5.83546102e-01 9.07156914e-02 -2.98747152e-01 -8.56928051e-01
-1.33712363e+00 3.57845634e-01 4.55866791e-02 3.47683370e-01
-2.14678898e-01 -3.10224622e-01 6.10362053e-01 1.66361168e-01
3.92130077e-01 8.24859142e-01 -6.62734985e-01 1.91679895e-02
-6.69400513e-01 3.87808949e-01 7.65289307e-01 6.67879760e-01
6.69453621e-01 -4.49302405e-01 4.90728676e-01 6.63593829e-01
-4.58781421e-02 3.67358208e-01 4.96945918e-01 -9.39162612e-01
1.07164674e-01 3.64502221e-01 2.66030520e-01 -7.24615932e-01
-7.27773368e-01 2.20186353e-01 -8.21159840e-01 4.10241574e-01
6.43710017e-01 3.10230196e-01 -7.51721501e-01 1.20118058e+00
5.01372993e-01 -2.06197470e-01 -2.82011807e-01 8.93629491e-01
6.96172833e-01 5.42041361e-01 -1.04400434e-01 -6.01630986e-01
1.34050333e+00 -7.24454999e-01 -4.12924945e-01 1.74735054e-01
5.90203702e-01 -9.80449796e-01 9.03503835e-01 4.97979552e-01
-9.70401406e-01 -3.57810706e-01 -1.02490389e+00 1.39618769e-01
-6.89691484e-01 1.50604531e-01 5.27977288e-01 4.90910321e-01
-8.07377696e-01 9.27019298e-01 -1.06060135e+00 -6.32811248e-01
1.92313358e-01 4.34978813e-01 -8.74514699e-01 2.19923943e-01
-2.71354824e-01 1.13162994e+00 5.70369884e-02 -4.45841640e-01
-8.93128574e-01 -7.83969820e-01 -3.63974750e-01 -3.28116745e-01
-1.78450480e-01 -5.41948140e-01 8.81228387e-01 -1.73664823e-01
-1.17605424e+00 1.32547629e+00 -5.56053579e-01 -2.20809147e-01
1.74930543e-01 3.29536766e-01 -2.10833043e-01 3.49356443e-01
1.66470602e-01 2.91835099e-01 3.53939325e-01 -1.42843461e+00
-2.15896964e-01 -6.71182334e-01 -5.76921880e-01 -1.60951734e-01
3.12693536e-01 3.04079741e-01 7.16607198e-02 -4.74837348e-02
7.27169991e-01 -8.24419975e-01 -1.37547791e-01 -1.40841275e-01
-2.46142417e-01 -7.56185176e-03 1.01374018e+00 -5.29659927e-01
4.20533955e-01 -1.45331895e+00 3.14529032e-01 -8.32878202e-02
5.40547967e-01 1.22339651e-01 2.67535746e-01 7.09383905e-01
-2.91000128e-01 -3.66178490e-02 2.25957427e-02 -4.35280263e-01
-2.08162755e-01 -1.31790936e-01 1.32687679e-02 9.65751112e-01
-1.31798893e-01 6.28531039e-01 -6.79408729e-01 -7.09695399e-01
4.46157932e-01 5.22229612e-01 -8.95083547e-02 3.11058789e-01
-1.04309984e-01 6.83503032e-01 -1.11830734e-01 1.03452265e+00
8.93079877e-01 -6.23599052e-01 8.67898405e-01 -5.93613744e-01
-5.08241296e-01 3.32461625e-01 -6.04923785e-01 1.36990893e+00
-7.17942268e-02 4.93001014e-01 4.85338300e-01 -1.02589142e+00
8.17848086e-01 1.65939644e-01 6.24961197e-01 -8.04140449e-01
-1.37876555e-01 1.69754431e-01 -2.17105858e-02 -4.10360485e-01
-5.83898276e-04 -4.23512936e-01 4.91339475e-01 6.77854478e-01
2.21642524e-01 -1.68107837e-01 1.53963327e-01 7.90475216e-03
1.20455277e+00 2.28774741e-01 1.76647559e-01 -7.37620831e-01
3.85500014e-01 1.97735935e-01 3.47968161e-01 4.70821679e-01
-2.76602894e-01 7.26862788e-01 2.10012674e-01 -1.12832069e+00
-1.86152947e+00 -1.05212510e+00 -6.54020190e-01 1.05387855e+00
1.33289350e-02 -5.42162538e-01 -9.41963851e-01 -1.62947387e-01
1.48978457e-03 -3.42565328e-01 -5.40839791e-01 3.27112973e-01
-7.36178100e-01 -1.32808232e+00 1.77218840e-01 2.06191182e-01
-1.99312061e-01 -9.81175125e-01 -8.33936512e-01 1.95782050e-01
7.05313310e-03 -1.11699378e+00 8.94851703e-03 6.45062149e-01
-9.76706803e-01 -1.47798455e+00 -5.70584714e-01 -7.93021739e-01
8.08549225e-01 3.16882610e-01 1.31270683e+00 1.69274345e-01
-7.45005906e-01 1.42202750e-01 -1.69202402e-01 -1.58478186e-01
-5.02593100e-01 2.37484574e-02 6.36727750e-01 -5.51087976e-01
4.07978803e-01 -7.74365366e-01 -5.41167080e-01 7.37648010e-01
-3.83783787e-01 2.01630546e-03 3.53899896e-01 1.00445497e+00
1.12637877e+00 -2.89740235e-01 1.86747462e-01 -8.27470243e-01
4.32404280e-01 1.06903333e-02 -6.52776837e-01 4.41768438e-01
-7.58426368e-01 -9.92401466e-02 8.49360406e-01 -2.80297667e-01
-6.87589228e-01 1.66202441e-01 -1.45855218e-01 -2.43805543e-01
-3.89927536e-01 8.17370880e-03 -1.94175959e-01 -6.27787292e-01
7.08081305e-01 5.09211659e-01 2.78849036e-01 -6.48694038e-01
-3.96969140e-01 4.08678502e-01 4.93718773e-01 -9.28521752e-01
2.25716218e-01 8.19703400e-01 2.66554356e-01 -9.36671913e-01
-3.44307274e-01 -6.20018065e-01 -1.04989338e+00 -2.30175301e-01
6.31839633e-01 -5.74836075e-01 -1.24044776e+00 4.26443905e-01
-1.09064913e+00 -1.55862197e-01 5.20370960e-01 5.52512348e-01
-8.66555870e-01 5.89478612e-01 -8.57124448e-01 -5.71876645e-01
-4.60439742e-01 -1.37245750e+00 1.30291557e+00 -1.55368432e-01
-2.78121866e-02 -9.28033292e-01 5.29988170e-01 2.73164719e-01
4.04528469e-01 2.96069205e-01 1.04626882e+00 -3.44013005e-01
-4.87920582e-01 1.81336254e-01 -1.97713584e-01 -3.05525154e-01
9.76978913e-02 3.06797829e-02 -9.09222126e-01 -7.85335541e-01
2.56922632e-01 -5.73516369e-01 7.34534800e-01 4.34081376e-01
9.53775108e-01 2.26395100e-01 -8.60638678e-01 8.12356830e-01
1.52971375e+00 3.51233751e-01 8.38296235e-01 7.37660050e-01
4.88315165e-01 6.37466431e-01 5.92567742e-01 3.08745533e-01
-1.69759631e-01 8.88903558e-01 2.39645794e-01 -5.57671413e-02
2.63003588e-01 1.66034088e-01 2.88933795e-02 8.32000732e-01
-5.70419729e-01 1.01278089e-01 -1.04039574e+00 5.70338964e-02
-1.52025688e+00 -1.16633332e+00 -1.83121134e-02 2.03249693e+00
8.53145540e-01 -6.55132011e-02 3.09940994e-01 1.92517252e-03
6.49368227e-01 -2.35901430e-01 -6.84574127e-01 -1.04312912e-01
-3.87045234e-01 3.11687887e-01 3.43244553e-01 1.12558231e-01
-1.18766379e+00 6.07887387e-01 8.11974812e+00 5.52945256e-01
-1.32351768e+00 1.57580033e-01 5.94523430e-01 1.40333131e-01
3.02506059e-01 -1.02077641e-01 -8.98946762e-01 5.56201994e-01
1.20741558e+00 3.03562712e-02 2.63811767e-01 8.07360828e-01
2.71709889e-01 -2.68479168e-01 -1.19885361e+00 9.62747097e-01
5.71652502e-03 -1.99988675e+00 -1.56972677e-01 2.38196135e-01
3.99365097e-01 3.79625589e-01 -3.73710364e-01 -3.33865792e-01
-8.14969689e-02 -1.17795670e+00 4.02780831e-01 7.37940669e-01
8.83598685e-01 -5.92161596e-01 7.96264708e-01 3.60402852e-01
-1.31901205e+00 4.02619988e-01 -1.03499806e+00 1.12915464e-01
1.96335658e-01 5.30763030e-01 -9.12807822e-01 3.22949499e-01
1.05261326e+00 5.09740770e-01 -4.17598963e-01 8.75308692e-01
5.88196754e-01 2.48862058e-01 -2.45151788e-01 -6.89611435e-02
-3.31893265e-01 -6.29069984e-01 2.38376111e-01 1.51632464e+00
4.13544523e-03 4.94296961e-02 2.99179792e-01 8.70944321e-01
6.84171915e-02 1.06922537e-03 -6.75441086e-01 -1.63680408e-02
5.69765031e-01 1.38593018e+00 -1.01595342e+00 -2.31849506e-01
-8.31085593e-02 6.76488400e-01 5.35318911e-01 -3.98164019e-02
-4.67140675e-01 -1.62778035e-01 6.13704026e-01 5.00969768e-01
3.34425658e-01 -4.98452723e-01 -2.78731287e-02 -9.80831146e-01
-1.45389244e-01 -9.36592162e-01 -3.20694037e-02 -9.54516292e-01
-1.50343192e+00 5.43910682e-01 -1.98484078e-01 -8.93333554e-01
5.30873053e-02 -1.19372809e+00 -4.51730400e-01 7.43112862e-01
-8.47616136e-01 -1.02502751e+00 -1.35325119e-01 1.44422442e-01
1.93205133e-01 -3.43629569e-01 1.20996463e+00 2.64782697e-01
-2.88150728e-01 1.76580191e-01 7.49947667e-01 -3.68766844e-01
9.51061845e-01 -1.40620828e+00 3.80265921e-01 1.85603753e-01
-1.52631894e-01 9.74255264e-01 8.98327351e-01 -6.89831436e-01
-1.88461387e+00 -7.59129226e-01 6.24114037e-01 -9.59273636e-01
3.62358361e-01 -7.50882328e-01 -1.00918853e+00 4.78745013e-01
1.37076691e-01 1.05470777e-01 7.47238278e-01 -9.21773389e-02
-2.28193492e-01 3.26156884e-01 -1.44505048e+00 5.36001846e-02
8.89168739e-01 -8.43556821e-01 -4.24253404e-01 7.31430650e-01
2.00873718e-01 -4.94794428e-01 -1.38396335e+00 3.10887784e-01
1.00881863e+00 -1.41267669e+00 1.21850669e+00 -8.43234181e-01
9.99015197e-02 -3.65307748e-01 -2.17253655e-01 -9.97547209e-01
-5.04232585e-01 -3.30300152e-01 1.58275470e-01 5.35275161e-01
3.42841923e-01 -6.19701743e-01 1.15475023e+00 1.95849910e-01
-2.40953788e-01 -1.04927766e+00 -1.04846823e+00 -8.27012658e-01
1.78139210e-01 2.10015744e-01 3.53095025e-01 9.09994423e-01
2.50471085e-01 3.25232714e-01 -2.12009307e-02 -7.20273927e-02
8.78116429e-01 4.53571588e-01 8.39516938e-01 -1.29428899e+00
-8.17908794e-02 1.66680366e-01 -6.75575793e-01 -7.42995083e-01
1.49328917e-01 -6.81521714e-01 -7.95643777e-04 -1.13509965e+00
9.45230424e-01 -2.89117366e-01 1.29435226e-01 -4.61489558e-02
4.40301090e-01 3.03577930e-01 -2.11787179e-01 6.14913404e-01
-8.23397517e-01 9.26377401e-02 1.30995429e+00 7.23094344e-02
3.46201152e-01 -4.78241652e-01 1.83906481e-01 6.17250323e-01
7.55607426e-01 -4.94793117e-01 1.63704947e-01 9.71422866e-02
2.70558838e-02 -1.42077476e-01 3.34717512e-01 -1.25733459e+00
3.78841430e-01 -6.48696274e-02 7.82986283e-01 -1.05108130e+00
6.13379896e-01 -4.57174808e-01 4.30970341e-01 3.77182931e-01
1.85335785e-01 6.40141606e-01 -1.88236944e-02 4.20390040e-01
7.21320286e-02 -9.36640352e-02 1.17581236e+00 -7.31059492e-01
-1.01791695e-01 4.29681910e-04 -7.76363313e-01 -2.01664597e-01
7.81108856e-01 -3.47886831e-01 -7.29391098e-01 2.74280757e-01
-9.38680589e-01 -2.50811636e-01 1.34182680e+00 -4.93817747e-01
7.81540811e-01 -9.21466708e-01 -2.82154381e-01 2.17980728e-01
1.36886844e-02 -4.62499619e-01 3.92983295e-02 1.05186427e+00
-9.35994089e-01 8.66251409e-01 -7.81719744e-01 -1.04767728e+00
-1.59293497e+00 5.65628648e-01 6.39761627e-01 -2.67000407e-01
-5.75161099e-01 4.85243857e-01 5.97914197e-02 -9.75013316e-01
-1.44509777e-01 -1.29428312e-01 -4.60147187e-02 -3.42293948e-01
6.56500936e-01 3.86158049e-01 4.98167992e-01 -7.13697553e-01
-4.13779348e-01 7.42512286e-01 -2.79813498e-01 2.91648656e-01
1.57728469e+00 -2.10574381e-02 -5.68837702e-01 4.11880344e-01
1.02641392e+00 -2.88085788e-01 -8.48214686e-01 9.72679034e-02
9.30400565e-02 -4.51819211e-01 -3.31651837e-01 -5.68692923e-01
-5.50731957e-01 7.95272291e-01 9.44150567e-01 1.17232665e-01
6.30976021e-01 3.54213506e-01 6.21447861e-01 7.41277337e-01
9.00305271e-01 -9.75646675e-01 1.77509747e-02 4.13466305e-01
5.55862308e-01 -1.17437446e+00 4.54245269e-01 -4.54358160e-01
8.82345065e-02 1.20701301e+00 6.20823860e-01 1.74277630e-02
3.91364366e-01 7.11333275e-01 -1.94449238e-02 -7.41950691e-01
-1.18187487e+00 2.78105587e-01 -2.72317797e-01 1.02541709e+00
8.03904653e-01 -3.77339683e-02 -2.24286452e-01 1.54025912e-01
-1.34941190e-02 -2.87213951e-01 3.52542609e-01 1.37078643e+00
-7.44060814e-01 -1.36376548e+00 -6.49744928e-01 5.58198154e-01
-5.71884453e-01 1.59301355e-01 -7.40851283e-01 6.33384764e-01
-5.49741648e-02 4.53314960e-01 7.16445968e-02 -4.17358786e-01
1.09738074e-01 2.08252877e-01 1.01096952e+00 -5.06712735e-01
-2.94957906e-01 8.10422078e-02 9.17128175e-02 -6.35091007e-01
-7.52021670e-01 -8.26798975e-01 -1.34252942e+00 -5.38528383e-01
-5.85679650e-01 5.37621140e-01 8.50403965e-01 7.57382274e-01
5.02943814e-01 1.38126791e-01 4.26331908e-01 -1.52757394e+00
-1.88200668e-01 -1.03108478e+00 -8.06168616e-01 3.87278646e-01
7.74346218e-02 -8.65455329e-01 -3.48073423e-01 2.45847747e-01] | [13.401782989501953, -3.0903818607330322] |
a429e41c-ddd6-4442-b164-ece3e47cf53b | an-overview-of-healthcare-data-analytics-with | 2111.14623 | null | https://arxiv.org/abs/2111.14623v1 | https://arxiv.org/pdf/2111.14623v1.pdf | An Overview of Healthcare Data Analytics With Applications to the COVID-19 Pandemic | In the era of big data, standard analysis tools may be inadequate for making inference and there is a growing need for more efficient and innovative ways to collect, process, analyze and interpret the massive and complex data. We provide an overview of challenges in big data problems and describe how innovative analytical methods, machine learning tools and metaheuristics can tackle general healthcare problems with a focus on the current pandemic. In particular, we give applications of modern digital technology, statistical methods, data platforms and data integration systems to improve diagnosis and treatment of diseases in clinical research and novel epidemiologic tools to tackle infection source problems, such as finding Patient Zero in the spread of epidemics. We make the case that analyzing and interpreting big data is a very challenging task that requires a multi-disciplinary effort to continuously create more effective methodologies and powerful tools to transfer data information into knowledge that enables informed decision making. | ['Weng Kee Wong', 'Chee Wei Tan', 'Oleksandr Sverdlov', 'Yevgen Ryeznik', 'Zhe Fei'] | 2021-11-25 | null | null | null | null | ['data-integration'] | ['knowledge-base'] | [ 1.32727534e-01 -2.96355098e-01 -6.35410696e-02 5.48389144e-02
-2.94968218e-01 -1.05241120e-01 1.48266479e-01 8.07973683e-01
-3.91206741e-01 9.32643712e-01 2.17714339e-01 -3.94017547e-01
-7.99002528e-01 -9.68871057e-01 -1.89209610e-01 -8.41144264e-01
-3.36492836e-01 1.27174783e+00 -2.70209312e-01 -4.37451452e-01
1.11397445e-01 5.01008689e-01 -1.45474589e+00 2.94477820e-01
1.23451138e+00 7.95794964e-01 2.36937180e-01 8.59395862e-01
-4.02117968e-01 3.83726120e-01 -6.54328585e-01 -2.36749038e-01
7.58584365e-02 -2.93862790e-01 -7.68190801e-01 -2.93001652e-01
-8.45161200e-01 2.84988672e-01 6.30678892e-01 8.15409184e-01
7.59292603e-01 -1.01553425e-01 5.28092921e-01 -1.01516891e+00
-2.19010741e-01 1.72913089e-01 -4.74301696e-01 3.89961153e-01
3.90604347e-01 1.33282438e-01 1.41080841e-01 -4.09683228e-01
8.13907444e-01 1.23513842e+00 7.38127828e-01 2.70088106e-01
-8.41207445e-01 -7.39642829e-02 -3.77602279e-01 5.05316854e-01
-9.57961500e-01 -7.33975647e-03 1.91800386e-01 -6.28389060e-01
8.82039726e-01 6.32770717e-01 7.39007711e-01 9.63376939e-01
3.07754099e-01 2.24867821e-01 8.55226994e-01 -3.78110796e-01
5.93833625e-01 -2.13001668e-02 1.98082179e-01 4.10469264e-01
7.35724568e-01 -2.43449640e-02 -2.53671408e-01 -4.85424668e-01
2.06702024e-01 8.11665535e-01 -5.41861728e-03 1.70556188e-01
-1.32472086e+00 1.09136069e+00 -2.05568001e-02 6.83833003e-01
-1.16807067e+00 -4.45352256e-01 3.68074954e-01 3.08378667e-01
4.44984138e-01 5.49917877e-01 -7.51652777e-01 -3.82591754e-01
-5.55761576e-01 2.44041443e-01 8.90725374e-01 1.92901894e-01
2.89194316e-01 -1.74469724e-01 1.56721100e-01 6.43761158e-01
-2.28987746e-02 8.50651622e-01 3.95600826e-01 -9.14209843e-01
2.63271481e-01 9.17876780e-01 -7.46214539e-02 -1.22259796e+00
-8.90719056e-01 -2.74089098e-01 -1.54123747e+00 -1.74106777e-01
3.29030991e-01 -4.03391480e-01 -5.31234384e-01 9.69654322e-01
7.87135422e-01 -7.40166828e-02 1.83008850e-01 7.10499883e-01
4.65934277e-01 5.24342537e-01 7.75509626e-02 -7.61282623e-01
1.76559901e+00 -2.76122749e-01 -1.06996751e+00 4.15358432e-02
6.46435916e-01 -5.79994857e-01 5.90197206e-01 7.65638351e-01
-1.19922066e+00 -1.66355118e-01 -4.33836937e-01 2.95086324e-01
-1.02715552e+00 -6.04791403e-01 8.32251191e-01 5.15421152e-01
-7.26189375e-01 4.27377135e-01 -8.87484729e-01 -7.26402581e-01
5.86633563e-01 4.57198352e-01 -3.31449211e-01 -4.42630559e-01
-1.19446814e+00 8.38842452e-01 5.14708281e-01 1.94779173e-01
-2.62019187e-01 -1.17075193e+00 -5.15049934e-01 -2.05557778e-01
5.22354722e-01 -1.33956409e+00 3.96575272e-01 -3.01099151e-01
-8.60376477e-01 7.99496770e-01 -1.26244232e-01 -4.19891775e-01
3.18346173e-01 4.83976342e-02 -2.87411630e-01 1.23921083e-04
-2.85731822e-01 -2.14835986e-01 4.09015983e-01 -7.84722626e-01
-7.74010122e-01 -1.16194844e+00 -8.54064107e-01 -4.48258579e-01
-2.06280231e-01 3.11887115e-01 7.00864121e-02 -4.06103760e-01
-3.26261789e-01 -7.18664467e-01 -8.61091197e-01 -5.44595599e-01
-3.71373206e-01 -1.00324512e-01 6.40629411e-01 -6.14690483e-01
1.18893504e+00 -1.63721466e+00 5.52592576e-01 5.09404838e-01
4.82393593e-01 6.84525073e-01 1.49055660e-01 7.69413650e-01
3.77493769e-01 3.43811512e-01 -2.63606995e-01 8.56441930e-02
-5.20528138e-01 4.96044993e-01 1.25458181e-01 2.22554058e-01
1.83637738e-01 8.08843076e-01 -9.56148803e-01 -3.18001181e-01
4.02383924e-01 8.11387241e-01 -5.05665839e-01 4.33664173e-01
-2.91212261e-01 7.21052349e-01 -8.89586806e-01 7.61934161e-01
6.06397152e-01 -8.35882843e-01 4.03304935e-01 2.40414724e-01
-2.03928202e-01 -5.44797480e-01 -1.03931260e+00 1.07728660e+00
-4.42679286e-01 -9.10835713e-02 3.54973704e-01 -1.43859220e+00
6.21470451e-01 3.47862363e-01 9.64053750e-01 -7.95572400e-01
3.24852079e-01 3.28802735e-01 1.97375510e-02 -1.21316910e+00
-1.79078002e-02 -1.53194010e-01 1.88777864e-01 4.36085492e-01
-6.23802185e-01 9.66330171e-02 4.15402472e-01 -6.44588172e-02
1.28223979e+00 -8.35638523e-01 4.94197965e-01 -1.28349677e-01
4.34531510e-01 2.38962889e-01 6.23760045e-01 4.61812884e-01
6.88538551e-02 1.46770537e-01 5.56659400e-01 -1.00106239e+00
-1.08008134e+00 -6.95639074e-01 -1.90576345e-01 8.25339973e-01
-5.34862697e-01 -4.38232929e-01 -4.81220305e-01 -2.68124249e-02
1.87007427e-01 2.21491411e-01 -6.83386147e-01 -6.46287855e-03
-4.08336490e-01 -1.71078956e+00 -7.05476105e-02 -3.01819947e-02
9.19080749e-02 -1.05633318e+00 -8.51386487e-01 4.95186985e-01
-1.42521083e-01 -9.42884624e-01 4.12618726e-01 1.12968899e-01
-9.98918235e-01 -1.29771090e+00 -6.89370513e-01 -4.08725679e-01
6.46996617e-01 -1.80924386e-01 1.25252938e+00 4.77636218e-01
-1.11267292e+00 -4.69748229e-02 -4.69429314e-01 -1.00350153e+00
-5.97206175e-01 -1.04022250e-01 1.80579945e-01 -2.43379965e-01
3.82045120e-01 -3.79439116e-01 -8.31847310e-01 1.68041214e-01
-1.04401910e+00 -1.62601620e-01 5.04246593e-01 7.41972685e-01
7.03715205e-01 4.01879400e-01 9.87946391e-01 -1.05999613e+00
9.72172260e-01 -1.05146253e+00 -6.21219099e-01 4.03833091e-01
-7.91312337e-01 1.66049361e-01 7.55430758e-01 3.17076743e-02
-5.87151587e-01 -2.92061239e-01 -3.21597159e-01 -6.53834119e-02
-1.12752818e-01 6.81768715e-01 1.17383420e-01 3.26979667e-01
4.95003641e-01 -1.38970137e-01 5.32496333e-01 -6.63891971e-01
-1.52917774e-02 8.96726906e-01 -4.12690863e-02 -1.68998972e-01
1.99469462e-01 7.46069789e-01 4.64608490e-01 -1.05689871e+00
-5.62611163e-01 -6.43266082e-01 -4.18467373e-01 -1.72218867e-02
1.15531063e+00 -3.22268933e-01 -1.28748178e+00 2.99683481e-01
-8.49862993e-01 -3.26301754e-02 -2.54272431e-01 4.34173495e-01
-3.43520314e-01 1.16671093e-01 -4.16496158e-01 -8.02984536e-01
-7.08008349e-01 -1.02488875e+00 1.15918005e+00 1.72213271e-01
-1.36576384e-01 -1.39545834e+00 5.41377604e-01 8.38312030e-01
8.36005509e-01 8.44244123e-01 1.21693623e+00 -6.98786318e-01
-3.69507819e-01 -3.02594513e-01 1.99934710e-02 -1.95875913e-01
3.36639911e-01 2.72183090e-01 -6.92231894e-01 -1.51232369e-02
4.73536849e-02 -1.14985086e-01 5.39202213e-01 7.40942121e-01
1.31666315e+00 -5.57363927e-01 -6.26840055e-01 5.92022717e-01
1.23922813e+00 3.09065074e-01 5.86869001e-01 8.53381455e-02
5.19194365e-01 9.27414536e-01 3.89450431e-01 9.21218276e-01
4.07597780e-01 4.06590492e-01 2.43892968e-01 -1.44636452e-01
6.32589281e-01 5.61745882e-01 -4.01836753e-01 9.11580324e-01
-7.10094452e-01 -2.23356739e-01 -1.24613678e+00 2.30958924e-01
-1.98248148e+00 -1.28594244e+00 -4.24917966e-01 2.31688380e+00
8.36430371e-01 -1.88613027e-01 3.93786162e-01 2.75715202e-01
3.84601325e-01 -4.60044205e-01 -2.36589849e-01 -6.93996549e-01
-1.62503541e-01 6.39208481e-02 2.28252396e-01 1.11302868e-01
-6.99583650e-01 3.91425997e-01 7.11642456e+00 5.68414211e-01
-1.12750340e+00 6.32668585e-02 6.87366188e-01 -7.45920986e-02
-2.11680114e-01 -8.03088009e-01 -5.14310002e-01 6.10096276e-01
1.56381214e+00 -8.34283680e-02 5.91662228e-01 3.25712860e-01
8.81530941e-01 -1.86869606e-01 -6.00463569e-01 8.65308702e-01
-3.54000658e-01 -1.60979843e+00 -2.34409139e-01 5.87609947e-01
7.33531296e-01 1.30559072e-01 -1.32694840e-01 -1.80081010e-01
4.67438400e-01 -1.11142230e+00 -5.35762727e-01 8.82303298e-01
2.04250485e-01 -7.06359625e-01 1.05343306e+00 4.34305876e-01
-5.30354500e-01 -2.37013280e-01 -4.34058726e-01 -1.29270926e-01
4.36894000e-01 1.26133859e+00 -9.82774436e-01 7.87259996e-01
7.55666614e-01 4.96866077e-01 -3.32591772e-01 1.13688242e+00
6.57164991e-01 2.77980298e-01 -3.34495813e-01 -6.30672872e-02
-5.76992556e-02 -3.54423642e-01 3.33824396e-01 7.42374241e-01
4.23371136e-01 5.25190607e-02 1.59396604e-01 3.23732316e-01
4.67334300e-01 2.73718745e-01 -6.45451188e-01 -5.24245679e-01
6.33790419e-02 1.23580694e+00 -9.13397133e-01 -3.48469228e-01
-1.97168812e-01 2.80654490e-01 8.62076059e-02 3.08130924e-02
-2.82555252e-01 -7.59993121e-02 8.13487411e-01 2.62909234e-01
-2.42536701e-02 2.14392282e-02 -3.69232297e-01 -1.05934131e+00
-4.40497190e-01 -1.32294583e+00 1.19971490e+00 -3.02933753e-01
-1.32628918e+00 3.49083066e-01 -1.27414837e-01 -5.58497429e-01
-8.58775318e-01 -7.95475006e-01 -5.36150634e-01 4.73381698e-01
-1.30608892e+00 -7.11491823e-01 -2.72401959e-01 7.46791363e-01
2.27850214e-01 -1.08997732e-01 1.06908858e+00 4.36838657e-01
-7.25438535e-01 -2.52956182e-01 7.89452016e-01 -3.68498474e-01
1.21380523e-01 -1.02453506e+00 -2.26272821e-01 1.94744706e-01
-5.26077926e-01 5.14259458e-01 9.14760411e-01 -8.16092908e-01
-2.00999665e+00 -7.66757190e-01 7.75075972e-01 -4.40461010e-01
5.35205245e-01 -2.73810953e-01 -9.38387752e-01 1.53882071e-01
-3.10379695e-02 -1.30163684e-01 1.13072872e+00 4.01939988e-01
1.79091945e-01 -2.26282492e-01 -1.41901624e+00 2.52252758e-01
5.58273673e-01 1.36955366e-01 -3.97305161e-01 9.21926916e-01
4.18422282e-01 2.71008044e-01 -1.18213844e+00 4.15765703e-01
3.19044113e-01 -8.66655409e-01 1.38392222e+00 -1.15304208e+00
1.95739828e-02 1.47792041e-01 2.04237819e-01 -1.27562857e+00
1.54998139e-01 -7.53402650e-01 -2.61270665e-02 5.27956784e-01
1.56342968e-01 -7.48860002e-01 5.66360533e-01 5.75940311e-01
2.32133031e-01 -1.02901220e+00 -8.14648032e-01 -5.30221343e-01
-2.11008936e-02 -3.53099376e-01 6.94317698e-01 8.90389979e-01
2.22132146e-01 -8.25800076e-02 -3.87883633e-01 -1.32805884e-01
8.26224625e-01 1.91432565e-01 5.03394842e-01 -1.71952069e+00
4.09929045e-02 -2.67685145e-01 -5.48915207e-01 6.26451731e-01
-3.54273707e-01 -4.44095939e-01 -4.36912894e-01 -1.91088474e+00
2.80732721e-01 -4.98013347e-01 -1.75372735e-01 1.10908799e-01
-1.46043062e-01 4.95671369e-02 -2.35500962e-01 5.78918681e-02
-5.93211949e-01 1.05507702e-01 1.39906669e+00 2.32384298e-02
-2.54109472e-01 2.88193226e-01 -7.37317264e-01 6.84486985e-01
7.00646520e-01 -4.51117486e-01 -1.94008738e-01 -1.62315518e-01
8.84237170e-01 1.97621584e-01 1.81788832e-01 -7.24364460e-01
1.84749588e-01 -7.23503351e-01 4.06946510e-01 -5.05959690e-01
4.74575274e-02 -1.00628185e+00 6.02035284e-01 1.15963364e+00
6.21802695e-02 4.66233253e-01 -2.62181237e-02 4.18693185e-01
-5.44662029e-02 1.60471518e-02 6.04853511e-01 -3.06250691e-01
-3.42030972e-01 3.31177145e-01 -4.27884042e-01 5.45122027e-01
1.49653721e+00 1.74608737e-01 -5.04438758e-01 -1.29510820e-01
-8.17889631e-01 6.28247261e-01 9.24167112e-02 4.98726100e-01
5.74567258e-01 -8.20610106e-01 -8.97043407e-01 3.13816100e-01
-1.48180261e-01 1.68691024e-01 6.34213924e-01 1.24078941e+00
-8.14541042e-01 5.87408066e-01 -3.35003853e-01 -4.59137559e-01
-1.39829719e+00 1.21619046e+00 8.65255147e-02 -5.72855711e-01
-4.93326217e-01 2.44784728e-01 -2.67727941e-01 -3.18849713e-01
3.18646356e-02 -2.11178556e-01 -2.54118532e-01 4.85959917e-01
1.20914817e+00 7.94434726e-01 1.95414484e-01 -2.75274158e-01
-6.92022026e-01 6.17440522e-01 5.58123253e-02 5.59209824e-01
2.07343197e+00 -4.57583405e-02 -3.94719779e-01 2.72783518e-01
9.84690547e-01 -2.55208582e-01 -1.98459238e-01 3.83491486e-01
2.88490951e-01 -3.82998317e-01 -3.48814472e-04 -9.34606194e-01
-1.02027392e+00 1.06586480e+00 5.53465426e-01 8.43285024e-01
1.20777297e+00 3.74425173e-01 7.67897248e-01 3.46389651e-01
1.84890300e-01 -1.11886251e+00 -1.48656234e-01 2.55246639e-01
7.92660296e-01 -1.41193795e+00 -7.71627426e-02 -1.16576456e-01
-3.95120144e-01 9.35128272e-01 -1.25621602e-01 4.17957425e-01
9.47522283e-01 6.55376375e-01 5.72607182e-02 -5.89493752e-01
-9.86936748e-01 -2.73648262e-01 9.33397189e-03 8.11873615e-01
2.38200739e-01 -3.92880142e-02 -3.42577368e-01 4.36793685e-01
4.18574139e-02 6.25733912e-01 2.15694532e-01 9.11360741e-01
-4.94428307e-01 -1.37047601e+00 -8.14154685e-01 7.93219090e-01
-9.11829710e-01 -2.60186549e-02 -2.26595864e-01 5.25638223e-01
2.09731713e-01 9.70730424e-01 1.46395534e-01 2.77477950e-02
1.60312995e-01 3.62511054e-02 1.54354274e-01 -2.83231109e-01
-6.93544209e-01 -9.97382700e-02 -1.22753248e-01 -5.09957850e-01
-4.88878280e-01 -6.42677367e-01 -1.38137615e+00 -7.52693236e-01
1.06777020e-01 4.27524298e-01 1.15227973e+00 9.65680122e-01
9.34026778e-01 5.91355681e-01 3.93834203e-01 -4.04289037e-01
-3.41738671e-01 -4.46116567e-01 -5.45141459e-01 3.71320963e-01
3.85220617e-01 -3.80726784e-01 -2.43427843e-01 -1.07172273e-01] | [6.03144645690918, 4.973541259765625] |
46c7e92b-b0f6-4025-b74a-090ec52737ac | ame-cam-attentive-multiple-exit-cam-for | 2306.14505 | null | https://arxiv.org/abs/2306.14505v1 | https://arxiv.org/pdf/2306.14505v1.pdf | AME-CAM: Attentive Multiple-Exit CAM for Weakly Supervised Segmentation on MRI Brain Tumor | Magnetic resonance imaging (MRI) is commonly used for brain tumor segmentation, which is critical for patient evaluation and treatment planning. To reduce the labor and expertise required for labeling, weakly-supervised semantic segmentation (WSSS) methods with class activation mapping (CAM) have been proposed. However, existing CAM methods suffer from low resolution due to strided convolution and pooling layers, resulting in inaccurate predictions. In this study, we propose a novel CAM method, Attentive Multiple-Exit CAM (AME-CAM), that extracts activation maps from multiple resolutions to hierarchically aggregate and improve prediction accuracy. We evaluate our method on the BraTS 2021 dataset and show that it outperforms state-of-the-art methods. | ['Tsung-Yi Ho', 'Yiyu Shi', 'Xinrong Hu', 'Yu-Jen Chen'] | 2023-06-26 | null | null | null | null | ['weakly-supervised-segmentation', 'weakly-supervised-semantic-segmentation', 'tumor-segmentation', 'brain-tumor-segmentation'] | ['computer-vision', 'computer-vision', 'computer-vision', 'medical'] | [ 5.10715365e-01 3.86031151e-01 -2.47144982e-01 -5.37169695e-01
-1.12185442e+00 -1.48848221e-01 3.98484468e-01 2.33692780e-01
-7.41383553e-01 7.54249990e-01 1.97401658e-01 -1.73655644e-01
1.55843467e-01 -5.98657906e-01 -1.44394830e-01 -6.52149379e-01
3.21749181e-01 5.11976779e-01 7.13408768e-01 9.18316096e-02
1.36495791e-02 5.78413963e-01 -8.13050807e-01 7.17744470e-01
1.14217484e+00 9.92357552e-01 6.17279053e-01 2.90909767e-01
-3.51721078e-01 8.41364741e-01 -3.97332191e-01 4.51334454e-02
-9.12003592e-02 -4.98710513e-01 -1.21819305e+00 4.02374417e-02
1.39657199e-01 4.57865074e-02 -2.29061525e-02 1.27809656e+00
2.53263831e-01 -3.16249095e-02 6.93133831e-01 -7.53359377e-01
-1.37089893e-01 6.70032263e-01 -5.92616558e-01 4.51125473e-01
-3.09915543e-01 -9.70952362e-02 5.64601600e-01 -9.61764216e-01
6.77359402e-01 6.85335875e-01 5.75755179e-01 8.33347499e-01
-1.18124640e+00 -6.17738068e-01 1.16915092e-01 2.12614313e-01
-1.34907222e+00 -1.78311467e-01 4.73928511e-01 -4.72992986e-01
8.23357224e-01 2.06382558e-01 8.86163116e-01 7.95378625e-01
2.20524684e-01 1.05969596e+00 1.42912960e+00 -1.48065254e-01
5.06176293e-01 -2.56814003e-01 2.53637761e-01 8.18096399e-01
8.02385658e-02 -4.06144202e-01 -2.22594455e-01 1.11215003e-01
7.49393284e-01 1.10018387e-01 -2.28933394e-01 -1.49407417e-01
-1.24172211e+00 7.45433152e-01 9.51519668e-01 5.97603559e-01
-5.69847882e-01 -1.97755247e-02 2.28794649e-01 -4.88399118e-01
7.79538810e-01 5.04122376e-01 -2.54585624e-01 1.88885048e-01
-1.35956013e+00 -6.69079199e-02 2.82200634e-01 5.49931765e-01
3.20280641e-01 -1.58974454e-01 -4.60472256e-01 9.29134071e-01
1.53244361e-02 -1.34711325e-01 9.50553954e-01 -6.84569776e-01
9.99726355e-02 7.94540524e-01 -2.64162570e-01 -3.11250210e-01
-1.01945281e+00 -5.75456560e-01 -9.52466965e-01 1.38761058e-01
2.70808935e-01 3.37097049e-02 -1.71596467e+00 1.32376933e+00
1.33930489e-01 2.90734172e-01 1.07667251e-02 1.10356212e+00
1.14377022e+00 6.87159076e-02 6.10027969e-01 -1.92242250e-01
1.15640390e+00 -1.43048620e+00 -6.44654572e-01 -4.47704971e-01
9.64416921e-01 -3.30414653e-01 8.71415436e-01 1.35456026e-01
-1.06217968e+00 -1.29676521e-01 -8.92018914e-01 1.21832699e-01
-4.76643056e-01 6.23542443e-02 7.09662795e-01 6.44218624e-01
-1.09134924e+00 3.52443397e-01 -1.38237035e+00 -1.98889256e-01
1.23558354e+00 4.95493978e-01 -3.45816255e-01 -1.70981362e-02
-8.79619539e-01 1.04761004e+00 5.14435470e-01 2.49863956e-02
-8.75828445e-01 -8.54079962e-01 -6.24659240e-01 -2.27767732e-02
4.03129309e-01 -4.36385065e-01 1.34036660e+00 -9.26789165e-01
-1.48958540e+00 8.79253626e-01 -2.30771184e-01 -6.98725104e-01
4.88476515e-01 1.77059531e-01 -1.33157283e-01 4.50802028e-01
8.74973834e-02 1.30554426e+00 5.54942608e-01 -1.06041253e+00
-5.14189065e-01 -5.72289646e-01 -2.71281570e-01 3.87950331e-01
-2.60654509e-01 5.08075245e-02 -4.61616665e-01 -7.91043758e-01
6.08481050e-01 -8.52361500e-01 -7.55081654e-01 -4.05662537e-01
-5.51577747e-01 -7.62593597e-02 8.08960497e-01 -9.01079714e-01
1.01983702e+00 -1.92001390e+00 2.54810363e-01 1.76704302e-01
5.36388636e-01 3.32421809e-01 2.55162567e-01 -6.13211930e-01
-3.81776914e-02 2.04531193e-01 -8.49518895e-01 -4.65229541e-01
-4.81082350e-01 2.21536681e-01 2.52079785e-01 2.91415453e-01
7.69256055e-02 1.18813860e+00 -7.37891316e-01 -9.01807249e-01
3.03034127e-01 2.93276936e-01 -5.79432786e-01 9.84297022e-02
-1.06418267e-01 9.50411439e-01 -4.59023893e-01 9.42945361e-01
4.11504745e-01 -4.04072821e-01 -3.46588599e-03 -2.88000286e-01
6.58814162e-02 -5.41716628e-02 -3.50192040e-01 1.92050195e+00
-2.49620423e-01 4.54158664e-01 7.35015497e-02 -1.29666936e+00
5.11824191e-01 2.81337827e-01 9.89387691e-01 -7.04356134e-01
3.15048367e-01 2.74762005e-01 4.65110242e-02 -2.79804409e-01
1.43952519e-01 -1.83458790e-01 4.90025841e-02 -3.06786969e-02
5.69269154e-03 -4.62593645e-01 -2.54649352e-02 1.56229854e-01
1.18119216e+00 -1.14791818e-01 2.72806197e-01 -4.37237382e-01
3.98094058e-01 3.17149520e-01 6.79476321e-01 4.66817498e-01
-5.91153562e-01 8.99935484e-01 3.27438682e-01 -3.37163001e-01
-7.38569140e-01 -9.96772170e-01 -2.99967229e-01 5.45523226e-01
7.78886080e-02 -2.33689517e-01 -1.40640938e+00 -1.03326130e+00
-3.90933484e-01 5.36840260e-01 -7.22192347e-01 5.27800657e-02
-5.43805063e-01 -1.14934993e+00 3.54459614e-01 8.92247379e-01
7.47632265e-01 -1.11819160e+00 -8.51875067e-01 2.98966467e-01
-4.03363496e-01 -1.44734955e+00 -3.19364220e-01 2.10012391e-01
-1.09395075e+00 -9.95616198e-01 -1.03989315e+00 -7.11820185e-01
1.14355052e+00 -6.77114679e-03 7.92288184e-01 -5.80459870e-02
-6.33716643e-01 5.54185808e-02 -2.19206646e-01 -4.35704887e-01
-2.30340421e-01 3.42563927e-01 -4.20155168e-01 -1.24684446e-01
2.54343480e-01 -1.41502321e-01 -8.31427872e-01 3.07026595e-01
-9.11030710e-01 7.88702965e-01 8.24778020e-01 9.02757823e-01
1.10747242e+00 -1.89458832e-01 5.35062313e-01 -1.35398626e+00
4.09745902e-01 -2.39929259e-01 -4.89142269e-01 2.98195183e-01
-5.67801416e-01 -2.87585497e-01 2.77602524e-01 -1.13270544e-01
-1.10432482e+00 4.30620402e-01 -2.82054305e-01 -2.48154104e-01
-4.13893074e-01 5.42285860e-01 1.90795690e-01 -4.82540458e-01
5.36156952e-01 -4.32623401e-02 -8.55925083e-02 -2.81751007e-01
1.29351184e-01 5.71414053e-01 6.93273187e-01 -1.13348149e-01
1.15353659e-01 5.90932250e-01 9.01364684e-02 -7.21533895e-01
-1.12951303e+00 -4.16779786e-01 -8.04586470e-01 -3.83193403e-01
1.42048609e+00 -7.14323759e-01 -1.05156913e-01 4.63498771e-01
-9.64144468e-01 -6.07524753e-01 -1.53002083e-01 6.10419810e-01
-6.79880500e-01 -4.58878494e-04 -8.57662380e-01 -3.69416714e-01
-5.13633907e-01 -1.66777754e+00 9.69290257e-01 4.56165344e-01
-2.77192295e-01 -6.97206378e-01 -4.25814360e-01 7.31306493e-01
6.75771654e-01 3.51768374e-01 8.77728760e-01 -5.64477026e-01
-6.16064191e-01 -9.63106081e-02 -5.03192306e-01 2.60612726e-01
6.08463138e-02 -5.36907792e-01 -9.80042756e-01 -1.43294871e-01
-3.12708348e-01 -1.98879302e-01 1.10502303e+00 9.24084485e-01
2.00392437e+00 8.85863975e-02 -6.61429226e-01 7.39503503e-01
1.15352011e+00 2.84136444e-01 6.17359638e-01 3.15075994e-01
8.44674289e-01 4.74839509e-01 3.70714277e-01 -4.03988361e-02
2.41741419e-01 5.19810021e-01 4.19314682e-01 -6.65240467e-01
-3.16378772e-01 2.16566965e-01 -2.09125236e-01 6.89623177e-01
-1.05811059e-01 1.77396849e-01 -1.24491882e+00 5.75889766e-01
-1.81637216e+00 -4.80480641e-01 4.54901084e-02 1.68716240e+00
8.40323627e-01 2.40081519e-01 -1.02616183e-01 2.72579342e-02
6.80509031e-01 -2.05072910e-01 -5.91010809e-01 3.38921063e-02
1.02931000e-01 6.69368744e-01 6.85994804e-01 4.13699031e-01
-1.21058846e+00 1.25498438e+00 6.61896086e+00 1.03010881e+00
-1.10290253e+00 7.81203151e-01 1.12865520e+00 8.11560429e-04
1.28165841e-01 -2.67782211e-01 -5.99237382e-01 3.69614929e-01
6.56432390e-01 2.97297508e-01 1.13165431e-01 7.98586726e-01
1.00808041e-02 -4.81398106e-01 -7.11969495e-01 8.76161993e-01
3.09456736e-02 -1.47376657e+00 -1.89087644e-01 -2.80675851e-02
8.98129880e-01 2.19377950e-01 -1.36603862e-01 9.12537798e-02
1.71666101e-01 -1.19147038e+00 4.47158128e-01 4.91084486e-01
6.84180975e-01 -5.72716534e-01 9.40520406e-01 1.87044069e-01
-1.03376055e+00 -5.17042875e-02 -6.74987733e-02 6.29622281e-01
2.15035081e-01 4.82476950e-01 -9.69413400e-01 3.69141012e-01
7.21152663e-01 5.40955007e-01 -5.69596767e-01 1.24454224e+00
-1.45317763e-01 7.21255183e-01 -2.33232901e-01 2.31809735e-01
4.38492894e-01 1.82044581e-01 2.90646553e-01 1.11650348e+00
2.32456237e-01 5.35907447e-01 4.27521437e-01 8.55845153e-01
-7.39168599e-02 1.58469960e-01 5.80949802e-03 -9.22062173e-02
4.96950597e-02 1.41648626e+00 -1.55487859e+00 -4.89998132e-01
-2.90945798e-01 9.33350027e-01 3.14775467e-01 2.49432951e-01
-6.64491773e-01 2.49017756e-02 2.06414506e-01 2.33634740e-01
-9.14227590e-02 -2.12194443e-01 -7.72855639e-01 -8.69225621e-01
-5.34784019e-01 -3.32605928e-01 5.93426228e-01 -7.08938181e-01
-1.00118339e+00 8.73327732e-01 7.69762322e-02 -9.14124668e-01
1.96141496e-01 -5.32928228e-01 -4.99892771e-01 6.76768005e-01
-1.57635736e+00 -1.15009129e+00 -6.25810504e-01 5.00964880e-01
7.95463622e-01 -1.15062177e-01 7.23957300e-01 3.08461308e-01
-6.76843584e-01 4.09362406e-01 -5.71657300e-01 1.29106462e-01
2.88323879e-01 -1.16984594e+00 7.69352764e-02 7.07227349e-01
-9.38201472e-02 -1.49083333e-02 1.46138608e-01 -6.69163644e-01
-5.96001804e-01 -1.30199945e+00 4.70484883e-01 -5.93631081e-02
3.58203351e-01 -7.15132104e-03 -1.01393604e+00 4.31355476e-01
4.77531664e-02 4.94178623e-01 7.36965418e-01 -4.88820791e-01
2.96661437e-01 9.29945409e-02 -1.42252088e+00 5.59862256e-01
8.09165716e-01 -3.32268357e-01 -3.66871715e-01 3.40167552e-01
6.61592126e-01 -7.12822735e-01 -8.67671967e-01 9.56341028e-01
1.45430043e-01 -6.93043172e-01 9.44117904e-01 -3.81359696e-01
2.49387279e-01 -1.46921217e-01 1.41044557e-01 -1.44665504e+00
-3.71765405e-01 1.04081444e-01 1.57563090e-01 4.50198561e-01
5.44756770e-01 -5.34843206e-01 1.14786172e+00 7.56545186e-01
-6.08477473e-01 -1.11384559e+00 -9.95771289e-01 -4.47818995e-01
1.00043073e-01 -3.28921646e-01 4.40623671e-01 9.58518624e-01
3.40741053e-02 -2.23772395e-02 3.34646553e-01 9.03073624e-02
6.09792352e-01 -1.67943567e-01 -5.69742732e-02 -1.16778040e+00
4.39361110e-02 -9.08818007e-01 -4.45298702e-01 -5.77081621e-01
1.95870310e-01 -1.11686146e+00 1.40115097e-01 -1.92783284e+00
2.95846075e-01 -5.77839792e-01 -6.73629761e-01 7.89414287e-01
-1.81134745e-01 7.17156351e-01 -1.03235528e-01 3.22008371e-01
-5.99177718e-01 3.49028349e-01 1.42322147e+00 -2.95930266e-01
-1.49697319e-01 -1.45864591e-01 -3.28493297e-01 1.04200673e+00
9.62577462e-01 -3.77129853e-01 -2.55583704e-01 -3.19912076e-01
-4.65803385e-01 1.52421013e-01 3.50037664e-01 -1.21734202e+00
3.73180687e-01 -2.00223148e-01 6.74382627e-01 -7.38257170e-01
2.36630142e-01 -5.87560654e-01 -1.59933880e-01 6.39182627e-01
-5.29840231e-01 -1.54673636e-01 1.40457183e-01 2.41784751e-01
-4.08995807e-01 -1.56948254e-01 1.15274096e+00 -4.10298288e-01
-6.98280215e-01 7.14705348e-01 -4.30032879e-01 -1.13857351e-01
1.16374958e+00 -6.41028583e-02 -1.93130195e-01 1.19317129e-01
-1.21829629e+00 1.47412196e-01 1.26049489e-01 1.47132680e-01
7.04104722e-01 -1.09828913e+00 -5.05299091e-01 3.67915072e-02
8.50185007e-02 4.21909511e-01 3.77272338e-01 1.32717669e+00
-6.93709195e-01 5.09997785e-01 -2.52733439e-01 -6.89838707e-01
-1.19763434e+00 7.54293101e-03 6.18021131e-01 -4.63082790e-01
-9.17956352e-01 9.77642596e-01 3.28518003e-01 -1.96243390e-01
2.41943434e-01 -4.10949588e-01 -5.92482150e-01 -2.81683773e-01
3.19450319e-01 6.06331974e-03 4.53223914e-01 -6.02426827e-01
-3.34637403e-01 3.36138695e-01 -3.58435124e-01 -1.28280669e-01
1.12394404e+00 2.05477446e-01 -1.34118438e-01 -1.15503088e-01
8.44447672e-01 -6.48536861e-01 -1.08845043e+00 -1.98172465e-01
2.83926159e-01 -2.03263566e-01 7.15305090e-01 -8.67426276e-01
-1.45270169e+00 8.99665475e-01 9.68171239e-01 -1.19308420e-01
1.26067865e+00 1.16569757e-01 9.97048855e-01 1.20007321e-01
3.93418163e-01 -1.26399088e+00 9.73463245e-03 2.22666830e-01
5.50342441e-01 -1.31745899e+00 -3.10328215e-01 -7.76847184e-01
-7.70832360e-01 8.88314307e-01 9.67708588e-01 2.02788755e-01
7.55137205e-01 5.02549767e-01 5.19822687e-02 -3.03755939e-01
-2.61595815e-01 -4.05871123e-01 3.85603577e-01 3.59507710e-01
3.03937107e-01 4.80870098e-01 -4.22604412e-01 8.16861629e-01
7.11813793e-02 1.26035959e-01 2.28685662e-01 1.03752828e+00
-5.47120571e-01 -8.96129191e-01 -7.18089268e-02 8.41184855e-01
-6.03923440e-01 -2.19907656e-01 -3.62790972e-01 3.72185141e-01
1.47519307e-02 5.65264761e-01 6.40856847e-02 -1.75372556e-01
5.80440536e-02 1.18024364e-01 5.61664581e-01 -8.03329289e-01
-4.45682019e-01 2.92986065e-01 -1.74835384e-01 -5.07431984e-01
-4.69621420e-01 -4.22744632e-01 -1.79318154e+00 4.29154009e-01
-2.00429276e-01 4.92722727e-02 7.88107574e-01 1.24609554e+00
1.53972059e-01 9.78758931e-01 2.76152909e-01 -7.92254746e-01
2.10912153e-02 -1.00608289e+00 -2.68037319e-01 3.22289973e-01
-8.61695930e-02 -7.80090630e-01 1.39188737e-01 -7.92495087e-02] | [14.726114273071289, -2.1976664066314697] |
6a960d3b-6452-4954-9ba6-274c46d3ca32 | joint-symmetry-detection-and-shape-matching | 2112.02713 | null | https://arxiv.org/abs/2112.02713v2 | https://arxiv.org/pdf/2112.02713v2.pdf | Joint Symmetry Detection and Shape Matching for Non-Rigid Point Cloud | Despite the success of deep functional maps in non-rigid 3D shape matching, there exists no learning framework that models both self-symmetry and shape matching simultaneously. This is despite the fact that errors due to symmetry mismatch are a major challenge in non-rigid shape matching. In this paper, we propose a novel framework that simultaneously learns both self symmetry as well as a pairwise map between a pair of shapes. Our key idea is to couple a self symmetry map and a pairwise map through a regularization term that provides a joint constraint on both of them, thereby, leading to more accurate maps. We validate our method on several benchmarks where it outperforms many competitive baselines on both tasks. | ['Maks Ovsjanikov', 'Abhishek Sharma'] | 2021-12-05 | null | null | null | null | ['symmetry-detection'] | ['computer-vision'] | [-6.76626712e-03 -1.40477687e-01 -1.50371029e-03 -5.96444130e-01
-8.41137886e-01 -8.72839928e-01 7.22728431e-01 1.40541136e-01
-2.53487825e-01 1.55202717e-01 4.27001685e-01 -7.30719343e-02
-1.26768366e-01 -8.06055903e-01 -8.86259377e-01 -4.77302372e-01
4.01851803e-01 6.99837208e-01 2.49784097e-01 -3.11523229e-01
1.49908304e-01 8.75375032e-01 -1.09912205e+00 1.50041685e-01
7.61283934e-01 7.83167124e-01 -1.66309938e-01 4.78159972e-02
-5.28807640e-02 2.47870591e-02 -4.05956479e-03 -6.25031948e-01
7.41055012e-01 -9.29216295e-02 -8.02927077e-01 -1.61623165e-01
1.02156746e+00 -1.92692742e-01 -6.43185675e-01 1.02105224e+00
6.42689824e-01 3.04174840e-01 7.65383422e-01 -9.42383826e-01
-6.42934561e-01 2.73618102e-01 -6.71180904e-01 -1.59770876e-01
3.54848832e-01 2.90329512e-02 1.25518167e+00 -8.15785885e-01
8.12174559e-01 1.35855877e+00 8.84263337e-01 3.60033303e-01
-1.38855064e+00 -6.54514790e-01 -6.40981197e-02 -2.48232484e-01
-1.26059091e+00 -5.34620702e-01 1.01120913e+00 -4.91715938e-01
9.29010749e-01 5.14899418e-02 4.98738766e-01 7.31768966e-01
1.04104057e-01 3.95258397e-01 8.44259501e-01 -8.52980185e-03
-8.35404545e-02 -3.83171707e-01 -3.48050922e-01 5.48322678e-01
-3.07344552e-02 3.02706361e-01 -3.02123189e-01 -2.00554505e-01
1.03447640e+00 2.04560414e-01 -4.01502140e-02 -7.87380695e-01
-1.12448549e+00 5.85469067e-01 8.99789870e-01 3.11183453e-01
-1.90060884e-01 1.70885772e-01 7.88350925e-02 2.32435986e-01
4.58660662e-01 4.97390121e-01 -1.62958488e-01 8.07829276e-02
-8.63919020e-01 6.19531095e-01 5.54555833e-01 7.26067781e-01
8.44693065e-01 -2.95094639e-01 -2.26481453e-01 7.50171304e-01
3.84875596e-01 5.25399685e-01 1.23357773e-01 -6.79883599e-01
5.13721049e-01 9.67191696e-01 -1.23893330e-02 -1.27013981e+00
-4.31217939e-01 -5.09136975e-01 -8.09912026e-01 2.96872944e-01
4.60622787e-01 5.05404234e-01 -8.99039328e-01 1.99687433e+00
3.90740007e-01 2.32676700e-01 -3.16608578e-01 1.10148180e+00
9.19341445e-01 1.45701826e-01 -1.52441338e-01 4.78331864e-01
9.88610506e-01 -9.02235806e-01 -2.68734664e-01 -3.59951220e-02
2.42129028e-01 -1.10859692e+00 7.43596852e-01 -3.04683089e-01
-1.37408090e+00 -3.72461349e-01 -9.84443486e-01 -5.23726285e-01
-1.02587804e-01 -1.75261587e-01 5.46322107e-01 7.13629797e-02
-9.91574943e-01 8.02204132e-01 -8.08104575e-01 -2.38523051e-01
3.34599018e-01 4.81999934e-01 -7.22614348e-01 -7.97219388e-03
-7.87723839e-01 9.80753183e-01 -5.25541268e-02 3.92714031e-02
-2.75209844e-01 -9.19939458e-01 -1.04781699e+00 -1.76676773e-02
5.77382036e-02 -8.74709845e-01 9.03558254e-01 -4.97217566e-01
-1.33395207e+00 1.38713884e+00 -2.61680812e-01 -1.81942016e-01
1.03673160e+00 -8.10302421e-03 7.49280229e-02 -2.24499404e-01
-8.03282410e-02 8.80913377e-01 6.50216699e-01 -1.21325147e+00
5.22212908e-02 -6.03606105e-01 1.42090768e-01 3.78679693e-01
2.82179564e-02 -3.87027897e-02 -5.51992357e-01 -7.55026519e-01
6.57108903e-01 -9.57457364e-01 -1.84375733e-01 2.92333543e-01
-4.01361048e-01 -4.19667423e-01 6.54961050e-01 -6.88917100e-01
7.84265578e-01 -1.97347093e+00 5.62137485e-01 4.81218100e-01
3.48501176e-01 5.93242869e-02 -4.12990391e-01 3.54409575e-01
-1.97945595e-01 5.80946505e-02 -4.65303153e-01 -5.24633467e-01
2.02405170e-01 7.40249082e-02 -3.77748251e-01 6.12338126e-01
3.18194002e-01 1.39691460e+00 -7.52138436e-01 -2.56825890e-02
2.52408832e-01 7.94321358e-01 -6.00249887e-01 2.56286263e-01
-2.08736956e-02 6.50489450e-01 -2.28352368e-01 6.78360462e-01
1.02197337e+00 -7.82662705e-02 -2.06403621e-02 -5.86701512e-01
-7.21277520e-02 5.49372971e-01 -1.18884206e+00 1.95636344e+00
-2.28723750e-01 8.13161433e-02 8.67282525e-02 -9.19773877e-01
1.27539396e+00 1.99146330e-01 7.27973104e-01 -9.46925700e-01
4.55135629e-02 4.54286486e-01 -9.25351679e-02 6.86702132e-02
2.43341461e-01 -3.08290631e-01 2.67841399e-01 5.77593863e-01
-8.70908648e-02 -4.34524715e-01 -8.19108188e-02 -1.74661726e-01
1.00246429e+00 4.08883363e-01 5.44447042e-02 -4.30103093e-01
4.35291529e-01 -5.36889434e-01 6.22366726e-01 3.59624922e-01
1.63537003e-02 1.10709047e+00 2.36270964e-01 -8.42140317e-01
-1.36811280e+00 -1.27109122e+00 -2.26810709e-01 4.56084311e-01
3.43525082e-01 -2.89768696e-01 -4.56866860e-01 -6.11153901e-01
5.98832965e-01 -9.30621102e-02 -5.14780283e-01 -1.73069492e-01
-9.88272786e-01 -3.86953115e-01 3.16950738e-01 6.26420557e-01
3.24300706e-01 -8.48105013e-01 -1.25208750e-01 1.02820948e-01
2.57270355e-02 -1.21795046e+00 -1.00630653e+00 -1.79635882e-01
-1.00036335e+00 -1.11829412e+00 -7.37023771e-01 -6.64315224e-01
6.72734737e-01 4.74085122e-01 1.16504645e+00 4.43112254e-01
-1.60647616e-01 1.49062164e-02 1.98075339e-01 -3.32556143e-02
-3.82916540e-01 2.42460966e-01 -2.59544626e-02 -7.17272237e-02
1.28724307e-01 -9.59961593e-01 -6.84393942e-01 5.60729861e-01
-8.33755910e-01 1.44676283e-01 4.75378454e-01 5.10487378e-01
8.23404491e-01 -4.71048385e-01 2.80716270e-01 -5.48539340e-01
3.15159857e-01 -1.55413747e-01 -7.53141642e-01 3.46063614e-01
-4.57872480e-01 3.39105994e-01 2.34315977e-01 -1.14503734e-01
-6.43563926e-01 3.53353739e-01 -4.13972318e-01 -5.55312514e-01
-2.81231496e-02 1.50466174e-01 -2.57596970e-01 -5.24676204e-01
4.20137435e-01 -1.01505024e-02 1.19656302e-01 -8.36847663e-01
2.01319933e-01 1.79801241e-01 6.49689853e-01 -7.82356143e-01
1.24382305e+00 6.00511730e-01 3.53207648e-01 -3.18585634e-01
-7.58285940e-01 -5.12240648e-01 -9.39279020e-01 2.56690532e-01
6.38360500e-01 -9.41486001e-01 -7.71133780e-01 5.25675178e-01
-1.21951854e+00 -1.25449181e-01 -1.91379219e-01 2.11795971e-01
-6.45854115e-01 5.00706732e-01 -4.72594768e-01 -2.08067432e-01
-5.57854652e-01 -1.23161209e+00 1.30182779e+00 1.45899802e-01
-1.28463417e-01 -8.93444836e-01 3.57186228e-01 3.23136449e-01
6.11088097e-01 3.81661743e-01 8.37732017e-01 -4.53214258e-01
-6.68720841e-01 -4.11723346e-01 -5.69646657e-01 -9.35540907e-03
3.81925166e-01 1.48170395e-02 -6.90178573e-01 -5.31574011e-01
-1.44881055e-01 -1.70071676e-01 1.01162493e+00 1.28118694e-01
8.76117229e-01 1.49722379e-02 -2.52790242e-01 1.06586516e+00
1.28495347e+00 -1.33276314e-01 6.22861803e-01 1.69210285e-01
9.99701500e-01 5.15957654e-01 3.93056646e-02 -1.27010196e-02
5.11182606e-01 1.26376569e+00 4.15686011e-01 -2.71285623e-01
-2.67980188e-01 -4.71976846e-01 -1.26055196e-01 8.89316201e-01
-3.99952203e-01 2.17451274e-01 -9.22957540e-01 3.87128502e-01
-1.96624553e+00 -7.12778270e-01 -6.36014864e-02 2.48937678e+00
8.37302983e-01 -2.95235459e-02 7.60800615e-02 -2.58686066e-01
6.66413784e-01 1.97133824e-01 -6.52228177e-01 -6.93258718e-02
-3.78359288e-01 4.03973699e-01 2.81061023e-01 6.42893493e-01
-1.25780296e+00 7.69846141e-01 6.49205637e+00 5.06356418e-01
-1.18926561e+00 -9.64235738e-02 2.36057416e-01 1.50678337e-01
-7.79314816e-01 1.19885176e-01 -5.86007118e-01 2.91601151e-01
1.86440516e-02 -4.42976095e-02 5.14263988e-01 4.98193234e-01
-6.85306340e-02 4.60775673e-01 -1.16770029e+00 9.97245729e-01
1.89690031e-02 -1.36260200e+00 2.96535075e-01 1.70827150e-01
9.31363642e-01 6.52993619e-02 -5.22954315e-02 -1.67399943e-01
1.48437440e-01 -1.20844722e+00 6.71329498e-01 4.20965433e-01
6.62060678e-01 -5.99395812e-01 5.52091479e-01 4.63453643e-02
-1.41564500e+00 4.84642595e-01 -5.24146318e-01 1.89017132e-01
1.18322052e-01 3.98452848e-01 -2.13379458e-01 7.96878159e-01
4.37385738e-01 8.69043171e-01 -3.33655864e-01 1.26811004e+00
-3.63040328e-01 -1.38131887e-01 -4.16958153e-01 7.17455029e-01
9.49125960e-02 -5.06292164e-01 6.88080609e-01 9.64007318e-01
3.05622399e-01 -2.15560365e-02 2.74896383e-01 1.22941864e+00
-3.60144049e-01 1.40481398e-01 -6.15753412e-01 1.95417508e-01
4.05184209e-01 1.33671653e+00 -5.71086824e-01 1.81611329e-01
-3.78955692e-01 9.39285517e-01 6.75422370e-01 2.12410048e-01
-5.26237190e-01 1.50279522e-01 1.03123021e+00 3.07920426e-01
1.20520152e-01 -4.94874060e-01 -4.17577595e-01 -1.24454331e+00
3.34587753e-01 -6.67459309e-01 2.09519476e-01 -5.13260603e-01
-1.53091836e+00 3.96563411e-01 -3.97177309e-01 -1.13863158e+00
-3.01147662e-02 -4.92139280e-01 -8.20052326e-01 1.19740105e+00
-1.70522416e+00 -1.58965266e+00 -3.03210169e-01 5.01165032e-01
-1.74661987e-02 3.76249515e-02 8.17627311e-01 5.28933108e-01
-9.97388512e-02 7.81718075e-01 -1.24618798e-01 1.55845568e-01
9.41064000e-01 -1.31334913e+00 9.29694891e-01 5.04743457e-01
3.35625798e-01 8.35940540e-01 4.18012947e-01 -5.21954417e-01
-1.52932453e+00 -7.15273321e-01 9.90172446e-01 -5.20590723e-01
3.25695485e-01 -3.85078758e-01 -1.29207861e+00 5.51315188e-01
-1.67726219e-01 3.47922653e-01 1.88938737e-01 9.86029059e-02
-9.25922334e-01 -5.45870103e-02 -1.06552017e+00 3.42137903e-01
1.40621698e+00 -7.33105481e-01 -5.70200861e-01 9.68352705e-02
5.60959637e-01 -8.72414708e-01 -9.32791293e-01 8.32239747e-01
7.74805188e-01 -9.31955159e-01 1.24471748e+00 -6.65650308e-01
2.73604959e-01 -3.77676755e-01 -1.40658110e-01 -1.17222834e+00
-3.52311671e-01 -5.33573687e-01 1.18410483e-01 1.10732949e+00
1.72172606e-01 -5.15729785e-01 8.45374584e-01 7.95374095e-01
-1.87016636e-01 -7.06232309e-01 -1.08121419e+00 -7.68834889e-01
4.02966619e-01 9.49114561e-03 7.93580413e-01 1.02337503e+00
-3.95574450e-01 1.07050389e-01 -2.60356188e-01 -1.23752274e-01
6.31790936e-01 5.71445405e-01 8.31896126e-01 -1.46042943e+00
-1.38095319e-01 -8.57447445e-01 -6.48726165e-01 -9.95945275e-01
3.52120161e-01 -1.27839959e+00 -5.69449253e-02 -1.38900566e+00
4.87197310e-01 -6.70493901e-01 -2.93231398e-01 6.49654567e-01
-9.70696807e-02 4.97324973e-01 1.48557901e-01 3.39064479e-01
-2.51827866e-01 7.03241944e-01 1.44638550e+00 -1.19705871e-01
-2.47547090e-01 -3.22746225e-02 -4.74973261e-01 6.27551973e-01
5.61986625e-01 -4.49750960e-01 -6.77975118e-02 -6.84295237e-01
3.76593351e-01 -1.97139546e-01 5.82519889e-01 -7.07237422e-01
2.63002425e-01 -5.91480127e-03 2.57025331e-01 -5.44203162e-01
2.34976664e-01 -7.45549619e-01 2.76986390e-01 2.46283680e-01
-1.23142160e-01 2.86649108e-01 1.54502809e-01 3.44561487e-01
-2.57205337e-01 6.36148974e-02 8.19116831e-01 -4.17555235e-02
-2.56191343e-01 7.09045768e-01 7.21864045e-01 1.90678254e-01
4.92872298e-01 -1.14445686e-01 -1.93492085e-01 -2.45359182e-01
-5.57646632e-01 1.38124868e-01 9.36418653e-01 8.01888704e-01
5.65635741e-01 -1.72061265e+00 -8.69050860e-01 3.82777989e-01
5.54671884e-02 2.32330427e-01 1.50947720e-01 8.51660430e-01
-3.55950475e-01 3.15352440e-01 -3.03928971e-01 -6.43105149e-01
-1.23410141e+00 2.53153235e-01 7.53747344e-01 -2.36435890e-01
-8.41536939e-01 5.88834047e-01 4.26717013e-01 -1.05869782e+00
2.43611127e-01 -1.19559132e-01 1.25971392e-01 -3.56289059e-01
2.68961370e-01 1.63985804e-01 3.42405766e-01 -1.06818461e+00
-4.30925369e-01 1.26569021e+00 1.27774447e-01 7.58197084e-02
1.41854477e+00 1.82183787e-01 -3.54688942e-01 1.38347089e-01
1.43304801e+00 1.16993703e-01 -1.38807929e+00 -6.28634095e-01
-6.28633648e-02 -6.47699058e-01 -2.91620791e-01 -5.87264597e-01
-1.18758774e+00 9.30113316e-01 4.31294084e-01 -2.23600239e-01
6.30838692e-01 1.14489824e-01 1.01005101e+00 2.46874645e-01
2.07877830e-01 -6.13718212e-01 7.17223063e-02 6.54889405e-01
1.31192982e+00 -1.29135334e+00 4.78336178e-02 -5.04845202e-01
-1.39385700e-01 1.09712076e+00 5.53051174e-01 -4.38177913e-01
6.37465119e-01 2.10174933e-01 -1.78200584e-02 -3.29994410e-01
-3.02370548e-01 -2.58572757e-01 1.02534986e+00 3.34272057e-01
6.88080490e-01 2.14304477e-02 -3.21175635e-01 2.05433130e-01
-2.78480440e-01 -2.14547515e-01 -2.54003882e-01 6.19699299e-01
-8.58098045e-02 -1.60314226e+00 -1.35156319e-01 1.33866861e-01
-3.64278316e-01 1.89564094e-01 -6.43997848e-01 5.43465197e-01
-1.65045306e-01 3.61713320e-01 1.54742375e-01 -3.86819333e-01
6.88540578e-01 -1.21265575e-01 7.95906007e-01 -4.34303731e-01
-7.05318153e-01 1.66400760e-01 -4.03312176e-01 -6.76253140e-01
-4.51496929e-01 -5.45252621e-01 -1.23208177e+00 -5.19362211e-01
4.59280014e-02 -2.57899374e-01 7.00429976e-01 9.34230745e-01
4.46333081e-01 5.02636172e-02 4.97016579e-01 -9.65250731e-01
-7.81461596e-01 -6.35413945e-01 -3.07651311e-01 1.04379010e+00
1.06775142e-01 -7.91612566e-01 -2.12570161e-01 -4.52675164e-01] | [8.444252967834473, -2.6582179069519043] |
e9e08a0a-1a0c-440f-a95b-3afc37a6302c | hdr-video-reconstruction-with-a-large-dynamic | 2304.04773 | null | https://arxiv.org/abs/2304.04773v2 | https://arxiv.org/pdf/2304.04773v2.pdf | HDR Video Reconstruction with a Large Dynamic Dataset in Raw and sRGB Domains | High dynamic range (HDR) video reconstruction is attracting more and more attention due to the superior visual quality compared with those of low dynamic range (LDR) videos. The availability of LDR-HDR training pairs is essential for the HDR reconstruction quality. However, there are still no real LDR-HDR pairs for dynamic scenes due to the difficulty in capturing LDR-HDR frames simultaneously. In this work, we propose to utilize a staggered sensor to capture two alternate exposure images simultaneously, which are then fused into an HDR frame in both raw and sRGB domains. In this way, we build a large scale LDR-HDR video dataset with 85 scenes and each scene contains 60 frames. Based on this dataset, we further propose a Raw-HDRNet, which utilizes the raw LDR frames as inputs. We propose a pyramid flow-guided deformation convolution to align neighboring frames. Experimental results demonstrate that 1) the proposed dataset can improve the HDR reconstruction performance on real scenes for three benchmark networks; 2) Compared with sRGB inputs, utilizing raw inputs can further improve the reconstruction quality and our proposed Raw-HDRNet is a strong baseline for raw HDR reconstruction. Our dataset and code will be released after the acceptance of this paper. | ['Jingyu Yang', 'Zhenyu Zhou', 'Xuanwu Yin', 'Biting Yu', 'Yubo Peng', 'Huanjing Yue'] | 2023-04-10 | null | null | null | null | ['video-reconstruction', 'hdr-reconstruction'] | ['computer-vision', 'computer-vision'] | [ 2.49107152e-01 -3.97926062e-01 1.73615664e-02 -2.88073361e-01
-4.40889776e-01 -1.86375871e-01 3.47142369e-01 -7.46068358e-01
-3.31706971e-01 6.42682731e-01 4.18369114e-01 -1.01376869e-01
1.02389172e-01 -8.08504164e-01 -8.90698075e-01 -7.43897796e-01
2.56208241e-01 -1.18847407e-01 5.41075468e-01 -3.58572513e-01
-1.66519150e-01 5.91706038e-01 -1.47852361e+00 3.19431633e-01
7.05448270e-01 1.02738822e+00 7.92121887e-01 7.01198518e-01
1.80317491e-01 1.10720849e+00 -2.45381385e-01 -1.05980337e-01
6.91667497e-01 -4.44540232e-01 -5.03681302e-01 -8.55947379e-03
3.49339992e-01 -1.08494639e+00 -1.13796067e+00 8.60843897e-01
6.62765980e-01 4.39156324e-01 2.50351569e-03 -9.48951244e-01
-8.05490077e-01 4.69683498e-01 -8.38690042e-01 3.43899637e-01
5.93953431e-01 4.81447130e-01 6.25747621e-01 -8.06931853e-01
8.75706375e-01 1.32089400e+00 4.20660734e-01 5.87648928e-01
-9.12955523e-01 -9.36732948e-01 -2.34005619e-02 2.62540430e-01
-1.32040596e+00 -5.81300139e-01 7.69952476e-01 1.29108317e-03
8.82225931e-01 1.75895631e-01 7.83373356e-01 1.07637715e+00
-5.90118393e-02 5.91148973e-01 1.15431070e+00 1.15614384e-02
-1.12789541e-01 -4.92877245e-01 -2.41208255e-01 3.37602645e-01
1.35792673e-01 5.48916280e-01 -4.06225443e-01 5.02728522e-01
1.40459538e+00 4.80994225e-01 -8.39442253e-01 7.37862289e-02
-1.49391568e+00 3.83514345e-01 6.26924217e-01 2.37521991e-01
-1.27809107e-01 1.84650034e-01 2.66319990e-01 4.41007435e-01
2.72555411e-01 -4.46073562e-02 -2.58499105e-02 -7.29902461e-02
-7.89609909e-01 6.57678470e-02 2.89419860e-01 9.92757857e-01
7.59008527e-01 1.80294380e-01 -7.76758268e-02 9.64565337e-01
3.49729151e-01 8.17397058e-01 4.03079748e-01 -1.17066789e+00
6.67115450e-01 2.09032893e-01 1.25679702e-01 -1.20619333e+00
-2.14268535e-01 2.02004090e-02 -1.33848035e+00 5.70411682e-02
1.21856794e-01 2.07204640e-01 -8.42493534e-01 1.35073090e+00
1.95439130e-01 4.61627305e-01 2.90040225e-01 1.47091222e+00
1.23780394e+00 1.22523212e+00 -3.06523710e-01 -4.83243287e-01
9.73789632e-01 -8.64826262e-01 -8.99426579e-01 9.05066077e-03
1.22839056e-01 -7.83961117e-01 1.14038920e+00 3.73034924e-01
-1.16092908e+00 -8.70183289e-01 -1.05646646e+00 -4.12810296e-01
2.53674686e-01 -9.61058587e-02 3.76114786e-01 2.67293572e-01
-1.02731538e+00 4.13332075e-01 -7.55927920e-01 -8.81186798e-02
2.91823328e-01 1.58017412e-01 -5.52628100e-01 -8.00128281e-01
-1.45552361e+00 5.74698389e-01 5.20550609e-01 3.87879729e-01
-1.00433910e+00 -7.13423908e-01 -7.89914131e-01 -2.29583949e-01
2.70617783e-01 -4.59168404e-01 7.27274954e-01 -6.91734314e-01
-1.58703375e+00 6.19984329e-01 1.88988134e-01 -2.70046294e-01
6.14049077e-01 -1.40282497e-01 -5.39280117e-01 5.04979670e-01
-1.78914487e-01 7.51677275e-01 7.06833780e-01 -1.31504595e+00
-5.92326283e-01 -1.46985933e-01 5.31034693e-02 2.90835798e-01
-9.17161033e-02 7.20389187e-02 -8.07359874e-01 -8.09929609e-01
1.44433632e-01 -8.37523937e-01 -2.52361353e-02 5.10550477e-02
-2.24773705e-01 2.43381351e-01 1.18893158e+00 -9.28895354e-01
1.03208578e+00 -2.18830276e+00 1.27898872e-01 -1.70864999e-01
4.01771456e-01 5.02990484e-01 -2.45284036e-01 5.87785505e-02
-1.68574661e-01 -1.64493978e-01 -1.63002517e-02 -2.83311848e-02
-3.75391454e-01 1.52642220e-01 -6.14348650e-01 6.57145917e-01
-1.25680566e-01 1.02240896e+00 -7.94772804e-01 -4.81133848e-01
7.73292720e-01 7.93634176e-01 -3.13153028e-01 7.19115973e-01
-9.74392053e-04 8.21198344e-01 -2.75248736e-01 4.48366880e-01
1.06487811e+00 -1.58520713e-01 -1.85910109e-02 -7.58992255e-01
-1.40513569e-01 -1.51697233e-01 -1.12410736e+00 1.76254249e+00
-5.49557149e-01 8.04092586e-01 -1.26770079e-01 -6.81379735e-01
1.24943340e+00 4.66824658e-02 6.13795817e-01 -1.05135965e+00
1.91696212e-02 2.98198313e-01 -6.92573264e-02 -6.19830310e-01
5.90724826e-01 -1.12230100e-01 1.69992134e-01 2.23484322e-01
-2.50416070e-01 1.20743213e-03 1.83524907e-01 1.92537159e-01
1.07649231e+00 2.03862920e-01 1.38777450e-01 2.30969369e-01
5.07102787e-01 -5.80436707e-01 7.59355545e-01 3.21418345e-01
-1.27683640e-01 1.10575128e+00 1.52253568e-01 -6.30138040e-01
-1.40545678e+00 -1.19020700e+00 -7.65093863e-02 6.36736214e-01
7.44329751e-01 -1.55017555e-01 -1.69635966e-01 -2.31549993e-01
-3.14658731e-01 2.61402041e-01 -3.32185954e-01 -1.19118802e-01
-1.01724744e+00 -5.40527225e-01 4.36199009e-01 4.96283650e-01
1.23423135e+00 -1.16349649e+00 -6.11837685e-01 6.87383488e-02
-5.48461795e-01 -1.37056899e+00 -6.56029165e-01 -2.54471928e-01
-7.48046219e-01 -8.49745214e-01 -1.03870821e+00 -6.33723378e-01
3.08369190e-01 8.78459156e-01 9.64491665e-01 1.26425698e-01
-2.12778494e-01 8.16635713e-02 -6.10601962e-01 3.97771776e-01
-4.65449631e-01 -3.68693560e-01 -7.06033707e-02 -3.33101414e-02
-9.29329619e-02 -6.91930890e-01 -9.87874687e-01 6.19865119e-01
-1.26823664e+00 5.41465044e-01 6.91198409e-01 6.50444269e-01
7.43575752e-01 1.29706830e-01 3.98189396e-01 -5.46088219e-01
1.49013745e-02 -3.19633454e-01 -5.17774403e-01 2.20314518e-01
-2.07583845e-01 -2.76235580e-01 9.04378831e-01 -4.63218361e-01
-1.31504500e+00 -1.46570906e-01 -3.67400408e-01 -1.11274421e+00
-5.01966886e-02 -1.29424214e-01 -3.54135007e-01 -1.04026916e-02
1.87690124e-01 4.61086094e-01 1.04929902e-01 -3.89309883e-01
3.16794455e-01 6.97190344e-01 7.30651200e-01 -4.20820922e-01
8.80915105e-01 5.90839446e-01 -1.54584572e-01 -7.63050795e-01
-5.03489733e-01 -3.67220402e-01 -6.93182573e-02 -4.28500652e-01
1.05276060e+00 -1.31317186e+00 -7.19281912e-01 7.16993511e-01
-1.00739956e+00 -5.01098871e-01 -1.70955345e-01 6.59852803e-01
-5.24111569e-01 5.61647236e-01 -9.95418131e-01 -3.69875729e-01
-4.07623410e-01 -1.38015437e+00 1.14119792e+00 3.63187045e-01
5.73979735e-01 -6.53894782e-01 -1.15226312e-02 4.39162612e-01
5.32388389e-01 2.58494884e-01 2.98417896e-01 8.65331888e-02
-9.96746898e-01 2.97209471e-01 -6.70024574e-01 3.30653220e-01
1.38060674e-01 -1.28929019e-01 -6.92357659e-01 -5.23376107e-01
6.53632507e-02 -3.69021654e-01 9.39504445e-01 4.49441850e-01
1.27070904e+00 -2.26332590e-01 -8.37368742e-02 1.17659926e+00
1.60517573e+00 3.33420813e-01 1.27374816e+00 4.14140671e-01
1.19432569e+00 3.42989475e-01 9.59209681e-01 4.14800227e-01
2.54204988e-01 1.00084376e+00 3.89288247e-01 -3.17514420e-01
-6.25245631e-01 -3.40840131e-01 6.39968574e-01 1.03579974e+00
-3.49066138e-01 -3.76481175e-01 -4.95834351e-01 1.25965446e-01
-1.61564183e+00 -1.18174708e+00 -1.59161508e-01 1.98314989e+00
8.52525473e-01 -2.11582169e-01 -5.59353605e-02 -9.51422453e-02
8.41320992e-01 6.59792721e-01 -7.38849759e-01 2.02050105e-01
-4.30494666e-01 -8.19641352e-02 5.21818638e-01 2.43287787e-01
-7.22226620e-01 6.45650625e-01 5.20108652e+00 9.49718595e-01
-1.34168398e+00 7.26715326e-02 8.20942521e-01 -2.32600316e-01
-3.58011365e-01 -1.81951940e-01 -7.67264426e-01 7.35599637e-01
6.52336061e-01 1.69961434e-02 6.99719667e-01 5.22550404e-01
5.14700234e-01 2.36089639e-02 -8.27902019e-01 1.51917267e+00
7.80595541e-02 -1.30728328e+00 1.17629096e-01 8.51688012e-02
9.27478433e-01 -9.80798304e-02 1.21292762e-01 1.54660761e-01
2.59673655e-01 -1.00257301e+00 5.97181857e-01 4.83305454e-01
1.35434973e+00 -7.48315692e-01 6.19501472e-01 1.15698893e-02
-1.45637715e+00 2.19442416e-02 -7.90242255e-01 2.90070862e-01
4.60963964e-01 6.58576965e-01 -1.97318137e-01 7.50165820e-01
1.05934739e+00 1.42568135e+00 -4.99632865e-01 5.90427816e-01
6.44092485e-02 2.00137019e-01 -1.01314813e-01 5.47629654e-01
-1.82266906e-01 -3.55034709e-01 4.13603485e-01 9.43226397e-01
4.87393439e-01 5.09028852e-01 1.36618689e-01 7.14162886e-01
-2.04137787e-01 -1.42258942e-01 -7.63038516e-01 2.46548474e-01
4.09302026e-01 1.23423254e+00 -4.95124727e-01 -2.01160684e-01
-5.66015244e-01 1.10728526e+00 -6.85705710e-03 3.86240304e-01
-1.12962627e+00 -3.45738977e-01 5.82643390e-01 1.05381772e-01
3.82860899e-01 -1.06398001e-01 2.94247836e-01 -1.61472094e+00
-4.96039540e-02 -1.07606876e+00 1.88759267e-01 -1.17555964e+00
-1.16937602e+00 7.64600396e-01 -7.84199983e-02 -1.57048154e+00
-5.30505925e-02 -2.34439984e-01 -2.98896879e-01 5.93328953e-01
-1.71427512e+00 -9.61235583e-01 -9.26945567e-01 8.74995112e-01
6.55787528e-01 -6.16828948e-02 6.65227175e-02 7.18129516e-01
-5.90754211e-01 3.20660025e-01 2.20623791e-01 2.07881883e-01
8.55684102e-01 -7.97093928e-01 2.91120648e-01 9.93156135e-01
-1.90880790e-01 5.16566694e-01 5.97880781e-01 -6.60601199e-01
-1.71233964e+00 -1.41092014e+00 -9.49993059e-02 3.70045900e-02
1.86672345e-01 -8.33318979e-02 -1.05427897e+00 6.17996395e-01
4.10254151e-02 4.80622143e-01 1.29974306e-01 -7.51616061e-01
-2.99327016e-01 -6.07777417e-01 -1.13362861e+00 4.10692871e-01
1.32847178e+00 -4.16742921e-01 -3.84778649e-01 -2.54493579e-02
1.24069715e+00 -6.45210862e-01 -1.17202210e+00 4.64655370e-01
4.98854369e-01 -1.33814180e+00 1.25783157e+00 3.04415315e-01
8.60450029e-01 -7.23331153e-01 -4.62667614e-01 -1.00723672e+00
-3.38210501e-02 -4.78921533e-01 -1.11961320e-01 1.32572496e+00
-4.65718269e-01 -5.80847919e-01 5.08546948e-01 4.02212709e-01
-1.35009512e-01 -7.15203881e-01 -6.31483316e-01 -7.40411401e-01
-3.71964663e-01 -2.17732221e-01 6.42996490e-01 8.15320671e-01
-6.23954594e-01 -1.15031302e-02 -9.82900560e-01 -2.75435764e-02
7.15149403e-01 3.84466678e-01 9.31924164e-01 -5.79853714e-01
-5.50081193e-01 8.16872716e-02 -5.12335658e-01 -1.48773122e+00
1.36107961e-02 -6.34718537e-01 1.75372392e-01 -1.47345233e+00
3.99319321e-01 -5.80285549e-01 -8.64021778e-02 5.47829308e-02
-1.79136142e-01 6.78276539e-01 4.40648615e-01 5.13204992e-01
-7.60641634e-01 7.40240455e-01 1.67475033e+00 1.06396876e-01
-1.83706015e-01 -5.34059346e-01 -2.68925667e-01 3.84930789e-01
4.62137997e-01 -1.49676185e-02 -4.50252742e-01 -4.77300316e-01
5.09934686e-02 5.12060404e-01 3.77604961e-01 -1.15848470e+00
-8.46316293e-02 -1.45436168e-01 6.96424723e-01 -6.26165807e-01
3.08870882e-01 -8.56435359e-01 8.53716910e-01 3.91804993e-01
-2.46371329e-01 4.30317149e-02 -1.85855925e-01 7.15627491e-01
-3.49060893e-01 2.79195547e-01 1.12579131e+00 -3.05130109e-02
-1.10371745e+00 6.89066947e-01 2.42176458e-01 3.24350260e-02
1.15386748e+00 -2.45463148e-01 -6.94528997e-01 -4.25871223e-01
-2.02103510e-01 -4.29012850e-02 8.07980120e-01 5.54939985e-01
1.15464568e+00 -1.50731146e+00 -7.24564135e-01 3.22253078e-01
-1.56237677e-01 3.30776066e-01 7.97890842e-01 7.64120102e-01
-9.97545362e-01 1.47102967e-01 -6.78088784e-01 -6.64566100e-01
-1.19908977e+00 7.89824009e-01 2.05665782e-01 -1.81304857e-01
-1.28867292e+00 3.45337152e-01 5.63560188e-01 -3.75494957e-02
4.77870405e-02 -2.18080282e-01 -1.94843799e-01 -4.09872383e-01
8.49250972e-01 5.06300867e-01 -3.03865224e-01 -8.29624832e-01
-7.82378614e-02 7.83191741e-01 -9.37226862e-02 5.27665503e-02
1.47788250e+00 -6.57871544e-01 -3.43927220e-02 2.75206685e-01
1.57798457e+00 -1.51076719e-01 -1.66340208e+00 -8.47022310e-02
-7.92662740e-01 -1.18297350e+00 7.83476382e-02 -2.01790705e-01
-1.76953125e+00 7.46555567e-01 8.30988050e-01 -1.67011634e-01
1.61132419e+00 -2.86548417e-02 1.40895355e+00 1.24199905e-01
5.12441099e-01 -5.88737607e-01 3.38330775e-01 8.80617425e-02
8.22923005e-01 -1.23783040e+00 9.40509811e-02 -4.05831695e-01
-6.58598244e-01 1.22842133e+00 6.86663508e-01 -2.02062637e-01
1.90147296e-01 1.45716086e-01 -9.51277930e-03 1.60788700e-01
-6.86375916e-01 -2.67731659e-02 -3.36898118e-02 6.09958827e-01
1.91503093e-01 -3.55582297e-01 -1.72461808e-01 5.78234270e-02
1.06842563e-01 2.38478169e-01 7.27858961e-01 4.52120006e-01
-3.02203268e-01 -8.35673928e-01 -4.05171603e-01 4.26240236e-01
-2.68129528e-01 5.99485673e-02 3.19257110e-01 7.51728058e-01
-2.63046264e-03 8.87197673e-01 2.38018576e-02 -5.87739587e-01
3.24622154e-01 -8.24991226e-01 5.24222732e-01 -1.07367970e-01
-1.31743386e-01 6.04737885e-02 -2.32903644e-01 -9.39644516e-01
-6.90361500e-01 -2.77547091e-01 -1.25863957e+00 -7.91412115e-01
-1.66309606e-02 -3.35703701e-01 3.12196225e-01 5.43319166e-01
1.94455639e-01 5.22280991e-01 9.15422440e-01 -1.06845391e+00
-2.28865057e-01 -6.05447233e-01 -7.90968299e-01 7.65776515e-01
4.74473029e-01 -5.55286884e-01 -4.33599025e-01 2.83794105e-01] | [10.887472152709961, -2.1245927810668945] |
5b2e7c64-b76b-4a5e-8381-17c7f7421a0e | indiscernible-object-counting-in-underwater | 2304.11677 | null | https://arxiv.org/abs/2304.11677v1 | https://arxiv.org/pdf/2304.11677v1.pdf | Indiscernible Object Counting in Underwater Scenes | Recently, indiscernible scene understanding has attracted a lot of attention in the vision community. We further advance the frontier of this field by systematically studying a new challenge named indiscernible object counting (IOC), the goal of which is to count objects that are blended with respect to their surroundings. Due to a lack of appropriate IOC datasets, we present a large-scale dataset IOCfish5K which contains a total of 5,637 high-resolution images and 659,024 annotated center points. Our dataset consists of a large number of indiscernible objects (mainly fish) in underwater scenes, making the annotation process all the more challenging. IOCfish5K is superior to existing datasets with indiscernible scenes because of its larger scale, higher image resolutions, more annotations, and denser scenes. All these aspects make it the most challenging dataset for IOC so far, supporting progress in this area. For benchmarking purposes, we select 14 mainstream methods for object counting and carefully evaluate them on IOCfish5K. Furthermore, we propose IOCFormer, a new strong baseline that combines density and regression branches in a unified framework and can effectively tackle object counting under concealed scenes. Experiments show that IOCFormer achieves state-of-the-art scores on IOCfish5K. | ['Luc van Gool', 'Deng-Ping Fan', 'Christos Sakaridis', 'Ce Liu', 'Yun Liu', 'Zhaochong An', 'Guolei Sun'] | 2023-04-23 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Sun_Indiscernible_Object_Counting_in_Underwater_Scenes_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Sun_Indiscernible_Object_Counting_in_Underwater_Scenes_CVPR_2023_paper.pdf | cvpr-2023-1 | ['object-counting'] | ['computer-vision'] | [-6.37184829e-02 -5.75259984e-01 1.99733824e-01 -1.54173076e-01
-5.21490932e-01 -6.57764852e-01 6.98242903e-01 2.87578255e-02
-1.12202334e+00 4.32455063e-01 2.48064891e-01 1.19286805e-01
9.33961272e-02 -7.83378005e-01 -8.01811874e-01 -6.01524174e-01
-2.07267538e-01 5.95723629e-01 5.25201976e-01 9.06522851e-03
2.65161544e-01 3.19650531e-01 -1.89619875e+00 1.08931065e-01
7.81038165e-01 9.54045773e-01 4.74420696e-01 7.95873463e-01
-1.82438400e-02 9.99554694e-01 -5.99965632e-01 -8.82514417e-01
8.70437101e-02 -4.71534528e-04 -8.21201503e-01 -2.64300406e-01
1.09657180e+00 -6.61948919e-01 -1.52961373e-01 1.37788713e+00
3.98040354e-01 -1.32546529e-01 7.75643229e-01 -1.02547598e+00
-6.05433643e-01 3.58245224e-01 -8.25960100e-01 5.05718589e-01
-3.68728973e-02 1.29834235e-01 9.16792154e-01 -1.02862585e+00
4.05412950e-02 1.47845638e+00 9.06364024e-01 6.07657313e-01
-8.73319328e-01 -1.02483356e+00 1.03151895e-01 -2.97547840e-02
-1.42704475e+00 -2.48135731e-01 8.31967741e-02 -4.19538528e-01
6.35891974e-01 1.85244456e-01 8.82501960e-01 7.54362524e-01
-4.36499923e-01 9.83187258e-01 1.11039329e+00 -2.10649177e-01
3.87024641e-01 -5.83714843e-01 2.89897233e-01 7.94927895e-01
8.67899001e-01 2.77406480e-02 -4.76560265e-01 -1.00141242e-01
8.22823703e-01 2.52497047e-01 -9.70443338e-02 2.90171187e-02
-1.33090675e+00 8.86605561e-01 5.10998487e-01 -1.93051204e-01
5.54631241e-02 4.69846308e-01 3.80442768e-01 -3.36473644e-01
4.31116372e-01 1.96683347e-01 -3.72467607e-01 -2.33973533e-01
-5.28184652e-01 4.43468362e-01 8.62843394e-01 1.17051840e+00
5.20874143e-01 -1.62840486e-01 -2.72121400e-01 9.05140221e-01
2.53745914e-01 1.23541641e+00 -2.40457319e-02 -1.26098204e+00
8.11794400e-01 7.12367117e-01 3.27666909e-01 -8.34974229e-01
-4.16134059e-01 -1.85612515e-01 -9.00274813e-01 2.30937913e-01
5.84860921e-01 8.58577155e-03 -8.52856636e-01 1.40735304e+00
2.19415933e-01 5.05645335e-01 9.31326151e-02 1.15435994e+00
1.29093325e+00 5.10006428e-01 2.17006877e-01 3.84905517e-01
1.74686134e+00 -1.03405285e+00 -3.91925335e-01 -3.82714808e-01
4.30212229e-01 -5.75425684e-01 9.71078694e-01 2.99080700e-01
-7.74983287e-01 -1.07350200e-01 -8.03828895e-01 -2.13022485e-01
-2.49192476e-01 4.06453669e-01 9.68592525e-01 4.97467160e-01
-6.77034080e-01 2.94560850e-01 -8.98437321e-01 -8.71099234e-02
9.58139479e-01 9.45431143e-02 -4.23606545e-01 -3.72208476e-01
-6.13007247e-01 7.33631968e-01 2.46170774e-01 4.68476526e-02
-9.61065948e-01 -6.49557054e-01 -1.04262328e+00 -2.47817077e-02
2.98703253e-01 -4.04396266e-01 1.30972672e+00 -7.71174654e-02
-8.36640120e-01 1.09211743e+00 1.77741721e-01 -4.82823789e-01
7.65232265e-01 -4.49137360e-01 2.54235268e-02 1.74990267e-01
5.40332854e-01 8.98282707e-01 4.40031797e-01 -1.20869589e+00
-1.20710397e+00 -5.41406333e-01 2.63779968e-01 -3.90362255e-02
-3.39702845e-01 1.11427560e-01 -9.37326014e-01 -6.08013690e-01
-1.26482189e-01 -8.66131604e-01 3.78714246e-03 3.86281699e-01
-1.89421028e-01 -4.44400311e-01 7.45775998e-01 -3.84786315e-02
9.15991247e-01 -1.95315659e+00 -1.24970123e-01 -6.94547534e-01
5.39367020e-01 4.88642573e-01 -2.71099627e-01 4.24084626e-02
4.44175541e-01 1.69458985e-02 -3.84300798e-01 -9.12967324e-01
-2.32820287e-01 5.90573490e-01 -1.64778858e-01 7.02545941e-01
2.82268316e-01 9.28053200e-01 -1.26238728e+00 -9.50288653e-01
4.11793381e-01 5.02414823e-01 -6.63307011e-01 1.31767437e-01
1.70862023e-02 3.63763005e-01 -4.29613948e-01 1.02225173e+00
1.11578262e+00 -2.52835780e-01 -4.98902202e-01 6.99920999e-03
-3.52442205e-01 -3.89353842e-01 -1.01804602e+00 1.51588404e+00
-4.27854270e-01 5.07442415e-01 1.94389392e-02 -5.42667270e-01
7.67608643e-01 -3.55076075e-01 2.65904665e-01 -5.70321739e-01
2.72702247e-01 2.49143615e-01 -3.18740219e-01 -4.60729808e-01
5.82976639e-01 -1.08812064e-01 -2.53731877e-01 5.99567108e-02
2.64290124e-01 -2.14379638e-01 5.91186285e-01 3.44134569e-01
1.02627778e+00 -2.60267347e-01 2.72964746e-01 -4.45301265e-01
2.81944543e-01 -7.55557371e-03 6.49576426e-01 1.08686745e+00
-4.89612877e-01 1.03894699e+00 2.86416888e-01 -6.98377550e-01
-7.32294023e-01 -9.44041252e-01 -5.97404659e-01 1.02682972e+00
5.67312479e-01 -3.03594470e-01 -7.94243872e-01 -5.74430823e-01
2.42604166e-01 1.00612320e-01 -1.00695670e+00 3.77299964e-01
-6.34079754e-01 -9.94018435e-01 8.86095405e-01 9.77491140e-01
1.07528615e+00 -1.05247974e+00 -8.19814086e-01 2.09338814e-02
-5.67035437e-01 -1.53344655e+00 -4.36735272e-01 1.50171831e-01
-8.06531191e-01 -1.46532762e+00 -9.86536622e-01 -5.57592869e-01
3.60316783e-01 8.94039750e-01 1.72010303e+00 4.14623618e-01
-4.72826779e-01 2.20624238e-01 -5.51289618e-01 -9.76115346e-01
3.59397717e-02 -1.84274629e-01 -1.64065082e-02 -2.62201250e-01
6.23067379e-01 -7.18825608e-02 -8.67162824e-01 5.10389268e-01
-8.57122481e-01 -2.20018372e-01 3.76325727e-01 7.26560414e-01
4.58088666e-01 -3.45044434e-01 3.22566509e-01 -7.26697326e-01
-6.20797202e-02 -3.78564984e-01 -1.23370647e+00 5.66069921e-03
1.00102194e-01 -7.03170374e-02 4.42955285e-01 -4.22242254e-01
-7.17111290e-01 -8.38805083e-03 -1.75791502e-01 -3.40336502e-01
1.04084894e-01 -3.62923071e-02 3.02205328e-02 -1.74699992e-01
5.05780160e-01 1.62875831e-01 -1.52632669e-01 -6.34086788e-01
5.07648960e-02 4.91873413e-01 1.04172254e+00 -6.50356412e-01
6.56818151e-01 9.35732424e-01 -1.18212968e-01 -8.50289106e-01
-1.47782433e+00 -1.00141609e+00 -5.27642787e-01 -3.22233945e-01
1.06752646e+00 -1.39275444e+00 -1.25520337e+00 9.82738435e-01
-1.30183470e+00 -1.97155192e-01 -1.21501140e-01 5.45428753e-01
-2.28863239e-01 3.75316054e-01 -5.75009108e-01 -1.03473473e+00
-4.77664262e-01 -1.03362966e+00 1.81557226e+00 3.01845133e-01
4.96325821e-01 -6.66778147e-01 5.29091880e-02 4.91920739e-01
2.13405848e-01 2.42549807e-01 2.82632202e-01 -2.02083945e-01
-7.69782424e-01 -3.56042147e-01 -8.08283746e-01 4.18747574e-01
-1.50317371e-01 -8.59234631e-02 -1.01542187e+00 -1.70567676e-01
-4.45825160e-01 -6.57139957e-01 1.55863619e+00 2.34223187e-01
1.34586263e+00 -3.27075832e-02 -1.35379687e-01 8.65355670e-01
1.48228729e+00 -4.08586890e-01 6.53041542e-01 4.20331657e-01
8.35683405e-01 3.06136161e-01 7.59031951e-01 7.02183604e-01
8.07822168e-01 6.60079420e-01 1.23668647e+00 -1.47200793e-01
-2.03799397e-01 -2.56927818e-01 -1.41910434e-01 7.08512068e-01
-7.54604340e-01 -2.36345395e-01 -9.13075507e-01 7.39927649e-01
-1.67486954e+00 -1.04621005e+00 -4.62656438e-01 1.94215333e+00
6.07527018e-01 -2.82211900e-01 1.88229814e-01 6.29814714e-02
7.04657078e-01 1.46285817e-01 -3.52973551e-01 5.20455897e-01
-2.98405677e-01 -5.30330651e-02 9.36084807e-01 6.20548166e-02
-1.62495160e+00 8.28025699e-01 5.87276793e+00 9.80879307e-01
-4.90964860e-01 1.93340108e-02 5.39832890e-01 1.36905923e-01
3.37620139e-01 -3.51420373e-01 -1.37983620e+00 6.83222294e-01
2.16653153e-01 4.46456790e-01 2.07982808e-01 1.11762500e+00
-2.28291422e-01 -3.50902915e-01 -7.27236092e-01 1.33518100e+00
2.20588714e-01 -1.38263166e+00 -1.22202307e-01 -1.18387572e-01
8.38422775e-01 5.60800254e-01 -4.76036631e-02 5.06240606e-01
7.11188316e-01 -1.06367338e+00 1.05624259e+00 2.49713480e-01
1.00404513e+00 -4.86121744e-01 1.16454518e+00 5.48115075e-01
-1.40168059e+00 -1.45167708e-01 -9.57706749e-01 -3.45645815e-01
3.85928061e-03 4.48576361e-01 -7.78302643e-03 4.25973348e-02
1.48116231e+00 8.38957191e-01 -7.72090852e-01 1.43193519e+00
-7.76880905e-02 6.15432262e-01 -6.16850793e-01 -2.45224610e-01
3.10675532e-01 -1.23551309e-01 5.63595966e-02 1.46891701e+00
2.47979105e-01 4.05456156e-01 1.83016747e-01 6.16480231e-01
-4.10850346e-01 -1.96756095e-01 -4.72795486e-01 4.53484416e-01
6.30843699e-01 1.36260521e+00 -9.51923668e-01 -5.25509417e-01
-6.08606040e-01 6.61679745e-01 4.30226237e-01 -1.34343714e-01
-1.08521807e+00 -5.70304953e-02 7.06494868e-01 -1.97507218e-01
3.60124528e-01 -1.44495919e-01 -9.02146697e-02 -1.49984372e+00
-5.66105470e-02 -4.91373926e-01 4.50596064e-01 -5.89132667e-01
-1.57699502e+00 4.96674985e-01 2.98507601e-01 -1.37568748e+00
4.92892623e-01 -8.98618698e-01 -4.49720502e-01 4.27138209e-01
-2.01350808e+00 -1.21401906e+00 -1.07957006e+00 4.75110620e-01
7.59082258e-01 2.11496763e-02 6.40271664e-01 4.45036769e-01
-3.53499085e-01 2.67262220e-01 1.24160498e-01 5.93233645e-01
3.74295950e-01 -1.37837029e+00 5.76973617e-01 7.58317828e-01
2.58193105e-01 3.79566163e-01 2.28495181e-01 -4.52514231e-01
-1.34919405e+00 -1.53346884e+00 3.90331298e-01 -1.12056005e+00
6.22740269e-01 -5.63885331e-01 -7.44434476e-01 2.95524210e-01
-4.71566319e-01 9.91126597e-01 2.25743890e-01 -1.77047953e-01
-5.42193413e-01 6.24969453e-02 -8.68099093e-01 1.98046744e-01
1.34278274e+00 -1.70401216e-01 -4.20527816e-01 3.11784685e-01
5.59228659e-01 -5.68148196e-01 -5.57587445e-01 6.20920777e-01
6.04524612e-01 -1.07564080e+00 1.31439304e+00 -2.40226798e-02
6.89306676e-01 -4.98592734e-01 -2.98037976e-01 -6.92462206e-01
-4.33545336e-02 6.21530153e-02 -9.78400111e-02 1.16375101e+00
-1.67648211e-01 -3.90279889e-01 8.82559896e-01 9.79644507e-02
-2.39957437e-01 -4.67041731e-01 -9.31983650e-01 -1.01584995e+00
3.32977295e-01 -5.12517750e-01 6.71696126e-01 5.35674036e-01
-6.72999680e-01 1.56407565e-01 -5.26530981e-01 3.30623865e-01
1.08927727e+00 2.08110839e-01 9.93828893e-01 -1.57261229e+00
8.37049484e-02 -2.45420143e-01 -7.10032761e-01 -1.46310270e+00
-1.06149562e-01 -6.85988963e-01 3.60911429e-01 -1.49780321e+00
7.31141448e-01 -6.17630899e-01 1.69055551e-01 5.75761259e-01
-5.03873527e-01 1.01001441e+00 4.78131205e-01 5.51676273e-01
-1.18338990e+00 7.17171669e-01 1.28081214e+00 -3.53760302e-01
4.42100197e-01 3.38260829e-02 -4.50027794e-01 1.05840898e+00
4.07331884e-01 -6.13132775e-01 1.89729586e-01 -9.43155766e-01
-2.04269458e-02 -1.38395429e-01 8.07114065e-01 -1.33510077e+00
2.12612465e-01 -3.10585964e-02 3.41227502e-01 -9.17855799e-01
3.41198474e-01 -7.47470021e-01 -3.95954490e-01 4.82877582e-01
9.85932574e-02 -1.03423640e-01 7.09291771e-02 9.55132902e-01
-3.07515830e-01 -3.66206884e-01 1.07808602e+00 -3.21977764e-01
-9.63835418e-01 5.10694325e-01 1.74629595e-02 5.37465572e-01
7.26856887e-01 -2.70096451e-01 -6.99420929e-01 1.93766117e-01
-1.43634491e-02 3.75899673e-01 3.90375733e-01 3.22792798e-01
5.30395150e-01 -1.02995539e+00 -1.00862956e+00 -8.00669789e-02
7.05546021e-01 6.73546195e-01 1.64848715e-01 6.94542229e-01
-9.26988006e-01 2.45598510e-01 8.46855938e-02 -1.02486849e+00
-1.25523984e+00 3.20686221e-01 7.53469691e-02 1.34017281e-02
-9.99181807e-01 1.17584193e+00 5.53952634e-01 -5.96483529e-01
2.87176400e-01 -6.08915806e-01 -4.96979982e-01 -8.51783529e-02
9.36539471e-01 4.92288172e-01 -1.20150046e-02 -7.26312876e-01
-5.34508944e-01 9.26766038e-01 1.90175131e-01 5.38176239e-01
1.49132168e+00 8.19363594e-02 3.23347412e-02 1.03630880e-02
9.99135911e-01 -1.01892471e-01 -1.79261410e+00 -2.72412777e-01
-1.10847011e-01 -1.00907516e+00 -6.60632700e-02 -6.03730202e-01
-1.31541049e+00 9.51061070e-01 5.76630771e-01 -1.99750140e-01
8.32283258e-01 4.01996881e-01 4.91171390e-01 5.33591270e-01
9.18210864e-01 -8.38421881e-01 4.23712403e-01 9.68806148e-01
7.33996511e-01 -1.79797184e+00 1.38301894e-01 -4.90834206e-01
-5.26973426e-01 9.25300181e-01 6.92926168e-01 -2.05537200e-01
3.09467465e-01 7.21835554e-01 4.20136191e-02 -4.36259896e-01
-3.13319504e-01 -6.64843261e-01 -7.15396693e-03 7.50992179e-01
1.31253660e-01 6.70790970e-02 3.73975337e-02 5.92087924e-01
-1.31528854e-01 -7.40629211e-02 4.87496048e-01 6.33166909e-01
-6.22011483e-01 -3.30225080e-01 -6.25064552e-01 5.27604938e-01
-4.45766598e-01 -1.94727987e-01 1.11092366e-01 9.57680941e-01
3.50682914e-01 9.70601201e-01 1.39692768e-01 1.09603576e-01
4.52642828e-01 -9.77136970e-01 3.46473336e-01 -4.61363554e-01
-1.77148610e-01 -2.44072467e-01 -1.45458996e-01 -4.48097289e-01
-8.16728830e-01 -6.22196972e-01 -1.11936009e+00 -5.49719930e-01
-7.41823435e-01 -8.97850394e-02 6.04668677e-01 7.49843419e-01
-1.87912464e-01 2.63489425e-01 1.79197952e-01 -1.22502661e+00
-5.01402915e-01 -1.00088274e+00 -6.46706522e-01 4.88563031e-01
2.80681252e-01 -1.14089632e+00 -4.21314865e-01 1.82938874e-02] | [8.779523849487305, 0.015572444535791874] |
8559bc13-1bb5-45af-a0b4-73a281d72cb2 | a-1d-cnn-based-deep-learning-technique-for | 2105.00528 | null | https://arxiv.org/abs/2105.00528v1 | https://arxiv.org/pdf/2105.00528v1.pdf | A 1D-CNN Based Deep Learning Technique for Sleep Apnea Detection in IoT Sensors | Internet of Things (IoT) enabled wearable sensors for health monitoring are widely used to reduce the cost of personal healthcare and improve quality of life. The sleep apnea-hypopnea syndrome, characterized by the abnormal reduction or pause in breathing, greatly affects the quality of sleep of an individual. This paper introduces a novel method for apnea detection (pause in breathing) from electrocardiogram (ECG) signals obtained from wearable devices. The novelty stems from the high resolution of apnea detection on a second-by-second basis, and this is achieved using a 1-dimensional convolutional neural network for feature extraction and detection of sleep apnea events. The proposed method exhibits an accuracy of 99.56% and a sensitivity of 96.05%. This model outperforms several lower resolution state-of-the-art apnea detection methods. The complexity of the proposed model is analyzed. We also analyze the feasibility of model pruning and binarization to reduce the resource requirements on a wearable IoT device. The pruned model with 80\% sparsity exhibited an accuracy of 97.34% and a sensitivity of 86.48%. The binarized model exhibited an accuracy of 75.59% and sensitivity of 63.23%. The performance of low complexity patient-specific models derived from the generic model is also studied to analyze the feasibility of retraining existing models to fit patient-specific requirements. The patient-specific models on average exhibited an accuracy of 97.79% and sensitivity of 92.23%. The source code for this work is made publicly available. | ['Deepu John', 'Barry Cardiff', 'Arlene John'] | 2021-05-02 | null | null | null | null | ['sleep-apnea-detection'] | ['medical'] | [ 2.93384522e-01 -8.51347074e-02 -6.70749918e-02 -3.50638419e-01
-2.97705799e-01 -5.29478453e-02 -6.01589680e-01 4.60867375e-01
-5.46459138e-01 7.15724647e-01 3.42761613e-02 -1.57727778e-01
-3.45100433e-01 -7.81514764e-01 -1.01792552e-01 -6.39660120e-01
-3.91799271e-01 -4.22904640e-02 -1.73655733e-01 2.01603904e-01
-1.94031596e-01 3.41004640e-01 -1.53339672e+00 2.25727618e-01
8.21280003e-01 1.40772319e+00 -1.18066475e-01 6.49783850e-01
3.92266244e-01 2.66590547e-02 -9.69537318e-01 8.59243274e-02
2.69430816e-01 -5.21878362e-01 -1.99096441e-01 -3.52670729e-01
7.49772713e-02 -2.95214504e-01 -5.94972037e-02 6.58984661e-01
9.27680850e-01 -1.01425320e-01 2.25511752e-02 -9.39741492e-01
-2.48948619e-01 4.47585404e-01 -9.51929539e-02 8.44715893e-01
3.02684218e-01 -9.86690819e-02 5.11184394e-01 -4.76607233e-01
-2.77802497e-01 3.28742117e-01 1.04740465e+00 6.64412379e-01
-9.04713631e-01 -8.05031955e-01 -4.79493707e-01 1.67384818e-01
-1.66766620e+00 -3.92781973e-01 6.92732155e-01 -4.75464351e-02
1.23255527e+00 5.59571743e-01 1.26616263e+00 6.80725813e-01
4.59451705e-01 -8.06294903e-02 7.10996270e-01 -2.09699780e-01
3.37197989e-01 -1.84385866e-01 2.22465634e-01 8.89933169e-01
1.01225400e+00 -3.02195288e-02 -4.93698359e-01 -3.10445845e-01
5.09433806e-01 7.04061806e-01 -1.38450444e-01 3.22189957e-01
-9.38508391e-01 6.19644105e-01 4.71335649e-01 6.58907592e-01
-5.28307736e-01 -7.07859769e-02 3.57156456e-01 -1.57237664e-01
3.13010514e-02 6.50518239e-01 -5.52660823e-01 -3.75389159e-01
-1.21761131e+00 -2.12949634e-01 8.97504985e-01 5.92663467e-01
3.05407673e-01 3.99706751e-01 -2.36782476e-01 7.54530251e-01
2.60532469e-01 5.92672110e-01 8.77257586e-01 -7.80803561e-01
1.84517741e-01 1.02185833e+00 -7.38409981e-02 -7.02652872e-01
-1.06669354e+00 -6.96490467e-01 -1.31157207e+00 -4.50399250e-01
-4.17466238e-02 -4.15445089e-01 -8.63405108e-01 1.36916089e+00
1.50919110e-01 6.94087520e-02 -5.53746335e-02 4.69984442e-01
9.28772688e-01 4.47582960e-01 -3.85690443e-02 -6.17679298e-01
1.51057744e+00 -4.68214512e-01 -8.54160547e-01 -1.72170684e-01
1.82563826e-01 -3.73219013e-01 8.31489563e-01 4.00455892e-01
-1.17788541e+00 -6.19166315e-01 -1.46602619e+00 2.77475357e-01
-2.88977325e-01 2.13845685e-01 2.00132951e-01 1.19706452e+00
-8.81844401e-01 6.74223185e-01 -1.29823935e+00 -5.67001820e-01
6.97108209e-01 8.21062148e-01 2.50013739e-01 2.48504683e-01
-9.53548014e-01 5.81478655e-01 2.16910899e-01 7.20640123e-02
-4.78227437e-01 -8.08886290e-01 -7.08215833e-01 2.81151175e-01
-1.03487007e-01 -9.47117984e-01 8.80160332e-01 -3.58813524e-01
-1.23899984e+00 4.86605942e-01 -3.29499871e-01 -8.27122986e-01
5.60845882e-02 -3.69557410e-01 -9.37114656e-01 3.96842152e-01
-1.50601402e-01 5.36736622e-02 6.59016550e-01 -3.92432272e-01
-5.25106907e-01 -6.03820622e-01 -3.10771495e-01 -1.21447124e-01
-6.55713797e-01 -1.26414225e-01 -6.53065145e-02 -4.65160728e-01
4.41997573e-02 -8.88123572e-01 -2.39072829e-01 -1.28190935e-01
-3.32725406e-01 1.49008021e-01 6.27233446e-01 -8.38859856e-01
1.87305951e+00 -2.03825808e+00 -5.48279881e-01 2.14119598e-01
5.31700313e-01 5.96637309e-01 4.96200025e-01 2.82617182e-01
-2.11473610e-02 2.17565000e-01 -4.77186888e-01 -3.02359968e-01
-5.41606724e-01 4.21503603e-01 3.99481863e-01 4.79374945e-01
-1.78747594e-01 6.88746750e-01 -5.42839885e-01 -4.16771650e-01
5.07716954e-01 8.52827072e-01 -5.16722023e-01 3.27149302e-01
5.20800531e-01 4.39061970e-01 -2.77692497e-01 9.72806692e-01
1.65420443e-01 -3.90640020e-01 3.61800194e-02 -4.89365906e-01
-1.52910963e-01 3.11792254e-01 -9.94144499e-01 1.60410523e+00
-3.30590785e-01 2.54422396e-01 -3.10917318e-01 -8.04007530e-01
1.05177879e+00 5.33835113e-01 1.04633653e+00 -7.03920901e-01
3.34728450e-01 7.39759207e-02 3.31951559e-01 -9.10188437e-01
8.73986445e-03 -1.47370353e-01 1.33227301e-03 1.26574889e-01
-2.93633521e-01 4.53830957e-01 1.99866936e-01 -2.99464524e-01
1.39928377e+00 -4.97157723e-01 8.97922277e-01 -3.57991487e-01
2.43939519e-01 -4.09803540e-01 7.27261424e-01 7.69533575e-01
-3.93714130e-01 4.66068894e-01 -3.40221703e-01 -8.50374401e-01
-7.76989579e-01 -1.07582784e+00 -3.26070309e-01 2.80835927e-01
-1.16360702e-01 -6.04898453e-01 -6.28753841e-01 -2.49587253e-01
-9.78988335e-02 5.15987754e-01 -5.33408523e-01 -5.00781894e-01
-7.46815860e-01 -1.19586086e+00 8.42930079e-01 7.83086598e-01
8.62678230e-01 -1.04011106e+00 -1.61664367e+00 2.73851275e-01
-3.50121260e-01 -1.00404751e+00 -3.36939692e-02 2.70882219e-01
-1.36333776e+00 -1.16917872e+00 -3.08832288e-01 -4.54314500e-01
4.08195615e-01 -1.49528340e-01 9.87029731e-01 3.03243309e-01
-8.14419091e-01 2.56159425e-01 -2.43087918e-01 -7.35259473e-01
-3.53873968e-02 8.80486444e-02 4.90246624e-01 -6.17004521e-02
5.99321008e-01 -9.35524285e-01 -1.32430792e+00 1.28889978e-01
-6.94429040e-01 -4.70951557e-01 5.28020442e-01 4.54037815e-01
8.44315112e-01 9.19927582e-02 7.08245754e-01 -3.36149037e-01
6.12783194e-01 -4.15890306e-01 -2.05525830e-01 -1.04300626e-01
-1.17967546e+00 -2.67831504e-01 7.65170515e-01 -4.00907367e-01
-3.42525482e-01 2.00780213e-01 -4.04989481e-01 -3.96855772e-01
-3.47292334e-01 1.97566852e-01 1.07032470e-01 1.15932234e-01
6.66046739e-01 1.17890716e-01 -7.04967603e-02 -6.03261888e-01
-4.36657101e-01 7.73835838e-01 4.38255727e-01 7.49060959e-02
3.33713800e-01 4.66974139e-01 9.76803377e-02 -1.16041219e+00
-7.78824866e-01 -6.89230263e-01 -4.49534744e-01 -9.76135023e-04
1.05017865e+00 -8.98719668e-01 -8.29376042e-01 7.54703879e-02
-5.61421096e-01 -5.97262830e-02 -6.90420926e-01 5.64538181e-01
-6.76080063e-02 2.17309654e-01 -2.12943465e-01 -9.34765637e-01
-1.09446502e+00 -6.92782283e-01 7.00562239e-01 5.18089056e-01
-6.50841713e-01 -6.19119644e-01 5.25082722e-02 2.82393515e-01
7.35106826e-01 6.83703184e-01 6.49846494e-01 -9.57637727e-01
2.92007141e-02 -4.12628770e-01 2.02688470e-01 4.59903866e-01
6.74010217e-01 -3.40507269e-01 -8.98247898e-01 -4.47461516e-01
4.17847663e-01 3.51544321e-01 5.77960670e-01 6.35851681e-01
1.39084816e+00 -5.23480415e-01 -3.90031934e-01 7.33161032e-01
1.53697634e+00 7.00671494e-01 7.12744534e-01 1.74579203e-01
5.16434550e-01 -1.12557791e-01 1.63357139e-01 7.35434055e-01
1.62643105e-01 3.65573436e-01 6.02476358e-01 -1.43343240e-01
-1.59728602e-01 2.95530200e-01 6.57849088e-02 9.67933536e-01
-4.27717149e-01 -3.42591926e-02 -1.00862062e+00 6.35619223e-01
-1.43944311e+00 -8.72119486e-01 -5.83389737e-02 2.35550356e+00
5.79697847e-01 -2.82903071e-02 5.35263479e-01 6.11645639e-01
6.29453003e-01 -9.76704508e-02 -7.72328675e-01 -5.55517375e-01
2.84804821e-01 7.25035608e-01 4.68476415e-01 -1.20523479e-02
-9.78096366e-01 -6.39647320e-02 6.15563583e+00 -3.24593391e-03
-9.73664522e-01 1.85503662e-01 3.59950066e-01 -5.18911839e-01
6.13011301e-01 -5.39641380e-01 -7.81612456e-01 8.16339910e-01
1.51834190e+00 4.97428998e-02 4.70576018e-01 7.93809474e-01
4.43034589e-01 5.18628582e-02 -9.74957526e-01 1.45320189e+00
2.30495274e-01 -1.14344645e+00 -3.42725843e-01 8.62608999e-02
3.83254319e-01 1.69181064e-01 -1.91079617e-01 -8.91191736e-02
-5.57241321e-01 -9.67793405e-01 2.37829909e-02 5.62223196e-01
1.09136534e+00 -7.08190918e-01 8.77357364e-01 4.35518436e-02
-1.45838118e+00 -7.36001253e-01 -1.49111658e-01 -3.51879686e-01
1.80792868e-01 8.74140978e-01 -9.50807273e-01 3.39584261e-01
1.35515738e+00 6.03980124e-01 -6.26071990e-01 1.43155277e+00
2.20027223e-01 9.92976189e-01 -7.60329247e-01 -1.61378220e-01
-4.36805367e-01 8.98132846e-02 5.61145306e-01 1.22168362e+00
6.15805805e-01 2.53365755e-01 5.10774069e-02 6.10828757e-01
3.18121612e-02 -1.80652943e-02 -4.79742408e-01 2.35782638e-01
5.33650398e-01 1.04648113e+00 -5.29620528e-01 -3.46862555e-01
-3.07137936e-01 5.60815811e-01 -3.28669518e-01 -5.01270592e-03
-1.07815182e+00 -4.97037172e-01 4.47342455e-01 2.61938661e-01
3.86600673e-01 9.60391983e-02 -6.58589780e-01 -8.69841337e-01
8.79366621e-02 -7.05389500e-01 6.61979318e-01 -3.19211185e-01
-9.90820825e-01 7.02468336e-01 -2.61360891e-02 -1.29590809e+00
-1.88790762e-03 -1.27911657e-01 -6.63995028e-01 4.96697485e-01
-1.12442446e+00 -6.43556178e-01 -8.22981536e-01 7.94813573e-01
4.75146741e-01 1.62648000e-02 1.27772295e+00 6.29453897e-01
-1.00058270e+00 7.52985179e-01 -1.37127131e-01 4.48008031e-02
2.52147496e-01 -1.02431381e+00 -7.26539735e-03 8.72031987e-01
-1.87160745e-01 8.70895624e-01 4.77567196e-01 -6.69194221e-01
-1.08522999e+00 -1.42555082e+00 9.47101474e-01 -1.68897197e-01
4.13950346e-02 1.09724710e-02 -7.02135026e-01 3.42390180e-01
-6.04454465e-02 1.46634206e-01 1.26617730e+00 -1.58647537e-01
2.21678674e-01 -7.09636033e-01 -1.62916362e+00 2.65768647e-01
9.08799231e-01 -1.39237314e-01 -6.82822526e-01 6.51803762e-02
4.84318018e-01 -2.43589163e-01 -1.33060813e+00 7.10036576e-01
8.82482827e-01 -1.02451801e+00 1.16511917e+00 -1.61706418e-01
-1.04566790e-01 -2.36337855e-01 -1.70628220e-01 -6.15268767e-01
-3.99718493e-01 -6.32764280e-01 -7.95510173e-01 8.52632701e-01
2.52665043e-01 -7.49437034e-01 6.83669984e-01 5.61230600e-01
-3.24199736e-01 -1.11111391e+00 -1.11913669e+00 -6.48811400e-01
-6.07937634e-01 -1.83510765e-01 6.56807780e-01 4.72667277e-01
-7.49170482e-02 2.87219286e-01 -2.44805098e-01 3.34294051e-01
4.62207198e-01 1.31589232e-03 1.43696502e-01 -1.49983490e+00
-9.94948000e-02 5.30148484e-02 -3.66845697e-01 -2.82324940e-01
-9.10407484e-01 -4.60289925e-01 -2.19843298e-01 -1.72164083e+00
1.57269333e-02 -1.26005456e-01 -1.00978839e+00 7.24896014e-01
2.80487686e-02 6.24432385e-01 -1.40438050e-01 1.05081961e-01
-4.31426615e-01 1.22923322e-01 6.47009671e-01 1.00648962e-01
-8.26795340e-01 3.86901587e-01 -6.94496810e-01 8.60972881e-01
1.67251265e+00 -7.24992394e-01 -3.60707760e-01 3.42165865e-02
1.18227296e-01 -2.36314777e-02 3.05818886e-01 -1.62385952e+00
-1.91239745e-03 4.68133628e-01 8.02271068e-01 -5.21341324e-01
5.86858034e-01 -1.11442721e+00 3.16672504e-01 1.22378469e+00
3.08740027e-02 4.81495589e-01 3.32271844e-01 5.34333766e-01
2.24246174e-01 7.95442313e-02 8.28620613e-01 -1.33059025e-01
-1.34579286e-01 2.12386832e-01 -3.66651237e-01 -1.13152355e-01
9.60161805e-01 -6.27148688e-01 -1.14230730e-01 2.80733611e-02
-8.71240795e-01 -2.98951298e-01 1.06832609e-01 1.99176028e-01
9.37688649e-01 -9.70897496e-01 -2.09330618e-01 5.51070631e-01
7.99366459e-02 -1.78976640e-01 2.26608023e-01 1.04115367e+00
-4.58299458e-01 4.43073303e-01 -3.81198764e-01 -5.58078885e-01
-1.56652403e+00 5.41142523e-01 5.43311000e-01 -2.41807312e-01
-7.95542717e-01 4.82000560e-01 -6.15180969e-01 3.64954472e-01
3.84279460e-01 -8.32617223e-01 -3.70614499e-01 -2.72553712e-01
7.17714667e-01 9.33577776e-01 4.04341549e-01 -2.49630675e-01
-8.73179317e-01 5.79885364e-01 4.40965235e-01 5.68244636e-01
1.37643349e+00 -1.08921669e-01 -6.36405423e-02 4.96759117e-01
1.01384711e+00 4.91228923e-02 -6.24783099e-01 4.21576440e-01
-3.78798455e-01 5.15563088e-03 -1.25140145e-01 -1.11612308e+00
-1.10613346e+00 6.65734947e-01 1.46193147e+00 4.39395040e-01
1.67556500e+00 -2.83153087e-01 1.17621183e+00 4.73984152e-01
1.21329889e-01 -8.43727112e-01 -9.83753875e-02 -1.48029074e-01
4.62673843e-01 -9.41424429e-01 2.99944520e-01 -6.10702969e-02
-2.09102452e-01 1.01785743e+00 3.88524592e-01 -2.55420536e-01
9.17436600e-01 2.20896795e-01 -1.27644196e-01 -3.64186704e-01
-9.80325192e-02 -7.47606205e-03 2.62094826e-01 7.50268698e-01
2.24179715e-01 1.17666349e-01 -4.36562538e-01 9.21374559e-01
-3.77517104e-01 2.63784021e-01 2.34025672e-01 1.08558750e+00
-4.93142933e-01 -6.94567919e-01 -1.74738944e-01 9.75731730e-01
-1.02916074e+00 -3.22438419e-01 -2.81036831e-02 4.86452729e-01
6.52151644e-01 1.43869448e+00 7.74913430e-02 -5.36252201e-01
3.96059036e-01 2.84374058e-01 2.03802183e-01 -6.08180285e-01
-8.99114728e-01 4.49045002e-02 1.08612604e-01 -7.34112561e-01
-6.58245921e-01 -4.01906163e-01 -1.27489984e+00 1.85292408e-01
2.15108646e-03 1.28133535e-01 7.55972564e-01 7.30218828e-01
8.75176787e-01 1.08959985e+00 2.47534990e-01 -3.67748886e-01
-1.78525060e-01 -8.91327798e-01 -4.36291754e-01 1.19454637e-01
7.63148069e-01 -3.21255505e-01 -2.77696818e-01 1.94606349e-01] | [13.77737045288086, 3.3968377113342285] |
165125ae-994c-4ef9-b906-902279abe1ed | unsupervised-cross-lingual-information | 1805.00879 | null | http://arxiv.org/abs/1805.00879v1 | http://arxiv.org/pdf/1805.00879v1.pdf | Unsupervised Cross-Lingual Information Retrieval using Monolingual Data Only | We propose a fully unsupervised framework for ad-hoc cross-lingual
information retrieval (CLIR) which requires no bilingual data at all. The
framework leverages shared cross-lingual word embedding spaces in which terms,
queries, and documents can be represented, irrespective of their actual
language. The shared embedding spaces are induced solely on the basis of
monolingual corpora in two languages through an iterative process based on
adversarial neural networks. Our experiments on the standard CLEF CLIR
collections for three language pairs of varying degrees of language similarity
(English-Dutch/Italian/Finnish) demonstrate the usefulness of the proposed
fully unsupervised approach. Our CLIR models with unsupervised cross-lingual
embeddings outperform baselines that utilize cross-lingual embeddings induced
relying on word-level and document-level alignments. We then demonstrate that
further improvements can be achieved by unsupervised ensemble CLIR models. We
believe that the proposed framework is the first step towards development of
effective CLIR models for language pairs and domains where parallel data are
scarce or non-existent. | ['Ivan Vulić', 'Goran Glavaš', 'Simone Paolo Ponzetto', 'Robert Litschko'] | 2018-05-02 | null | null | null | null | ['cross-lingual-information-retrieval'] | ['natural-language-processing'] | [-1.69056103e-01 -2.66764313e-01 -4.72408712e-01 -2.40016907e-01
-1.45125747e+00 -1.02717412e+00 1.09658837e+00 2.10130155e-01
-8.49799871e-01 4.13918644e-01 4.82019246e-01 -5.41513324e-01
-1.38564005e-01 -4.46636230e-01 -4.77843165e-01 -3.03466529e-01
1.01325303e-01 7.75369406e-01 -2.84954697e-01 -6.25163496e-01
-2.40314398e-02 2.83006281e-01 -1.12925744e+00 3.02138627e-01
9.51110125e-01 4.14544582e-01 2.51161680e-02 4.67464685e-01
-3.06595206e-01 3.86493087e-01 -2.37853378e-01 -5.84980667e-01
4.50312674e-01 -2.45377254e-02 -7.82147586e-01 -4.51366603e-01
4.87507641e-01 -8.89901593e-02 -4.54990029e-01 8.83306623e-01
6.53843641e-01 7.87861571e-02 1.03180945e+00 -1.01735580e+00
-1.45369530e+00 7.59619296e-01 -2.87690282e-01 -2.74417065e-02
3.23707759e-01 -2.49635965e-01 1.65149212e+00 -1.15056288e+00
7.97506690e-01 1.20512688e+00 4.96401906e-01 4.16359484e-01
-1.28424180e+00 -6.60410345e-01 -1.31840063e-02 1.95371121e-01
-1.68450260e+00 -2.78061599e-01 7.60376871e-01 -4.62667763e-01
9.98870075e-01 1.01974107e-01 -1.12316407e-01 1.27042150e+00
5.56159876e-02 7.99834967e-01 1.01934314e+00 -9.78524506e-01
-2.20376790e-01 5.58687031e-01 3.99067312e-01 2.67000228e-01
1.80742905e-01 9.81632769e-02 -3.87975454e-01 -3.02673161e-01
2.23180413e-01 -8.20518583e-02 -3.69866528e-02 -4.73091990e-01
-1.30922759e+00 1.08673275e+00 1.97169587e-01 7.61969745e-01
-6.81050122e-02 -2.32797354e-01 6.48951650e-01 6.25388980e-01
6.08322680e-01 7.06443608e-01 -6.05094790e-01 3.90142918e-01
-8.48069489e-01 2.81740073e-02 6.24492764e-01 1.08090854e+00
7.40426540e-01 -2.71962851e-01 -1.24059871e-01 1.22442722e+00
3.38467807e-01 7.32542336e-01 7.31462717e-01 -5.57197928e-01
3.88848484e-01 4.07015592e-01 1.49539724e-01 -8.31290960e-01
2.04241484e-01 -2.22916842e-01 -3.58674198e-01 -2.88129300e-01
2.75915042e-02 -4.54067811e-02 -5.01247108e-01 1.74505293e+00
-6.02375250e-03 -2.32532337e-01 6.21701598e-01 5.78433454e-01
7.88768411e-01 4.68809456e-01 -2.50730645e-02 1.25188380e-01
1.53183258e+00 -1.11002207e+00 -7.57820368e-01 -2.31473818e-01
8.48617315e-01 -1.08150411e+00 1.22703373e+00 -9.91081595e-02
-7.15605021e-01 -5.43376625e-01 -9.73731279e-01 -4.88230944e-01
-9.58318710e-01 2.51016706e-01 4.19009984e-01 5.70510328e-01
-1.09544945e+00 -1.01258405e-01 -4.53869253e-01 -3.75826955e-01
-2.99553305e-01 7.99536631e-02 -5.70403337e-01 -3.78963053e-01
-1.65896893e+00 9.38307941e-01 4.07430112e-01 -3.08156759e-02
-6.34847343e-01 -6.33996248e-01 -1.03501689e+00 -2.71349013e-01
-2.42499765e-02 -2.98864216e-01 7.76263714e-01 -9.55843687e-01
-1.00184345e+00 1.23255181e+00 -1.14516445e-01 -2.24503189e-01
2.67361552e-01 -4.18280721e-01 -6.99276209e-01 -1.92823201e-01
3.75861466e-01 5.79812706e-01 4.06989604e-01 -1.28760004e+00
-2.26381600e-01 -3.58350039e-01 1.81021297e-03 3.43961477e-01
-8.22939157e-01 4.40116048e-01 -6.69004738e-01 -7.51688421e-01
-3.28654826e-01 -1.06630862e+00 -1.32968919e-02 -3.86615336e-01
-2.16574460e-01 -5.45692265e-01 5.06045043e-01 -8.71359050e-01
1.19672072e+00 -1.95906746e+00 1.18382148e-01 2.21128799e-02
-3.07255685e-01 1.52713403e-01 -6.24878347e-01 1.02824378e+00
7.07364082e-02 3.53523880e-01 1.51416391e-01 -2.81906962e-01
3.20740908e-01 1.82380378e-01 -5.80956936e-01 1.71541825e-01
2.23361135e-01 1.03370619e+00 -1.07485843e+00 -4.89410579e-01
-4.81104068e-02 6.49848819e-01 -4.82482255e-01 2.94227660e-01
-1.86600145e-02 2.73914039e-01 -4.76291597e-01 5.15434384e-01
2.74564922e-01 1.45725816e-01 5.03029406e-01 -2.61506617e-01
-4.18284126e-02 6.19908810e-01 -5.08075833e-01 2.00350523e+00
-8.67392719e-01 5.90598464e-01 -2.73923516e-01 -9.18012321e-01
9.83574986e-01 4.68123078e-01 4.44729686e-01 -7.10985124e-01
-5.55886850e-02 5.07962286e-01 -7.32381642e-02 -2.39578649e-01
8.68734062e-01 9.40789431e-02 -6.01828277e-01 8.32594275e-01
2.46300176e-01 6.17715493e-02 2.57788956e-01 2.64346331e-01
9.63709354e-01 1.73516750e-01 3.30819637e-02 -5.63966572e-01
6.54974401e-01 1.01091564e-01 2.64131278e-01 7.20108926e-01
-1.86892241e-01 2.95907974e-01 -1.85254440e-01 -3.13124359e-01
-1.22273910e+00 -1.26821923e+00 -6.34330869e-01 1.52079952e+00
-6.05285875e-02 -3.70219499e-01 -2.76970565e-01 -7.52980769e-01
-1.90660916e-02 8.94967675e-01 -5.79606950e-01 6.41688854e-02
-4.67939347e-01 -4.73750293e-01 8.35459173e-01 1.95509538e-01
-7.67486766e-02 -1.11529231e+00 3.74088377e-01 1.46381736e-01
-2.41004437e-01 -1.44407320e+00 -8.71473968e-01 2.04421692e-02
-4.48350608e-01 -7.86637604e-01 -7.64494538e-01 -1.20856869e+00
6.10378146e-01 2.78248996e-01 1.42644930e+00 -2.53768206e-01
-3.13495547e-01 7.79050410e-01 -5.34038126e-01 -1.65585488e-01
-4.84828889e-01 2.47973323e-01 3.40522647e-01 -1.77714661e-01
1.25113773e+00 -3.46625000e-01 -4.63145643e-01 2.06361547e-01
-1.04694211e+00 -4.58170593e-01 3.73747259e-01 9.64420736e-01
5.22731960e-01 -3.46718669e-01 6.24677241e-01 -1.03008199e+00
9.58867133e-01 -6.11419022e-01 -3.59661222e-01 8.19203198e-01
-5.48069417e-01 4.60799664e-01 4.91511583e-01 -4.64127541e-01
-8.91921282e-01 -4.69045907e-01 8.14240500e-02 -4.05607611e-01
-8.65285099e-02 5.56173861e-01 9.75396186e-02 2.48372078e-01
6.87574744e-01 2.62291461e-01 -4.28977132e-01 -5.33492804e-01
9.87466693e-01 1.15875554e+00 3.81580174e-01 -9.84098196e-01
7.99541950e-01 1.14858776e-01 -7.62605131e-01 -5.80927014e-01
-6.23885036e-01 -8.16852212e-01 -9.39335585e-01 2.98902929e-01
1.10439825e+00 -1.34400141e+00 -1.97999435e-03 -2.86461115e-02
-1.29750693e+00 8.55168402e-02 5.35968952e-02 5.47790647e-01
-1.02526836e-01 3.60863507e-01 -7.54708171e-01 -6.58501565e-01
-5.40128231e-01 -1.11008692e+00 1.17530215e+00 -3.76774609e-01
-3.14525306e-01 -1.59193456e+00 7.26419985e-01 5.91282666e-01
3.47581625e-01 -3.15473288e-01 1.11951971e+00 -1.20230198e+00
-2.27770284e-01 -2.90015221e-01 -3.73194143e-02 5.86400092e-01
5.03066003e-01 6.03689514e-05 -8.37061405e-01 -5.26206613e-01
-5.44698417e-01 -7.10628510e-01 5.85976541e-01 -2.46234298e-01
5.65212190e-01 -2.43615627e-01 -1.48493350e-01 2.76558310e-01
1.62400842e+00 -1.78605150e-02 2.59810925e-01 4.28330332e-01
7.14202464e-01 8.22686732e-01 3.17478538e-01 -1.27125561e-01
6.73802853e-01 7.62501121e-01 -2.87369668e-01 -6.94972053e-02
-3.80133390e-02 -4.65398908e-01 6.87382519e-01 1.77989793e+00
3.24329644e-01 -5.93023151e-02 -1.05553544e+00 1.20514238e+00
-1.42520201e+00 -7.77405202e-01 3.89479280e-01 2.23573184e+00
1.12429607e+00 -3.42537075e-01 -3.09921533e-01 -4.70110089e-01
6.12676144e-01 7.79774487e-02 -3.26766580e-01 -7.20560730e-01
-2.30696797e-01 4.74535078e-01 4.08963203e-01 7.57910967e-01
-1.09875298e+00 1.22656047e+00 6.34394169e+00 7.83623636e-01
-8.85605812e-01 3.82562488e-01 1.35602728e-01 2.22487420e-01
-8.37736845e-01 -8.67217034e-02 -9.82166290e-01 2.20404819e-01
1.01971853e+00 -3.82293403e-01 4.23156589e-01 6.52121067e-01
-2.41207987e-01 6.83009326e-01 -1.18592203e+00 6.65898383e-01
4.66823071e-01 -9.05876100e-01 2.22840965e-01 -7.01372996e-02
1.03085792e+00 5.36737025e-01 2.16264293e-01 5.43703496e-01
1.04648864e+00 -9.54151213e-01 9.44194719e-02 6.03244007e-02
1.07624519e+00 -6.42158926e-01 7.12221444e-01 1.72514051e-01
-1.02127993e+00 4.02132452e-01 -4.80554014e-01 5.42444229e-01
-5.96155003e-02 5.69424853e-02 -7.50302196e-01 8.28554273e-01
5.60575008e-01 7.46745408e-01 -5.14155865e-01 4.61894870e-02
-2.69645184e-01 3.31604332e-01 -1.38007686e-01 1.09949544e-01
2.95291603e-01 -3.32109511e-01 4.02070403e-01 1.54422474e+00
2.42375091e-01 -5.88572204e-01 3.02241236e-01 6.32619321e-01
-2.94140488e-01 7.83542156e-01 -1.14825976e+00 -4.31399763e-01
6.25243723e-01 1.18167067e+00 -4.61861603e-02 -1.14472993e-01
-8.78719568e-01 1.05905116e+00 6.35028303e-01 5.86416960e-01
-2.60605991e-01 -4.09998149e-01 8.02196741e-01 -2.27425531e-01
2.21900180e-01 -3.24433386e-01 6.04031160e-02 -1.60393679e+00
3.62931900e-02 -1.13836050e+00 5.82416177e-01 -5.27668893e-01
-2.02764130e+00 1.00054049e+00 -9.79948491e-02 -1.25500596e+00
-6.62531793e-01 -7.47726202e-01 -1.68791011e-01 1.13090134e+00
-1.89027500e+00 -1.51180696e+00 3.66057903e-01 7.24101961e-01
5.43640614e-01 -7.36424625e-01 1.39606953e+00 4.17008698e-01
-2.46443942e-01 1.08079946e+00 8.41959178e-01 5.61059058e-01
1.19896603e+00 -1.01429999e+00 1.92878142e-01 7.58886516e-01
8.75044107e-01 1.10738552e+00 3.62104654e-01 -3.21708798e-01
-1.42592347e+00 -1.09836447e+00 1.53980899e+00 -8.78337204e-01
1.22711575e+00 -6.72239900e-01 -8.18312824e-01 9.27750647e-01
7.14324832e-01 2.99065039e-02 1.30190575e+00 6.25176966e-01
-9.98817801e-01 -4.68928814e-02 -7.66510904e-01 7.30592728e-01
7.14843750e-01 -1.28615880e+00 -1.02795076e+00 6.36433661e-01
9.82852697e-01 2.35696986e-01 -1.23868704e+00 2.85772949e-01
5.88815808e-01 -3.75311881e-01 1.13652539e+00 -9.88842785e-01
4.74383593e-01 -9.05747861e-02 -5.76837897e-01 -1.26874304e+00
-1.50666267e-01 -2.64576465e-01 4.49466616e-01 1.29327512e+00
5.27221441e-01 -6.28002048e-01 5.62676974e-02 3.06381077e-01
1.30164847e-01 -2.75559098e-01 -8.80961537e-01 -9.27102804e-01
8.37975621e-01 -6.34250641e-01 3.55721831e-01 1.29835594e+00
1.12043455e-01 6.24297976e-01 -2.17229992e-01 1.69298351e-01
6.10523045e-01 6.83936626e-02 6.71901286e-01 -8.91625226e-01
-4.30420280e-01 -2.43321821e-01 -3.20701480e-01 -9.12455797e-01
9.85552311e-01 -1.46076345e+00 -8.95467773e-03 -1.11939967e+00
3.12894255e-01 -6.64276481e-01 -9.86988544e-01 4.40939277e-01
-3.46457094e-01 3.80177617e-01 -6.76578507e-02 4.58926052e-01
-5.73078811e-01 6.69483066e-01 7.15730906e-01 -3.79459888e-01
4.54055741e-02 -7.10732877e-01 -7.10601926e-01 3.42723131e-01
4.93209302e-01 -4.45713013e-01 -4.15801197e-01 -9.25901353e-01
-1.62616912e-02 -2.76493162e-01 -1.65607542e-01 -3.28315705e-01
1.86418563e-01 6.82810042e-03 -1.75097048e-01 -2.55637348e-01
1.48783028e-01 -8.59933794e-01 -2.10859567e-01 -4.66392748e-02
-8.03866386e-01 3.92787963e-01 2.86998212e-01 4.00044054e-01
-5.65417886e-01 -2.27925315e-01 2.38120615e-01 1.31962851e-01
-6.68410301e-01 2.37857118e-01 -1.47937819e-01 4.26236153e-01
7.02497721e-01 3.95739943e-01 -1.21780284e-01 -3.28648239e-01
-4.01653886e-01 2.06360117e-01 5.19421935e-01 9.85707104e-01
1.91997856e-01 -1.76903892e+00 -1.16079760e+00 3.47462714e-01
8.37184370e-01 -7.19466209e-01 -1.57784477e-01 2.76853234e-01
-1.77175537e-01 9.69466865e-01 4.45974991e-02 -3.34094614e-01
-1.07095981e+00 6.58841789e-01 -5.78208603e-02 -8.33867908e-01
-2.09348164e-02 5.34390390e-01 3.46218556e-01 -1.30478382e+00
-4.68391068e-02 4.33522344e-01 -1.19476102e-01 -8.37267488e-02
3.24060977e-01 -1.36188999e-01 1.52904928e-01 -1.04922521e+00
-3.65456283e-01 6.89918339e-01 -5.39797306e-01 -5.01688778e-01
1.19715798e+00 -2.28050098e-01 -2.12143049e-01 7.39975989e-01
1.48888397e+00 4.40544248e-01 -3.86236131e-01 -7.81298578e-01
2.29397461e-01 -2.05757156e-01 1.18906111e-01 -6.62285447e-01
-7.24397659e-01 9.82888520e-01 4.99412686e-01 -2.62352347e-01
8.26609075e-01 4.61803265e-02 7.73182571e-01 6.02094173e-01
4.40361857e-01 -1.03740084e+00 -1.11703590e-01 6.71417236e-01
8.31065357e-01 -1.40372181e+00 -2.68357128e-01 5.98094463e-02
-5.26609182e-01 8.65289569e-01 2.77130753e-01 -2.93224543e-01
7.03325391e-01 1.18952096e-01 5.83895683e-01 8.42925981e-02
-8.14983249e-01 -3.10205281e-01 8.26454401e-01 5.72657526e-01
1.06503642e+00 8.22450370e-02 -7.33502924e-01 5.04621923e-01
-1.26973480e-01 -4.31853801e-01 -1.51245937e-01 7.06952095e-01
1.50025740e-01 -1.91347551e+00 -1.04308993e-01 1.99811403e-02
-7.24811196e-01 -7.54940867e-01 -3.30549359e-01 7.98008084e-01
-1.90812960e-01 8.80858183e-01 2.30037883e-01 -2.54683495e-01
-3.48334201e-02 4.61616188e-01 3.65914732e-01 -6.88051283e-01
-4.92895246e-01 1.19275730e-02 3.30212116e-02 -1.31677225e-01
-3.51932645e-01 -5.10803938e-01 -5.83356738e-01 5.88289909e-02
-1.28024340e-01 3.67444426e-01 6.69920146e-01 8.38033736e-01
5.24494052e-01 3.39671522e-02 7.15151429e-01 -3.29136431e-01
-5.66750824e-01 -1.12791753e+00 -3.50448489e-01 8.02399218e-01
-5.83940595e-02 -3.59218329e-01 -4.91331518e-01 1.59576088e-01] | [11.242423057556152, 9.839028358459473] |
9dfc7324-89b3-4c2d-ae96-336c71f6239a | math-agents-computational-infrastructure | 2307.02502 | null | https://arxiv.org/abs/2307.02502v1 | https://arxiv.org/pdf/2307.02502v1.pdf | Math Agents: Computational Infrastructure, Mathematical Embedding, and Genomics | The advancement in generative AI could be boosted with more accessible mathematics. Beyond human-AI chat, large language models (LLMs) are emerging in programming, algorithm discovery, and theorem proving, yet their genomics application is limited. This project introduces Math Agents and mathematical embedding as fresh entries to the "Moore's Law of Mathematics", using a GPT-based workflow to convert equations from literature into LaTeX and Python formats. While many digital equation representations exist, there's a lack of automated large-scale evaluation tools. LLMs are pivotal as linguistic user interfaces, providing natural language access for human-AI chat and formal languages for large-scale AI-assisted computational infrastructure. Given the infinite formal possibility spaces, Math Agents, which interact with math, could potentially shift us from "big data" to "big math". Math, unlike the more flexible natural language, has properties subject to proof, enabling its use beyond traditional applications like high-validation math-certified icons for AI alignment aims. This project aims to use Math Agents and mathematical embeddings to address the ageing issue in information systems biology by applying multiscalar physics mathematics to disease models and genomic data. Generative AI with episodic memory could help analyse causal relations in longitudinal health records, using SIR Precision Health models. Genomic data is suggested for addressing the unsolved Alzheimer's disease problem. | ['Renato P. dos Santos', 'Eric Roland', 'Takashi Kido', 'Melanie Swan'] | 2023-07-04 | null | null | null | null | ['automated-theorem-proving', 'automated-theorem-proving'] | ['miscellaneous', 'reasoning'] | [-1.94522738e-01 5.10910094e-01 4.30166610e-02 -2.08520323e-01
-6.00812137e-01 -5.78023076e-01 5.39379060e-01 6.51816070e-01
-2.30738297e-01 6.09992981e-01 3.97596151e-01 -8.50302577e-01
-5.76850951e-01 -1.15566623e+00 -8.94183695e-01 -1.37638971e-01
-2.18195379e-01 1.17793489e+00 -2.20048398e-01 -4.78898704e-01
-9.59356651e-02 1.47061810e-01 -1.22561347e+00 2.42841035e-01
9.08907950e-01 4.68084574e-01 -3.87448748e-03 8.82421613e-01
-4.74482656e-01 9.69644070e-01 -3.29843640e-01 -9.27196443e-01
-3.92507128e-02 -3.62031609e-01 -7.41512835e-01 -1.19252503e+00
2.57113785e-01 -8.34847763e-02 -2.21501678e-01 9.51318860e-01
5.75228155e-01 -3.43699127e-01 3.60218614e-01 -1.66966605e+00
-1.01609266e+00 1.04958832e+00 2.24726513e-01 1.55255020e-01
6.97641850e-01 4.81606275e-01 1.04580998e+00 -2.77007371e-01
9.97992635e-01 1.59719527e+00 1.14784825e+00 4.88394320e-01
-1.54186583e+00 -5.55313170e-01 -5.44954300e-01 1.07240014e-01
-1.18342495e+00 7.74598718e-02 3.74611199e-01 -8.38018954e-01
1.35339117e+00 5.88218391e-01 1.26809239e+00 1.05897117e+00
6.60566390e-01 2.68818259e-01 9.13060308e-01 -2.07389355e-01
3.60021025e-01 2.66727835e-01 4.22088057e-01 1.15180302e+00
3.11785072e-01 2.00577140e-01 -4.89057720e-01 -6.09402359e-01
6.46145761e-01 -9.58773121e-02 1.88460842e-01 2.38577172e-01
-1.58155465e+00 8.49697709e-01 1.97941914e-01 2.20281973e-01
-4.27470773e-01 6.21255159e-01 4.25058663e-01 3.44076127e-01
3.96833688e-01 7.91216493e-01 -5.82206845e-01 -3.12328845e-01
-8.23289990e-01 8.68518651e-01 1.00446546e+00 8.35502744e-01
4.22047436e-01 -1.79159880e-01 7.67161623e-02 3.14612001e-01
6.39866054e-01 6.02146745e-01 2.32435733e-01 -1.03500938e+00
4.31318209e-02 1.16771817e+00 -5.05934298e-01 -1.16950786e+00
-7.82344818e-01 -2.80353636e-01 -8.76956284e-01 4.44209427e-02
4.91620630e-01 1.32201195e-01 -4.96048540e-01 1.68398058e+00
5.36520720e-01 5.61398789e-02 1.09710626e-01 7.18323767e-01
9.93438601e-01 6.54758990e-01 3.66506666e-01 3.17614555e-01
1.68157530e+00 8.64364300e-03 -5.09967923e-01 5.61088249e-02
1.00121939e+00 -2.67283827e-01 9.69530463e-01 4.46976691e-01
-1.33316517e+00 -8.53701010e-02 -7.00023651e-01 -5.57500064e-01
-8.54550064e-01 -4.22463030e-01 1.22884083e+00 6.19367123e-01
-1.11557448e+00 4.22663748e-01 -1.00984585e+00 -2.97023594e-01
6.38483465e-01 4.00759250e-01 -3.03017646e-01 4.49328274e-02
-1.65947390e+00 1.13493347e+00 2.50602037e-01 -7.24227950e-02
-4.81893361e-01 -1.79066527e+00 -9.46839511e-01 -6.84787482e-02
-7.81729072e-02 -1.21219659e+00 6.97468936e-01 -5.51976681e-01
-9.13191557e-01 7.36192524e-01 1.45769760e-01 -8.45364869e-01
3.41650724e-01 1.17076196e-01 -3.53491485e-01 -5.34109846e-02
4.45535854e-02 8.27497303e-01 8.56807306e-02 -4.57501024e-01
9.95903686e-02 -4.53332126e-01 2.07229748e-01 -4.17889565e-01
-5.19519299e-02 2.04722106e-01 2.12973967e-01 -5.61627626e-01
-1.90835327e-01 -7.24616349e-01 -3.15691233e-01 3.10522407e-01
-1.68910831e-01 -3.13541144e-01 2.17689037e-01 -1.03903091e+00
1.26107955e+00 -1.66275823e+00 4.04029757e-01 3.62124026e-01
8.52832496e-01 -1.35883167e-01 1.84585407e-01 4.27175224e-01
3.10258977e-02 5.79281092e-01 -1.55379042e-01 2.92832255e-01
5.52482903e-01 1.47832632e-01 -2.08763495e-01 2.10196823e-01
4.34180111e-01 1.54188240e+00 -1.01717722e+00 -6.36090755e-01
1.03051804e-01 7.09160686e-01 -1.04132688e+00 -2.19788015e-01
-8.64638448e-01 -4.10785638e-02 -2.61887938e-01 4.54157531e-01
4.77107972e-01 -3.43804926e-01 2.02823550e-01 -2.99916603e-02
-1.39668614e-01 3.72964531e-01 -1.11170638e+00 1.71916294e+00
-3.35801750e-01 6.58249557e-01 -1.30833194e-01 -7.39047945e-01
7.12516189e-01 6.72218129e-02 4.58657742e-01 -6.96754754e-01
3.53389718e-02 4.45487089e-02 4.87428427e-01 -5.56055605e-01
1.48684621e-01 -2.94873238e-01 -1.21235743e-01 5.53667367e-01
1.04865856e-01 -4.10756975e-01 2.20574886e-01 6.99630558e-01
1.46830571e+00 2.72396743e-01 -2.60823388e-02 -4.02356893e-01
1.15711458e-01 5.34316659e-01 3.67584735e-01 4.39697206e-01
1.45309046e-01 5.23187742e-02 7.50469506e-01 -7.17807174e-01
-1.20211470e+00 -1.25091672e+00 -3.76677692e-01 8.69104207e-01
-5.55870891e-01 -1.08020473e+00 -8.19212556e-01 1.43395932e-02
2.81388462e-01 8.60323310e-01 -4.78525192e-01 -2.28501558e-01
-4.17855948e-01 -9.49930847e-01 1.25173867e+00 2.89132029e-01
-4.32340391e-02 -9.33097780e-01 -7.10556090e-01 3.23512644e-01
-7.83073530e-02 -6.40535593e-01 5.63616976e-02 -8.61719996e-02
-6.10089183e-01 -1.05922735e+00 -3.81908447e-01 -2.52354264e-01
3.74925494e-01 -8.79052460e-01 1.24620199e+00 5.80732748e-02
-1.04916108e+00 5.50556123e-01 4.64384146e-02 -9.47262943e-01
-1.29567230e+00 -2.40357786e-01 3.56196880e-01 -8.57896149e-01
3.93001288e-01 -5.93029797e-01 -3.98037493e-01 -2.55275398e-01
-1.04717255e+00 4.81249124e-01 3.68524581e-01 7.59359837e-01
3.86079282e-01 -2.09612727e-01 5.21712303e-01 -6.64219916e-01
6.24030232e-01 -7.32415974e-01 -6.53703451e-01 2.87364990e-01
-7.64890671e-01 4.59956855e-01 6.71028137e-01 -3.71818453e-01
-5.44777215e-01 -3.99980038e-01 -1.39194682e-01 -6.36333376e-02
9.43004489e-02 9.20714021e-01 2.43186373e-02 4.26394671e-01
8.51999104e-01 -1.39413267e-01 4.64064032e-01 -1.40758783e-01
6.05614126e-01 4.15498853e-01 3.75830859e-01 -7.29469478e-01
5.60850263e-01 8.23960602e-02 1.57539725e-01 -8.64199877e-01
-1.01341560e-01 2.70749599e-01 -2.74653941e-01 5.97135089e-02
1.10559642e+00 -6.62794173e-01 -1.24606264e+00 -2.55522262e-02
-1.27823544e+00 -4.93269920e-01 -5.21427393e-01 4.95920211e-01
-4.80768949e-01 1.55321462e-02 -6.04703248e-01 -6.68544531e-01
-4.92200077e-01 -9.69973862e-01 9.80201364e-01 5.78982718e-02
-8.70737374e-01 -1.19202161e+00 5.39508283e-01 4.15251166e-01
4.72709239e-01 2.89525539e-01 1.52259672e+00 -7.11739480e-01
-7.09588647e-01 -1.22753799e-01 -1.01552218e-01 -3.22337039e-02
-4.84520763e-01 2.23025128e-01 -6.29553854e-01 5.29738009e-01
-4.14871126e-01 -2.38310203e-01 2.69639343e-01 2.32308611e-01
8.25556159e-01 -5.16318142e-01 -3.69711518e-01 5.50043344e-01
1.11163378e+00 -1.83774501e-01 8.60486269e-01 2.57714897e-01
5.48278749e-01 4.73165095e-01 -2.22833715e-02 1.92757189e-01
8.44251931e-01 2.66761214e-01 6.10146448e-02 1.11749150e-01
2.33892407e-02 -2.71977633e-01 5.81829488e-01 7.15678632e-01
3.44761908e-02 7.92986453e-02 -1.63870180e+00 1.31773606e-01
-1.68509066e+00 -1.23484182e+00 -5.46184063e-01 1.78759921e+00
1.27681434e+00 1.47314608e-01 5.21979332e-02 -1.42522529e-01
7.36386478e-02 -5.45255721e-01 -5.15914321e-01 -6.94685459e-01
-1.88502267e-01 6.96830213e-01 3.03120077e-01 4.70049560e-01
-4.85357046e-01 5.65339863e-01 5.89125919e+00 6.02312386e-01
-7.43985713e-01 2.62355357e-01 3.95534635e-01 -7.27429390e-02
-9.54913139e-01 1.04207709e-01 -5.11869490e-01 5.26903868e-01
1.77381730e+00 -3.76602799e-01 7.85515308e-01 4.65319782e-01
1.54991359e-01 -1.04630962e-02 -1.37985313e+00 8.64090145e-01
-2.19917119e-01 -2.09825516e+00 -5.44075966e-02 6.14952110e-02
9.98523384e-02 1.89886823e-01 -2.23979712e-01 3.26579541e-01
6.52998030e-01 -1.44132829e+00 7.15333045e-01 1.24238265e+00
8.37387562e-01 -5.21909595e-01 4.54030186e-01 2.54867315e-01
-8.29001248e-01 -1.25828348e-02 -2.07817495e-01 -8.42980072e-02
1.49024665e-01 7.01275826e-01 -9.98979390e-01 4.57181036e-01
5.83999574e-01 3.46226186e-01 -7.40002573e-01 6.05155408e-01
3.39222491e-01 6.50441051e-01 -6.15623057e-01 -3.58924180e-01
-3.48403871e-01 -2.18957111e-01 5.66983402e-01 1.27819228e+00
1.98463723e-01 1.19574048e-01 -8.50391611e-02 1.84336460e+00
2.51077741e-01 -1.02153458e-01 -7.17955053e-01 -8.87396395e-01
2.56519586e-01 1.10609615e+00 -5.80027997e-01 -2.53778785e-01
-1.63511559e-01 6.55864179e-01 -9.13431868e-02 -1.80302128e-01
-1.10937095e+00 -3.62396479e-01 7.28270769e-01 3.97453666e-01
-6.61479235e-01 -3.24387074e-01 -3.11216235e-01 -8.18158567e-01
-2.50576973e-01 -1.28845263e+00 2.95307159e-01 -1.00011909e+00
-1.27053213e+00 1.89503357e-02 1.82604700e-01 -5.13429046e-01
-4.06131685e-01 -8.31187725e-01 -3.86371821e-01 9.20208395e-01
-6.04133129e-01 -1.34924340e+00 1.08067475e-01 2.47363802e-02
-6.66674301e-02 -2.11584911e-01 1.25165427e+00 3.32819223e-01
-3.11960280e-01 5.75302541e-01 -2.12801620e-02 1.67824328e-01
3.89152557e-01 -1.37305939e+00 6.64730430e-01 1.63981840e-01
1.08666964e-01 1.15500689e+00 7.05949008e-01 -1.08578849e+00
-2.29215622e+00 -8.32313061e-01 8.80530775e-01 -1.08952844e+00
1.03074813e+00 -7.53447056e-01 -7.26015747e-01 6.54802859e-01
-1.49916843e-01 -3.36179614e-01 8.62904847e-01 9.54843462e-02
-4.33878213e-01 9.59023833e-02 -1.21558869e+00 7.22350717e-01
9.73897994e-01 -7.32088327e-01 -5.56332529e-01 6.49445593e-01
1.21608067e+00 -4.47591096e-01 -1.41283572e+00 -6.42432198e-02
7.55790830e-01 -3.41850817e-01 1.21803331e+00 -1.03097987e+00
7.14079559e-01 -1.87490180e-01 -6.49360269e-02 -9.84004080e-01
-1.46099970e-01 -8.67236793e-01 -1.92563906e-01 1.18281817e+00
5.46226501e-01 -7.45525002e-01 5.10613203e-01 1.08739972e+00
8.66838265e-03 -6.64442778e-01 -8.48970771e-01 -4.94058788e-01
4.84118611e-01 -1.03668523e+00 7.11059034e-01 1.04832351e+00
3.36067498e-01 2.64730781e-01 4.19962317e-01 1.14572234e-01
3.61618608e-01 -4.12296742e-01 6.89485669e-01 -1.63562036e+00
-5.29533505e-01 -5.18033445e-01 -9.34552848e-01 8.11340436e-02
2.19726115e-01 -1.91921985e+00 -4.83120650e-01 -1.70117342e+00
2.49362037e-01 -5.70703447e-01 2.79927611e-01 5.24139762e-01
2.95806855e-01 -1.23572886e-01 6.45071492e-02 -4.05312687e-01
-1.32874012e-01 1.37338489e-01 5.75953007e-01 -3.90584797e-01
-1.90445587e-01 -6.64827347e-01 -7.98861682e-01 6.77903891e-01
5.96302509e-01 -5.29309571e-01 -2.20166430e-01 -2.92261690e-01
1.27848577e+00 9.62761603e-03 1.16128957e+00 -8.95687044e-01
3.20760936e-01 -1.62611812e-01 3.31218302e-01 -2.75778413e-01
1.65159977e-03 -6.17432177e-01 8.28909039e-01 7.08665431e-01
-4.94547874e-01 5.38837671e-01 6.47614181e-01 2.76501596e-01
5.52044451e-01 -1.20808883e-02 2.25218832e-01 -2.28773251e-01
7.58415461e-02 2.38022506e-02 -3.98709297e-01 3.57847124e-01
7.33058393e-01 1.17387794e-01 -6.83048725e-01 -9.98610929e-02
-7.68865287e-01 3.99999648e-01 3.44409078e-01 2.20697448e-01
6.15467668e-01 -9.98970032e-01 -7.97203302e-01 1.47817954e-01
-2.12624431e-01 6.80310279e-02 2.06310600e-01 1.06008613e+00
-8.98038387e-01 4.13944781e-01 -1.45005375e-01 -3.88339549e-01
-9.81148839e-01 6.38278365e-01 3.31458747e-01 -2.51468688e-01
-7.01971054e-01 7.02210844e-01 -2.67374605e-01 -9.11153793e-01
9.73832533e-02 -8.04466486e-01 4.64104742e-01 -1.77897196e-02
7.11386740e-01 6.39065444e-01 -1.64652526e-01 -6.37916327e-02
-5.30836701e-01 6.29139915e-02 3.17827165e-01 -2.54304141e-01
1.72776556e+00 3.93281490e-01 -7.60532618e-01 8.01426172e-01
8.24432015e-01 -7.86313601e-03 -2.67480403e-01 4.09779936e-01
1.02623895e-01 3.51870298e-01 -1.27389535e-01 -1.28130317e+00
-2.64121711e-01 1.01576316e+00 7.52645075e-01 2.15389043e-01
5.04311621e-01 3.36533874e-01 6.84972346e-01 5.75549424e-01
1.32567406e-01 -8.64600420e-01 -3.36264700e-01 3.89986962e-01
9.47902918e-01 -8.75910819e-01 8.89972895e-02 1.90768540e-01
-3.42715770e-01 1.26457179e+00 2.92942762e-01 1.42249212e-01
6.77212298e-01 8.47398460e-01 -3.29286844e-01 -6.90846205e-01
-9.18148816e-01 9.51925740e-02 1.96759731e-01 5.85031986e-01
6.67945027e-01 2.08099395e-01 -1.46988124e-01 1.19795609e+00
-8.51355910e-01 5.72397947e-01 4.08944219e-01 4.67950732e-01
-1.00113610e-02 -1.16854894e+00 -5.18157184e-01 8.13708067e-01
-3.05510849e-01 -5.62294185e-01 -4.25766170e-01 7.69770861e-01
4.47027177e-01 4.08073455e-01 3.29324812e-01 -1.64737180e-01
-2.37049256e-02 6.77308500e-01 6.27087712e-01 -3.68437707e-01
-7.44763672e-01 -5.55352747e-01 1.30906612e-01 -5.30009687e-01
-4.39132005e-02 -6.42795801e-01 -1.88451791e+00 -7.92418540e-01
5.37707238e-03 -1.07071556e-01 8.22601140e-01 8.60782504e-01
7.24361897e-01 6.85292423e-01 -5.18021464e-01 -3.33305001e-01
-3.36475283e-01 -5.48256457e-01 2.71837246e-02 -6.10695928e-02
7.45490193e-02 -2.44213402e-01 7.75931329e-02 9.90767330e-02] | [9.381101608276367, 7.22774076461792] |
5912659a-85d3-488b-8ef6-68719befb435 | reconsider-improved-re-ranking-using-span | null | null | https://aclanthology.org/2021.naacl-main.100 | https://aclanthology.org/2021.naacl-main.100.pdf | RECONSIDER: Improved Re-Ranking using Span-Focused Cross-Attention for Open Domain Question Answering | State-of-the-art Machine Reading Comprehension (MRC) models for Open-domain Question Answering (QA) are typically trained for span selection using distantly supervised positive examples and heuristically retrieved negative examples. This training scheme possibly explains empirical observations that these models achieve a high recall amongst their top few predictions, but a low overall accuracy, motivating the need for answer re-ranking. We develop a successful re-ranking approach (RECONSIDER) for span-extraction tasks that improves upon the performance of MRC models, even beyond large-scale pre-training. RECONSIDER is trained on positive and negative examples extracted from high confidence MRC model predictions, and uses in-passage span annotations to perform span-focused re-ranking over a smaller candidate set. As a result, RECONSIDER learns to eliminate close false positives, achieving a new extractive state of the art on four QA tasks, with 45.5{\%} Exact Match accuracy on Natural Questions with real user questions, and 61.7{\%} on TriviaQA. We will release all related data, models, and code. | ['Wen-tau Yih', 'Yashar Mehdad', 'Sewon Min', 'Srinivasan Iyer'] | 2021-06-01 | null | null | null | naacl-2021-4 | ['triviaqa'] | ['miscellaneous'] | [ 4.15606111e-01 6.59770191e-01 -2.87049613e-03 -4.31194901e-01
-1.99143350e+00 -8.32646132e-01 2.53061235e-01 5.97522020e-01
-5.74594080e-01 9.23234344e-01 5.63289225e-01 -5.36159992e-01
-2.96039402e-01 -6.66037858e-01 -7.45254695e-01 1.30861029e-01
9.75092724e-02 9.60530400e-01 7.40497410e-01 -7.36647666e-01
5.86121440e-01 -1.93316564e-01 -1.69647682e+00 1.08951533e+00
1.26604617e+00 8.64433646e-01 4.94968928e-02 1.15103412e+00
-2.15469033e-01 1.37354839e+00 -7.94471383e-01 -8.49762976e-01
-3.30945015e-01 -5.29195905e-01 -1.76844299e+00 -6.74704790e-01
1.05271816e+00 -2.74287403e-01 -1.71914205e-01 2.14682922e-01
5.65222919e-01 3.90800953e-01 6.33714080e-01 -5.98568618e-01
-7.85516977e-01 6.02424383e-01 1.70910154e-02 5.96655130e-01
1.20415413e+00 1.86767817e-01 1.72583282e+00 -9.38418686e-01
5.16179264e-01 1.02760673e+00 6.22337937e-01 7.88541675e-01
-1.07459533e+00 7.79867824e-03 -3.29718180e-02 6.43000007e-01
-1.02891946e+00 -5.84139109e-01 3.51249009e-01 -8.67323279e-02
1.48538220e+00 7.08421052e-01 2.95773298e-01 9.97131824e-01
-1.79526225e-01 1.04242814e+00 1.06170690e+00 -8.76924753e-01
1.73464399e-02 -9.37757418e-02 6.94914460e-01 6.95825577e-01
-2.41280630e-01 -2.04755053e-01 -7.71956861e-01 -5.52830696e-01
-9.82920229e-02 -7.59788334e-01 -7.86947250e-01 1.00878388e-01
-1.05097663e+00 7.82463729e-01 3.10164303e-01 -9.69641954e-02
-7.02536628e-02 -3.96317959e-01 3.21755081e-01 8.09589088e-01
3.07425588e-01 1.15066087e+00 -1.10999393e+00 -3.26274991e-01
-8.27410698e-01 7.09431767e-01 1.19848526e+00 7.85516560e-01
4.34060067e-01 -6.96117699e-01 -8.37513685e-01 1.22706616e+00
-8.99775028e-02 4.72840279e-01 5.46291411e-01 -1.02390575e+00
7.46290684e-01 5.87172687e-01 1.63980216e-01 -5.33417821e-01
-5.69869399e-01 -5.95660746e-01 2.62380503e-02 -5.38040638e-01
7.73664534e-01 1.91773579e-01 -6.90852165e-01 1.77033103e+00
1.79358512e-01 -3.54262859e-01 2.40424111e-01 6.17463529e-01
1.03622985e+00 5.53172708e-01 2.30385199e-01 -1.98769987e-01
1.61025989e+00 -1.04946923e+00 -4.14784670e-01 -3.89201373e-01
1.04941761e+00 -7.18628347e-01 1.80947566e+00 4.77482140e-01
-1.15167260e+00 -4.10599589e-01 -7.04178631e-01 -5.12795985e-01
-2.55118106e-02 -5.57923391e-02 2.25207612e-01 2.77455032e-01
-8.46877038e-01 4.22029346e-01 -1.33687645e-01 -1.93237439e-01
2.08092555e-01 -1.47784576e-02 -6.06280304e-02 -5.19795895e-01
-1.50602794e+00 1.40506506e+00 3.16879272e-01 -5.17041981e-01
-7.81350374e-01 -9.99364734e-01 -7.79870391e-01 1.87589020e-01
4.67032671e-01 -9.05182719e-01 1.87574387e+00 -5.70889592e-01
-1.28526723e+00 1.07471263e+00 -3.95824015e-01 -6.58743560e-01
1.18999094e-01 -7.83320963e-01 -5.61976612e-01 6.03341401e-01
1.08471125e-01 6.94440246e-01 6.44382894e-01 -9.12211597e-01
-5.62599242e-01 -2.46967450e-01 2.12020367e-01 3.82498085e-01
-8.90349001e-02 -2.88275350e-02 1.52692199e-01 -2.45285824e-01
1.64839357e-01 -5.85800767e-01 2.24624395e-01 -5.41447520e-01
-9.43290591e-02 -1.04361308e+00 -9.84888989e-03 -1.20393384e+00
1.64764619e+00 -1.55929267e+00 3.80289294e-02 -1.95768133e-01
2.71017164e-01 1.94494605e-01 -5.84455013e-01 6.76354766e-01
1.27266556e-01 -9.92582589e-02 -3.30706954e-01 5.81658073e-02
-3.85525376e-02 -1.39750436e-01 -7.09965050e-01 -1.78363353e-01
5.07406592e-01 1.10375249e+00 -1.21089935e+00 -6.28544211e-01
-3.22620004e-01 -4.24247444e-01 -6.00332916e-01 5.79478443e-01
-8.82864237e-01 1.30974278e-01 -1.61033317e-01 5.73570788e-01
1.23942420e-01 -4.95159268e-01 -1.17994674e-01 8.58331248e-02
4.98308182e-01 1.30187011e+00 -4.34839159e-01 1.58679748e+00
-4.87183779e-01 5.08068740e-01 -5.31682730e-01 -5.85752666e-01
7.95657456e-01 2.86994487e-01 -8.09684694e-02 -1.11467254e+00
-2.18723103e-01 5.06399155e-01 2.18909949e-01 -8.80893052e-01
6.70907378e-01 -9.80191212e-03 -2.28282418e-02 4.96902704e-01
4.06583846e-01 -3.91239583e-01 3.22562844e-01 6.28192961e-01
1.49905658e+00 1.65514246e-01 2.82055676e-01 -1.76718324e-01
7.44826615e-01 4.16488171e-01 9.95645002e-02 1.01453459e+00
-1.68184966e-01 7.36775219e-01 5.00611007e-01 -1.72372445e-01
-7.97585130e-01 -1.08639252e+00 -1.18907250e-01 1.69947100e+00
-1.52125135e-01 -6.93094373e-01 -7.96757638e-01 -1.18390191e+00
-1.96341306e-01 1.30475163e+00 -5.03055751e-01 -2.62927264e-01
-8.00063550e-01 -1.18527338e-01 9.12064612e-01 3.45678210e-01
1.06458500e-01 -1.15488970e+00 -5.71329474e-01 2.34641463e-01
-9.76306975e-01 -8.04574490e-01 -5.98488972e-02 2.00099181e-02
-8.82482708e-01 -1.36634886e+00 -5.37953794e-01 -8.47407043e-01
2.27669373e-01 2.73316987e-02 2.09299755e+00 3.98870349e-01
1.36967987e-01 6.35752022e-01 -7.74362206e-01 -3.90074462e-01
-3.97629261e-01 4.77694184e-01 -2.79765487e-01 -6.76905453e-01
7.05350459e-01 -2.60307491e-01 -5.00399053e-01 2.82889396e-01
-4.77707863e-01 -6.68548718e-02 4.48154181e-01 1.18336499e+00
4.71521765e-01 -8.16185772e-01 1.30022728e+00 -6.88607037e-01
9.74485457e-01 -5.83911657e-01 -3.47087383e-02 8.67296815e-01
-4.99563575e-01 2.16036409e-01 5.55171907e-01 -4.02426273e-01
-1.13083291e+00 -5.89061737e-01 -5.63246667e-01 2.02037260e-01
-1.51168585e-01 6.08608484e-01 1.93521693e-01 2.80819446e-01
1.33991981e+00 1.11262798e-01 -1.94477782e-01 -5.79385936e-01
5.75876832e-01 7.22504914e-01 4.62562650e-01 -7.23805785e-01
4.07159358e-01 -2.32939094e-01 -5.68598449e-01 -5.23685992e-01
-1.87196743e+00 -5.52023709e-01 -3.23850691e-01 -2.82483660e-02
4.83640403e-01 -7.78359473e-01 -4.90218937e-01 2.39596814e-02
-1.19411469e+00 -3.70303810e-01 -5.08634508e-01 2.86692083e-01
-6.87286735e-01 2.91358769e-01 -7.31045485e-01 -7.41285324e-01
-5.88452399e-01 -3.25968355e-01 1.01054573e+00 1.78791717e-01
-9.38768744e-01 -7.63363779e-01 4.18412268e-01 1.18665481e+00
3.13103825e-01 -4.69039947e-01 1.42995203e+00 -1.26376438e+00
-2.92547941e-01 -2.14761626e-02 -1.86603032e-02 2.89345533e-01
-4.52724040e-01 -5.64130843e-01 -1.10929108e+00 -8.28369707e-02
-4.78512384e-02 -1.19245279e+00 1.00845921e+00 -1.79759040e-01
1.13830507e+00 -3.48457575e-01 3.57461311e-02 -1.75056294e-01
8.99906993e-01 -3.97826821e-01 6.46752000e-01 3.02381366e-01
2.17032626e-01 9.86025810e-01 1.03884649e+00 6.82072118e-02
6.10530376e-01 2.65455246e-01 3.15440029e-01 4.94659901e-01
-1.59193605e-01 -5.69415748e-01 3.32825363e-01 8.28305781e-01
2.65558571e-01 -3.60965818e-01 -1.05230165e+00 9.85339403e-01
-1.44977260e+00 -1.03664601e+00 -3.29260439e-01 2.20395970e+00
1.29173243e+00 2.80461639e-01 4.59342077e-02 1.07017189e-01
1.44283757e-01 8.93452838e-02 -4.65323210e-01 -4.05015469e-01
-1.93261087e-01 8.16331983e-01 -1.37840331e-01 7.40730166e-01
-7.74068832e-01 9.00742769e-01 6.32011318e+00 9.97187018e-01
-3.89152706e-01 1.95879728e-01 6.13388538e-01 1.02950854e-03
-6.85993433e-01 9.90937948e-02 -7.41568029e-01 -2.36506597e-03
1.35607827e+00 8.14087018e-02 2.01198459e-01 7.89057970e-01
-4.81434166e-01 -4.02762383e-01 -1.31503844e+00 6.18013501e-01
4.38831061e-01 -1.24377418e+00 3.28160286e-01 -6.40104175e-01
4.82868612e-01 4.69989218e-02 -2.13158131e-01 1.03673148e+00
1.04959093e-01 -1.27185309e+00 4.29285020e-01 7.20572531e-01
6.11034811e-01 -6.65566325e-01 8.22622955e-01 8.48542452e-01
-4.61501867e-01 -2.99650520e-01 -3.59441608e-01 -3.36875021e-01
3.33372116e-01 2.78789282e-01 -9.99142647e-01 4.43488300e-01
6.91896021e-01 4.70931344e-02 -1.01816738e+00 9.72122848e-01
-5.94658434e-01 1.26297069e+00 -2.35162094e-01 -5.03924191e-01
-1.19769365e-01 5.02710342e-01 6.17251754e-01 1.10743201e+00
-1.17526665e-01 5.51078081e-01 -1.86890990e-01 4.50931221e-01
-1.96933940e-01 3.30665171e-01 -6.81185499e-02 2.11919680e-01
6.41938925e-01 7.94862807e-01 1.04466043e-01 -5.23060143e-01
-2.99254596e-01 9.36647654e-01 1.11662793e+00 2.67950259e-02
-6.47824526e-01 -5.76095223e-01 1.40689105e-01 1.72335297e-01
1.21197365e-01 1.40919566e-01 -2.00322241e-01 -1.15908527e+00
3.70219350e-01 -1.31174946e+00 9.80621397e-01 -9.33308363e-01
-1.65876043e+00 6.64668500e-01 -4.55559194e-02 -9.74981427e-01
-4.18755114e-01 -5.37620902e-01 -5.05396843e-01 7.88199782e-01
-1.57866991e+00 -8.00744414e-01 -9.24151465e-02 3.94814938e-01
8.51889670e-01 6.33391887e-02 1.10412788e+00 -1.07236989e-01
6.19142652e-02 8.73328626e-01 -2.13474602e-01 -8.72974619e-02
9.19349551e-01 -1.51578224e+00 3.36691469e-01 6.19624376e-01
3.56775939e-01 8.12881291e-01 6.26671970e-01 -4.42937076e-01
-9.79311228e-01 -6.45474970e-01 1.63479137e+00 -1.32902944e+00
5.02147496e-01 -3.57880890e-02 -1.34683335e+00 6.30916119e-01
3.60496551e-01 -3.31408441e-01 9.06491697e-01 5.33866227e-01
-6.76118195e-01 1.26746252e-01 -1.02216566e+00 5.13019741e-01
1.14758575e+00 -7.20559180e-01 -1.67543435e+00 6.42263234e-01
1.21170580e+00 -5.43076456e-01 -8.69778752e-01 6.60358667e-01
2.90123165e-01 -9.09779489e-01 9.64927495e-01 -1.15492427e+00
5.91999114e-01 2.93099228e-02 -9.02004838e-02 -1.14687896e+00
-1.53593093e-01 -4.65558738e-01 -8.75594914e-01 9.24785256e-01
8.58163714e-01 -3.25744152e-01 4.44390416e-01 5.23088515e-01
-2.72879213e-01 -1.26415765e+00 -1.05111122e+00 -5.45344293e-01
2.95655847e-01 -4.80417192e-01 3.06722671e-01 7.00142205e-01
4.13496405e-01 1.14622796e+00 5.46738617e-02 7.24174529e-02
2.48472437e-01 1.10035814e-01 5.12848496e-01 -1.14232588e+00
-5.35278738e-01 -1.25871420e-01 1.87044516e-01 -1.75610960e+00
8.51393417e-02 -6.89694822e-01 1.39410034e-01 -1.60885227e+00
3.12939212e-02 -3.87698174e-01 -1.90597460e-01 2.45309189e-01
-6.76649153e-01 7.64407590e-02 -5.14516607e-02 4.12920453e-02
-1.25082302e+00 7.18747556e-01 1.21275270e+00 4.73093092e-02
-8.31122398e-02 9.49618965e-02 -8.83639097e-01 4.81435180e-01
8.37413251e-01 -3.38327944e-01 -4.93979275e-01 -5.70897043e-01
6.08302534e-01 3.77099991e-01 2.69279152e-01 -9.75092053e-01
1.21898642e-02 -5.31813912e-02 2.49064520e-01 -7.21965790e-01
3.14635068e-01 -1.21685609e-01 -8.31846654e-01 3.03951591e-01
-9.99036193e-01 1.40339717e-01 9.34855342e-02 4.11250710e-01
-2.41074666e-01 -6.31748080e-01 4.89215672e-01 -3.18359792e-01
-6.30075514e-01 -3.45829040e-01 -2.39956662e-01 1.00829089e+00
3.93575132e-01 1.57005012e-01 -9.28480804e-01 -6.74912870e-01
-7.42687821e-01 5.63871562e-01 -6.48029372e-02 5.95999002e-01
7.78144598e-01 -7.95512080e-01 -1.14260364e+00 -2.27976114e-01
6.27462029e-01 1.20443098e-01 3.49447012e-01 7.52285659e-01
-2.97749877e-01 6.04156613e-01 2.80150354e-01 -3.93666178e-01
-1.36404359e+00 3.84995461e-01 2.64356911e-01 -7.80273318e-01
-2.78660208e-01 1.23773909e+00 -4.33563501e-01 -9.36008215e-01
1.98644117e-01 -2.02832296e-01 -5.37762225e-01 -7.56037235e-02
7.94321120e-01 5.36994576e-01 4.76697445e-01 -1.49678692e-01
-7.31816515e-02 9.61659551e-02 -2.85787672e-01 -2.99476326e-01
9.02642310e-01 -1.55966178e-01 -1.77534625e-01 4.24202740e-01
8.79920065e-01 4.15213227e-01 -6.22865081e-01 -4.96897668e-01
3.99469346e-01 -2.13850692e-01 -5.69829941e-01 -1.55646551e+00
1.27009600e-01 8.64732265e-01 1.61873713e-01 9.39287692e-02
1.00873327e+00 3.74961406e-01 1.07309437e+00 9.63608682e-01
2.95041144e-01 -1.08253646e+00 5.05745053e-01 1.12098205e+00
1.13027310e+00 -1.23440337e+00 -1.52253255e-01 -3.22564095e-01
-6.30383432e-01 9.45016205e-01 1.18754983e+00 1.04431085e-01
-3.27132121e-02 -6.92882478e-01 1.50593221e-02 -2.65495479e-01
-1.23973131e+00 -2.54648358e-01 6.67884111e-01 5.35291851e-01
7.02438951e-01 -9.83302519e-02 -5.53340256e-01 8.43791664e-01
-6.39757574e-01 -4.42291945e-01 2.85184950e-01 1.01745021e+00
-8.98496032e-01 -7.77223170e-01 -2.54008293e-01 9.16985154e-01
-3.76299381e-01 -5.92758298e-01 -5.60311139e-01 5.31665921e-01
-2.56665856e-01 1.36335647e+00 2.35308930e-02 -2.48748019e-01
4.09398317e-01 6.86313808e-01 7.72809029e-01 -6.71026587e-01
-8.36668015e-01 -9.35698390e-01 8.72925878e-01 -4.89689231e-01
4.72807884e-03 -5.75676143e-01 -1.14888752e+00 1.51098073e-01
-6.56475246e-01 6.55765533e-01 -8.06732401e-02 1.20498502e+00
5.61592877e-01 2.38017604e-01 3.27443242e-01 1.24783263e-01
-1.18645871e+00 -1.53769922e+00 1.93715826e-01 5.70445657e-01
4.05320048e-01 -2.50955105e-01 -3.72185975e-01 -2.02111050e-01] | [11.301006317138672, 8.008299827575684] |
d39956bd-54ef-4dd9-8481-83c8627050e0 | one-shot-hyperspectral-imaging-using-faced | null | null | http://openaccess.thecvf.com/content_cvpr_2017/html/Takatani_One-Shot_Hyperspectral_Imaging_CVPR_2017_paper.html | http://openaccess.thecvf.com/content_cvpr_2017/papers/Takatani_One-Shot_Hyperspectral_Imaging_CVPR_2017_paper.pdf | One-Shot Hyperspectral Imaging Using Faced Reflectors | Hyperspectral imaging is a useful technique for various computer vision tasks such as material recognition. However, such technique usually requires an expensive and professional setup and is time-consuming because a conventional hyperspectral image consists of a large number of observations. In this paper, we propose a novel technique of one-shot hyperspectral imaging using faced reflectors on which color filters are attached. The key idea is based on the principle that each of multiple reflections on the filters has a different spectrum, which allows us to observe multiple intensities through different spectra. Our technique can be implemented either by a coupled mirror or a kaleidoscope geometry. Experimental results show that our technique is capable of accurately capturing a hyperspectral image by using a coupled mirror setup which is readily available.
| ['Tsuyoshi Takatani', 'Takahito Aoto', 'Yasuhiro Mukaigawa'] | 2017-07-01 | null | null | null | cvpr-2017-7 | ['material-recognition'] | ['computer-vision'] | [ 1.00764287e+00 -6.95994198e-01 4.81209606e-01 -1.12797907e-02
-2.03104019e-01 -6.64434910e-01 2.80802965e-01 -3.55201632e-01
-3.30016881e-01 5.39399326e-01 -5.57087958e-01 -2.30087474e-01
-1.41420767e-01 -1.06737208e+00 -4.12965566e-01 -1.11514461e+00
6.26721919e-01 1.52323470e-01 2.23780885e-01 -2.19517708e-01
1.78847671e-01 6.45094931e-01 -1.71942961e+00 9.21923388e-03
8.98368418e-01 9.99842405e-01 6.65402234e-01 3.47767860e-01
-3.83739471e-02 1.30290151e-01 -4.11605626e-01 5.60581423e-02
5.95934510e-01 -6.55686557e-01 -2.33156726e-01 4.91235524e-01
3.14395726e-01 -3.77975941e-01 -1.52034014e-01 1.51771164e+00
1.31885231e-01 2.83705145e-01 7.07663715e-01 -7.81399906e-01
-5.80386400e-01 2.24072859e-01 -1.07444882e+00 -1.01872303e-01
2.77455568e-01 -8.30272734e-02 6.43723071e-01 -7.00989306e-01
5.79785220e-02 5.97698331e-01 3.05088848e-01 2.57159799e-01
-9.48721647e-01 -5.93166232e-01 -4.53097284e-01 3.04987341e-01
-1.17078364e+00 -3.51381958e-01 9.78373289e-01 -3.29948813e-01
4.47052151e-01 2.94372439e-01 7.67336786e-01 3.85228246e-01
6.45937107e-04 -9.71778408e-02 1.72274375e+00 -6.50550246e-01
1.84845492e-01 2.68365264e-01 9.96708721e-02 5.97801149e-01
5.66737652e-01 3.51772457e-01 -6.91403747e-02 1.09528720e-01
7.32039988e-01 7.31042385e-01 -8.31040919e-01 -8.61779377e-02
-7.41214812e-01 4.31930900e-01 3.06103319e-01 1.86736166e-01
-2.77779013e-01 -5.92367649e-01 -3.30227554e-01 4.23162699e-01
2.91475087e-01 5.53491235e-01 1.33742332e-01 4.55018789e-01
-7.69045651e-01 -4.87693846e-01 5.84417164e-01 3.56613100e-01
8.83303285e-01 -1.95797920e-01 4.95720416e-01 1.06091774e+00
3.66600603e-01 1.10507846e+00 2.91066349e-01 -5.52361250e-01
5.26289381e-02 4.26239878e-01 3.93568933e-01 -6.02038860e-01
-3.18047523e-01 9.04227570e-02 -5.62425196e-01 8.05400968e-01
4.37078685e-01 -2.68055856e-01 -8.52760434e-01 9.97341156e-01
2.83191293e-01 1.77709952e-01 1.74828812e-01 1.17633963e+00
7.01847613e-01 9.94201481e-01 -7.26239860e-01 -6.62937164e-01
1.18402719e+00 -8.57500553e-01 -5.10795891e-01 4.37052287e-02
2.97387987e-02 -1.08308935e+00 9.62495327e-01 7.37243891e-01
-9.65980828e-01 -1.19374655e-01 -1.20192158e+00 2.92092979e-01
-2.38200039e-01 1.64502844e-01 2.69812137e-01 1.01177788e+00
-4.64510202e-01 6.60511792e-01 -5.95954835e-01 -3.39509577e-01
-2.51734376e-01 1.42385885e-01 -3.96145970e-01 -1.46919727e-01
-5.32240331e-01 5.07787883e-01 2.77317852e-01 1.89885795e-01
-2.02286765e-01 -3.34487170e-01 -4.71227348e-01 1.86378673e-01
5.66030502e-01 -7.10454658e-02 8.41011584e-01 -8.38995039e-01
-2.30123615e+00 8.74777973e-01 -2.41660848e-01 2.58414477e-01
4.64808419e-02 -8.96905810e-02 -6.07498288e-01 6.32909656e-01
-4.15040404e-01 -4.78450388e-01 8.99325013e-01 -1.09373367e+00
-4.39739287e-01 -7.13619888e-01 1.49170563e-01 2.94637024e-01
-4.38427478e-01 1.21721193e-01 1.13791591e-02 1.00149773e-01
5.27342021e-01 -1.08212948e+00 1.31094098e-01 -6.15888722e-02
-5.30076087e-01 3.92839491e-01 1.14309597e+00 -2.16730669e-01
5.79325795e-01 -2.19492912e+00 -2.24427983e-01 3.09222817e-01
-2.65208515e-03 6.65830135e-01 7.95094147e-02 8.01331997e-01
-1.65100455e-01 -2.15083167e-01 -3.13392073e-01 2.86411077e-01
-5.93164444e-01 -4.27800745e-01 -2.17464745e-01 6.82265520e-01
-4.57496583e-01 2.28583783e-01 -6.84462726e-01 4.72742133e-02
4.37932611e-01 3.98367971e-01 -8.21394622e-02 2.54593909e-01
1.49671296e-02 5.41973531e-01 -3.46198469e-01 5.95126808e-01
1.11239123e+00 -3.25497568e-01 2.62983561e-01 -2.97363371e-01
-7.61069596e-01 -5.88162839e-02 -1.12891996e+00 1.16534305e+00
-1.95542112e-01 3.87711585e-01 3.47408414e-01 -9.31006908e-01
1.06240118e+00 3.02754402e-01 3.79449129e-01 -5.28663158e-01
9.29095894e-02 3.30723554e-01 -3.50177556e-01 -7.29154408e-01
3.11957926e-01 -8.05115640e-01 6.18020236e-01 6.53842628e-01
-5.43233514e-01 -4.20987487e-01 1.60379186e-01 -4.27422225e-01
5.48011363e-01 -1.52972639e-01 3.42368931e-01 -7.89772943e-02
7.18271077e-01 -9.90947187e-02 3.17862153e-01 5.13313591e-01
2.26884410e-01 5.05259395e-01 -1.42898634e-01 -2.60782450e-01
-9.44255054e-01 -8.28927398e-01 -2.35245526e-01 4.22869831e-01
6.39818013e-01 1.47976652e-01 -4.14884806e-01 5.59035465e-02
-1.54519111e-01 3.37933898e-01 -1.41047284e-01 2.69981865e-02
-3.01955808e-02 -7.07007587e-01 -1.95810981e-02 3.12389079e-02
6.85489058e-01 -8.96914542e-01 -6.41318500e-01 -1.94737166e-02
7.96029791e-02 -8.80372107e-01 -7.86817446e-02 -1.88718125e-01
-9.52562213e-01 -1.46327114e+00 -7.34282970e-01 -5.26053190e-01
5.96151650e-01 1.44881785e+00 4.02125210e-01 -6.07458167e-02
-4.91764665e-01 3.58117610e-01 -4.51326162e-01 -5.84758580e-01
-8.74826536e-02 -4.21739370e-01 -2.02437833e-01 5.20369589e-01
4.28183705e-01 -6.07635975e-01 -7.52710700e-01 2.86476642e-01
-9.38061535e-01 2.83053238e-02 6.13003492e-01 6.35877490e-01
6.19728982e-01 4.60123450e-01 1.42431989e-01 -1.12052810e+00
3.32090586e-01 -1.93184420e-01 -1.21205842e+00 3.72048497e-01
-5.26934624e-01 -3.63651007e-01 9.21588182e-01 -3.41374874e-02
-1.49531972e+00 -1.20144017e-01 2.19108030e-01 -3.21794838e-01
-3.74356270e-01 6.63638890e-01 -3.48624417e-05 -5.72024107e-01
6.73794806e-01 3.16938668e-01 2.02637389e-01 -6.03306711e-01
-7.03909621e-02 9.87437606e-01 2.97599226e-01 -1.58515081e-01
9.10886645e-01 9.88931477e-01 4.51404721e-01 -1.49670339e+00
-9.33271110e-01 -8.87718678e-01 -3.24096531e-01 -3.64144892e-01
7.13623166e-01 -7.06348062e-01 -9.57388401e-01 9.49115753e-01
-7.33790874e-01 3.54750305e-02 2.68451899e-01 9.28294122e-01
-1.06835917e-01 6.13607407e-01 -6.61019444e-01 -1.01051891e+00
-2.69097418e-01 -8.31447840e-01 7.55577862e-01 7.06865191e-01
7.09089279e-01 -9.14958775e-01 1.79428428e-01 6.00024402e-01
2.10201457e-01 2.69327730e-01 7.65496373e-01 2.16922730e-01
-6.34929776e-01 -2.46747062e-01 -4.68900979e-01 1.56897694e-01
5.46161175e-01 3.46396774e-01 -1.07554102e+00 -4.46488142e-01
5.16966045e-01 -4.65916276e-01 8.58209193e-01 3.91798586e-01
1.19688237e+00 3.00778300e-01 -3.57388407e-01 7.79994249e-01
2.01539803e+00 6.04533076e-01 8.53777647e-01 3.06734324e-01
6.36371672e-01 5.85336685e-01 5.62131584e-01 3.40998799e-01
-3.39078754e-01 3.80106539e-01 4.16710258e-01 -2.93211639e-01
5.45751810e-01 2.54242152e-01 1.31182998e-01 5.19802272e-01
-6.22780323e-01 -4.80816990e-01 -5.31204045e-01 -2.29812600e-02
-1.43576145e+00 -1.36933577e+00 -6.73889577e-01 2.66439629e+00
5.29442310e-01 -4.93546277e-01 -1.57887012e-01 2.21856788e-01
8.98890018e-01 -4.43308577e-02 -3.80244464e-01 4.80018463e-03
-1.03414152e-02 5.99941730e-01 5.32904804e-01 5.33938229e-01
-7.14797556e-01 3.85974616e-01 6.77039862e+00 8.68256837e-02
-1.75668013e+00 -1.63055226e-01 -1.61563039e-01 -6.66623488e-02
-2.58036852e-01 -4.54876907e-02 -4.43488061e-01 4.45715249e-01
3.22025388e-01 -2.00904474e-01 5.51164627e-01 2.16325909e-01
2.96771109e-01 -6.01638496e-01 -6.11094952e-01 1.20105159e+00
4.01521288e-02 -9.17537570e-01 1.35203600e-02 1.64438203e-01
6.63776755e-01 -1.52457133e-01 2.00678751e-01 -5.25460005e-01
-3.15706193e-01 -6.75895631e-01 1.61417842e-01 4.20292974e-01
9.42840755e-01 -5.81632793e-01 2.36435235e-01 3.65175545e-01
-9.93191957e-01 -1.41892163e-02 -7.11389601e-01 -2.60592759e-01
-8.55181366e-03 9.31147933e-01 -4.96029615e-01 6.24250174e-01
4.94835496e-01 5.34139812e-01 2.74362892e-01 1.25675738e+00
-2.70508945e-01 5.91962814e-01 -3.07821989e-01 -1.74305849e-02
1.14311673e-01 -1.36464226e+00 5.30110300e-01 5.59717298e-01
6.14969492e-01 6.67901814e-01 2.86428958e-01 9.26689923e-01
8.88958499e-02 9.76980627e-02 -8.35774779e-01 -3.28599691e-01
1.79110304e-01 1.49090290e+00 -7.48367548e-01 -2.35544309e-01
-9.33040798e-01 8.71626198e-01 -7.13444278e-02 5.37468135e-01
-4.91422147e-01 -6.71402633e-01 3.87026668e-01 5.06168902e-02
3.86138529e-01 -1.22761287e-01 1.10697769e-01 -1.46940172e+00
-9.48858336e-02 -4.93044555e-01 2.08007887e-01 -1.07342505e+00
-9.40099537e-01 2.37785771e-01 -3.57421428e-01 -1.32254505e+00
5.53300261e-01 -9.92529392e-01 -7.74399161e-01 1.13370502e+00
-1.94986880e+00 -6.55988932e-01 -8.92429173e-01 9.57068384e-01
4.84494530e-02 1.51911616e-01 9.34192777e-01 -7.07603525e-03
-7.00932086e-01 -1.65953621e-01 7.28525102e-01 -3.76876324e-01
8.33063424e-01 -1.02979946e+00 -5.00767052e-01 7.47442245e-01
-8.84430036e-02 4.92943823e-01 5.35289586e-01 -3.57649028e-01
-1.51824415e+00 -5.80814242e-01 1.81822270e-01 4.11262661e-01
3.97666812e-01 2.70665407e-01 -7.92846859e-01 4.01149035e-01
3.36497158e-01 -2.16760635e-01 1.32100880e+00 -2.17741117e-01
-3.09388161e-01 -2.89383858e-01 -1.22162628e+00 2.07447216e-01
7.30564177e-01 -5.41558087e-01 -4.02615368e-01 7.39114046e-01
-1.16202086e-02 -1.28493547e-01 -6.01682603e-01 2.67066717e-01
6.88788235e-01 -1.37112689e+00 5.79651892e-01 5.99436054e-04
4.56795841e-01 -5.37865162e-01 1.56530216e-02 -1.35550034e+00
-4.07795757e-01 -4.72814262e-01 6.84733331e-01 6.89642429e-01
2.85794258e-01 -1.28240705e+00 7.42851138e-01 3.44283760e-01
-8.43901038e-02 -2.14524969e-01 -1.31692618e-01 -9.22489047e-01
-4.73001182e-01 4.49851602e-01 4.71954942e-01 9.71720040e-01
2.02607498e-01 3.18742216e-01 -2.34803200e-01 5.95812976e-01
1.03500068e+00 8.17124903e-01 3.60238642e-01 -1.66201890e+00
-5.35995483e-01 -2.00160444e-01 -6.25029346e-03 -7.83933938e-01
-9.65800509e-02 -6.64001644e-01 -8.33925530e-02 -1.38320863e+00
3.15808147e-01 -3.20439339e-01 -2.96764612e-01 9.69148427e-02
-1.28072379e-02 4.20379788e-01 2.64377147e-02 5.30995548e-01
7.65179992e-02 3.29353720e-01 1.41794717e+00 -1.02330163e-01
-4.28733021e-01 6.79109544e-02 -4.20844138e-01 7.11307108e-01
8.34617317e-01 -2.87332296e-01 -6.37644708e-01 -3.85651290e-01
2.20981374e-01 1.15921900e-01 2.63696164e-01 -1.11218846e+00
3.20102781e-01 -4.44714189e-01 2.66891807e-01 -3.13905478e-01
6.46517277e-01 -9.40987825e-01 4.66725856e-01 4.77743149e-01
4.49853182e-01 -5.18734872e-01 -8.55062604e-02 5.10194421e-01
-2.14605913e-01 -7.17476070e-01 1.02692735e+00 -6.17246330e-01
-7.28095233e-01 2.27947161e-01 -2.29524016e-01 -5.60827374e-01
1.26056039e+00 -2.90197760e-01 -6.15266502e-01 -1.21089712e-01
-2.57313937e-01 -2.18209639e-01 1.04790735e+00 -3.74984056e-01
6.36258602e-01 -9.35153961e-01 -2.60776639e-01 2.56753504e-01
1.39406577e-01 -3.27216923e-01 3.35507035e-01 6.64217055e-01
-1.00345218e+00 3.08685690e-01 -4.13324118e-01 -4.66443419e-01
-1.56053185e+00 3.13035727e-01 5.45861781e-01 2.33347997e-01
-7.97643244e-01 7.02496171e-01 3.02896619e-01 -1.70021698e-01
-5.04363120e-01 2.66331017e-01 -4.33455467e-01 -9.21929702e-02
8.44028294e-01 5.05187690e-01 4.88180965e-02 -3.38715196e-01
-8.67342693e-04 1.11957192e+00 5.09164631e-02 -2.81053692e-01
1.41164041e+00 -3.39889824e-02 -5.24728060e-01 5.36563396e-01
8.96031439e-01 2.82796353e-01 -1.02952123e+00 -2.30404750e-01
-8.81138146e-01 -8.55789185e-01 2.02782676e-01 -4.48651105e-01
-1.01651025e+00 8.81816447e-01 2.84586757e-01 5.88505268e-01
1.39597034e+00 -4.57107991e-01 6.45787537e-01 6.24773502e-01
7.09329307e-01 -1.05965269e+00 -1.31920293e-01 1.40942127e-01
5.39730966e-01 -1.30122936e+00 8.85442644e-02 -9.58095133e-01
-1.44928038e-01 1.52649021e+00 4.20936942e-01 7.42960498e-02
7.13731647e-01 -1.00225195e-01 3.73153627e-01 -5.67908823e-01
-2.58118689e-01 -1.85545325e-01 7.33425617e-02 4.58016187e-01
3.97106975e-01 2.22186849e-01 -3.97980630e-01 -2.85838842e-01
1.90737411e-01 1.84438303e-02 1.06901538e+00 9.38177407e-01
-7.71535695e-01 -1.04308665e+00 -6.94534302e-01 5.43260396e-01
-3.57925415e-01 3.48901190e-02 -3.63558203e-01 2.16548413e-01
-2.76908666e-01 1.19770896e+00 -1.91111788e-01 -1.86050922e-01
3.51139605e-01 5.01896851e-02 7.60184526e-01 -8.40451419e-01
1.31901070e-01 1.65735945e-01 -3.52295876e-01 -2.56850004e-01
-9.84581769e-01 -6.66939497e-01 -1.02331448e+00 -2.53166586e-01
-8.33340049e-01 1.62597939e-01 8.37562501e-01 8.77449811e-01
-3.02036349e-02 -6.23626783e-02 9.81090963e-01 -6.92654133e-01
-5.06966770e-01 -7.61457920e-01 -1.47079897e+00 3.66213560e-01
2.55596817e-01 -6.62134349e-01 -5.79510510e-01 -1.01294033e-01] | [10.278006553649902, -2.407461643218994] |
7c3d8463-0419-4b87-9d91-dd0f3faa2939 | multi-view-interactive-collaborative | 2305.18306 | null | https://arxiv.org/abs/2305.18306v1 | https://arxiv.org/pdf/2305.18306v1.pdf | Multi-View Interactive Collaborative Filtering | In many scenarios, recommender system user interaction data such as clicks or ratings is sparse, and item turnover rates (e.g., new articles, job postings) high. Given this, the integration of contextual "side" information in addition to user-item ratings is highly desirable. Whilst there are algorithms that can handle both rating and contextual data simultaneously, these algorithms are typically limited to making only in-sample recommendations, suffer from the curse of dimensionality, and do not incorporate multi-armed bandit (MAB) policies for long-term cumulative reward optimization. We propose multi-view interactive topic regression (MV-ICTR) a novel partially online latent factor recommender algorithm that incorporates both rating and contextual information to model item-specific feature dependencies and users' personal preferences simultaneously, with multi-armed bandit policies for continued online personalization. The result is significantly increased performance on datasets with high percentages of cold-start users and items. | ['Umashanger Thayasivam', 'Maria Lentini'] | 2023-05-14 | null | null | null | null | ['collaborative-filtering'] | ['miscellaneous'] | [-8.57795700e-02 -3.16015869e-01 -9.55154657e-01 -5.68523824e-01
-7.76813030e-01 -4.48415846e-01 3.05296421e-01 -1.90263703e-01
-4.03702229e-01 7.20047295e-01 6.39280260e-01 -5.57071209e-01
-6.96935117e-01 -5.02545595e-01 -2.37878248e-01 -4.25526410e-01
7.23952353e-02 8.22671711e-01 -2.49513939e-01 -3.02079409e-01
4.84407187e-01 1.85824156e-01 -1.50558007e+00 3.62389952e-01
9.98239040e-01 7.17959225e-01 3.14231306e-01 7.14959085e-01
-2.61177216e-02 4.17087883e-01 -1.22935928e-01 -5.04969656e-01
3.15792561e-01 4.59217727e-02 -3.96585554e-01 1.69589724e-02
1.31139800e-01 -4.57053095e-01 -2.64421910e-01 4.23546344e-01
3.96765560e-01 8.40369701e-01 8.42564583e-01 -9.49081898e-01
-1.05808485e+00 5.88900089e-01 -8.56971085e-01 4.03760880e-01
3.33878726e-01 -2.60575026e-01 1.47221053e+00 -8.90926540e-01
3.30519080e-01 1.08300436e+00 3.59191895e-01 3.91986370e-01
-1.41676390e+00 -4.06671733e-01 8.70461762e-01 -3.66129540e-03
-5.65647781e-01 -5.76685667e-02 5.86998045e-01 -5.55736959e-01
9.13271368e-01 5.91993988e-01 5.82166314e-01 1.33633363e+00
6.94532171e-02 1.03025973e+00 9.21792030e-01 -1.04174949e-01
2.77602226e-01 5.42320490e-01 6.86352909e-01 -1.84579909e-01
1.49185658e-01 -3.01979762e-02 -5.73112547e-01 -5.88591099e-01
6.30462170e-01 1.04470360e+00 2.23285839e-01 -4.73051161e-01
-7.92781115e-01 1.07987928e+00 -1.74117088e-03 -6.95148483e-02
-7.74367094e-01 -2.21489832e-01 1.43361554e-01 4.23586100e-01
9.51692164e-01 4.49156046e-01 -8.25066745e-01 -2.99027294e-01
-1.01810348e+00 3.47555161e-01 7.91651249e-01 8.77681255e-01
7.17554092e-01 -1.06722035e-01 -3.27182800e-01 1.38141489e+00
3.09580594e-01 6.07492149e-01 6.36758149e-01 -8.41530740e-01
5.53021014e-01 2.92362183e-01 6.26796782e-01 -7.58435428e-01
-3.44598204e-01 -8.87320995e-01 -5.08344948e-01 -3.35098505e-01
2.18645409e-01 -3.34607750e-01 -7.42201328e-01 1.47883463e+00
2.82920212e-01 -1.29312277e-01 -2.57221222e-01 9.11738694e-01
1.73235416e-01 5.70461214e-01 1.56674922e-01 -6.91519797e-01
1.17738116e+00 -1.06074846e+00 -6.35917723e-01 -3.25543672e-01
4.47358847e-01 -5.75068295e-01 1.22469747e+00 9.04231250e-01
-1.23986781e+00 -4.23534840e-01 -4.83085632e-01 5.48820496e-02
-1.59957543e-01 1.32783368e-01 1.10087836e+00 9.31317627e-01
-5.07768095e-01 5.56491256e-01 -7.25432217e-01 -1.05517082e-01
1.95535064e-01 5.33537686e-01 2.14996383e-01 -2.10097432e-01
-8.62842381e-01 5.78814209e-01 -4.33459491e-01 -9.08273235e-02
-3.97801518e-01 -1.01355457e+00 6.70540845e-03 2.07860425e-01
6.20301723e-01 -8.03497851e-01 1.22243166e+00 -8.73859406e-01
-1.60696924e+00 -1.28957585e-01 -2.43299767e-01 -2.06022635e-01
3.17965716e-01 -7.60441124e-01 -4.53429312e-01 -4.89668280e-01
-1.56739980e-01 -6.81458712e-02 8.67243946e-01 -9.71487701e-01
-9.45892572e-01 -8.32988441e-01 1.39032051e-01 5.86698949e-01
-6.98465049e-01 6.38166443e-02 -4.71408159e-01 -6.31886542e-01
-1.80029988e-01 -1.24628019e+00 -5.73293269e-01 -1.10708869e+00
9.56843719e-02 -3.56027514e-01 5.56607187e-01 -7.81857669e-01
1.52005577e+00 -1.94878042e+00 2.28713810e-01 2.77813375e-01
-2.00440809e-01 2.08720807e-02 -6.19474053e-02 5.12382388e-01
1.41241802e-02 1.89419389e-01 6.30802214e-01 -5.09746790e-01
-5.55881448e-02 8.47766027e-02 -5.36427796e-01 1.84610367e-01
-6.34687126e-01 6.73026264e-01 -7.66233981e-01 1.47526875e-01
-5.82573265e-02 2.24450398e-02 -1.08960533e+00 1.67864025e-01
-2.23113820e-01 3.75983715e-01 -6.66798592e-01 4.98903513e-01
3.57185811e-01 -4.12949413e-01 2.32922539e-01 2.38248751e-01
3.79389748e-02 4.24590200e-01 -1.20493245e+00 1.44303417e+00
-9.29957688e-01 1.26140742e-02 6.94212914e-02 -6.74291372e-01
6.40311956e-01 2.61426896e-01 8.44664395e-01 -6.30964637e-01
-2.49121308e-01 -2.10602596e-01 -1.74682304e-01 -3.02419156e-01
1.06219721e+00 6.07639998e-02 -5.44795208e-02 9.37046826e-01
-1.44222453e-01 7.15474308e-01 -1.30718620e-02 6.01425350e-01
9.16953743e-01 3.72481979e-02 4.31649201e-02 -1.17220229e-03
-2.58828044e-01 -1.82357728e-01 6.03070498e-01 1.06883585e+00
3.78623575e-01 3.82243663e-01 1.84117511e-01 -2.37097248e-01
-9.15836275e-01 -7.29709625e-01 6.45297393e-02 2.20291400e+00
-4.02761810e-02 -3.32114667e-01 -9.81534943e-02 -8.22618306e-01
4.42062765e-01 9.64714468e-01 -6.42975152e-01 1.12732410e-01
-1.34511039e-01 -9.90728796e-01 -4.42732185e-01 5.41018069e-01
-5.42562068e-01 -7.86082327e-01 2.36929115e-02 6.13878429e-01
-9.54593047e-02 -5.46104252e-01 -8.70629013e-01 1.09941959e-01
-1.17907369e+00 -6.02338552e-01 -7.19411373e-01 7.48121515e-02
5.51657379e-01 9.03506815e-01 1.09435987e+00 -3.69093657e-01
7.10606501e-02 8.46401274e-01 -5.46629250e-01 -2.31199309e-01
1.94305420e-01 2.78146356e-01 3.51607203e-01 2.07112715e-01
2.74350882e-01 -7.14056671e-01 -1.01391995e+00 8.06609213e-01
-6.05821908e-01 -5.01819551e-02 6.09841704e-01 1.09953046e+00
4.17123556e-01 -4.23866510e-01 1.04167020e+00 -1.78978908e+00
8.87994468e-01 -8.39215100e-01 -2.93510646e-01 3.36169392e-01
-1.31449902e+00 -3.24157685e-01 4.02801514e-01 -9.81422603e-01
-1.61234248e+00 -4.32931542e-01 1.11172058e-01 -1.35610029e-01
-6.07784875e-02 8.33232403e-01 2.74291962e-01 5.65253675e-01
7.10999906e-01 -1.81106567e-01 -3.38537604e-01 -8.46082032e-01
5.77251136e-01 1.00599694e+00 3.69555615e-02 -4.48404282e-01
2.85061955e-01 2.76836902e-01 -5.39563715e-01 -5.76000690e-01
-1.09648693e+00 -1.21069539e+00 -3.35609764e-01 -1.43410251e-01
2.62942761e-01 -1.01335800e+00 -7.38972723e-01 -2.38354325e-01
-4.43244070e-01 -2.32442334e-01 -2.78280973e-01 9.84290302e-01
-6.20079696e-01 2.60896713e-01 -6.26361787e-01 -1.44434202e+00
-4.71938580e-01 -8.75268638e-01 6.05151832e-01 2.53876537e-01
-1.10268794e-01 -9.42573369e-01 1.50717169e-01 1.08897674e+00
3.82606626e-01 -5.45134962e-01 7.73702919e-01 -9.45560277e-01
-4.63231117e-01 -5.55952072e-01 3.51727866e-02 1.57603934e-01
1.13600053e-01 -2.18668774e-01 -5.47900200e-01 -5.79112232e-01
-2.43238971e-01 -1.56976983e-01 7.26095676e-01 7.85671413e-01
1.04234588e+00 -3.95046562e-01 -3.79109651e-01 1.25936911e-01
1.05814755e+00 3.48977655e-01 3.45973879e-01 -8.55754837e-02
6.34716570e-01 6.05174005e-01 1.16316319e+00 9.34589326e-01
3.37658137e-01 8.59972954e-01 2.65253872e-01 1.21378772e-01
5.38515151e-01 -2.26131916e-01 4.03804213e-01 6.97217941e-01
-4.71749157e-01 -2.67126352e-01 -3.55244726e-01 4.11219358e-01
-2.32872605e+00 -1.09984708e+00 -3.70052248e-01 2.41751981e+00
3.94119114e-01 -9.13858637e-02 7.52724826e-01 -3.58568281e-01
5.81903040e-01 -2.24257186e-01 -1.03744388e+00 -5.29409289e-01
3.64704341e-01 -2.97685176e-01 8.34129035e-01 2.67753720e-01
-7.28027761e-01 6.89864695e-01 6.32385111e+00 6.92480147e-01
-4.66628551e-01 2.95482785e-01 6.54907048e-01 -7.77593791e-01
-6.55613601e-01 1.99541990e-02 -1.20501578e+00 4.74923044e-01
1.16620457e+00 1.19134761e-01 9.64062274e-01 1.10382760e+00
4.82569009e-01 -1.16443135e-01 -8.07265759e-01 8.40872943e-01
-2.10200790e-02 -9.76987183e-01 -3.21325868e-01 7.12695360e-01
1.19232893e+00 2.18661249e-01 5.16473711e-01 6.39954567e-01
9.70447063e-01 -5.02882302e-01 2.17594430e-01 8.17350388e-01
4.28865194e-01 -9.47237313e-01 3.69495004e-01 8.02292168e-01
-4.47577566e-01 -7.52460659e-01 -5.83640635e-01 -5.16777672e-02
2.35612988e-01 6.04263663e-01 -5.44598579e-01 5.19515216e-01
6.84278190e-01 7.21676826e-01 -2.12837860e-01 9.98706162e-01
3.66872102e-01 9.22900856e-01 -2.84811974e-01 5.98116824e-03
-4.95733507e-02 -6.52849853e-01 2.75967211e-01 7.87702382e-01
4.13245767e-01 1.45942405e-01 2.88830638e-01 1.32856891e-01
3.41182463e-02 5.63683212e-01 -3.09995174e-01 2.59249564e-02
2.59635150e-01 1.51159561e+00 -4.13522243e-01 -2.02063322e-01
-8.08009148e-01 6.31743789e-01 3.16766530e-01 6.39029443e-01
-5.11587083e-01 3.46951693e-01 7.10378826e-01 3.19739610e-01
4.57203895e-01 -2.76629776e-01 -3.83284301e-01 -1.25338709e+00
-3.50390643e-01 -8.88069570e-01 6.69075906e-01 -5.37936151e-01
-1.53600645e+00 -3.78762335e-02 -1.55218050e-01 -8.66241217e-01
-2.94473886e-01 -1.84103653e-01 -3.33281010e-01 7.78481185e-01
-1.29331076e+00 -1.15939867e+00 2.50044674e-01 6.38795018e-01
8.90780628e-01 -2.34677985e-01 5.83010256e-01 5.00232279e-01
-5.86947143e-01 6.58429682e-01 9.30501044e-01 -6.97232068e-01
1.09485328e+00 -1.36444747e+00 1.08331323e-01 1.98896229e-01
1.96511105e-01 1.00624740e+00 6.82645380e-01 -8.82591188e-01
-1.63068259e+00 -8.01842690e-01 3.56018633e-01 -7.35664368e-01
5.56150854e-01 -4.50889498e-01 -6.82090104e-01 8.28459740e-01
-1.98288038e-01 -4.37638998e-01 1.38774931e+00 1.53955746e+00
-3.09456736e-01 -1.20528102e-01 -7.43742406e-01 4.99416202e-01
1.09751236e+00 -3.26975524e-01 -1.80024132e-01 5.59323370e-01
6.15981221e-01 4.85745035e-02 -1.30273497e+00 1.29918322e-01
8.94483745e-01 -7.64123917e-01 1.03086150e+00 -1.11958969e+00
5.44422306e-02 2.60364175e-01 -7.93892518e-02 -1.42085016e+00
-9.29614246e-01 -7.31252849e-01 -4.32229191e-01 1.15760767e+00
8.24447453e-01 -3.93172652e-01 1.13246262e+00 1.33923662e+00
-7.35211894e-02 -6.84202015e-01 -5.48814297e-01 -4.58215326e-01
-2.69026905e-01 -3.07907015e-01 4.31505948e-01 8.21013629e-01
3.06819320e-01 7.18168676e-01 -1.44363642e+00 -1.13982625e-01
3.54841948e-01 5.15016973e-01 9.90547121e-01 -1.61513138e+00
-8.07196915e-01 -1.89416781e-01 6.38176024e-01 -1.43413019e+00
-2.02041820e-01 -5.82885921e-01 -4.16035414e-01 -1.41935933e+00
5.33667445e-01 -7.33831108e-01 -7.89870262e-01 8.38938728e-02
-3.57636690e-01 -8.50323662e-02 -8.54917988e-02 5.41569233e-01
-1.02733314e+00 4.48392779e-01 1.07757282e+00 1.65004805e-01
-8.05449009e-01 9.98013556e-01 -1.15924752e+00 3.40195417e-01
6.23522699e-01 -4.70754057e-01 -5.90350330e-01 -1.31752238e-01
8.97609055e-01 6.57297790e-01 -3.93453598e-01 -1.68639302e-01
3.03018540e-01 -6.79233074e-01 3.67020488e-01 -9.59865272e-01
5.58422387e-01 -7.59808481e-01 3.36402208e-01 -2.15743408e-01
-9.02905464e-01 7.08626062e-02 -4.76432890e-01 1.44454193e+00
3.45894247e-01 -1.26523003e-01 3.17748457e-01 -6.03990853e-02
2.96534505e-02 6.63088083e-01 -6.13191068e-01 -2.60349572e-01
6.10205591e-01 -2.95076847e-01 -1.06038548e-01 -7.08043873e-01
-1.20068824e+00 3.19778681e-01 2.26296663e-01 6.52557075e-01
1.14263318e-01 -1.01773131e+00 -4.05874491e-01 -1.70535907e-01
7.78001174e-02 -4.57725167e-01 8.57168496e-01 6.55943692e-01
6.70040607e-01 4.78406876e-01 1.32785752e-01 3.41352671e-02
-1.30189288e+00 9.12721157e-01 -4.45153862e-01 -7.47303247e-01
-1.45521268e-01 6.46960855e-01 1.59387842e-01 -4.28477436e-01
2.18666390e-01 2.21244201e-01 -7.37926006e-01 5.68111241e-01
3.57190996e-01 8.01339388e-01 8.32183380e-03 -2.26600990e-01
3.50400776e-01 -2.96725482e-02 -8.00765336e-01 -1.36409327e-01
1.77283633e+00 -6.01439655e-01 2.12869480e-01 7.13204384e-01
6.91174030e-01 2.61054784e-01 -1.31515431e+00 -7.17467606e-01
2.91182157e-02 -8.73067319e-01 3.05841357e-01 -9.31403637e-01
-9.16708946e-01 5.12794554e-01 4.87530857e-01 5.38303673e-01
7.38623619e-01 -2.00931117e-01 7.15993404e-01 4.09202129e-01
3.41202646e-01 -1.62134945e+00 2.87133425e-01 2.50532597e-01
4.54361826e-01 -1.14284337e+00 3.17840874e-01 -3.76181193e-02
-1.11943161e+00 5.41248858e-01 5.29441774e-01 -4.68305536e-02
9.89249468e-01 -2.82467544e-01 -3.06644559e-01 2.22556546e-01
-1.36744308e+00 8.79085660e-02 4.47387785e-01 3.69339108e-01
4.80135560e-01 4.03826863e-01 -3.92489523e-01 1.00724161e+00
1.17080569e-01 -1.34703413e-01 5.02527893e-01 5.69570482e-01
-4.01018023e-01 -1.33183432e+00 -1.00716598e-01 1.41979444e+00
-9.50897694e-01 -1.35042444e-01 1.05660200e-01 4.18311507e-02
-5.39205372e-01 1.15799129e+00 -1.70365691e-01 -2.96777010e-01
3.54884773e-01 6.61072731e-02 1.95602998e-02 -8.68267119e-01
-8.93895566e-01 5.70419788e-01 2.02112481e-01 -3.25688034e-01
-7.82904252e-02 -1.08290255e+00 -4.18333292e-01 -3.10970306e-01
-1.01582420e+00 4.50007886e-01 9.72997665e-01 7.49304950e-01
7.67154515e-01 4.90170956e-01 1.07997942e+00 -7.00936258e-01
-8.15338433e-01 -1.08249629e+00 -9.69941616e-01 4.68204111e-01
3.07234507e-02 -6.95368290e-01 -1.82427600e-01 -2.64855266e-01] | [9.968794822692871, 5.588112831115723] |
dfddf069-8625-4703-b72e-35384debd85a | on-the-prime-number-divisibility-by-deep | 2304.01333 | null | https://arxiv.org/abs/2304.01333v2 | https://arxiv.org/pdf/2304.01333v2.pdf | Classification of integers based on residue classes via modern deep learning algorithms | Computing the residue when dividing a given integer by prime numbers like 2, 3, or others may appear trivial to human beings, but it can be less straightforward for computers in the absence of pre-defined algorithms. In this paper, we tested multiple deep learning architectures and feature engineering approaches on classifying large finite integers (up to $2^{32}$) based on their residues when divided by small prime numbers. It turns out that, regardless of the network architectures (CNN, RNN, Transformer, etc.) or the complexity of the networks, the ability of classification critically depends on the feature space fed into the deep learning models. We also evaluated commercially available Automated Machine Learning (AutoML) pipelines from Amazon, Google and Microsoft, and found that they failed to address this issue unless appropriately engineered features were provided. Furthermore, we introduced a method that utilizes linear regression on Fourier series basis vectors, and successfully demonstrated its effectiveness in the general case. Finally, we evaluated prompt-based learning approaches using GPT-J, Falcon-40B, and LLaMA and demonstrated its apparent failures. To conclude, feature engineering remains an important task to improve the performance, increase the interpretability, and reduce the complexity of models, even in the era of AutoML and Large Language Models (LLMs). | ['Kai Wang', 'Mian Umair Ahsan', 'Jingye Yang', 'Da Wu'] | 2023-04-03 | null | null | null | null | ['feature-engineering', 'automl'] | ['methodology', 'methodology'] | [-8.68237913e-02 -2.00049013e-01 1.66180402e-01 -2.65885502e-01
-7.52582729e-01 -8.01020563e-01 5.23161232e-01 4.10419405e-01
-4.98450637e-01 6.20845258e-01 -4.34812993e-01 -9.39427137e-01
-1.18290074e-01 -1.05602610e+00 -7.59896338e-01 -2.72799611e-01
-4.05123740e-01 2.85966814e-01 -1.06339782e-01 -4.10308033e-01
3.75183731e-01 7.12487578e-01 -1.56536269e+00 3.05756927e-01
5.94616354e-01 1.36086535e+00 -2.89577514e-01 8.70904922e-01
-2.51091480e-01 1.06619596e+00 -7.17970788e-01 -6.07275486e-01
5.37028670e-01 7.71803483e-02 -6.59136355e-01 -5.26667655e-01
6.51135147e-01 -2.26508111e-01 -1.83058619e-01 8.64626169e-01
3.42067152e-01 -1.98604345e-01 4.75469023e-01 -1.41135812e+00
-5.53773940e-01 6.40987396e-01 -1.55687734e-01 2.72969365e-01
3.18828970e-01 3.34503323e-01 1.21964097e+00 -7.66917586e-01
1.44937098e-01 1.31873918e+00 1.27457142e+00 9.52864438e-02
-1.22613323e+00 -8.65285695e-01 -2.63551235e-01 1.79163858e-01
-1.29394805e+00 -2.11777717e-01 2.87723243e-01 -4.75260198e-01
1.75067103e+00 2.26939529e-01 9.42697048e-01 7.49965072e-01
5.86147368e-01 2.83250332e-01 9.90956664e-01 -6.47002757e-01
8.44811499e-02 -1.55025572e-01 2.96709746e-01 1.01571357e+00
2.71791458e-01 6.35472909e-02 -4.04678792e-01 -2.89792508e-01
7.04821289e-01 -2.61309981e-01 2.20343992e-01 2.12202966e-01
-1.11691070e+00 8.54912162e-01 1.61684126e-01 3.30837011e-01
-2.86262453e-01 5.21490335e-01 5.52522838e-01 5.11835635e-01
3.15709054e-01 9.00166512e-01 -9.36338961e-01 -3.10929030e-01
-1.02606106e+00 2.28814691e-01 9.68033493e-01 7.90746748e-01
8.80923986e-01 2.18539715e-01 2.79813051e-01 2.35627174e-01
-8.77765566e-02 1.46230012e-01 4.24987942e-01 -8.58639598e-01
3.63617659e-01 6.72098637e-01 -1.37209669e-01 -1.15283787e+00
-7.43084967e-01 -5.94322860e-01 -8.23144555e-01 1.21213272e-01
6.51144326e-01 -1.90940097e-01 -6.82543814e-01 1.47476292e+00
-2.17798397e-01 -1.88709274e-01 2.23081671e-02 3.77223581e-01
6.53589904e-01 6.87928081e-01 6.80520013e-02 2.75922328e-01
1.35192835e+00 -7.59649158e-01 -2.27132171e-01 -3.60869735e-01
9.21925008e-01 -9.19189870e-01 9.88069832e-01 7.03033447e-01
-9.20677185e-01 -6.94290638e-01 -1.38249886e+00 -2.64352173e-01
-7.04320550e-01 1.94919378e-01 1.10498953e+00 7.51874864e-01
-1.28121042e+00 9.71398234e-01 -7.31563210e-01 -7.02529624e-02
2.18209609e-01 8.23939264e-01 -3.47844243e-01 2.13458344e-01
-1.48985767e+00 1.04169214e+00 3.95584494e-01 3.57529707e-03
-3.24284315e-01 -5.82921028e-01 -8.33982229e-01 3.52027208e-01
-2.04410911e-01 -4.71050739e-01 1.27983689e+00 -1.19966793e+00
-1.23597860e+00 6.83739185e-01 1.76321194e-01 -9.43145514e-01
3.26350331e-01 -7.49609321e-02 -3.89068842e-01 -9.64392442e-03
-1.87422812e-01 8.48057926e-01 7.36454070e-01 -2.83146560e-01
-6.75445080e-01 -1.77989289e-01 3.80694777e-01 -2.96153456e-01
-3.68787855e-01 7.96117783e-02 1.58482805e-01 -4.34512287e-01
-2.70905495e-02 -8.64561617e-01 -3.10239613e-01 -1.99396163e-01
-2.05240995e-01 -4.58824873e-01 1.97678000e-01 -6.42410457e-01
1.18462622e+00 -2.29319310e+00 -4.77635711e-01 3.91403288e-01
2.08976224e-01 2.29565099e-01 -1.46092430e-01 2.92619646e-01
-4.28386241e-01 5.14496446e-01 1.65840447e-01 1.73659593e-01
1.27279550e-01 -1.67359635e-01 -1.89486668e-01 5.25231302e-01
6.54263616e-01 7.31080830e-01 -7.61077881e-01 -1.96336508e-01
3.93689312e-02 5.41611493e-01 -6.48912191e-01 -1.99422061e-01
-2.28397787e-01 -2.02353075e-01 -3.05635314e-02 4.93271738e-01
4.67401057e-01 -5.02703071e-01 1.98608488e-01 -3.22414875e-01
-3.27878624e-01 7.20016181e-01 -1.19863963e+00 1.12731338e+00
-5.41417599e-01 9.88009691e-01 -3.11715215e-01 -9.72585797e-01
1.01979017e+00 2.49454886e-01 4.77002293e-01 -7.53121316e-01
2.28606071e-02 6.03693783e-01 3.69849771e-01 -9.43199992e-02
5.12528539e-01 9.25787315e-02 -4.10419554e-02 4.42673832e-01
1.31323278e-01 -1.97593644e-01 3.82247806e-01 -2.32092738e-01
1.28177166e+00 -1.67049855e-01 4.60453242e-01 -1.57269388e-01
3.43297809e-01 1.36082664e-01 3.58437926e-01 7.55653560e-01
-2.08414756e-02 4.53266084e-01 6.58436179e-01 -9.86939192e-01
-1.10680509e+00 -6.86285615e-01 -2.72828251e-01 1.25731194e+00
-5.24339318e-01 -8.73095989e-01 -5.84585726e-01 -3.23665291e-01
-5.30954525e-02 5.25534570e-01 -2.24006191e-01 -1.94164827e-01
-6.44145489e-01 -7.63716936e-01 1.02425611e+00 5.42924047e-01
4.56546634e-01 -9.56323266e-01 -8.12804937e-01 4.80663419e-01
1.72174245e-01 -1.03025961e+00 -1.30159274e-01 8.09149027e-01
-8.47257078e-01 -1.12090003e+00 -1.95269719e-01 -7.99028456e-01
4.33084160e-01 -1.82334870e-01 1.34586024e+00 1.26341760e-01
-2.99092025e-01 2.63097882e-01 -4.94629741e-02 -4.40325111e-01
-3.79548311e-01 3.40558410e-01 6.08982295e-02 -6.32487178e-01
6.63449287e-01 -5.35656512e-01 -4.34324831e-01 -1.20161297e-02
-7.31082439e-01 -3.98900062e-02 6.68482363e-01 8.68126929e-01
3.29858899e-01 2.86720246e-01 4.83260423e-01 -4.63068783e-01
7.23313332e-01 -1.93105429e-01 -6.60050750e-01 1.80738598e-01
-5.55612803e-01 3.14273924e-01 1.01704192e+00 -4.18213189e-01
-9.15759355e-02 1.36535510e-01 -2.16068447e-01 -2.78882217e-02
2.24298611e-01 6.74306512e-01 1.02893807e-01 -2.85925180e-01
7.95321524e-01 3.35475691e-02 -1.41693801e-01 -2.22868755e-01
2.54071563e-01 6.28648579e-01 3.72523695e-01 -5.77578783e-01
4.21457738e-01 9.42111108e-03 1.44270480e-01 -5.88457644e-01
-4.38035578e-01 1.58136010e-01 -5.34143507e-01 2.48698160e-01
5.01916945e-01 -1.01531363e+00 -9.71964478e-01 5.83690345e-01
-1.41883516e+00 -1.23964131e-01 -1.89591348e-01 3.73939902e-01
-4.37519222e-01 2.69816905e-01 -8.42438936e-01 -5.02584994e-01
-7.43146241e-01 -1.23823953e+00 6.40673995e-01 1.32841498e-01
-5.92733264e-01 -9.03512955e-01 -4.52121019e-01 1.11468406e-02
6.74463034e-01 3.42869796e-02 1.46712863e+00 -1.14627492e+00
-5.44064939e-01 -5.62123537e-01 -2.46283531e-01 6.53456867e-01
7.65099972e-02 3.78762186e-01 -9.01980996e-01 -1.49971038e-01
-6.90017566e-02 -3.54858458e-01 5.81391156e-01 9.58013535e-02
1.29523253e+00 -5.40426910e-01 1.91691160e-01 5.96791923e-01
1.10264671e+00 1.29886612e-01 5.31863749e-01 6.48458481e-01
2.42432624e-01 2.84752160e-01 1.73665836e-01 4.49327946e-01
2.98251987e-01 2.02896178e-01 2.23366886e-01 8.56534243e-02
1.17610291e-01 -5.35366051e-02 5.06260157e-01 8.02167714e-01
-1.21005937e-01 1.00477904e-01 -1.15626705e+00 9.53266323e-02
-1.25178587e+00 -7.89398670e-01 7.76293874e-02 2.07508349e+00
9.57243383e-01 6.85117602e-01 -1.90067053e-01 3.92285675e-01
4.94581848e-01 -9.56446826e-02 -4.33543354e-01 -7.71597564e-01
-2.00277910e-01 7.90185332e-01 6.18875265e-01 1.78657979e-01
-1.08500159e+00 8.53773475e-01 7.43689823e+00 9.10104811e-01
-1.48083961e+00 -2.42509693e-01 9.39384043e-01 1.24242514e-01
-1.45935267e-01 -1.26486778e-01 -8.30626190e-01 2.04049617e-01
1.44701552e+00 -2.36823454e-01 6.63711131e-01 9.66140449e-01
-1.43415511e-01 9.17399153e-02 -1.53774607e+00 1.03330219e+00
-3.13209921e-01 -1.33516550e+00 -3.88474651e-02 -1.61201015e-01
2.26148188e-01 6.29487038e-02 2.36535788e-01 6.44653022e-01
1.80817798e-01 -1.53303194e+00 7.94497073e-01 3.23255837e-01
8.00920606e-01 -8.52776766e-01 6.22890651e-01 2.47586176e-01
-1.08069170e+00 -1.98497668e-01 -4.03979778e-01 -5.66487432e-01
-5.72328329e-01 5.44529080e-01 -1.01462793e+00 1.09304316e-01
5.07900476e-01 5.88858902e-01 -9.36189175e-01 8.00763369e-01
2.84702815e-02 5.25079131e-01 -7.11622357e-01 -2.14751735e-01
3.49147677e-01 1.15989596e-01 2.11869087e-02 1.35781693e+00
4.77734029e-01 -1.46483094e-01 4.70028073e-02 6.50762260e-01
-7.56110400e-02 6.42736778e-02 -4.97423172e-01 -4.58456546e-01
4.32292789e-01 1.31839442e+00 -7.43480980e-01 -2.60264397e-01
-5.89304090e-01 3.86020660e-01 3.03995788e-01 1.36787280e-01
-8.30453098e-01 -5.64405918e-01 6.20791495e-01 1.56203836e-01
2.31247276e-01 -4.72441733e-01 -3.24480057e-01 -1.06373131e+00
8.69322643e-02 -1.45628524e+00 1.57457411e-01 -6.32072151e-01
-1.27223527e+00 7.13238299e-01 -3.80823761e-01 -1.11018860e+00
-5.89020789e-01 -1.06975806e+00 -3.47737938e-01 9.30376828e-01
-1.12537909e+00 -9.08765376e-01 2.71615297e-01 2.89218217e-01
1.92765906e-01 -2.79731333e-01 1.10985684e+00 3.48262638e-01
-2.55156189e-01 7.78366506e-01 5.95359057e-02 3.10177207e-01
4.50211853e-01 -1.10451424e+00 7.93769896e-01 4.30475980e-01
2.79094130e-01 8.40936363e-01 6.19351506e-01 -2.17868224e-01
-1.64266062e+00 -7.85226405e-01 9.71189082e-01 -1.78098768e-01
9.37182128e-01 -3.82719457e-01 -6.55941069e-01 7.50076056e-01
1.06158793e-01 1.84895143e-01 8.20651531e-01 1.03648879e-01
-6.38719201e-01 -1.96492732e-01 -1.02219486e+00 4.82261986e-01
7.21357882e-01 -6.87828541e-01 -3.67189676e-01 3.37172359e-01
7.80594647e-01 -3.93328398e-01 -9.49400246e-01 4.14670616e-01
6.52602971e-01 -1.06285000e+00 1.08437741e+00 -9.43273902e-01
4.64640170e-01 7.20617920e-02 -3.63302410e-01 -1.19422686e+00
-2.92733759e-01 -5.55953205e-01 -1.31404117e-01 8.93676460e-01
6.12191260e-01 -6.96622491e-01 6.86144233e-01 7.87636578e-01
9.44947749e-02 -9.68020558e-01 -9.61767077e-01 -7.11090803e-01
3.58056188e-01 -8.63359332e-01 8.30022156e-01 8.38549256e-01
-1.05035834e-01 2.68422067e-01 5.50164953e-02 -5.21169566e-02
4.28692289e-02 9.05271992e-02 6.35270834e-01 -1.37241554e+00
-4.68923301e-01 -6.88705206e-01 -8.32710266e-01 -7.71208823e-01
1.88331485e-01 -9.75793004e-01 -4.28699940e-01 -1.08488941e+00
-2.39603013e-01 -6.42847717e-01 -3.55830073e-01 8.36491525e-01
1.67448133e-01 8.69166926e-02 1.81034476e-01 -7.74039179e-02
-4.09742534e-01 1.03235237e-01 7.55649567e-01 -3.49180907e-01
9.86946672e-02 4.23694029e-02 -7.83141196e-01 1.00600600e+00
8.76674533e-01 -3.78299445e-01 1.15485698e-01 -4.41351056e-01
8.16534281e-01 -1.71518818e-01 4.36053455e-01 -1.31935596e+00
3.00498515e-01 1.34742200e-01 7.11062908e-01 -4.25742328e-01
1.21616453e-01 -7.32171416e-01 -1.99253168e-02 5.43535352e-01
-3.40302080e-01 6.31572187e-01 5.91412067e-01 -1.43709937e-02
-3.53054181e-02 -3.23028564e-01 4.37977284e-01 -1.98671281e-01
-5.45052290e-01 4.31063883e-02 -5.40525317e-01 1.30743710e-02
5.84049344e-01 9.82330292e-02 -4.09534931e-01 -2.27763921e-01
-4.43725407e-01 -3.65302749e-02 1.88678756e-01 2.29574531e-01
4.05045003e-01 -1.19074607e+00 -4.73052025e-01 4.68126625e-01
-2.54288226e-01 4.35171369e-03 -3.37665707e-01 6.03818238e-01
-1.01350939e+00 6.92958415e-01 -2.99096018e-01 -3.86855930e-01
-1.07172620e+00 3.14283073e-01 6.27442658e-01 -5.43403804e-01
-2.41424873e-01 5.73933303e-01 -4.47602093e-01 -5.85800231e-01
3.74630719e-01 -8.57268274e-01 5.39587289e-02 7.95599297e-02
4.45545882e-01 2.60532379e-01 5.64665079e-01 -2.09749669e-01
-3.15739959e-01 3.39650899e-01 -5.16520217e-02 1.78736359e-01
1.45402765e+00 5.52816510e-01 -3.05912226e-01 3.84563833e-01
1.25572038e+00 -1.91766679e-01 -8.38981926e-01 5.92152253e-02
1.68152779e-01 -5.42423315e-03 -2.00221255e-01 -5.78306258e-01
-9.40074325e-01 1.13858616e+00 5.32003522e-01 5.29293895e-01
1.04940546e+00 -5.05767524e-01 7.73761272e-01 1.02285910e+00
5.02926767e-01 -8.79270434e-01 -9.68592986e-02 1.12558913e+00
5.36123395e-01 -1.08220720e+00 4.79064919e-02 -4.90395688e-02
-4.54025017e-03 1.83582664e+00 4.90465283e-01 -1.48304984e-01
6.96060896e-01 5.29812098e-01 -1.80553302e-01 -1.54545475e-02
-9.03633535e-01 1.74130782e-01 2.08969101e-01 1.76040560e-01
9.43543494e-01 2.23546047e-02 -8.31630901e-02 8.96237314e-01
-9.50270653e-01 8.79065022e-02 4.39189732e-01 9.25311506e-01
-4.01813537e-01 -9.89424586e-01 -5.15625238e-01 6.58789039e-01
-7.54602790e-01 -6.65723860e-01 -1.59165755e-01 8.65083158e-01
3.29556972e-01 7.10322320e-01 2.86180139e-01 -5.52866578e-01
-1.37298964e-02 3.19529384e-01 5.32159865e-01 -5.46297789e-01
-1.01047385e+00 -3.54484826e-01 1.57693505e-01 -2.58941829e-01
8.09924603e-02 -4.37957346e-01 -1.34780812e+00 -5.69633603e-01
-1.55020535e-01 -1.63920633e-02 7.48478591e-01 9.42544460e-01
4.00338322e-01 3.22310150e-01 4.16831136e-01 -7.47841477e-01
-8.95493269e-01 -8.59742105e-01 -3.79110903e-01 -1.15916468e-01
3.19360137e-01 -1.55254617e-01 -5.38646519e-01 -3.36739458e-02] | [8.738572120666504, 7.099165439605713] |
33e0eaf5-f306-4805-8e6c-619dfe59d0a3 | neural-user-simulation-for-corpus-based-1 | 1805.06966 | null | http://arxiv.org/abs/1805.06966v1 | http://arxiv.org/pdf/1805.06966v1.pdf | Neural User Simulation for Corpus-based Policy Optimisation for Spoken Dialogue Systems | User Simulators are one of the major tools that enable offline training of
task-oriented dialogue systems. For this task the Agenda-Based User Simulator
(ABUS) is often used. The ABUS is based on hand-crafted rules and its output is
in semantic form. Issues arise from both properties such as limited diversity
and the inability to interface a text-level belief tracker. This paper
introduces the Neural User Simulator (NUS) whose behaviour is learned from a
corpus and which generates natural language, hence needing a less labelled
dataset than simulators generating a semantic output. In comparison to much of
the past work on this topic, which evaluates user simulators on corpus-based
metrics, we use the NUS to train the policy of a reinforcement learning based
Spoken Dialogue System. The NUS is compared to the ABUS by evaluating the
policies that were trained using the simulators. Cross-model evaluation is
performed i.e. training on one simulator and testing on the other. Furthermore,
the trained policies are tested on real users. In both evaluation tasks the NUS
outperformed the ABUS. | ['Milica Gasic', 'Florian Kreyssig', 'Inigo Casanueva', 'Pawel Budzianowski'] | 2018-05-17 | null | null | null | null | ['user-simulation'] | ['natural-language-processing'] | [ 2.57411867e-01 7.06141710e-01 1.91203237e-01 -4.57350850e-01
-5.40338218e-01 -5.61143160e-01 9.51532841e-01 -4.16317210e-03
-7.83686996e-01 1.19130623e+00 2.07486212e-01 -5.25910497e-01
3.05522114e-01 -4.85322446e-01 -5.33426642e-01 -3.87716085e-01
9.94253010e-02 1.09118462e+00 4.04363722e-01 -7.39436388e-01
2.31561139e-01 -6.89769909e-02 -1.48119426e+00 1.89695746e-01
1.06942987e+00 6.41166210e-01 6.82060421e-01 1.22617829e+00
-8.15495625e-02 9.71449494e-01 -1.05055988e+00 -3.43719162e-02
-7.54126236e-02 -1.01478076e+00 -1.06922460e+00 -8.23341161e-02
-9.67413858e-02 -3.97481352e-01 4.46976162e-02 8.42428923e-01
6.16944671e-01 5.26809692e-01 5.77183604e-01 -1.26287818e+00
8.03814530e-02 4.26489025e-01 4.12032872e-01 -1.70415804e-01
8.16333532e-01 3.63703519e-01 7.09400415e-01 -3.04260492e-01
6.61875010e-01 1.69989204e+00 3.64029318e-01 1.21254551e+00
-1.28019643e+00 -2.12832108e-01 -5.35744168e-02 -4.02732611e-01
-9.30299222e-01 -4.75903958e-01 2.46215403e-01 -4.12827671e-01
1.15061927e+00 2.15622813e-01 4.57283020e-01 1.39386189e+00
-3.99615616e-02 9.99196947e-01 1.46802747e+00 -7.89237618e-01
6.63073123e-01 9.02317107e-01 -2.72845924e-02 7.87758112e-01
-2.84990191e-01 5.77958822e-01 -2.91359603e-01 -3.84330601e-01
7.40722775e-01 -8.01113784e-01 -3.94295365e-01 -3.48658979e-01
-8.93821836e-01 1.06138301e+00 -3.87477800e-02 2.33861968e-01
-4.56047565e-01 -1.14998452e-01 6.24129534e-01 6.21332049e-01
1.04787506e-01 9.62795615e-01 -5.79873502e-01 -5.92883468e-01
-6.46171808e-01 6.04788661e-01 1.72682405e+00 8.26919019e-01
3.02724093e-01 3.00012231e-01 -4.43325669e-01 9.25511718e-01
5.13121367e-01 3.51698369e-01 8.48742306e-01 -9.57233489e-01
2.35060468e-01 4.01658386e-01 4.89370108e-01 -2.21336842e-01
-3.85937452e-01 1.34625524e-01 -2.00395539e-01 3.85511369e-01
5.56753695e-01 -8.33079755e-01 -7.95128047e-01 1.86685801e+00
2.80614018e-01 6.77291080e-02 5.63122690e-01 7.89182723e-01
8.12671065e-01 8.48432422e-01 3.57912034e-01 -5.99935986e-02
1.11892664e+00 -9.16721821e-01 -7.02779114e-01 -1.82507977e-01
9.43863809e-01 -3.59065622e-01 1.21520543e+00 5.21902800e-01
-1.11726367e+00 -5.53028286e-01 -9.32738841e-01 4.36951101e-01
-4.67100054e-01 -2.26884365e-01 2.65292853e-01 7.41476178e-01
-1.42083561e+00 6.38407409e-01 -5.83584011e-01 -6.29632056e-01
-3.52546245e-01 4.98352498e-01 7.54930153e-02 4.30586249e-01
-1.77257061e+00 1.56401098e+00 8.18479836e-01 -4.25292075e-01
-1.16845405e+00 -1.14171676e-01 -1.01783264e+00 -7.26557374e-02
3.30097347e-01 -4.57083583e-01 2.39327192e+00 -1.14275873e+00
-2.47964168e+00 5.83358824e-01 2.17240065e-01 -6.92109346e-01
6.86257064e-01 -3.64813916e-02 -1.94041342e-01 -4.30110618e-02
-8.64953920e-02 8.38713348e-01 5.21380007e-01 -1.44634664e+00
-8.94684196e-01 2.07422316e-01 3.49158168e-01 7.23413229e-01
7.47824609e-02 -7.41722435e-02 -1.57252610e-01 -2.06825778e-01
-8.11849713e-01 -1.09107161e+00 -2.31962726e-01 -7.10259616e-01
-2.52391785e-01 -5.44125378e-01 7.46848822e-01 -5.21148622e-01
1.16518414e+00 -1.65431225e+00 2.07318813e-01 1.18339106e-01
-3.49526465e-01 6.97314084e-01 -5.47666065e-02 7.29860485e-01
2.27781907e-01 -1.74656913e-01 -1.54909194e-01 -2.84163058e-01
1.55133560e-01 5.75098932e-01 -1.71628237e-01 -1.73597723e-01
3.17830853e-02 5.79919517e-01 -1.16202128e+00 -3.65654796e-01
2.36472011e-01 1.02900036e-01 -5.91303110e-01 9.38798726e-01
-8.41782033e-01 7.06176221e-01 -4.72305775e-01 -1.66083872e-01
-1.09429888e-01 -7.26536214e-02 3.76550436e-01 4.14194584e-01
-8.51061344e-02 6.60631716e-01 -9.36499059e-01 1.68861008e+00
-9.38501894e-01 3.71634722e-01 1.66123018e-01 -7.85738349e-01
9.04435933e-01 8.58968496e-01 -2.05941334e-01 -7.96929479e-01
2.77671665e-01 1.90852568e-01 3.78392696e-01 -5.74456036e-01
6.21903658e-01 -2.38086507e-01 -2.15046972e-01 8.43710899e-01
5.36308765e-01 -4.10459548e-01 2.48818606e-01 4.16202873e-01
8.52230012e-01 5.88536263e-01 4.69912142e-01 -3.78493637e-01
7.24136531e-01 3.04378808e-01 1.09145686e-01 1.17187536e+00
-1.11279331e-01 2.56065160e-01 5.39643049e-01 -3.51271741e-02
-9.90425229e-01 -6.03249371e-01 1.39672667e-01 1.48768926e+00
-1.23824276e-01 -2.68304408e-01 -1.39071023e+00 -8.12786281e-01
-2.14518726e-01 1.54775977e+00 -4.43002194e-01 -1.81635112e-01
-4.36294556e-01 -2.81375647e-01 6.22711182e-01 2.26468191e-01
4.34280455e-01 -1.57742858e+00 -9.72470820e-01 5.73189080e-01
9.61449370e-03 -9.23435867e-01 -2.47833908e-01 1.46719664e-01
-6.47166967e-01 -8.53922963e-01 -6.72310829e-01 -5.87806523e-01
4.01985228e-01 -2.42797017e-01 1.35154092e+00 1.36574268e-01
2.82828897e-01 6.89546168e-01 -3.86475891e-01 -9.25759852e-01
-1.40973938e+00 1.41367346e-01 1.72354266e-01 -2.76077360e-01
2.72745639e-01 -1.46513015e-01 -2.21356735e-01 3.19247305e-01
-8.58883917e-01 3.62256318e-01 4.81296718e-01 1.20036709e+00
-2.16074228e-01 -2.95046359e-01 8.50394130e-01 -1.35858083e+00
1.43210709e+00 -3.48084390e-01 -6.42945766e-01 2.64986217e-01
-7.35167623e-01 3.67987216e-01 7.66640007e-01 -3.80669147e-01
-1.36667097e+00 5.09195216e-02 -2.71316171e-01 4.21450101e-02
-5.66751063e-01 3.61900985e-01 -3.47681306e-02 2.71661729e-01
9.57884729e-01 2.60794818e-01 4.35117900e-01 -2.98243791e-01
1.93167344e-01 1.09158683e+00 3.11893374e-01 -8.73169422e-01
1.96400210e-01 -3.68648052e-01 -6.42404020e-01 -1.13005555e+00
-6.62976146e-01 -1.74348533e-01 -2.48800844e-01 -3.58081251e-01
6.98680878e-01 -6.23356462e-01 -7.36666858e-01 3.61161739e-01
-1.14997566e+00 -1.10123575e+00 -2.00814933e-01 3.82796675e-01
-1.00737250e+00 -7.52477273e-02 -4.66410190e-01 -1.29348791e+00
-3.06824476e-01 -1.25632143e+00 9.31467891e-01 4.40882415e-01
-7.91735768e-01 -1.39699984e+00 4.73357111e-01 6.49320930e-02
3.87417853e-01 -5.80100976e-02 7.96240747e-01 -1.45424342e+00
1.90106347e-01 -1.29810899e-01 3.06308448e-01 4.36085224e-01
-1.74390435e-01 -3.63090038e-01 -1.18980360e+00 -3.37295979e-01
7.03788325e-02 -9.47734416e-01 1.00011513e-01 5.48528135e-02
4.56896365e-01 -4.67181534e-01 -1.01898208e-01 -1.36023030e-01
8.97419631e-01 6.84893847e-01 3.39972407e-01 3.34514380e-01
1.44785076e-01 8.95339131e-01 6.49831116e-01 1.56145796e-01
4.06570137e-01 6.91920102e-01 9.57524255e-02 -3.95443775e-02
3.65052849e-01 -4.80560929e-01 6.30958259e-01 5.32892585e-01
1.63247645e-01 -4.46200758e-01 -8.95667315e-01 2.14288116e-01
-2.07789898e+00 -6.33661389e-01 4.10750657e-01 2.21394348e+00
1.18958819e+00 3.52007478e-01 5.33151686e-01 -1.84384093e-01
5.25329173e-01 -7.29579926e-02 -4.39142704e-01 -1.06445944e+00
4.14731145e-01 3.82800549e-01 1.82919607e-01 1.06113887e+00
-7.45565712e-01 1.06226659e+00 5.98057699e+00 3.75794709e-01
-9.00394201e-01 7.43159791e-03 4.16598141e-01 2.89362520e-01
7.01501295e-02 -9.81970727e-02 -8.68608892e-01 4.97654915e-01
1.70503354e+00 -1.53865665e-01 5.07952809e-01 8.77854645e-01
5.26314795e-01 -5.60321391e-01 -1.15922606e+00 3.33278686e-01
-5.36772273e-02 -9.10262704e-01 -1.12812899e-01 -1.48831218e-01
5.17757654e-01 -1.83271378e-01 -3.17185670e-01 1.02566361e+00
1.02624094e+00 -1.15873742e+00 5.39733410e-01 4.34574187e-01
6.01814330e-01 -6.50998414e-01 8.72141123e-01 9.75255609e-01
-1.71582356e-01 2.71943450e-01 -2.96099801e-02 -6.93838447e-02
2.04121590e-01 -4.62624997e-01 -1.72156513e+00 2.44655207e-01
1.68728963e-01 -8.70253742e-02 -1.63636535e-01 7.02352107e-01
-2.28944793e-01 8.24223280e-01 -3.42246443e-01 -6.22201145e-01
6.97335958e-01 -3.14636171e-01 4.26545769e-01 1.41261566e+00
-1.56109214e-01 1.32589936e-01 5.54177165e-01 6.04605973e-01
2.75234938e-01 1.20738782e-02 -8.46143723e-01 -1.90857902e-01
3.27548355e-01 1.00168145e+00 -2.25576103e-01 -6.28571093e-01
-2.42365584e-01 1.02978456e+00 1.11729927e-01 2.71428138e-01
-4.60556746e-01 -3.32660854e-01 3.38510424e-01 -6.93387389e-02
-5.79604320e-02 1.61722556e-01 2.38101929e-01 -7.21801758e-01
-8.10281873e-01 -1.34739447e+00 1.88543856e-01 -7.62542784e-01
-9.42665458e-01 9.40070271e-01 2.98460960e-01 -7.92368770e-01
-1.19854355e+00 -6.16337717e-01 -5.86612463e-01 1.31778026e+00
-1.08816230e+00 -4.14861262e-01 -8.42185467e-02 3.44428152e-01
8.35727513e-01 -5.27449369e-01 1.39881563e+00 -2.80815929e-01
-4.88685906e-01 4.73841995e-01 -6.23375028e-02 -4.80997264e-02
4.98607159e-01 -1.60850537e+00 4.55937624e-01 1.45333886e-01
-1.77203029e-01 4.91005808e-01 1.22182965e+00 -6.10643387e-01
-9.37109232e-01 -7.50582337e-01 6.79899454e-01 -4.86654222e-01
4.79903907e-01 -3.79507512e-01 -1.03937054e+00 5.99056602e-01
6.18411243e-01 -6.84626818e-01 6.19784117e-01 -1.67406008e-01
3.66158128e-01 5.84935963e-01 -1.24851346e+00 7.34961450e-01
4.44697320e-01 -4.06300336e-01 -8.50194991e-01 4.40886050e-01
4.76723075e-01 -8.30292881e-01 -7.44279802e-01 2.25687353e-03
3.55786055e-01 -1.00217688e+00 4.88808811e-01 -1.00567091e+00
1.90167248e-01 5.01688756e-02 1.36087731e-01 -1.97927976e+00
2.15795457e-01 -9.33011353e-01 -8.09393227e-02 9.74056244e-01
5.97586513e-01 -6.70621693e-01 5.84355891e-01 7.51160622e-01
-3.28591140e-03 -6.79324269e-01 -4.80825126e-01 -6.88360155e-01
-3.51565029e-03 1.64409522e-02 3.85575891e-01 5.97576380e-01
1.26243189e-01 1.02192712e+00 -3.76472414e-01 -2.61420190e-01
1.57691449e-01 -4.80141163e-01 9.19954300e-01 -1.16734612e+00
-4.98317391e-01 -3.27565461e-01 1.59136266e-01 -1.09636807e+00
2.51296103e-01 -5.57549357e-01 5.67882180e-01 -1.25454938e+00
-5.05480170e-01 -3.37814689e-01 3.96420360e-02 1.11023448e-01
-1.32236764e-01 -5.43687940e-01 1.34394497e-01 -1.44273907e-01
-5.53455174e-01 6.75217688e-01 1.21992469e+00 4.05872852e-01
-5.89141965e-01 3.13390255e-01 -4.62480128e-01 8.83919299e-01
1.07225049e+00 -3.88030946e-01 -8.59770894e-01 1.37400270e-01
-1.14288747e-01 5.16389847e-01 4.14301716e-02 -8.54775190e-01
2.57646650e-01 -1.60599515e-01 -5.62571594e-03 4.60621789e-02
3.20254564e-01 -4.89800334e-01 -3.13048095e-01 5.61580837e-01
-8.31801951e-01 8.19485784e-02 2.56206095e-01 4.40201908e-01
-1.24411322e-01 -8.17445576e-01 8.58599484e-01 -6.41853571e-01
-7.37755120e-01 -2.99366742e-01 -8.40104818e-01 2.72949040e-01
9.58098650e-01 -2.09508255e-01 1.11245960e-01 -1.01463866e+00
-6.06333435e-01 4.61484551e-01 2.21546575e-01 4.10822183e-01
3.82419050e-01 -7.77170420e-01 -6.57015681e-01 1.69024870e-01
7.72218630e-02 -6.42976724e-03 -2.61627942e-01 3.98547500e-01
-6.03517354e-01 7.18438864e-01 -2.08760470e-01 -4.68636006e-01
-1.07075953e+00 1.00509919e-01 6.95472419e-01 -4.92462426e-01
-3.86397749e-01 6.26554608e-01 1.45052880e-01 -1.09599102e+00
4.91947681e-01 1.43367797e-01 -5.24052978e-01 -1.51378915e-01
4.13882494e-01 4.87494059e-02 -9.23218280e-02 -4.18170780e-01
2.14227065e-01 -5.53890705e-01 -1.18644029e-01 -9.41026866e-01
1.05854511e+00 1.66047975e-01 4.53523993e-01 7.57760525e-01
6.39054537e-01 -5.05660594e-01 -1.28230119e+00 -2.48242229e-01
2.83865213e-01 -6.72676135e-03 -1.86554324e-02 -1.26837254e+00
-1.40764818e-01 7.63664961e-01 5.25524318e-01 6.09080732e-01
4.62726891e-01 -5.34863293e-01 4.66425419e-01 6.73709512e-01
5.13269067e-01 -1.53642154e+00 1.17486380e-02 8.54845464e-01
1.03855526e+00 -1.28028405e+00 -4.09722298e-01 2.12308168e-01
-1.42704189e+00 8.32219899e-01 1.02112496e+00 -1.09535046e-01
1.71811089e-01 1.63252577e-01 3.35812420e-01 -1.19036287e-02
-1.11881566e+00 -1.84843928e-01 -4.40312885e-02 7.08941579e-01
5.29608011e-01 4.50044163e-02 -2.35907063e-01 3.09671611e-01
-4.22380954e-01 1.67030692e-01 6.05824351e-01 1.16352725e+00
-5.25969863e-01 -1.35622978e+00 -2.66808063e-01 5.21502137e-01
-1.25022650e-01 2.51241941e-02 -6.86317205e-01 9.68315005e-01
-4.60117698e-01 1.03020310e+00 -8.28216448e-02 -3.48205268e-01
5.99050820e-01 5.68447769e-01 3.87973070e-01 -1.08674192e+00
-1.16279805e+00 -2.16167003e-01 6.98967457e-01 -4.65613306e-01
-1.10998854e-01 -5.40820301e-01 -1.14487255e+00 9.73253548e-02
-3.34960461e-01 8.03732753e-01 8.11100066e-01 9.78591859e-01
6.16257265e-02 5.02711356e-01 6.90889120e-01 -8.59875441e-01
-1.19419146e+00 -1.44172287e+00 -3.74342591e-01 3.81280154e-01
2.44426936e-01 -6.55481279e-01 -1.54614478e-01 -2.38738731e-01] | [13.0616455078125, 8.032208442687988] |
b0fa1189-f969-4efd-bbcc-bd4a022003bc | are-alphazero-like-agents-robust-to | 2211.03769 | null | https://arxiv.org/abs/2211.03769v1 | https://arxiv.org/pdf/2211.03769v1.pdf | Are AlphaZero-like Agents Robust to Adversarial Perturbations? | The success of AlphaZero (AZ) has demonstrated that neural-network-based Go AIs can surpass human performance by a large margin. Given that the state space of Go is extremely large and a human player can play the game from any legal state, we ask whether adversarial states exist for Go AIs that may lead them to play surprisingly wrong actions. In this paper, we first extend the concept of adversarial examples to the game of Go: we generate perturbed states that are ``semantically'' equivalent to the original state by adding meaningless moves to the game, and an adversarial state is a perturbed state leading to an undoubtedly inferior action that is obvious even for Go beginners. However, searching the adversarial state is challenging due to the large, discrete, and non-differentiable search space. To tackle this challenge, we develop the first adversarial attack on Go AIs that can efficiently search for adversarial states by strategically reducing the search space. This method can also be extended to other board games such as NoGo. Experimentally, we show that the actions taken by both Policy-Value neural network (PV-NN) and Monte Carlo tree search (MCTS) can be misled by adding one or two meaningless stones; for example, on 58\% of the AlphaGo Zero self-play games, our method can make the widely used KataGo agent with 50 simulations of MCTS plays a losing action by adding two meaningless stones. We additionally evaluated the adversarial examples found by our algorithm with amateur human Go players and 90\% of examples indeed lead the Go agent to play an obviously inferior action. Our code is available at \url{https://PaperCode.cc/GoAttack}. | ['Cho-Jui Hsieh', 'I-Chen Wu', 'Meng-Yu Tsai', 'Ti-Rong Wu', 'huan zhang', 'Li-Cheng Lan'] | 2022-11-07 | null | null | null | null | ['game-of-go', 'board-games'] | ['playing-games', 'playing-games'] | [ 2.60455072e-01 6.23734593e-01 1.81765780e-01 4.40085709e-01
-1.02440417e+00 -1.18989909e+00 4.17381853e-01 -6.46209180e-01
-6.28065526e-01 1.06006789e+00 -2.55035996e-01 -7.80342400e-01
-2.62760162e-01 -1.09756231e+00 -9.93494749e-01 -8.44231486e-01
-2.93064237e-01 6.75820291e-01 4.16648984e-01 -8.21649671e-01
1.11836232e-01 1.85686052e-01 -1.12174273e+00 -8.01029280e-02
7.56832242e-01 6.13767505e-01 -1.68952659e-01 1.21394610e+00
5.04234254e-01 1.10791433e+00 -9.19304311e-01 -6.03412032e-01
8.82388175e-01 -7.92070627e-01 -8.86804938e-01 -7.05250978e-01
1.75164342e-01 -5.50502300e-01 -5.70453703e-01 1.45271277e+00
6.45171106e-01 2.84871608e-01 3.54041159e-01 -1.36115456e+00
-1.08268887e-01 9.29259896e-01 -8.64819586e-02 1.05100103e-01
2.25778252e-01 8.23463142e-01 9.17639077e-01 2.35715702e-01
5.50505519e-01 1.09234822e+00 5.57689607e-01 1.07502139e+00
-1.04544270e+00 -1.01297879e+00 -1.08714990e-01 6.21160902e-02
-1.08615339e+00 -2.08687931e-01 4.89474028e-01 -1.58301473e-01
8.09448600e-01 6.39517665e-01 7.96957314e-01 1.37413657e+00
3.00675184e-01 5.31111300e-01 1.23221731e+00 -2.56371319e-01
5.40344775e-01 -6.83479548e-01 -5.38112998e-01 7.53736436e-01
1.16246238e-01 1.00330412e+00 -1.17357709e-01 -4.09735769e-01
7.54299879e-01 -4.74331021e-01 -1.78735070e-02 -9.75446776e-02
-9.20250654e-01 9.98665571e-01 5.26311994e-01 6.33725226e-02
-3.43110561e-01 1.03254974e+00 2.17556924e-01 6.55089319e-01
-2.31515616e-01 1.14016294e+00 -3.25674117e-01 -5.49885452e-01
-5.13037384e-01 7.96577215e-01 1.08013308e+00 3.42193842e-01
3.35147649e-01 4.10194188e-01 4.64365892e-02 -3.88127491e-02
-1.91301242e-01 4.36698109e-01 3.77239347e-01 -1.67742622e+00
3.86475176e-01 1.08860858e-01 2.31520146e-01 -6.71237826e-01
-3.17958415e-01 -4.93129492e-01 -5.62727869e-01 1.05709326e+00
9.69779313e-01 -6.73434496e-01 -8.28048110e-01 2.15911579e+00
2.09810778e-01 1.39213949e-01 2.81761050e-01 6.62749231e-01
1.66108832e-01 5.00159800e-01 -2.57228345e-01 2.19061151e-01
1.11878240e+00 -5.70685923e-01 -1.83638752e-01 -6.08571172e-01
7.97951043e-01 -1.31960034e-01 1.06611896e+00 5.67404687e-01
-1.44839096e+00 2.41298929e-01 -1.05995727e+00 7.05593348e-01
-1.22158043e-01 -7.17311621e-01 6.85995698e-01 8.51243198e-01
-9.10283744e-01 8.13011408e-01 -9.21752930e-01 8.46973271e-04
5.02767801e-01 7.45811343e-01 -2.66670972e-01 8.85478631e-02
-1.79510188e+00 9.95875359e-01 6.51936829e-01 2.68675275e-02
-1.53445351e+00 -2.83808529e-01 -6.26477480e-01 1.26093775e-01
1.17289603e+00 -6.60029411e-01 1.34898460e+00 -1.05474210e+00
-1.67962945e+00 5.44472218e-01 5.28203011e-01 -9.07239914e-01
9.76614058e-01 2.28335224e-02 2.11555716e-02 9.35555771e-02
5.81372194e-02 4.15487856e-01 6.00303531e-01 -9.11892474e-01
-4.87704337e-01 -2.60282278e-01 8.86398077e-01 4.45113868e-01
1.83194607e-01 -4.38663065e-02 5.62843159e-02 -5.21493435e-01
-2.89863378e-01 -1.22790289e+00 -6.82641566e-01 -4.31761563e-01
-5.95082521e-01 4.02026996e-02 3.78163382e-02 -3.99410188e-01
1.07486260e+00 -1.83280241e+00 1.43232167e-01 4.64040548e-01
3.92105311e-01 2.49372974e-01 -2.17534259e-01 3.37909728e-01
3.58588211e-02 4.12906587e-01 -2.10973948e-01 4.53809977e-01
2.99283683e-01 3.19408298e-01 -3.63809913e-01 3.64457905e-01
-2.98721969e-01 1.18776762e+00 -9.73600924e-01 9.96563304e-03
-2.55247891e-01 -2.65207916e-01 -1.00011635e+00 -4.44248766e-02
-3.75420958e-01 4.79635209e-01 -6.50099695e-01 3.58136505e-01
1.39309332e-01 1.98762327e-01 6.80154487e-02 6.16503954e-01
2.82135487e-01 2.01635927e-01 -1.10913348e+00 1.12778187e+00
-2.87839379e-02 3.29280794e-01 6.00571372e-02 -8.73264730e-01
3.41702938e-01 2.05089971e-01 4.90616821e-02 -5.11623263e-01
4.78835464e-01 4.15715843e-01 6.12544239e-01 -2.60833353e-02
3.54745388e-01 -4.84070331e-01 -7.69365370e-01 4.69640106e-01
-2.70998508e-01 -6.38794005e-01 -3.92805263e-02 9.75564644e-02
1.76047707e+00 8.26366767e-02 1.03262559e-01 -1.23718772e-02
1.51786998e-01 3.49883735e-01 6.17550611e-01 1.53169513e+00
-5.59627235e-01 2.48413295e-01 1.09660387e+00 -2.56135404e-01
-9.33978677e-01 -1.14011538e+00 5.97581863e-01 1.14208019e+00
2.21648872e-01 -2.68633187e-01 -1.10494792e+00 -7.50970542e-01
-2.03655049e-01 9.98995066e-01 -8.03309798e-01 -7.86983192e-01
-8.05573523e-01 -3.71790916e-01 1.42636251e+00 3.49999726e-01
7.42679894e-01 -1.33781445e+00 -7.78733313e-01 2.42345497e-01
-7.43627995e-02 -5.63364625e-01 -3.75834316e-01 4.16710258e-01
-4.16702837e-01 -1.09430039e+00 -4.66603726e-01 -3.87682378e-01
1.97385415e-01 -4.92691457e-01 8.80651295e-01 2.66493022e-01
3.32840420e-02 2.38947514e-02 -8.74362588e-02 -3.41871411e-01
-9.19730127e-01 1.46823734e-01 1.98924258e-01 -8.59921098e-01
4.10292391e-03 -5.79896152e-01 -5.67342818e-01 4.03164536e-01
-9.12133515e-01 -1.20575719e-01 2.62719512e-01 9.37613130e-01
1.19231209e-01 4.07246739e-01 4.88285571e-01 -7.72103012e-01
8.27326953e-01 -4.11855400e-01 -9.25784767e-01 1.59847476e-02
-1.55834764e-01 2.45784715e-01 9.63285029e-01 -6.28848433e-01
-5.84041834e-01 -1.54216468e-01 -3.81042361e-01 -3.19905519e-01
7.51059502e-02 2.87212610e-01 -1.58541992e-01 -2.87878513e-01
1.20235777e+00 1.54732391e-01 -9.08538420e-03 8.20340067e-02
4.96822566e-01 7.84750879e-02 8.45597088e-01 -9.39293921e-01
1.23407805e+00 2.20662862e-01 2.39183411e-01 5.51401963e-03
-4.89076525e-01 4.81934845e-01 1.90874085e-01 -2.08814293e-01
8.36743832e-01 -5.08759558e-01 -1.45974565e+00 6.56078160e-01
-7.40981519e-01 -1.00164866e+00 -5.87738752e-01 2.83707738e-01
-1.09118485e+00 2.56512433e-01 -6.02228701e-01 -6.96536124e-01
-7.97967240e-02 -1.37921119e+00 3.23740155e-01 2.69457161e-01
-3.24767858e-01 -6.29125297e-01 3.46335880e-02 3.93104970e-01
4.42911059e-01 5.14887691e-01 7.37971783e-01 -1.02013493e+00
-5.03398597e-01 -4.05194908e-01 5.02716303e-01 1.89200684e-01
-3.66897374e-01 -3.51110041e-01 -6.09561741e-01 -3.30075890e-01
7.90725872e-02 -6.75804973e-01 5.90451717e-01 3.10224742e-01
8.29626381e-01 -6.68101549e-01 -1.35668978e-01 6.29942894e-01
1.08487892e+00 6.33553147e-01 7.51401901e-01 8.89750421e-01
4.51618433e-01 4.32735771e-01 4.75839555e-01 1.38150856e-01
-2.37850342e-02 5.70921540e-01 1.07166791e+00 3.83786231e-01
3.40807587e-01 -4.59334552e-01 6.42576218e-01 -1.72927771e-02
-3.11802536e-01 -4.25189763e-01 -9.73410070e-01 2.80244350e-01
-1.69995844e+00 -1.24525905e+00 2.19868109e-01 2.25256515e+00
1.00390577e+00 7.88004100e-01 2.74125427e-01 1.29153207e-01
5.42274714e-01 -7.28513300e-02 -8.03686202e-01 -7.55198479e-01
1.34008024e-02 5.16934574e-01 1.04667222e+00 7.45647728e-01
-8.28407228e-01 1.43020177e+00 6.19085121e+00 1.38643527e+00
-7.01183617e-01 -2.77886959e-03 4.99297500e-01 -5.08447051e-01
-3.36962223e-01 2.21160546e-01 -4.57814008e-01 4.30889338e-01
1.03289247e+00 -4.70180720e-01 9.94740903e-01 8.36515427e-01
-3.06816213e-02 -1.18193157e-01 -8.30780089e-01 3.61531734e-01
-4.27143812e-01 -1.29362845e+00 -3.50907266e-01 4.31961238e-01
7.25280941e-01 -5.53781018e-02 1.85041070e-01 7.12234139e-01
1.60306084e+00 -1.45816267e+00 9.25897717e-01 2.05649987e-01
7.70816207e-01 -1.11969280e+00 6.73842371e-01 6.59707308e-01
-6.32095456e-01 -3.19605649e-01 -1.19308770e-01 -4.54894215e-01
-3.04919686e-02 -3.15343887e-01 -5.58330834e-01 3.35101724e-01
5.08100629e-01 -3.95496577e-01 -1.63844258e-01 6.98462784e-01
-6.05685711e-01 8.34942818e-01 -7.07545638e-01 -1.31766975e-01
8.17609370e-01 -9.98490676e-02 9.92566884e-01 1.65834531e-01
2.65261024e-01 3.45465720e-01 -7.19789490e-02 1.01430357e+00
-3.44960876e-02 -6.00688338e-01 -8.71590674e-01 -2.85548605e-02
4.37374562e-01 6.26701951e-01 -4.83387738e-01 -2.04899564e-01
3.89071703e-01 1.03705180e+00 2.45600060e-01 2.57465005e-01
-1.06897855e+00 -5.55472016e-01 8.85176837e-01 -1.91863045e-01
1.56508952e-01 9.95773301e-02 -1.88845575e-01 -9.48207378e-01
-2.58416593e-01 -1.40941989e+00 3.84930193e-01 -6.97300255e-01
-8.41365397e-01 5.34902275e-01 -1.14631757e-01 -8.68716121e-01
-8.51853371e-01 -3.49624306e-01 -9.07314420e-01 8.53996336e-01
-7.55176425e-01 -6.58281386e-01 1.81537628e-01 5.89697540e-01
2.36510754e-01 -3.23791981e-01 6.16135061e-01 -2.45442972e-01
-4.29739058e-01 9.09240544e-01 1.38253912e-01 1.84037283e-01
1.47879586e-01 -1.31201851e+00 7.38945723e-01 1.01974022e+00
-8.28408673e-02 4.04800057e-01 1.05324435e+00 -6.22049809e-01
-1.17686176e+00 -7.17324138e-01 2.07862854e-01 -6.21046662e-01
9.94060814e-01 -1.60717383e-01 -4.96462852e-01 9.03207183e-01
-2.05548212e-01 -2.34443381e-01 -4.42229444e-03 -3.38086754e-01
1.01992507e-02 3.38269562e-01 -1.22588301e+00 1.38996208e+00
1.31476355e+00 -4.38853681e-01 -6.48343444e-01 1.20837398e-01
7.11378455e-01 -6.70516193e-01 -3.64342123e-01 2.87749887e-01
5.23103714e-01 -8.95460129e-01 9.71884191e-01 -1.03944933e+00
4.24408466e-01 -2.74086237e-01 -6.94467276e-02 -1.51072085e+00
-1.04988962e-01 -1.30024314e+00 -3.20451558e-02 6.34006262e-01
1.74023569e-01 -9.34495747e-01 1.12041032e+00 6.09837234e-01
1.39291044e-02 -7.38083482e-01 -1.40253890e+00 -1.07905209e+00
9.45538521e-01 -7.05662370e-01 6.26266420e-01 5.42847335e-01
2.30622347e-02 -1.28829673e-01 -5.38378358e-01 6.37259036e-02
4.95193690e-01 -4.51848179e-01 8.17682445e-01 -7.26336956e-01
-8.58597040e-01 -8.06501687e-01 -4.70413387e-01 -9.32597160e-01
1.81718081e-01 -8.42581034e-01 2.69154608e-01 -9.82290864e-01
-2.23478287e-01 -4.76185083e-01 -4.17372398e-02 6.86535954e-01
-2.79264212e-01 2.23107025e-01 5.66678941e-01 1.17968142e-01
-3.89592886e-01 2.16457963e-01 1.37642753e+00 -5.64058721e-02
-4.45010792e-03 2.35604793e-01 -9.21637058e-01 9.96317029e-01
9.76915061e-01 -6.80609345e-01 -2.41652757e-01 4.24587764e-02
8.29871416e-01 4.20938402e-01 6.02513134e-01 -1.14265609e+00
1.56151354e-01 -3.56661201e-01 -2.32834861e-01 1.50358364e-01
2.42880672e-01 -6.21843994e-01 3.69151652e-01 1.19157612e+00
-4.39158469e-01 -1.24841191e-01 2.06636801e-01 5.38158834e-01
2.79190630e-01 -4.77189273e-01 7.21123576e-01 -4.17077154e-01
-3.94927949e-01 5.42016774e-02 -8.28667343e-01 4.78902966e-01
1.23448157e+00 -3.63754869e-01 -4.51324970e-01 -8.60271931e-01
-8.91345859e-01 3.68663460e-01 6.72534943e-01 -1.90501839e-01
1.67318463e-01 -1.05263388e+00 -5.72583318e-01 -1.42022027e-02
-3.54612589e-01 5.86728845e-03 2.07252890e-01 5.20175755e-01
-8.81487072e-01 4.95054983e-02 -3.27317774e-01 1.59854710e-01
-1.03203821e+00 4.34660196e-01 1.00187528e+00 -6.90263629e-01
-4.72604781e-01 9.90747631e-01 4.43604976e-01 -6.26019955e-01
5.40211797e-02 -7.66585097e-02 1.61591634e-01 -3.86610985e-01
1.96753055e-01 3.65478784e-01 -2.71181315e-01 -2.71575421e-01
-2.40595803e-01 2.02883065e-01 1.02149233e-01 -5.74484169e-01
1.01285088e+00 3.27495754e-01 3.91039491e-01 -6.68002516e-02
6.92495167e-01 9.64302868e-02 -1.49632716e+00 1.84306696e-01
-3.46748084e-01 -1.87654555e-01 -3.27363014e-01 -9.32215631e-01
-9.68409836e-01 6.25347853e-01 3.03256840e-01 4.40716684e-01
8.83640528e-01 -1.00163646e-01 8.18851173e-01 6.57281399e-01
7.43420243e-01 -8.32944691e-01 4.82592545e-02 7.85905242e-01
5.69068253e-01 -7.46079862e-01 -4.69938427e-01 1.44872025e-01
-7.14892864e-01 6.76481783e-01 7.00975120e-01 -5.56195080e-01
-1.36581287e-01 2.70862699e-01 -6.40720204e-02 -2.60817915e-01
-7.77705371e-01 -2.41566807e-01 -3.31140399e-01 6.35024011e-01
-5.76303959e-01 1.31516635e-01 -1.61042616e-01 8.08848381e-01
-9.06272233e-01 -1.50717637e-02 8.81427288e-01 9.62183893e-01
-4.34780955e-01 -1.02929544e+00 -5.95551252e-01 3.92225981e-01
-8.53895605e-01 -1.87763005e-01 -4.75268096e-01 9.66181874e-01
-1.03721851e-02 8.11907053e-01 -2.01448813e-01 -5.57340384e-01
2.50844866e-01 5.37778577e-03 4.70634073e-01 -2.61764556e-01
-1.05326366e+00 -4.39152777e-01 3.27854931e-01 -8.32406938e-01
3.89013439e-01 -3.61419916e-01 -1.31105745e+00 -9.50696349e-01
-3.00574243e-01 1.51039124e-01 9.30191949e-02 9.70368624e-01
-4.80487309e-02 5.86263120e-01 5.13498366e-01 -6.46917105e-01
-1.12329113e+00 -5.72593570e-01 -5.23537219e-01 1.37165457e-01
1.38488486e-01 -5.19485354e-01 -8.62020731e-01 -5.63635945e-01] | [3.581098794937134, 1.568150281906128] |
12c31e17-8c8c-4c07-9a20-a1b474555dc7 | fine-tuning-of-convolutional-neural-networks | null | null | https://aclanthology.org/2022.sltat-1.5 | https://aclanthology.org/2022.sltat-1.5.pdf | Fine-tuning of Convolutional Neural Networks for the Recognition of Facial Expressions in Sign Language Video Samples | In this paper, we investigate the capability of convolutional neural networks to recognize in sign language video frames the six basic Ekman facial expressions for ‘fear’, ‘disgust’, ‘surprise’, ‘sadness’, ‘happiness’, ‘anger’ along with the ‘neutral’ class. Given the limited amount of annotated facial expression data for the sign language domain, we started from a model pre-trained on general-purpose facial expression datasets and we applied various machine learning techniques such as fine-tuning, data augmentation, class balancing, as well as image preprocessing to reach a better accuracy. The models were evaluated using K-fold cross-validation to get more accurate conclusions. It is experimentally demonstrated that fine-tuning a pre-trained model along with data augmentation by horizontally flipping images and image normalization, helps in providing the best accuracy on the sign language dataset. The best setting achieves satisfactory classification accuracy, comparable to state-of-the-art systems in generic facial expression recognition. Experiments were performed using different combinations of the above-mentioned techniques based on two different architectures, namely MobileNet and EfficientNet, and is deemed that both architectures seem equally suitable for the purpose of fine-tuning, whereas class balancing is discouraged. | ['Eleftherios Avramidis', 'Fabrizio Nunnari', 'Neha Deshpande'] | null | null | null | null | sltat-lrec-2022-6 | ['facial-expression-recognition'] | ['computer-vision'] | [ 2.19810158e-01 -9.52645242e-02 -5.93718477e-02 -8.71645629e-01
-1.89166620e-01 -1.73267588e-01 7.37343848e-01 -3.58578503e-01
-8.78164589e-01 6.07326090e-01 1.09992802e-01 3.37165147e-02
-2.63833776e-02 -3.87433022e-01 -3.02422464e-01 -7.95665026e-01
-3.10686707e-01 8.09156597e-02 -1.20838158e-01 -4.37792480e-01
1.78570479e-01 8.41258705e-01 -1.96946740e+00 2.91632295e-01
3.18093419e-01 1.29526818e+00 -6.31227553e-01 5.81207991e-01
-1.14626423e-01 1.23563421e+00 -7.40955114e-01 -5.43395102e-01
-6.66847965e-03 -3.68092358e-01 -8.41164768e-01 2.53312051e-01
4.89834905e-01 -3.04283530e-01 7.79484734e-02 8.40982318e-01
5.40125370e-01 4.91407961e-02 5.97564042e-01 -1.52926290e+00
-2.24041492e-01 -1.63833965e-02 -4.75397795e-01 -9.54649225e-02
1.57142058e-01 2.28883848e-01 6.86962366e-01 -7.23040342e-01
8.22036624e-01 1.17556632e+00 6.21496141e-01 9.53725755e-01
-8.93358469e-01 -8.69790018e-01 -1.55842258e-02 3.06923181e-01
-1.27477956e+00 -5.41939795e-01 8.82517159e-01 -3.93794805e-01
1.04179263e+00 2.46159256e-01 7.60327160e-01 1.20469487e+00
-2.17647746e-01 7.43168831e-01 1.38970935e+00 -6.14907742e-01
2.68649399e-01 3.25707227e-01 -4.41193022e-02 4.63671446e-01
-3.30237001e-01 -3.80808413e-02 -3.32334161e-01 -1.28508344e-01
5.72846174e-01 -5.15190244e-01 -3.12162906e-01 -3.53110194e-01
-8.10391486e-01 6.90697134e-01 4.97628719e-01 8.43822539e-01
-6.37921691e-01 1.48288250e-01 8.72661710e-01 5.47165692e-01
3.25149745e-01 3.39533538e-01 -5.43509185e-01 -3.92818749e-01
-8.88821781e-01 3.05024713e-01 7.38189459e-01 3.87184948e-01
6.42822623e-01 1.85282946e-01 -2.46363897e-02 1.01022542e+00
1.44243091e-01 1.06850855e-01 7.32340276e-01 -7.73105919e-01
-1.36560639e-02 5.50740302e-01 7.04197271e-04 -8.06361496e-01
-6.39862895e-01 -1.44680351e-01 -7.21888185e-01 9.19618130e-01
7.03286111e-01 -4.12623793e-01 -1.29171574e+00 2.10689378e+00
-5.85945919e-02 2.25163102e-02 2.98185676e-01 1.20564520e+00
8.84109974e-01 4.17175233e-01 5.78453481e-01 -1.11448236e-01
1.36088216e+00 -6.90247595e-01 -6.98336303e-01 -2.20727455e-03
7.00300992e-01 -7.44156480e-01 1.20462108e+00 5.06285548e-01
-8.36609423e-01 -5.63679874e-01 -8.05662751e-01 2.71086276e-01
-4.22022849e-01 4.38896984e-01 7.47234106e-01 8.07642102e-01
-1.12004864e+00 4.95029211e-01 -5.25054932e-01 -6.43172085e-01
4.84487206e-01 4.27440077e-01 -8.38484883e-01 2.76664585e-01
-1.05902338e+00 1.06320000e+00 2.50495635e-02 4.76122916e-01
-4.51468289e-01 -2.30170950e-01 -5.83042741e-01 -1.86855003e-01
-9.79210902e-03 -2.94134617e-01 1.17483842e+00 -2.05726862e+00
-1.84617507e+00 1.60709381e+00 1.27722677e-02 -2.35208020e-01
5.97486138e-01 -1.18763134e-01 -5.35893023e-01 7.50766397e-02
-6.02435172e-01 8.98009062e-01 1.09459162e+00 -1.04806387e+00
-2.57659435e-01 -3.27139437e-01 1.17047809e-01 3.64507810e-04
-2.05897167e-01 6.71748579e-01 -1.27928123e-01 -4.70700651e-01
-2.43134886e-01 -9.85404253e-01 -1.13311358e-01 5.22636995e-02
1.17467448e-01 -1.58282161e-01 1.00774360e+00 -5.38005888e-01
9.42886889e-01 -2.07431412e+00 3.76217812e-02 5.47568262e-01
-2.49639377e-01 7.43107915e-01 -2.50179470e-01 8.63156393e-02
-6.19145632e-01 -1.48724645e-01 -1.33043319e-01 -1.53980181e-01
-6.16210178e-02 4.14199293e-01 7.49344081e-02 4.07581717e-01
6.36352181e-01 7.69376576e-01 -5.81987381e-01 -4.17290777e-01
3.42943668e-01 8.26727808e-01 -3.99372727e-01 7.37246200e-02
4.23411652e-03 4.74481046e-01 -2.66619653e-01 8.03893209e-01
6.19181514e-01 1.00102641e-01 1.30619347e-01 -3.99576664e-01
9.60923582e-02 -2.34016314e-01 -1.02767718e+00 1.28026366e+00
-4.91723210e-01 8.98125648e-01 2.47383833e-01 -1.14196682e+00
1.29984796e+00 4.50847536e-01 4.79402155e-01 -7.06277311e-01
7.29600132e-01 3.76280427e-01 2.63923198e-01 -8.43716145e-01
1.72997579e-01 -4.78582680e-01 9.09555256e-02 3.07095170e-01
3.06501567e-01 1.19946994e-01 1.27868950e-01 -3.24132234e-01
8.31962645e-01 2.35907286e-01 1.92205638e-01 -1.63795948e-01
1.07976258e+00 -3.41185033e-01 7.35688657e-02 2.09068507e-01
-6.77852213e-01 3.06418628e-01 7.66972482e-01 -4.12010401e-01
-7.75507808e-01 -6.17112219e-01 1.17112501e-02 1.21805084e+00
-4.38172787e-01 -1.62712678e-01 -1.07307947e+00 -7.41735816e-01
-3.25536847e-01 4.02019709e-01 -8.63482594e-01 -8.04961622e-02
-6.00366116e-01 -9.15300310e-01 7.78308868e-01 4.96110111e-01
6.67008698e-01 -1.60206437e+00 -1.06111002e+00 -1.45607129e-01
1.33011781e-03 -9.58659232e-01 2.66739190e-01 2.09638372e-01
-5.32369733e-01 -1.02710867e+00 -1.01639664e+00 -7.14231312e-01
6.82493687e-01 -5.56605697e-01 1.03666019e+00 2.27047056e-01
-2.96091199e-01 3.09456825e-01 -6.22755468e-01 -4.21772271e-01
-4.98325884e-01 -2.68798769e-01 -2.01811120e-01 3.93055111e-01
7.39037871e-01 -4.42900747e-01 -3.95387739e-01 2.71735191e-01
-1.11928046e+00 -4.02105033e-01 6.11563563e-01 8.79408777e-01
-5.56158498e-02 -4.29433376e-01 2.99505562e-01 -4.10897195e-01
7.83659637e-01 1.99653864e-01 -1.94793299e-01 6.01838939e-02
-2.08465278e-01 -5.44384718e-02 3.36013734e-01 -6.03320599e-01
-9.98979330e-01 2.09586509e-02 -5.64711094e-01 -4.78072882e-01
-5.65600276e-01 2.85136908e-01 -8.77992362e-02 -5.33625066e-01
8.06271553e-01 -9.46510881e-02 4.62248236e-01 -9.20345262e-02
2.40488574e-01 8.32704961e-01 5.17709017e-01 -2.85685122e-01
2.46313348e-01 4.52908397e-01 1.22363389e-01 -8.38977039e-01
-4.56718266e-01 -3.40444267e-01 -7.82433152e-01 -6.96718335e-01
7.60869145e-01 -6.45988286e-01 -8.87277186e-01 1.10745180e+00
-8.01858246e-01 -3.78560662e-01 -1.93408161e-01 3.43493670e-01
-6.51886165e-01 1.76512584e-01 -3.26745272e-01 -8.58700633e-01
-3.54278237e-01 -9.61820245e-01 8.54397178e-01 1.09403148e-01
-6.36272669e-01 -7.52330780e-01 -6.48971424e-02 1.70864224e-01
7.82951176e-01 6.20303810e-01 5.96125722e-01 -3.96948546e-01
1.20191649e-01 -4.72920209e-01 -3.33187401e-01 8.80150437e-01
1.26099914e-01 4.55999792e-01 -1.22643840e+00 -1.45589001e-02
-1.83937982e-01 -9.05715406e-01 7.28742182e-01 2.30218932e-01
1.06724286e+00 -1.72624633e-01 1.86209455e-01 4.62564170e-01
1.12101114e+00 1.16702832e-01 1.03885221e+00 6.43562019e-01
1.92432940e-01 8.28742743e-01 6.23015523e-01 5.24123609e-01
-4.64718454e-02 8.29134941e-01 5.23335576e-01 -2.68524557e-01
1.57721564e-01 2.67131835e-01 2.23582998e-01 -8.12959373e-02
-6.21025026e-01 5.69902314e-03 -7.96163261e-01 3.88168991e-01
-1.67394829e+00 -1.02978957e+00 2.29826570e-01 1.80639935e+00
6.38128996e-01 -3.83946635e-02 4.42889154e-01 5.44876158e-01
3.57110530e-01 3.77260447e-01 -1.48232058e-01 -9.71257806e-01
-2.49594584e-01 6.00737095e-01 3.13176125e-01 3.63981187e-01
-1.42897284e+00 1.07680261e+00 5.98851919e+00 5.77359200e-01
-1.96555340e+00 -1.25413880e-01 7.81221867e-01 -7.37503171e-02
3.44171643e-01 -3.99460196e-01 -4.05341476e-01 2.42463693e-01
9.38044667e-01 4.07361746e-01 2.49777406e-01 1.07307684e+00
2.80945390e-01 3.05640157e-02 -6.41032040e-01 1.18315387e+00
3.11420172e-01 -9.20218408e-01 -1.13475407e-02 -3.02596390e-01
4.48353857e-01 -7.39850029e-02 -2.53890842e-01 3.86066377e-01
-1.06529966e-01 -1.06850231e+00 7.56839812e-01 5.87931573e-01
6.85705423e-01 -7.62512207e-01 9.59992409e-01 -2.08136495e-02
-6.34641469e-01 -1.53659210e-01 8.83503258e-02 -3.17378134e-01
2.11110696e-01 8.57916921e-02 -4.66207296e-01 7.32918233e-02
9.53678131e-01 4.52172220e-01 -4.60393071e-01 6.02216601e-01
-3.63179356e-01 5.06332397e-01 -4.35093075e-01 -3.41705203e-01
5.93382359e-01 -1.38131604e-02 2.48625278e-01 1.45341873e+00
5.98470308e-02 -2.14691795e-02 -5.45276701e-01 4.55461681e-01
2.46024475e-01 3.92125219e-01 -2.14165524e-01 6.12498634e-03
-3.16395074e-01 1.43144047e+00 -6.05004191e-01 -3.72536451e-01
-1.03184260e-01 9.68510747e-01 9.71627235e-02 4.05742079e-01
-8.75228465e-01 -2.29189560e-01 9.40190911e-01 7.75811216e-03
2.34474212e-01 2.63839997e-02 -2.60095689e-02 -9.35778737e-01
6.24055676e-02 -9.03380275e-01 4.22852695e-01 -1.08222997e+00
-9.12466705e-01 8.37066412e-01 -1.43365681e-01 -9.39852893e-01
-7.66634226e-01 -1.09135759e+00 -6.64860308e-01 7.48573720e-01
-1.52547312e+00 -1.24334621e+00 -6.95127070e-01 7.68016338e-01
-1.55079007e-01 -1.89063206e-01 1.00310552e+00 4.20572609e-01
-2.09202990e-01 7.42656827e-01 -5.44710875e-01 3.44418675e-01
6.88902199e-01 -8.30410361e-01 -2.24400133e-01 5.18225670e-01
-6.10168539e-02 2.61171192e-01 7.65547395e-01 5.24562001e-02
-7.53169596e-01 -7.67708600e-01 9.02403593e-01 3.87029201e-02
5.42942464e-01 -1.00166641e-01 -8.66175771e-01 4.26621586e-01
4.97108623e-02 9.74648595e-02 6.68019056e-01 -7.39015713e-02
-4.31792676e-01 -1.17455035e-01 -1.27620184e+00 5.64228952e-01
7.31631100e-01 -3.17947090e-01 -4.80932862e-01 3.39939408e-02
-3.32811564e-01 -2.80806214e-01 -7.23970234e-01 6.63039148e-01
9.47334588e-01 -1.27326012e+00 7.23514020e-01 -9.55572605e-01
4.88969684e-01 -1.24948375e-01 -5.16607352e-02 -1.05783832e+00
2.63259783e-02 -3.82594854e-01 3.85172695e-01 1.14930189e+00
2.12181836e-01 -4.79958326e-01 9.71985459e-01 6.22720718e-01
2.85424948e-01 -8.04430842e-01 -1.03407729e+00 -5.51159143e-01
-6.31407797e-02 -6.10277534e-01 4.29952860e-01 8.96165788e-01
-2.19805446e-02 1.31620541e-01 -5.03103554e-01 -4.32505220e-01
1.21164456e-01 -1.59897313e-01 1.29151976e+00 -1.05196428e+00
1.15694091e-01 -8.55715871e-01 -1.04543674e+00 -2.48791724e-01
4.38681364e-01 -5.42725563e-01 -2.20217839e-01 -1.07408929e+00
-2.49388561e-01 -1.82592764e-01 -4.19423997e-01 9.45497632e-01
2.73583740e-01 6.82996392e-01 2.24643141e-01 -2.18046814e-01
-2.15625435e-01 3.25041384e-01 1.02414346e+00 1.31143302e-01
-1.42067596e-01 -1.52265774e-02 -4.97053832e-01 8.59095573e-01
6.95085883e-01 1.82305053e-01 -1.03226818e-01 3.07537112e-02
-9.31042433e-03 -3.27942461e-01 7.18286991e-01 -9.47801054e-01
-2.64128685e-01 1.21577688e-01 3.87126893e-01 -1.74908102e-01
6.40668809e-01 -9.23035324e-01 -6.28431067e-02 4.87230510e-01
-3.55718136e-01 -3.24466199e-01 4.59462225e-01 -2.57412553e-01
-6.65766835e-01 -1.94196582e-01 1.39032590e+00 -1.08270414e-01
-1.27941823e+00 -8.32588673e-02 -3.95318687e-01 -2.71755725e-01
1.08028638e+00 -4.50369537e-01 7.39191324e-02 -8.07222307e-01
-1.07093644e+00 -1.46201953e-01 3.20304483e-01 5.71543515e-01
4.69326735e-01 -1.17223561e+00 -5.92932224e-01 4.37436253e-01
3.08115929e-01 -7.09535897e-01 2.68906355e-01 9.83574867e-01
-6.08466864e-01 1.98054299e-01 -7.69788504e-01 -7.19697356e-01
-1.87706780e+00 1.23902328e-01 6.59859478e-01 -1.36876047e-01
-2.75529146e-01 8.65522504e-01 -3.96561652e-01 -3.92362565e-01
4.27957207e-01 -5.18074691e-01 -4.15959030e-01 1.90279618e-01
4.45411891e-01 3.91868502e-02 1.14866778e-01 -1.15451789e+00
-5.45283973e-01 8.54483306e-01 2.58770108e-01 -9.11116153e-02
1.12576020e+00 2.58929551e-01 -1.13123432e-01 2.84036815e-01
1.28051782e+00 -3.97644907e-01 -9.42541242e-01 1.60689622e-01
-2.51302291e-02 -4.90301251e-01 6.55089468e-02 -1.04920757e+00
-1.21006513e+00 8.60668421e-01 8.09325337e-01 -1.45424604e-01
1.71228731e+00 -1.26152858e-01 2.23929852e-01 2.84666717e-01
1.70242727e-01 -1.09845269e+00 -1.33045390e-01 4.85283107e-01
1.07115221e+00 -1.19654930e+00 -2.43604973e-01 -1.03234708e-01
-8.08914840e-01 1.23024964e+00 6.33596718e-01 -2.15718925e-01
7.74404585e-01 2.78167576e-01 7.44687974e-01 -2.33669207e-01
-4.97724563e-01 -4.40611541e-01 3.23001802e-01 3.51789266e-01
5.96571684e-01 -1.50572211e-01 -4.15722251e-01 1.84713051e-01
-2.87568629e-01 7.15811193e-01 2.32034996e-01 8.17493021e-01
-1.59366563e-01 -9.89797592e-01 -3.60940814e-01 2.65345901e-01
-7.18898594e-01 3.76118839e-01 -6.95242941e-01 1.34321284e+00
2.87254333e-01 6.40312910e-01 1.38653666e-01 -2.16761857e-01
5.37418425e-01 2.97311664e-01 7.52693832e-01 1.64124921e-01
-8.11197877e-01 -8.56209248e-02 3.68141174e-01 -7.21610188e-01
-1.08129096e+00 -6.58160448e-01 -1.03777075e+00 -4.33040746e-02
-2.07082815e-02 -1.94303408e-01 9.25609589e-01 9.89109576e-01
-2.52946559e-03 1.96065977e-01 3.56430650e-01 -1.11931777e+00
-4.70467955e-01 -1.23010254e+00 -4.37403560e-01 9.79454875e-01
1.83204934e-01 -6.75895751e-01 -4.63139981e-01 5.98863736e-02] | [13.530777931213379, 1.8690687417984009] |
5f1dc885-9738-410d-847d-0622fee60759 | adversary-aware-partial-label-learning-with | 2304.00498 | null | https://arxiv.org/abs/2304.00498v1 | https://arxiv.org/pdf/2304.00498v1.pdf | Adversary-Aware Partial label learning with Label distillation | To ensure that the data collected from human subjects is entrusted with a secret, rival labels are introduced to conceal the information provided by the participants on purpose. The corresponding learning task can be formulated as a noisy partial-label learning problem. However, conventional partial-label learning (PLL) methods are still vulnerable to the high ratio of noisy partial labels, especially in a large labelling space. To learn a more robust model, we present Adversary-Aware Partial Label Learning and introduce the $\textit{rival}$, a set of noisy labels, to the collection of candidate labels for each instance. By introducing the rival label, the predictive distribution of PLL is factorised such that a handy predictive label is achieved with less uncertainty coming from the transition matrix, assuming the rival generation process is known. Nonetheless, the predictive accuracy is still insufficient to produce an sufficiently accurate positive sample set to leverage the clustering effect of the contrastive loss function. Moreover, the inclusion of rivals also brings an inconsistency issue for the classifier and risk function due to the intractability of the transition matrix. Consequently, an adversarial teacher within momentum (ATM) disambiguation algorithm is proposed to cope with the situation, allowing us to obtain a provably consistent classifier and risk function. In addition, our method has shown high resiliency to the choice of the label noise transition matrix. Extensive experiments demonstrate that our method achieves promising results on the CIFAR10, CIFAR100 and CUB200 datasets. | ['Ivor W. Tsang', 'Yueming Lyu', 'Cheng Chen'] | 2023-04-02 | null | null | null | null | ['partial-label-learning'] | ['methodology'] | [ 2.15032786e-01 2.43786946e-01 -2.28512540e-01 -2.91870028e-01
-1.33589721e+00 -6.65729046e-01 4.53829467e-01 1.19076356e-01
-5.14900684e-01 9.23252523e-01 -4.14948046e-01 -6.94299564e-02
-2.14086071e-01 -4.27649230e-01 -7.13288903e-01 -1.18277061e+00
5.01490235e-02 4.27522898e-01 -2.90310532e-01 1.78693175e-01
-1.07916884e-01 1.30933017e-01 -9.11845684e-01 2.12984625e-03
8.55806589e-01 1.23069620e+00 -1.25227824e-01 1.28027171e-01
2.94239551e-01 7.91963100e-01 -6.94526672e-01 -7.62143850e-01
6.14061475e-01 -3.89307618e-01 -7.71177292e-01 6.65182471e-02
2.37889871e-01 -7.07808733e-02 -7.61899278e-02 1.55882370e+00
5.25470853e-01 2.23697186e-01 6.22102380e-01 -1.34568143e+00
-1.74313098e-01 7.30905533e-01 -5.24179697e-01 -2.04076595e-03
5.17725013e-02 1.86935842e-01 1.21079791e+00 -6.56349242e-01
5.00716388e-01 1.04891753e+00 5.65497279e-01 5.73052287e-01
-1.53656805e+00 -1.12528813e+00 3.54336470e-01 9.75892842e-02
-1.58089209e+00 -1.94713980e-01 1.02745640e+00 -3.87645423e-01
-1.33266822e-01 5.02906859e-01 4.03237343e-01 1.36387312e+00
-1.21325832e-02 7.47029305e-01 1.61075759e+00 -2.09340215e-01
4.68665421e-01 7.52366483e-01 1.85472593e-01 5.62302589e-01
3.05925012e-02 3.03715825e-01 -5.30695319e-01 -5.34534395e-01
1.34678721e-01 -1.07014433e-01 -3.97055089e-01 -4.16030467e-01
-7.19604492e-01 7.02258766e-01 4.76706058e-01 4.77600992e-02
-8.42960253e-02 5.82644083e-02 2.99700439e-01 3.52224886e-01
5.75389326e-01 4.83211160e-01 -4.94276941e-01 2.65276998e-01
-8.03074181e-01 2.42237851e-01 6.70541346e-01 6.20271385e-01
7.52006471e-01 -1.35879025e-01 -3.34140420e-01 5.13801277e-01
1.93642408e-01 4.23212111e-01 2.88628250e-01 -9.29287970e-01
5.90360284e-01 3.54673654e-01 1.91484779e-01 -1.02435923e+00
-3.23279947e-01 -8.92763197e-01 -9.51954424e-01 8.60963538e-02
6.25729859e-01 -1.83740839e-01 -6.59624636e-01 2.09849072e+00
6.00919187e-01 4.09099519e-01 -9.16819051e-02 1.01553023e+00
2.62250036e-01 3.12362432e-01 1.55794755e-01 -3.65624309e-01
1.35974228e+00 -4.90242392e-01 -7.20942080e-01 -3.89034688e-01
7.49842703e-01 -4.50187296e-01 8.52822781e-01 4.65706766e-01
-6.20153308e-01 -2.03180134e-01 -1.08317101e+00 2.98982978e-01
2.95774192e-02 9.09247994e-03 4.38156605e-01 8.86607885e-01
-4.96629000e-01 4.93677318e-01 -6.78055763e-01 5.60671091e-01
7.44350195e-01 4.81043279e-01 -4.56131816e-01 -3.30866665e-01
-1.39287758e+00 6.06307983e-01 4.03541654e-01 1.75666928e-01
-8.60964596e-01 -7.30729461e-01 -5.22159934e-01 -2.96752788e-02
6.80383861e-01 -3.23068857e-01 9.38758135e-01 -9.11334634e-01
-1.18100250e+00 6.99301183e-01 1.32538080e-01 -5.81850529e-01
9.71440971e-01 1.32014707e-01 -6.14561290e-02 8.55558664e-02
1.48819372e-01 3.51583242e-01 1.15631211e+00 -1.41283619e+00
-5.48583865e-01 -3.57576251e-01 -2.94019398e-03 1.82956785e-01
-1.19223572e-01 -2.97990173e-01 -3.07324976e-02 -8.09799492e-01
6.94134533e-02 -1.30751705e+00 -2.98191577e-01 -2.54266083e-01
-7.81019926e-01 -1.08269043e-01 5.49823403e-01 -4.55259174e-01
9.79974329e-01 -2.40753198e+00 -2.38047436e-01 4.99114156e-01
2.51721740e-01 2.68315673e-01 -2.12886045e-03 -1.12240463e-01
-9.90488306e-02 3.79327655e-01 -2.15264961e-01 -7.05186129e-01
-2.96912119e-02 1.49116278e-01 -3.95696670e-01 7.32611835e-01
5.20172864e-02 6.99618399e-01 -8.18804324e-01 -3.85244817e-01
-1.53785199e-01 3.14187974e-01 -4.98999089e-01 2.07316922e-03
-8.13005418e-02 7.95295537e-01 -7.85757482e-01 4.91655350e-01
7.38222301e-01 -2.34134838e-01 1.34098664e-01 -2.08237264e-02
6.00117147e-01 1.59292579e-01 -1.37080157e+00 1.19531655e+00
-2.39253446e-01 -3.20974700e-02 2.90139616e-01 -1.07613552e+00
6.50099933e-01 5.32212973e-01 4.54304546e-01 -3.87884289e-01
2.95702338e-01 2.52808958e-01 -1.50729090e-01 -2.02724352e-01
7.74593130e-02 -4.13264573e-01 -3.17006022e-01 4.78725433e-01
-5.49799465e-02 7.53723010e-02 -3.00736189e-01 1.83033228e-01
9.60393727e-01 -2.98180729e-01 -7.77952373e-03 -1.22014336e-01
4.31633681e-01 -3.21193457e-01 1.02219236e+00 9.53523040e-01
-3.86044115e-01 4.49278653e-01 5.67299604e-01 -2.43726969e-01
-6.23360991e-01 -7.52108276e-01 -4.45560545e-01 6.38609171e-01
8.48597363e-02 -1.75201923e-01 -6.43037438e-01 -1.07629764e+00
-1.98618054e-01 5.31338990e-01 -6.33619905e-01 -3.94305706e-01
-2.83846885e-01 -1.05661952e+00 4.78938550e-01 1.31638488e-02
3.94583166e-01 -6.43727541e-01 -7.45415092e-02 1.08839892e-01
-3.95575076e-01 -1.07169354e+00 -6.57052457e-01 5.70193589e-01
-4.06704485e-01 -1.07940221e+00 -2.42549911e-01 -5.60222626e-01
8.48505497e-01 -4.10845801e-02 5.90690970e-01 -1.86986700e-01
6.51925728e-02 1.14160903e-01 -1.87935084e-01 -2.77235746e-01
-3.87933224e-01 -7.24899247e-02 1.44978642e-01 3.50371659e-01
3.40519808e-02 -5.08297920e-01 -4.51162219e-01 2.28777334e-01
-8.37325096e-01 -3.10898513e-01 2.62359917e-01 1.35036516e+00
7.90549517e-01 5.92461228e-01 7.95226216e-01 -1.21055448e+00
3.44409585e-01 -5.33847690e-01 -6.57384932e-01 2.16766968e-01
-8.38027298e-01 7.26527050e-02 7.64787972e-01 -8.85214448e-01
-1.01771152e+00 2.15062931e-01 1.43489316e-01 -4.17480558e-01
5.83798699e-02 3.05944979e-01 -4.72969383e-01 -2.17297181e-01
5.79791546e-01 -8.80719870e-02 -7.12470785e-02 -3.70584399e-01
4.82742876e-01 4.52610105e-01 3.52021098e-01 -6.96126819e-01
9.26224709e-01 4.43017095e-01 2.07451254e-01 -8.45902562e-02
-1.14941096e+00 -2.49242142e-01 -1.74029604e-01 2.96663735e-02
4.95763451e-01 -1.03903747e+00 -7.76162386e-01 4.72806990e-01
-8.37297201e-01 -1.12110689e-01 -4.98843819e-01 5.02220809e-01
-3.00520062e-01 2.34341651e-01 -7.48204947e-01 -9.98728216e-01
-1.50696307e-01 -1.43527210e+00 7.09291518e-01 -6.86985701e-02
-2.71157194e-02 -8.85090113e-01 -2.39386484e-01 7.22343028e-01
-1.40640105e-03 3.22363615e-01 9.55139339e-01 -1.01844716e+00
-7.55344093e-01 -7.01304734e-01 1.05128422e-01 5.71045339e-01
3.98260169e-02 -6.36939287e-01 -1.31122208e+00 -5.41405082e-01
5.65097511e-01 -6.53545082e-01 7.98258126e-01 2.26326317e-01
1.15551865e+00 -8.63927126e-01 -2.18779489e-01 4.84807283e-01
1.22783148e+00 1.99503034e-01 3.14201191e-02 2.10612968e-01
7.46930659e-01 6.52343869e-01 5.98711193e-01 4.32483256e-01
3.17030787e-01 6.03650868e-01 3.94997180e-01 7.34432340e-02
2.67898351e-01 -2.49072239e-01 2.78860033e-01 3.99730414e-01
5.39190412e-01 -1.69054702e-01 -5.50814867e-01 1.20891176e-01
-1.62652528e+00 -7.53592253e-01 1.88085660e-01 2.37935638e+00
1.26140952e+00 1.87520951e-01 -1.71631783e-01 3.52413356e-01
7.16871560e-01 1.24971449e-01 -6.01872325e-01 9.53167826e-02
-1.24208987e-01 -4.45791893e-02 5.82012832e-01 4.55493957e-01
-1.29309595e+00 6.22798562e-01 5.54123020e+00 1.40184283e+00
-1.16096437e+00 3.23536634e-01 1.16192377e+00 -2.81802207e-01
-1.87489510e-01 -2.21180846e-03 -7.87403166e-01 7.73758471e-01
7.43258059e-01 -2.22414602e-02 4.75704104e-01 6.91616356e-01
8.69183708e-03 -8.11021179e-02 -9.76354718e-01 9.10894811e-01
-1.12858616e-01 -8.13173413e-01 -4.24403787e-01 1.99517816e-01
8.83710742e-01 -1.64412394e-01 5.59272349e-01 3.09594870e-01
4.69667375e-01 -9.66349304e-01 8.78939211e-01 3.48550737e-01
7.07419693e-01 -9.21266794e-01 8.63808572e-01 7.26074755e-01
-6.59365952e-01 -3.50798994e-01 -1.57270536e-01 1.95730865e-01
-8.48617777e-02 1.09557331e+00 -8.37388158e-01 5.54447889e-01
4.90057498e-01 1.66682333e-01 -5.24699390e-01 7.28035748e-01
-3.51785064e-01 8.05094063e-01 -4.53700989e-01 3.63650978e-01
7.37350062e-02 -2.15411156e-01 6.57540500e-01 6.31267190e-01
5.58690913e-02 6.75811917e-02 5.57121873e-01 7.86737084e-01
-5.45928359e-01 1.49894223e-01 -4.12219554e-01 1.72916427e-01
6.97912276e-01 1.25443614e+00 -5.97134173e-01 2.98110377e-02
1.48986168e-02 6.82891011e-01 3.81219715e-01 4.41601485e-01
-6.55489028e-01 3.00399303e-01 2.44670466e-01 -1.29721478e-01
3.25053297e-02 3.16726089e-01 -4.10663724e-01 -1.06703651e+00
3.47517908e-01 -1.05874598e+00 4.79522854e-01 -1.58890367e-01
-1.58347261e+00 5.15444398e-01 -2.18948781e-01 -1.30266178e+00
-1.02155007e-01 -6.03003949e-02 -3.32570255e-01 8.49847853e-01
-1.45361471e+00 -1.07209849e+00 3.47494066e-01 4.64882851e-01
-3.92057672e-02 -1.17854021e-01 7.21118033e-01 3.64913583e-01
-7.94730008e-01 1.08293509e+00 1.32818177e-01 1.56478837e-01
8.32163990e-01 -1.29337156e+00 -2.66600400e-01 7.48755276e-01
1.35491163e-01 4.58828092e-01 5.62335789e-01 -5.46137929e-01
-8.80708933e-01 -1.28903162e+00 7.18495190e-01 -5.45944273e-01
6.78256273e-01 -5.47251642e-01 -9.53830481e-01 5.21207154e-01
-4.61845219e-01 5.36059976e-01 7.63727963e-01 1.96588561e-02
-6.31664693e-01 -2.86698103e-01 -1.44435894e+00 3.46083462e-01
5.33746600e-01 -8.02663088e-01 -1.60192236e-01 4.40980345e-01
7.44901896e-01 -4.18105096e-01 -7.66164660e-01 3.62384468e-01
1.45937607e-01 -5.23141384e-01 8.62031877e-01 -4.02890414e-01
-5.94659299e-02 -2.15241268e-01 4.37467999e-04 -1.24067307e+00
-2.29418844e-01 -6.72108173e-01 3.75121199e-02 1.63872600e+00
5.17401874e-01 -7.24927962e-01 8.54528964e-01 9.82268572e-01
3.91061842e-01 -8.64490628e-01 -1.43418694e+00 -8.35163474e-01
9.77263898e-02 -5.48298836e-01 4.62708354e-01 1.20239508e+00
-1.61769539e-01 2.03303099e-01 -7.89914846e-01 2.93198258e-01
8.72717142e-01 -4.10037637e-02 3.51765215e-01 -1.22246230e+00
-3.80059689e-01 -1.18343674e-01 -1.56078145e-01 -6.00257933e-01
5.45077384e-01 -1.22174060e+00 1.96154356e-01 -5.48053145e-01
2.60009170e-01 -1.25706816e+00 -6.14736259e-01 6.20903254e-01
-5.79793870e-01 2.92303711e-01 3.41268152e-01 3.45723778e-01
-6.51768386e-01 5.60131729e-01 1.29616106e+00 -3.70273113e-01
-1.87351882e-01 4.94380027e-01 -7.97602892e-01 6.27158523e-01
6.67858005e-01 -1.14485598e+00 -4.36876535e-01 1.70005515e-01
2.49626085e-01 1.05984285e-01 5.22715867e-01 -6.15758598e-01
2.62762666e-01 -6.50777481e-03 1.12417847e-01 -1.86242431e-01
3.81469965e-01 -9.40826237e-01 2.75159329e-01 4.23594594e-01
-6.75349772e-01 -6.06313765e-01 -2.48453721e-01 8.95471454e-01
-2.09364127e-02 -3.17553014e-01 9.06141579e-01 1.28305063e-01
1.05093583e-01 4.27143097e-01 -7.04243705e-02 2.98265606e-01
8.11567426e-01 2.72665888e-01 -1.94711223e-01 -2.36552030e-01
-8.02653015e-01 4.11482275e-01 1.88672274e-01 9.04781744e-02
1.36354133e-01 -1.38622987e+00 -5.16353130e-01 1.66808546e-01
-6.42474815e-02 2.45634824e-01 3.38042259e-01 8.19813013e-01
3.72783780e-01 1.66621372e-01 3.30114126e-01 -3.49120706e-01
-9.94122148e-01 7.44121194e-01 3.22241485e-01 -6.57293856e-01
-5.21758080e-01 9.92004097e-01 3.27136904e-01 -3.18716198e-01
4.62380350e-01 2.52541229e-02 2.13593803e-02 3.00227493e-01
4.14718390e-01 2.72749931e-01 1.19535297e-01 -6.13928378e-01
-2.53378212e-01 1.47519901e-01 -1.90508619e-01 -9.66342837e-02
8.84000540e-01 -3.02547216e-01 6.04569018e-02 3.55112523e-01
1.20324469e+00 2.10123777e-01 -1.48970699e+00 -5.79699874e-01
9.22503397e-02 -3.24129760e-01 1.54368222e-01 -8.45857203e-01
-1.21494031e+00 5.07516146e-01 7.57207751e-01 1.24120690e-01
1.05428779e+00 -1.94740534e-01 6.65855050e-01 1.72933429e-01
5.13854563e-01 -1.09072578e+00 3.52331214e-02 1.10097349e-01
3.97210419e-01 -1.40460801e+00 -7.75182769e-02 -4.16858882e-01
-8.01761508e-01 3.87952238e-01 2.53604829e-01 1.41976193e-01
7.05647230e-01 2.97971278e-01 1.81159332e-01 1.03922658e-01
-5.93983948e-01 1.69390053e-01 2.36183867e-01 4.32076722e-01
-3.62763494e-01 2.98871577e-01 -1.81485593e-01 1.21702993e+00
-1.77158594e-01 -3.27384889e-01 4.31078762e-01 5.88319540e-01
2.10354440e-02 -1.33662641e+00 -5.04894137e-01 3.80721122e-01
-1.01050687e+00 -2.82601118e-02 -1.18121870e-01 4.20988083e-01
4.90540564e-01 9.86912251e-01 -5.14801025e-01 -3.81815225e-01
4.69159670e-02 2.23387897e-01 7.95813464e-03 -4.81428653e-01
-6.37086511e-01 1.73229739e-01 -1.31741777e-01 -3.86941522e-01
-2.25304469e-01 -6.00506008e-01 -9.99577463e-01 -6.32463470e-02
-7.54893482e-01 4.85788018e-01 3.93025935e-01 1.14205015e+00
8.82510990e-02 2.62099564e-01 1.05388951e+00 -4.48147148e-01
-1.05129635e+00 -6.57240868e-01 -9.75886583e-01 5.82295656e-01
4.77768570e-01 -6.20698512e-01 -8.54037523e-01 -1.64506957e-01] | [9.377683639526367, 3.9858951568603516] |
6d188968-a8f1-476c-8592-c81b404475c4 | pain-evaluation-in-video-using-extended | null | null | http://proceedings.mlr.press/v116/xu20a.html | http://proceedings.mlr.press/v116/xu20a/xu20a.pdf | Pain Evaluation in Video using Extended Multitask Learning from Multidimensional Measurements | Previous work on automated pain detection from facial expressions has primarily focused on frame-level pain metrics based on specific facial muscle activations, such as Prkachin and Solomon Pain Intensity (PSPI). However, the current gold standard pain metric is the patient's self-reported visual analog scale (VAS) level which is a video-level measure. In this work, we propose a multitask multidimensional-pain model to directly predict VAS from video. Our model consists of three stages: (1) a VGGFace neural network model trained to predict frame-level PSPI, where multitask learning is applied, i.e. individual facial action units are predicted together with PSPI, to improve the learning of PSPI; (2) a fully connected neural network to estimate sequence-level pain scores from frame-level PSPI predictions, where again we use multitask learning to learn multidimensional pain scales instead of VAS alone; and (3) an optimal linear combination of the multidimensional pain predictions to obtain a final estimation of VAS. We show on the UNBC-McMaster Shoulder Pain dataset that our multitask multidimensional-pain method achieves state-of-the-art performance with a mean absolute error (MAE) of 1.95 and an intraclass correlation coefficient (ICC) of 0.43. While still not as good as trained human observer predictions provided with the dataset, when we average our estimates with those human estimates, our model improves their MAE from 1.76 to 1.58. Trained on the UNBC-McMaster dataset and applied directly with no further training or fine-tuning on a separate dataset of facial videos recorded during post-appendectomy physical exams, our model also outperforms previous work by 6% on the Area under the ROC curve metric (AUC). | ['Virginia R. de Sa', 'Jeannie S. Huang', 'Xiaojing Xu'] | 2019-12-13 | null | null | null | null | ['pain-intensity-regression'] | ['medical'] | [ 2.37732932e-01 -4.18922342e-02 -6.73547924e-01 -3.38893592e-01
-1.45578337e+00 -2.43503079e-01 -4.05877046e-02 5.28032556e-02
-7.70139933e-01 5.93149722e-01 2.14945197e-01 5.57446182e-02
-6.17006980e-02 -3.80600899e-01 -5.63246429e-01 -5.26248157e-01
-3.14853400e-01 9.74913910e-02 -1.44463912e-01 1.03083238e-01
-2.73950002e-03 3.07888865e-01 -1.14456177e+00 7.24152625e-01
3.37557882e-01 1.66140485e+00 -1.30455196e-01 9.02190089e-01
3.36962044e-01 8.35751891e-01 -5.41275680e-01 -2.38734052e-01
1.38261229e-01 -3.89207929e-01 -8.27311635e-01 2.06047744e-02
4.36793029e-01 -6.29645228e-01 -1.86604392e-02 6.76569223e-01
5.50971091e-01 -1.52618796e-01 4.79989588e-01 -1.04435265e+00
-1.26314625e-01 -1.59087792e-01 -9.35931265e-01 1.93901435e-01
3.61435682e-01 5.51593788e-02 8.41215491e-01 -6.52519226e-01
7.31939554e-01 1.08420765e+00 6.96367204e-01 3.98496807e-01
-1.26414716e+00 -8.71194899e-01 -3.05661827e-01 1.49136722e-01
-1.40443504e+00 1.05701849e-01 6.34841919e-01 -2.91670263e-01
8.48771811e-01 9.00409669e-02 7.96479046e-01 1.15118647e+00
9.02621210e-01 6.27202749e-01 1.52019298e+00 -1.05400302e-01
1.77132502e-01 -3.27236056e-01 -4.78014320e-01 1.00222492e+00
-4.64171112e-01 1.07971147e-01 -6.52353585e-01 -2.72768766e-01
1.01317108e+00 -8.43593106e-02 -2.93684732e-02 -1.13478601e-01
-9.25205410e-01 1.11649144e+00 6.04718268e-01 1.72856867e-01
-6.93250418e-01 6.25151277e-01 5.91166973e-01 3.56214911e-01
4.95744944e-01 2.52221018e-01 -4.64005232e-01 -4.16557819e-01
-1.08043349e+00 1.93251237e-01 5.54776430e-01 1.49376407e-01
6.27747357e-01 -3.59022170e-01 -1.74619690e-01 1.07336819e+00
-1.73168540e-01 2.48146906e-01 7.07249522e-01 -1.48272288e+00
-5.31457886e-02 2.00065151e-01 -7.85755143e-02 -9.97186601e-01
-8.35919678e-01 -3.99484724e-01 -5.13905466e-01 6.32716477e-01
3.71369153e-01 -4.67497200e-01 -1.05235898e+00 1.89647961e+00
6.90804273e-02 1.47724420e-01 -2.69191563e-01 1.20704687e+00
6.56411111e-01 4.23031598e-01 4.27048296e-01 -4.47778195e-01
1.56346035e+00 -8.89432847e-01 -4.90550816e-01 -1.42776772e-01
9.66909647e-01 -6.88150406e-01 9.68788803e-01 9.25815403e-01
-1.01883936e+00 -4.78662044e-01 -8.33724916e-01 1.59043700e-01
4.74122912e-02 2.64999002e-01 5.87368727e-01 4.40695435e-01
-1.06976414e+00 7.41001308e-01 -8.59470010e-01 -1.73253283e-01
6.17917120e-01 6.07944250e-01 -7.46884644e-01 -2.04034932e-02
-9.44356918e-01 1.02636755e+00 1.15321763e-01 -8.95349383e-02
-8.03310752e-01 -5.79707086e-01 -6.49874687e-01 -9.53525454e-02
5.47517955e-01 -8.05840373e-01 1.38625491e+00 -1.42913568e+00
-1.56517458e+00 1.03374791e+00 -3.70050259e-02 -4.62866962e-01
3.71064395e-01 -2.07216054e-01 -1.50985122e-01 6.95097864e-01
-7.25524724e-02 1.23456085e+00 8.91933918e-01 -8.75961542e-01
-5.24259269e-01 -3.06045771e-01 -5.24665937e-02 4.29392993e-01
-9.51427817e-02 3.11845988e-02 -5.88817596e-01 -6.07317686e-01
-2.78769076e-01 -1.27981722e+00 -2.62155056e-01 5.80159247e-01
1.46262810e-01 -1.65692553e-01 5.67238688e-01 -8.48676264e-01
1.04026854e+00 -2.11487389e+00 2.12612540e-01 2.39484221e-01
3.16105545e-01 2.93062869e-02 -3.96248370e-01 -1.08012676e-01
-2.59420305e-01 -9.88350511e-02 -1.59743633e-02 -3.22585076e-01
-6.86995387e-01 2.67332643e-01 4.07047838e-01 4.56999540e-01
3.38263124e-01 9.30228710e-01 -7.54191458e-01 -6.97532296e-01
2.53266841e-01 3.79638344e-01 -6.59639180e-01 5.10474220e-02
4.66259867e-02 4.30510223e-01 -2.14941934e-01 6.57923818e-01
3.45292032e-01 -2.17337385e-01 -5.63407466e-02 -1.86025903e-01
4.16691661e-01 -3.00141454e-01 -6.63637161e-01 2.39136314e+00
-6.74796700e-01 3.55640650e-01 2.14244306e-01 -8.79319072e-01
7.97789633e-01 5.07945240e-01 1.10517812e+00 -7.74452806e-01
3.83021057e-01 1.94704324e-01 1.38792634e-01 -4.61301774e-01
-1.28562823e-01 -7.22080052e-01 -2.89981097e-01 2.33574077e-01
3.21877748e-01 -1.41972870e-01 -1.36932656e-01 1.99207328e-02
1.42584383e+00 1.48736136e-02 3.95494372e-01 5.83899133e-02
2.55423099e-01 -8.06250572e-02 3.03378344e-01 3.88975561e-01
-4.11279291e-01 7.52157450e-01 9.60688710e-01 -4.13565516e-01
-6.27232969e-01 -1.10976756e+00 -1.35150596e-01 9.37967241e-01
-1.81044564e-01 -4.04393643e-01 -6.01209700e-01 -7.59710371e-01
1.72510535e-01 2.14092836e-01 -1.11214161e+00 -2.54709780e-01
-2.47370228e-01 -4.99649227e-01 5.11142910e-01 5.59147716e-01
2.52125002e-02 -1.14707661e+00 -8.48594308e-01 2.70892054e-01
-2.46352315e-01 -1.01959407e+00 -3.18333000e-01 4.01001304e-01
-7.40701497e-01 -9.49041784e-01 -1.02501965e+00 -4.20573860e-01
1.11287586e-01 -1.59851938e-01 7.51429737e-01 -7.11811334e-02
-5.81701875e-01 3.96166921e-01 -1.81237385e-01 -4.38736796e-01
1.52125116e-03 -1.74482405e-01 -1.19562313e-01 -2.41505969e-02
8.76401588e-02 -5.02022028e-01 -8.94263625e-01 1.59366131e-01
-8.85198653e-01 1.08358987e-01 9.07841146e-01 9.57760513e-01
8.83897543e-01 -4.91294891e-01 4.47642446e-01 -5.26414454e-01
7.05931187e-01 -4.05171722e-01 -1.83442444e-01 -1.05573818e-01
-4.53824162e-01 -2.64761060e-01 2.67181814e-01 -6.77245080e-01
-6.80290818e-01 2.75419950e-01 -2.25540861e-01 -1.15256011e+00
-7.09072948e-02 7.81110525e-01 8.93890321e-01 -2.63533294e-01
6.40331268e-01 -3.69507700e-01 4.45869952e-01 -4.50544395e-02
2.28460357e-01 6.02687776e-01 7.91756511e-01 -4.17753637e-01
-4.78677861e-02 3.72658908e-01 3.92267555e-01 -5.21543443e-01
-1.04981816e+00 -4.86014903e-01 -4.49089229e-01 -4.79435384e-01
1.26313853e+00 -1.14807141e+00 -8.99942577e-01 2.04263598e-01
-8.09264004e-01 -5.90620577e-01 -6.52731434e-02 7.45386541e-01
-1.04117322e+00 1.42475545e-01 -1.00897336e+00 -6.61757588e-01
-5.51318944e-01 -1.20046031e+00 1.44397199e+00 -7.97828063e-02
-7.50791550e-01 -7.21581519e-01 2.39483461e-01 4.31555957e-01
1.91783264e-01 8.41526270e-01 8.37898433e-01 -1.35300010e-01
1.24380320e-01 -3.87990445e-01 -2.97939330e-01 4.47509199e-01
1.67663410e-01 -2.32550815e-01 -8.82201314e-01 -1.47913650e-01
5.65220080e-02 -9.85196054e-01 7.21574187e-01 8.31880569e-01
1.67851949e+00 -3.62696797e-02 -8.01660419e-02 5.70365667e-01
1.56770873e+00 2.09401727e-01 5.86458147e-01 2.48102129e-01
1.88788727e-01 3.79433423e-01 7.03294098e-01 4.71190065e-01
1.47816716e-02 9.06590939e-01 7.04573035e-01 -3.65434945e-01
-9.33553576e-02 9.15624276e-02 1.89315736e-01 2.91237801e-01
-2.46264175e-01 2.16758892e-01 -7.35183358e-01 2.06357673e-01
-1.73632514e+00 -6.81793869e-01 2.53363878e-01 1.90837729e+00
6.76530778e-01 9.65064913e-02 4.55493212e-01 -6.85553551e-02
2.49908254e-01 2.40831319e-02 -4.53974396e-01 -9.39902008e-01
2.41344869e-01 8.22481096e-01 3.83894295e-01 4.01003212e-01
-1.14262724e+00 6.48610950e-01 6.05607462e+00 9.47971284e-01
-1.45250058e+00 2.67439336e-01 8.64889085e-01 -7.47699916e-01
3.33443552e-01 -3.92259568e-01 7.56056309e-02 4.38145220e-01
1.09052265e+00 1.55743420e-01 4.73973364e-01 9.09210443e-01
2.29726374e-01 -5.80254734e-01 -7.77136028e-01 1.27434504e+00
1.09932050e-01 -1.10496569e+00 -3.64578009e-01 2.61477768e-01
4.09340441e-01 -8.49143416e-03 8.09348524e-02 3.89530897e-01
-1.08646765e-01 -1.27624905e+00 4.18488950e-01 5.25465786e-01
1.16550779e+00 -8.59688640e-01 8.72541785e-01 1.64284751e-01
-7.69309163e-01 -1.20972693e-01 1.01715727e-02 -4.99311835e-02
1.99369341e-01 3.25322121e-01 -6.25165582e-01 2.78238118e-01
7.31679261e-01 4.98433083e-01 -2.12991551e-01 8.27287078e-01
-6.99844733e-02 2.80319244e-01 -2.94205457e-01 1.11331090e-01
6.29950464e-01 1.98545933e-01 8.54057819e-03 1.17175770e+00
3.22494626e-01 2.17127323e-01 2.26464555e-01 3.53412211e-01
-8.49229842e-02 2.29959950e-01 -2.61188626e-01 1.92561403e-01
-3.58364850e-01 1.74666524e+00 -5.79160750e-01 -2.21096322e-01
-5.34102619e-01 1.13043821e+00 4.97386366e-01 6.63302466e-02
-9.52582955e-01 -6.41718656e-02 6.05659842e-01 -3.14912852e-03
1.81980759e-01 6.09630905e-02 -3.20042968e-01 -7.30934203e-01
-2.10018784e-01 -8.51561904e-01 7.51233220e-01 -1.15091252e+00
-1.07557583e+00 4.19059992e-01 -5.03785908e-02 -1.18323147e+00
-4.04810816e-01 -7.33115971e-01 -5.05559266e-01 7.84293234e-01
-1.20510566e+00 -9.71475542e-01 -3.40247124e-01 6.69508457e-01
3.95337850e-01 3.54853839e-01 1.12655282e+00 9.42593217e-02
-2.80415297e-01 5.14426410e-01 -5.78813493e-01 1.46470979e-01
9.33411956e-01 -9.79806304e-01 -5.56197047e-01 9.01779607e-02
-7.79604772e-03 5.47678769e-02 4.10296738e-01 -5.60032070e-01
-1.09705186e+00 -8.00076425e-01 3.76259685e-01 -1.95069127e-02
7.44056344e-01 3.55460756e-02 -7.32573986e-01 4.56529498e-01
-3.07484455e-02 1.85358033e-01 7.81270385e-01 1.36253193e-01
-2.56407976e-01 -1.49618208e-01 -1.21298504e+00 3.25410664e-01
5.47867060e-01 -4.80398744e-01 -2.56956041e-01 4.38546717e-01
2.38918364e-01 -7.09319174e-01 -1.35299969e+00 6.19617403e-01
1.10355508e+00 -8.84012282e-01 7.17423975e-01 -5.34019887e-01
8.68294775e-01 1.80631891e-01 -1.14181206e-01 -1.19556749e+00
-3.04738611e-01 -2.02221930e-01 -2.26326790e-02 2.31613383e-01
3.55407059e-01 -1.06920423e-02 1.03383207e+00 5.26539564e-01
2.19544023e-02 -1.61036718e+00 -1.31806672e+00 -5.08262038e-01
1.47128433e-01 -6.85500979e-01 -3.75214458e-01 6.17981970e-01
2.09477678e-01 3.27938080e-01 -6.39348745e-01 -3.13286006e-01
1.05907626e-01 3.19171846e-02 6.74604058e-01 -8.35812390e-01
-5.34377038e-01 -3.70172799e-01 -5.77445984e-01 -5.31665385e-01
1.84988081e-02 -7.77669907e-01 -1.10876329e-01 -1.41780853e+00
4.45491284e-01 -3.74035574e-02 -6.78197920e-01 8.38411570e-01
1.10618863e-02 5.56746244e-01 2.80081570e-01 4.53969501e-02
-4.52762365e-01 1.75140873e-01 1.35318828e+00 2.20016390e-01
-2.81894892e-01 -2.05011010e-01 -4.57046688e-01 6.85885489e-01
6.51155531e-01 -7.12858438e-01 -1.80757314e-01 -1.46611497e-01
-9.17306393e-02 8.23825598e-01 4.27169144e-01 -1.17928934e+00
-2.22382799e-01 -1.04559742e-01 8.40918481e-01 -2.16956556e-01
9.26047027e-01 -5.99461377e-01 3.23247910e-02 9.91121233e-01
-2.21078485e-01 -9.88205522e-03 3.80264312e-01 5.10636449e-01
-2.53542483e-01 1.01054110e-01 9.26054060e-01 -3.06107283e-01
-5.86984992e-01 3.54423732e-01 -5.88447273e-01 -3.55524302e-01
1.11801863e+00 2.38821190e-02 -4.84693563e-03 -8.69245708e-01
-1.20022035e+00 7.23339245e-02 2.32664198e-01 4.15824503e-01
5.43646991e-01 -1.34859312e+00 -5.02971172e-01 -1.29033506e-01
6.70697168e-02 -4.57957268e-01 3.24629813e-01 1.38815415e+00
-4.29476440e-01 3.34979504e-01 -4.64060754e-01 -9.27960455e-01
-1.45191884e+00 5.07205427e-01 3.73972654e-01 -6.27198040e-01
-3.07943881e-01 7.82603204e-01 1.54496670e-01 -4.46592225e-03
7.80321360e-02 -3.69940579e-01 1.76937371e-01 1.39179662e-01
2.50499070e-01 -2.03918964e-02 1.34469420e-01 -6.24863803e-01
-4.02000666e-01 7.62077808e-01 6.32635951e-02 -3.75656694e-01
1.18797958e+00 3.70403171e-01 1.50479943e-01 3.13257724e-01
1.67962086e+00 -3.74768883e-01 -1.13712132e+00 1.14417270e-01
-3.16607177e-01 -2.73384005e-01 3.73889148e-01 -1.08154106e+00
-1.05004251e+00 8.64643693e-01 1.05256116e+00 -4.18675959e-01
1.56874156e+00 7.30485618e-02 8.97212923e-01 8.49964172e-02
3.56618494e-01 -1.06812787e+00 6.15170479e-01 2.59468760e-02
9.73465562e-01 -1.11654043e+00 1.13023177e-01 -2.96966732e-01
-1.06861639e+00 1.15324259e+00 4.83544886e-01 -4.29191381e-01
6.47312105e-01 3.48359317e-01 3.69265854e-01 -2.42965698e-01
-8.75777841e-01 -1.18844934e-01 3.82840365e-01 1.79225087e-01
4.27240878e-01 1.02726616e-01 -6.25410855e-01 5.90059996e-01
-1.12728015e-01 7.40218103e-01 1.55436054e-01 9.23013926e-01
-4.43494767e-01 -9.15541768e-01 -1.00377940e-01 8.75333965e-01
-9.36491251e-01 3.42952549e-01 -1.57322094e-01 8.41991067e-01
2.13160306e-01 6.80014312e-01 1.47085547e-01 -5.65191448e-01
1.98455349e-01 1.07702978e-01 6.24782503e-01 -5.39769948e-01
-5.62869072e-01 4.52402860e-01 1.74489558e-01 -1.14998996e+00
-5.86090326e-01 -4.23658073e-01 -1.20597935e+00 -4.55164380e-04
3.40480208e-02 -1.96839184e-01 7.52810776e-01 8.80536377e-01
1.30564809e-01 2.92184651e-01 4.83059794e-01 -8.80467772e-01
-1.63655296e-01 -8.65239501e-01 -6.77533865e-01 5.48425794e-01
3.28019917e-01 -7.91279733e-01 -9.08296332e-02 -2.93637365e-01] | [13.599246978759766, 1.998848795890808] |
ad000ef1-f6a0-4121-b775-a07e08be7dd0 | an-entire-renal-anatomy-extraction-network | 2305.13616 | null | https://arxiv.org/abs/2305.13616v1 | https://arxiv.org/pdf/2305.13616v1.pdf | An Entire Renal Anatomy Extraction Network for Advanced CAD During Partial Nephrectomy | Partial nephrectomy (PN) is common surgery in urology. Digitization of renal anatomies brings much help to many computer-aided diagnosis (CAD) techniques during PN. However, the manual delineation of kidney vascular system and tumor on each slice is time consuming, error-prone, and inconsistent. Therefore, we proposed an entire renal anatomies extraction method from Computed Tomographic Angiographic (CTA) images fully based on deep learning. We adopted a coarse-to-fine workflow to extract target tissues: first, we roughly located the kidney region, and then cropped the kidney region for more detail extraction. The network we used in our workflow is based on 3D U-Net. To dealing with the imbalance of class contributions to loss, we combined the dice loss with focal loss, and added an extra weight to prevent excessive attention. We also improved the manual annotations of vessels by merging semi-trained model's prediction and original annotations under supervision. We performed several experiments to find the best-fitting combination of variables for training. We trained and evaluated the models on our 60 cases dataset with 3 different sources. The average dice score coefficient (DSC) of kidney, tumor, cyst, artery, and vein, were 90.9%, 90.0%, 89.2%, 80.1% and 82.2% respectively. Our modulate weight and hybrid strategy of loss function increased the average DSC of all tissues about 8-20%. Our optimization of vessel annotation improved the average DSC about 1-5%. We proved the efficiency of our network on renal anatomies segmentation. The high accuracy and fully automation make it possible to quickly digitize the personal renal anatomies, which greatly increases the feasibility and practicability of CAD application on urology surgery. | ['Dongkai Zhou', 'Ying Yang', 'Nan Ma'] | 2023-05-23 | null | null | null | null | ['anatomy'] | ['miscellaneous'] | [-1.90484628e-01 3.62119615e-01 -1.23747177e-01 -4.94820863e-01
-3.40094894e-01 -4.87183899e-01 4.74907234e-02 1.06720515e-01
-6.21492147e-01 8.38418722e-01 -1.26312494e-01 -3.81161153e-01
-1.34844720e-01 -9.59721565e-01 -4.58288938e-01 -6.57864034e-01
-1.40542328e-01 5.60282588e-01 1.89140383e-02 2.35746756e-01
2.99497455e-01 1.01612163e+00 -7.43347526e-01 -2.02656403e-01
1.41346574e+00 7.90325940e-01 3.48472893e-01 3.28757435e-01
-6.71531558e-01 8.23433459e-01 -2.28024617e-01 -6.48663938e-01
6.54290795e-01 -4.31195915e-01 -6.73796177e-01 -1.51205420e-01
4.17319030e-01 -4.07802463e-01 -4.24117029e-01 1.25013912e+00
6.00618482e-01 -3.21066827e-01 7.10606813e-01 -7.89204240e-01
-5.58923542e-01 9.67018366e-01 -9.17974055e-01 1.75015837e-01
-5.11008561e-01 2.25041136e-01 4.20504510e-01 -3.19019496e-01
8.19084704e-01 9.32525396e-01 8.05511415e-01 5.59445083e-01
-1.00158679e+00 -9.58308101e-01 -3.20948839e-01 -1.13745131e-01
-1.24169993e+00 3.29193771e-02 4.16450411e-01 -6.22894645e-01
1.57146111e-01 7.14728534e-02 9.12820041e-01 1.91549122e-01
4.97439921e-01 6.34543300e-01 1.03818023e+00 -2.66204715e-01
-2.88963526e-01 2.17799038e-01 1.43432856e-01 1.02346504e+00
4.68196094e-01 1.23952366e-02 5.36439419e-01 1.87165603e-01
1.49819589e+00 3.22189212e-01 -7.46268257e-02 -4.57043111e-01
-9.60017383e-01 6.08751237e-01 8.28965664e-01 3.39412332e-01
-3.88109267e-01 1.13411024e-01 5.68939149e-01 4.91680056e-02
-8.39347765e-02 3.84608895e-01 -2.39051312e-01 1.65254220e-01
-6.41807854e-01 -1.11605935e-01 6.20229483e-01 1.02238750e+00
4.95346993e-01 -2.08981305e-01 -2.29456618e-01 9.57080007e-01
7.64083043e-02 3.49925578e-01 5.49593091e-01 -9.45990980e-01
2.68806070e-01 8.93607855e-01 -4.03931856e-01 -6.12874568e-01
-5.81313848e-01 -7.67872214e-01 -1.29665065e+00 2.71206141e-01
6.93381846e-01 -2.58342594e-01 -1.43392658e+00 1.32641435e+00
2.58456916e-01 6.70811161e-02 -3.80939215e-01 1.04348147e+00
8.05709064e-01 4.46735546e-02 4.69764590e-01 -1.23852998e-01
1.31026101e+00 -8.44480217e-01 -7.07303703e-01 2.82710671e-01
8.44192684e-01 -7.71447062e-01 6.61015451e-01 -2.24482808e-02
-1.05839396e+00 -3.60269547e-01 -9.17576551e-01 -4.79553491e-02
6.94043711e-02 5.62739551e-01 7.93676972e-01 6.97286189e-01
-5.88973045e-01 1.07469642e+00 -1.09554529e+00 -2.56855935e-02
9.71961856e-01 5.31141579e-01 -6.08529985e-01 1.65054530e-01
-8.21948528e-01 1.22053969e+00 2.09005982e-01 2.82267213e-01
-5.31434298e-01 -1.01429367e+00 -7.11916089e-01 1.23789184e-01
1.57356098e-01 -8.25930476e-01 9.83395040e-01 -7.05389321e-01
-1.43455970e+00 1.08052742e+00 1.79961279e-01 -4.22083765e-01
9.44651186e-01 5.56541197e-02 -1.04760530e-03 7.72315636e-02
2.92423810e-03 3.43063474e-01 5.37495017e-02 -1.01464832e+00
-5.53947866e-01 -6.63372099e-01 -3.29296261e-01 1.59872681e-01
1.32380441e-01 -2.54360110e-01 -4.75522816e-01 -3.74860704e-01
5.78116596e-01 -8.53714406e-01 -7.75393963e-01 5.57557404e-01
-4.36248451e-01 2.91431695e-01 2.91256666e-01 -9.92242336e-01
1.10255349e+00 -2.04663372e+00 -1.27962917e-01 5.16976833e-01
4.96914744e-01 3.10706317e-01 4.71935838e-01 -2.64886260e-01
-3.46050300e-02 2.47980326e-01 -3.40558618e-01 -9.08423364e-02
-4.57398832e-01 2.49051824e-01 3.13496947e-01 5.37367940e-01
-2.12211251e-01 7.83559442e-01 -6.94444418e-01 -1.15708327e+00
4.23205912e-01 3.55942637e-01 -3.99946243e-01 4.72219177e-02
4.67891634e-01 4.21754897e-01 -5.88694990e-01 7.51586914e-01
1.06517351e+00 -2.01196253e-01 2.11160585e-01 -4.23171133e-01
-2.67995037e-02 -2.31162712e-01 -1.17171109e+00 1.85647357e+00
-5.70043206e-01 2.58570045e-01 1.97354451e-01 -7.36740053e-01
1.17189741e+00 1.18315667e-01 7.48990715e-01 -7.38800228e-01
2.96160191e-01 5.08803606e-01 3.39124292e-01 -7.99099147e-01
-6.34292886e-02 -4.57003713e-01 2.57855058e-01 2.85046026e-02
1.62223384e-01 -5.18724285e-02 1.19290955e-01 1.71402276e-01
8.15360129e-01 5.06659448e-02 4.94198173e-01 -3.74424696e-01
5.97994685e-01 1.16037510e-01 7.19109118e-01 4.31538999e-01
-5.94811380e-01 7.19098210e-01 6.38814509e-01 -5.65298140e-01
-1.19914067e+00 -1.13836384e+00 -7.57762372e-01 8.17260444e-02
4.80563462e-01 9.78239775e-02 -5.05816400e-01 -7.88292408e-01
4.34740454e-01 2.18547970e-01 -6.00396752e-01 1.52155235e-01
-8.39167774e-01 -8.31304550e-01 7.10247338e-01 6.05327606e-01
5.34303129e-01 -9.14451361e-01 -2.65715837e-01 1.05308704e-01
1.00829698e-01 -6.38241947e-01 -1.59031853e-01 6.15114719e-02
-1.35929883e+00 -1.31554604e+00 -8.64478946e-01 -8.02996159e-01
1.05502784e+00 -1.93608209e-01 7.47357428e-01 1.96065843e-01
-8.60797048e-01 -4.92989749e-01 2.40527660e-01 -2.02541694e-01
-3.01368237e-01 1.57842249e-01 -4.02329355e-01 -5.14763415e-01
-2.83097960e-02 -5.06783903e-01 -7.78593540e-01 2.34133676e-01
-4.32437152e-01 -9.01658554e-03 9.85379338e-01 7.46914804e-01
6.67477667e-01 -2.30090559e-01 3.80656660e-01 -1.32050502e+00
3.48976463e-01 -2.87053496e-01 -7.26629496e-01 1.62757441e-01
-7.47128606e-01 9.28842798e-02 7.05682397e-01 -2.01150969e-01
-1.27004337e+00 3.52587104e-01 -4.99095395e-02 -5.29357851e-01
-4.00804952e-02 2.69987196e-01 2.24233642e-01 -2.88794607e-01
6.05407715e-01 -1.22984767e-01 4.16694254e-01 -3.40815485e-01
3.57611507e-01 4.18298632e-01 6.54612482e-01 -4.38351393e-01
6.70672238e-01 3.47220927e-01 3.77955437e-01 -3.37403327e-01
-3.17930967e-01 -3.09301466e-01 -7.58245647e-01 -1.17776752e-01
8.84097457e-01 -7.97449350e-01 -8.94628882e-01 1.57055452e-01
-8.01886380e-01 -7.71818385e-02 -2.80503482e-01 8.59568357e-01
-1.20672196e-01 6.24806106e-01 -1.11332357e+00 -5.32460809e-01
-7.33390868e-01 -1.16477430e+00 3.39861691e-01 7.57973075e-01
8.95984173e-02 -8.95766973e-01 -5.42185903e-02 1.94756359e-01
6.68718159e-01 3.40718865e-01 9.47201073e-01 -4.47475493e-01
-6.90437376e-01 -2.98026741e-01 -6.58285022e-01 3.00116479e-01
9.15665403e-02 2.45409340e-01 -7.88840890e-01 2.52612889e-01
-2.23431125e-01 7.42529407e-02 9.38466251e-01 7.30329096e-01
1.48504448e+00 1.68826565e-01 -5.40911436e-01 1.09100270e+00
1.82483470e+00 4.28197086e-01 7.12308586e-01 3.54292154e-01
7.04636574e-01 3.40591252e-01 5.23697734e-01 2.08376899e-01
9.41024423e-02 6.46652430e-02 5.00719726e-01 -3.90409917e-01
-1.50661722e-01 -9.19963568e-02 -4.41560298e-01 6.21093810e-01
-4.82363522e-01 4.05179739e-01 -8.73807251e-01 5.18475175e-01
-1.57665706e+00 -8.30114782e-01 -4.06289160e-01 2.13666344e+00
9.47248399e-01 1.89670354e-01 -2.42145479e-01 -5.17039478e-01
8.36522460e-01 -4.35921490e-01 -5.05548120e-01 -1.04576491e-01
1.99532241e-01 3.00139576e-01 1.15821910e+00 5.00050783e-01
-1.11129510e+00 5.92118919e-01 5.37545776e+00 7.02976048e-01
-1.23435152e+00 -2.18982145e-01 7.90859461e-01 -4.57672887e-02
-5.33353537e-02 1.07493542e-01 -7.62513220e-01 4.95869637e-01
3.17098260e-01 -3.82546455e-01 2.69521438e-02 9.13189411e-01
6.26655146e-02 -1.96029499e-01 -7.54932940e-01 8.25258911e-01
-4.27231252e-01 -1.45525753e+00 -2.21399084e-01 1.12693138e-01
5.36587298e-01 2.08223630e-02 -3.96924615e-01 3.38547498e-01
1.92410171e-01 -1.15934598e+00 -1.22406222e-01 7.05938101e-01
7.50698924e-01 -3.92910838e-01 1.28031600e+00 2.62211949e-01
-1.00781310e+00 9.23755392e-02 -4.97039676e-01 2.30479196e-01
1.73332363e-01 6.95881307e-01 -1.04934382e+00 6.34904504e-01
4.43322212e-01 5.05150795e-01 -3.06597650e-01 1.42776048e+00
-4.03510407e-02 1.43133208e-01 -5.43761551e-01 4.93676588e-02
6.91296756e-02 -5.59379578e-01 2.83667762e-02 1.11070502e+00
2.63730943e-01 2.16016084e-01 -5.35495430e-02 1.08022344e+00
-2.73627520e-01 5.33696353e-01 -3.39773864e-01 4.60571557e-01
5.94722390e-01 1.63337100e+00 -8.07475865e-01 -3.71465594e-01
-1.75287038e-01 4.10239697e-01 2.46214494e-01 -1.01181082e-01
-8.85890007e-01 -5.97977281e-01 1.41805127e-01 8.12331289e-02
-1.87320158e-01 2.60561079e-01 -9.10365760e-01 -1.08111346e+00
8.60011280e-02 -8.96855369e-02 4.47106600e-01 -2.17296615e-01
-1.26764798e+00 3.83409679e-01 -3.09962064e-01 -1.25249982e+00
3.22750628e-01 -5.09168148e-01 -8.13299537e-01 1.24899304e+00
-1.36741805e+00 -8.24804544e-01 -7.39196718e-01 1.45365164e-01
1.67179450e-01 7.01023564e-02 6.04916215e-01 7.36074507e-01
-7.59337783e-01 5.31773746e-01 7.01310188e-02 5.25651157e-01
8.90531421e-01 -1.46856344e+00 -2.14077190e-01 4.80005413e-01
-7.19069183e-01 8.26143384e-01 2.42075443e-01 -6.63416624e-01
-9.88424778e-01 -9.69092488e-01 7.40752399e-01 -5.12973405e-02
3.36551815e-01 3.09001893e-01 -8.30321908e-01 8.81648540e-01
-3.61705571e-02 2.57382125e-01 6.04757428e-01 -2.05916576e-02
1.74863994e-01 -2.70350695e-01 -1.49198115e+00 6.08772218e-01
8.02585185e-01 5.92691787e-02 -4.35735434e-01 1.19985677e-01
3.23755860e-01 -8.86690676e-01 -1.40072167e+00 5.54577947e-01
9.17110622e-01 -7.56740808e-01 9.79058266e-01 -4.48075563e-01
7.91291714e-01 -2.53541559e-01 3.10464799e-01 -9.46917117e-01
-3.70150864e-01 -6.89329356e-02 3.63805175e-01 1.00852203e+00
5.05749524e-01 -5.27286291e-01 1.05992234e+00 8.46877277e-01
-3.83087337e-01 -9.78200376e-01 -6.25177562e-01 -2.67768741e-01
3.16443592e-01 1.87720954e-01 2.84734905e-01 1.20791638e+00
5.93826771e-02 -8.05983618e-02 -1.47534862e-01 4.16348688e-02
9.13579583e-01 1.63963810e-01 7.34203637e-01 -1.31316853e+00
1.99857149e-02 -7.23963559e-01 -5.32776296e-01 -6.56337619e-01
-3.67408425e-01 -1.29073095e+00 -4.50231373e-01 -1.76904404e+00
3.84595484e-01 -8.68848860e-01 -3.06435555e-01 4.78136361e-01
-1.44851923e-01 -1.35974541e-01 1.44174367e-01 3.28264505e-01
1.82543725e-01 2.85565853e-01 2.06025362e+00 9.48491246e-02
-3.24092895e-01 2.36393362e-01 -5.92689216e-01 6.51767075e-01
8.49258840e-01 -4.44280773e-01 -1.39310926e-01 -1.80306882e-01
-2.79473841e-01 3.03677469e-01 9.35301781e-02 -6.65009081e-01
4.77662534e-01 -4.71624546e-02 8.84885311e-01 -5.47079146e-01
-1.89360213e-02 -9.44999039e-01 1.86148942e-01 1.05102932e+00
-2.07416177e-01 -6.09045804e-01 1.37060598e-01 7.39675537e-02
-1.63694695e-01 -3.93586099e-01 1.10312617e+00 -6.47497535e-01
-4.10520941e-01 4.54752922e-01 2.06100374e-01 -2.84095585e-01
1.27350461e+00 -2.74673164e-01 -2.24520385e-01 1.01973899e-01
-1.13621497e+00 5.74798763e-01 2.58141309e-01 -2.44298413e-01
3.68553430e-01 -1.04452169e+00 -6.31748557e-01 2.44767576e-01
-3.14343899e-01 4.83338028e-01 6.41201198e-01 1.25849187e+00
-1.41199076e+00 3.42477202e-01 -6.24390066e-01 -7.34238088e-01
-1.12055278e+00 2.04165801e-01 8.30343723e-01 -4.42879140e-01
-9.36029553e-01 7.13543832e-01 1.20706566e-01 -6.99406207e-01
2.87693799e-01 -5.71266353e-01 -2.38895357e-01 -2.28366613e-01
7.14605954e-03 4.72638041e-01 -1.75391451e-01 -1.13255866e-01
-2.69858748e-01 7.81971693e-01 -3.49641025e-01 5.09548008e-01
1.13270855e+00 -9.37419664e-03 -1.59126893e-01 -1.63756520e-01
1.08857262e+00 -5.00690192e-02 -9.66015041e-01 -1.64375171e-01
-2.98264712e-01 -5.51847517e-01 2.13477351e-02 -8.95816565e-01
-1.53647041e+00 7.75572896e-01 7.33378291e-01 -1.56228632e-01
8.97179961e-01 -2.83160180e-01 7.13980734e-01 -2.60276627e-03
2.96648771e-01 -7.43878007e-01 -7.47233808e-01 1.62909910e-01
3.26144785e-01 -1.34328985e+00 3.76885951e-01 -7.52713144e-01
-6.95805967e-01 1.50530636e+00 9.21627939e-01 -4.94108140e-01
7.88725674e-01 4.03436452e-01 4.47174579e-01 -1.87344819e-01
-1.71438396e-01 2.16523856e-01 2.05544792e-02 9.81192812e-02
6.72357619e-01 3.03990722e-01 -7.60917008e-01 5.84567845e-01
-2.65537024e-01 3.02933156e-01 5.60401976e-01 8.68590772e-01
-5.53703964e-01 -8.38343263e-01 -2.11122949e-02 7.38852918e-01
-7.21458077e-01 1.80315226e-01 1.73256233e-01 1.17145264e+00
1.82629511e-01 -1.59762911e-02 -6.09449036e-02 8.07626322e-02
3.75586838e-01 -1.79140553e-01 5.38129628e-01 -2.46639490e-01
-7.26170540e-01 2.71224529e-01 -1.55227572e-01 -4.58958000e-01
-5.04929796e-02 -4.18603957e-01 -1.75481224e+00 -4.15700376e-01
-5.34269512e-01 8.39073285e-02 9.53716755e-01 6.66039348e-01
-2.57233027e-02 7.84404933e-01 5.31955242e-01 -2.79941976e-01
-7.03841746e-01 -9.87203181e-01 -8.17929208e-01 3.70265990e-01
8.79689083e-02 -5.10098100e-01 -3.49174887e-01 -5.09570679e-03] | [14.57791519165039, -2.6438276767730713] |
2b4879d7-d5a5-4273-8e3d-be31e430f00e | improving-compositional-generalization-in-2 | 2209.01352 | null | https://arxiv.org/abs/2209.01352v1 | https://arxiv.org/pdf/2209.01352v1.pdf | Improving Compositional Generalization in Math Word Problem Solving | Compositional generalization refers to a model's capability to generalize to newly composed input data based on the data components observed during training. It has triggered a series of compositional generalization analysis on different tasks as generalization is an important aspect of language and problem solving skills. However, the similar discussion on math word problems (MWPs) is limited. In this manuscript, we study compositional generalization in MWP solving. Specifically, we first introduce a data splitting method to create compositional splits from existing MWP datasets. Meanwhile, we synthesize data to isolate the effect of compositions. To improve the compositional generalization in MWP solving, we propose an iterative data augmentation method that includes diverse compositional variation into training data and could collaborate with MWP methods. During the evaluation, we examine a set of methods and find all of them encounter severe performance loss on the evaluated datasets. We also find our data augmentation method could significantly improve the compositional generalization of general MWP methods. Code is available at https://github.com/demoleiwang/CGMWP. | ['Ee-Peng Lim', 'Jing Jiang', 'Lei Wang', 'Yunshi Lan'] | 2022-09-03 | null | null | null | null | ['math-word-problem-solving', 'math-word-problem-solving', 'math-word-problem-solving'] | ['knowledge-base', 'reasoning', 'time-series'] | [ 3.15945834e-01 -4.14740741e-02 -3.45590591e-01 -4.52732056e-01
-4.04733509e-01 -7.56904364e-01 3.77942979e-01 1.04969226e-01
-1.62192822e-01 6.37959421e-01 2.26720616e-01 -4.98243153e-01
-2.59698629e-01 -8.80833030e-01 -6.66254818e-01 -4.48798299e-01
2.19188884e-01 4.18383658e-01 6.75483420e-02 -4.85057801e-01
4.82311100e-01 4.25827920e-01 -1.33317530e+00 7.70571113e-01
1.53877497e+00 7.33629286e-01 3.38351488e-01 3.34014535e-01
-4.73925471e-01 5.24727345e-01 -6.34299636e-01 -4.76450771e-01
3.58122706e-01 -3.85419339e-01 -1.16137397e+00 -2.18858346e-01
5.41359425e-01 8.41889307e-02 -1.70063794e-01 8.14963222e-01
4.38697428e-01 3.83079529e-01 5.10471284e-01 -1.56382251e+00
-1.14148903e+00 1.12360096e+00 -3.67457062e-01 3.52406293e-01
2.71898121e-01 2.13559777e-01 8.74327302e-01 -1.08447182e+00
4.15505439e-01 1.10084307e+00 6.97590888e-01 7.50383615e-01
-1.15049303e+00 -1.06522536e+00 5.61574459e-01 2.82399356e-01
-1.27618289e+00 -1.22523867e-01 9.67524230e-01 -2.97307521e-01
6.74115419e-01 4.45298761e-01 3.54272604e-01 1.23999548e+00
-1.80986047e-01 1.07432377e+00 1.23047876e+00 -3.71492386e-01
-1.03890695e-01 1.34509310e-01 5.50814986e-01 4.84362572e-01
5.02261110e-02 -2.21573159e-01 -6.63010359e-01 6.03022752e-04
3.59885842e-01 -8.48898888e-02 -4.55285162e-01 -2.90691927e-02
-1.10331702e+00 5.48406243e-01 3.83952141e-01 2.71897703e-01
8.59618634e-02 -1.58420578e-01 3.10902417e-01 4.10847753e-01
4.14558023e-01 9.41891611e-01 -7.11654305e-01 -1.02795668e-01
-7.92044520e-01 5.08579075e-01 5.70741832e-01 1.08418250e+00
6.48831308e-01 8.84418935e-02 -3.02442491e-01 1.00233328e+00
-1.77005172e-01 2.57843584e-01 8.48177969e-01 -7.89355159e-01
8.30139399e-01 1.10928845e+00 -5.62232196e-01 -7.86217988e-01
-4.80062246e-01 -6.99796677e-01 -8.64149570e-01 -3.12644571e-01
2.84260839e-01 -1.43988524e-02 -8.96176875e-01 1.90341723e+00
1.91748992e-01 9.48544741e-02 1.44045889e-01 7.08519101e-01
1.29620850e+00 7.63868809e-01 9.63214189e-02 3.65803391e-02
1.02178371e+00 -1.25854814e+00 -5.86497664e-01 -2.19059333e-01
8.33675265e-01 -6.11986637e-01 1.74807036e+00 5.64505219e-01
-1.03756011e+00 -7.60651052e-01 -1.01976609e+00 -2.26927549e-01
-5.54815710e-01 1.04416966e-01 5.67309797e-01 6.20634675e-01
-7.44420052e-01 6.77947581e-01 -5.07716417e-01 -2.20670685e-01
4.17606503e-01 2.88899720e-01 -2.10356548e-01 -1.61632046e-01
-1.30937243e+00 7.77787924e-01 7.12613046e-01 -1.61015660e-01
-6.20315075e-01 -1.30803359e+00 -7.65734613e-01 1.10926502e-03
4.82048154e-01 -5.95937669e-01 1.20230198e+00 -6.98530853e-01
-1.39596069e+00 5.87402701e-01 -1.97681829e-01 -2.13225916e-01
2.93931454e-01 -2.36872956e-01 -2.77978271e-01 -3.08072865e-01
-1.25613213e-01 6.93483949e-01 4.16858763e-01 -1.41662478e+00
-3.49415869e-01 -2.13190123e-01 8.32674354e-02 2.18529224e-01
-6.05772018e-01 -3.08865011e-01 -2.19496161e-01 -1.05430388e+00
2.63518631e-01 -7.78383374e-01 -1.45976841e-01 -5.61600864e-01
-3.43546867e-01 -3.91885489e-01 5.68181038e-01 -6.14772439e-01
1.61621583e+00 -2.11583400e+00 3.29102606e-01 1.96204051e-01
1.29933029e-01 3.91346723e-01 -5.42940617e-01 6.61461473e-01
-3.96312475e-01 4.80553031e-01 -4.32079852e-01 -3.77276152e-01
4.07432355e-02 2.85681486e-01 -7.58615375e-01 -7.53839388e-02
3.19873571e-01 1.11638510e+00 -7.45801926e-01 -1.82068959e-01
-2.56151378e-01 -1.10493645e-01 -7.73948610e-01 1.00590743e-01
-4.72216427e-01 5.67558885e-01 -2.82253981e-01 6.76828563e-01
9.00494158e-01 -1.39293358e-01 -1.25979248e-03 -1.21011220e-01
6.46468997e-02 4.44782346e-01 -1.07951367e+00 1.95707214e+00
-4.20876712e-01 3.05818170e-01 -5.02418220e-01 -1.26350236e+00
1.04422987e+00 1.90008618e-02 8.70940015e-02 -6.13702953e-01
5.87307103e-02 2.31667176e-01 2.83402413e-01 -5.62101364e-01
8.99109542e-01 -6.11561500e-02 5.44623062e-02 5.22767186e-01
8.77476707e-02 -2.01533094e-01 5.68125665e-01 2.62248784e-01
8.71697426e-01 1.64968982e-01 -8.52800310e-02 -5.06629467e-01
7.56865740e-01 1.25300854e-01 7.60289788e-01 6.33339882e-01
1.22215852e-01 5.55245996e-01 4.61302072e-01 -3.25957805e-01
-8.09909761e-01 -1.24055517e+00 -1.53932467e-01 1.32583630e+00
1.08251438e-01 -7.55252481e-01 -4.64783192e-01 -6.09265447e-01
-4.93431985e-02 8.87842715e-01 -5.52321494e-01 -4.92470443e-01
-8.17305028e-01 -9.59237039e-01 6.11480832e-01 8.04756343e-01
6.25222385e-01 -1.05595827e+00 1.56253472e-01 -2.36452535e-01
-3.73555630e-01 -9.33233619e-01 -4.43335801e-01 5.42408638e-02
-1.00005960e+00 -9.95603025e-01 -4.59141433e-01 -1.00962663e+00
8.62797976e-01 2.10824147e-01 1.10591376e+00 2.46135443e-01
-5.08590005e-02 2.79725105e-01 -7.47396767e-01 -5.28727353e-01
-3.90589774e-01 5.45000494e-01 1.64803803e-01 -4.39180434e-01
4.25960600e-01 -7.53988266e-01 -3.34511697e-01 2.30798855e-01
-1.06445575e+00 2.89437056e-01 3.76590341e-01 5.71312368e-01
3.68217438e-01 2.21564710e-01 6.99314654e-01 -1.06614912e+00
1.20261574e+00 -5.29977441e-01 -2.25886017e-01 3.99265021e-01
-6.40859902e-01 4.46253307e-02 7.75629282e-01 -9.96689200e-01
-9.28486049e-01 -3.82850021e-01 -9.35351327e-02 -3.19867820e-01
4.86057326e-02 7.38924742e-01 -2.29515612e-01 5.07429102e-03
7.91482210e-01 4.88740772e-01 -1.06870241e-01 -6.45611942e-01
3.83794606e-01 4.00240123e-01 2.96456277e-01 -1.43054056e+00
8.88362229e-01 -1.62802991e-02 -2.45341659e-01 -6.50359988e-01
-9.33461845e-01 -1.33874491e-01 -4.50241923e-01 1.80728644e-01
3.49367142e-01 -6.12655878e-01 -5.80718338e-01 4.55932826e-01
-1.09106088e+00 -5.81101000e-01 -3.97222787e-01 3.24394703e-01
-1.68958843e-01 3.82496417e-01 -6.22256577e-01 -4.33232814e-01
-3.15426797e-01 -1.17100978e+00 4.56344306e-01 2.39330202e-01
-5.86836159e-01 -8.12787831e-01 7.06559867e-02 6.21899068e-01
3.87951016e-01 -1.56046644e-01 1.41106999e+00 -1.01548672e+00
-3.10020268e-01 9.98597592e-02 -1.08089864e-01 5.64966381e-01
1.60346553e-01 -8.46138522e-02 -8.05750549e-01 -2.00910464e-01
-8.21752008e-03 -2.61008918e-01 8.53527069e-01 5.62621281e-02
1.76039159e+00 -4.36213672e-01 -1.27614945e-01 8.07966828e-01
1.02182329e+00 1.76177099e-01 4.74975526e-01 4.03780311e-01
6.40559018e-01 7.40643084e-01 3.46527785e-01 9.64699537e-02
2.94799328e-01 3.16011876e-01 1.00944526e-01 2.14112014e-01
-2.08794758e-01 -3.97589505e-01 2.92236567e-01 1.05387974e+00
-5.46322316e-02 -2.97910243e-01 -1.26985359e+00 5.12791753e-01
-1.60428524e+00 -5.67847311e-01 -1.65421888e-01 1.96922386e+00
1.12934458e+00 7.19289761e-03 -9.84642282e-03 2.25956962e-01
4.25958425e-01 -2.01037806e-02 -3.40909272e-01 -5.04080594e-01
-9.66218859e-02 6.33572996e-01 1.26116872e-01 4.21488613e-01
-6.98134363e-01 1.09545577e+00 5.72529030e+00 1.00275922e+00
-9.33824360e-01 -2.19854470e-02 5.89949191e-01 -1.33691013e-01
-5.48991144e-01 -2.28217095e-01 -8.18058193e-01 4.09455925e-01
5.55556834e-01 -3.80259663e-01 4.72159505e-01 5.60743093e-01
-1.50563672e-01 1.71423525e-01 -1.29600692e+00 8.40089500e-01
6.33275136e-02 -1.22104967e+00 6.34629250e-01 -2.46460527e-01
1.01561129e+00 -2.99644440e-01 4.24017966e-01 7.47339189e-01
1.88659266e-01 -1.20098090e+00 3.16503525e-01 2.57999361e-01
2.69129992e-01 -7.25442290e-01 5.69486260e-01 5.78450978e-01
-9.73634303e-01 -2.71922946e-01 -3.00744385e-01 -3.48115176e-01
-1.18938982e-01 3.79716128e-01 -5.57672858e-01 7.70863473e-01
6.53045893e-01 4.99571890e-01 -8.58869076e-01 8.40427697e-01
-4.77863193e-01 7.10203171e-01 -5.84477521e-02 6.39021173e-02
1.21577233e-01 -3.76365960e-01 3.61238033e-01 9.80364740e-01
4.02884305e-01 3.59959692e-01 1.50921822e-01 1.07867575e+00
-2.32515439e-01 3.55343997e-01 -3.34335476e-01 -2.71191776e-01
5.64783335e-01 9.46197629e-01 -4.00568008e-01 -2.92265385e-01
-3.12259108e-01 6.97319806e-01 5.24017453e-01 5.38462639e-01
-8.25267434e-01 -1.17180072e-01 5.97807527e-01 2.61056185e-01
-2.03510940e-01 -3.71415049e-01 -6.65186226e-01 -1.14701796e+00
2.88249016e-01 -1.22545695e+00 4.97739166e-01 -6.37846172e-01
-1.54094934e+00 4.73958671e-01 3.02324444e-01 -1.03411138e+00
4.08332795e-01 -7.32461810e-01 -7.93409586e-01 1.08211470e+00
-1.16930735e+00 -1.01902914e+00 -4.75869685e-01 3.43584061e-01
7.18625605e-01 -2.95312643e-01 6.14040911e-01 2.22890288e-01
-8.19533885e-01 8.42756510e-01 -1.69095039e-01 2.13046148e-01
7.16420352e-01 -9.81485248e-01 4.21858013e-01 9.00329232e-01
-2.44350582e-02 1.07179403e+00 5.01828909e-01 -6.63492084e-01
-1.05098796e+00 -1.09626913e+00 7.51639128e-01 -5.13284922e-01
7.37475574e-01 -3.26398879e-01 -1.43574500e+00 8.80060136e-01
2.79225081e-01 -3.69748026e-01 9.87291694e-01 3.19697797e-01
-4.63171780e-01 -6.47478402e-02 -8.04789305e-01 9.51676190e-01
1.28088593e+00 -1.61528558e-01 -1.00808930e+00 2.60749578e-01
1.01486897e+00 -6.02243602e-01 -7.48932362e-01 9.75712955e-01
2.28118166e-01 -5.23921132e-01 1.04544830e+00 -1.13845444e+00
9.16584134e-01 -1.39267638e-01 -1.03904672e-01 -1.50271475e+00
-3.96024525e-01 -2.41911858e-01 -7.71663338e-02 1.41943359e+00
7.65567899e-01 -8.19431245e-01 9.27414179e-01 7.47317612e-01
-3.86714667e-01 -1.09572351e+00 -4.36804175e-01 -9.62310135e-01
7.12897658e-01 -5.50441623e-01 8.96887779e-01 1.34651434e+00
2.71081924e-01 2.78472245e-01 4.67153639e-03 2.30793342e-01
9.98781919e-02 3.50331426e-01 9.16718423e-01 -9.07973289e-01
-2.67503500e-01 -7.65629172e-01 5.40564582e-02 -1.22064459e+00
3.41636121e-01 -1.49545908e+00 -4.04712290e-01 -1.41089404e+00
2.36177683e-01 -5.69637001e-01 -2.83864796e-01 6.55839980e-01
-5.24022818e-01 3.11404164e-03 2.55337477e-01 4.40278165e-02
-2.00623587e-01 7.54012942e-01 1.58692241e+00 -2.79358029e-01
-4.64270800e-01 -2.04574037e-02 -9.46183026e-01 5.65717280e-01
1.42057562e+00 -3.37535143e-01 -8.62908721e-01 -7.02970088e-01
4.38783705e-01 -5.16928196e-01 1.15520783e-01 -9.44265783e-01
1.94892272e-01 -3.92669320e-01 2.30926543e-01 -5.61567247e-01
2.05013618e-01 -5.74638486e-01 -4.54976596e-02 5.05713224e-01
-5.13065696e-01 3.28156829e-01 7.05276072e-01 2.00214935e-03
-2.33162001e-01 -4.91331607e-01 4.59102660e-01 -1.89559996e-01
-7.08258331e-01 3.05790037e-01 -1.64290905e-01 4.48353589e-01
9.38509822e-01 -1.59750611e-01 -6.53875709e-01 6.87344000e-02
-6.87166452e-01 6.34754360e-01 3.18118691e-01 6.59112990e-01
6.08334363e-01 -1.48453045e+00 -7.47572482e-01 1.53551430e-01
1.50778040e-01 3.67292404e-01 3.56953233e-01 7.97278285e-01
-5.69820702e-01 4.13442522e-01 -4.67140488e-02 -2.58585751e-01
-1.30837381e+00 8.49480212e-01 1.13106988e-01 -2.65887409e-01
-3.81279677e-01 1.17626560e+00 4.00792569e-01 -1.05309534e+00
2.28386179e-01 -7.33838677e-01 -1.57476366e-01 -1.05355249e-03
4.94078338e-01 4.80226278e-01 -1.47735060e-03 -5.41840233e-02
-9.66241211e-02 4.36997175e-01 -3.20666701e-01 2.90290743e-01
1.35377359e+00 1.58895552e-01 -3.10571760e-01 5.48632383e-01
9.68533397e-01 -7.18062278e-03 -6.58731043e-01 -4.96879399e-01
3.17488872e-02 -4.08980191e-01 -5.36092162e-01 -8.73839200e-01
-1.03970933e+00 8.45261097e-01 1.18539758e-01 1.72233041e-02
1.26979589e+00 1.61555465e-02 7.62121677e-01 4.14513469e-01
6.23922003e-03 -9.03244257e-01 1.60904720e-01 7.05278575e-01
1.16402817e+00 -1.19167757e+00 1.23769447e-01 -8.17095280e-01
-6.33404970e-01 9.95813727e-01 1.17406094e+00 1.36269778e-01
4.73048210e-01 1.36796281e-01 -1.36816844e-01 -4.48267199e-02
-7.33903944e-01 1.66860938e-01 4.69918162e-01 3.39528710e-01
5.27175486e-01 -1.06940374e-01 -7.52951860e-01 1.09509790e+00
-7.56321311e-01 -9.83808413e-02 3.85340095e-01 8.32962751e-01
-2.36473426e-01 -1.46689343e+00 -2.38338619e-01 3.59580368e-01
-3.82637158e-02 -5.73950112e-01 -5.39209425e-01 8.17777634e-01
3.30693871e-01 6.74716473e-01 -8.71609747e-02 -4.67490613e-01
4.58363384e-01 4.31580991e-01 6.65915906e-01 -8.12275946e-01
-6.30085886e-01 -8.06183219e-01 -1.07040673e-01 -4.35081303e-01
-2.09404945e-01 -3.63021135e-01 -1.26328266e+00 -5.51961541e-01
8.18692967e-02 3.70966882e-01 3.19570243e-01 8.83304715e-01
1.13056242e-01 7.85735786e-01 2.51125485e-01 -3.93118978e-01
-4.54660147e-01 -1.17576098e+00 -2.76108623e-01 4.76622820e-01
-5.78858666e-02 -5.23071349e-01 -3.35215092e-01 -2.75187027e-02] | [9.988358497619629, 7.670563697814941] |
15191457-cab7-4799-ac49-5e4671be48ea | learning-ontologies-with-epistemic-reasoning | 1902.03273 | null | http://arxiv.org/abs/1902.03273v1 | http://arxiv.org/pdf/1902.03273v1.pdf | Learning Ontologies with Epistemic Reasoning: The EL Case | We investigate the problem of learning description logic ontologies from
entailments via queries, using epistemic reasoning. We introduce a new learning
model consisting of epistemic membership and example queries and show that
polynomial learnability in this model coincides with polynomial learnability in
Angluin's exact learning model with membership and equivalence queries. We then
instantiate our learning framework to EL and show some complexity results for
an epistemic extension of EL where epistemic operators can be applied over the
axioms. Finally, we transfer known results for EL ontologies and its fragments
to our learning model based on epistemic reasoning. | ['Nicolas Troquard', 'Ana Ozaki'] | 2019-02-08 | null | null | null | null | ['epistemic-reasoning'] | ['miscellaneous'] | [-2.60531269e-02 1.17392480e+00 4.96311449e-02 -5.56017280e-01
-8.45977008e-01 -7.89955020e-01 5.84428906e-01 2.03910634e-01
-1.66575655e-01 1.05151892e+00 4.75439616e-02 -4.23945814e-01
-9.00435328e-01 -1.36741781e+00 -1.21880698e+00 -3.33143562e-01
-6.78804874e-01 8.38311851e-01 7.57782638e-01 -3.86509806e-01
-2.00623885e-01 2.29097232e-01 -1.80918968e+00 4.74959165e-01
7.42877543e-01 8.00293386e-01 -5.32081127e-01 6.01155221e-01
-2.63398886e-01 1.73397398e+00 -7.75880143e-02 -5.64991474e-01
4.08503301e-02 -2.42164582e-01 -1.66195142e+00 -1.50324434e-01
5.50248981e-01 -1.03098795e-01 -3.94497901e-01 1.47215033e+00
3.10381204e-02 1.26547083e-01 5.19438088e-01 -1.68323755e+00
-8.35727334e-01 1.43874776e+00 7.09592462e-01 -3.04520130e-01
9.47968125e-01 -2.65685856e-01 1.47517180e+00 -3.08835417e-01
6.90266728e-01 1.47399569e+00 8.21343780e-01 7.89256454e-01
-1.43316936e+00 7.09142582e-03 -1.35748401e-01 8.91488612e-01
-1.36976755e+00 -3.46697688e-01 2.51255095e-01 -4.91804481e-01
7.59927988e-01 4.28423703e-01 5.02600491e-01 4.19950545e-01
2.14803685e-02 8.57361376e-01 1.44269490e+00 -1.00616491e+00
4.43119049e-01 4.59322155e-01 6.36105776e-01 9.94326115e-01
5.79477549e-01 2.41490886e-01 -5.13303697e-01 -2.43523851e-01
3.75368148e-01 -5.25055826e-01 -4.16817188e-01 -5.20847678e-01
-5.91951489e-01 9.05343592e-01 1.86623894e-02 1.01393841e-01
2.63719801e-02 4.57676947e-01 1.28933534e-01 9.85018015e-01
-2.42582336e-01 5.38375974e-01 -6.26415372e-01 3.01151037e-01
-2.29408994e-01 4.33199823e-01 1.68760777e+00 1.15584874e+00
8.37069452e-01 -3.10668021e-01 3.44166845e-01 6.71176165e-02
6.00977123e-01 7.64888883e-01 5.76859005e-02 -1.80271804e+00
-3.27968478e-01 3.62312794e-01 5.29423177e-01 -1.09061442e-01
-2.91003883e-01 -1.98456831e-02 3.38383287e-01 4.47947830e-01
5.77710211e-01 4.24734876e-02 -1.21436961e-01 2.17181182e+00
1.79807425e-01 3.82610291e-01 9.76536751e-01 2.39784777e-01
7.26006746e-01 4.47828561e-01 4.47144965e-03 -4.29963619e-01
1.37905407e+00 -3.95726889e-01 -7.86953032e-01 3.11387926e-01
1.07257271e+00 2.68482924e-01 9.81533170e-01 6.93675816e-01
-1.23909461e+00 2.55124420e-01 -9.89890397e-01 -1.87570110e-01
-3.19231331e-01 -6.28448665e-01 1.06036878e+00 5.17386675e-01
-1.06687462e+00 5.53806782e-01 -7.08958030e-01 -1.47836924e-01
1.39830753e-01 6.09960437e-01 -5.16747177e-01 -2.93280482e-01
-1.91647947e+00 1.02714503e+00 1.13012850e+00 -2.04648852e-01
-1.25953043e+00 -2.73993671e-01 -1.18432367e+00 -2.47107800e-02
9.39559698e-01 -6.18655682e-01 1.64528942e+00 -9.14164066e-01
-1.68315279e+00 8.15586030e-01 2.30542511e-01 -8.01658213e-01
2.60513484e-01 6.71809167e-02 -7.39279211e-01 2.01745167e-01
-1.15859769e-01 -1.41634285e-01 1.12895936e-01 -1.23075640e+00
-9.21454728e-01 -3.79838496e-01 1.56559217e+00 -9.12764519e-02
1.76922485e-01 -7.86404461e-02 5.28021872e-01 5.90892196e-01
2.12465316e-01 -9.22628701e-01 1.32679686e-01 -3.07639949e-02
3.04481518e-02 -5.85028648e-01 3.58128846e-01 -1.33955535e-02
8.39603603e-01 -1.85349584e+00 8.75177830e-02 4.83528674e-01
1.56567484e-01 -5.77200234e-01 1.27231330e-01 2.90396661e-01
-7.10652694e-02 9.10880044e-02 -5.84099591e-02 4.71927196e-01
1.05724037e+00 7.62191892e-01 -5.17786026e-01 4.39764917e-01
-2.24612117e-01 9.06172097e-01 -8.48223865e-01 -7.35105574e-01
-3.87786217e-02 -1.95911918e-02 -9.53136623e-01 4.27816920e-02
-1.09868813e+00 -1.70277193e-01 -4.60047483e-01 4.85993505e-01
3.19192171e-01 -4.54012007e-01 6.21081412e-01 2.78237462e-01
1.96274608e-01 3.06794196e-01 -1.61927259e+00 1.54717481e+00
-6.09168291e-01 2.43634388e-01 -2.88381856e-02 -6.04822636e-01
2.97105879e-01 8.14557076e-01 -1.45989269e-01 9.08559188e-02
6.55312687e-02 3.32557470e-01 -3.52455676e-02 -1.08745658e+00
-5.65192066e-02 -8.02916229e-01 -4.75247711e-01 4.69855458e-01
2.38697723e-01 -2.39529043e-01 5.42059764e-02 2.82522827e-01
8.88514221e-01 3.44696522e-01 2.41903007e-01 -5.29522538e-01
9.16175961e-01 1.25741199e-01 5.36174417e-01 9.47012365e-01
1.31151900e-01 -5.40190995e-01 8.97906363e-01 -3.83884549e-01
-4.81169790e-01 -1.36812341e+00 -5.27142704e-01 1.28207958e+00
1.91436350e-01 -6.74421370e-01 -4.54840600e-01 -6.40728951e-01
-7.65547901e-02 1.19840205e+00 -6.79221034e-01 -3.12199712e-01
-7.86000565e-02 -8.62857476e-02 1.03459311e+00 3.62820268e-01
4.26430821e-01 -8.75028372e-01 -5.15422106e-01 -1.33800192e-03
-1.91128209e-01 -8.38844240e-01 5.06036639e-01 4.09285367e-01
-7.46187329e-01 -1.40389860e+00 5.84958851e-01 -6.89650893e-01
3.30345273e-01 -1.02785277e+00 9.58570123e-01 -9.85803604e-02
5.26162446e-01 8.88795137e-01 -1.20106243e-01 -5.97594798e-01
-7.36550450e-01 -2.37482220e-01 9.18922126e-02 -3.50530386e-01
6.43358886e-01 -4.51286912e-01 3.44985366e-01 -2.22176433e-01
-1.30972219e+00 -3.16797167e-01 -6.27831444e-02 7.35504568e-01
3.52131695e-01 5.67927659e-01 4.06976908e-01 -1.31168032e+00
3.02637726e-01 -4.67533320e-01 -1.07353461e+00 7.77322710e-01
-6.46209061e-01 7.84085870e-01 5.34064829e-01 -2.89534420e-01
-1.20109046e+00 6.62454171e-03 1.83749706e-01 2.50777483e-01
-4.14529741e-02 1.03757286e+00 -8.30146372e-01 -2.08837524e-01
9.77466524e-01 -3.24274302e-02 -1.52331173e-01 -6.42234907e-02
5.23899078e-01 6.10552311e-01 6.47739232e-01 -1.45759165e+00
6.37406170e-01 5.80288589e-01 6.26837671e-01 -3.79408747e-01
-1.39407349e+00 3.73606294e-01 -3.57721567e-01 -6.63873693e-03
4.59922314e-01 -8.49271655e-01 -1.38724756e+00 -2.80278409e-03
-8.12222540e-01 -5.25155246e-01 -9.77864742e-01 7.38537014e-01
-1.35391164e+00 4.75900084e-01 -7.59070694e-01 -1.16568935e+00
5.30048430e-01 -9.12448943e-01 4.49082702e-01 -1.18885010e-01
-3.89340967e-02 -1.30005670e+00 2.85305738e-01 1.36643797e-01
9.99315307e-02 1.62034392e-01 1.34050548e+00 -1.27211225e+00
-6.73314214e-01 -2.93997347e-01 2.50310153e-01 3.12246293e-01
-4.05906647e-01 -2.87543803e-01 -1.06260562e+00 1.44595608e-01
-9.79261473e-02 -7.62459219e-01 2.69983649e-01 -2.21593250e-02
5.77068150e-01 -7.54345834e-01 -1.46462200e-02 1.47878736e-01
1.90473056e+00 -1.80378288e-01 7.89167166e-01 6.04984283e-01
-5.81821762e-02 5.56700766e-01 3.22207212e-01 1.24201015e-01
6.12362921e-01 2.76518524e-01 3.21605027e-01 1.05085516e+00
4.41796273e-01 -1.07873887e-01 2.96842009e-01 2.86137283e-01
-5.28962135e-01 4.10514772e-01 -8.40896070e-01 3.74186814e-01
-2.19315100e+00 -1.44153523e+00 -1.03999622e-01 2.21522927e+00
1.67923117e+00 2.81435475e-02 -2.73405015e-02 2.00907946e-01
4.27593559e-01 -6.02716684e-01 -1.37635112e-01 -2.19253391e-01
-5.07586479e-01 4.34005529e-01 2.02949554e-01 1.45457721e+00
-7.71268487e-01 7.90107846e-01 7.67968559e+00 2.78544933e-01
-3.19263071e-01 4.56138998e-01 -5.88957310e-01 7.12960884e-02
-8.67462158e-01 6.50867939e-01 -5.64196765e-01 -1.26499776e-02
1.45462334e+00 -4.02971178e-01 7.35604584e-01 1.06957102e+00
-5.95037520e-01 -2.60441247e-02 -1.76134408e+00 3.48653764e-01
-9.84478965e-02 -1.19523966e+00 4.77901325e-02 -1.33429572e-01
6.81174159e-01 -4.56926793e-01 -3.69369775e-01 5.44903994e-01
1.11250865e+00 -1.05603755e+00 9.32632685e-01 8.90650451e-01
5.97192824e-01 -8.93308401e-01 1.10178530e+00 3.25323403e-01
-7.47952521e-01 -6.40812397e-01 -3.77100646e-01 -3.07503074e-01
-2.67253211e-03 5.97125180e-02 -2.23601624e-01 6.29160821e-01
1.47940189e-01 2.89338827e-01 -1.87699303e-01 1.06878161e+00
-7.71276534e-01 4.66954082e-01 -5.54675281e-01 2.35691965e-01
-2.12341711e-01 -6.22223802e-02 3.56193662e-01 8.96735489e-01
9.82938036e-02 5.00941098e-01 4.56319936e-02 8.16606104e-01
-1.15271755e-01 -3.38426650e-01 -5.08486509e-01 3.13911706e-01
5.31410456e-01 4.77752388e-01 1.74616992e-01 -7.78777957e-01
-4.85423297e-01 2.63824642e-01 2.43933052e-01 2.53171891e-01
-6.50882483e-01 -7.14009583e-01 3.54146659e-01 -7.34017044e-02
3.90883461e-02 3.42440754e-01 4.88184482e-01 -1.30290937e+00
-1.86130583e-01 -7.88383901e-01 9.53380764e-01 -9.06821966e-01
-1.16254580e+00 2.71753341e-01 6.86830223e-01 -6.01844907e-01
-3.48850250e-01 -8.67062092e-01 -1.22981975e-02 5.06222904e-01
-1.58512580e+00 -1.17359900e+00 5.91453016e-02 8.31044912e-01
-3.11564922e-01 2.34751374e-01 1.39347422e+00 -1.99457973e-01
2.41971478e-01 3.22680593e-01 -2.51799673e-01 -2.29229301e-01
2.24850297e-01 -1.76865315e+00 -8.31788063e-01 5.89778423e-01
-4.43866216e-02 7.08495677e-01 1.16183949e+00 -2.35035166e-01
-1.63576651e+00 -9.88633931e-01 1.15436888e+00 -8.13463449e-01
1.08602071e+00 1.20476387e-01 -9.47348595e-01 1.89846134e+00
2.27855667e-01 1.81460842e-01 8.40986311e-01 3.27374339e-01
-8.10010672e-01 -1.44682527e-01 -1.39626169e+00 4.50371712e-01
1.05776906e+00 -1.03046978e+00 -1.57306445e+00 5.15343368e-01
9.58234012e-01 -3.88246626e-01 -1.36967444e+00 3.21800292e-01
7.80009210e-01 -5.80117404e-01 4.99069661e-01 -1.17078578e+00
-1.89589038e-01 -8.04581463e-01 -7.41673589e-01 -7.82770813e-01
-2.92231500e-01 -5.62253714e-01 -4.08067882e-01 6.98721111e-01
5.72197556e-01 -1.18318367e+00 3.04413885e-01 1.09913826e+00
9.40090418e-02 -1.38881296e-01 -1.20518577e+00 -9.40627098e-01
4.32185382e-01 -7.27175117e-01 7.98455358e-01 8.29971552e-01
1.01632559e+00 2.71146983e-01 1.42713170e-02 7.11405456e-01
8.42034161e-01 1.82924300e-01 2.22730473e-01 -1.84616065e+00
-8.49971175e-01 5.71401753e-02 -6.42129779e-01 -4.65831697e-01
8.16960275e-01 -1.39554179e+00 1.82561591e-01 -1.43002725e+00
2.07288176e-01 -2.22955301e-01 -5.86196668e-02 9.03985620e-01
4.95445818e-01 -6.18630797e-02 -8.09168220e-02 1.18019216e-01
-1.25899255e+00 2.34814525e-01 6.62999809e-01 -1.78162053e-01
2.13091180e-01 -1.11833721e-01 -7.26424992e-01 1.07463288e+00
4.84188259e-01 -4.44809437e-01 -5.98236978e-01 4.56528924e-02
1.24855840e+00 8.42306018e-02 6.21144474e-01 -9.67659533e-01
4.27406192e-01 -6.25007451e-01 -5.53061247e-01 3.66880327e-01
1.01273628e-02 -1.09368908e+00 6.28148615e-01 4.68414485e-01
-8.61941576e-01 -5.51499844e-01 -4.54468764e-02 5.26872873e-01
4.86823209e-02 -9.46441293e-01 5.46446502e-01 -2.20609680e-01
-8.73051107e-01 -1.75719693e-01 -6.17734969e-01 3.65799248e-01
1.08528769e+00 1.21949412e-01 -4.15983826e-01 -3.24487507e-01
-1.74144387e+00 2.05006078e-01 6.67294800e-01 -6.40766919e-01
2.86982328e-01 -1.18829691e+00 -4.58613038e-01 -1.85325116e-01
3.81267756e-01 -4.80980128e-02 7.32393265e-02 9.03735220e-01
-5.33710957e-01 3.75043511e-01 -4.56726877e-03 -5.29755950e-02
-1.13953280e+00 5.65529287e-01 7.79759884e-01 1.36585906e-01
-3.38257462e-01 5.32790482e-01 -1.83061704e-01 -7.60179102e-01
4.58754361e-01 -4.36253339e-01 -6.47654235e-02 -5.63835144e-01
6.53068781e-01 2.72566855e-01 -3.80535781e-01 -1.97398320e-01
-5.10511816e-01 2.89785147e-01 3.68881762e-01 -3.90177339e-01
1.14164042e+00 -2.37353295e-01 -5.84487796e-01 9.63563263e-01
5.29859543e-01 8.75333324e-02 -7.63211489e-01 -6.28346384e-01
3.07371110e-01 6.55143559e-02 -3.57736230e-01 -7.47495353e-01
-2.43699655e-01 3.60164315e-01 3.33122492e-01 4.66007203e-01
8.01731288e-01 8.44666719e-01 -2.39864402e-02 1.28203177e+00
8.90541077e-01 -1.01242602e+00 -5.73690593e-01 6.18970513e-01
5.80715775e-01 -6.93471670e-01 -6.78353128e-04 -4.27146584e-01
-2.44092494e-01 1.20859408e+00 1.11113451e-01 -3.37800346e-02
6.41964018e-01 5.95793188e-01 -1.61038369e-01 -4.01559204e-01
-1.22076607e+00 -3.40341747e-01 -3.48012418e-01 7.62677789e-01
5.47896978e-03 2.43072450e-01 -1.55530870e-01 9.21703100e-01
-2.90999234e-01 6.62152886e-01 1.04588735e+00 1.31987667e+00
-8.25125575e-01 -1.13255870e+00 -2.67971158e-01 1.26771882e-01
-5.77225089e-01 -6.27429709e-02 -2.73838583e-02 9.41068292e-01
3.31462115e-01 8.63086343e-01 -3.52360249e-01 -1.00400068e-01
1.82024196e-01 5.73710144e-01 1.19548428e+00 -8.71109426e-01
8.52383450e-02 -6.37748897e-01 4.62438911e-01 -3.25958520e-01
-8.31672132e-01 -5.97371697e-01 -1.70070314e+00 -3.96125853e-01
-6.37109339e-01 8.91057491e-01 2.07692105e-02 1.23224163e+00
-2.80968100e-01 1.48336917e-01 1.53150424e-01 2.39074543e-01
-1.31462669e+00 -7.14022458e-01 -1.29071391e+00 2.79753990e-02
3.94992352e-01 -3.98024499e-01 -9.50977981e-01 4.33787525e-01] | [8.626334190368652, 6.686056613922119] |
3cef6dd2-ad2a-404e-908d-91032c9adfce | natural-questions-in-icelandic | null | null | https://aclanthology.org/2022.lrec-1.477 | https://aclanthology.org/2022.lrec-1.477.pdf | Natural Questions in Icelandic | We present the first extractive question answering (QA) dataset for Icelandic, Natural Questions in Icelandic (NQiI). Developing such datasets is important for the development and evaluation of Icelandic QA systems. It also aids in the development of QA methods that need to work for a wide range of morphologically and grammatically different languages in a multilingual setting. The dataset was created by asking contributors to come up with questions they would like to know the answer to. Later, they were tasked with finding answers to each others questions following a previously published methodology. The questions are Natural in the sense that they are real questions posed out of interest in knowing the answer. The complete dataset contains 18 thousand labeled entries of which 5,568 are directly suitable for training an extractive QA system for Icelandic. The dataset is a valuable resource for Icelandic which we demonstrate by creating and evaluating a system capable of extractive QA in Icelandic. | ['Hafsteinn Einarsson', 'Vésteinn Snæbjarnarson'] | null | null | null | null | lrec-2022-6 | ['natural-questions'] | ['miscellaneous'] | [-7.13083670e-02 3.98264021e-01 6.79241419e-01 -5.20548940e-01
-1.52240312e+00 -1.31028092e+00 5.90854466e-01 4.04179752e-01
-7.07076848e-01 9.99836266e-01 6.10178351e-01 -7.33983874e-01
-1.86204031e-01 -9.13375020e-01 -5.38943648e-01 -1.42666042e-01
3.06971490e-01 1.38282430e+00 -1.85717810e-02 -1.05531597e+00
1.79194778e-01 9.60860252e-02 -1.21946371e+00 4.79715616e-01
1.35100865e+00 5.19773543e-01 1.88881889e-01 9.12063420e-01
-3.13126385e-01 9.56178010e-01 -9.07056868e-01 -8.58461320e-01
2.84165025e-01 -4.92751926e-01 -1.77642143e+00 -4.80867654e-01
1.02184641e+00 -4.88416329e-02 3.43575105e-02 8.11564207e-01
4.09566790e-01 2.37802099e-02 7.23380625e-01 -7.64158607e-01
-5.67858517e-01 4.54817116e-01 7.87891328e-01 3.90066743e-01
9.63608742e-01 1.20660380e-01 1.49773192e+00 -9.46744323e-01
8.97166193e-01 1.18697226e+00 2.29120567e-01 7.78857350e-01
-6.91367626e-01 -2.04997003e-01 -5.41075826e-01 4.89950806e-01
-8.50884199e-01 -4.42529708e-01 3.43125105e-01 -3.23385030e-01
1.01587212e+00 6.38572633e-01 1.55344397e-01 5.82503080e-01
-1.30340252e-02 7.01422215e-01 1.23696721e+00 -7.00434327e-01
6.85573369e-02 2.85575211e-01 1.81316763e-01 6.71316981e-01
1.71835005e-01 -1.46603361e-01 -4.65507925e-01 -2.17843860e-01
3.49992216e-02 -7.52093434e-01 -3.30048382e-01 2.70643979e-01
-1.09089196e+00 1.03861380e+00 6.23916574e-02 3.15969229e-01
-1.68673873e-01 -2.08732143e-01 5.53585947e-01 1.03540492e+00
1.06593415e-01 9.69482243e-01 -1.13843477e+00 -1.75131232e-01
-6.23141944e-01 8.39896560e-01 1.39004838e+00 8.41426790e-01
7.66730726e-01 -2.45242625e-01 1.40171006e-01 5.22211790e-01
3.56824607e-01 9.98681724e-01 4.11559314e-01 -1.22960925e+00
1.10314775e+00 1.08169842e+00 2.72679955e-01 -7.69420087e-01
-3.46539408e-01 -4.47263457e-02 -2.35171840e-01 -6.49848357e-02
8.29836190e-01 -4.11635220e-01 -7.52111852e-01 1.32219660e+00
3.86372387e-01 -1.12380373e+00 7.97388911e-01 8.09223711e-01
1.67525005e+00 8.06353807e-01 6.69435486e-02 1.85267434e-01
1.96566057e+00 -6.15209520e-01 -7.51898706e-01 -3.50477874e-01
7.91057050e-01 -1.00116122e+00 1.23974478e+00 2.45158687e-01
-8.87439728e-01 -5.15117347e-01 -6.77529156e-01 -5.37219048e-01
-5.90907276e-01 9.70992818e-02 3.63072991e-01 9.63109851e-01
-1.01317096e+00 -2.74103973e-02 -1.85490757e-01 -3.37914050e-01
-2.91867703e-01 2.58208752e-01 -6.09014928e-01 -3.57934564e-01
-1.70212817e+00 1.23064160e+00 3.81017566e-01 3.69871110e-02
-7.86171257e-01 -3.95022869e-01 -1.25642538e+00 -9.98289790e-03
3.57318491e-01 -6.41122520e-01 1.45310605e+00 -8.01410735e-01
-1.00281441e+00 1.09712064e+00 -2.03051165e-01 -4.60916400e-01
1.52619034e-01 -4.03536439e-01 -5.72749615e-01 4.80993658e-01
6.60282552e-01 6.66156352e-01 4.56906796e-01 -1.18373227e+00
-6.34071290e-01 -6.16452217e-01 5.54798961e-01 4.22046810e-01
1.05801383e-02 1.49434417e-01 -8.36986210e-03 -3.63909483e-01
1.65338248e-01 -7.28808343e-01 4.22965586e-02 -7.25332737e-01
2.31868345e-02 -3.88068259e-01 5.32527745e-01 -1.59252834e+00
9.51347053e-01 -1.47327173e+00 -1.66727886e-01 -2.06835512e-02
1.58740863e-01 4.04075712e-01 -2.13969812e-01 1.14308012e+00
2.17145488e-01 1.88821405e-01 -5.33415258e-01 4.27321017e-01
2.74688572e-01 5.47677279e-01 -4.11168158e-01 -1.22416466e-01
4.91909504e-01 9.79664087e-01 -1.11008871e+00 -8.41016352e-01
-1.39776930e-01 -1.86013475e-01 -3.66766483e-01 2.98873067e-01
-4.98395652e-01 4.81810659e-01 -5.04276574e-01 6.24027371e-01
5.01430929e-01 2.96419680e-01 1.90180480e-01 9.98050943e-02
6.90794364e-02 7.74179280e-01 -7.97483385e-01 1.40964854e+00
-6.53449535e-01 7.00671732e-01 -6.62834719e-02 -5.09997904e-01
1.13367164e+00 6.30460322e-01 -5.43186702e-02 -9.78749335e-01
-4.47726436e-02 6.83286071e-01 -1.84639983e-04 -9.32535112e-01
1.03211915e+00 -4.99345630e-01 -5.75126171e-01 4.32300478e-01
3.93255323e-01 -6.32485032e-01 7.21327662e-01 5.09513259e-01
1.26782131e+00 2.08704565e-02 1.12793759e-01 -5.28820753e-01
1.13365960e+00 7.78293669e-01 4.27179307e-01 4.67942536e-01
-1.42947644e-01 4.34818506e-01 2.85628498e-01 -6.45590007e-01
-8.79887342e-01 -9.03811455e-01 6.74834773e-02 8.93263817e-01
-2.41831228e-01 -5.58408916e-01 -7.34014034e-01 -1.02256989e+00
-4.89021450e-01 8.21908355e-01 -4.30588543e-01 4.12264138e-01
-1.03831530e+00 4.48035076e-02 8.97716284e-01 1.39160454e-01
6.64637804e-01 -1.38506889e+00 -7.11149275e-01 3.47874224e-01
-1.13910377e+00 -1.02557325e+00 -1.67040810e-01 1.71429530e-01
-8.20659876e-01 -1.48450553e+00 -6.84149444e-01 -1.13440216e+00
3.34401339e-01 -3.76331180e-01 1.86405504e+00 2.13712737e-01
3.34807709e-02 7.12542653e-01 -7.21959352e-01 -5.58098197e-01
-9.61300313e-01 4.34438825e-01 -5.07729590e-01 -6.48718417e-01
6.58364117e-01 8.57311394e-03 -2.51125544e-01 1.94397539e-01
-1.20672703e+00 -5.58602452e-01 3.71128350e-01 6.62842035e-01
6.70729280e-02 -4.12530541e-01 7.56882727e-01 -1.31885386e+00
9.41664577e-01 -3.88082713e-01 -5.11406481e-01 5.72366178e-01
-9.24987122e-02 4.62674558e-01 8.35994959e-01 5.68686306e-01
-1.32236624e+00 5.18358126e-03 -8.18135142e-01 7.99245656e-01
-4.36283588e-01 5.90908110e-01 -3.49426806e-01 7.23465532e-03
1.14780307e+00 6.33233935e-02 -3.79089594e-01 -3.87653381e-01
5.83231390e-01 8.08478057e-01 6.95501506e-01 -8.75027239e-01
9.88359869e-01 2.47535445e-02 -2.30164140e-01 -8.23233783e-01
-6.32553995e-01 -8.88608515e-01 -7.48793125e-01 -2.83702463e-01
6.51879251e-01 -8.66573036e-01 -5.82080841e-01 2.78355144e-02
-1.23798430e+00 -6.19180724e-02 -3.75843763e-01 4.30762433e-02
-5.34812868e-01 4.54474479e-01 -5.00299811e-01 -6.94826424e-01
-7.71041334e-01 -6.80606067e-01 9.33310509e-01 2.03897461e-01
-3.73150259e-01 -1.28046179e+00 6.70479119e-01 1.12210727e+00
1.34277031e-01 1.42173186e-01 1.44875574e+00 -8.49033952e-01
-9.82984900e-02 -1.26760651e-03 -2.80253440e-02 4.48086083e-01
6.90066367e-02 -5.46989799e-01 -7.50014842e-01 1.33474190e-02
3.13855588e-01 -7.99361765e-01 4.75854516e-01 -4.03285325e-01
-3.88489403e-02 -5.14366329e-01 4.53548729e-01 -3.40134591e-01
1.58198380e+00 4.49946932e-02 9.45696473e-01 5.76576948e-01
1.96642160e-01 1.13600862e+00 7.48372853e-01 -2.53017813e-01
1.03195453e+00 2.44838089e-01 2.93927610e-01 6.45824149e-02
-1.88691184e-01 -9.46176425e-02 4.72331464e-01 1.12295604e+00
1.43289998e-01 -1.32705644e-01 -1.54356730e+00 1.19029689e+00
-1.35476577e+00 -6.65860236e-01 -7.07204461e-01 1.81741798e+00
1.08398092e+00 -4.45031315e-01 -1.45415872e-01 1.96655154e-01
1.84626281e-01 -2.56998986e-01 4.02520783e-02 -7.41458476e-01
-2.72635460e-01 8.62201273e-01 7.09597841e-02 7.23175228e-01
-9.53117669e-01 1.12928903e+00 5.97731543e+00 4.33297724e-01
-4.09293503e-01 -5.08244075e-02 2.62693197e-01 8.32897961e-01
-6.93906069e-01 3.19725603e-01 -8.22303534e-01 -1.82235211e-01
1.34780920e+00 1.91115290e-02 2.43878998e-02 2.92851031e-01
-5.65670319e-02 -3.90855312e-01 -7.96896040e-01 4.49076951e-01
3.63589138e-01 -1.09190178e+00 3.30191821e-01 -4.21894222e-01
6.34164333e-01 -4.66513336e-02 -3.70785475e-01 5.05080998e-01
6.40130937e-01 -1.14218128e+00 5.58221519e-01 3.82090211e-01
5.33727348e-01 -8.20763409e-01 1.31318581e+00 4.97115582e-01
-9.27896857e-01 6.14358578e-03 -4.36043262e-01 -1.45806521e-01
2.02511296e-01 2.11703435e-01 -1.00618601e+00 1.00948322e+00
6.18966937e-01 -2.12805867e-01 -9.55509901e-01 8.21324170e-01
-7.23457754e-01 9.22395766e-01 -2.30347961e-01 -3.14564288e-01
6.17669880e-01 -3.98281723e-01 3.44429910e-01 1.13286352e+00
7.16221854e-02 2.62633026e-01 -1.53624967e-01 3.79873872e-01
-2.23902375e-01 7.03733504e-01 -4.35052216e-01 -1.19798696e-02
1.41259149e-01 1.21532547e+00 -3.73549551e-01 -2.92747885e-01
-2.60796905e-01 8.05630565e-01 2.36438155e-01 9.04399157e-02
-1.60902902e-01 -9.90921497e-01 -5.74985344e-04 7.90516064e-02
-1.87162921e-01 -3.19825381e-01 -6.14867546e-02 -1.05553067e+00
4.68812585e-01 -1.75068259e+00 8.41799974e-01 -8.20866585e-01
-1.24551880e+00 9.43011165e-01 -1.65521696e-01 -8.54701519e-01
-6.70186341e-01 -8.94987702e-01 -2.43847042e-01 1.01851976e+00
-1.60502970e+00 -1.22203541e+00 -1.92305163e-01 6.30508125e-01
5.86609006e-01 -8.61813203e-02 1.14981365e+00 4.44669306e-01
4.84675497e-01 1.43143767e-02 -1.67156175e-01 5.59884608e-01
8.59127879e-01 -1.64563358e+00 5.33905923e-01 7.14778125e-01
6.21749163e-01 6.18835151e-01 6.51337028e-01 -6.55508518e-01
-1.30686283e+00 -7.89421022e-01 2.03574133e+00 -8.74884009e-01
5.35783052e-01 -2.43628174e-01 -8.89758646e-01 4.64082748e-01
6.18461311e-01 -7.32688844e-01 9.59057391e-01 8.00536051e-02
-1.09306365e-01 6.87999502e-02 -1.18965316e+00 1.93168879e-01
3.66637707e-01 -7.51045227e-01 -1.48034072e+00 6.00651324e-01
4.82157558e-01 -3.31451893e-01 -9.00561988e-01 1.39202312e-01
1.03948869e-01 -6.70110285e-01 5.35958171e-01 -9.14756000e-01
3.85440439e-01 -4.26763743e-01 -2.40840092e-01 -1.24658597e+00
3.48699450e-01 -5.78483880e-01 4.36430454e-01 1.24441755e+00
1.02498364e+00 -5.61618567e-01 8.30141962e-01 6.72297418e-01
-2.69945025e-01 -8.07386488e-02 -1.09455729e+00 -6.12830818e-01
6.94561243e-01 -2.62168229e-01 5.85797608e-01 7.21707106e-01
-4.87229973e-02 7.03486145e-01 1.03882030e-01 9.60316062e-02
1.28105164e-01 7.45844841e-02 7.87595153e-01 -1.30451453e+00
4.12410051e-02 3.59025240e-01 -1.48745462e-01 -1.02103627e+00
2.13945545e-02 -9.19283330e-01 3.22647125e-01 -1.90595663e+00
-2.38193899e-01 -4.09907371e-01 6.70824170e-01 3.88807207e-01
-3.93592119e-01 1.67469129e-01 1.37850121e-01 1.09243624e-01
-4.45691586e-01 4.73332703e-01 1.16697025e+00 -1.17662556e-01
1.79136008e-01 -5.99547923e-02 -7.78755128e-01 4.67334151e-01
8.62477660e-01 -7.04288661e-01 -2.76285827e-01 -7.15395689e-01
8.58551502e-01 1.80151567e-01 2.31960148e-01 -9.00706947e-01
2.70632833e-01 6.22625165e-02 4.75393608e-02 -6.38942897e-01
3.00936546e-04 -7.43955910e-01 -2.76524127e-01 5.75751185e-01
-8.56922790e-02 6.34983838e-01 1.92675456e-01 -5.97415268e-02
-8.91196430e-01 -9.47160900e-01 2.51311809e-01 -6.35202825e-01
-9.07444537e-01 -1.66579604e-01 -7.17122316e-01 1.01928830e+00
4.57953215e-01 1.15308486e-01 -3.57956946e-01 -7.47485697e-01
-5.47651112e-01 5.40989399e-01 8.92020464e-02 7.16457069e-02
6.73583686e-01 -1.02959740e+00 -1.31067240e+00 -2.27638319e-01
4.75679666e-01 -1.69934571e-01 5.97996973e-02 1.11690117e-02
-1.32754040e+00 9.01852906e-01 -2.63029695e-01 -4.42793258e-02
-1.16294575e+00 -2.29641218e-02 2.29381055e-01 -7.74348021e-01
-1.42098173e-01 6.70951247e-01 -4.41625208e-01 -1.42251301e+00
-4.96957362e-01 -1.06778532e-01 -6.38421357e-01 2.26042122e-01
3.61842126e-01 1.67480856e-01 6.07787848e-01 -8.66095006e-01
-2.87436575e-01 3.98058861e-01 1.27811387e-01 -5.88763237e-01
1.20697773e+00 -2.12407112e-01 -3.07536274e-01 1.73187062e-01
7.54339099e-01 3.21449548e-01 -3.06454241e-01 -1.47039235e-01
3.42506260e-01 -2.47500569e-01 -4.69201833e-01 -1.30414355e+00
-4.28392112e-01 9.25596893e-01 4.23241973e-01 1.92431062e-01
9.26435649e-01 2.36273661e-01 1.14471114e+00 1.06056452e+00
4.24620956e-01 -1.11013293e+00 -7.38308057e-02 1.18962443e+00
9.89301801e-01 -1.26588559e+00 -2.86385357e-01 -3.24473679e-01
-6.14896178e-01 1.36093354e+00 3.86048853e-01 1.87473223e-01
7.53549933e-02 -1.26903430e-01 9.59137976e-01 -7.01297343e-01
-3.66414905e-01 -5.17450869e-01 5.61292768e-01 9.14561808e-01
4.90846932e-01 2.91900896e-03 -7.63017952e-01 5.02843380e-01
-9.84483063e-01 -5.16105652e-01 6.99705005e-01 9.73804951e-01
-5.21690905e-01 -1.57032311e+00 -5.11543095e-01 3.00419152e-01
-5.44840634e-01 -5.08406162e-01 -8.62103462e-01 1.07935500e+00
4.02884223e-02 1.46754766e+00 -4.92984623e-01 6.05070442e-02
4.42202866e-01 3.79392028e-01 4.25734282e-01 -7.88167059e-01
-1.27336740e+00 -8.37457001e-01 8.14709902e-01 1.70307860e-01
-3.10912102e-01 -4.94446456e-01 -1.15482426e+00 -1.24630071e-01
-2.30095029e-01 9.42098737e-01 7.01492131e-01 1.17111158e+00
-5.30878492e-02 -5.75236343e-02 2.76742458e-01 2.77103037e-01
-4.52842087e-01 -9.77576435e-01 -1.88021138e-01 5.49513280e-01
2.59411074e-02 1.30394697e-01 -7.89088011e-02 3.40867192e-01] | [11.357394218444824, 8.145805358886719] |
0b665820-a4dc-48f8-a4eb-7fa2aa73acd5 | lifelong-unsupervised-domain-adaptive-person | 2112.06632 | null | https://arxiv.org/abs/2112.06632v2 | https://arxiv.org/pdf/2112.06632v2.pdf | Lifelong Unsupervised Domain Adaptive Person Re-identification with Coordinated Anti-forgetting and Adaptation | Unsupervised domain adaptive person re-identification (ReID) has been extensively investigated to mitigate the adverse effects of domain gaps. Those works assume the target domain data can be accessible all at once. However, for the real-world streaming data, this hinders the timely adaptation to changing data statistics and sufficient exploitation of increasing samples. In this paper, to address more practical scenarios, we propose a new task, Lifelong Unsupervised Domain Adaptive (LUDA) person ReID. This is challenging because it requires the model to continuously adapt to unlabeled data in the target environments while alleviating catastrophic forgetting for such a fine-grained person retrieval task. We design an effective scheme for this task, dubbed CLUDA-ReID, where the anti-forgetting is harmoniously coordinated with the adaptation. Specifically, a meta-based Coordinated Data Replay strategy is proposed to replay old data and update the network with a coordinated optimization direction for both adaptation and memorization. Moreover, we propose Relational Consistency Learning for old knowledge distillation/inheritance in line with the objective of retrieval-based tasks. We set up two evaluation settings to simulate the practical application scenarios. Extensive experiments demonstrate the effectiveness of our CLUDA-ReID for both scenarios with stationary target streams and scenarios with dynamic target streams. | ['Zheng-Jun Zha', 'Zicheng Liu', 'Jiang Wang', 'Quanzeng You', 'Peng Chu', 'Wenjun Zeng', 'Cuiling Lan', 'Zhizheng Zhang', 'Zhipeng Huang'] | 2021-12-13 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Huang_Lifelong_Unsupervised_Domain_Adaptive_Person_Re-Identification_With_Coordinated_Anti-Forgetting_and_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Huang_Lifelong_Unsupervised_Domain_Adaptive_Person_Re-Identification_With_Coordinated_Anti-Forgetting_and_CVPR_2022_paper.pdf | cvpr-2022-1 | ['person-retrieval'] | ['computer-vision'] | [-5.32058403e-02 -4.26251769e-01 6.09498192e-03 -3.92681032e-01
-3.34794343e-01 -1.34823546e-01 4.70269203e-01 2.66531557e-01
-7.74960101e-01 9.10560668e-01 2.14425042e-01 3.80153090e-01
-4.71445352e-01 -7.35244334e-01 -4.88801718e-01 -7.73994446e-01
3.70849855e-02 8.02759528e-01 3.21108341e-01 -2.27750406e-01
-2.08873842e-02 2.40778148e-01 -1.59436858e+00 -1.64021209e-01
1.16549659e+00 5.43532908e-01 2.16634169e-01 5.33971310e-01
-1.70616835e-01 3.71529758e-01 -6.87092245e-01 -4.22123969e-01
2.67355949e-01 -1.85504094e-01 -5.00364661e-01 1.90021798e-01
1.58921629e-01 -5.18275023e-01 -5.13517737e-01 8.23099434e-01
1.00891662e+00 6.20281994e-01 4.78526205e-01 -1.45412946e+00
-6.66049063e-01 6.02507710e-01 -7.08893538e-01 6.14503860e-01
2.20750809e-01 1.75345466e-01 1.70172095e-01 -7.46244788e-01
3.86342078e-01 1.39169466e+00 5.75001478e-01 8.79646301e-01
-8.75758767e-01 -9.60103452e-01 7.92077422e-01 5.68326652e-01
-1.59920192e+00 -7.08614826e-01 7.13298738e-01 -2.15329334e-01
4.57115442e-01 -1.52290910e-01 5.95718980e-01 1.18113518e+00
-2.40501493e-01 7.66663671e-01 6.21236324e-01 -2.20796660e-01
3.94860357e-01 3.01503986e-01 4.45837647e-01 1.58058405e-01
6.23795748e-01 1.98618062e-02 -9.49584126e-01 -1.29794389e-01
4.36584473e-01 3.02647084e-01 -7.44620636e-02 -2.59937406e-01
-1.06452394e+00 2.25394741e-01 -1.12298168e-02 8.26578140e-02
-3.64900440e-01 -3.28419298e-01 5.92666149e-01 6.52587652e-01
5.16863406e-01 -1.74094558e-01 -4.39912498e-01 -9.62579474e-02
-7.35801816e-01 5.82787156e-01 6.40841603e-01 1.27135849e+00
5.06139934e-01 1.58020139e-01 -2.99097687e-01 1.07388222e+00
1.34643108e-01 6.12665176e-01 8.74676883e-01 -4.38104957e-01
5.92083931e-01 5.19136310e-01 3.36697400e-01 -9.48693633e-01
-4.63778526e-01 -6.79444790e-01 -1.18793774e+00 -4.63165820e-01
4.57052737e-01 -1.36173502e-01 -6.65421069e-01 2.07682443e+00
7.77251780e-01 7.26720929e-01 2.20434561e-01 8.49433780e-01
5.10700345e-01 4.77205455e-01 4.63287920e-01 -6.53027236e-01
1.09708655e+00 -6.73336267e-01 -8.13259125e-01 -1.35980785e-01
2.02842414e-01 -3.84477466e-01 1.00632119e+00 3.52464825e-01
-9.31210697e-01 -8.25460136e-01 -1.08069062e+00 3.28747779e-01
-3.35736394e-01 -2.21723825e-01 1.84913605e-01 5.81771374e-01
-6.29632056e-01 2.41179585e-01 -5.90960979e-01 -7.50261128e-01
1.86992303e-01 3.81603569e-01 -2.05679268e-01 -2.13666186e-01
-1.54462755e+00 5.02552271e-01 7.68181503e-01 1.25307187e-01
-7.75724173e-01 -7.65049160e-01 -4.87856537e-01 -3.13045382e-02
5.75460792e-01 -9.17451918e-01 1.13286126e+00 -7.81384766e-01
-1.27654409e+00 5.46849430e-01 -2.26423129e-01 -5.24087608e-01
7.08127022e-01 -4.82980132e-01 -8.32863867e-01 -2.74493814e-01
1.10371538e-01 2.33985245e-01 1.03547275e+00 -1.24683785e+00
-8.63445997e-01 -6.65474117e-01 -3.69068086e-01 5.83934367e-01
-1.00076687e+00 -2.93003440e-01 -6.88101232e-01 -9.42528009e-01
-7.98241496e-02 -6.35033250e-01 7.95927346e-02 -3.15430284e-01
-6.24783747e-02 -3.72770786e-01 8.34992528e-01 -6.59981430e-01
1.56168926e+00 -2.08450437e+00 3.82893942e-02 2.76276588e-01
4.20923121e-02 3.60432804e-01 -2.20621631e-01 3.52756441e-01
8.49001408e-02 -3.19362313e-01 -1.88130081e-01 -6.77455604e-01
-2.18073115e-01 2.53171682e-01 -2.92810827e-01 4.34004456e-01
-2.42948189e-01 3.81578863e-01 -9.41370368e-01 -6.42054677e-01
-1.04333699e-01 2.06823498e-01 -3.30307245e-01 4.48872149e-01
-3.05916611e-02 6.29879892e-01 -5.14820278e-01 6.15162849e-01
9.94651914e-01 6.46339431e-02 1.11346496e-02 1.21415459e-01
1.35590315e-01 -3.91975999e-01 -1.71363533e+00 1.71289313e+00
-1.86096698e-01 -5.11291288e-02 3.77950771e-03 -1.00705326e+00
9.17846620e-01 -4.01306152e-03 5.20849228e-01 -9.37095642e-01
-1.38655066e-01 -5.52835874e-02 -2.48098776e-01 -5.25179148e-01
6.90222740e-01 1.56281829e-01 -7.27602914e-02 5.35030603e-01
-7.98217356e-02 6.59539104e-01 1.12050541e-01 4.53671724e-01
9.41733956e-01 6.92879781e-02 1.52393267e-01 -3.09657548e-02
7.62972653e-01 -1.19822145e-01 1.03711987e+00 1.10053337e+00
-5.12271523e-01 2.97078937e-01 -2.40940973e-01 -4.18107152e-01
-9.43447590e-01 -1.09819257e+00 2.26456970e-01 1.38918817e+00
7.00338066e-01 -2.65473425e-01 -6.66210234e-01 -6.57640576e-01
1.98144704e-01 5.31832814e-01 -2.58082271e-01 -6.16870642e-01
-7.11830795e-01 -9.11752343e-01 4.84934479e-01 3.22471589e-01
1.06148767e+00 -8.87886763e-01 -2.75667816e-01 4.45272475e-01
-3.41616213e-01 -8.77362907e-01 -7.69672811e-01 -2.85069495e-01
-6.67773724e-01 -9.99358416e-01 -9.25363958e-01 -7.64749169e-01
6.32976174e-01 5.58499455e-01 7.64571846e-01 3.36039588e-02
-4.80678789e-02 6.45713687e-01 -4.28643018e-01 -4.23324287e-01
-2.10555673e-01 3.31248641e-01 7.60132551e-01 3.69389355e-01
5.72796643e-01 -6.78034544e-01 -6.82916462e-01 4.38167661e-01
-1.13106346e+00 -1.71098232e-01 4.47999060e-01 1.01552999e+00
4.47064936e-01 4.23665345e-01 1.31484139e+00 -8.16820562e-01
9.56863523e-01 -8.16138685e-01 -2.31759846e-01 5.97498000e-01
-9.06735241e-01 -3.17420885e-02 5.89246213e-01 -1.03187633e+00
-1.41914499e+00 -1.97799295e-01 1.67291492e-01 -4.53480095e-01
-1.33882985e-01 2.84548998e-01 -4.55276936e-01 2.85317510e-01
5.65248966e-01 6.10550463e-01 -9.64904800e-02 -5.35276651e-01
2.04263985e-01 9.44554687e-01 1.01558459e+00 -8.40763807e-01
1.02902424e+00 3.24584544e-01 -4.61084008e-01 -7.32484400e-01
-5.05885422e-01 -6.84563994e-01 -6.05095029e-01 -3.27381402e-01
3.08197379e-01 -1.07728207e+00 -7.49214172e-01 9.67110932e-01
-9.71414983e-01 -5.40992990e-02 -2.85779148e-01 2.82358974e-01
-2.88045019e-01 6.36732757e-01 -2.53477812e-01 -9.52364504e-01
-6.07555568e-01 -5.15362322e-01 5.45189619e-01 5.79437733e-01
-3.83513384e-02 -6.61243200e-01 1.74016163e-01 4.94634248e-02
4.89970297e-01 -7.68186301e-02 7.14361548e-01 -9.84242737e-01
-3.11063111e-01 4.39790860e-02 -1.43571138e-01 5.97217269e-02
3.02164882e-01 -7.01280594e-01 -8.01550031e-01 -7.43922591e-01
-6.83885142e-02 -1.67094454e-01 7.29119897e-01 -9.87272635e-02
9.94608462e-01 -4.66321826e-01 -4.12294507e-01 3.97977024e-01
9.99131858e-01 2.49974206e-01 2.59270519e-01 5.60829461e-01
4.89661932e-01 6.50537133e-01 1.05332363e+00 9.67452347e-01
7.77822793e-01 6.69022381e-01 -5.14842942e-02 3.84823769e-01
-1.52246386e-01 -5.02308905e-01 1.95555761e-01 9.02347922e-01
2.19015703e-01 -4.47482318e-01 -8.00609946e-01 7.60128140e-01
-2.32818055e+00 -8.93742919e-01 3.88462216e-01 2.57070470e+00
9.97818410e-01 1.11384206e-01 2.28598058e-01 1.18965760e-01
1.11159670e+00 -1.68392763e-01 -1.17755628e+00 3.22480857e-01
-2.29552060e-01 -2.83139855e-01 3.12414646e-01 1.53290987e-01
-9.64826286e-01 7.99265146e-01 5.20671511e+00 8.60163093e-01
-7.73363233e-01 4.47883755e-01 3.56802970e-01 -1.60468847e-01
9.01180580e-02 -2.59015054e-01 -1.00363171e+00 9.05525386e-01
8.00056100e-01 -4.67764586e-01 4.61533248e-01 6.55660331e-01
1.99387342e-01 -1.44576371e-01 -9.53499913e-01 1.37895048e+00
1.33461148e-01 -6.83814347e-01 2.59135157e-01 -4.75218594e-01
5.14309227e-01 -4.94254678e-01 -1.42439716e-02 5.31896412e-01
2.68494725e-01 -2.90313423e-01 5.56639791e-01 9.98560071e-01
6.62489951e-01 -9.23782349e-01 5.66141307e-01 6.73893511e-01
-1.19500613e+00 -4.09306407e-01 -4.60115254e-01 3.02381277e-01
1.99516460e-01 4.39138740e-01 -6.13283873e-01 6.96067274e-01
9.22442615e-01 7.33848691e-01 -7.04685152e-01 1.17683315e+00
3.24376792e-01 3.70541990e-01 -3.64741236e-01 3.31699938e-01
-4.64742631e-01 1.49986058e-01 7.72558868e-01 1.12716568e+00
3.18992823e-01 2.38847420e-01 3.23154300e-01 2.44535297e-01
8.04415569e-02 1.02063179e-01 -3.76465738e-01 3.15002739e-01
1.16181886e+00 7.14332879e-01 -2.17074350e-01 -4.88668710e-01
-9.00356472e-03 1.09117591e+00 3.64841163e-01 5.78283012e-01
-6.83057547e-01 -3.73024434e-01 5.41991115e-01 2.37016559e-01
-2.57545173e-01 -2.57158041e-01 1.03727281e-02 -1.31963277e+00
1.32202864e-01 -6.95194125e-01 1.04070532e+00 -3.99919480e-01
-1.80451059e+00 2.01363057e-01 2.91822881e-01 -1.12433076e+00
-1.32230386e-01 2.76833147e-01 -5.77689946e-01 6.55748844e-01
-1.70316350e+00 -1.03289413e+00 -6.83279991e-01 1.20110357e+00
7.48818338e-01 -4.68842447e-01 4.33973700e-01 7.46605158e-01
-8.63557696e-01 1.15318298e+00 2.10897729e-01 -2.08872750e-01
1.42878354e+00 -8.37938964e-01 1.63338959e-01 9.88409281e-01
-5.17153740e-01 7.23308384e-01 7.99379826e-01 -1.11155701e+00
-1.48762906e+00 -1.37820613e+00 5.67928791e-01 -1.63744856e-02
1.99899673e-01 -2.83336818e-01 -1.24865294e+00 3.80653441e-01
-1.29659250e-01 -2.37842306e-01 3.65030199e-01 4.35252972e-02
-1.89753950e-01 -7.20511973e-01 -1.39303517e+00 4.30525899e-01
1.42002261e+00 -1.09192595e-01 -5.83788574e-01 2.83280462e-01
8.76274765e-01 -3.20871621e-01 -6.45488739e-01 3.68886799e-01
4.18175370e-01 -6.22411132e-01 1.01626670e+00 -6.09526098e-01
-4.93401080e-01 -4.46068645e-01 1.67457402e-01 -1.26640344e+00
-2.77213424e-01 -6.62272513e-01 -4.47456837e-01 1.76397908e+00
-3.38592470e-01 -8.31827343e-01 6.20369136e-01 8.39551806e-01
2.24174589e-01 -1.06498197e-01 -9.68758881e-01 -9.53261971e-01
-3.27486277e-01 -5.11467569e-02 1.02094698e+00 8.84747148e-01
-2.23267272e-01 7.91047215e-02 -6.47738457e-01 4.46085781e-01
8.48910093e-01 -3.44371587e-01 1.10690153e+00 -1.27453995e+00
-1.93969101e-01 9.63943917e-03 -1.54540986e-01 -1.09630740e+00
5.87846749e-02 -5.45107961e-01 -7.04564080e-02 -1.08245420e+00
3.57777089e-01 -8.03453088e-01 -5.84825337e-01 2.51982629e-01
-4.04488236e-01 -4.63717431e-01 4.94599827e-02 6.32749319e-01
-9.86878514e-01 7.69529045e-01 7.69726515e-01 -2.86097139e-01
-6.38978958e-01 1.60167143e-01 -6.71467483e-01 2.88397700e-01
7.84899712e-01 -4.80743170e-01 -7.93578088e-01 -4.83810723e-01
-6.71691746e-02 5.93497939e-02 2.75693774e-01 -1.27987683e+00
8.76695156e-01 -2.79610962e-01 4.41368788e-01 -7.24280953e-01
1.71606332e-01 -7.83205152e-01 2.48427957e-01 2.56522357e-01
-3.17278564e-01 2.24307805e-01 7.65331462e-02 1.08553374e+00
-4.20503654e-02 2.22626291e-02 7.41553843e-01 -5.39653674e-02
-1.03124714e+00 6.98745310e-01 5.25112487e-02 9.80513170e-02
1.13763404e+00 -1.76785335e-01 -4.88280654e-01 -3.95749867e-01
-8.66873085e-01 9.27872300e-01 3.76075208e-01 6.69777453e-01
8.18209529e-01 -1.33664322e+00 -7.87830412e-01 3.12255025e-01
3.72087747e-01 1.91015750e-01 8.78151774e-01 4.48121279e-01
9.54307839e-02 9.13841184e-03 -1.38540432e-01 -3.80143285e-01
-1.13215530e+00 8.55258524e-01 2.55047679e-01 -2.15983704e-01
-5.22065520e-01 4.85008210e-01 1.22120883e-02 -2.37171516e-01
7.25453198e-01 3.21983159e-01 -4.04447794e-01 5.42986766e-02
8.40055466e-01 6.19269729e-01 -2.76791696e-02 -3.11613709e-01
-2.34915704e-01 2.80281812e-01 -4.76048082e-01 -9.00144782e-03
1.14008081e+00 -8.68375063e-01 6.44616783e-02 3.61432642e-01
5.00166833e-01 -3.20719630e-01 -1.30509269e+00 -9.03581440e-01
3.56228389e-02 -4.78648454e-01 -4.22263771e-01 -7.58286059e-01
-7.65622795e-01 3.87212723e-01 9.71820176e-01 -1.21781468e-01
1.29059410e+00 -3.63040775e-01 1.05846524e+00 6.70149267e-01
5.64329505e-01 -1.51043904e+00 1.56175494e-01 3.17180395e-01
5.77936411e-01 -1.12415099e+00 2.84978766e-02 -2.04820540e-02
-7.05028176e-01 7.19464839e-01 1.10267675e+00 1.63179964e-01
6.46028996e-01 -1.12176932e-01 -2.66020060e-01 2.10976541e-01
-7.66114116e-01 -1.96948916e-01 -1.32063895e-01 8.57373357e-01
-4.42690998e-01 -8.57119262e-02 -3.71403456e-01 9.67219353e-01
1.76147476e-01 2.14132845e-01 2.79844612e-01 1.06151628e+00
-4.28190649e-01 -1.16534948e+00 -5.00642657e-01 3.05514067e-01
-6.45289943e-03 2.85856903e-01 2.52917614e-02 5.72930217e-01
3.05657715e-01 1.02855325e+00 -3.50095034e-02 -3.30657899e-01
5.04894972e-01 2.43748620e-01 1.56184211e-01 -3.50400031e-01
-3.31508279e-01 -1.86504409e-01 -2.81562895e-01 -3.95710208e-02
-4.43224907e-01 -7.72162318e-01 -1.17198324e+00 -4.57201988e-01
-1.23612605e-01 2.08923072e-01 2.25883797e-01 8.61958086e-01
6.10529304e-01 3.39378923e-01 6.36547625e-01 -2.86342263e-01
-6.71291888e-01 -1.01175511e+00 -6.73686862e-01 5.65271378e-01
2.37624407e-01 -8.32571387e-01 5.36954813e-02 1.93371668e-01] | [14.772913932800293, 1.2338731288909912] |
2f7583eb-08b0-4018-8c0d-6ea5e3748a6d | styletrf-stylizing-tensorial-radiance-fields | 2212.09330 | null | https://arxiv.org/abs/2212.09330v1 | https://arxiv.org/pdf/2212.09330v1.pdf | StyleTRF: Stylizing Tensorial Radiance Fields | Stylized view generation of scenes captured casually using a camera has received much attention recently. The geometry and appearance of the scene are typically captured as neural point sets or neural radiance fields in the previous work. An image stylization method is used to stylize the captured appearance by training its network jointly or iteratively with the structure capture network. The state-of-the-art SNeRF method trains the NeRF and stylization network in an alternating manner. These methods have high training time and require joint optimization. In this work, we present StyleTRF, a compact, quick-to-optimize strategy for stylized view generation using TensoRF. The appearance part is fine-tuned using sparse stylized priors of a few views rendered using the TensoRF representation for a few iterations. Our method thus effectively decouples style-adaption from view capture and is much faster than the previous methods. We show state-of-the-art results on several scenes used for this purpose. | ['P. J. Narayanan', 'Saurabh Saini', 'Sirikonda Dhawal', 'Rahul Goel'] | 2022-12-19 | null | null | null | null | ['image-stylization'] | ['computer-vision'] | [ 2.95605510e-01 4.36566845e-02 1.97662205e-01 -4.23021942e-01
-3.97049785e-01 -7.11784959e-01 7.32241273e-01 -8.47549796e-01
-6.21698871e-02 6.02033317e-01 1.53221473e-01 2.31780428e-02
3.35849166e-01 -6.71885252e-01 -1.04375911e+00 -5.72589219e-01
6.48573697e-01 4.75856692e-01 -3.58606130e-02 -1.15081236e-01
1.16797544e-01 7.84765303e-01 -1.40744758e+00 3.12822074e-01
6.55268729e-01 8.05595756e-01 3.27926725e-01 8.54415834e-01
-3.27280521e-01 7.38808990e-01 -4.23103333e-01 -6.44064844e-01
4.06292588e-01 -3.35832953e-01 -6.92172825e-01 7.68471658e-01
1.11716855e+00 -6.40792191e-01 -3.35190922e-01 8.66178691e-01
3.22710663e-01 2.33589470e-01 6.45173788e-01 -9.07250702e-01
-8.78524244e-01 3.98277104e-01 -8.87568533e-01 -3.73520881e-01
3.41141760e-01 5.72275883e-03 7.63063967e-01 -1.13996339e+00
8.83278131e-01 1.56136191e+00 3.93928677e-01 7.66068935e-01
-1.60245466e+00 -4.26345497e-01 5.34003556e-01 -3.25341970e-01
-9.02693093e-01 -3.07082891e-01 1.15115619e+00 -4.42895323e-01
6.69606924e-01 1.77776143e-01 8.35268080e-01 1.30349052e+00
1.22056171e-01 9.53594744e-01 1.03304148e+00 -4.20590609e-01
6.81228340e-02 -1.39848650e-01 -4.30170059e-01 9.58663881e-01
-8.57507586e-02 -2.28123404e-02 -5.85834205e-01 -1.67304307e-01
1.58235097e+00 3.42605144e-01 -3.79674643e-01 -6.71701670e-01
-1.30915308e+00 8.69505167e-01 4.92860854e-01 -1.04480073e-01
-2.97393262e-01 6.85822427e-01 5.96479475e-02 1.50242224e-01
8.14238310e-01 5.82897305e-01 -3.75263274e-01 1.76412389e-02
-1.05449605e+00 2.88468689e-01 5.81199586e-01 1.06484318e+00
8.42900038e-01 5.26608050e-01 -2.26077307e-02 9.70156968e-01
2.10647181e-01 7.63334870e-01 3.88634726e-02 -1.38297403e+00
2.53031433e-01 2.76309162e-01 1.97297484e-01 -7.64767647e-01
3.40352347e-03 -2.16024429e-01 -9.61493969e-01 4.79784697e-01
3.57324779e-01 -3.07011306e-01 -1.12985611e+00 1.50399733e+00
3.82098436e-01 2.33165532e-01 -2.46789202e-01 8.26549172e-01
8.27501357e-01 7.38433838e-01 -3.52576077e-01 7.86669552e-02
1.31220114e+00 -1.23469746e+00 -7.87021756e-01 -1.41220629e-01
2.22990531e-02 -1.18342221e+00 1.22467005e+00 6.64617121e-01
-1.42604482e+00 -6.13949001e-01 -7.38273799e-01 -3.02714646e-01
-3.43890153e-02 4.20500070e-01 7.66024172e-01 4.71996188e-01
-1.22335649e+00 7.70783782e-01 -6.70836508e-01 -3.83727878e-01
5.34560859e-01 1.49687633e-01 -4.01874721e-01 -1.07206404e-01
-6.27530456e-01 4.86149430e-01 -8.47845897e-02 6.92792833e-02
-9.52694416e-01 -7.13243008e-01 -9.08908665e-01 -1.60314724e-01
3.71994674e-01 -1.55995715e+00 1.20835638e+00 -1.27504098e+00
-2.32733250e+00 9.78604496e-01 -2.94353664e-01 -2.99524739e-02
5.52050173e-01 -7.14127243e-01 6.86357319e-02 1.71607137e-01
-6.76830485e-03 8.48592103e-01 1.46206045e+00 -1.83832443e+00
-2.10460678e-01 -1.82221770e-01 3.69199455e-01 3.50848526e-01
1.27624452e-01 -1.22675329e-01 -8.55457604e-01 -1.16127586e+00
1.18542939e-01 -1.06054294e+00 -3.10040027e-01 3.03527445e-01
-5.47466993e-01 2.27632999e-01 9.52963710e-01 -4.59067881e-01
7.43812501e-01 -1.76759219e+00 6.18332744e-01 -7.16156140e-02
3.53273571e-01 1.08644061e-01 -1.69210449e-01 3.61185551e-01
-3.36895853e-01 -1.02916948e-01 -2.09578082e-01 -8.87467325e-01
-1.80349365e-01 4.92673457e-01 -6.14002764e-01 3.49019438e-01
1.22575372e-01 8.88842702e-01 -9.54701066e-01 -2.01804563e-01
4.88752693e-01 5.80030739e-01 -8.52504492e-01 6.02787673e-01
-3.87799114e-01 6.82417214e-01 -2.84550577e-01 4.23709959e-01
7.50004649e-01 -2.97084510e-01 -1.53333813e-01 -7.07510948e-01
-4.59144376e-02 -2.90971071e-01 -1.15670204e+00 2.25561380e+00
-8.22087467e-01 4.42154467e-01 1.80125266e-01 -5.06651998e-01
7.22512484e-01 1.71250969e-01 3.93439829e-01 -2.14257427e-02
1.36005744e-01 -4.07458432e-02 -7.16907859e-01 -4.01412159e-01
6.99449658e-01 -2.95025885e-01 1.96874917e-01 6.61618888e-01
1.94675773e-01 -7.46449471e-01 -8.62689689e-02 2.58891344e-01
5.35336256e-01 9.99628782e-01 1.57645881e-01 -2.09891479e-02
3.81677657e-01 -3.63763750e-01 1.87988758e-01 4.71777290e-01
6.95406020e-01 1.12533438e+00 2.31009975e-01 -6.94204390e-01
-1.16107774e+00 -1.21295512e+00 1.93598881e-01 8.97512496e-01
-3.31026129e-02 -4.02345568e-01 -1.06667054e+00 -6.84780359e-01
-2.38811195e-01 8.41952562e-01 -7.87258863e-01 1.61843032e-01
-6.63863301e-01 -3.20547581e-01 1.00016013e-01 5.83582461e-01
4.38865900e-01 -1.03770137e+00 -3.75479788e-01 2.66528297e-02
-8.74624699e-02 -1.25258088e+00 -9.21279073e-01 3.67970727e-02
-1.04743576e+00 -6.78355038e-01 -1.14449620e+00 -4.29114908e-01
1.16036463e+00 5.03895164e-01 1.42478561e+00 -1.36809975e-01
2.01364532e-02 6.21932626e-01 -1.63718686e-01 -1.73014000e-01
-2.47846708e-01 -3.15284729e-02 2.88917199e-02 5.52783012e-01
-4.36585456e-01 -9.57528591e-01 -5.57859600e-01 2.58035243e-01
-1.17768097e+00 5.37410975e-01 4.89008904e-01 8.94025505e-01
8.32329810e-01 -4.77311939e-01 -2.87361890e-01 -1.26399374e+00
3.19454819e-01 1.91491425e-01 -7.91020870e-01 1.15148745e-01
-3.84704471e-02 4.19766337e-01 7.59602427e-01 -4.23888236e-01
-1.47495770e+00 3.22859138e-01 5.45919798e-02 -1.16602790e+00
-2.26356909e-01 -3.92375737e-02 -3.03364307e-01 -2.46739417e-01
4.36628878e-01 -2.49164943e-02 -9.83810425e-02 -7.62134135e-01
9.80240166e-01 3.98142077e-02 5.62838554e-01 -8.42212260e-01
1.28573573e+00 9.06664789e-01 1.33669630e-01 -7.85678327e-01
-1.21107948e+00 -1.83246329e-01 -9.49117541e-01 -3.14020455e-01
9.91206825e-01 -8.63209665e-01 -5.64578235e-01 6.22696996e-01
-1.54366279e+00 -6.07572436e-01 -5.32488942e-01 2.35401422e-01
-7.50787973e-01 4.40606952e-01 -4.71130371e-01 -4.83628660e-01
-3.37269127e-01 -1.18850195e+00 1.67532921e+00 3.36243063e-02
-5.96191958e-02 -1.09124637e+00 1.53676689e-01 3.56155813e-01
1.82945579e-01 5.01125991e-01 5.69597542e-01 5.33408761e-01
-8.26055646e-01 2.40288764e-01 -3.46460372e-01 3.32927257e-01
4.01471108e-01 1.91636816e-01 -1.28216827e+00 -1.84702858e-01
-1.54329780e-02 -8.29455182e-02 8.47793818e-01 6.13702059e-01
1.49132323e+00 -2.66185313e-01 -9.24297795e-02 1.30440354e+00
1.58768308e+00 -1.18007220e-01 5.88724792e-01 -4.70067859e-02
1.53149188e+00 5.12792885e-01 1.60953298e-01 3.21488112e-01
1.15587451e-01 8.26756895e-01 5.06648958e-01 -4.62742358e-01
-2.85592526e-01 -2.94341862e-01 3.29374939e-01 7.63679743e-01
-5.17533422e-01 -2.62829155e-01 -3.19324285e-01 2.64831305e-01
-1.78656709e+00 -1.04866314e+00 -9.56471562e-02 1.97657692e+00
6.14536583e-01 -1.68199241e-01 -1.74322147e-02 -3.91690016e-01
4.61828500e-01 4.45055306e-01 -5.15078247e-01 -5.20473421e-01
-9.03369561e-02 2.07592845e-01 4.16988999e-01 6.96838379e-01
-9.21091318e-01 1.16905594e+00 6.35042810e+00 7.46105731e-01
-1.18568194e+00 -1.36834914e-02 7.07250953e-01 -2.77415812e-01
-5.74862897e-01 3.40358121e-03 -5.71460187e-01 6.35195076e-02
2.29533434e-01 3.37659717e-01 8.96112978e-01 8.98826957e-01
2.65100062e-01 -5.44704795e-02 -1.16405749e+00 1.33015144e+00
4.01662111e-01 -1.49899220e+00 5.20216584e-01 1.07251957e-01
1.17347920e+00 -1.35117337e-01 2.24973172e-01 -2.17599139e-01
5.05607843e-01 -8.81490886e-01 9.45058703e-01 9.67277169e-01
1.14397633e+00 -6.50308490e-01 1.86169013e-01 -6.57157451e-02
-1.13945210e+00 4.07396913e-01 -4.00014818e-01 2.21340403e-01
6.12282217e-01 8.14859033e-01 -2.19610363e-01 6.28245890e-01
6.64915979e-01 8.34078372e-01 -4.40023512e-01 4.67600048e-01
-2.82844394e-01 3.87188286e-01 -2.34959155e-01 3.70362997e-01
3.82148415e-01 -6.35307372e-01 7.13783681e-01 9.55143809e-01
3.83682698e-01 -4.79391105e-02 1.22557841e-01 1.14776886e+00
-1.55227125e-01 -1.53735593e-01 -8.07547212e-01 2.73011953e-01
-3.03496152e-01 1.54404008e+00 -6.79480731e-01 -4.67283517e-01
-3.72780859e-01 1.42536950e+00 3.87628496e-01 6.32488132e-01
-8.46916080e-01 -1.00562014e-01 5.07675529e-01 2.54006982e-01
6.32550597e-01 -3.12361568e-01 -1.80542618e-01 -1.74151909e+00
-2.01604590e-01 -7.69027472e-01 -3.10144901e-01 -1.50319660e+00
-1.38859808e+00 8.77830863e-01 1.28934920e-01 -1.36253488e+00
-3.00196439e-01 -7.99754679e-01 -7.63496280e-01 9.82560635e-01
-1.31766903e+00 -1.58910787e+00 -5.78062832e-01 6.99162245e-01
8.01049709e-01 -1.26617566e-01 7.72150457e-01 -8.20334479e-02
-5.14068246e-01 2.92140722e-01 1.26116112e-01 -1.41716562e-02
8.71403635e-01 -1.56939757e+00 7.42246270e-01 7.47671843e-01
2.44962752e-01 6.43130779e-01 7.01569974e-01 -5.42402267e-01
-1.33456159e+00 -1.20365369e+00 3.09185743e-01 -6.40720487e-01
4.30171460e-01 -4.43285406e-01 -5.73365569e-01 7.30343103e-01
6.22738600e-01 -3.48216295e-02 3.90423000e-01 -9.46475565e-02
-2.61621207e-01 -1.40168115e-01 -8.59801471e-01 1.02447140e+00
1.24261725e+00 -5.50078332e-01 -3.70519042e-01 2.18916744e-01
5.85580111e-01 -6.97003901e-01 -6.06322169e-01 -1.09757401e-01
5.45313954e-01 -1.05293047e+00 1.02557147e+00 -4.30882841e-01
5.56518972e-01 -5.47776818e-01 -1.53166011e-01 -1.74953902e+00
-5.10313213e-01 -1.19232452e+00 -2.06951126e-01 1.11378288e+00
1.51775405e-01 -3.22636396e-01 8.07230771e-01 3.82663012e-01
-1.23031959e-01 -7.33532131e-01 -3.38376164e-01 -4.13400441e-01
-2.37718642e-01 -1.53211042e-01 6.01113260e-01 8.28565240e-01
-8.11790705e-01 7.44082928e-01 -7.53011346e-01 -1.45255193e-01
9.52793717e-01 4.53949213e-01 1.26862955e+00 -1.08806086e+00
-6.62913978e-01 -1.39369711e-01 1.42418385e-01 -1.44889903e+00
1.62058055e-01 -5.50803781e-01 -1.49177298e-01 -1.49334216e+00
1.33564457e-01 4.73614670e-02 4.91509199e-01 2.05678716e-01
-2.83912957e-01 3.15586120e-01 3.40799093e-01 1.22627258e-01
-1.30752712e-01 7.48671353e-01 1.98750103e+00 2.06561014e-02
-2.92205922e-02 -6.79856539e-02 -4.80771661e-01 1.07506692e+00
5.14456689e-01 -2.45209709e-01 -6.64383471e-01 -8.42268288e-01
3.41609597e-01 7.49967396e-02 4.45348799e-01 -7.38147318e-01
-1.69946015e-01 -2.29745090e-01 6.35624290e-01 -8.09810281e-01
7.94663787e-01 -8.32505405e-01 5.06021738e-01 -9.60200876e-02
-1.09313056e-01 3.60746920e-01 1.07244320e-01 6.93264723e-01
1.31358150e-02 -2.06619903e-01 9.45578933e-01 -5.11204600e-01
-3.14591259e-01 6.53637588e-01 -2.38494262e-01 -5.72979264e-03
5.43588638e-01 -3.59472841e-01 2.33138353e-01 -6.78124547e-01
-8.25751364e-01 -2.30637714e-01 7.83723533e-01 2.81620771e-01
7.18603015e-01 -1.57264698e+00 -5.35772443e-01 2.85210818e-01
-1.43919662e-01 4.34016138e-01 3.59551460e-01 4.29722369e-01
-8.37832034e-01 6.41568974e-02 -2.67123193e-01 -6.83406830e-01
-1.20909739e+00 6.61496460e-01 4.52428490e-01 -3.45148504e-01
-6.79913104e-01 9.30461824e-01 1.03224897e+00 -4.99702305e-01
-6.73719123e-02 -4.98107761e-01 -3.89497541e-02 -1.15430489e-01
3.14874530e-01 1.25258505e-01 -3.30344588e-01 -7.27164209e-01
2.41254002e-01 1.22460747e+00 -8.73569101e-02 -3.86857152e-01
1.44406724e+00 -1.31334633e-01 -1.79256603e-01 4.82613027e-01
1.17582798e+00 1.64269447e-01 -1.75173819e+00 -2.26284236e-01
-8.98091435e-01 -6.29175842e-01 2.45823070e-01 -4.13636237e-01
-1.43769860e+00 9.27049994e-01 2.14946046e-01 -7.89686292e-02
1.02185941e+00 -2.49754637e-01 8.53486478e-01 3.61923963e-01
2.29200408e-01 -9.46821928e-01 4.28845376e-01 5.01039922e-01
1.34134924e+00 -9.22770798e-01 1.77871332e-01 -5.65195203e-01
-7.40507066e-01 1.24973810e+00 5.85483313e-01 -4.81234968e-01
6.69823647e-01 1.45605683e-01 2.02263147e-01 -3.54399353e-01
-3.98435682e-01 1.23098783e-01 5.77439368e-01 5.36830485e-01
4.11745131e-01 -3.62916924e-02 4.19268638e-01 -2.03689653e-03
-3.10867459e-01 -2.04204410e-01 4.88801956e-01 5.04521906e-01
1.30496278e-01 -1.23906136e+00 -6.15401030e-01 2.09484950e-01
-3.63576025e-01 -2.62342185e-01 -5.52141249e-01 5.20456254e-01
-5.42241596e-02 4.68765527e-01 1.62125137e-02 -1.48482218e-01
4.55206573e-01 -2.33431861e-01 9.87881362e-01 -7.13827252e-01
-5.63725352e-01 5.90646088e-01 9.17242691e-02 -8.96985114e-01
-7.47827172e-01 -5.64316988e-01 -5.94976306e-01 -3.25062245e-01
-2.89266735e-01 -2.86614805e-01 3.46287787e-01 7.18965113e-01
1.84017763e-01 6.69799030e-01 7.83716917e-01 -1.70542371e+00
-6.91478103e-02 -7.18867719e-01 -5.52679896e-01 4.97292727e-01
2.66964495e-01 -7.85510898e-01 -1.58586144e-01 6.43532515e-01] | [9.32887077331543, -3.203643560409546] |
fa8152fa-5e5e-4dba-ba91-538fe1be36c7 | shadfa-0-1-the-iranian-movie-knowledge-graph | 2210.07822 | null | https://arxiv.org/abs/2210.07822v1 | https://arxiv.org/pdf/2210.07822v1.pdf | Shadfa 0.1: The Iranian Movie Knowledge Graph and Graph-Embedding-Based Recommender System | Movies are a great source of entertainment. However, the problem arises when one is trying to find the desired content within this vast amount of data which is significantly increasing every year. Recommender systems can provide appropriate algorithms to solve this problem. The content_based technique has found popularity due to the lack of available user data in most cases. Content_based recommender systems are based on the similarity of items' demographic information; Term Frequency _ Inverse Document Frequency (TF_IDF) and Knowledge Graph Embedding (KGE) are two approaches used to vectorize data to calculate these similarities. In this paper, we propose a weighted content_based movie RS by combining TF_IDF which is an appropriate approach for embedding textual data such as plot/description, and KGE which is used to embed named entities such as the director's name. The weights between features are determined using a Genetic algorithm. Additionally, the Iranian movies dataset is created by scraping data from movie_related websites. This dataset and the structure of the FarsBase KG are used to create the MovieFarsBase KG which is a component in the implementation process of the proposed content_based RS. Using precision, recall, and F1 score metrics, this study shows that the proposed approach outperforms the conventional approach that uses TF_IDF for embedding all attributes. | ['Mohammad-R. Akbarzadeh-T', 'Mohammad Karrabi', 'Hannane Ebrahimian', 'Hadi Kalamati', 'Rayhane Pouyan'] | 2022-10-14 | null | null | null | null | ['knowledge-graph-embedding'] | ['graphs'] | [-3.41826260e-01 -4.21832055e-01 -3.13519150e-01 -2.48958230e-01
-7.07786605e-02 -4.99890745e-01 3.49042356e-01 5.51110387e-01
-5.58452368e-01 5.06656408e-01 7.02770174e-01 4.33365665e-02
-7.66948640e-01 -1.08899641e+00 -1.40103519e-01 -4.50919271e-01
6.21544942e-02 2.81492174e-02 2.56520212e-01 -5.19158840e-01
1.07593977e+00 2.89266557e-01 -1.81392288e+00 3.08960855e-01
7.69396305e-01 9.48905170e-01 3.96850497e-01 3.42549682e-01
-7.67822564e-01 7.42158353e-01 -5.41389346e-01 -3.90089303e-01
2.47273013e-01 -3.13609809e-01 -4.64445293e-01 -1.40793815e-01
6.32440522e-02 -1.59437954e-01 -2.19657645e-01 7.61331975e-01
3.69791090e-01 8.92801523e-01 8.67529511e-01 -1.18737721e+00
-9.63887393e-01 5.58484018e-01 -4.02028352e-01 4.67777997e-01
5.36628664e-01 -6.64487362e-01 1.07033062e+00 -9.83242810e-01
7.54889369e-01 9.71114159e-01 4.68671203e-01 -5.52601367e-02
-5.65774858e-01 -3.02833021e-01 -2.25812614e-01 5.62377274e-01
-1.50177598e+00 1.75252840e-01 9.32663023e-01 -6.48087263e-01
1.01463187e+00 2.80815303e-01 6.46820605e-01 2.72297949e-01
2.68904001e-01 1.25498429e-01 9.43767488e-01 -5.96118689e-01
8.73572156e-02 8.26691866e-01 3.98053199e-01 4.48745430e-01
1.88433230e-01 -2.69523323e-01 -3.73125106e-01 -1.97246253e-01
3.64581257e-01 4.12469834e-01 -3.70234512e-02 5.37200645e-02
-7.82217920e-01 1.29780614e+00 4.43928659e-01 6.10228062e-01
-3.69496644e-01 -4.27125782e-01 5.36954641e-01 2.77246088e-01
2.98005730e-01 6.97220385e-01 -1.21269241e-01 -2.06346571e-01
-8.74613583e-01 1.91113159e-01 7.72377253e-01 8.02587211e-01
4.71520752e-01 7.53105581e-02 6.83857948e-02 9.97425616e-01
6.51954114e-01 2.19947517e-01 7.85255134e-01 -5.46885788e-01
2.32418075e-01 1.00054681e+00 -7.29684234e-02 -1.93822157e+00
-1.34765118e-01 -3.36510688e-01 -3.82137746e-01 -2.63834119e-01
-5.91168255e-02 -2.16724146e-02 -5.13734579e-01 1.19507182e+00
4.56116617e-01 -4.28676605e-04 -2.70938175e-03 7.33341575e-01
1.26930285e+00 1.12108397e+00 8.03099200e-02 -2.74498522e-01
1.37164330e+00 -8.16958189e-01 -9.18510735e-01 4.97233003e-01
4.83185381e-01 -1.07586873e+00 7.51925647e-01 4.02183950e-01
-5.86467147e-01 -6.79566860e-01 -9.09347177e-01 1.83271840e-01
-1.03699553e+00 3.36738229e-01 5.19904435e-01 7.16998041e-01
-8.60539913e-01 6.12212539e-01 7.84994513e-02 -6.67083144e-01
-1.93212375e-01 4.33459818e-01 -5.15886486e-01 2.13310705e-03
-1.35271418e+00 1.02319813e+00 5.42708576e-01 -4.32409465e-01
-7.51131400e-02 -3.14225048e-01 -6.32426739e-01 -1.12139350e-02
2.48293176e-01 -3.01226139e-01 1.08397372e-01 -8.88384044e-01
-1.12953007e+00 2.45318040e-01 2.83391267e-01 1.09602390e-02
-3.98811013e-01 1.87111333e-01 -1.03741193e+00 2.26702794e-01
-5.10258824e-02 6.52465671e-02 6.58511221e-01 -7.63674319e-01
-8.37416530e-01 -4.60851550e-01 2.67568707e-01 4.77650464e-01
-8.74555469e-01 4.21430379e-01 -2.63238400e-01 -8.46362531e-01
1.93835124e-01 -8.39349449e-01 -2.61786766e-02 -4.52694416e-01
-6.43801168e-02 -3.39904457e-01 7.21581459e-01 -9.33812737e-01
1.89778483e+00 -2.08968234e+00 2.10770935e-01 6.68396592e-01
-1.36357313e-02 2.08380073e-01 1.72389910e-01 1.05977452e+00
1.30139023e-01 1.41406059e-01 2.61292636e-01 5.45826256e-01
-1.17280520e-01 -9.80100781e-02 1.77801862e-01 9.09785479e-02
-4.43261981e-01 1.07284553e-01 -6.06965482e-01 -6.47768617e-01
2.85230458e-01 5.57642221e-01 -5.88573754e-01 2.29847804e-01
1.55806795e-01 -8.80893990e-02 -6.25821888e-01 3.83958548e-01
3.35998684e-01 -1.77464057e-02 1.42937586e-01 -6.79728627e-01
-2.76665837e-01 -1.88743435e-02 -1.52648556e+00 1.20599365e+00
-3.18194509e-01 4.30922836e-01 -4.27605122e-01 -1.08803022e+00
1.32087708e+00 3.18637788e-01 6.86678767e-01 -3.11567485e-01
5.57404876e-01 -1.83433499e-02 6.96681589e-02 -9.78248358e-01
9.53738630e-01 6.88567385e-02 -6.05467334e-02 4.03639227e-01
8.94401595e-02 2.70774871e-01 4.04611826e-01 4.77847308e-01
9.21150982e-01 -3.34558934e-02 4.58487004e-01 -1.87959537e-01
7.94427872e-01 2.68939972e-01 3.64700407e-01 1.04036026e-01
2.56060183e-01 1.75591007e-01 9.34550613e-02 -1.14649862e-01
-1.18086052e+00 -6.71788454e-01 -9.80634317e-02 9.14212584e-01
-4.44907211e-02 -7.91783631e-01 -4.75259423e-01 -5.63923180e-01
1.30495682e-01 7.77200341e-01 -6.42339528e-01 -1.75937429e-01
-1.26872323e-02 -4.42953855e-01 1.87938437e-02 2.13473439e-01
4.16556478e-01 -8.39641929e-01 -2.63829641e-02 3.40532929e-01
-8.25992040e-03 -3.62807721e-01 -3.85834038e-01 -3.98619592e-01
-7.86597431e-01 -7.67945349e-01 -4.58260566e-01 -6.72248006e-01
7.20778823e-01 4.25043881e-01 6.48230731e-01 9.08762589e-02
-2.47152135e-01 3.70426208e-01 -1.10970843e+00 -2.18479514e-01
-1.17635690e-01 3.35900746e-02 7.10647255e-02 5.17419018e-02
6.54409051e-01 -5.39280891e-01 -6.59496486e-01 2.81384557e-01
-1.09353518e+00 -5.33596754e-01 3.44060451e-01 5.91407895e-01
4.41565007e-01 6.64066315e-01 7.80358732e-01 -9.47520196e-01
1.16429484e+00 -1.12767494e+00 -2.52614528e-01 3.49022835e-01
-8.87353957e-01 -8.50034431e-02 7.48475015e-01 -3.55397671e-01
-7.75289714e-01 -3.68752241e-01 -1.82825536e-01 -1.30355805e-01
1.01886369e-01 9.45311487e-01 1.03642009e-01 -1.42014086e-01
4.52946991e-01 9.66892466e-02 4.84962836e-02 -5.93718588e-01
2.12926328e-01 1.11102521e+00 -2.12046519e-01 -2.03607351e-01
4.41838890e-01 -7.54892975e-02 -1.32908851e-01 -8.73199224e-01
-3.52145672e-01 -8.23547304e-01 -3.27237666e-01 -6.04914427e-01
9.59621847e-01 -4.85255599e-01 -6.25638485e-01 -2.33042538e-01
-5.88449478e-01 8.70067418e-01 3.87681387e-02 1.02725744e+00
-4.84898388e-02 3.45009089e-01 -3.08634043e-01 -8.19262624e-01
-4.01974201e-01 -8.88459980e-01 -6.56992123e-02 4.63991463e-01
-1.07486032e-01 -9.62405503e-01 5.03952429e-02 5.46574593e-01
6.49263740e-01 2.79392451e-01 1.19182301e+00 -9.51316893e-01
-1.48149192e-01 -3.51387292e-01 1.07747458e-01 3.71141076e-01
3.34229290e-01 2.00162485e-01 -7.45888203e-02 -1.11497946e-01
-4.35565673e-02 1.87695771e-01 3.54678750e-01 2.16957971e-01
8.05131912e-01 -6.28755510e-01 -1.05292097e-01 1.62111923e-01
1.93604946e+00 6.86589777e-01 5.83511591e-01 5.28508961e-01
6.87512815e-01 7.01950014e-01 1.10253239e+00 7.99750030e-01
4.80990261e-01 7.36095309e-01 3.85173671e-02 3.85674477e-01
1.41302511e-01 -3.14175814e-01 2.04882249e-01 1.46022081e+00
-3.49209577e-01 -2.65859812e-01 -4.26602274e-01 3.98804456e-01
-1.64431548e+00 -1.19538713e+00 -1.56738758e-01 2.18288708e+00
4.27811086e-01 -2.11544096e-01 3.02026331e-01 4.84918296e-01
6.17538989e-01 -5.03541194e-02 1.81034543e-02 -8.82086754e-01
1.65236384e-01 7.80480430e-02 5.95801890e-01 2.18377188e-01
-7.53889084e-01 6.66825354e-01 4.84366179e+00 9.75758851e-01
-9.95952487e-01 5.19062346e-03 1.51951984e-01 3.74656320e-02
-5.99323630e-01 1.39763147e-01 -8.04429770e-01 8.54589045e-01
9.87205863e-01 -4.85960335e-01 5.44267058e-01 9.84902978e-01
3.56870413e-01 -2.66461790e-01 -2.07295388e-01 8.72231185e-01
5.75974643e-01 -1.33764505e+00 1.14675872e-01 3.97055112e-02
6.03905976e-01 -6.52036011e-01 -2.00697891e-02 1.15048304e-01
1.32241294e-01 -5.79044819e-01 9.29962322e-02 6.38199568e-01
4.52458769e-01 -1.00796795e+00 8.38541031e-01 1.58484653e-01
-1.23936319e+00 -3.16169292e-01 -8.23410809e-01 8.75436664e-02
1.20266035e-01 3.30352813e-01 -7.66811252e-01 5.65745533e-01
6.50051653e-01 6.80833101e-01 -4.63360369e-01 1.17491138e+00
1.91861436e-01 4.37384397e-01 -4.97162342e-02 -5.56477726e-01
2.65725285e-01 -6.90488815e-01 4.21684861e-01 9.02643859e-01
8.54668379e-01 4.14232105e-01 -1.78416058e-01 3.54731798e-01
1.67847469e-01 1.19662809e+00 -7.41420865e-01 -1.90300614e-01
7.63229251e-01 1.26746106e+00 -7.83355236e-01 -1.34729683e-01
-5.11617303e-01 5.81695855e-01 5.82669973e-02 -6.95111081e-02
-8.16455841e-01 -8.62758994e-01 2.19759107e-01 4.85370576e-01
4.11097020e-01 -1.61472470e-01 3.82174015e-01 -7.05276906e-01
-4.04635221e-01 -8.56076837e-01 6.16850913e-01 -7.28318751e-01
-1.26642036e+00 8.29653919e-01 2.43550017e-01 -1.45289147e+00
8.13515857e-02 -3.43382597e-01 -3.10003579e-01 7.26128221e-01
-9.34103787e-01 -9.55011964e-01 -1.47516131e-01 7.53767073e-01
4.86441344e-01 -5.75954974e-01 8.64509821e-01 6.62075937e-01
-3.38038951e-01 4.73142415e-01 7.36123145e-01 -7.71110430e-02
6.43753648e-01 -9.10591006e-01 -6.02007508e-01 5.98944128e-01
1.80522472e-01 9.22836244e-01 6.95992053e-01 -9.03814375e-01
-1.48283958e+00 -5.98629653e-01 1.11901903e+00 5.83506525e-02
5.96140444e-01 2.49662176e-01 -6.19065166e-01 2.35774204e-01
2.25899175e-01 -4.38267052e-01 1.36087394e+00 8.74219462e-03
-1.70764044e-01 -3.85338545e-01 -1.56719017e+00 2.80588984e-01
4.54341888e-01 -1.41393960e-01 -6.90180004e-01 3.51383716e-01
4.82580245e-01 3.02803457e-01 -1.56471658e+00 -1.27955124e-01
5.53221762e-01 -8.31721902e-01 9.07003462e-01 -6.02991343e-01
4.57734019e-01 -5.26789963e-01 -4.09974575e-01 -1.34905958e+00
-6.99412167e-01 1.42766923e-01 1.81905702e-01 1.61975002e+00
5.26267469e-01 -4.71362591e-01 5.58646381e-01 7.04257727e-01
1.92701638e-01 -6.83089316e-01 -4.26344156e-01 -4.46874678e-01
-5.16781509e-01 1.37530193e-01 5.72258353e-01 1.18781078e+00
2.00282782e-01 3.95312250e-01 -4.99181241e-01 -2.34522745e-01
3.61510366e-01 2.70670317e-02 4.27082151e-01 -1.17210901e+00
-2.82727003e-01 -2.41172779e-02 -7.57672191e-01 -2.77900279e-01
-3.55572045e-01 -1.03684521e+00 -6.21647596e-01 -1.72671056e+00
1.24202348e-01 -7.22023308e-01 -4.38730776e-01 -1.62714228e-01
6.65756539e-02 1.05936952e-01 3.31243068e-01 2.94993341e-01
-1.74092367e-01 1.94290757e-01 1.01862347e+00 1.46101490e-01
-2.99131513e-01 -4.37702350e-02 -6.68068409e-01 3.93332839e-01
8.14630091e-01 -5.23925960e-01 -7.80270576e-01 -1.96602479e-01
6.44724548e-01 9.48540345e-02 -6.18839622e-01 -8.08219910e-01
3.02806973e-01 -3.48164976e-01 3.45830768e-01 -5.71088493e-01
4.27845836e-01 -1.21317291e+00 6.35024250e-01 -1.81373581e-02
-5.35696149e-01 6.07387304e-01 -1.55600980e-01 5.50976098e-01
-3.21976393e-01 -7.57632434e-01 4.31097209e-01 -1.29515052e-01
-9.65027452e-01 2.53865749e-01 -4.56727773e-01 -3.15495253e-01
1.16604626e+00 -5.44753730e-01 -1.71440780e-01 -4.39758658e-01
-6.95501864e-01 -3.00943255e-01 3.57406855e-01 5.82550764e-01
8.81252348e-01 -1.53647017e+00 -5.51128983e-01 -2.57419348e-01
1.28462359e-01 -1.04930365e+00 5.51692128e-01 6.18284047e-01
-6.76265121e-01 2.99881637e-01 -6.30539536e-01 3.69706839e-01
-1.66198719e+00 7.57836878e-01 -4.84461576e-01 -9.88800675e-02
-2.03405902e-01 6.60727143e-01 -5.13753653e-01 -1.16468549e-01
-1.11488216e-01 2.77064145e-01 -1.44233155e+00 4.04580832e-01
4.12326664e-01 7.15108871e-01 -2.23966241e-01 -1.15015841e+00
-2.53840685e-01 7.63797224e-01 -9.72327888e-02 -1.96919262e-01
1.48341835e+00 -2.10793629e-01 -3.36390167e-01 2.18284085e-01
1.36952448e+00 2.87214369e-01 -1.21790335e-01 -2.04020664e-01
-2.10098643e-02 -9.56196070e-01 3.28264028e-01 -6.52777255e-01
-1.21789932e+00 4.78666753e-01 5.73483407e-01 2.99838781e-01
1.12021160e+00 -3.32676589e-01 7.75284410e-01 8.80893022e-02
2.59751260e-01 -1.60787904e+00 -8.23862851e-02 1.65076211e-01
7.71812439e-01 -9.38362062e-01 3.42457533e-01 -4.16504264e-01
-8.49397361e-01 1.18310666e+00 4.67256129e-01 -1.89998567e-01
1.26379979e+00 -1.44339964e-01 -2.86791384e-01 -8.89113247e-02
-4.93989497e-01 -8.55737701e-02 5.52945614e-01 3.67425919e-01
6.71093583e-01 1.15712769e-02 -1.38497615e+00 9.14032102e-01
-2.56745815e-01 -1.45692423e-01 5.54132044e-01 9.25928950e-01
-7.78554022e-01 -1.36235762e+00 -2.29295209e-01 9.94804382e-01
-7.41683781e-01 -1.86687738e-01 -1.93901926e-01 5.51043034e-01
3.86789680e-01 1.24060690e+00 -2.51843542e-01 -1.01417601e+00
8.26105848e-02 -2.41630405e-01 2.04781622e-01 -5.76144695e-01
-8.31247985e-01 -3.12046617e-01 3.32151800e-01 -9.20285583e-02
-4.66499686e-01 -4.22848850e-01 -1.18292284e+00 -5.90694487e-01
-6.93202674e-01 7.50520349e-01 1.34013915e+00 6.31568432e-01
3.84376496e-01 2.14230999e-01 1.08781433e+00 -1.35723740e-01
-2.50144452e-01 -8.63267004e-01 -7.81291008e-01 5.94063103e-01
-6.43119752e-01 -9.04650807e-01 -3.42515230e-01 -4.08651493e-02] | [10.118931770324707, 5.911219120025635] |
f4b902db-26ea-47d3-b129-76a330cb37e8 | fdn-finite-difference-network-with | 2108.07974 | null | https://arxiv.org/abs/2108.07974v2 | https://arxiv.org/pdf/2108.07974v2.pdf | FDN: Finite Difference Network with Hierarchical Convolutional Features for Text-independent Speaker Verification | In recent years, using raw waveforms as input for deep networks has been widely explored for the speaker verification system. For example, RawNet and RawNet2 extracted speaker's feature embeddings from waveforms automatically for recognizing their voice, which can vastly reduce the front-end computation and obtain state-of-the-art performance. However, these models do not consider the speaker's high-level behavioral features, such as intonation, indicating each speaker's universal style, rhythm, \textit{etc}. This paper presents a novel network that can handle the intonation information by computing the finite difference of different speakers' utterance variations. Furthermore, a hierarchical way is also designed to enhance the intonation property from coarse to fine to improve the system accuracy. The high-level intonation features are then fused with the low-level embedding features. Experimental results on official VoxCeleb1 test data, VoxCeleb1-E, and VoxCeleb-H protocols show our method outperforms and robustness existing state-of-the-art systems. To facilitate further research, code is available at https://github.com/happyjin/FDN | ['Lan Wang', 'Nan Yan', 'Jin Li'] | 2021-08-18 | null | null | null | null | ['text-independent-speaker-verification'] | ['speech'] | [-5.55849075e-01 -3.81443918e-01 -1.15074642e-01 -7.76611090e-01
-6.32600307e-01 -3.74846965e-01 2.54234493e-01 -3.02093059e-01
-2.74060786e-01 7.34933689e-02 5.61979592e-01 -2.95305878e-01
2.68935055e-01 -4.49512422e-01 -1.73269898e-01 -6.62225306e-01
-2.30891742e-02 -1.76804438e-01 -1.19135432e-01 -4.07006860e-01
-5.35898656e-02 5.01242936e-01 -1.50377977e+00 1.13243155e-01
4.55324113e-01 1.26116645e+00 -2.71807492e-01 7.26740181e-01
-1.34225219e-01 3.77967268e-01 -7.35921323e-01 -4.16428715e-01
3.82032664e-03 -4.63434517e-01 -1.95192471e-01 -3.58921289e-01
3.11939389e-01 -2.53696769e-01 -7.27667153e-01 1.21195757e+00
1.13328731e+00 -7.12532923e-02 3.18622112e-01 -1.46197343e+00
-6.93502784e-01 1.03163910e+00 -2.24047914e-01 3.41337949e-01
1.49698481e-01 4.08173561e-01 1.15237427e+00 -1.08493257e+00
1.10734917e-01 1.38655782e+00 7.36958802e-01 7.73733437e-01
-6.75172508e-01 -1.40882373e+00 2.58767959e-02 6.21569633e-01
-1.71989262e+00 -1.02591217e+00 1.27209198e+00 -1.90503910e-01
6.35170817e-01 4.89381015e-01 4.97678787e-01 1.05356503e+00
-2.65298545e-01 8.44597340e-01 6.86763287e-01 -8.27773139e-02
2.98770387e-02 1.65821835e-01 4.02534127e-01 4.53972340e-01
-2.80685484e-01 2.52002060e-01 -8.17596674e-01 -1.36230337e-02
2.70829618e-01 -2.87213683e-01 -5.12613714e-01 4.79658782e-01
-9.57277536e-01 7.18171358e-01 3.52981240e-01 6.14535630e-01
-2.58609891e-01 -1.89157669e-03 6.84077442e-01 3.62107962e-01
2.76498288e-01 -1.86597690e-01 -3.99742067e-01 -5.20029426e-01
-9.76842821e-01 4.17086342e-03 7.79351115e-01 7.38161445e-01
6.76894844e-01 7.84690499e-01 7.62409791e-02 1.11943805e+00
6.47543788e-01 6.75523221e-01 8.64121497e-01 -4.51834470e-01
4.90306497e-01 2.81477869e-01 -4.55196321e-01 -1.10209382e+00
-2.99070388e-01 -4.56282645e-01 -1.05441427e+00 -2.70644516e-01
-7.15890452e-02 -2.16958791e-01 -9.48768437e-01 1.99578714e+00
2.64391214e-01 3.47754925e-01 1.14699028e-01 9.71576333e-01
1.47669935e+00 9.07047212e-01 -1.84794575e-01 -2.06183046e-01
1.57563114e+00 -7.00821221e-01 -1.03062248e+00 -8.42062980e-02
3.15189928e-01 -1.02091026e+00 1.00286388e+00 3.13907087e-01
-7.94147730e-01 -8.00504267e-01 -1.09755111e+00 1.68320701e-01
-2.99942940e-01 2.30228111e-01 3.19149137e-01 1.24646497e+00
-1.10353994e+00 1.14923127e-01 -6.22338772e-01 5.80812581e-02
2.67908692e-01 3.30811292e-01 -2.84625649e-01 3.06193203e-01
-1.75219381e+00 6.08335555e-01 9.26795006e-02 6.15766764e-01
-9.12959039e-01 -7.58939087e-01 -1.14066887e+00 1.50075302e-01
-1.10276364e-01 1.13248520e-01 1.34109545e+00 -5.78716159e-01
-1.92280269e+00 5.18000007e-01 -3.21670383e-01 -2.28308111e-01
1.75648600e-01 2.21371621e-01 -1.15260267e+00 -1.53217420e-01
-3.62718880e-01 5.00925124e-01 7.53839076e-01 -7.42748618e-01
-2.45485365e-01 -2.03585878e-01 -3.76571506e-01 -1.33814812e-01
-6.21724069e-01 4.26988572e-01 -4.43785608e-01 -6.06214285e-01
3.50474268e-01 -7.98375070e-01 1.71147734e-01 -2.09713280e-01
-5.50647736e-01 -6.30361497e-01 8.96685243e-01 -8.94428015e-01
1.34744298e+00 -2.64960051e+00 -2.51622289e-01 3.35961044e-01
9.89602655e-02 5.43834567e-01 -3.10783178e-01 2.73951679e-01
-9.43922400e-02 2.36810967e-01 7.53674209e-02 -3.32257867e-01
3.38573992e-01 -8.36326405e-02 -3.01017493e-01 7.08701611e-01
8.02351460e-02 6.59924805e-01 -2.46575341e-01 -4.56212968e-01
2.15626597e-01 8.79592180e-01 -5.33204734e-01 1.97335675e-01
4.20539796e-01 1.95786655e-01 -1.98756292e-01 6.12919092e-01
1.03949332e+00 5.26687562e-01 -1.62422210e-01 -3.53795201e-01
-1.56642556e-01 7.80878067e-01 -1.31871104e+00 1.33333385e+00
-5.43500781e-01 9.04164493e-01 3.05824399e-01 -5.84415555e-01
1.36033344e+00 6.88059628e-01 1.28947601e-01 -5.39707005e-01
4.91853237e-01 2.57912964e-01 4.81525928e-01 -4.13420051e-01
4.15250808e-01 -1.97610542e-01 -2.84922481e-01 3.83574784e-01
1.50728762e-01 -2.36767232e-02 -2.69335568e-01 4.54606116e-02
6.22855961e-01 -7.58815408e-01 1.38340354e-01 -1.28166243e-01
8.47011089e-01 -8.58672202e-01 9.65854347e-01 2.76718855e-01
-6.55010700e-01 5.48429132e-01 2.84698039e-01 -7.38603845e-02
-6.62887692e-01 -1.07155609e+00 -3.04227173e-01 1.05763090e+00
3.86157050e-03 -3.89109671e-01 -7.91519582e-01 -6.23416007e-02
-8.08876604e-02 5.78515172e-01 -5.61052740e-01 -2.69071102e-01
-5.81494927e-01 -4.02880520e-01 1.42128706e+00 5.46555936e-01
5.80774665e-01 -1.23687673e+00 1.55503787e-02 1.46342024e-01
-3.04875165e-01 -1.08323538e+00 -1.11803746e+00 -2.34024242e-01
-2.78570294e-01 -4.44108009e-01 -6.35365665e-01 -9.44475114e-01
1.23471975e-01 3.75330681e-04 5.77877998e-01 1.83867201e-01
-1.39202431e-01 -2.80917957e-02 -3.37184817e-01 -5.01143336e-01
-3.78397822e-01 1.14351846e-01 5.83268285e-01 3.93747747e-01
6.20580673e-01 -6.56666696e-01 -4.39552844e-01 3.89060706e-01
-6.55018628e-01 -4.36984479e-01 3.22623163e-01 7.18628049e-01
-4.78804968e-02 -2.80383341e-02 9.37984526e-01 -1.86932907e-01
7.63995588e-01 -1.97478935e-01 -2.49254897e-01 -1.24310993e-01
-2.22438857e-01 -2.84497738e-01 5.71317136e-01 -7.35438049e-01
-9.59398270e-01 -3.76041114e-01 -9.60394442e-01 -4.30296659e-01
-1.63100407e-01 4.15003836e-01 -7.04678059e-01 5.02933934e-02
2.62147158e-01 6.17636323e-01 1.07418955e-03 -5.22765100e-01
2.93833077e-01 1.43844736e+00 6.49755180e-01 -2.10437953e-01
8.95821095e-01 -1.08174374e-02 -8.07296991e-01 -1.18915248e+00
-3.52405578e-01 -4.92701590e-01 -3.41800511e-01 -2.55276293e-01
6.16505563e-01 -9.79648709e-01 -1.24218690e+00 8.93051445e-01
-1.34659624e+00 1.20846704e-01 1.10770002e-01 8.05223405e-01
1.75200764e-03 2.67091185e-01 -9.76253688e-01 -9.86507237e-01
-5.90441227e-01 -1.27298665e+00 6.14902377e-01 4.78654325e-01
-1.44035727e-01 -6.78355277e-01 1.01711981e-01 2.08349884e-01
7.43091464e-01 -4.27893639e-01 3.97741050e-01 -8.70271266e-01
-1.76276818e-01 -1.19490221e-01 -6.04026839e-02 6.80545449e-01
1.38586581e-01 1.45762429e-01 -1.54906249e+00 -1.45046830e-01
2.65959024e-01 -6.92954510e-02 6.74807847e-01 1.91195726e-01
1.08568656e+00 -5.10999680e-01 1.97521523e-01 8.18911433e-01
7.61087298e-01 2.75446385e-01 7.61217237e-01 -1.58616349e-01
6.74905300e-01 3.62288475e-01 2.23128676e-01 2.81172603e-01
5.90789318e-01 6.05254769e-01 9.40272659e-02 -1.32945171e-02
-2.99434841e-01 -2.31909111e-01 7.81201959e-01 1.71613050e+00
1.87097877e-01 -1.74433202e-01 -7.15411782e-01 6.91658556e-01
-1.15369046e+00 -1.15037215e+00 3.23540419e-02 1.83702469e+00
1.01174140e+00 3.52624692e-02 7.90610164e-02 5.68049967e-01
1.05199897e+00 6.45027816e-01 -5.14203608e-01 -6.59289718e-01
-1.08677693e-01 1.26787454e-01 3.53037901e-02 4.97490853e-01
-7.65347183e-01 8.95423770e-01 5.66232538e+00 1.08419836e+00
-1.80737484e+00 3.21749181e-01 4.70247358e-01 -4.55343574e-02
-2.72890627e-01 -4.74763036e-01 -1.00147140e+00 6.43045783e-01
1.20091546e+00 -2.93219596e-01 3.52747738e-01 7.39397168e-01
3.05853665e-01 6.73597097e-01 -9.60184216e-01 1.25850022e+00
3.82556289e-01 -9.87074077e-01 -3.36173475e-01 3.25215459e-02
2.44471997e-01 1.58565879e-01 2.26021528e-01 6.10333145e-01
-1.09142728e-01 -1.00569773e+00 8.62105131e-01 2.72806622e-02
9.45333838e-01 -8.40810597e-01 7.58488774e-01 3.82792391e-02
-1.63732684e+00 1.51498333e-01 -2.65000343e-01 7.20856860e-02
3.22005987e-01 4.94175583e-01 -9.06090379e-01 2.96409160e-01
6.80382252e-01 4.67659265e-01 -2.55204111e-01 8.92644167e-01
-2.77259588e-01 1.15656877e+00 -4.25063640e-01 -3.20927799e-01
9.64940786e-02 3.14977199e-01 5.61633587e-01 1.50937176e+00
2.51559466e-01 1.32596463e-01 -1.74542144e-01 6.63686395e-01
-2.88708210e-01 2.24602237e-01 -1.70972094e-01 -7.71128833e-02
9.56821144e-01 1.22051489e+00 -1.79445431e-01 -2.72039592e-01
-1.02903828e-01 4.59627658e-01 -1.20833635e-01 2.82407254e-01
-9.20805991e-01 -8.60603631e-01 1.14975131e+00 -2.71678090e-01
2.31757641e-01 -2.49898598e-01 -1.97430566e-01 -1.03510630e+00
7.45332167e-02 -1.04831183e+00 5.80737479e-02 -4.30414289e-01
-1.26909363e+00 1.00829768e+00 -5.93563914e-01 -1.29556382e+00
-1.99519724e-01 -4.35119838e-01 -1.07227421e+00 1.02272213e+00
-1.39801478e+00 -1.08419454e+00 -2.30522335e-01 5.64610064e-01
6.10604465e-01 -4.41927552e-01 7.20654726e-01 8.01283479e-01
-7.71869779e-01 1.32365680e+00 -3.16328734e-01 8.02030444e-01
6.94976151e-01 -6.91949308e-01 6.67166889e-01 9.61907983e-01
2.19896227e-01 7.39233315e-01 6.04564071e-01 -8.38429481e-02
-1.30275679e+00 -9.03101444e-01 1.05078936e+00 1.16464794e-01
6.88612580e-01 -6.26482308e-01 -1.00528502e+00 4.91787016e-01
3.70567173e-01 1.08595528e-01 9.14212704e-01 2.25645468e-01
-7.29002476e-01 -6.20166898e-01 -1.05808437e+00 5.36915720e-01
7.74366319e-01 -9.91659343e-01 -6.79361463e-01 -2.97787249e-01
8.67555678e-01 -3.12727004e-01 -7.90183067e-01 2.40370870e-01
7.98421681e-01 -9.87866938e-01 7.50143349e-01 -2.90139079e-01
-1.43284231e-01 -3.64916891e-01 -3.22115064e-01 -1.40461576e+00
-1.01717718e-01 -7.63773501e-01 2.15130642e-01 1.87157023e+00
4.10763830e-01 -9.23596025e-01 4.00857776e-01 2.35467821e-01
-2.24496320e-01 -4.91388887e-01 -1.24249578e+00 -6.38397515e-01
-3.88947614e-02 -9.79115248e-01 1.17172229e+00 9.72424567e-01
1.57496795e-01 2.28300124e-01 -5.99955499e-01 4.18876231e-01
4.01550472e-01 -1.55105099e-01 7.23691523e-01 -1.08209860e+00
5.38370498e-02 -6.31754696e-01 -8.71243715e-01 -1.09936464e+00
4.16807681e-01 -8.70248318e-01 3.25066626e-01 -9.52532172e-01
-1.77719817e-01 -4.39272463e-01 -5.34574330e-01 4.58670527e-01
5.03972732e-03 4.64708000e-01 3.42414677e-01 -1.97170034e-01
-1.45224571e-01 8.71082783e-01 9.06626403e-01 -3.03093344e-01
-2.62637526e-01 -2.33367342e-03 -7.78998196e-01 6.42949224e-01
1.01598072e+00 -4.95432466e-01 -1.19910561e-01 -3.14989597e-01
-3.99198234e-01 1.26969755e-01 1.05120108e-01 -1.03420460e+00
5.59538543e-01 9.06977877e-02 3.10433581e-02 -7.85679817e-01
6.82887077e-01 -3.42664510e-01 -1.11862287e-01 4.30497736e-01
-4.44185942e-01 -6.32345453e-02 2.62166768e-01 4.69123684e-02
-6.11784518e-01 -7.52566457e-02 8.25970113e-01 4.05023247e-01
-4.81110692e-01 5.43362617e-01 -4.17554617e-01 8.20369944e-02
4.53093946e-01 1.33083835e-01 -2.56232589e-01 -6.12097085e-01
-3.94139528e-01 -1.14640640e-02 -7.37607032e-02 7.16213524e-01
7.25332737e-01 -1.60230982e+00 -1.03491783e+00 5.98082006e-01
4.63616960e-02 -3.71215701e-01 6.35281920e-01 7.49278426e-01
-2.72150666e-01 2.60062844e-01 -6.63370639e-02 -7.67377913e-01
-1.48072994e+00 2.18849778e-02 5.09658754e-01 3.45576584e-01
-2.29997531e-01 1.05301714e+00 1.59068510e-01 -7.63322234e-01
4.21781003e-01 -3.37817907e-01 -1.38793513e-01 1.86559036e-01
8.31899762e-01 2.56257057e-01 9.79905948e-02 -1.11427534e+00
-7.24082232e-01 5.37001908e-01 -6.65187612e-02 -2.87187219e-01
1.21203864e+00 -1.86980531e-01 -9.69908535e-02 5.04604697e-01
1.45147252e+00 4.23995942e-01 -6.89292729e-01 -2.87602514e-01
-3.33788574e-01 -3.80738556e-01 2.38328278e-01 -5.39857686e-01
-1.49192619e+00 1.38257670e+00 6.91791773e-01 1.67307287e-01
1.07838595e+00 -1.23575442e-01 1.12470043e+00 2.27792427e-01
9.46234539e-02 -8.55208158e-01 -1.70681983e-01 5.33619106e-01
8.75754416e-01 -9.85905528e-01 -4.67290610e-01 -1.99074730e-01
-6.02704108e-01 9.51418877e-01 5.68919897e-01 1.77422985e-01
1.06622112e+00 5.56551635e-01 7.29633331e-01 2.38236159e-01
-6.79513097e-01 -9.62543562e-02 3.21155071e-01 4.49924797e-01
5.71962833e-01 3.71643573e-01 -1.06860489e-01 1.05110490e+00
-9.25107598e-01 -4.94522601e-01 4.85961914e-01 2.53069788e-01
-4.61329579e-01 -8.96616817e-01 -5.80066204e-01 1.34542599e-01
-5.03415346e-01 -2.55224556e-01 -2.38197833e-01 4.32090908e-01
2.28930097e-02 1.20251381e+00 8.65666345e-02 -8.92642677e-01
3.63493741e-01 9.94382203e-02 -2.51965001e-02 -3.22468609e-01
-6.85404003e-01 1.39844924e-01 2.00237148e-02 -2.36134499e-01
-5.22118062e-03 -6.10639215e-01 -1.25839520e+00 -5.32895267e-01
-4.13628846e-01 3.65107089e-01 9.95487034e-01 7.60934770e-01
2.08318010e-01 5.40970266e-01 1.05085099e+00 -9.13874745e-01
-6.63209677e-01 -1.41689992e+00 -7.54566133e-01 1.46101460e-01
7.52759695e-01 -3.62710953e-01 -7.43877232e-01 -1.04710490e-01] | [14.36009693145752, 5.987492561340332] |
3f34be35-d24f-4c8d-a17c-dd8179162383 | cardiologist-level-arrhythmia-detection-and | null | null | https://doi.org/10.1038/s41591-018-0268-3 | https://arxiv.org/pdf/1707.01836.pdf | Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network | Computerized electrocardiogram (ECG) interpretation plays a critical role in the clinical ECG workflow. Widely available digital ECG data and the algorithmic paradigm of deep learning present an opportunity to substantially improve the accuracy and scalability of automated ECG analysis. However, a comprehensive evaluation of an end-to-end deep learning approach for ECG analysis across a wide variety of diagnostic classes has not been previously reported. Here, we develop a deep neural network (DNN) to classify 12 rhythm classes using 91,232 single-lead ECGs from 53,549 patients who used a single-lead ambulatory ECG monitoring device. When validated against an independent test dataset annotated by a consensus committee of board-certified practicing cardiologists, the DNN achieved an average area under the receiver operating characteristic curve (ROC) of 0.97. The average F1 score, which is the harmonic mean of the positive predictive value and sensitivity, for the DNN (0.837) exceeded that of average cardiologists (0.780). With specificity fixed at the average specificity achieved by cardiologists, the sensitivity of the DNN exceeded the average cardiologist sensitivity for all rhythm classes. These findings demonstrate that an end-to-end deep learning approach can classify a broad range of distinct arrhythmias from single-lead ECGs with high diagnostic performance similar to that of cardiologists. If confirmed in clinical settings, this approach could reduce the rate of misdiagnosed computerized ECG interpretations and improve the efficiency of expert human ECG interpretation by accurately triaging or prioritizing the most urgent conditions.
Dataset available from: https://irhythm.github.io/cardiol_test_set/ | ['Andrew Y. Ng', 'Mintu P. Turakhia', 'Masoumeh Haghpanahi', 'Codie Bourn', 'Pranav Rajpurkar', 'Geoffrey H. Tison', 'Awni Y. Hannun'] | 2019-01-07 | null | null | null | nature-medicine-2019-1 | ['arrhythmia-detection', 'electrocardiography-ecg'] | ['medical', 'methodology'] | [ 2.28202343e-01 -2.11423174e-01 2.92080175e-02 -4.36331004e-01
-1.08497679e+00 -9.49824691e-01 -4.55879092e-01 4.30719882e-01
-4.34937924e-01 6.62765801e-01 -9.38656926e-02 -9.28412139e-01
-5.67393005e-01 -5.49476266e-01 -2.77390629e-01 -4.63865697e-01
-4.09651458e-01 8.15545619e-01 -5.41005909e-01 4.12258446e-01
-2.57381797e-01 5.03634810e-01 -6.50630355e-01 4.27856803e-01
6.44962907e-01 1.16509521e+00 -3.65914345e-01 1.10037482e+00
6.45863593e-01 4.10579085e-01 -9.01749790e-01 -1.40115798e-01
3.50667566e-01 -1.03024340e+00 -4.04911727e-01 -2.54674405e-01
2.53986418e-01 -5.88173926e-01 -1.71966895e-01 4.07828718e-01
1.41738570e+00 -4.02130216e-01 4.72554922e-01 -6.55312955e-01
-7.46688023e-02 7.52536178e-01 -2.39260018e-01 7.04460502e-01
1.53558357e-02 5.53768456e-01 7.97692060e-01 -4.06851679e-01
3.05580646e-01 1.62296817e-01 1.24727046e+00 1.82462901e-01
-1.18003547e+00 -5.13735652e-01 -5.38786590e-01 -2.01347962e-01
-1.48731768e+00 -1.22981936e-01 5.06446779e-01 -3.13987851e-01
5.48766255e-01 2.77026772e-01 1.20951438e+00 6.25456035e-01
4.73368376e-01 -7.45878089e-04 7.92150915e-01 -1.68344170e-01
-1.43389720e-02 -2.60549635e-01 1.23137608e-01 4.01269525e-01
6.40835881e-01 6.51107505e-02 -1.41159669e-01 -5.86110592e-01
8.90571177e-01 1.93057582e-01 -4.16963607e-01 4.18543480e-02
-1.34865105e+00 4.34043735e-01 2.79735681e-02 2.57984340e-01
-6.55891001e-01 3.84746417e-02 8.30379009e-01 5.31075060e-01
2.32134424e-02 7.14182675e-01 -7.37417281e-01 -4.54117358e-01
-1.06253111e+00 3.17725278e-02 7.83959925e-01 1.97215185e-01
2.20439769e-02 2.83219010e-01 -1.63139507e-01 7.39965141e-01
-1.24694981e-01 7.18080938e-01 3.82205486e-01 -1.16809034e+00
1.40739918e-01 6.16819322e-01 1.39995500e-01 -7.03354955e-01
-8.96218240e-01 -1.23086643e+00 -1.23886895e+00 -7.66299516e-02
5.67070901e-01 -6.85434282e-01 -6.77032173e-01 1.28952885e+00
-1.69776618e-01 -2.23322138e-01 2.58634565e-04 1.06695461e+00
6.20274603e-01 3.76895331e-02 1.65555030e-01 -3.54079247e-01
1.40886199e+00 1.27405167e-01 -3.68424892e-01 9.01495442e-02
9.71641362e-01 -3.67507517e-01 7.52090156e-01 7.26710439e-01
-1.09532595e+00 -5.80429494e-01 -1.10440528e+00 4.32125926e-01
4.04323697e-01 4.36076254e-01 1.86857313e-01 8.66203845e-01
-1.01055968e+00 7.21997797e-01 -9.20606434e-01 -5.53146191e-02
7.98349917e-01 4.80721086e-01 -2.13846117e-01 1.41937047e-01
-1.26356328e+00 6.74738646e-01 2.18633220e-01 2.84276962e-01
-1.04757524e+00 -8.89301777e-01 -4.24509048e-01 1.45688623e-01
-1.26302810e-02 -1.08146191e+00 1.08599389e+00 -9.10721779e-01
-8.33440661e-01 1.00756514e+00 1.95700362e-01 -7.57571042e-01
7.53193438e-01 -1.13124952e-01 -3.48282635e-01 4.23552781e-01
8.81391540e-02 3.71021777e-02 4.79225010e-01 -6.87539816e-01
-5.03246129e-01 -6.24214351e-01 -1.81505248e-01 1.63396612e-01
2.78021637e-02 -1.66115940e-01 1.78330228e-01 -5.17830551e-01
9.28143859e-02 -8.89033198e-01 -3.16510260e-01 -1.58420712e-01
-1.01885080e-01 3.01873207e-01 2.81426281e-01 -8.43424022e-01
1.48603058e+00 -2.38751125e+00 -1.45238623e-01 4.49581414e-01
8.54081690e-01 5.34030378e-01 3.37842792e-01 2.59709299e-01
-2.98760653e-01 3.36835921e-01 -4.70638454e-01 3.00472349e-01
-4.76038307e-01 -6.95069656e-02 8.97291750e-02 6.77240372e-01
-2.37422362e-02 1.11605930e+00 -7.87698328e-01 -1.40805900e-01
3.32092792e-01 3.62757266e-01 -2.04505891e-01 1.26101509e-01
4.48069334e-01 7.70146310e-01 -1.36384204e-01 3.82292271e-01
2.75755882e-01 -6.68858409e-01 8.45573783e-01 -5.65551594e-02
4.03369099e-01 2.11666495e-01 -9.35993910e-01 1.56868935e+00
-1.42933547e-01 5.69934607e-01 -2.08230883e-01 -9.91539001e-01
1.08311176e+00 6.70175374e-01 7.58273780e-01 -3.37568581e-01
2.64965296e-01 3.50163162e-01 8.90260577e-01 -3.90090883e-01
-6.46280527e-01 -4.76789445e-01 -3.83916944e-02 7.27502048e-01
-2.67960161e-01 1.50906071e-01 -1.77499980e-01 3.12380865e-03
1.37244391e+00 -3.15700620e-01 5.69963753e-01 -3.32031518e-01
2.66173422e-01 -3.12130600e-02 7.41004050e-01 1.29130006e+00
-3.98483425e-01 9.64595199e-01 7.02563524e-01 -1.20675409e+00
-8.61516654e-01 -1.36242473e+00 -4.80005413e-01 3.79577816e-01
-4.83964056e-01 -3.44371229e-01 -5.29920697e-01 -5.30707181e-01
-2.65334249e-02 1.08186573e-01 -3.31651688e-01 -3.95647705e-01
-5.57674646e-01 -1.00000751e+00 1.24600530e+00 8.41975272e-01
2.54613936e-01 -1.03681946e+00 -1.38555121e+00 6.34844184e-01
-3.52932364e-01 -7.73198247e-01 -3.33740786e-02 2.63270199e-01
-1.03454626e+00 -1.29633343e+00 -9.04557586e-01 -4.64401245e-01
1.26899049e-01 -5.13379097e-01 1.52455091e+00 1.77196115e-01
-5.89171469e-01 2.30039567e-01 -9.42242742e-02 -7.92168975e-01
-5.31386793e-01 1.92809567e-01 9.74069983e-02 -9.22374129e-02
3.46391916e-01 -6.73146904e-01 -1.12968218e+00 8.17365050e-02
-3.86262298e-01 -3.42887193e-01 5.71232915e-01 9.50408816e-01
6.17683768e-01 -3.39755446e-01 1.17192435e+00 -8.32279503e-01
7.21249819e-01 -3.05190563e-01 -3.14809799e-01 -1.70501564e-02
-8.84927750e-01 -6.26942217e-01 6.42605066e-01 -1.79980949e-01
-2.11873099e-01 5.09137660e-03 -1.63209781e-01 -3.72061253e-01
-3.57336789e-01 7.07774043e-01 2.86218882e-01 4.03389037e-01
9.29067612e-01 1.05623052e-01 1.08865522e-01 -5.63306548e-02
-4.00282413e-01 6.41829669e-01 6.08317614e-01 -3.90136987e-01
1.95232540e-01 2.04347864e-01 2.27365971e-01 -5.93606651e-01
-6.47808850e-01 -3.44614118e-01 -5.86787522e-01 -1.79084733e-01
8.23998988e-01 -8.56087565e-01 -7.86005855e-01 3.46758276e-01
-9.19263721e-01 -3.05315524e-01 -5.81936896e-01 6.25025094e-01
-3.16299170e-01 3.85643572e-01 -4.85601187e-01 -6.84876919e-01
-9.87734795e-01 -8.99039924e-01 8.54154944e-01 -2.74436384e-01
-9.37226713e-01 -9.10733998e-01 7.63827413e-02 2.71911230e-02
2.23895952e-01 8.47458899e-01 1.00913835e+00 -9.43141699e-01
-1.50575163e-02 -5.81010461e-01 6.22520735e-03 6.58092201e-01
4.12044376e-01 -1.83384746e-01 -7.45455384e-01 -1.93560660e-01
6.00575004e-03 -3.69420275e-02 4.80518520e-01 9.55267191e-01
1.18036616e+00 3.29990149e-01 1.07189892e-02 7.00850070e-01
1.17747462e+00 6.19131804e-01 6.40327454e-01 -9.00993403e-03
6.06435955e-01 -1.84651643e-01 -4.05873870e-03 6.30143344e-01
1.73365995e-01 1.70369614e-02 1.35871973e-02 -4.62918222e-01
2.09225118e-01 1.83720306e-01 -2.06712589e-01 3.52841079e-01
-3.81756097e-01 -1.34947792e-01 -1.56141925e+00 4.15953159e-01
-1.51418018e+00 -7.84471095e-01 -3.48463118e-01 2.42261004e+00
4.97240633e-01 3.69095385e-01 4.62671906e-01 6.35794878e-01
6.73632920e-01 -2.08580673e-01 -8.69916320e-01 -6.46312356e-01
-1.97881669e-01 4.65029746e-01 2.72581965e-01 9.18043554e-02
-8.40788007e-01 6.12834767e-02 6.63175154e+00 -8.68789256e-02
-1.22160876e+00 6.80425018e-02 1.19595265e+00 -1.71950504e-01
2.82515794e-01 -3.02011400e-01 -7.49620646e-02 3.64243329e-01
1.48314202e+00 -3.22368830e-01 1.55603826e-01 5.25486588e-01
4.85463947e-01 2.15661809e-01 -1.16464388e+00 1.20545733e+00
-1.20869659e-01 -1.49210536e+00 -3.29557896e-01 8.44218582e-02
4.00001198e-01 1.69450223e-01 -2.22181268e-02 1.18190594e-01
-4.23260808e-01 -1.08710098e+00 2.35860258e-01 5.21212220e-01
1.69427764e+00 -7.17747867e-01 1.24694693e+00 1.77486539e-01
-7.99402475e-01 -2.45002836e-01 1.93991438e-01 -4.80889887e-01
1.25215963e-01 7.59248614e-01 -9.29681540e-01 4.26800340e-01
7.75764585e-01 8.89268875e-01 -2.78368205e-01 8.92370880e-01
5.55499531e-02 1.08965039e+00 -3.47268790e-01 4.49165374e-01
-4.60862033e-02 1.82821155e-01 4.25134629e-01 1.03611946e+00
2.16122754e-02 3.99872065e-01 1.12808742e-01 6.03716791e-01
-2.14417785e-01 1.25264645e-01 -6.77920103e-01 3.45314890e-02
3.12170595e-01 1.06652808e+00 -8.05828631e-01 -6.46313071e-01
-1.33528873e-01 3.52014542e-01 -1.84099197e-01 3.16067398e-01
-8.75828266e-01 -5.88855922e-01 1.14470780e-01 5.02836764e-01
-2.49436557e-01 2.99812376e-01 -1.19284606e+00 -7.80934453e-01
1.80709973e-01 -1.29390371e+00 7.43796110e-01 -4.07078654e-01
-1.14978695e+00 7.58624077e-01 -2.53223062e-01 -1.47216153e+00
-3.92891258e-01 -3.32708895e-01 -5.73558450e-01 1.14673650e+00
-7.47746587e-01 -4.22266960e-01 -2.63818979e-01 1.72805309e-01
1.31919056e-01 -1.65551826e-01 1.22436142e+00 4.52717692e-01
-3.09110016e-01 6.97925568e-01 -1.74747273e-01 5.58716357e-01
6.81451499e-01 -1.20435321e+00 2.23553821e-01 5.61609209e-01
-2.23917127e-01 7.36456692e-01 1.12692066e-01 -4.94055450e-01
-9.38446879e-01 -9.78017986e-01 7.64076531e-01 -6.84522808e-01
1.46838903e-01 2.21925467e-01 -7.83418715e-01 5.97744763e-01
-1.30917028e-01 9.82468873e-02 1.19915164e+00 1.80035219e-01
1.30080968e-01 -4.78530884e-01 -1.00536823e+00 1.66080892e-01
7.24770129e-01 -5.13783753e-01 -4.79182243e-01 1.21686764e-01
-7.33881742e-02 -5.33855915e-01 -1.44175172e+00 7.77573228e-01
1.17375326e+00 -9.58991945e-01 7.79588878e-01 -6.59979761e-01
4.36623812e-01 -9.10088271e-02 3.55734855e-01 -1.04980731e+00
-4.33914691e-01 -7.53871858e-01 2.14614384e-02 5.48893034e-01
6.09669685e-01 -8.89840841e-01 5.96712112e-01 5.18036962e-01
-1.98390663e-01 -1.14390337e+00 -7.41069317e-01 -3.31098616e-01
5.04133217e-02 -5.16058207e-01 3.16021413e-01 9.02644634e-01
-1.95205286e-01 2.39386708e-01 -2.81325191e-01 1.13355465e-01
5.48600793e-01 1.13947660e-01 3.58647019e-01 -1.46791339e+00
-4.69455957e-01 -2.68147051e-01 -6.75197184e-01 -1.60202637e-01
-3.96408617e-01 -1.18095613e+00 -4.90531981e-01 -1.74931240e+00
1.92261726e-01 -5.17578244e-01 -9.64738786e-01 5.17959595e-01
-2.58440733e-01 5.09005070e-01 4.10212241e-02 1.32828906e-01
-2.71485418e-01 -3.28892201e-01 9.66095269e-01 1.19041398e-01
-5.22519469e-01 2.14533269e-01 -9.30610061e-01 8.27424109e-01
1.10780251e+00 -7.01082289e-01 -4.81447846e-01 -4.17888939e-01
2.04264686e-01 6.12682819e-01 4.04539764e-01 -1.25729024e+00
-2.75055885e-01 3.72354418e-01 1.02654254e+00 -3.77911806e-01
-1.24468498e-01 -5.52930415e-01 5.08587301e-01 9.88402963e-01
-3.67797345e-01 5.51608145e-01 4.24320489e-01 2.93444127e-01
-3.61639336e-02 4.11176562e-01 7.39307344e-01 -3.22367847e-01
1.06946692e-01 3.57648909e-01 -7.80932128e-01 6.04554296e-01
7.77247965e-01 -3.37769687e-01 5.91068491e-02 -4.35082823e-01
-1.19094205e+00 -2.73214672e-02 1.31421974e-02 5.15678041e-02
7.25070477e-01 -9.01166618e-01 -1.13395536e+00 2.42328629e-01
8.00299272e-02 -2.43551489e-02 5.19092202e-01 1.18986738e+00
-9.62887228e-01 1.47246376e-01 -2.75913566e-01 -1.01935375e+00
-9.78953123e-01 -3.23204696e-02 9.50282097e-01 -1.87863499e-01
-9.34574068e-01 4.23791498e-01 -1.23497166e-01 -1.52344201e-02
1.03022933e-01 -3.27897459e-01 2.52850592e-01 -9.56616700e-02
5.00207841e-01 4.79749382e-01 3.60536605e-01 -7.32509121e-02
-6.39032900e-01 4.21310365e-01 1.01100147e-01 9.01524946e-02
1.06514037e+00 1.70121267e-01 3.75778154e-02 6.44516051e-01
8.51829350e-01 -4.06762958e-01 -6.41048312e-01 4.13740456e-01
-4.12306756e-01 -9.86332372e-02 -4.61679930e-03 -1.31730199e+00
-1.05128407e+00 1.08623147e+00 1.17234659e+00 1.01759493e-01
1.25578737e+00 -3.39430332e-01 6.83470666e-01 3.08859169e-01
2.32576713e-01 -5.68694413e-01 -4.05110151e-01 8.01638176e-04
3.82589161e-01 -1.04314375e+00 2.00044531e-02 2.51820713e-01
-9.25926387e-01 1.06440544e+00 1.30630136e-01 -2.01800317e-01
7.40520537e-01 2.05576509e-01 6.67155802e-01 -2.46065348e-01
-5.64425886e-01 4.72541749e-01 -7.25394934e-02 5.61685145e-01
7.09197164e-01 4.12700206e-01 -5.75739026e-01 8.94344270e-01
9.52494666e-02 5.01926899e-01 5.46120584e-01 8.33376348e-01
7.48400390e-02 -7.43367553e-01 3.62549461e-02 1.04799104e+00
-9.90036666e-01 -1.91360027e-01 -2.26822913e-01 3.89435798e-01
2.10562065e-01 9.01225328e-01 -3.79706360e-02 -2.25516528e-01
3.78522754e-01 4.69590634e-01 3.57600182e-01 -6.29667222e-01
-1.04101801e+00 1.43758774e-01 1.12861477e-01 -1.00451998e-01
-3.48755755e-02 -6.35766208e-01 -1.25527287e+00 1.14444099e-01
1.50488511e-01 6.52375072e-02 2.69038767e-01 7.26950467e-01
7.46105134e-01 8.36948097e-01 3.65000367e-01 -2.56717175e-01
-5.73669553e-01 -9.43915248e-01 -8.06402624e-01 2.57426947e-01
5.69636583e-01 -6.69392496e-02 -3.04582566e-01 2.10167870e-01] | [14.38037395477295, 3.3278591632843018] |
258b9308-b062-443a-8595-235027fd5d07 | em-based-bounding-of-unidentifiable-queries | 2011.02912 | null | https://arxiv.org/abs/2011.02912v3 | https://arxiv.org/pdf/2011.02912v3.pdf | Causal Expectation-Maximisation | Structural causal models are the basic modelling unit in Pearl's causal theory; in principle they allow us to solve counterfactuals, which are at the top rung of the ladder of causation. But they often contain latent variables that limit their application to special settings. This appears to be a consequence of the fact, proven in this paper, that causal inference is NP-hard even in models characterised by polytree-shaped graphs. To deal with such a hardness, we introduce the causal EM algorithm. Its primary aim is to reconstruct the uncertainty about the latent variables from data about categorical manifest variables. Counterfactual inference is then addressed via standard algorithms for Bayesian networks. The result is a general method to approximately compute counterfactuals, be they identifiable or not (in which case we deliver bounds). We show empirically, as well as by deriving credible intervals, that the approximation we provide becomes accurate in a fair number of EM runs. These results lead us finally to argue that there appears to be an unnoticed limitation to the trending idea that counterfactual bounds can often be computed without knowledge of the structural equations. | ['Rafael Cabañas', 'Alessandro Antonucci', 'Marco Zaffalon'] | 2020-11-04 | null | null | null | null | ['counterfactual-inference'] | ['miscellaneous'] | [ 2.96297699e-01 8.60018015e-01 -2.24285409e-01 -9.78541970e-02
-5.50621986e-01 -6.15855694e-01 9.67204452e-01 3.79108898e-02
-2.95184493e-01 1.18702590e+00 5.13812065e-01 -8.69695842e-01
-7.60066271e-01 -9.75006640e-01 -7.71425724e-01 -7.56631374e-01
-4.42991883e-01 6.46538973e-01 1.11620501e-01 1.90160051e-01
4.23889905e-01 3.24806213e-01 -1.37897813e+00 -5.34822270e-02
7.03698277e-01 1.99408546e-01 -2.48323992e-01 7.04472840e-01
-2.38908753e-02 8.80180478e-01 -2.99686819e-01 -7.28148162e-01
-9.01965573e-02 -5.60543716e-01 -1.13712025e+00 -3.16342711e-01
-7.05889836e-02 -3.54361355e-01 -1.20301366e-01 1.11248028e+00
1.47625268e-01 -2.31842041e-01 9.71702456e-01 -1.41404486e+00
-2.45954648e-01 1.17647052e+00 -6.60715163e-01 1.16315842e-01
4.46150959e-01 -2.22550780e-01 1.18592834e+00 -1.85980812e-01
5.61582804e-01 1.66074729e+00 6.87306046e-01 2.40274578e-01
-1.66367221e+00 -3.56876850e-01 1.55550227e-01 1.17043033e-01
-1.07548428e+00 -3.01933229e-01 5.12584984e-01 -4.43256646e-01
5.82380593e-01 6.58596635e-01 5.53224921e-01 1.17344642e+00
3.28826845e-01 4.65759307e-01 1.38298225e+00 -7.26815343e-01
4.19037521e-01 -2.81566143e-01 1.13373389e-02 3.63297284e-01
8.58776450e-01 6.05831206e-01 -2.87358344e-01 -7.41127133e-01
7.36555398e-01 -3.32396805e-01 -3.90024960e-01 -4.10933584e-01
-1.15502453e+00 1.14381337e+00 5.22201955e-02 6.37946203e-02
-2.52708346e-01 5.45650899e-01 3.51276189e-01 1.93500981e-01
2.94477820e-01 1.22935928e-01 -3.49379033e-01 6.87204162e-03
-7.14681923e-01 5.83127141e-01 1.10640907e+00 6.26338422e-01
4.07650732e-02 -2.77957112e-01 3.17411631e-01 2.08013415e-01
6.27609909e-01 3.19417715e-01 -9.16039720e-02 -1.11703026e+00
3.12927574e-01 1.35946544e-02 4.59579736e-01 -8.02179813e-01
-2.87811488e-01 4.50220369e-02 -7.70275772e-01 3.07800740e-01
9.29905593e-01 -2.83121437e-01 -4.51440006e-01 2.12579060e+00
3.78944993e-01 2.91537970e-01 -1.82720035e-01 5.05113482e-01
-2.34961174e-02 4.73560512e-01 3.98786187e-01 -9.63705659e-01
1.07426357e+00 1.62303925e-01 -7.88209319e-01 2.06683740e-01
6.41217291e-01 -6.63775265e-01 6.46370053e-01 6.13449335e-01
-9.83800769e-01 1.68444082e-01 -8.87311518e-01 3.48703444e-01
5.36453053e-02 -9.19874966e-01 1.22410607e+00 1.04447591e+00
-8.10929954e-01 8.70505095e-01 -8.80554438e-01 -2.41402894e-01
1.46992415e-01 2.22224891e-01 -9.25386027e-02 7.38107227e-03
-1.26884913e+00 9.44688022e-01 5.69421589e-01 1.87431574e-01
-7.47261167e-01 -4.82682526e-01 -4.83494312e-01 3.71185429e-02
6.76628649e-01 -9.48377788e-01 1.34789515e+00 -9.48510349e-01
-9.55677271e-01 5.74912965e-01 -8.69281664e-02 -4.86795604e-01
8.38177145e-01 2.26272449e-01 -2.03298241e-01 -5.44495434e-02
1.16738938e-01 8.31962228e-02 4.78135049e-01 -1.34299672e+00
-5.52685380e-01 -5.54326892e-01 3.05229098e-01 -7.81768188e-02
5.89507937e-01 1.63959309e-01 2.97614008e-01 -4.43271637e-01
3.51629525e-01 -7.89138675e-01 -4.42298323e-01 -4.61125523e-01
-7.61589170e-01 -4.81694847e-01 1.94200754e-01 -3.02700758e-01
1.19035196e+00 -1.63379908e+00 -5.38758226e-02 4.43491101e-01
2.26440430e-01 -5.31548679e-01 3.78309876e-01 6.46859527e-01
-5.73565543e-01 4.73309755e-01 -4.93974656e-01 2.33619675e-01
4.32950258e-01 2.91466296e-01 -6.30302787e-01 7.67991722e-01
-1.96107611e-01 8.69169831e-01 -9.01559412e-01 -6.95231438e-01
1.43703163e-01 -1.09990649e-02 -4.78662401e-01 -1.52727962e-01
-3.22006434e-01 2.55587939e-02 -3.79240900e-01 1.05919726e-01
6.31572366e-01 -1.22565433e-01 7.79485762e-01 2.96797901e-01
-3.64632577e-01 5.34863710e-01 -1.62352204e+00 1.23485208e+00
6.98988959e-02 2.51069516e-01 -5.41048199e-02 -1.11465025e+00
2.06334636e-01 5.97214818e-01 2.51078278e-01 -2.17621282e-01
1.58734858e-01 1.04054481e-01 2.12283045e-01 -3.56943876e-01
1.02315485e-01 -9.74158704e-01 -3.75851214e-01 6.32535279e-01
-3.07963461e-01 -3.21606584e-02 6.45759106e-02 2.06922859e-01
9.82637227e-01 1.44925535e-01 6.07096732e-01 -5.52128494e-01
-2.22791787e-02 9.50246304e-02 5.21630943e-01 1.06272709e+00
1.99562505e-01 1.27750531e-01 1.17622352e+00 -2.39473939e-01
-1.13375747e+00 -1.41844630e+00 -5.32817125e-01 3.50704640e-01
-2.06510052e-01 -1.86003804e-01 -7.85287559e-01 -5.13960123e-01
-1.01251140e-01 1.26341510e+00 -8.68681371e-01 1.37806848e-01
-3.36715102e-01 -1.30351305e+00 3.21380496e-01 3.19460213e-01
-3.92028913e-02 -7.50187993e-01 -7.38395214e-01 3.08070004e-01
-1.80397496e-01 -3.43063831e-01 3.67400646e-01 3.41728747e-01
-1.10131693e+00 -1.47338653e+00 -3.63311470e-01 9.90000293e-02
3.69260520e-01 -1.72019050e-01 1.09907019e+00 -1.23045832e-01
1.07412681e-01 1.66339472e-01 -5.96417561e-02 -4.23618585e-01
-6.88371241e-01 -5.29727757e-01 2.23235171e-02 -4.94797140e-01
4.74187732e-01 -1.02726853e+00 -3.44201952e-01 5.53764810e-04
-8.80788982e-01 1.68500524e-02 3.99663806e-01 7.42030501e-01
2.39652961e-01 4.58722055e-01 5.06225407e-01 -1.16910434e+00
6.23749018e-01 -6.04967415e-01 -9.71292555e-01 2.41816491e-01
-6.73218846e-01 2.39420399e-01 3.06264162e-01 -7.52429441e-02
-1.30786312e+00 -2.38688514e-01 7.69995302e-02 2.80725807e-01
-4.06752408e-01 7.97951400e-01 -4.77536529e-01 5.95281541e-01
6.17423236e-01 -3.42620462e-01 -2.13789791e-01 -4.46058273e-01
6.58766270e-01 4.79807943e-01 4.66594756e-01 -7.87003756e-01
6.78110600e-01 6.97187841e-01 5.08588433e-01 -5.19324243e-01
-8.74053180e-01 1.26234308e-01 -4.41127777e-01 1.91798760e-03
7.51736164e-01 -5.46498597e-01 -1.33277786e+00 -2.04831943e-01
-1.20861518e+00 -4.03487086e-01 -1.58121750e-01 6.23331904e-01
-9.21685874e-01 4.11611021e-01 -4.78432804e-01 -1.47088945e+00
4.71565485e-01 -6.64411604e-01 5.77890754e-01 -3.22767347e-01
-5.01677811e-01 -1.16222286e+00 2.64520943e-01 3.13512906e-02
-3.07991147e-01 7.17644513e-01 1.26758504e+00 -2.38486439e-01
-5.02116501e-01 -7.36801773e-02 -1.07014604e-01 -2.75894493e-01
-1.10916294e-01 2.46263310e-01 -9.86022711e-01 8.83112848e-03
2.63789803e-01 1.48973599e-01 7.78392076e-01 8.62850428e-01
8.92046928e-01 -6.68475091e-01 -5.22456467e-01 -8.53088796e-02
1.62876964e+00 1.88088883e-02 8.64090085e-01 7.19331056e-02
2.61898220e-01 9.92311180e-01 3.96070242e-01 2.88332522e-01
2.80252099e-01 6.45001352e-01 4.67136830e-01 3.60426664e-01
4.34346616e-01 -5.04550219e-01 1.16533361e-01 3.47389758e-01
-2.00639427e-01 -9.79440287e-02 -9.10917878e-01 5.88870406e-01
-1.96359408e+00 -1.55456281e+00 -9.53261912e-01 2.57610464e+00
9.85940635e-01 4.78117317e-01 3.91332269e-01 3.50264341e-01
7.10464895e-01 -2.19373941e-01 -3.35944258e-02 -5.32961547e-01
5.62676042e-02 -6.68226555e-03 8.27375054e-01 8.49620104e-01
-7.50551164e-01 3.06873620e-01 7.31981802e+00 7.60517776e-01
-2.47962952e-01 2.39700288e-01 5.15306354e-01 -2.10758001e-02
-8.25330496e-01 6.76008105e-01 -3.42982173e-01 6.47907734e-01
1.26001215e+00 -5.06674170e-01 3.46757919e-01 4.39075589e-01
5.21368861e-01 -7.35388815e-01 -1.28170061e+00 3.15348566e-01
-5.73572934e-01 -1.11050677e+00 -1.71761349e-01 4.35141504e-01
6.41902328e-01 -4.28087294e-01 -3.62917751e-01 -1.86547264e-01
1.20885754e+00 -1.27619982e+00 7.94576824e-01 4.51817751e-01
6.49017572e-01 -1.08124340e+00 5.87233543e-01 6.30869091e-01
-5.81764102e-01 -1.40870646e-01 -3.83877873e-01 -4.83183116e-01
4.92408097e-01 9.25845623e-01 -5.47225356e-01 7.49849677e-01
2.43132234e-01 2.72301827e-02 8.07355717e-02 1.07879639e+00
-6.45363927e-01 9.08871949e-01 -4.17546511e-01 1.81986302e-01
3.13209221e-02 -2.87629217e-01 4.86513495e-01 9.02236521e-01
2.57366002e-01 3.19212198e-01 -5.70921361e-01 1.13754380e+00
3.96902144e-01 -4.10997689e-01 -8.60474646e-01 5.78550287e-02
4.73513365e-01 5.99306762e-01 -8.03073764e-01 -9.75379944e-02
-4.03443396e-01 4.23110574e-01 -4.64710295e-02 2.53517359e-01
-7.82245994e-01 8.40729252e-02 2.15106055e-01 1.10163055e-01
-1.69996712e-02 -2.11892445e-02 -5.29698074e-01 -9.97636497e-01
-1.98521569e-01 -7.00956762e-01 6.21355414e-01 -6.27132356e-01
-1.22486007e+00 -2.87970603e-01 6.27251983e-01 -5.69309950e-01
-7.23862350e-01 -4.06985372e-01 -5.15349269e-01 9.96369541e-01
-9.81541157e-01 -7.75549531e-01 5.28117537e-01 2.90500432e-01
-2.67943293e-02 7.72346616e-01 8.89283776e-01 -2.96074629e-01
-3.14265430e-01 -3.10484543e-02 1.44751981e-01 -2.59563535e-01
1.55118406e-01 -1.67129648e+00 2.26273656e-01 1.02255344e+00
2.08939433e-01 8.61758769e-01 1.41074002e+00 -9.97826815e-01
-1.19568706e+00 -4.76162434e-01 1.12263072e+00 -6.55632496e-01
1.03153849e+00 -3.28491300e-01 -5.97676337e-01 9.59967256e-01
1.22017950e-01 -5.99048018e-01 5.08150995e-01 7.48099327e-01
-3.22461247e-01 3.52461040e-01 -8.20090055e-01 7.50266254e-01
1.03881979e+00 -2.16992378e-01 -1.08491719e+00 2.08095923e-01
6.38841033e-01 1.10038608e-01 -7.46293008e-01 2.67459482e-01
6.37809336e-01 -1.22607958e+00 7.95630991e-01 -7.24325359e-01
6.13168120e-01 -3.40843409e-01 -1.05819553e-01 -1.02561390e+00
-1.07288495e-01 -8.44237208e-01 2.72055902e-02 1.23518527e+00
3.21313679e-01 -6.83160305e-01 5.10946691e-01 9.25356925e-01
3.38755667e-01 -2.79188395e-01 -1.25681102e+00 -7.85307348e-01
4.42523569e-01 -8.81182373e-01 5.50616026e-01 1.30177677e+00
4.18474823e-01 3.53095204e-01 -1.67473704e-01 2.53306031e-01
1.24163091e+00 2.24004358e-01 4.86770660e-01 -1.70635533e+00
-5.59923947e-01 -5.06547391e-01 -1.29520983e-01 -6.03050888e-01
1.68101221e-01 -4.91553128e-01 -1.74840793e-01 -1.42584872e+00
6.12782776e-01 -5.16481102e-01 2.41333768e-02 1.19657651e-01
-1.40941054e-01 -2.78529495e-01 -1.35867983e-01 -3.85364424e-03
-1.81227833e-01 1.62462473e-01 9.68140304e-01 3.11766148e-01
1.09895304e-01 2.34467492e-01 -9.55408692e-01 1.06359065e+00
6.37294829e-01 -8.29335690e-01 -5.36868513e-01 1.43882751e-01
8.70961785e-01 7.09193230e-01 1.06798196e+00 -2.20878497e-01
5.61243156e-03 -5.50066173e-01 2.03874618e-01 -6.29633784e-01
-1.79138348e-01 -8.02999318e-01 8.36399674e-01 5.89784145e-01
-4.66468453e-01 -1.78736225e-01 1.08871892e-01 8.28191936e-01
2.37221807e-01 -6.80350006e-01 4.60551500e-01 -2.17635840e-01
-1.50566429e-01 -2.52939165e-01 -4.12185878e-01 -5.89286052e-02
7.18787134e-01 -7.99998417e-02 -2.74741262e-01 -4.78015900e-01
-8.36716115e-01 4.76253480e-02 5.12540221e-01 -2.23498926e-01
1.36734858e-01 -1.14694417e+00 -7.14260697e-01 -5.87150097e-01
-2.57783771e-01 -1.44868046e-01 3.26670408e-01 1.08949673e+00
-8.04989263e-02 6.18761182e-01 2.39323169e-01 -2.27350101e-01
-9.50756967e-01 9.94016051e-01 3.29926282e-01 -1.47860542e-01
-5.64446688e-01 5.49903750e-01 4.07262474e-01 1.94237251e-02
-9.84160975e-02 -5.24792038e-02 1.26701385e-01 4.02430221e-02
5.47356963e-01 5.24874628e-01 -4.35553789e-01 -1.41723081e-01
-2.52982020e-01 -7.81648904e-02 3.22863221e-01 -5.36864758e-01
1.36932111e+00 -3.96738440e-01 -5.33026457e-01 7.40968347e-01
5.66685855e-01 2.34513044e-01 -1.18116975e+00 2.64223754e-01
4.72056001e-01 -4.74136412e-01 -1.03020012e-01 -9.21325684e-01
-2.38689482e-01 7.71404386e-01 2.16395333e-01 8.27464700e-01
9.41550434e-01 1.75241366e-01 -6.77168071e-02 -1.53589807e-02
5.83778381e-01 -8.11522901e-01 -9.54656959e-01 -2.14706868e-01
8.25361848e-01 -8.18467498e-01 2.79207587e-01 -4.80266064e-01
1.63843874e-02 9.09783483e-01 -2.45059192e-01 -2.75677051e-02
3.07934433e-01 3.29157531e-01 -5.20982862e-01 -3.58042806e-01
-9.94012654e-01 -7.59464875e-02 -1.48905098e-01 6.02019072e-01
5.49907923e-01 5.18731296e-01 -7.82586455e-01 4.46267366e-01
-4.49918777e-01 -1.74750220e-02 8.64930809e-01 4.35044974e-01
-4.24149066e-01 -1.14225686e+00 -6.74403667e-01 4.62013364e-01
-8.23311090e-01 -1.38700023e-01 -3.49557579e-01 1.29292035e+00
-1.50395691e-01 1.25181985e+00 -8.13506097e-02 3.20904583e-01
5.71741238e-02 7.57555068e-02 8.15764725e-01 -3.02748889e-01
5.31354435e-02 3.39012980e-01 4.52562332e-01 -5.51896214e-01
-5.77724278e-01 -9.10402536e-01 -8.99020195e-01 -8.69528651e-01
-3.97919029e-01 4.20350730e-01 3.50485295e-01 1.15921700e+00
-4.18944985e-01 3.00514549e-01 2.88989633e-01 -3.96222562e-01
-9.47883666e-01 -8.55507612e-01 -9.71229434e-01 -3.79275233e-02
8.14223513e-02 -7.10525513e-01 -7.07031250e-01 -5.37745692e-02] | [8.1520357131958, 5.676864147186279] |
c14c6c71-bc48-4aff-b518-f7985119a4b8 | gumdrop-at-the-disrpt2019-shared-task-a-model | 1904.10419 | null | https://arxiv.org/abs/1904.10419v2 | https://arxiv.org/pdf/1904.10419v2.pdf | GumDrop at the DISRPT2019 Shared Task: A Model Stacking Approach to Discourse Unit Segmentation and Connective Detection | In this paper we present GumDrop, Georgetown University's entry at the DISRPT 2019 Shared Task on automatic discourse unit segmentation and connective detection. Our approach relies on model stacking, creating a heterogeneous ensemble of classifiers, which feed into a metalearner for each final task. The system encompasses three trainable component stacks: one for sentence splitting, one for discourse unit segmentation and one for connective detection. The flexibility of each ensemble allows the system to generalize well to datasets of different sizes and with varying levels of homogeneity. | ['Amir Zeldes', 'YIlun Zhu', 'Yue Yu', 'Siyao Peng', 'Mackenzie Gong', 'Yang Liu', 'Yan Liu'] | 2019-04-23 | gumdrop-at-the-disrpt2019-shared-task-a-model-1 | https://aclanthology.org/W19-2717 | https://aclanthology.org/W19-2717.pdf | ws-2019-6 | ['discourse-segmentation', 'connective-detection'] | ['natural-language-processing', 'natural-language-processing'] | [ 4.76760238e-01 6.44223809e-01 -2.14188561e-01 -3.71690303e-01
-9.65667784e-01 -8.22550058e-01 7.99768388e-01 3.12577486e-01
-2.05807507e-01 8.39469433e-01 6.10163987e-01 -7.66030014e-01
3.71721685e-01 -4.13187593e-01 -4.00409311e-01 -2.85857707e-01
-3.54153998e-02 7.11327791e-01 5.85025489e-01 -3.31851840e-01
3.66562873e-01 -2.00901628e-01 -1.23259914e+00 1.28083241e+00
6.98989332e-01 8.09437335e-01 1.72009494e-03 1.25800133e+00
-1.29382193e-01 1.55673563e+00 -9.30567026e-01 -2.24260449e-01
-8.87399390e-02 -2.67107695e-01 -1.46968865e+00 1.19102991e-03
2.55546629e-01 -1.61426574e-01 -4.61668558e-02 5.07529378e-01
2.91575342e-01 -1.02412812e-02 7.05716193e-01 -8.43866289e-01
-2.39374444e-01 1.43455172e+00 -1.19750872e-01 3.79266649e-01
6.79905772e-01 -6.21227548e-02 1.38903141e+00 -7.64650166e-01
9.31674242e-01 1.39807630e+00 6.30590618e-01 5.09036005e-01
-1.26477456e+00 -1.73823714e-01 3.93442214e-01 1.86564356e-01
-7.94442952e-01 -8.53843689e-01 5.00303030e-01 -7.64705539e-01
1.52065861e+00 5.58440030e-01 4.01362717e-01 1.24923372e+00
-5.92696033e-02 1.08630478e+00 1.09894931e+00 -6.41320646e-01
2.33033031e-01 -7.20853731e-02 6.49059117e-01 4.50811684e-01
-2.46182218e-01 -5.00021577e-01 -3.95248204e-01 -1.17793843e-01
2.33509168e-01 -8.65835249e-01 -2.97686726e-01 2.31880501e-01
-1.06258440e+00 7.65482068e-01 9.48246270e-02 5.28290510e-01
1.42776504e-01 -1.84349597e-01 7.17901230e-01 3.32628548e-01
6.99203014e-01 6.30933344e-01 -4.42888230e-01 -3.38279784e-01
-9.55678284e-01 6.45349562e-01 1.33478713e+00 9.98558581e-01
1.23213120e-01 -3.25099409e-01 -4.65053707e-01 6.55505538e-01
3.73818099e-01 -3.66536349e-01 5.02753437e-01 -1.10351300e+00
8.46278787e-01 8.53677988e-01 -5.89246526e-02 -5.64974010e-01
-5.81733823e-01 -2.89324448e-02 -2.74509192e-01 -1.47407711e-01
4.91145283e-01 -5.76244056e-01 -7.97164917e-01 1.38903165e+00
1.06227033e-01 -1.04270041e-01 6.12566806e-02 5.07315397e-01
1.17750335e+00 4.19481039e-01 -1.72501393e-02 -1.32542044e-01
1.31005239e+00 -1.34704494e+00 -5.93210757e-01 -3.13605905e-01
9.82494891e-01 -1.12328792e+00 7.55386114e-01 5.04508078e-01
-1.48376930e+00 -2.16367558e-01 -1.34166753e+00 -5.64733863e-01
-3.00834417e-01 -1.41490206e-01 4.79942769e-01 4.22564775e-01
-1.10685158e+00 3.76040757e-01 -8.58257174e-01 -1.14008099e-01
1.92218944e-01 2.43522227e-01 -1.37572046e-02 2.61179715e-01
-1.12636387e+00 1.25237989e+00 8.01600933e-01 -7.77694210e-02
-4.41681743e-01 -3.86566639e-01 -9.62507188e-01 -1.52213678e-01
2.71498144e-01 -6.24165475e-01 1.83114958e+00 -7.04420269e-01
-1.68888605e+00 1.05014443e+00 -2.10408419e-01 -7.60941684e-01
5.97785950e-01 -1.39394268e-01 -6.75990954e-02 -8.57007876e-02
3.59776244e-02 3.46178740e-01 4.19774175e-01 -1.17891741e+00
-7.91541934e-01 -6.26666024e-02 4.29626703e-01 3.02574933e-01
1.49080366e-01 4.19223994e-01 -3.29126000e-01 -6.14156365e-01
1.67661548e-01 -5.78573227e-01 -2.08768863e-02 -1.00484419e+00
-9.57430482e-01 -4.35766637e-01 6.35075986e-01 -9.98400331e-01
1.68883073e+00 -1.58631742e+00 8.69381666e-01 -1.10770129e-01
4.34004992e-01 1.53156355e-01 1.21389158e-01 4.24953014e-01
-1.54563725e-01 1.89379513e-01 -3.42949718e-01 -9.19614971e-01
8.57510269e-02 1.54677406e-01 -2.91854404e-02 -2.72339396e-02
4.26457256e-01 8.36251974e-01 -7.25408137e-01 -7.15845346e-01
-2.38650435e-04 5.70940785e-02 -5.90470731e-01 1.47817552e-01
-7.36022532e-01 3.48576725e-01 -2.21367404e-01 6.37473583e-01
2.02671885e-01 -1.89872608e-01 6.74952567e-01 1.41781241e-01
-3.07110667e-01 9.89049494e-01 -1.05024135e+00 1.41861129e+00
-1.74463496e-01 8.67898762e-01 6.09751821e-01 -8.70347679e-01
4.46164250e-01 4.39028680e-01 1.41969383e-01 -8.56890902e-02
4.86798406e-01 2.38851145e-01 2.13493809e-01 -5.14661849e-01
9.74580109e-01 -5.75228594e-02 -3.78701389e-01 5.91358125e-01
3.08590680e-01 -1.40044853e-01 8.01138699e-01 4.38146710e-01
1.35012412e+00 -3.80645581e-02 1.20948248e-01 -4.35324788e-01
5.35815656e-01 7.65321031e-02 3.96809280e-01 2.70296752e-01
-3.26180123e-02 8.11545193e-01 8.71105909e-01 -1.30955875e-01
-1.02057254e+00 -9.13987279e-01 -2.68617392e-01 1.45631337e+00
-4.46123958e-01 -7.36539423e-01 -8.91622305e-01 -6.44994080e-01
-6.48393780e-02 6.91939592e-01 -6.38607502e-01 2.53997177e-01
-1.19243884e+00 -6.50474906e-01 6.12925410e-01 5.11501849e-01
3.01153243e-01 -1.13395905e+00 -4.56494242e-01 1.72900528e-01
-5.32807887e-01 -1.11577785e+00 -1.77754670e-01 4.40314770e-01
-4.17897224e-01 -1.15730762e+00 -2.81647652e-01 -9.98501360e-01
2.93475568e-01 -2.07488984e-01 1.58053923e+00 9.19261724e-02
-2.44562235e-02 8.86643976e-02 -4.81683642e-01 -4.78506446e-01
-9.25412297e-01 7.01658785e-01 -3.37880999e-01 -3.80314887e-01
2.14074373e-01 -2.08125234e-01 -1.35112599e-01 -2.47580975e-01
-6.33330345e-01 2.12107703e-01 2.66816076e-02 6.69796705e-01
9.85450745e-02 -4.97299254e-01 3.45982164e-01 -1.30587637e+00
1.00971901e+00 -5.40254593e-01 -2.57283002e-01 3.42286289e-01
-2.10376363e-02 -3.36647034e-01 2.91520745e-01 9.50522348e-02
-9.91900265e-01 -3.00820529e-01 -2.13864401e-01 2.88907349e-01
-2.17277370e-02 6.36613309e-01 -2.42880583e-01 4.38230604e-01
6.08154058e-01 -3.91130865e-01 -1.56398669e-01 -5.85796595e-01
5.68896294e-01 7.87427425e-01 5.83800435e-01 -6.26853287e-01
1.86909437e-01 -2.32321024e-01 -6.36135578e-01 -7.95434237e-01
-9.90957975e-01 -3.87362897e-01 -9.05575395e-01 -5.77278323e-02
1.19731975e+00 -9.26872194e-01 -3.71640176e-01 6.23384297e-01
-1.38139355e+00 -6.84006870e-01 -3.23170453e-01 -8.44949335e-02
-4.43633497e-01 1.52598649e-01 -1.14818096e+00 -6.24947548e-01
-3.90435636e-01 -1.02783251e+00 8.34466577e-01 3.15326899e-01
-7.80463934e-01 -1.21991134e+00 2.98230201e-01 7.11828828e-01
7.80216679e-02 3.00827652e-01 1.09217107e+00 -9.72963452e-01
-3.42425704e-01 -1.32359579e-01 -1.16788194e-01 4.44033921e-01
-2.46155281e-02 4.49567974e-01 -1.10371816e+00 -1.12919092e-01
-2.33967341e-02 -5.77153385e-01 1.05568671e+00 4.13060963e-01
6.78084493e-01 -3.93340856e-01 -4.72319126e-01 1.20724581e-01
7.53930748e-01 9.43709239e-02 4.19256091e-01 4.56670851e-01
6.81868315e-01 5.64276874e-01 2.24157900e-01 1.12200096e-01
7.93104172e-01 5.27597189e-01 2.85761356e-01 1.78746954e-02
-9.50125828e-02 2.81618565e-01 3.49723667e-01 1.17459214e+00
-5.53769097e-02 -5.73492289e-01 -1.31645155e+00 4.95559037e-01
-2.00962782e+00 -9.88336742e-01 -2.14160949e-01 1.57887340e+00
1.20749319e+00 5.56454837e-01 4.24906820e-01 2.98527598e-01
5.91621161e-01 3.27632964e-01 -7.36886114e-02 -7.26085126e-01
-2.96818435e-01 1.99474722e-01 -1.07886821e-01 1.03992498e+00
-1.37130427e+00 9.42616880e-01 7.46325684e+00 4.48581725e-01
-7.51503468e-01 1.93134576e-01 7.45155871e-01 -3.03530067e-01
-3.08134168e-01 6.65801540e-02 -7.00147271e-01 3.74192268e-01
1.01227808e+00 -2.66441971e-01 6.68406785e-02 4.65352952e-01
-2.11522475e-01 -3.35070908e-01 -1.13957238e+00 1.31233677e-03
9.66030806e-02 -1.72880852e+00 -2.86462218e-01 -2.67176658e-01
6.97051585e-01 3.32394093e-01 -3.67912412e-01 5.10926902e-01
7.68801212e-01 -1.26090288e+00 8.04672956e-01 3.11217397e-01
2.95943230e-01 -5.78592002e-01 7.31073380e-01 4.20611084e-01
-9.17349875e-01 -7.17344806e-02 2.96792060e-01 -5.13856828e-01
3.61282080e-01 2.43673027e-01 -1.08100545e+00 5.31690061e-01
1.18308373e-01 4.06127334e-01 -5.06191075e-01 7.17948854e-01
-2.71153331e-01 7.00700223e-01 -3.59173983e-01 -1.08490102e-02
1.83178693e-01 1.55396983e-01 7.79870033e-01 1.70707321e+00
-3.08140457e-01 9.97799486e-02 6.37638032e-01 4.41882282e-01
-2.06876934e-01 -6.83123544e-02 -3.01505566e-01 1.76355734e-01
8.49023938e-01 1.25483775e+00 -8.16170514e-01 -6.01945221e-01
-1.04506195e-01 6.04087055e-01 5.23268759e-01 1.11144871e-01
-6.81130111e-01 -2.72030801e-01 3.58878672e-01 -1.20701671e-01
2.71225154e-01 -3.20694417e-01 -7.36855865e-01 -1.09982204e+00
7.99037665e-02 -1.06551671e+00 5.22689044e-01 -2.97625363e-01
-1.14562273e+00 9.45645511e-01 2.29205802e-01 -6.25134826e-01
-5.89271843e-01 -7.88092911e-01 -9.90485072e-01 8.44086945e-01
-9.29645061e-01 -1.34599233e+00 -1.50007373e-02 1.68557033e-01
7.30649352e-01 -6.53897896e-02 9.91104186e-01 1.07867144e-01
-9.82204497e-01 3.69263232e-01 -2.05637485e-01 3.16151738e-01
6.78484499e-01 -1.85347581e+00 6.48467541e-01 7.14059949e-01
-1.94899604e-01 5.79020262e-01 9.48270202e-01 -4.72682089e-01
-7.77075171e-01 -8.31201553e-01 1.37803352e+00 -8.67613494e-01
1.02471459e+00 -6.22856915e-01 -8.56026530e-01 1.18639076e+00
7.49705970e-01 -6.67058349e-01 8.51608694e-01 5.93010783e-01
-1.97290346e-01 6.45280659e-01 -8.33140612e-01 5.05723178e-01
7.70068944e-01 -4.96578783e-01 -1.07399571e+00 6.53730690e-01
8.02204847e-01 -1.09771240e+00 -1.05119312e+00 3.49319458e-01
3.62649649e-01 -1.01400244e+00 5.45109630e-01 -7.29582608e-01
9.48373914e-01 3.96680571e-02 -3.10493529e-01 -9.87923324e-01
-1.90933421e-01 -6.93384767e-01 -4.98470545e-01 1.41257501e+00
9.96621251e-01 -3.10654819e-01 5.76570094e-01 5.94102323e-01
-6.44111156e-01 -9.77667272e-01 -1.01815689e+00 -2.63199627e-01
5.30707777e-01 -3.01137984e-01 1.97428823e-01 9.28908050e-01
7.27219880e-01 1.07845438e+00 1.61705196e-01 -4.42395478e-01
2.27135926e-01 9.65414867e-02 5.72868705e-01 -1.14944160e+00
-6.00621104e-01 -8.70568693e-01 3.93279046e-02 -1.01927018e+00
5.04569449e-02 -1.04264176e+00 3.30160893e-02 -1.66449785e+00
1.16723202e-01 -7.73288757e-02 6.64585307e-02 4.44655776e-01
-3.84301007e-01 7.44148344e-02 5.42090416e-01 2.02363104e-01
-8.18532288e-01 1.71489403e-01 8.37945223e-01 -2.35159323e-01
-4.18410122e-01 7.96684623e-03 -7.71551192e-01 1.00505066e+00
8.61420393e-01 3.54296602e-02 -2.80328333e-01 -7.60140181e-01
-4.10237461e-02 -1.32610038e-01 1.85823534e-02 -8.88956547e-01
7.55965561e-02 2.97877878e-01 2.01566115e-01 -6.68000638e-01
3.95776898e-01 6.55595735e-02 -3.78746808e-01 1.07479677e-01
-5.39438605e-01 2.30153948e-01 4.28275079e-01 3.90081927e-02
-1.70942605e-01 -2.60079026e-01 5.99523842e-01 -3.88145298e-01
-1.89709648e-01 -4.56026405e-01 -6.41478837e-01 3.58016968e-01
1.16620088e+00 -5.03919758e-02 -9.34137404e-01 -1.74825899e-02
-1.12849915e+00 6.72369361e-01 4.07570958e-01 4.36388940e-01
1.32333726e-01 -9.81285453e-01 -8.84940863e-01 -8.97177607e-02
-3.55448037e-01 4.12106007e-01 -6.36227950e-02 8.12457561e-01
-4.98900145e-01 4.69723761e-01 9.15480778e-02 -6.34445667e-01
-1.59387100e+00 2.29031071e-02 3.01749974e-01 -9.69944656e-01
-3.81802827e-01 1.38398659e+00 -4.09041107e-01 -4.65314895e-01
2.87524581e-01 -5.00712335e-01 -4.51556712e-01 5.90415299e-01
5.09603560e-01 3.11320037e-01 3.05573434e-01 -5.85007191e-01
-2.45960906e-01 -4.59195673e-02 -5.84692240e-01 -3.65430743e-01
1.21049774e+00 -3.61582041e-02 -5.03977418e-01 9.71517146e-01
7.81593025e-01 -1.37490809e-01 -8.88364732e-01 2.26249616e-03
6.70692921e-01 2.43980378e-01 -6.98321760e-02 -9.39202130e-01
-3.06493402e-01 5.75771213e-01 -1.63469493e-01 8.74394357e-01
7.69640148e-01 -5.20480759e-02 7.47819126e-01 1.17943011e-01
1.47867039e-01 -1.35120010e+00 -1.87650040e-01 1.20422268e+00
9.56288815e-01 -1.00311267e+00 1.62343442e-01 -5.26391625e-01
-5.28039157e-01 1.05345070e+00 6.08481705e-01 -1.81759298e-01
4.25899476e-01 6.64802313e-01 2.14311436e-01 -1.55125797e-01
-1.28515494e+00 -4.45690602e-02 2.79185683e-01 4.51460510e-01
1.16451025e+00 2.18437895e-01 -3.91234577e-01 7.61675239e-01
-5.41341007e-01 -2.20063522e-01 5.38059294e-01 1.34440863e+00
-6.45746350e-01 -1.38385606e+00 -2.65965849e-01 5.64128757e-01
-3.81752551e-01 -1.64490446e-01 -1.04688919e+00 9.32028353e-01
1.66465491e-01 1.05802226e+00 2.67880738e-01 -5.23879290e-01
3.68987113e-01 4.59488451e-01 6.35751724e-01 -1.13653493e+00
-1.14060044e+00 1.03754736e-01 1.07999408e+00 -3.04942220e-01
-3.27040136e-01 -9.62761998e-01 -1.34247065e+00 -3.88803154e-01
-1.86938256e-01 1.20014928e-01 1.62005007e-01 1.21860611e+00
9.76611674e-02 9.33470011e-01 2.61584640e-01 -1.10451365e+00
-3.74356806e-01 -1.51977146e+00 -5.75625971e-02 5.42453900e-02
3.16497564e-01 -3.25379193e-01 -1.41474336e-01 2.00709313e-01] | [10.894290924072266, 9.471261978149414] |
2b5d7ecf-a508-476d-8256-015865d71e4f | feed-forward-source-free-latent-domain | 2207.07624 | null | https://arxiv.org/abs/2207.07624v1 | https://arxiv.org/pdf/2207.07624v1.pdf | Feed-Forward Source-Free Latent Domain Adaptation via Cross-Attention | We study the highly practical but comparatively under-studied problem of latent-domain adaptation, where a source model should be adapted to a target dataset that contains a mixture of unlabelled domain-relevant and domain-irrelevant examples. Furthermore, motivated by the requirements for data privacy and the need for embedded and resource-constrained devices of all kinds to adapt to local data distributions, we focus on the setting of feed-forward source-free domain adaptation, where adaptation should not require access to the source dataset, and also be back propagation-free. Our solution is to meta-learn a network capable of embedding the mixed-relevance target dataset and dynamically adapting inference for target examples using cross-attention. The resulting framework leads to consistent improvement on strong ERM baselines. We also show that our framework sometimes even improves on the upper bound of domain-supervised adaptation, where only domain-relevant instances are provided for adaptation. This suggests that human annotated domain labels may not always be optimal, and raises the possibility of doing better through automated instance selection. | ['Timothy Hospedales', 'Shell Xu Hu', 'Da Li', 'Ondrej Bohdal'] | 2022-07-15 | null | null | null | null | ['source-free-domain-adaptation'] | ['computer-vision'] | [ 5.90650320e-01 6.14262223e-01 -5.20594120e-01 -7.46343911e-01
-1.15665066e+00 -7.00974882e-01 5.96897423e-01 4.13585752e-02
-6.38232887e-01 1.14329767e+00 3.57382223e-02 -6.87746182e-02
-1.70965761e-01 -5.72993755e-01 -8.87811124e-01 -6.76902592e-01
1.43967569e-01 9.39517975e-01 5.56352176e-02 9.65566188e-02
-7.26784766e-02 3.68084610e-01 -1.30707467e+00 1.08714849e-01
7.31987894e-01 9.83814597e-01 -1.20073318e-01 6.28387749e-01
-2.97631472e-02 3.26238364e-01 -5.61521173e-01 -6.05974257e-01
4.12157834e-01 -3.12515914e-01 -9.55585301e-01 2.90118665e-01
5.20248771e-01 -1.85817704e-01 3.83761898e-02 1.02752566e+00
5.56800783e-01 3.12076449e-01 8.93686652e-01 -1.31136954e+00
-9.05017674e-01 4.74390775e-01 -4.17340398e-01 9.65983272e-02
-2.45742798e-02 5.21336049e-02 7.36853540e-01 -4.42290962e-01
7.69442916e-01 9.15026248e-01 3.69264305e-01 1.10807431e+00
-1.51600182e+00 -6.26091242e-01 6.05314851e-01 -6.80739209e-02
-1.05680788e+00 -8.41844499e-01 6.33754134e-01 -2.63991505e-01
6.91331089e-01 1.30755380e-01 -1.71783045e-01 1.63207924e+00
-4.13631529e-01 6.17203236e-01 8.28234196e-01 -6.61525130e-01
6.54570043e-01 1.00552678e+00 -9.23478752e-02 3.18513736e-02
2.87345827e-01 -6.30558729e-02 -3.44725013e-01 -4.78129804e-01
5.14387548e-01 -1.87416628e-01 -1.39985427e-01 -9.94232059e-01
-1.08190274e+00 7.23274410e-01 4.84070890e-02 1.05149381e-01
-3.37647378e-01 -2.82480687e-01 4.98783946e-01 5.75822771e-01
6.70126200e-01 5.36578953e-01 -1.11112225e+00 1.74459442e-01
-7.61112809e-01 1.35955811e-01 8.85957062e-01 1.35913324e+00
8.47000420e-01 -2.50240266e-01 7.63473287e-03 9.75236297e-01
5.06219007e-02 4.65516210e-01 5.25252759e-01 -1.01768970e+00
7.33315170e-01 3.41978937e-01 4.91443515e-01 -3.60653579e-01
-8.26103091e-02 -3.61533940e-01 -6.86022401e-01 1.72199637e-01
6.67657673e-01 -4.45191979e-01 -8.78274977e-01 2.30095768e+00
5.26799917e-01 -1.17902823e-01 2.80670434e-01 6.82312012e-01
1.00095063e-01 3.51264179e-01 4.78526175e-01 -3.24218690e-01
1.05645013e+00 -6.73649609e-01 -4.93404269e-01 -6.35990739e-01
4.77178007e-01 -1.99254319e-01 1.26476753e+00 2.17296287e-01
-9.81069505e-01 -3.28729540e-01 -1.00295079e+00 6.20637871e-02
-5.62060952e-01 -6.32018372e-02 3.53764474e-01 7.58171320e-01
-7.86195695e-01 3.75885189e-01 -5.99007368e-01 -6.13633037e-01
6.81501985e-01 5.66858470e-01 -7.05855727e-01 -1.76704884e-01
-1.15567172e+00 9.35838997e-01 6.15093350e-01 -2.77147293e-01
-6.61801994e-01 -6.08398736e-01 -7.72269070e-01 -4.50128168e-02
5.05298615e-01 -8.40654612e-01 1.18495774e+00 -1.51645136e+00
-1.55892861e+00 1.07593536e+00 -2.21767455e-01 -4.90160108e-01
7.43814945e-01 -1.08598202e-01 -6.46063566e-01 5.21531589e-02
7.73326680e-02 6.07184947e-01 1.05038428e+00 -1.14050746e+00
-8.11073124e-01 -5.15011489e-01 2.51470674e-02 3.43287528e-01
-6.67995155e-01 -2.46047303e-01 -3.18965852e-01 -2.48123020e-01
-4.01855618e-01 -9.15783346e-01 -3.53916675e-01 8.03390369e-02
-3.10565829e-01 5.59843592e-02 7.17050552e-01 -5.06396592e-01
9.27699387e-01 -2.32382536e+00 -2.52561532e-02 2.66689330e-01
-1.48695335e-01 2.49541715e-01 -2.14494362e-01 4.76600341e-02
-2.04363286e-01 1.47768095e-01 -4.37087864e-01 -4.55686331e-01
2.14161053e-01 2.74441749e-01 -4.61659312e-01 4.41313237e-01
4.00515944e-01 6.63557708e-01 -8.49757552e-01 -4.74722773e-01
4.90836538e-02 3.32627386e-01 -5.81769705e-01 2.86884129e-01
-5.67155182e-01 6.16817176e-01 -5.89740515e-01 4.37745601e-01
6.31529570e-01 -4.08090532e-01 2.44946435e-01 2.66130209e-01
4.78023499e-01 2.65226662e-01 -1.24961996e+00 1.71578312e+00
-6.84237719e-01 3.35487574e-01 3.40096653e-01 -1.15326488e+00
7.17210531e-01 3.44812959e-01 2.89393932e-01 -4.38521981e-01
-1.88309416e-01 1.46728456e-01 -3.86984706e-01 -3.62694412e-01
2.05661356e-01 -2.00041845e-01 -2.14019775e-01 4.47161645e-01
2.24313900e-01 2.96849757e-01 -2.74518341e-01 -3.53884026e-02
1.00104809e+00 3.16016078e-01 4.54430908e-01 -1.48491219e-01
4.92688239e-01 6.07259646e-02 6.86921775e-01 7.43720114e-01
-2.55326986e-01 5.59929132e-01 3.20315868e-01 -1.63300514e-01
-9.60599065e-01 -1.08373725e+00 -2.45068908e-01 1.52661896e+00
-1.24756023e-01 3.15417349e-01 -8.02001536e-01 -1.26478553e+00
2.85484944e-03 9.61954355e-01 -5.67232907e-01 -3.70254159e-01
-4.52754378e-01 -7.65366256e-01 1.39261141e-01 4.38447386e-01
2.13322923e-01 -8.88922930e-01 -3.77595991e-01 2.36939639e-01
-8.22823048e-02 -9.38239634e-01 -6.03411376e-01 7.47947812e-01
-8.70569348e-01 -8.02383065e-01 -6.90585613e-01 -6.01122141e-01
9.72660184e-01 -4.21958894e-01 1.19132543e+00 -6.07915998e-01
2.16495186e-01 6.03016376e-01 -5.23620173e-02 -3.75686049e-01
-4.83834296e-01 6.10450089e-01 1.61257267e-01 2.58322686e-01
6.95629478e-01 -7.65761971e-01 -4.07099545e-01 2.71203011e-01
-8.17452610e-01 -6.13940597e-01 6.51599288e-01 8.85922790e-01
5.63951194e-01 6.45750985e-02 1.03013289e+00 -1.49755943e+00
4.33815449e-01 -7.11249232e-01 -6.18036270e-01 4.74132806e-01
-8.74442279e-01 2.52262175e-01 7.82336831e-01 -9.26171780e-01
-1.20390558e+00 4.91834700e-01 1.90579876e-01 -4.44472790e-01
-6.20161295e-01 -1.59218565e-01 -8.71560752e-01 1.02491096e-01
1.05491304e+00 -1.68635696e-02 -2.40233704e-01 -4.39284295e-01
5.75856507e-01 1.00800455e+00 6.63494170e-01 -6.81963742e-01
7.30414689e-01 4.14355069e-01 -3.94585073e-01 -6.52510643e-01
-7.67387807e-01 -3.13644052e-01 -8.87339473e-01 4.50491965e-01
5.51892877e-01 -9.82094705e-01 -2.77114838e-01 -2.69784089e-02
-9.30274427e-01 -5.37253320e-01 -8.24022233e-01 1.96099937e-01
-8.06936145e-01 1.51825577e-01 -1.12886928e-01 -8.29725564e-01
-4.24588062e-02 -8.09544206e-01 9.69786942e-01 -2.27134172e-02
-4.77731884e-01 -1.22138774e+00 7.60990754e-02 2.08566487e-01
4.72962767e-01 1.54055238e-01 9.50657785e-01 -1.37242985e+00
-4.19847935e-01 -4.20416862e-01 -2.80187611e-04 4.19447631e-01
1.89937055e-01 -6.04005039e-01 -1.33255804e+00 -3.25724959e-01
-1.01091929e-01 -4.80535597e-01 6.43541694e-01 2.64751375e-01
1.12717402e+00 -6.32343531e-01 -5.69148779e-01 6.13217831e-01
1.22856176e+00 -3.87189933e-03 3.71110618e-01 4.05453444e-01
4.65640604e-01 7.82937825e-01 5.26388288e-01 3.45411658e-01
4.06904131e-01 7.16970623e-01 1.77841336e-01 6.25369102e-02
1.67165563e-01 -4.15101320e-01 1.19726926e-01 1.97586659e-02
4.89146739e-01 -4.22771275e-01 -6.74026847e-01 9.64876950e-01
-1.78991961e+00 -8.02673399e-01 4.53874350e-01 2.66458082e+00
1.14348269e+00 1.07642777e-01 4.26290542e-01 -3.60698968e-01
8.44188094e-01 -2.26393625e-01 -1.18807530e+00 -3.58447224e-01
8.66472051e-02 1.49797708e-01 8.77767146e-01 5.21822512e-01
-1.37394559e+00 6.74931407e-01 6.90206242e+00 6.43719792e-01
-1.03157556e+00 3.83929193e-01 8.13083053e-01 -2.66579270e-01
-4.11480814e-01 -1.95289239e-01 -9.48364139e-01 5.83181024e-01
1.44232476e+00 -1.29730389e-01 4.32911515e-01 1.21035373e+00
-3.90749395e-01 1.61282748e-01 -1.48431933e+00 7.05731511e-01
-1.41735762e-01 -7.62420654e-01 -2.63167143e-01 2.61647016e-01
7.01072633e-01 -1.26934469e-01 2.65771568e-01 5.10544181e-01
6.83997750e-01 -6.99866712e-01 2.83326864e-01 2.82989472e-01
1.09603643e+00 -7.65118897e-01 5.39762735e-01 6.28359437e-01
-5.06149054e-01 -2.59710670e-01 -5.20119548e-01 3.61794204e-01
-7.31518120e-02 6.03572309e-01 -8.66459668e-01 1.65383890e-01
5.54566205e-01 3.74539793e-01 -2.72491157e-01 6.92970395e-01
-1.60739776e-02 5.36773682e-01 -5.74886441e-01 3.23727876e-01
-2.71992773e-01 1.95613056e-01 3.53841960e-01 1.25627542e+00
1.41202435e-01 -1.29267707e-01 6.80490630e-03 7.48887062e-01
-1.81821629e-01 3.65992822e-02 -8.39196563e-01 1.47965429e-02
7.18428075e-01 9.48512912e-01 -1.88432023e-01 -2.92359084e-01
-3.30901206e-01 1.29860079e+00 5.40328622e-01 5.46094656e-01
-5.83208561e-01 -2.59985328e-01 7.72774637e-01 2.59998769e-01
4.17238325e-01 2.84275085e-01 -3.49964470e-01 -1.18395722e+00
1.41398311e-01 -8.07314932e-01 8.38098347e-01 -4.12667572e-01
-1.67049611e+00 4.56438899e-01 1.81232676e-01 -1.17002046e+00
-6.98812425e-01 -3.41113359e-01 -1.87491685e-01 1.07934082e+00
-1.62445772e+00 -1.03644180e+00 2.79335052e-01 7.79597938e-01
2.43121058e-01 -2.61124134e-01 1.18525863e+00 3.50495189e-01
-3.25580508e-01 1.06586123e+00 1.76689774e-01 -1.28556728e-01
1.10772407e+00 -1.37882602e+00 3.92397821e-01 7.72035182e-01
-1.93056948e-02 6.46334231e-01 6.58374906e-01 -5.44385731e-01
-8.66396248e-01 -1.48160481e+00 9.68469620e-01 -9.96614039e-01
3.36870342e-01 -6.56746268e-01 -1.19478643e+00 1.03905571e+00
-7.52255619e-02 2.95391917e-01 7.78832316e-01 3.37818086e-01
-5.41785479e-01 -2.78145403e-01 -1.68750989e+00 3.91486704e-01
1.07683790e+00 -6.30420327e-01 -4.83801961e-01 4.37368602e-01
6.99811101e-01 -2.91912824e-01 -6.83308244e-01 1.52984917e-01
1.61557958e-01 -4.72947747e-01 1.01978362e+00 -9.70677674e-01
7.41362423e-02 -1.31054536e-01 -2.32625097e-01 -1.30872083e+00
-3.98044497e-01 -5.54769695e-01 -2.51845032e-01 1.48485160e+00
7.78968453e-01 -7.42128551e-01 9.91919339e-01 1.42097604e+00
4.24787074e-01 -1.41941279e-01 -1.03067350e+00 -8.21709216e-01
3.20485920e-01 -1.85826063e-01 6.88601077e-01 1.22666848e+00
-4.04527970e-02 3.50126624e-01 -4.60413992e-01 5.54834604e-01
6.89275324e-01 -8.22719783e-02 6.70976520e-01 -1.16585743e+00
-5.20174384e-01 -1.36962399e-01 -1.19555853e-01 -9.51257825e-01
4.94635314e-01 -6.87771261e-01 1.80327684e-01 -1.06029665e+00
-4.38329168e-02 -5.79185665e-01 -6.00998521e-01 5.66235363e-01
-1.39409751e-01 2.24374961e-02 -3.97346497e-01 7.72273540e-02
-7.90298998e-01 2.14817509e-01 7.44245589e-01 -1.23761430e-01
-3.87700081e-01 4.05878842e-01 -1.18224299e+00 3.33430052e-01
8.04340601e-01 -7.06362784e-01 -7.37087667e-01 -2.43842676e-01
2.45718490e-02 -1.85490269e-02 1.83022499e-01 -6.86107278e-01
2.01396808e-01 -4.30952519e-01 4.69922423e-01 -3.38811837e-02
2.56178588e-01 -1.43698990e+00 1.69668004e-01 -1.60582811e-01
-9.49266613e-01 -5.21085382e-01 1.10723265e-01 9.21525419e-01
1.79625362e-01 -3.09998900e-01 9.84291553e-01 -2.53837049e-01
-6.01217389e-01 4.03190285e-01 1.18770972e-02 3.13721895e-01
1.10038900e+00 -2.84826458e-01 -2.26939917e-01 -3.55472296e-01
-9.23554957e-01 1.64722726e-01 6.93918586e-01 2.73165226e-01
-3.18965875e-02 -1.18033361e+00 -5.19008100e-01 3.50800484e-01
5.70029438e-01 1.78298011e-01 1.10734634e-01 2.33627498e-01
4.39218134e-01 2.25367427e-01 -2.63104793e-02 -3.66345316e-01
-8.30079913e-01 9.56626356e-01 4.31639999e-01 -1.24774709e-01
-4.31544423e-01 9.95977640e-01 3.84328246e-01 -9.18046176e-01
4.37005162e-01 -3.10454741e-02 6.32614642e-02 -1.53574005e-01
5.27847528e-01 4.16933931e-02 1.31898597e-01 -3.03700805e-01
-3.99719685e-01 1.88584253e-01 -2.91608453e-01 -2.16000557e-01
1.23322535e+00 -5.36647141e-01 2.87396491e-01 4.55150425e-01
1.15561438e+00 -1.30187403e-02 -1.64556205e+00 -5.41348517e-01
2.05945909e-01 -3.54118228e-01 -2.98316240e-01 -1.23031223e+00
-8.90528858e-01 8.12484682e-01 5.87131858e-01 -3.62410322e-02
1.28275406e+00 1.16711054e-02 4.99020785e-01 4.88822311e-01
3.78398776e-01 -1.27148414e+00 -3.93050045e-01 1.46660119e-01
4.78230029e-01 -1.46956444e+00 -9.02305767e-02 -1.57296367e-04
-8.91411245e-01 6.10396206e-01 7.15854943e-01 9.77260470e-02
5.09874761e-01 1.54009044e-01 1.19780257e-01 2.42787600e-01
-9.49881613e-01 3.17469202e-02 3.00494820e-01 1.33100820e+00
7.27467090e-02 1.99300535e-02 4.50142950e-01 7.02860713e-01
5.57693988e-02 8.46438482e-03 1.52086079e-01 5.94968498e-01
-1.59049764e-01 -1.40329814e+00 -2.68213123e-01 4.85648751e-01
-5.36495030e-01 6.55115694e-02 -4.48073775e-01 7.83016324e-01
9.46483687e-02 8.13442647e-01 -7.98905343e-02 7.83590674e-02
3.30679595e-01 6.21244133e-01 2.50512958e-01 -7.44157493e-01
-3.58802050e-01 -1.49494186e-01 1.14726245e-01 -3.03316057e-01
-2.20955417e-01 -8.08554828e-01 -8.03447604e-01 6.03162237e-02
-2.71481901e-01 1.86836094e-01 5.14343917e-01 8.44861031e-01
6.77388668e-01 1.78506196e-01 4.48473871e-01 -4.52744871e-01
-8.11465561e-01 -6.76183462e-01 -4.78558898e-01 5.34616351e-01
7.22624362e-01 -3.88876438e-01 -5.31329811e-01 3.31457615e-01] | [10.350144386291504, 3.2447855472564697] |
fc59d399-c6df-488c-a5af-fc72bee14e2a | unexpected-sawtooth-artifact-in-beat-to-beat | 1809.01722 | null | https://arxiv.org/abs/1809.01722v2 | https://arxiv.org/pdf/1809.01722v2.pdf | Unexpected sawtooth artifact in beat-to-beat pulse transit time measured from patient monitor data | Object: It is increasingly popular to collect as much data as possible in the hospital setting from clinical monitors for research purposes. However, in this setup the data calibration issue is often not discussed and, rather, implicitly assumed, while the clinical monitors might not be designed for the data analysis purpose. We hypothesize that this calibration issue for a secondary analysis may become an important source of artifacts in patient monitor data. We test an off-the-shelf integrated photoplethysmography (PPG) and electrocardiogram (ECG) monitoring device for its ability to yield a reliable pulse transit time (PTT) signal. Approach: This is a retrospective clinical study using two databases: one containing 35 subjects who underwent laparoscopic cholecystectomy, another containing 22 subjects who underwent spontaneous breathing test in the intensive care unit. All data sets include recordings of PPG and ECG using a commonly deployed patient monitor. We calculated the PTT signal offline. Main Results: We report a novel constant oscillatory pattern in the PTT signal and identify this pattern as a sawtooth artifact. We apply an approach based on the de-shape method to visualize, quantify and validate this sawtooth artifact. Significance: The PPG and ECG signals not designed for the PTT evaluation may contain unwanted artifacts. The PTT signal should be calibrated before analysis to avoid erroneous interpretation of its physiological meaning. | ['Martin G. Frasch', 'Hau-Tieng Wu', 'Chen-Yun Lin', 'Yu-Lun Lo', 'Yu-Ting Lin'] | 2018-08-27 | null | null | null | null | ['photoplethysmography-ppg'] | ['medical'] | [ 4.72370356e-01 6.94729984e-02 3.36376876e-01 -3.25081982e-02
-4.00330365e-01 -5.53708911e-01 -8.82215425e-02 4.81476605e-01
-3.93714309e-01 7.16137767e-01 -1.19716436e-01 -6.03553057e-01
-1.37426630e-01 -3.30529422e-01 -4.22384351e-01 -8.26281846e-01
-4.73232716e-01 2.56611139e-01 2.06720717e-02 3.33422959e-01
2.12528929e-01 5.30555129e-01 -8.74293864e-01 1.41897574e-01
9.49702978e-01 1.00806630e+00 -4.94413897e-02 7.76179254e-01
2.08880290e-01 4.29105163e-01 -9.26661611e-01 1.16473762e-02
5.18586099e-01 -1.04202676e+00 -2.06792593e-01 -2.15958312e-01
1.25781074e-01 -2.82831788e-01 2.96735257e-01 6.84881389e-01
1.03926766e+00 -2.84418941e-01 2.74836272e-01 -1.02325571e+00
1.62157640e-01 5.00364661e-01 -4.54517633e-01 4.25583154e-01
2.72444069e-01 3.26509356e-01 -2.80105751e-02 -4.12495881e-01
5.00391424e-01 3.22185576e-01 1.03447592e+00 2.63983101e-01
-1.65255904e+00 -4.49925184e-01 -7.75258362e-01 4.55321781e-02
-1.39327633e+00 -2.42302254e-01 9.18663025e-01 -5.19243956e-01
4.51021910e-01 5.95286250e-01 1.22654712e+00 9.17183220e-01
9.07738566e-01 -2.01708078e-01 1.60065997e+00 -4.53373462e-01
1.70695424e-01 5.57158709e-01 1.27767041e-01 3.12333316e-01
8.79041672e-01 9.73402038e-02 -2.99561769e-01 -5.30595005e-01
1.05371606e+00 -2.07004309e-01 -1.01454353e+00 -2.25170016e-01
-1.09640288e+00 3.42815608e-01 -4.48308587e-01 7.83290267e-01
-7.40985632e-01 4.35823016e-02 4.20939177e-01 4.91140723e-01
-5.82239069e-02 7.34176815e-01 -3.40111345e-01 -7.45189965e-01
-8.80193710e-01 -4.42064911e-01 1.17534161e+00 5.70948243e-01
8.73492062e-02 -1.43965751e-01 -1.18675120e-01 4.44960386e-01
1.77414060e-01 3.93395782e-01 7.83020735e-01 -8.70757401e-01
-7.63801038e-02 4.26434845e-01 2.57476628e-01 -1.00006521e+00
-6.23177469e-01 -1.85421452e-01 -7.20535457e-01 1.92719206e-01
6.00604236e-01 -2.00409546e-01 -4.83018249e-01 1.12440932e+00
2.40632802e-01 2.91831136e-01 -7.29488255e-03 1.06248558e+00
8.06404114e-01 -1.11036689e-03 1.15063051e-02 -9.23254132e-01
1.57419288e+00 -1.78264424e-01 -1.03750622e+00 3.56783301e-01
5.56151867e-01 -7.41720736e-01 1.02154028e+00 8.49742830e-01
-1.04659331e+00 -3.52119416e-01 -9.89030242e-01 4.32731748e-01
1.18750349e-01 1.53458923e-01 1.17533267e-01 1.10361910e+00
-7.64621258e-01 1.05727422e+00 -1.01799464e+00 -4.39990819e-01
-1.73607305e-01 9.20794532e-02 -3.39050829e-01 5.48899114e-01
-9.40499783e-01 1.12571204e+00 8.90398026e-02 3.83209616e-01
-1.90587059e-01 -8.99678230e-01 -6.66157782e-01 -6.97382167e-02
-8.13278258e-02 -4.29250866e-01 8.45718563e-01 -7.59687722e-01
-1.73797476e+00 9.50972438e-01 -6.59965575e-02 -2.29546770e-01
6.07267499e-01 -1.33310566e-02 -5.68691909e-01 5.45301318e-01
-2.98284411e-01 -6.53722763e-01 6.44892097e-01 -1.11022890e+00
1.69002518e-01 -3.48937005e-01 -8.16139340e-01 -1.48391411e-01
2.31454462e-01 -1.21798396e-01 1.00457929e-01 -4.44581538e-01
3.87703419e-01 -7.38316476e-01 -6.39962628e-02 -1.47113964e-01
2.58841738e-02 5.13234794e-01 2.40788728e-01 -7.57963419e-01
1.47894895e+00 -2.05572224e+00 -5.99023819e-01 4.86601651e-01
1.33227840e-01 1.87622920e-01 4.22276407e-01 5.37605762e-01
-3.54427457e-01 2.66277939e-01 -3.95361066e-01 1.18963391e-01
-3.52198392e-01 3.46503407e-02 -5.14983125e-02 1.07718396e+00
-2.61806935e-01 7.22810507e-01 -9.61846530e-01 -6.02040172e-01
3.88493210e-01 3.38408530e-01 7.97737688e-02 4.05594528e-01
6.79337263e-01 1.01036763e+00 3.92543487e-02 4.72671002e-01
7.99686670e-01 8.07017535e-02 5.44002652e-01 -5.64655244e-01
-4.08585995e-01 2.31134981e-01 -1.03638327e+00 1.61637306e+00
-1.81213155e-01 7.00350702e-01 -5.97348772e-02 -7.48646080e-01
1.27664196e+00 9.46410894e-01 8.38629305e-01 -8.75451982e-01
3.56184334e-01 5.66779137e-01 4.28376883e-01 -1.18849218e+00
-2.39272177e-01 -7.80726671e-01 5.46688735e-01 3.75944257e-01
-2.92933077e-01 -2.72414625e-01 -1.81630060e-01 -2.68704683e-01
8.67476940e-01 1.79221720e-01 7.24718094e-01 -7.39081085e-01
4.75477397e-01 -9.50541124e-02 5.09289443e-01 7.46447086e-01
-4.72847462e-01 9.40680981e-01 8.98160696e-01 -5.06706238e-01
-6.47819102e-01 -7.14882255e-01 -6.62956178e-01 -2.36649513e-01
1.13918543e-01 -1.44150436e-01 -4.07045960e-01 -8.90045464e-02
4.87453938e-02 6.19632125e-01 -6.04322970e-01 -1.14190958e-01
-5.49326956e-01 -7.81947434e-01 6.78410470e-01 2.39613682e-01
-1.00330478e-02 -9.18313265e-01 -1.49541640e+00 6.98282897e-01
-1.74450338e-01 -6.79011106e-01 -2.43209302e-01 1.45588219e-01
-1.29898977e+00 -1.35190070e+00 -8.31274152e-01 -1.23603232e-01
5.82848847e-01 -2.85513699e-01 1.05458379e+00 -7.14590475e-02
-7.85255730e-01 5.92021465e-01 -2.80389786e-01 -7.83280134e-01
-5.56584001e-01 -7.03051209e-01 2.44506672e-02 2.58038908e-01
5.23173094e-01 -9.38620567e-01 -1.02689993e+00 2.37010330e-01
-5.76623440e-01 -2.94769228e-01 4.74046320e-01 3.20691496e-01
5.28833628e-01 -5.64821899e-01 4.19246346e-01 -6.76685035e-01
9.59455192e-01 -3.43966991e-01 -4.91988957e-01 -1.58511758e-01
-9.07350957e-01 -2.45508075e-01 5.66500723e-01 -6.57563567e-01
-5.77484787e-01 -4.40506369e-01 1.62560716e-01 -4.46071535e-01
-2.14315251e-01 3.82208318e-01 1.92234814e-01 -1.15202874e-01
1.03984749e+00 2.22901374e-01 5.57939470e-01 -3.84226143e-01
-5.61191022e-01 5.51669180e-01 6.51785910e-01 -4.48712349e-01
2.83172935e-01 3.46987426e-01 3.00094754e-01 -8.88956308e-01
-1.24486694e-02 -5.72616518e-01 -4.95218903e-01 -4.08766985e-01
5.28680086e-01 -3.93266350e-01 -1.05865383e+00 2.28779271e-01
-1.03767002e+00 -3.59900713e-01 -5.70381165e-01 1.10249281e+00
-3.40602994e-01 7.23114371e-01 -4.78727072e-01 -1.10164344e+00
-7.38116384e-01 -8.24322164e-01 3.62839133e-01 1.75337732e-01
-5.19444287e-01 -1.08247209e+00 5.94617307e-01 -2.93407291e-01
7.01428175e-01 1.18671608e+00 4.16615695e-01 -4.34764206e-01
1.33119136e-01 -4.49566573e-01 2.98156083e-01 2.11024821e-01
5.20393670e-01 7.27787241e-02 -9.81166542e-01 -2.04950392e-01
8.87754202e-01 3.66207868e-01 7.61826336e-02 6.45934105e-01
9.06016469e-01 -1.23883680e-01 -6.80202916e-02 6.02409720e-01
1.69446623e+00 7.77412772e-01 1.03234649e+00 7.64707401e-02
1.23535044e-01 5.18223941e-01 3.09994698e-01 5.05079925e-01
-2.11242780e-01 3.82442266e-01 1.60755645e-02 -1.35375112e-01
2.21950844e-01 -1.03090527e-02 3.06880891e-01 6.67706311e-01
-5.26846886e-01 1.76994026e-01 -7.86480129e-01 4.40148890e-01
-1.33695006e+00 -5.88653743e-01 -1.01830995e+00 2.85671282e+00
8.49343419e-01 -1.16499037e-01 -2.92757414e-02 1.69310212e-01
6.36657655e-01 -4.64753777e-01 -1.03499301e-01 -7.98409820e-01
1.76739752e-01 5.84348321e-01 5.78048408e-01 2.99550533e-01
-3.87258112e-01 -3.53584975e-01 6.45723534e+00 -2.88885117e-01
-1.64055228e+00 1.18002407e-01 3.11751038e-01 2.20883489e-01
-1.40789241e-01 6.55069426e-02 -8.20909590e-02 7.61405885e-01
1.24645483e+00 -5.54202259e-01 -1.66131809e-01 4.58905578e-01
6.82563007e-01 -6.17145419e-01 -1.18217647e+00 1.33743584e+00
-4.72057834e-02 -9.55608428e-01 -5.84286511e-01 9.49484184e-02
-1.35786012e-01 -5.89285851e-01 -4.94483292e-01 -2.27635220e-01
-9.76004660e-01 -9.61574316e-01 2.96208978e-01 8.47419024e-01
1.14810157e+00 -1.16707794e-01 9.09113407e-01 6.17387928e-02
-8.87849808e-01 5.29545188e-01 -1.62649661e-01 -6.18190430e-02
2.18876645e-01 9.09489572e-01 -8.55016887e-01 4.95136291e-01
2.78450876e-01 3.20828825e-01 -3.17582130e-01 1.57791150e+00
-8.13205391e-02 8.53986204e-01 -6.14581227e-01 2.28409156e-01
-3.71179879e-01 -4.91153300e-01 7.79516637e-01 8.87084603e-01
6.51887953e-01 5.22425413e-01 -6.10613763e-01 1.22823548e+00
5.76154411e-01 3.24561656e-01 -7.02492535e-01 6.73281178e-02
2.70803094e-01 1.27370286e+00 -9.40498412e-01 -2.45796084e-01
-4.71797675e-01 6.82547987e-01 -7.64152765e-01 2.34772623e-01
-7.37815857e-01 -6.51583135e-01 -4.25127186e-02 5.48620760e-01
-3.12988609e-01 2.85500258e-01 -6.69238329e-01 -1.04606903e+00
3.79795521e-01 -6.29859030e-01 3.85360837e-01 -6.12506211e-01
-9.00872648e-01 5.96095264e-01 -1.46450137e-03 -1.64641285e+00
-3.32576409e-02 -3.14601690e-01 -9.26838160e-01 1.37748349e+00
-1.10769963e+00 -1.05973721e-01 -6.97660327e-01 3.16900730e-01
-2.64736235e-01 6.32316887e-01 1.07904780e+00 2.83263206e-01
-1.27909452e-01 2.37930432e-01 -1.69817194e-01 -2.31878534e-01
9.20562983e-01 -1.37753749e+00 -3.67367208e-01 5.34177363e-01
-7.39783049e-01 9.88769293e-01 8.97160053e-01 -7.57273495e-01
-1.30222988e+00 -3.57659519e-01 7.99273849e-01 -5.36791027e-01
4.10263866e-01 8.21198672e-02 -1.04636192e+00 2.99543023e-01
3.34696293e-01 2.11769819e-01 9.77413654e-01 -6.70269072e-01
2.53426552e-01 -2.97457755e-01 -1.50214112e+00 1.05512321e-01
1.67386428e-01 -4.06353623e-02 -1.04783022e+00 -2.28753000e-01
1.14047289e-01 -5.72868824e-01 -1.67886674e+00 3.95515054e-01
9.88813102e-01 -9.89629030e-01 3.63296509e-01 -4.91429400e-03
5.49023375e-02 -2.73310989e-01 7.47816563e-01 -1.10357869e+00
3.57733518e-02 -1.32669389e+00 1.74109235e-01 8.79567742e-01
1.91192850e-01 -1.31966162e+00 3.96374851e-01 1.11236751e+00
-6.40617609e-02 -4.71699297e-01 -7.73374438e-01 -6.64300978e-01
-2.84888536e-01 -1.06380783e-01 7.53728822e-02 8.88824582e-01
7.56763041e-01 -1.00954764e-01 -2.50546247e-01 -2.18801219e-02
7.34736741e-01 1.23130128e-01 4.75098789e-01 -1.23907435e+00
-2.21975088e-01 -3.07954196e-02 -3.76248717e-01 -1.26450524e-01
-9.95507538e-01 -3.86261821e-01 5.85765192e-05 -1.40156388e+00
-2.07714289e-01 -4.10416245e-01 -4.57839459e-01 5.45123033e-02
-9.46106166e-02 7.01205060e-02 -6.19861186e-02 1.53516471e-01
4.32511061e-01 -1.49783656e-01 1.29090297e+00 5.92141271e-01
-8.41179371e-01 -8.54683518e-02 -4.29994434e-01 2.37735733e-01
5.96746981e-01 -7.82926202e-01 -4.04728830e-01 4.37489331e-01
1.35471866e-01 5.27366817e-01 3.42315108e-01 -9.88882184e-01
1.62994578e-01 7.07440674e-02 4.16603208e-01 -4.36077476e-01
-1.80355221e-01 -1.11640942e+00 1.01973701e+00 9.82297122e-01
2.00025082e-01 4.10646424e-02 4.16149050e-01 -2.51243785e-02
-2.93203503e-01 -3.48385096e-01 9.03555751e-01 -4.35228199e-01
1.43833205e-01 -2.49406677e-02 -5.37711501e-01 3.95650677e-02
9.90538895e-01 -7.89620876e-01 -3.20112735e-01 -3.15382630e-01
-7.67263532e-01 -1.81990877e-01 6.17917120e-01 -1.97226420e-01
5.66156745e-01 -1.11588156e+00 -6.04852796e-01 4.00731862e-01
3.62149663e-02 -1.74688548e-01 4.67693329e-01 2.11211729e+00
-1.11206973e+00 2.33174384e-01 -1.92769453e-01 -6.23114884e-01
-1.00800538e+00 4.74710882e-01 7.79913366e-01 1.44769652e-02
-1.08855367e+00 3.09041858e-01 -1.62304029e-01 4.24125046e-01
6.58195391e-02 -5.91675699e-01 -2.72210270e-01 1.90754727e-01
5.72170258e-01 3.25977921e-01 2.75257647e-01 -5.66332825e-02
-3.99493843e-01 6.22840226e-01 7.26045609e-01 -1.10271208e-01
8.84961128e-01 -3.30549538e-01 -3.29909772e-01 8.74178171e-01
1.01933563e+00 2.36011073e-01 -6.19024038e-01 3.74103218e-01
-7.89244026e-02 -6.19310081e-01 -2.96789944e-01 -9.04527545e-01
-8.75730097e-01 6.36995256e-01 9.02264893e-01 6.44078672e-01
1.32110727e+00 -5.86053193e-01 4.10990238e-01 -1.72982767e-01
6.02318823e-01 -9.87644851e-01 -3.64956915e-01 -4.10694063e-01
9.69261527e-01 -6.91453278e-01 -1.10680357e-01 -3.58214617e-01
-5.80866933e-01 1.45091426e+00 -3.27046309e-03 -5.79870604e-02
6.91629052e-01 4.89789307e-01 5.20853162e-01 -1.68934152e-01
-4.92852628e-01 3.05501580e-01 -2.62076091e-02 5.57222366e-01
7.72136927e-01 -3.09527609e-02 -1.33804059e+00 8.00620854e-01
7.26116300e-02 6.19336605e-01 1.04480803e+00 1.12716043e+00
-1.55890524e-01 -7.80551136e-01 -6.11235678e-01 4.63229239e-01
-8.69438887e-01 -1.73226837e-02 -7.30733201e-02 8.34278107e-01
-8.69435538e-03 8.66131306e-01 -1.29819617e-01 2.19259694e-01
6.82760358e-01 4.36347842e-01 6.73947811e-01 -2.14006871e-01
-9.16398823e-01 3.70718360e-01 -5.87547794e-02 -6.75054669e-01
-5.69646120e-01 -3.95810515e-01 -1.10068583e+00 2.51824617e-01
-2.48311639e-01 6.11654818e-01 8.94417048e-01 3.24661970e-01
3.35815728e-01 4.90815163e-01 4.14980531e-01 -3.46096665e-01
-3.72662306e-01 -1.28707290e+00 -1.14140546e+00 5.81274867e-01
4.10964638e-01 -2.79831976e-01 -8.60824406e-01 3.05491507e-01] | [14.05119514465332, 2.9988596439361572] |
4b7ab365-85ae-4b0d-b336-2f1631e69435 | leave-one-out-kernel-optimization-for-shadow | null | null | http://openaccess.thecvf.com/content_iccv_2015/html/Vicente_Leave-One-Out_Kernel_Optimization_ICCV_2015_paper.html | http://openaccess.thecvf.com/content_iccv_2015/papers/Vicente_Leave-One-Out_Kernel_Optimization_ICCV_2015_paper.pdf | Leave-One-Out Kernel Optimization for Shadow Detection | The objective of this work is to detect shadows in images. We pose this as the problem of labeling image regions, where each region corresponds to a group of superpixels. To predict the label of each region, we train a kernel Least-Squares SVM for separating shadow and non-shadow regions. The parameters of the kernel and the classifier are jointly learned to minimize the leave-one-out cross validation error. Optimizing the leave-one-out cross validation error is typically difficult, but it can be done efficiently in our framework. Experiments on two challenging shadow datasets, UCF and UIUC, show that our region classifier outperforms more complex methods. We further enhance the performance of the region classifier by embedding it in an MRF framework and adding pairwise contextual cues. This leads to a method that significantly outperforms the state-of-the-art. | ['Tomas F. Yago Vicente', 'Minh Hoai', 'Dimitris Samaras'] | 2015-12-01 | null | null | null | iccv-2015-12 | ['shadow-detection'] | ['computer-vision'] | [ 5.62812567e-01 1.33489013e-01 -4.25003529e-01 -6.94442511e-01
-8.64185512e-01 -6.80572808e-01 4.60631520e-01 -1.23446226e-01
-3.44394535e-01 6.56098008e-01 -2.21629679e-01 -3.66125524e-01
3.93242180e-01 -4.44925874e-01 -8.32948208e-01 -8.82086635e-01
-1.35356337e-02 2.83628255e-01 8.95488381e-01 3.85436475e-01
2.95940697e-01 5.37280321e-01 -1.32898271e+00 3.74814153e-01
8.55901361e-01 1.28106248e+00 3.37747246e-01 7.05834687e-01
2.49605566e-01 7.59974301e-01 -4.23819005e-01 2.21036300e-01
1.96091473e-01 -2.77143896e-01 -8.90002847e-01 5.13476729e-01
7.85655439e-01 -2.76180655e-01 -2.06323117e-02 6.95746183e-01
-1.28037035e-02 2.11263627e-01 8.98547232e-01 -1.30769658e+00
-3.49094540e-01 6.70871735e-02 -7.20337451e-01 -4.99353893e-02
1.66819189e-02 -3.86652768e-01 1.07414258e+00 -8.28764081e-01
5.82904279e-01 9.12918448e-01 5.69631696e-01 1.19406573e-01
-1.53455281e+00 -3.70248020e-01 3.57246101e-01 -8.39004386e-03
-1.34445512e+00 -1.12685025e-01 4.49154973e-01 -5.40459573e-01
5.80129266e-01 2.27422595e-01 5.51665008e-01 7.66956925e-01
2.64237076e-01 1.06777823e+00 1.51074290e+00 -5.21420002e-01
4.10491407e-01 3.85739267e-01 2.96050996e-01 1.29859436e+00
-1.06157102e-01 4.13429551e-02 -4.53318298e-01 -4.10249650e-01
7.54046857e-01 5.29773682e-02 -3.37928474e-01 -9.64069545e-01
-1.01406682e+00 8.51229966e-01 5.27688324e-01 -1.56957135e-01
6.20399415e-02 3.74028981e-01 -8.10783505e-02 -1.90653592e-01
6.62623525e-01 1.37068495e-01 -5.12811542e-01 4.23387021e-01
-1.26194644e+00 1.15896896e-01 8.28902960e-01 7.42746592e-01
1.05056131e+00 -4.14357662e-01 -1.98787048e-01 9.23460960e-01
2.59035110e-01 5.24116933e-01 -1.35188416e-01 -1.14424968e+00
6.16278388e-02 5.59781194e-01 2.00590193e-01 -6.02225304e-01
-2.74348229e-01 -2.90513188e-01 -3.15358937e-01 6.20470583e-01
6.33930862e-01 -7.11281504e-03 -1.25976849e+00 1.30196154e+00
5.86504221e-01 4.03102547e-01 -1.88921958e-01 9.42070961e-01
3.78854781e-01 6.16548896e-01 -2.89932400e-01 6.88964650e-02
1.21222985e+00 -1.57145774e+00 -3.90195638e-01 -5.78059256e-01
4.08058614e-01 -7.48686492e-01 1.04001486e+00 2.84182459e-01
-4.89684969e-01 -4.21367675e-01 -1.00087917e+00 -5.37776239e-02
-4.01859611e-01 6.42721772e-01 6.15246356e-01 4.28211510e-01
-7.23679364e-01 5.54427743e-01 -8.73522997e-01 -1.72906414e-01
5.78288972e-01 3.79016757e-01 -3.96164834e-01 -5.00466786e-02
-5.23064554e-01 8.10081899e-01 2.81536072e-01 -2.84799915e-02
-8.97634983e-01 -6.11972868e-01 -9.54218626e-01 -1.29257098e-01
7.79468596e-01 -1.41687453e-01 1.18868721e+00 -1.06000483e+00
-1.19056273e+00 1.04208386e+00 -4.23449934e-01 -3.44303489e-01
3.22970778e-01 -1.74488559e-01 5.45256846e-02 6.39357194e-02
2.35547408e-01 7.30908573e-01 9.97896314e-01 -1.84240210e+00
-8.70896339e-01 -2.28666112e-01 6.57324940e-02 3.05106975e-02
7.75618355e-06 -1.07166015e-01 -8.74996305e-01 -5.09026885e-01
1.82932854e-01 -1.46809399e+00 -2.31552586e-01 1.89202547e-01
-5.41397750e-01 -1.55244181e-02 1.20257056e+00 -7.69536972e-01
9.72916424e-01 -2.19901562e+00 -1.81476161e-01 2.94867098e-01
-4.41730134e-02 -1.11397162e-01 1.82083994e-01 -1.21721849e-01
2.59410322e-01 -2.77537167e-01 -6.38738096e-01 -4.64701563e-01
-1.39053032e-01 4.32529151e-01 -3.79175097e-01 7.02846885e-01
2.72305578e-01 7.12684035e-01 -7.70140111e-01 -7.79567540e-01
1.93901896e-01 4.15229052e-01 -1.24796830e-01 1.75813317e-01
-2.65726119e-01 2.30867684e-01 -2.73369968e-01 7.93453872e-01
6.53189242e-01 -5.05247056e-01 9.13810506e-02 -1.67398244e-01
-1.04575634e-01 -1.69892721e-02 -1.10304749e+00 1.50243437e+00
-3.14462930e-01 9.57275569e-01 1.03381805e-01 -8.09035480e-01
5.98531544e-01 -1.87984452e-01 2.93229133e-01 -1.42002597e-01
-1.06193520e-01 1.10420957e-01 -3.95992219e-01 -2.92402357e-01
4.67068881e-01 8.26163888e-02 -1.12108923e-01 3.74289721e-01
-6.40921891e-02 -4.64124940e-02 -1.66944861e-01 1.28007680e-01
1.03099585e+00 4.39989656e-01 3.22730601e-01 -4.05195296e-01
2.72140831e-01 6.25867546e-02 5.15101373e-01 8.31613719e-01
-2.31254682e-01 8.13120782e-01 5.02835751e-01 -1.63159996e-01
-6.48919880e-01 -1.15149689e+00 -2.87534058e-01 1.23229396e+00
3.84511769e-01 -2.11370550e-02 -8.94391000e-01 -1.39703655e+00
4.32155490e-01 6.02515817e-01 -8.64835858e-01 2.19410315e-01
-4.94782388e-01 -5.42086899e-01 1.80662185e-01 8.79034162e-01
3.44337761e-01 -7.50169456e-01 -7.71600008e-01 -3.34795415e-01
-4.92985509e-02 -1.43209922e+00 -7.08206296e-01 6.63184285e-01
-6.41140461e-01 -1.09538031e+00 -6.41466379e-01 -9.34442580e-01
9.39777911e-01 5.44919074e-01 9.31385040e-01 9.91421845e-03
-4.89355057e-01 2.36843705e-01 -2.75448859e-01 -3.61608595e-01
-6.43188879e-02 8.18498507e-02 -2.59755880e-01 7.74940848e-02
-1.28928021e-01 5.60108759e-02 -6.02809072e-01 5.57391524e-01
-6.32110178e-01 1.21693805e-01 7.11512387e-01 9.62634563e-01
9.62442100e-01 -9.59241316e-02 -3.17122415e-02 -1.20914018e+00
-9.53233168e-02 -1.96646541e-01 -9.11324382e-01 6.50848866e-01
-6.32997394e-01 1.91012084e-01 3.97749960e-01 -2.96956062e-01
-1.05829310e+00 7.42977798e-01 6.63258612e-01 -3.81443888e-01
-2.19467565e-01 1.13864662e-02 -1.12820819e-01 -4.77861524e-01
4.53821272e-01 1.30952895e-01 -3.40225041e-01 -2.90966153e-01
5.15804589e-01 7.11411655e-01 5.97669244e-01 -4.16782051e-01
8.75681460e-01 8.45388412e-01 2.37552792e-01 -9.60259676e-01
-1.36943746e+00 -9.76073861e-01 -9.86056089e-01 -3.59163016e-01
1.03777885e+00 -8.43917310e-01 -2.30195835e-01 1.55908540e-01
-9.26654279e-01 -8.95797670e-01 -2.95935068e-02 1.37597084e-01
-5.67720473e-01 4.04565543e-01 -3.11767668e-01 -9.63340461e-01
1.39316276e-01 -1.25278473e+00 1.58198559e+00 3.46914113e-01
-2.35337391e-02 -1.08878660e+00 -2.96706706e-02 6.40964150e-01
-2.14665551e-02 3.30078930e-01 6.66482627e-01 -3.84257227e-01
-7.89299428e-01 -1.54526904e-01 -5.96546173e-01 5.13693035e-01
8.62784758e-02 -3.66679952e-02 -1.25297999e+00 -1.44071728e-01
-4.13385153e-01 -4.51765835e-01 1.42310333e+00 3.90490055e-01
1.40214860e+00 -1.21828511e-01 -6.56875789e-01 5.72985590e-01
1.53558314e+00 -1.26103997e-01 3.57076466e-01 1.71578065e-01
8.22955191e-01 6.12870932e-01 1.19665229e+00 9.30400044e-02
3.34711045e-01 8.56602371e-01 3.49657744e-01 -5.18860579e-01
-1.84717953e-01 3.25407982e-02 3.26465279e-01 -2.79632777e-01
1.14750192e-01 -1.87289238e-01 -8.76227498e-01 3.58905047e-01
-2.13530564e+00 -7.18278646e-01 -2.56100833e-01 2.16766429e+00
6.85244143e-01 2.75430754e-02 6.27297610e-02 -1.14680961e-01
3.18565190e-01 3.17717761e-01 -5.27703524e-01 -1.86844228e-03
-9.36593935e-02 4.81588989e-02 9.55411613e-01 8.01157355e-01
-1.72534883e+00 1.31715298e+00 6.45556784e+00 8.28733921e-01
-1.03526616e+00 -8.42098147e-02 8.17277133e-01 1.20059617e-01
1.17797531e-01 3.69026095e-01 -9.18171942e-01 2.87683040e-01
5.57272673e-01 7.37626314e-01 5.90083301e-01 1.12347043e+00
1.51100516e-01 -1.01780903e+00 -9.91802156e-01 6.64902151e-01
4.70758647e-01 -1.07725775e+00 -6.83375716e-01 1.37070343e-01
1.12519526e+00 1.34565517e-01 -2.58175731e-02 6.48382772e-03
2.60066181e-01 -9.60613072e-01 7.45895445e-01 4.69974846e-01
6.48029804e-01 -3.81704867e-01 5.47892869e-01 2.28756413e-01
-1.18545318e+00 -7.25342408e-02 -3.00935924e-01 3.83871049e-01
-1.10740200e-01 8.15691292e-01 -1.09249592e+00 1.36964694e-01
8.77475381e-01 4.40193146e-01 -7.09397554e-01 9.68062818e-01
-4.74151701e-01 6.56447411e-01 -2.34602734e-01 5.20250387e-02
7.36093000e-02 -3.70041788e-01 5.33056520e-02 1.43895555e+00
-1.42349884e-01 -5.28054088e-02 6.97890520e-01 7.91134596e-01
-1.51894605e-02 -2.25442484e-01 -3.82702857e-01 1.50833458e-01
3.08165610e-01 1.46640062e+00 -1.31272471e+00 -3.67254019e-01
-4.14649308e-01 1.58627057e+00 5.04877090e-01 5.07615805e-01
-9.89206314e-01 -3.23749334e-01 4.79894459e-01 5.09212427e-02
6.76460743e-01 -2.84804314e-01 -3.51939619e-01 -9.89076972e-01
1.71405196e-01 -4.68581259e-01 1.73265368e-01 -7.81052411e-01
-9.82167482e-01 1.57770246e-01 6.68336302e-02 -9.54909325e-01
5.81567250e-02 -8.75973582e-01 -4.63613421e-01 5.38088500e-01
-1.73236680e+00 -1.41706491e+00 -5.72772563e-01 2.69986361e-01
4.54482377e-01 2.44370118e-01 7.41861343e-01 -2.19773442e-01
-6.08913004e-01 4.14662004e-01 3.90483618e-01 2.41068959e-01
9.52179492e-01 -1.64518714e+00 5.00416011e-03 8.18805695e-01
2.09820211e-01 3.02650481e-01 5.91638565e-01 -5.82543492e-01
-9.68962371e-01 -1.26309657e+00 8.08271825e-01 -4.98395890e-01
4.94774967e-01 -6.94850862e-01 -6.76012874e-01 6.66174889e-01
-1.04523137e-01 5.84578037e-01 7.48691618e-01 6.07231818e-02
-4.37699854e-01 -1.17440000e-01 -9.34848964e-01 3.44337463e-01
7.92938590e-01 -7.00654864e-01 -1.19596645e-01 6.09210968e-01
4.26850438e-01 -4.58054721e-01 -6.87860727e-01 2.24910587e-01
7.43390739e-01 -1.04854989e+00 1.03470659e+00 -3.67166966e-01
2.73318231e-01 -6.51916146e-01 -3.61030459e-01 -1.15235746e+00
-1.49744928e-01 -9.95768756e-02 -2.87936181e-01 1.10780764e+00
6.05589747e-01 -3.51048440e-01 1.17042243e+00 3.47954571e-01
-6.99391887e-02 -1.15599334e+00 -6.33389592e-01 -9.92798924e-01
-3.57652277e-01 -2.69909650e-01 -1.13238813e-02 5.72842538e-01
-3.39460969e-01 1.76404551e-01 -3.18214267e-01 5.67383230e-01
7.40662336e-01 6.42430305e-01 7.92499900e-01 -1.07612479e+00
-4.62793469e-01 -8.13199803e-02 -1.01308830e-01 -1.08020306e+00
5.00182390e-01 -6.78269207e-01 6.38392508e-01 -1.63068628e+00
5.71177423e-01 -6.09505177e-01 -1.17008530e-01 8.71224523e-01
-4.35341984e-01 4.75566894e-01 1.14852302e-01 1.80624932e-01
-8.31653297e-01 3.77305299e-01 9.37632740e-01 -1.64550364e-01
-1.60194486e-01 1.49319530e-01 -2.55529195e-01 8.48670363e-01
7.37775326e-01 -6.06432021e-01 -3.27018172e-01 -1.48217782e-01
-3.44498724e-01 -1.56630382e-01 6.47924364e-01 -9.22262847e-01
8.64794329e-02 -4.10385936e-01 7.04568684e-01 -6.37920916e-01
6.65396452e-01 -8.77191603e-01 -3.99244994e-01 2.81748742e-01
-2.28761226e-01 -4.89158154e-01 -8.33424330e-02 8.57180536e-01
2.77763326e-02 -1.25755668e-01 1.04478371e+00 8.61483440e-02
-7.17061102e-01 -3.70533066e-03 -1.61529884e-01 -1.27825320e-01
1.37108803e+00 -2.87632763e-01 -2.60362923e-01 -2.02481478e-01
-4.84561324e-01 2.75407195e-01 7.60157943e-01 1.55226380e-01
5.87585211e-01 -1.02355909e+00 -2.51022518e-01 2.09634945e-01
3.29782814e-01 3.35273445e-02 -1.70515791e-01 8.97363007e-01
-5.16701341e-01 2.53737181e-01 2.73487955e-01 -8.74299884e-01
-1.75731862e+00 2.82467961e-01 4.16459441e-01 -2.46981010e-01
-3.50996077e-01 1.01664257e+00 4.04128194e-01 -4.41486746e-01
4.34179664e-01 -4.06753361e-01 5.52727133e-02 -1.32528260e-01
2.37585977e-01 4.07012671e-01 -1.73798114e-01 -5.93340516e-01
-5.81185877e-01 6.14909351e-01 1.18786409e-01 -2.76887476e-01
1.05835557e+00 7.62237161e-02 -9.72717926e-02 5.64083874e-01
1.42799282e+00 1.52109593e-01 -1.70329368e+00 -1.01682842e-01
4.65667658e-02 -6.12274289e-01 2.51160771e-01 -8.39766324e-01
-8.59597087e-01 6.17140114e-01 7.27723539e-01 -8.53042752e-02
9.70859230e-01 3.39626282e-01 6.04135036e-01 3.26448709e-01
3.37240905e-01 -1.26041710e+00 2.51693010e-01 5.26389241e-01
5.16933382e-01 -1.65636659e+00 3.20209175e-01 -8.02994311e-01
-7.86757648e-01 1.10924053e+00 4.46868151e-01 -3.11608732e-01
8.48016798e-01 3.97574902e-01 2.12773457e-01 3.69241871e-02
-4.06444132e-01 -3.05709541e-01 7.43108511e-01 4.13142651e-01
3.12110245e-01 2.29060262e-01 8.81494060e-02 2.09818736e-01
3.51518728e-02 -1.91470146e-01 1.34653568e-01 1.01929510e+00
-8.13606501e-01 -1.06996799e+00 -5.97707510e-01 4.81220841e-01
-1.57704264e-01 1.23274013e-01 -7.51339257e-01 7.21508861e-01
3.33950847e-01 9.86573875e-01 3.13464776e-02 -3.76055509e-01
-4.58721146e-02 1.60119880e-03 5.84166765e-01 -6.44413054e-01
-2.57639855e-01 2.25381359e-01 4.16655429e-02 -7.15426564e-01
-3.23004186e-01 -8.62546980e-01 -1.42789125e+00 3.39922726e-01
-6.54679596e-01 -6.20114096e-02 7.02230155e-01 1.00987065e+00
1.67173725e-02 4.03266519e-01 7.34216094e-01 -9.40712452e-01
-3.96295965e-01 -5.84119558e-01 -7.24653184e-01 1.55843377e-01
4.67023998e-01 -6.35032058e-01 -2.99365729e-01 1.99183166e-01] | [9.408207893371582, 0.8174193501472473] |
ab7235ca-68ec-48cd-9884-5ed45ce49332 | taguchi-based-design-of-sequential | 2207.10992 | null | https://arxiv.org/abs/2207.10992v1 | https://arxiv.org/pdf/2207.10992v1.pdf | Taguchi based Design of Sequential Convolution Neural Network for Classification of Defective Fasteners | Fasteners play a critical role in securing various parts of machinery. Deformations such as dents, cracks, and scratches on the surface of fasteners are caused by material properties and incorrect handling of equipment during production processes. As a result, quality control is required to ensure safe and reliable operations. The existing defect inspection method relies on manual examination, which consumes a significant amount of time, money, and other resources; also, accuracy cannot be guaranteed due to human error. Automatic defect detection systems have proven impactful over the manual inspection technique for defect analysis. However, computational techniques such as convolutional neural networks (CNN) and deep learning-based approaches are evolutionary methods. By carefully selecting the design parameter values, the full potential of CNN can be realised. Using Taguchi-based design of experiments and analysis, an attempt has been made to develop a robust automatic system in this study. The dataset used to train the system has been created manually for M14 size nuts having two labeled classes: Defective and Non-defective. There are a total of 264 images in the dataset. The proposed sequential CNN comes up with a 96.3% validation accuracy, 0.277 validation loss at 0.001 learning rate. | ['Kayode Owa', 'Garima Joshi', 'Isibor Kennedy Ihianle', 'Renu Vig', 'Pushkar Bharadwaj', 'Tanya Aggarwal', 'Krishan Kumar Chauhan', 'Manjeet Kaur'] | 2022-07-22 | null | null | null | null | ['defect-detection'] | ['computer-vision'] | [ 1.57658309e-01 4.24056463e-02 2.47394100e-01 -2.10004643e-01
-1.34774387e-01 -2.69683450e-01 -1.01950698e-01 4.87348229e-01
-2.34675765e-01 4.19090718e-01 -5.83535373e-01 -1.70920312e-01
-3.59005839e-01 -9.87463117e-01 -4.73262101e-01 -7.59779096e-01
2.76585609e-01 1.05850741e-01 3.77219141e-01 -3.18715096e-01
7.32053697e-01 5.87681353e-01 -1.64112973e+00 2.12005436e-01
7.75799751e-01 1.14067388e+00 4.90922242e-01 4.42608386e-01
1.67584941e-01 5.36277175e-01 -6.82514191e-01 -2.06841856e-01
2.20740408e-01 -2.83712924e-01 -6.63516819e-01 2.94539481e-01
-1.49402201e-01 -4.15086061e-01 1.25299081e-01 9.88565087e-01
3.95637900e-01 2.27989390e-01 5.99839926e-01 -9.20031190e-01
-5.42269826e-01 1.13408200e-01 -5.26855409e-01 -9.76524949e-02
-1.71828151e-01 8.90790522e-02 5.74010253e-01 -6.25529706e-01
1.95908234e-01 9.27629471e-01 6.92842543e-01 2.34826297e-01
-8.34466994e-01 -5.17791688e-01 -6.58757091e-01 1.51969388e-01
-8.04407179e-01 3.03683046e-04 8.78611445e-01 -4.32763517e-01
8.01386774e-01 -8.68968368e-02 6.75333679e-01 1.82656631e-01
9.51220393e-01 3.41402292e-01 9.00693357e-01 -6.23017848e-01
1.85311958e-01 1.20710529e-01 2.42439345e-01 8.02085578e-01
6.63316309e-01 1.17734045e-01 -1.51614728e-03 2.72086143e-01
8.56403053e-01 8.13725218e-02 -8.43433216e-02 -1.36558235e-01
-4.95984942e-01 8.84151578e-01 3.55359703e-01 3.95922363e-01
-4.01859164e-01 -1.45518586e-01 7.34830081e-01 4.60469604e-01
2.10179642e-01 6.74260318e-01 -7.09257185e-01 -1.68858364e-01
-6.26064479e-01 -2.21597441e-02 4.60135162e-01 4.79089677e-01
4.38313156e-01 2.31818229e-01 3.97389114e-01 1.03626812e+00
3.44931573e-01 1.25773922e-01 4.79392201e-01 -7.66931593e-01
1.55311182e-01 9.31721866e-01 -6.14438914e-02 -1.46392882e+00
-4.03939098e-01 -2.32417405e-01 -9.59687412e-01 6.66948080e-01
2.17074081e-01 -1.67605266e-01 -1.04780817e+00 1.05553722e+00
3.22971940e-01 -8.10030222e-01 -1.19941972e-01 7.43509412e-01
5.25375903e-01 6.23557329e-01 -1.33145988e-01 1.10045247e-01
1.17790723e+00 -7.97585785e-01 -8.48966062e-01 1.70036539e-01
5.30075490e-01 -1.09535813e+00 8.83137763e-01 9.13239479e-01
-9.65047300e-01 -9.07379746e-01 -1.61050987e+00 2.03999177e-01
-4.43827301e-01 7.61884093e-01 6.23622894e-01 6.46791637e-01
-5.65442920e-01 9.54046249e-01 -1.04694319e+00 -1.59058958e-01
3.93505484e-01 6.53546333e-01 -4.43238199e-01 -2.11314678e-01
-6.82635546e-01 8.44621122e-01 3.58288825e-01 8.09731185e-01
-6.12635970e-01 -2.48611569e-01 -6.38982296e-01 -8.70807618e-02
1.00654483e-01 -2.75462270e-02 1.12036741e+00 -7.60088325e-01
-1.42406535e+00 4.88670945e-01 5.57150245e-01 -1.63600072e-01
2.80606568e-01 -4.21947062e-01 -2.70232856e-01 2.09386989e-01
-1.16847962e-01 1.83969215e-01 5.88762403e-01 -1.16633940e+00
-5.34456730e-01 -3.39615434e-01 9.34724286e-02 -3.78535390e-01
-1.67803913e-01 1.49757847e-01 7.11292103e-02 -5.74423015e-01
6.13993943e-01 -8.42053890e-01 -1.09772228e-01 1.28434747e-01
-4.48155880e-01 -1.95872590e-01 1.03634894e+00 -9.72480893e-01
7.51269698e-01 -1.99058306e+00 -4.61942911e-01 1.21022694e-01
-1.34831712e-01 4.72255588e-01 2.81390935e-01 5.12513876e-01
-1.14497110e-01 -7.34178051e-02 -1.64904237e-01 1.20651387e-01
-3.15328389e-01 7.41955414e-02 6.05243742e-01 5.05782068e-01
4.52629745e-01 2.18984306e-01 -4.07849401e-01 -2.98748642e-01
3.34963441e-01 3.94481987e-01 -4.12214726e-01 2.84373045e-01
1.31494582e-01 8.13039616e-02 -4.40906763e-01 1.00729489e+00
7.24238634e-01 8.40822011e-02 -2.25337625e-01 -6.24964595e-01
-7.82076046e-02 -1.79358482e-01 -1.16372788e+00 1.36886883e+00
-6.07371151e-01 5.14750659e-01 1.74485549e-01 -1.31948864e+00
1.29505038e+00 5.12567997e-01 3.45106184e-01 -5.02625823e-01
5.34048557e-01 5.30053794e-01 1.50128826e-01 -1.00159919e+00
4.28877741e-01 -2.03637108e-01 1.30664095e-01 5.09110726e-02
-3.93922254e-02 -1.54744714e-01 2.50523359e-01 -2.75731325e-01
8.31060708e-01 2.06487566e-01 -1.59214661e-01 -2.57408142e-01
3.32383871e-01 1.89533979e-01 6.53440654e-01 -6.76141400e-03
5.50246872e-02 6.13416672e-01 3.39147568e-01 -6.43189847e-01
-1.25910616e+00 -4.32060629e-01 -1.57991827e-01 1.70117304e-01
-3.73366997e-02 3.98777753e-01 -6.83676422e-01 -3.30913693e-01
-6.12641983e-02 3.39574844e-01 -4.90892231e-01 -5.37782729e-01
-4.82865840e-01 -6.05850637e-01 1.77570865e-01 5.87492585e-01
8.23520899e-01 -1.34013450e+00 -9.99530911e-01 4.36128855e-01
4.67857867e-01 -6.99701369e-01 8.41037631e-02 4.08342212e-01
-1.26699686e+00 -1.51667202e+00 -5.01859188e-01 -9.95005965e-01
1.02008283e+00 1.85973108e-01 6.77040040e-01 5.39449811e-01
-7.82459557e-01 -1.74630135e-01 -5.39556026e-01 -4.37078834e-01
-5.91818631e-01 -7.80522451e-02 -2.05631301e-01 -3.07589859e-01
2.69427538e-01 -2.29922593e-01 -7.29426205e-01 3.09469372e-01
-9.16506350e-01 -2.12923631e-01 8.94871831e-01 1.15452015e+00
2.62564003e-01 8.49437892e-01 8.78109872e-01 -8.10224593e-01
6.05375171e-01 -2.00414866e-01 -6.47814691e-01 2.22911742e-02
-8.66082013e-01 -2.22846001e-01 6.25125647e-01 -1.44914389e-01
-1.10066950e+00 -9.72748995e-02 -1.63986176e-01 -8.44712630e-02
-3.12032521e-01 6.94473505e-01 -1.43322840e-01 -9.41012576e-02
2.89559543e-01 -2.27094024e-01 3.33650559e-01 -7.70495474e-01
-5.39101481e-01 8.23475182e-01 4.77212876e-01 -3.11488271e-01
5.87833643e-01 -8.62970650e-02 1.21866554e-01 -9.39025879e-01
-4.91169453e-01 -3.27722013e-01 -5.73887944e-01 -4.47540700e-01
8.81404996e-01 -3.68580937e-01 -6.65072739e-01 8.85131478e-01
-1.07410336e+00 -4.58528548e-02 1.68736741e-01 6.65842772e-01
6.10686168e-02 4.30294603e-01 -8.85108292e-01 -8.79994214e-01
-6.14313722e-01 -1.26847744e+00 5.43869734e-01 3.70466113e-01
-1.06313460e-01 -5.74082077e-01 -4.17559296e-01 6.68734670e-01
4.56572294e-01 5.23033977e-01 1.15959275e+00 -2.39318401e-01
-3.03583056e-01 -7.83301890e-01 -1.48294181e-01 1.05006897e+00
6.54576659e-01 4.30004805e-01 -6.15301669e-01 -3.20921630e-01
3.35719496e-01 -5.42178094e-01 5.16862750e-01 4.48182404e-01
9.93279278e-01 -1.31106805e-02 9.88715142e-02 -4.92833368e-03
1.79888403e+00 9.79285300e-01 7.00832725e-01 4.89925563e-01
4.71204817e-01 8.07612777e-01 1.18187153e+00 2.75229305e-01
-2.63884574e-01 1.59528583e-01 6.98732555e-01 -2.40297094e-01
3.24479267e-02 1.15494341e-01 -8.22868496e-02 8.56127918e-01
-2.06258848e-01 -1.28207728e-01 -7.58982599e-01 7.46701002e-01
-1.11516237e+00 -6.39717698e-01 -4.18979228e-01 2.12776995e+00
7.53892899e-01 4.36399341e-01 -3.36204469e-01 1.11359811e+00
6.39324963e-01 -7.39718020e-01 -2.98661172e-01 -6.91713870e-01
3.11723471e-01 2.95847982e-01 4.44194913e-01 2.50134412e-02
-9.50545371e-01 3.05074334e-01 4.95078230e+00 5.21324992e-01
-1.27028799e+00 -4.32081550e-01 5.87381899e-01 2.44267628e-01
2.83649892e-01 -2.10894391e-01 -4.21495110e-01 4.62066680e-01
5.34060001e-01 7.07355440e-01 -2.56276503e-02 6.77566707e-01
3.40600640e-01 -6.06530249e-01 -5.78215539e-01 6.18397474e-01
-1.95660651e-01 -1.10413671e+00 -4.45343137e-01 -9.06185955e-02
6.92116618e-01 -4.69649017e-01 -2.22211048e-01 -5.29556395e-03
-1.50175184e-01 -7.91274369e-01 5.74774265e-01 3.05055022e-01
4.27476525e-01 -9.99819875e-01 1.44704521e+00 6.66678995e-02
-7.95025468e-01 -2.47610986e-01 -4.98411119e-01 -9.28625539e-02
-1.03451893e-01 7.68833935e-01 -6.30993068e-01 5.92437744e-01
8.87796283e-01 2.08412036e-01 -2.35841349e-01 1.27259064e+00
-1.56090995e-02 7.06404984e-01 -3.14796031e-01 -1.98571309e-02
1.31281137e-01 -3.72579366e-01 -6.03946149e-02 4.99625653e-01
4.46264386e-01 -1.12606950e-01 -2.38008667e-02 6.43320084e-01
2.55963385e-01 1.44950464e-01 -4.72715616e-01 -1.17391936e-01
3.16663921e-01 1.23280144e+00 -1.18442965e+00 2.16741443e-01
-2.71908820e-01 7.05205739e-01 -2.92897791e-01 -1.48609385e-01
-6.14744008e-01 -8.00653338e-01 1.97707042e-01 2.58387953e-01
3.49005848e-01 -2.85784364e-01 -4.18188840e-01 -2.36852169e-01
7.13313669e-02 -8.58370841e-01 -3.93422246e-02 -6.26645029e-01
-9.66855049e-01 3.36474895e-01 -1.70357391e-01 -1.15468657e+00
1.72566295e-01 -8.56717229e-01 -6.61381721e-01 6.79895997e-01
-1.08685720e+00 -1.02697611e+00 -4.36211497e-01 -9.07962546e-02
9.62321401e-01 -2.76808918e-01 6.44269228e-01 5.13758481e-01
-8.95947337e-01 3.05098355e-01 3.60412598e-02 1.24981299e-01
3.64348203e-01 -1.05179083e+00 -1.41463071e-01 7.56428599e-01
-5.59537411e-01 3.76046658e-01 8.35933328e-01 -6.79221034e-01
-1.46460974e+00 -7.95420647e-01 6.68754339e-01 4.58923578e-01
2.08562255e-01 9.08138305e-02 -9.47611392e-01 7.37678409e-02
1.83402225e-01 -1.39712632e-01 3.64753276e-01 -1.99241325e-01
4.05948609e-01 -1.24623939e-01 -1.51567006e+00 6.76291883e-02
8.10508579e-02 -6.64169565e-02 -3.41579676e-01 7.83294290e-02
2.64047176e-01 -2.73910373e-01 -1.15441501e+00 4.53858882e-01
6.43981516e-01 -9.26747084e-01 6.46650791e-01 -1.56229988e-01
8.48780036e-01 -1.99359015e-01 2.59500265e-01 -9.51117694e-01
-1.18669242e-01 -1.30102426e-01 4.40979570e-01 1.34181333e+00
4.27795798e-01 -5.44472158e-01 8.66397858e-01 5.65410554e-01
-5.98004222e-01 -1.21175230e+00 -4.63607669e-01 -6.05005264e-01
-2.01284736e-01 -6.87026829e-02 2.63838321e-01 7.70927489e-01
-3.31417710e-01 8.13032538e-02 5.52785322e-02 1.69357538e-01
4.75101650e-01 -8.23286250e-02 4.26657915e-01 -1.44770789e+00
-1.01538494e-01 -7.08136335e-02 -5.13173044e-01 -3.53691250e-01
-5.73550165e-01 -1.74539387e-01 2.60346025e-01 -1.56334341e+00
-1.27906561e-01 -6.48444891e-01 -2.16529280e-01 8.37316871e-01
1.31884992e-01 2.85674095e-01 -1.98886126e-01 -3.42160347e-03
2.67665416e-01 4.49355066e-01 1.64400959e+00 -2.18880251e-01
-7.10894093e-02 2.65764415e-01 -2.78627336e-01 5.92347682e-01
1.34514999e+00 -4.23567712e-01 -5.39333820e-01 -4.39471871e-01
2.15425193e-01 1.42078653e-01 2.24116474e-01 -1.14689171e+00
-1.42734945e-02 1.03632979e-01 3.67012829e-01 -7.70694494e-01
1.26656130e-01 -1.09179056e+00 1.18086368e-01 7.94625759e-01
-1.67371612e-02 1.53242171e-01 2.35099763e-01 2.46925473e-01
-4.85185862e-01 -7.95101166e-01 9.31308627e-01 -1.31580770e-01
-4.11913067e-01 -1.25595510e-01 -2.88158506e-01 -7.11637020e-01
9.71222579e-01 -7.62381792e-01 9.03380215e-02 6.63417801e-02
-4.20528650e-01 -7.51667097e-02 2.97910571e-01 2.34731749e-01
7.50636399e-01 -9.80651319e-01 -3.99488509e-01 4.04305577e-01
-2.58072793e-01 3.22849244e-01 4.69778478e-01 6.66845083e-01
-1.19236898e+00 2.38174930e-01 -6.33633673e-01 -3.41344684e-01
-1.30515218e+00 2.16841385e-01 2.31215477e-01 -8.51705074e-02
-4.66910064e-01 7.86277533e-01 -6.17034078e-01 -1.19203851e-01
-2.41201557e-02 -7.07039177e-01 -4.47471172e-01 -1.16279364e-01
-6.00413978e-02 7.14082301e-01 5.57865977e-01 -1.73535854e-01
1.04395963e-01 5.43277383e-01 -1.17527880e-01 5.06160140e-01
1.53491998e+00 1.72665522e-01 -9.71517414e-02 2.43714944e-01
1.19251871e+00 -2.47505978e-01 -1.11124825e+00 4.87101048e-01
-1.66554470e-02 -3.20911497e-01 4.01455909e-01 -8.79956901e-01
-1.55271995e+00 9.35636222e-01 8.47889245e-01 4.55086946e-01
1.18698382e+00 -4.59464639e-01 1.00067997e+00 2.87878931e-01
3.87817442e-01 -1.40049481e+00 1.21384047e-01 -7.15303868e-02
8.38484168e-01 -1.42585301e+00 3.61418128e-02 -3.68268639e-01
-2.23395452e-01 1.45376229e+00 7.33838558e-01 -3.56577814e-01
6.04135454e-01 2.91002750e-01 3.95911410e-02 -6.13273561e-01
-3.04647803e-01 4.23649669e-01 -1.70008540e-02 4.26735699e-01
7.37446725e-01 -2.35186532e-01 -5.75752616e-01 5.38244426e-01
-1.51483610e-01 1.92071229e-01 4.63407218e-01 1.57790875e+00
-5.94769835e-01 -9.82746184e-01 -4.77906495e-01 6.71726465e-01
-9.02718604e-01 4.44126427e-01 1.41730011e-01 9.86621857e-01
3.49056274e-01 1.25537646e+00 -1.21741608e-01 -3.60662133e-01
4.30392951e-01 -3.78726602e-01 4.83702511e-01 -4.75170761e-01
-7.24752784e-01 -9.55175534e-02 1.67687804e-01 -7.47257397e-02
-3.05826575e-01 -5.27756155e-01 -1.30310476e+00 -9.54171494e-02
-9.27099228e-01 1.02511100e-01 1.07192171e+00 8.95450592e-01
-1.20700940e-01 7.68200696e-01 8.18230331e-01 -5.62448680e-01
-7.80103564e-01 -1.25166893e+00 -7.87289679e-01 3.82493943e-01
-7.24112466e-02 -9.53220725e-01 -2.34277457e-01 2.42501780e-01] | [7.339228630065918, 1.8817611932754517] |
fc3b7743-fba0-4fba-a00b-135258b4b828 | document-expansion-by-query-prediction | 1904.08375 | null | https://arxiv.org/abs/1904.08375v2 | https://arxiv.org/pdf/1904.08375v2.pdf | Document Expansion by Query Prediction | One technique to improve the retrieval effectiveness of a search engine is to expand documents with terms that are related or representative of the documents' content.From the perspective of a question answering system, this might comprise questions the document can potentially answer. Following this observation, we propose a simple method that predicts which queries will be issued for a given document and then expands it with those predictions with a vanilla sequence-to-sequence model, trained using datasets consisting of pairs of query and relevant documents. By combining our method with a highly-effective re-ranking component, we achieve the state of the art in two retrieval tasks. In a latency-critical regime, retrieval results alone (without re-ranking) approach the effectiveness of more computationally expensive neural re-rankers but are much faster. | ['Rodrigo Nogueira', 'Kyunghyun Cho', 'Wei Yang', 'Jimmy Lin'] | 2019-04-17 | null | null | null | null | ['passage-re-ranking'] | ['natural-language-processing'] | [ 4.11623597e-01 -2.72949070e-01 -1.49982333e-01 -8.91826004e-02
-1.41099012e+00 -7.70045817e-01 9.99133170e-01 3.82941127e-01
-7.33156443e-01 4.61819917e-01 4.17108029e-01 -4.04913813e-01
-6.01721585e-01 -7.87752867e-01 -7.13805616e-01 -2.26827711e-02
8.96103233e-02 9.31414187e-01 6.46399856e-01 -6.80813253e-01
8.56861949e-01 5.22860587e-01 -1.98589897e+00 7.38158047e-01
5.91000199e-01 9.56176102e-01 3.53852868e-01 1.05818772e+00
-2.76617199e-01 8.41709495e-01 -6.83690071e-01 -2.49193847e-01
1.76244035e-01 -1.08843058e-01 -1.54553604e+00 -5.94831944e-01
6.70157135e-01 -6.48539305e-01 -6.32722318e-01 5.79813898e-01
4.74341959e-01 5.93355179e-01 4.47966754e-01 -5.09434879e-01
-7.94903696e-01 5.68555355e-01 -2.65672978e-04 5.80540597e-01
6.99913383e-01 -4.21548277e-01 1.32263386e+00 -7.27839351e-01
8.28983247e-01 1.10860384e+00 4.02621999e-02 4.59231228e-01
-1.03785574e+00 -1.97314411e-01 4.46435288e-02 5.86877823e-01
-1.25492489e+00 -5.08520365e-01 3.96309733e-01 1.20407686e-01
1.48111868e+00 8.76857042e-01 8.75803977e-02 7.00548708e-01
1.38677955e-01 7.37158358e-01 3.96103561e-01 -6.25510514e-01
9.49922726e-02 1.69547480e-02 5.76460123e-01 2.06060722e-01
-1.91129908e-01 8.06941539e-02 -3.62173140e-01 -6.11947954e-01
-1.50462240e-01 5.87157719e-02 -4.12662834e-01 -5.49270101e-02
-7.28179276e-01 8.16379011e-01 5.00103533e-01 7.03144312e-01
-5.01943231e-01 2.60738552e-01 3.68528605e-01 5.51412165e-01
3.79011720e-01 1.14781988e+00 -5.25469720e-01 5.73904850e-02
-1.12974215e+00 6.08866990e-01 1.03946185e+00 6.99612021e-01
7.79254794e-01 -8.54020119e-01 -8.95210743e-01 9.61460352e-01
1.22699462e-01 2.04761788e-01 9.33065414e-01 -9.42287505e-01
1.63197562e-01 5.44614494e-01 2.41154000e-01 -8.64910066e-01
-1.59876332e-01 -4.59381878e-01 -2.51287192e-01 -5.34703374e-01
-1.09581426e-01 4.53852147e-01 -1.01478803e+00 1.36422992e+00
3.80793139e-02 -1.93749323e-01 -2.55524684e-02 7.22643018e-01
5.72221220e-01 9.72966611e-01 -1.26876742e-01 -3.01719040e-01
1.43434811e+00 -1.25508821e+00 -3.88848305e-01 -2.39492580e-01
7.36472487e-01 -9.32315171e-01 9.58324850e-01 2.04231113e-01
-1.20326328e+00 -5.66383839e-01 -7.82199979e-01 -4.58673030e-01
-6.39387906e-01 -9.03725401e-02 2.90246934e-01 1.93891883e-01
-1.62114227e+00 7.46044695e-01 -2.27065101e-01 -4.28165108e-01
-2.30045632e-01 4.48514730e-01 1.58619434e-01 -3.18381429e-01
-1.40256023e+00 1.09277999e+00 4.20223564e-01 -2.41606504e-01
-8.95635784e-01 -6.80732608e-01 -9.83816907e-02 5.36906540e-01
6.20472372e-01 -8.71302903e-01 1.78510201e+00 -3.02240849e-01
-1.16747272e+00 5.88410735e-01 -5.29217064e-01 -4.35796916e-01
5.72007559e-02 -5.93019903e-01 -2.72098452e-01 4.77535874e-01
-2.18445882e-02 5.60957968e-01 7.66956687e-01 -1.01409400e+00
-6.63930953e-01 -3.24913651e-01 3.21373194e-01 2.39168003e-01
-3.81638855e-01 3.18165541e-01 -1.10108852e+00 -2.60685295e-01
-1.99048240e-02 -8.90012920e-01 -2.15990350e-01 -4.16198015e-01
-2.30475172e-01 -8.61806154e-01 6.26191318e-01 -7.87097514e-01
1.64069843e+00 -1.58064115e+00 2.07640883e-02 4.18292463e-01
2.15432808e-01 5.17079592e-01 -7.07956851e-01 9.73019958e-01
-1.86280510e-03 4.64269221e-01 1.97405994e-01 7.31336996e-02
3.72525118e-02 -1.70955837e-01 -9.24744010e-01 -2.55795509e-01
-8.41395184e-02 1.22798061e+00 -8.20641518e-01 -3.07334721e-01
-2.37336412e-01 2.12254122e-01 -6.47578835e-01 4.49770033e-01
-7.32631922e-01 -1.33742943e-01 -8.06269288e-01 3.11838329e-01
2.32489839e-01 -5.65010428e-01 1.12959005e-01 2.57135485e-03
2.57232755e-01 7.18833148e-01 -4.73095506e-01 1.56270981e+00
-6.35437191e-01 6.12466395e-01 -3.88665587e-01 -8.94234598e-01
6.73674881e-01 2.11449727e-01 2.47848779e-01 -1.29334831e+00
-2.66272962e-01 3.91159952e-01 -3.27676982e-01 -7.33092964e-01
1.04522932e+00 3.29152495e-01 2.24940553e-01 8.35182190e-01
-1.73484728e-01 -5.62395081e-02 6.89949036e-01 5.02904594e-01
1.63801825e+00 -1.49723683e-02 2.22623311e-02 3.33454907e-02
6.98771417e-01 5.59601374e-03 -5.99635184e-01 1.54840565e+00
4.61763889e-01 2.49505952e-01 1.32163465e-01 -3.31934631e-01
-1.15751922e+00 -6.70428574e-01 2.72804927e-02 1.95850575e+00
-5.12805693e-02 -5.86488843e-01 -4.78641361e-01 -6.49656832e-01
-1.09769637e-02 7.62064219e-01 -5.01097143e-01 -4.61787879e-01
-7.83088505e-01 -1.55159861e-01 5.30233145e-01 2.70180762e-01
-8.24836716e-02 -1.07597840e+00 -4.01035994e-01 2.18986124e-01
-2.81209499e-01 -7.02819645e-01 -5.35212219e-01 6.91101654e-03
-8.78415048e-01 -9.42083716e-01 -1.02854955e+00 -6.94864392e-01
3.31917882e-01 6.07084751e-01 1.58632290e+00 8.50550115e-01
-1.11844555e-01 5.73691547e-01 -6.51952982e-01 8.05135220e-02
-3.57650608e-01 5.36933184e-01 -4.19250250e-01 -6.12034857e-01
4.66419697e-01 -3.71818781e-01 -8.20684671e-01 1.98276356e-01
-1.42186356e+00 -5.19804358e-01 8.12402248e-01 8.37365448e-01
3.82329613e-01 -2.35934287e-01 5.66940784e-01 -7.46665597e-01
1.29916418e+00 -4.75651473e-01 -6.37370527e-01 1.01159680e+00
-1.14157724e+00 6.28341317e-01 4.62560922e-01 -5.31848013e-01
-7.79283047e-01 -5.21727145e-01 -2.11034611e-01 -2.65261501e-01
5.21426611e-02 8.24497879e-01 7.19054043e-01 1.00995556e-01
9.53189135e-01 5.21797061e-01 -4.78186548e-01 -6.90311074e-01
6.15848303e-01 7.38446653e-01 4.05485213e-01 -5.71822226e-01
6.04090154e-01 1.24232871e-02 -1.44925132e-01 -5.34409404e-01
-9.21012759e-01 -1.08745146e+00 -1.03981607e-01 6.12371229e-02
2.13656202e-01 -4.55867678e-01 -8.74677360e-01 -2.25838527e-01
-1.47522008e+00 2.52417713e-01 -3.12256277e-01 2.49028340e-01
-2.69467115e-01 5.77354431e-01 -7.17599630e-01 -7.73170948e-01
-8.80440235e-01 -9.13796723e-01 1.32418764e+00 8.41450319e-02
-2.39883885e-01 -5.13151884e-01 5.32950401e-01 5.48728943e-01
7.91449904e-01 -8.05349946e-01 1.35305381e+00 -1.40527844e+00
-7.80461609e-01 -5.98359287e-01 -4.31186527e-01 1.42963260e-01
-2.41838768e-01 -3.60593230e-01 -9.29750085e-01 -4.17924225e-01
-1.22778855e-01 -4.36954588e-01 1.23277545e+00 -3.53910327e-02
1.41207492e+00 -7.26465881e-01 -5.60036361e-01 8.98596738e-03
1.28607237e+00 3.96133751e-01 7.68152237e-01 3.66892070e-01
1.05050318e-01 9.10350382e-01 4.26774025e-01 1.45191550e-01
1.05755746e-01 8.22124720e-01 1.84210971e-01 4.21400577e-01
4.40106653e-02 -3.16112459e-01 3.33906412e-02 6.42555892e-01
1.93726614e-01 -7.46358156e-01 -8.63490224e-01 6.41694784e-01
-1.73136532e+00 -9.98621762e-01 2.94754475e-01 2.26498151e+00
9.31568682e-01 -1.43865615e-01 -1.77384481e-01 -2.23940372e-01
4.82838511e-01 1.71099216e-01 -6.31051481e-01 -2.89894015e-01
2.62107670e-01 6.55522585e-01 1.50326192e-01 6.66101038e-01
-6.22652173e-01 9.31207597e-01 7.35791779e+00 1.09772158e+00
-8.14547122e-01 -2.45808184e-01 4.63726968e-01 -1.66725069e-01
-6.17152214e-01 7.28618801e-02 -7.79651940e-01 9.45240036e-02
1.45118570e+00 -5.26865005e-01 8.36057544e-01 7.31767416e-01
-2.85718113e-01 -1.41844511e-01 -1.39186502e+00 5.38126886e-01
3.59731644e-01 -1.53657615e+00 5.02440333e-01 -1.54041946e-01
5.01648605e-01 2.54129082e-01 4.31782901e-02 6.88239694e-01
2.64134347e-01 -9.47740197e-01 1.72792286e-01 8.17005813e-01
3.31298232e-01 -4.82135326e-01 7.54546881e-01 4.99044150e-01
-7.79676795e-01 -2.85250455e-01 -5.82105875e-01 2.73968637e-01
-5.21362899e-03 2.19753116e-01 -1.08838034e+00 4.97091353e-01
7.30751157e-01 4.49644662e-02 -8.33116531e-01 1.19015157e+00
1.35807311e-02 3.43719035e-01 -3.84221673e-01 -5.57006657e-01
3.15968096e-01 2.22667351e-01 3.90641481e-01 1.14862835e+00
4.73161191e-01 1.23120975e-02 -1.93497777e-01 6.02707684e-01
-3.99325341e-01 3.33657652e-01 -4.95575517e-01 -3.56228083e-01
4.67752576e-01 1.05267513e+00 -2.90285647e-01 -7.43941665e-01
-4.48659174e-02 9.66434062e-01 5.10443151e-01 5.66631258e-01
-2.15137392e-01 -4.48365599e-01 1.04462877e-02 8.78402889e-02
2.68437326e-01 3.33376527e-01 5.06163895e-01 -9.19216692e-01
4.74842221e-01 -1.11856890e+00 5.97738326e-01 -1.21952593e+00
-1.17775035e+00 9.12575662e-01 1.64405212e-01 -8.70862365e-01
-1.01142454e+00 -4.59359318e-01 -2.07525179e-01 9.99644816e-01
-1.80831218e+00 -7.82943726e-01 3.02655399e-02 3.41790736e-01
5.01995146e-01 2.95227505e-02 8.97791207e-01 2.51803905e-01
5.69644459e-02 3.37675035e-01 4.60328102e-01 -2.97099978e-01
6.35074496e-01 -8.18984032e-01 3.95929188e-01 4.78407174e-01
3.96158814e-01 1.17670226e+00 4.91135120e-01 -4.88209784e-01
-1.40076458e+00 -6.46452785e-01 1.61286795e+00 -4.86930519e-01
7.28879631e-01 1.14203826e-01 -1.13792849e+00 3.43251646e-01
3.77370030e-01 -5.47927201e-01 5.40185690e-01 3.58825356e-01
-5.73882878e-01 -6.79717511e-02 -5.62579453e-01 8.06356072e-01
7.68965185e-01 -1.13447547e+00 -1.03188193e+00 7.42609143e-01
1.08914721e+00 -9.33494419e-02 -4.78403687e-01 4.51269984e-01
6.25091672e-01 -6.74679041e-01 1.30624318e+00 -1.12508523e+00
4.20132697e-01 1.25814360e-02 -2.28627577e-01 -8.90424311e-01
-3.46180499e-01 -5.74933767e-01 -4.81317073e-01 6.61100924e-01
5.13385653e-01 -3.63854319e-01 6.95514917e-01 8.06077302e-01
4.39623334e-02 -9.37767684e-01 -8.73280704e-01 -6.09279394e-01
1.33824110e-01 -1.98031023e-01 6.24236882e-01 1.99842036e-01
1.35647997e-01 6.71511769e-01 -1.16500899e-01 -7.49432445e-02
-1.30849928e-01 3.89694661e-01 3.48878413e-01 -1.29421926e+00
-3.72356683e-01 -6.48236513e-01 2.10224241e-01 -1.76167762e+00
1.77586332e-01 -1.00910342e+00 2.33105317e-01 -1.49702168e+00
4.55011696e-01 -5.54009937e-02 -5.10731101e-01 2.56200284e-01
-4.20712531e-01 -6.40695449e-03 5.65848574e-02 7.41673052e-01
-1.12672150e+00 3.74430776e-01 8.53931069e-01 -4.06025887e-01
-1.62083417e-01 6.26398176e-02 -7.66412854e-01 2.26609722e-01
3.66505027e-01 -6.35585964e-01 -3.52854162e-01 -6.43998325e-01
7.40587413e-01 3.71469051e-01 2.11782768e-01 -5.99566221e-01
8.19722831e-01 3.91682573e-02 7.38624036e-02 -9.36059415e-01
4.34195608e-01 -6.92560136e-01 -2.00475723e-01 5.67982852e-01
-1.30275345e+00 4.45840091e-01 9.87340435e-02 5.39107919e-01
-3.33349824e-01 -8.27247500e-01 1.10612027e-01 -9.82186496e-02
-6.12910092e-01 2.34616190e-01 -3.36537212e-01 9.78561025e-03
4.17202175e-01 1.55159652e-01 -6.72938108e-01 -7.41110384e-01
-3.13836843e-01 4.19569463e-01 3.42657752e-02 6.33343816e-01
7.40740359e-01 -1.03893447e+00 -7.30223596e-01 -2.40962446e-01
4.10134822e-01 -4.95448351e-01 1.61357686e-01 2.32760906e-01
-3.12685966e-01 1.37753630e+00 5.12092769e-01 -1.66807652e-01
-1.05104780e+00 8.99288118e-01 1.27342135e-01 -1.00959957e+00
-5.37595674e-02 8.98870468e-01 -9.51933395e-03 -3.90994847e-01
2.86381066e-01 2.47749519e-02 -6.13624275e-01 9.20730531e-02
8.62726927e-01 3.38391453e-01 4.65077460e-01 -1.38093725e-01
-2.00335681e-01 4.02033836e-01 -6.70855463e-01 -4.15809542e-01
1.02494693e+00 -9.69248936e-02 -4.34869111e-01 -7.68490136e-02
1.51326919e+00 -5.92044406e-02 -2.19903156e-01 -6.89137340e-01
4.57991987e-01 -3.50945443e-01 1.65720403e-01 -1.07391989e+00
-5.74408412e-01 6.55549347e-01 5.25970340e-01 5.98895609e-01
1.30877209e+00 3.10439378e-01 8.98511350e-01 1.41408968e+00
7.50276074e-02 -1.04125798e+00 3.45907360e-01 8.26434731e-01
1.04321945e+00 -9.38090444e-01 -4.20731073e-03 1.94095924e-01
-2.21829135e-02 1.11963475e+00 2.76564121e-01 5.11053056e-02
4.48009819e-01 -4.84008849e-01 -1.35559827e-01 -3.26588780e-01
-1.33605027e+00 -3.70440960e-01 1.03963387e+00 1.39795110e-01
4.10792381e-01 -6.18211210e-01 -6.52066231e-01 6.99237064e-02
4.37431596e-03 1.83617562e-01 -4.99737486e-02 9.16729391e-01
-6.36352420e-01 -1.28321707e+00 -5.92131205e-02 8.34681571e-01
-5.71867526e-01 -7.06657887e-01 -5.70312500e-01 2.30561078e-01
-6.24293268e-01 1.28722918e+00 5.13109379e-02 -5.29646933e-01
2.53487349e-01 3.46579343e-01 5.88885881e-02 -5.91318309e-01
-8.69386613e-01 6.39062226e-02 1.45007074e-01 -8.11653256e-01
-1.82916135e-01 -1.31940484e-01 -6.86615944e-01 -1.62135452e-01
-5.90217948e-01 7.98980713e-01 6.06675208e-01 1.06035388e+00
8.42692137e-01 3.52258772e-01 7.12842166e-01 -4.35294658e-01
-1.17206752e+00 -1.06592727e+00 -3.45451087e-02 3.32497954e-01
3.85097563e-01 5.25887869e-02 -2.58788466e-01 -2.92358279e-01] | [11.510384559631348, 7.558253288269043] |
7277b6a0-19d4-46c5-8aab-56ad9d3a2a46 | q-deckrec-a-fast-deck-recommendation-system | 1806.09771 | null | http://arxiv.org/abs/1806.09771v1 | http://arxiv.org/pdf/1806.09771v1.pdf | Q-DeckRec: A Fast Deck Recommendation System for Collectible Card Games | Deck building is a crucial component in playing Collectible Card Games
(CCGs). The goal of deck building is to choose a fixed-sized subset of cards
from a large card pool, so that they work well together in-game against
specific opponents. Existing methods either lack flexibility to adapt to
different opponents or require large computational resources, still making them
unsuitable for any real-time or large-scale application. We propose a new deck
recommendation system, named Q-DeckRec, which learns a deck search policy
during a training phase and uses it to solve deck building problem instances.
Our experimental results demonstrate Q-DeckRec requires less computational
resources to build winning-effective decks after a training phase compared to
several baseline methods. | ['Magy Seif El-Nasr', 'Truong-Huy Nguyen', 'Seth Cooper', 'Chris Amato', 'Zhengxing Chen', 'Yizhou Sun'] | 2018-06-26 | null | null | null | null | ['card-games'] | ['playing-games'] | [-2.33574629e-01 -5.55123866e-01 -3.47435743e-01 -9.50201675e-02
-8.23397934e-01 -1.09768653e+00 2.49088287e-01 -3.01089913e-01
-5.32364368e-01 7.91687012e-01 -1.40690356e-01 -5.97722471e-01
-5.85018516e-01 -1.11263680e+00 -4.97516721e-01 -4.41708893e-01
-8.87634382e-02 1.17708349e+00 6.20790243e-01 -6.64856911e-01
6.19279146e-01 1.19486757e-01 -1.36550713e+00 8.82451355e-01
6.88322008e-01 9.15594161e-01 9.83755440e-02 8.29967797e-01
2.15093851e-01 1.10506403e+00 -8.87771189e-01 -9.71415281e-01
9.11534250e-01 -5.42947710e-01 -8.59910011e-01 -4.97932822e-01
1.02415949e-01 -7.43118227e-01 -2.06869572e-01 9.60407078e-01
3.91211629e-01 2.48498499e-01 3.79972935e-01 -1.13033390e+00
1.30372643e-01 1.02965081e+00 -3.12431663e-01 9.26602259e-02
1.46957248e-01 4.78702933e-02 1.23105741e+00 -2.63057649e-01
4.34986889e-01 7.82916725e-01 5.62008142e-01 6.08586431e-01
-1.05086708e+00 -9.22827482e-01 -6.02935776e-02 3.54434729e-01
-1.05265152e+00 -1.97524562e-01 5.38798690e-01 -8.21440741e-02
8.52640390e-01 7.75917172e-01 9.69595671e-01 6.15415335e-01
8.52130204e-02 8.23413670e-01 1.31906283e+00 -2.02197567e-01
5.64951658e-01 -1.23840578e-01 8.33903924e-02 8.50315914e-02
5.77219784e-01 4.11063164e-01 -5.43833554e-01 -5.14247417e-01
9.21569824e-01 -1.37998551e-01 3.44156712e-01 -2.54148602e-01
-5.53809702e-01 1.15657818e+00 1.06522232e-01 -1.11439653e-01
-2.77167797e-01 2.70808280e-01 5.18967152e-01 6.66758001e-01
1.12486631e-01 1.02448249e+00 -2.73615718e-01 -8.43472183e-01
-9.91423786e-01 9.82057691e-01 9.32906687e-01 6.57551467e-01
6.40246689e-01 -3.66106391e-01 6.16222955e-02 7.65948951e-01
-2.21593320e-01 1.35934666e-01 3.46287072e-01 -1.19111872e+00
8.38540673e-01 5.31025410e-01 2.11018115e-01 -9.42272961e-01
-5.27562350e-02 -4.27470565e-01 -5.52633703e-01 3.77044767e-01
7.26682603e-01 -1.85530722e-01 -3.63710850e-01 1.31157982e+00
1.85359269e-01 2.99123824e-01 -1.40709251e-01 1.12983239e+00
3.00197393e-01 6.51767194e-01 -4.15907234e-01 1.83862165e-01
1.21512461e+00 -9.41468835e-01 2.89887100e-01 -4.27770615e-01
4.96715754e-01 -6.20653987e-01 7.23585784e-01 1.13302422e+00
-1.19757020e+00 -1.57386512e-01 -1.01816666e+00 2.87475079e-01
-8.95922258e-02 -3.82904219e-03 1.25705457e+00 1.21939182e+00
-7.08407104e-01 6.29209697e-01 -4.11250770e-01 5.06149113e-01
1.63458034e-01 6.41952515e-01 -9.14992839e-02 -3.48335594e-01
-1.23090065e+00 7.21794069e-01 7.04238713e-01 -1.06111035e-01
-9.88075852e-01 -4.66010958e-01 -2.67765820e-01 1.61552027e-01
9.22709227e-01 -2.97551274e-01 1.49445176e+00 -1.01507783e+00
-1.95097864e+00 6.95203662e-01 8.77040982e-01 -4.98832226e-01
5.21734357e-01 -2.98586816e-01 -1.07689396e-01 -2.21888497e-01
-1.14488244e-01 3.61720435e-02 7.43327975e-01 -9.71081436e-01
-9.98649478e-01 -8.53443965e-02 9.41494107e-01 5.58582723e-01
1.67265274e-02 3.80116016e-01 -7.87909746e-01 -6.47778749e-01
-1.78920344e-01 -1.06668770e+00 -6.08422101e-01 -9.32988286e-01
-2.28099287e-01 -2.11151272e-01 6.66073933e-02 -2.80573636e-01
1.62264824e+00 -1.64404166e+00 5.21711707e-02 8.61526906e-01
2.41716385e-01 5.84265828e-01 -2.57147580e-01 5.61194062e-01
2.47553900e-01 -3.52978289e-01 5.55565059e-01 7.39116669e-02
1.33777916e-01 1.69601351e-01 -4.84771878e-01 2.14886621e-01
-5.37322104e-01 6.78566277e-01 -9.10857558e-01 2.74253590e-03
2.80809738e-02 -5.32852888e-01 -1.41376472e+00 2.54724413e-01
-2.85094470e-01 -3.53260785e-01 -5.29066443e-01 4.13725793e-01
8.68211865e-01 -7.46390000e-02 7.85959601e-01 5.53005397e-01
1.89622790e-02 7.59301364e-01 -1.76259208e+00 1.50679410e+00
-2.06813082e-01 8.61625150e-02 1.13204174e-01 -1.20575416e+00
6.61214054e-01 -3.08161706e-01 1.77102655e-01 -7.86510766e-01
3.61950785e-01 5.04984617e-01 2.36602932e-01 4.20571834e-01
1.20905817e+00 -1.26967296e-01 -7.76160240e-01 8.26687455e-01
-2.83336043e-01 -2.48054922e-01 4.88580078e-01 5.83235145e-01
1.21963775e+00 -3.36516112e-01 2.53542334e-01 -2.30918542e-01
1.30356550e-01 3.67177516e-01 8.36596668e-01 1.28680968e+00
1.91505879e-01 3.94374520e-01 7.23763406e-01 -7.29427636e-01
-8.63202631e-01 -6.69699371e-01 4.17997211e-01 1.53641963e+00
5.83894670e-01 -1.01480067e+00 -7.00019717e-01 -7.10172415e-01
1.43630475e-01 2.70206660e-01 -4.32644278e-01 -1.17719822e-01
-7.55195558e-01 -4.63696808e-01 5.80030024e-01 4.05011237e-01
4.47683394e-01 -9.79160964e-01 -6.38893425e-01 6.94859743e-01
-8.39525089e-02 -6.51818752e-01 -7.77387738e-01 2.10782021e-01
-5.48798680e-01 -1.44090986e+00 -3.40082377e-01 -5.58894396e-01
3.63585204e-01 7.61398435e-01 1.36288106e+00 2.95275271e-01
-7.09422454e-02 -1.67467549e-01 -6.37486577e-01 -3.71454984e-01
-1.88670769e-01 1.33610860e-01 1.61080435e-01 -2.63773203e-01
3.94262552e-01 -4.47856158e-01 -7.88504660e-01 7.44145095e-01
-7.02011943e-01 2.70865560e-01 4.58299249e-01 9.51979816e-01
2.63801754e-01 6.60939753e-01 9.97512117e-02 -1.26955879e+00
9.68148053e-01 -4.04345125e-01 -1.11925411e+00 1.34612069e-01
-3.98521274e-01 -3.02984416e-01 1.05527091e+00 -6.90136671e-01
-6.39291704e-01 -3.80510718e-01 9.66688171e-02 -1.31473280e-02
4.63346541e-01 4.30288374e-01 -2.58986235e-01 -1.10257342e-01
6.70802772e-01 2.27467090e-01 -3.03439468e-01 -4.74258810e-01
3.24027330e-01 6.30882680e-01 5.42776227e-01 -1.22392905e+00
8.20265710e-01 1.93804316e-02 -4.80542511e-01 -4.57207067e-03
-5.47990918e-01 -6.19083881e-01 1.28288910e-01 -2.71007538e-01
1.89311802e-02 -6.90781713e-01 -1.28183925e+00 9.56345797e-01
-6.51735127e-01 -7.61013150e-01 -1.58254281e-01 3.76328588e-01
-4.86990601e-01 1.59911603e-01 -6.77975118e-01 -6.53829932e-01
-3.10553312e-01 -8.54760170e-01 5.75068831e-01 5.48061788e-01
1.35356873e-01 -4.95062321e-01 5.34451723e-01 8.28983188e-01
2.25489736e-01 -3.75697941e-01 3.61365497e-01 -4.70070481e-01
-6.95778668e-01 -5.77326357e-01 2.06600521e-02 1.94085374e-01
-4.20252234e-01 -4.82977122e-01 -3.51105809e-01 -6.00220561e-01
-6.51868701e-01 -8.50396752e-01 8.34354818e-01 2.85617739e-01
1.40023291e+00 -4.47701722e-01 1.40143558e-02 7.94611514e-01
1.32419288e+00 3.05037469e-01 1.08808172e+00 8.40529978e-01
1.59150913e-01 2.04383209e-01 8.88277173e-01 8.19180310e-01
3.00126106e-01 1.08593440e+00 2.65924841e-01 3.23745698e-01
4.78075355e-01 -6.49749577e-01 4.02882993e-01 5.65616727e-01
-6.63767815e-01 -3.38575900e-01 -4.54764009e-01 3.29289645e-01
-2.01391101e+00 -1.21895790e+00 3.03885520e-01 2.31796503e+00
8.58648539e-01 6.19979203e-01 7.42060721e-01 2.35217944e-01
4.47934091e-01 -1.22813202e-01 -4.58065122e-01 -7.02021718e-01
-6.43577948e-02 8.09662461e-01 9.71516132e-01 2.52904207e-01
-9.57667708e-01 1.38493383e+00 6.42127466e+00 1.78355777e+00
-7.16434300e-01 -2.85910785e-01 5.98857224e-01 -4.78190482e-01
-2.40475804e-01 4.56519336e-01 -7.61508524e-01 7.34828055e-01
6.26704991e-01 -2.43608430e-01 1.09674048e+00 1.13916838e+00
-1.77406237e-01 -3.26782763e-01 -5.50915241e-01 1.18408382e+00
-1.23022005e-01 -1.74690282e+00 -1.33989185e-01 3.24163586e-01
8.57414842e-01 -3.77801567e-01 9.17305276e-02 7.28301108e-01
1.48530853e+00 -1.15439939e+00 7.97893405e-01 -1.80045858e-01
8.74956787e-01 -9.41947937e-01 6.26800537e-01 5.92342198e-01
-1.14001167e+00 -3.26235443e-01 -8.06761146e-01 -5.06907940e-01
-3.26687172e-02 1.39762327e-01 -3.54013771e-01 4.29087311e-01
5.98584831e-01 1.19908482e-01 -1.86360255e-01 1.20359409e+00
-2.34027594e-01 9.56494153e-01 -3.72357339e-01 -3.27958971e-01
6.49576724e-01 -4.08743888e-01 3.52665693e-01 6.39940739e-01
2.10333511e-01 7.62873590e-01 2.74194568e-01 3.39804977e-01
-9.63094383e-02 1.85913280e-01 6.62356988e-02 5.86413741e-02
7.05218017e-01 1.15730774e+00 -7.91428983e-01 -3.11715931e-01
-3.51401307e-02 9.14918780e-01 5.00579298e-01 -1.83937728e-01
-1.00337899e+00 -4.38599646e-01 1.18615198e+00 1.77494556e-01
4.83952641e-01 -2.36101113e-02 -1.43878832e-01 -1.27319002e+00
-1.08577631e-01 -1.60720658e+00 8.70970011e-01 -2.68292218e-01
-1.17978668e+00 3.68369222e-01 -1.19619213e-01 -1.39134526e+00
-2.63110936e-01 -7.72709489e-01 -9.73913908e-01 5.18466830e-01
-1.01613498e+00 -9.27479208e-01 7.79497474e-02 6.89405560e-01
3.79492223e-01 -7.13927448e-01 6.58309281e-01 5.63486889e-02
-3.91702324e-01 1.11837626e+00 6.59360290e-01 6.03307895e-02
5.20708740e-01 -1.25111473e+00 4.17610556e-01 6.54508412e-01
3.26223552e-01 7.14645743e-01 3.70748997e-01 -4.42917496e-01
-1.53307509e+00 -5.29256105e-01 2.94574827e-01 -4.21660215e-01
3.90467763e-01 -5.01571238e-01 -3.14981461e-01 8.97959471e-02
-2.71115839e-01 -3.46353143e-01 8.84426773e-01 7.12053835e-01
-5.78564644e-01 -3.56533438e-01 -8.65591764e-01 4.22810525e-01
9.31529403e-01 -3.00698757e-01 -4.35875624e-01 -1.33300886e-01
-1.63841657e-02 -8.30536962e-01 -4.31958228e-01 -2.15819478e-01
9.47800100e-01 -1.17542148e+00 1.09823501e+00 -1.10075927e+00
4.30737287e-01 -1.69132426e-01 1.17545128e-01 -1.47542655e+00
-7.64593780e-01 -1.11717844e+00 4.98586856e-02 7.88355529e-01
2.32905850e-01 -4.76563126e-01 1.49235010e+00 8.34307969e-01
1.98492289e-01 -4.58488256e-01 -9.03251350e-01 -9.57959294e-01
4.01230961e-01 -7.45958090e-01 1.05909431e+00 8.37119639e-01
5.37056446e-01 -2.79040486e-02 -1.15735185e+00 -2.06986010e-01
6.16059303e-01 6.88890755e-01 1.56474304e+00 -9.33252215e-01
-1.29294372e+00 -5.44402599e-01 -3.93708885e-01 -1.49271297e+00
-3.30839247e-01 -7.92451382e-01 -1.46374390e-01 -9.28951740e-01
5.01621962e-01 -1.15722084e+00 -5.27982652e-01 3.59630346e-01
-2.71056026e-01 6.44821346e-01 5.70387781e-01 3.30210924e-01
-1.20476127e+00 -1.35668263e-01 1.08509278e+00 -1.05181031e-01
-4.94386524e-01 6.21041834e-01 -1.30812907e+00 4.43087369e-01
9.00522649e-01 -4.68111366e-01 -3.79796058e-01 -3.47958267e-01
6.86822474e-01 1.62077114e-01 -9.85919163e-02 -8.32363725e-01
3.35973293e-01 -8.76788020e-01 2.20047478e-02 -4.14317787e-01
1.78683102e-01 -4.18003470e-01 4.03717428e-01 4.72839803e-01
-8.83086026e-02 -4.90319319e-02 2.16361154e-02 2.86515176e-01
6.97394088e-02 -2.40201846e-01 5.60801864e-01 -1.08112216e-01
-8.48497331e-01 2.40504652e-01 -5.29738426e-01 3.09964508e-01
1.06991446e+00 -5.37347257e-01 -3.33879679e-01 -5.31902254e-01
8.11405014e-03 3.15873116e-01 5.17025054e-01 2.63926446e-01
4.18623745e-01 -1.25462401e+00 -8.56396914e-01 9.64209214e-02
9.01128724e-02 -2.36578777e-01 5.14670551e-01 7.44392723e-02
-1.08747709e+00 2.75197178e-01 -5.51643610e-01 3.26235712e-01
-1.48303068e+00 2.65080094e-01 2.50781089e-01 -1.21009696e+00
-5.49093485e-01 1.23497987e+00 -6.11668229e-02 -3.71488273e-01
4.51717190e-02 -2.40593292e-02 -3.56231295e-02 -4.01379555e-01
9.08993065e-01 3.71246845e-01 -2.43855622e-02 -1.68589830e-01
-3.80424559e-01 6.11106791e-02 -4.38539565e-01 -1.39326662e-01
1.45755649e+00 5.58568478e-01 5.69137298e-02 -5.74330986e-01
4.24559861e-01 3.03963453e-01 -1.12233818e+00 -3.32600325e-01
-3.76588613e-01 -1.09037876e+00 -2.41337344e-02 -7.66151071e-01
-1.18709326e+00 1.78122938e-01 -1.39353797e-01 9.58016962e-02
1.16056144e+00 -2.62231529e-01 1.05370247e+00 4.21938866e-01
1.14261103e+00 -1.47018611e+00 4.42439355e-02 7.58419693e-01
3.96495670e-01 -6.65045440e-01 2.41813153e-01 -1.05984002e-01
-9.91121411e-01 9.70363915e-01 8.03969562e-01 -5.77636540e-01
3.86693329e-01 3.58409971e-01 -1.03900991e-01 -1.50217593e-01
-1.03800023e+00 -1.12715699e-01 1.44691497e-01 5.69170654e-01
-2.98146993e-01 6.25789523e-01 -5.61388373e-01 1.37665260e+00
-7.83653259e-01 -4.62443642e-02 4.85451192e-01 9.01143610e-01
-4.72842693e-01 -1.78897941e+00 -5.40526569e-01 8.15865219e-01
-4.45733547e-01 -1.37223005e-01 -7.27353811e-01 3.67423683e-01
3.38612556e-01 1.00810683e+00 -6.89365566e-02 -9.53564286e-01
2.16863558e-01 -7.09786355e-01 4.41338956e-01 -6.27717435e-01
-1.14059699e+00 -1.17038168e-01 3.87887329e-01 -7.80592978e-01
2.22629592e-01 -4.15409327e-01 -5.02441883e-01 -1.33639443e+00
-7.75437772e-01 8.68968666e-01 8.21629092e-02 6.08738422e-01
1.25121534e-01 -3.56431492e-02 7.14503944e-01 -6.08495772e-01
-7.56549954e-01 -4.76678997e-01 -1.27515113e+00 4.64561850e-01
-6.36804044e-01 -7.11169302e-01 1.27292559e-01 -5.59187710e-01] | [3.493525505065918, 1.4773659706115723] |
17a0217f-db28-4394-87bb-97a5f6b4abcd | deep-learning-based-counting-methods-datasets | 2303.02632 | null | https://arxiv.org/abs/2303.02632v2 | https://arxiv.org/pdf/2303.02632v2.pdf | Deep-Learning-based Counting Methods, Datasets, and Applications in Agriculture -- A Review | The number of objects is considered an important factor in a variety of tasks in the agricultural domain. Automated counting can improve farmers decisions regarding yield estimation, stress detection, disease prevention, and more. In recent years, deep learning has been increasingly applied to many agriculture-related applications, complementing conventional computer-vision algorithms for counting agricultural objects. This article reviews progress in the past decade and the state of the art for counting methods in agriculture, focusing on deep-learning methods. It presents an overview of counting algorithms, metrics, platforms, and sensors, a list of all publicly available datasets, and an in-depth discussion of various deep-learning methods used for counting. Finally, it discusses open challenges in object counting using deep learning and gives a glimpse into new directions and future perspectives for counting research. The review reveals a major leap forward in object counting in agriculture in the past decade, led by the penetration of deep learning methods into counting platforms. | ['Yael Edan', 'Liu Huijun', 'Guy Farjon'] | 2023-03-05 | null | null | null | null | ['object-counting'] | ['computer-vision'] | [ 7.51801804e-02 -7.79390335e-01 -2.85576791e-01 -2.51937717e-01
1.36259273e-01 -7.98723459e-01 2.33553946e-01 8.79721999e-01
-6.91472769e-01 3.44836801e-01 -2.08466113e-01 -5.24026334e-01
2.20277771e-01 -1.36799026e+00 -4.81595039e-01 -7.50896871e-01
-1.76864028e-01 3.30936581e-01 -8.52774009e-02 -5.67232929e-02
2.77250797e-01 6.80933416e-01 -1.57679677e+00 2.48378500e-01
3.70106429e-01 9.14930999e-01 3.97782803e-01 1.06429839e+00
-4.83520657e-01 8.52570057e-01 -8.10626030e-01 -6.52111292e-01
-2.21501127e-01 -7.61999115e-02 -4.59341615e-01 -3.06709319e-01
5.18843532e-01 -9.54938948e-01 -3.63163440e-03 1.09200048e+00
6.26958966e-01 -4.36614305e-01 6.88645184e-01 -9.36655164e-01
-8.70887220e-01 5.57334960e-01 -1.09768951e+00 6.16633296e-01
-1.83350801e-01 -1.23589449e-01 4.95500714e-01 -6.79061234e-01
-1.34899616e-01 1.12186563e+00 1.15217853e+00 7.69276693e-02
-8.79827321e-01 -9.25959826e-01 -3.56078930e-02 -3.80145088e-02
-8.79362404e-01 -3.10818069e-02 1.38238758e-01 -5.18115997e-01
1.01477361e+00 1.25747547e-02 1.15435493e+00 6.27552569e-01
3.03319037e-01 1.16493213e+00 7.13851750e-01 -7.27441609e-01
2.95860052e-01 -6.52549922e-01 4.05770421e-01 8.34773481e-01
1.17267728e+00 1.04701854e-01 -2.23492131e-01 -1.62921652e-01
1.16915858e+00 1.67726174e-01 8.01334679e-01 -2.96920627e-01
-1.20317185e+00 1.49963605e+00 4.91292596e-01 3.47076297e-01
-3.34261805e-01 5.91466427e-01 7.74003923e-01 -1.82005808e-01
8.21621895e-01 4.75919187e-01 -9.16692555e-01 2.55020529e-01
-1.01639903e+00 5.01792073e-01 8.82699788e-01 9.70036983e-01
3.40561241e-01 4.84847069e-01 -5.27689576e-01 6.70647323e-01
1.24508608e-02 1.17147672e+00 -2.95304060e-01 -1.12994933e+00
1.25841469e-01 4.46872026e-01 3.34107697e-01 -1.15350091e+00
-6.40001416e-01 -1.04962237e-01 -9.76756096e-01 3.13585222e-01
5.66802204e-01 -2.32394770e-01 -6.81817472e-01 1.09010005e+00
1.90222859e-01 -1.40565410e-01 -4.71721202e-01 4.82970476e-01
1.19370997e+00 5.19565105e-01 6.21973634e-01 -1.21819712e-02
1.93668723e+00 -5.94639182e-01 -9.02752757e-01 -3.60864550e-01
5.90143800e-01 -8.01711321e-01 4.04522747e-01 1.53544873e-01
-7.79409170e-01 -3.21197212e-01 -9.52253222e-01 -2.94536889e-01
-1.09753096e+00 8.42828751e-01 1.51382649e+00 1.04037726e+00
-3.27276707e-01 5.73689580e-01 -1.32387388e+00 -7.86341667e-01
1.06189108e+00 2.29044542e-01 3.02349981e-02 8.69992301e-02
-6.01466656e-01 1.03188014e+00 3.00570667e-01 4.49699134e-01
-8.46948981e-01 -5.61594725e-01 -9.81219351e-01 7.33257756e-02
-2.21006363e-03 -4.14388299e-01 1.54967988e+00 -2.23397449e-01
-1.22536135e+00 1.38995039e+00 -1.76206082e-01 -4.44992870e-01
4.16102111e-01 -4.52056408e-01 5.93466721e-02 -2.84809053e-01
3.54458719e-01 7.15512931e-01 2.64409274e-01 -7.82173514e-01
-9.78112578e-01 -8.51999521e-01 8.37856531e-02 -3.10960114e-01
-3.97739798e-01 5.03418744e-01 5.11866868e-01 -5.24549365e-01
-1.40554696e-01 -3.87466878e-01 -1.20970279e-01 4.57213968e-01
8.29844177e-02 -2.03765124e-01 6.92174196e-01 -5.68291187e-01
9.15247083e-01 -1.47530997e+00 -4.70099926e-01 -6.09881222e-01
4.60741520e-01 8.30567598e-01 -3.33012581e-01 3.23125541e-01
4.49126989e-01 -2.05868050e-01 -2.06114978e-01 -3.47585194e-02
-3.81288052e-01 1.42065823e-01 -1.24092698e-01 7.02025712e-01
5.55785120e-01 1.32571006e+00 -1.60664797e+00 -4.20057714e-01
9.13287222e-01 4.15122151e-01 2.07547992e-02 4.54978980e-02
-1.50165439e-01 2.93111384e-01 -4.35496032e-01 1.20233119e+00
1.32563090e+00 1.73758008e-02 4.14878875e-02 -1.57461628e-01
-7.42512107e-01 -3.92189294e-01 -6.67923510e-01 1.24639499e+00
-4.74039197e-01 6.21008098e-01 -2.69301794e-02 -1.17428219e+00
9.84095752e-01 -1.63268179e-01 3.25107783e-01 -4.20570016e-01
5.47890961e-01 3.85676324e-01 -1.88418359e-01 -3.60667109e-01
4.68478471e-01 3.88610363e-01 -5.59403226e-02 1.34805471e-01
4.07877207e-01 -4.65216249e-01 5.45586407e-01 -5.12105286e-01
8.81683409e-01 1.00198627e-01 8.96877646e-01 -3.14841866e-01
1.65792301e-01 3.45136970e-01 9.88249108e-02 9.12622154e-01
-5.87099075e-01 1.07260801e-01 4.98847514e-01 -1.35475433e+00
-9.45587575e-01 -7.99354494e-01 -3.28739345e-01 1.67370951e+00
-3.29794347e-01 2.24090293e-01 -7.23667264e-01 -2.48370633e-01
5.02366841e-01 3.61191124e-01 -1.08960509e+00 2.77031749e-01
-7.73324609e-01 -1.39539206e+00 9.25600648e-01 1.19895041e+00
9.21035707e-01 -1.31439614e+00 -1.27722991e+00 3.99848104e-01
-3.50995868e-01 -1.29182792e+00 3.04086030e-01 7.39969254e-01
-1.30538225e+00 -1.22528279e+00 -1.04565644e+00 -9.41355467e-01
1.84615925e-01 8.79073083e-01 1.59389102e+00 6.80211633e-02
-8.30977380e-01 -2.10078478e-01 -2.51067042e-01 -1.63499749e+00
-1.84629917e-01 3.88929904e-01 -3.01184237e-01 -1.16716254e+00
1.54707134e+00 1.03350841e-01 -6.06008232e-01 -1.58923298e-01
-5.30101538e-01 -5.71958303e-01 3.64329606e-01 6.73042953e-01
3.04588109e-01 -2.99394280e-01 4.70312655e-01 -6.10668063e-01
2.65370518e-01 -3.44925940e-01 -1.39300835e+00 3.53698373e-01
6.85148034e-03 -2.99155354e-01 2.61195332e-01 -3.02935839e-01
-5.04765868e-01 1.10740572e-01 1.58616558e-01 3.91826481e-01
4.48907167e-02 2.75352836e-01 2.55023241e-01 -2.76493728e-01
6.14357054e-01 -2.23876357e-01 -6.82577342e-02 -1.58908993e-01
2.71590680e-01 3.00966889e-01 2.93531239e-01 -1.77424699e-01
1.84700876e-01 6.75243974e-01 2.72678167e-01 -1.20518076e+00
-1.27487528e+00 -6.14653528e-01 -1.05167031e+00 -1.40646875e-01
1.02981496e+00 -9.08884883e-01 -1.32467365e+00 1.14212000e+00
-1.64176881e+00 -4.08384860e-01 -7.86924735e-02 4.63488638e-01
-2.91919917e-01 2.98855845e-02 -9.53424633e-01 -1.07137728e+00
-7.35350907e-01 -7.81382501e-01 1.52688873e+00 4.26459312e-01
-3.38457972e-02 -9.68880594e-01 2.72495598e-01 8.56182724e-02
4.89480317e-01 6.23365879e-01 8.79582226e-01 -1.32669240e-01
-2.04392523e-02 -8.05437386e-01 -9.26872551e-01 2.11204410e-01
2.12916434e-01 3.57734174e-01 -1.34762049e+00 1.48484670e-02
-4.95755494e-01 -4.93208230e-01 1.18718302e+00 1.35356736e+00
1.45753455e+00 3.13952774e-01 -4.70857739e-01 5.60035765e-01
1.41673291e+00 -3.77041139e-02 3.12402666e-01 2.12749541e-01
6.22419000e-01 4.19823676e-01 6.11446321e-01 5.85843325e-01
3.53269756e-01 7.40364343e-02 8.01645517e-01 -4.26928729e-01
8.63933489e-02 -1.86586589e-01 -3.24651092e-01 5.45561671e-01
-7.16112852e-01 -3.38764876e-01 -6.76666081e-01 5.04644275e-01
-1.58197343e+00 -1.13382208e+00 -5.10318518e-01 2.09032798e+00
3.86447936e-01 -5.78322470e-01 2.88726002e-01 3.04223239e-01
8.30354691e-01 2.01349959e-01 -5.52587628e-01 -5.64594805e-01
-2.07672834e-01 6.99890137e-01 1.33807039e+00 8.86131227e-02
-1.82929099e+00 1.26430821e+00 7.82301092e+00 4.85362232e-01
-7.32226491e-01 8.09308290e-02 5.54512918e-01 4.60188955e-01
6.55592859e-01 -5.61552703e-01 -9.89241183e-01 6.13223277e-02
4.27309930e-01 3.43132943e-01 3.01202804e-01 8.96967828e-01
9.04855803e-02 -6.88096464e-01 -7.59216309e-01 5.42564332e-01
-9.00054127e-02 -1.32685184e+00 -5.96350618e-02 -1.37387976e-01
6.21913314e-01 2.27862120e-01 -7.97813162e-02 3.63012284e-01
6.93003654e-01 -1.01152360e+00 5.17065108e-01 -1.34957954e-01
9.84545290e-01 -7.23191559e-01 1.42839432e+00 1.10036969e-01
-1.61042726e+00 -2.88063675e-01 -1.28560829e+00 -6.26488388e-01
-1.86926901e-01 8.82390320e-01 -2.18460828e-01 6.81683645e-02
9.78381872e-01 5.24342597e-01 -3.76870036e-01 1.00982726e+00
-8.03713053e-02 4.67388809e-01 -3.34832668e-01 -5.56024492e-01
3.81448776e-01 -2.94577092e-01 -2.57934183e-01 1.74215269e+00
2.90301710e-01 1.94126502e-01 1.39538676e-01 8.23296249e-01
-7.93613046e-02 -1.48626357e-01 -8.34107816e-01 -4.06002223e-01
5.87313652e-01 1.38560402e+00 -1.38421977e+00 -1.36432812e-01
-5.06498277e-01 6.81404054e-01 8.68008658e-02 -7.80273601e-02
-9.11744893e-01 -5.48017561e-01 4.79102850e-01 -2.11438358e-01
2.39484385e-01 -4.58527058e-01 -5.23817480e-01 -7.15718627e-01
-3.77673358e-01 -5.13402641e-01 2.73478508e-01 -4.91537511e-01
-1.15202391e+00 -2.73976684e-01 1.94332495e-01 -7.71151543e-01
4.38349307e-01 -1.44528568e+00 -4.75937277e-01 5.95103621e-01
-1.73423147e+00 -1.37471700e+00 -9.15055513e-01 -3.73882711e-01
6.84139907e-01 -1.24516673e-01 1.53324389e+00 2.28334889e-01
-2.92965442e-01 2.76535809e-01 3.80270213e-01 6.92457974e-01
2.10437045e-01 -1.06449389e+00 9.96035397e-01 3.30075383e-01
-1.63060240e-02 1.45436734e-01 3.19747001e-01 -5.31575918e-01
-1.23956573e+00 -1.20132530e+00 9.44407403e-01 -6.47491693e-01
6.08276665e-01 -3.52481276e-01 -3.32355648e-01 6.36198938e-01
-2.34706745e-01 4.12420511e-01 7.22506106e-01 4.93947640e-02
-2.85036743e-01 6.11114837e-02 -1.39993298e+00 5.49332798e-02
8.62298846e-01 -1.40719995e-01 1.98871747e-01 5.34227490e-01
3.30473870e-01 -4.42412138e-01 -4.95625854e-01 3.58276814e-01
1.33053827e+00 -6.96746826e-01 1.08486712e+00 -6.99018419e-01
7.86248386e-01 8.27912390e-02 -1.37195721e-01 -9.14417148e-01
-7.63280213e-01 1.65053457e-01 -4.24782544e-01 8.29738379e-01
-3.47877473e-01 -3.36551487e-01 9.57923591e-01 -3.74399006e-01
3.46121907e-01 -3.22573096e-01 -4.18499559e-01 -7.05378950e-01
7.57122934e-01 -6.16340153e-02 8.24949205e-01 7.06882119e-01
-4.04428005e-01 -3.35941203e-02 1.43881561e-02 1.49602070e-01
8.98880243e-01 1.36393011e-02 5.76810002e-01 -1.54293525e+00
6.12747908e-01 -5.74552178e-01 -7.19574273e-01 -7.38568902e-01
-1.19301882e-02 -6.85702622e-01 -3.90513204e-02 -1.68649292e+00
6.03346288e-01 6.79171160e-02 1.79884389e-01 3.81951183e-01
-4.03475106e-01 3.47224861e-01 1.05120219e-01 -3.42626572e-01
-1.63309544e-01 -8.39466751e-02 1.13814354e+00 -5.70611179e-01
4.30245325e-02 4.26151425e-01 -3.80729318e-01 9.56846178e-01
1.05879271e+00 -5.10244071e-01 2.68250346e-01 -1.43590140e+00
3.62665921e-01 -4.26693231e-01 5.84690928e-01 -8.80652726e-01
-2.45643362e-01 -2.00732574e-01 1.13852990e+00 -1.10317886e+00
-1.17499359e-01 -6.91604197e-01 -7.57192194e-01 9.63352203e-01
1.77349022e-03 1.81348145e-01 6.52594030e-01 5.04832029e-01
1.99223816e-01 -3.68658513e-01 9.18607593e-01 -5.94291747e-01
-6.16476238e-01 1.97942629e-01 -6.24135733e-01 -2.05407590e-01
7.56846905e-01 -4.22782868e-01 -3.66839856e-01 2.91652590e-01
-3.71256620e-01 3.78021568e-01 -2.18877167e-01 2.42297515e-01
9.12329033e-02 -1.25404489e+00 -1.08968198e+00 -3.24984603e-02
4.48951721e-01 -7.72876963e-02 -2.27421954e-01 4.95697826e-01
-1.28199422e+00 7.03230619e-01 -4.81293827e-01 -6.94556832e-01
-1.00557625e+00 5.45604408e-01 1.07975997e-01 -4.10701722e-01
-6.98344260e-02 9.39412951e-01 1.90685727e-02 -6.99893355e-01
2.61011660e-01 -9.91919935e-01 -6.22465432e-01 2.19878912e-01
9.01328921e-01 1.15023124e+00 3.28426749e-01 -8.63578841e-02
-3.68453890e-01 7.15118885e-01 3.02645981e-01 6.65734947e-01
1.47803068e+00 3.08403701e-01 -2.35989541e-01 7.54625261e-01
7.32277215e-01 -6.78467572e-01 -8.49502087e-01 1.43227190e-01
1.04512267e-01 -3.09794486e-01 1.40439123e-01 -7.41192460e-01
-1.25476563e+00 1.35633302e+00 9.90437269e-01 1.96566328e-01
6.57472312e-01 -2.70273596e-01 5.84509313e-01 7.65806973e-01
8.69215548e-01 -9.83183146e-01 2.02106144e-02 9.15256143e-01
6.32092476e-01 -1.90392303e+00 3.98657680e-01 -3.37549627e-01
-1.68651640e-02 1.51856422e+00 3.82434279e-01 -2.66931176e-01
6.97619021e-01 9.04302418e-01 5.63210100e-02 -4.09060299e-01
2.22320080e-01 -4.19825137e-01 -5.99599183e-01 1.20115423e+00
1.24455690e+00 7.82312214e-01 -1.26946792e-01 1.07099541e-01
-1.44368783e-02 5.78393757e-01 1.22723751e-01 1.19283092e+00
-8.98567975e-01 -5.53736687e-01 -5.57173908e-01 8.64553273e-01
-5.09457111e-01 -1.33010000e-01 -3.26191574e-01 6.39289558e-01
3.30776602e-01 1.22452497e+00 4.95956033e-01 2.73985475e-01
2.41459519e-01 -4.08731282e-01 1.04725766e+00 -6.68820858e-01
-5.34530163e-01 -4.70307857e-01 -2.88885117e-01 -2.44753420e-01
-6.12258971e-01 -5.60229480e-01 -6.23346448e-01 -1.09588242e+00
-1.15972579e+00 -5.68220437e-01 8.83469045e-01 7.23404706e-01
-2.66073316e-01 5.00924408e-01 1.57320410e-01 -1.28637147e+00
-5.30868948e-01 -1.23228538e+00 -7.42459178e-01 -6.56187296e-01
2.24599078e-01 -8.11359167e-01 1.77304689e-02 1.51588097e-01] | [9.13414478302002, -1.4677510261535645] |
c479461d-b231-4783-ac32-d5e03e03dc54 | harp-personalized-hand-reconstruction-from-a | 2212.09530 | null | https://arxiv.org/abs/2212.09530v3 | https://arxiv.org/pdf/2212.09530v3.pdf | HARP: Personalized Hand Reconstruction from a Monocular RGB Video | We present HARP (HAnd Reconstruction and Personalization), a personalized hand avatar creation approach that takes a short monocular RGB video of a human hand as input and reconstructs a faithful hand avatar exhibiting a high-fidelity appearance and geometry. In contrast to the major trend of neural implicit representations, HARP models a hand with a mesh-based parametric hand model, a vertex displacement map, a normal map, and an albedo without any neural components. As validated by our experiments, the explicit nature of our representation enables a truly scalable, robust, and efficient approach to hand avatar creation. HARP is optimized via gradient descent from a short sequence captured by a hand-held mobile phone and can be directly used in AR/VR applications with real-time rendering capability. To enable this, we carefully design and implement a shadow-aware differentiable rendering scheme that is robust to high degree articulations and self-shadowing regularly present in hand motion sequences, as well as challenging lighting conditions. It also generalizes to unseen poses and novel viewpoints, producing photo-realistic renderings of hand animations performing highly-articulated motions. Furthermore, the learned HARP representation can be used for improving 3D hand pose estimation quality in challenging viewpoints. The key advantages of HARP are validated by the in-depth analyses on appearance reconstruction, novel-view and novel pose synthesis, and 3D hand pose refinement. It is an AR/VR-ready personalized hand representation that shows superior fidelity and scalability. | ['Siyu Tang', 'Otmar Hilliges', 'Sergey Prokudin', 'Korrawe Karunratanakul'] | 2022-12-19 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Karunratanakul_HARP_Personalized_Hand_Reconstruction_From_a_Monocular_RGB_Video_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Karunratanakul_HARP_Personalized_Hand_Reconstruction_From_a_Monocular_RGB_Video_CVPR_2023_paper.pdf | cvpr-2023-1 | ['3d-hand-pose-estimation', '3d-hand-pose-estimation'] | ['computer-vision', 'graphs'] | [ 3.52924801e-02 -3.93646099e-02 8.00893754e-02 7.03112334e-02
-4.79934454e-01 -6.61334932e-01 5.03826201e-01 -7.30745614e-01
9.59939416e-03 5.98163426e-01 3.86184782e-01 1.12364024e-01
1.01068869e-01 -5.49900472e-01 -7.16255963e-01 -6.50947869e-01
7.47259855e-02 8.41992259e-01 8.80414769e-02 -3.61370593e-01
-1.31537035e-01 1.16027617e+00 -1.66489291e+00 -1.01146989e-01
3.59046072e-01 6.37918413e-01 2.92571932e-01 1.04956865e+00
2.98496276e-01 5.42724371e-01 -4.75052238e-01 -4.33897525e-01
4.80444789e-01 -1.30951494e-01 -6.99474812e-01 1.65127486e-01
6.19049251e-01 -8.27288389e-01 -6.99701428e-01 3.22494477e-01
1.01990891e+00 4.46937978e-01 4.46468055e-01 -8.67986262e-01
-5.12508631e-01 -4.05179411e-02 -5.25901854e-01 -4.89142835e-01
8.73772085e-01 4.62164104e-01 6.77763820e-01 -9.38204110e-01
1.11575973e+00 1.39570773e+00 8.67324591e-01 8.75487983e-01
-1.34942365e+00 -4.05873746e-01 2.51773298e-01 -4.10990007e-02
-1.54841650e+00 -2.11494893e-01 9.63948905e-01 -3.52890074e-01
7.89685130e-01 6.89840853e-01 1.41615498e+00 1.57003820e+00
2.50489134e-02 7.16184080e-01 9.31156754e-01 -2.76151299e-01
-4.60007265e-02 -1.66916490e-01 -4.80678380e-01 8.71218801e-01
-4.11608875e-01 2.63935238e-01 -5.64452887e-01 -2.74811506e-01
1.83274007e+00 8.39355066e-02 -5.84101439e-01 -7.43282020e-01
-1.39303219e+00 2.47759059e-01 3.89065325e-01 -1.56214431e-01
-6.65764809e-01 5.88636875e-01 9.56098959e-02 -1.15038261e-01
2.62278676e-01 3.15370351e-01 -5.47067225e-01 -3.89228493e-01
-6.76774621e-01 5.96566916e-01 7.42575586e-01 1.10650420e+00
2.85594344e-01 4.68402207e-01 -2.69183993e-01 6.56297266e-01
2.62003809e-01 7.02213764e-01 -3.28324847e-02 -1.39849365e+00
-7.17153540e-03 1.89362645e-01 2.01135710e-01 -8.85318160e-01
-4.27811921e-01 -4.11385745e-01 -7.70210326e-01 7.86689162e-01
3.65819633e-01 1.42124742e-01 -7.71375895e-01 1.73487246e+00
6.66889608e-01 1.72092453e-01 -4.10982460e-01 1.21716273e+00
8.43004167e-01 3.19692880e-01 -1.41539544e-01 -9.79140475e-02
1.38298237e+00 -8.16521585e-01 -6.70253396e-01 3.04926902e-01
-3.44260186e-01 -6.67743266e-01 1.51045156e+00 5.88678181e-01
-1.21008861e+00 -4.68970537e-01 -6.59024000e-01 -3.01811606e-01
2.89591908e-01 -8.55283290e-02 7.42562711e-01 5.94091833e-01
-1.06733918e+00 8.25179636e-01 -9.38686669e-01 -1.66415364e-01
2.10374549e-01 2.40758881e-01 -4.77735370e-01 9.90465358e-02
-6.14091039e-01 7.03562081e-01 -2.23888785e-01 1.22726418e-01
-8.97735476e-01 -9.13874805e-01 -6.54748857e-01 -1.66057751e-01
2.45614931e-01 -1.19447160e+00 1.17396748e+00 -9.02491748e-01
-2.33912849e+00 7.05787659e-01 -1.71486199e-01 3.20891529e-01
1.00857055e+00 -3.46501380e-01 1.97320402e-01 1.42705590e-01
-5.40168881e-01 3.95034403e-01 1.19180286e+00 -1.68864226e+00
5.49161509e-02 -5.68609118e-01 8.87475535e-02 4.90164846e-01
7.59903491e-02 -1.42713323e-01 -6.50205433e-01 -9.90805507e-01
2.15497881e-01 -9.54674482e-01 -1.31425366e-01 4.50276852e-01
-3.55199009e-01 1.75693572e-01 8.43616724e-01 -1.19799042e+00
7.34194517e-01 -1.70319939e+00 7.55976737e-01 2.60001123e-01
1.89370602e-01 7.37193972e-02 -8.44495818e-02 2.65771717e-01
7.18645155e-02 -3.33951235e-01 9.88974571e-02 -6.19174898e-01
4.36352491e-02 3.25570285e-01 -2.27895811e-01 3.85615110e-01
-3.02723765e-01 1.03946400e+00 -9.17866468e-01 -2.81239599e-01
3.12686771e-01 1.39286757e+00 -5.52010834e-01 5.60440361e-01
-3.25262904e-01 1.22515595e+00 -2.36874938e-01 8.64509642e-01
3.66827399e-01 -1.49537235e-01 2.35432059e-01 -4.50319439e-01
3.79149392e-02 -1.36867678e-02 -1.35342240e+00 2.05320311e+00
-5.91678262e-01 4.43926007e-01 2.29013368e-01 8.24814811e-02
6.95501208e-01 5.76087415e-01 4.12640065e-01 -4.64950114e-01
2.17993885e-01 -3.52975018e-02 -6.22649968e-01 -4.41191763e-01
5.93075514e-01 -7.36934468e-02 5.85839629e-01 5.71552336e-01
-1.64790452e-01 -3.44329476e-01 -7.55317152e-01 -3.30819003e-02
6.93120301e-01 9.88415420e-01 1.51707996e-02 1.46612719e-01
2.10860655e-01 -3.02773505e-01 2.41900086e-01 2.51694262e-01
2.85316408e-01 1.20020819e+00 -9.36419889e-03 -6.94862366e-01
-1.38901985e+00 -1.31105077e+00 4.80492786e-02 1.10043538e+00
4.33460204e-03 -2.05857456e-01 -6.75935745e-01 -1.29317909e-01
-3.77832353e-02 4.67021614e-01 -6.37509108e-01 2.91182786e-01
-1.02783787e+00 -1.34868994e-01 2.83989191e-01 8.45415652e-01
1.33407146e-01 -1.17663360e+00 -8.43511045e-01 1.84951782e-01
-1.39783785e-01 -9.84015107e-01 -7.18021691e-01 -3.59289795e-01
-8.31461728e-01 -1.00372839e+00 -1.39601684e+00 -6.94474757e-01
5.03559947e-01 4.36608493e-02 1.10447502e+00 -2.79089008e-02
-4.98784810e-01 9.18973029e-01 -1.77439854e-01 -5.85537516e-02
-5.09285450e-01 -2.33110443e-01 2.95869410e-01 -1.94291305e-02
-7.53010273e-01 -1.06460726e+00 -8.78562152e-01 2.23392650e-01
-4.24049050e-01 3.34696651e-01 3.25153545e-02 7.26368368e-01
8.95345807e-01 -7.70677090e-01 1.58466753e-02 -2.52291203e-01
5.13143480e-01 1.96340710e-01 -4.12330806e-01 2.50351965e-01
-2.87137538e-01 -9.50384885e-02 4.64618385e-01 -9.09031808e-01
-1.28275287e+00 3.35493535e-01 -4.50367421e-01 -8.95469666e-01
7.60816261e-02 -2.35251009e-01 -2.54413068e-01 -2.17197835e-01
7.60983407e-01 3.30934972e-01 2.35066995e-01 -7.14548767e-01
6.72306776e-01 3.40589941e-01 8.14386666e-01 -8.73141110e-01
8.21960926e-01 6.60873532e-01 7.84306228e-02 -9.97499585e-01
3.03235073e-02 -1.06439516e-02 -9.73950088e-01 -4.45308536e-01
7.41811097e-01 -6.42860889e-01 -1.28410304e+00 7.25255191e-01
-1.39392912e+00 -7.90010929e-01 -5.97278118e-01 3.03631693e-01
-1.00217736e+00 6.36666179e-01 -6.21220469e-01 -1.04237413e+00
-5.61604977e-01 -1.23673940e+00 1.28741133e+00 3.31173800e-02
-4.71382380e-01 -8.11753273e-01 1.61386997e-01 1.40319005e-01
1.28225937e-01 7.35002458e-01 7.76433051e-01 3.25296938e-01
-7.76839972e-01 -5.00657000e-02 1.23548426e-01 1.34700492e-01
1.66769579e-01 1.85551971e-01 -1.12076974e+00 -4.54805583e-01
-4.68810529e-01 -5.36719151e-02 1.15606114e-01 2.99216747e-01
1.05660164e+00 -4.91040289e-01 -3.05399410e-02 1.11872804e+00
1.07367587e+00 9.72402319e-02 6.33193195e-01 1.94876045e-01
1.23728979e+00 5.12684226e-01 1.88937485e-01 7.52351344e-01
5.02465844e-01 1.21361232e+00 6.93933308e-01 -8.17496404e-02
-5.54276645e-01 -4.55613643e-01 9.19351578e-02 5.79157889e-01
-1.19974864e+00 1.58405036e-01 -6.54884219e-01 1.02829844e-01
-1.59041345e+00 -8.42981935e-01 -3.11425095e-03 2.35180259e+00
8.14437926e-01 -7.59837449e-01 5.49982786e-01 1.57705709e-01
2.63993174e-01 3.02153658e-02 -7.15290189e-01 -2.42025360e-01
-1.96005255e-01 5.56317687e-01 8.85345042e-02 6.18016779e-01
-4.83592451e-01 9.99558866e-01 6.38937712e+00 3.31704855e-01
-1.08425379e+00 2.17326075e-01 -6.72385991e-02 -2.25266173e-01
-6.88577533e-01 -3.01033944e-01 -1.17441699e-01 -1.82281390e-01
8.05025846e-02 2.80939013e-01 1.04394090e+00 7.86410987e-01
9.67833251e-02 3.82408530e-01 -1.10204113e+00 1.39868653e+00
1.10795595e-01 -1.42446148e+00 2.41138592e-01 7.08021596e-02
5.39289057e-01 -3.30048412e-01 9.27539542e-02 -2.48062178e-01
1.54077768e-01 -1.22107613e+00 1.17279541e+00 7.53163636e-01
1.36681211e+00 -6.38749063e-01 1.37029946e-01 3.21492374e-01
-1.22945035e+00 2.15969205e-01 1.52144864e-01 2.98766252e-02
4.05310720e-01 -2.81076044e-01 -6.69408321e-01 4.13713485e-01
8.25113773e-01 2.63924211e-01 -4.80918437e-02 5.72735369e-01
-2.71627516e-01 -1.58943515e-02 -2.86349714e-01 2.27872699e-01
-3.56692791e-01 -1.96908280e-01 9.10737574e-01 9.91149902e-01
8.61568972e-02 3.96508276e-01 -1.04071520e-01 8.95766973e-01
2.07781538e-01 7.16224760e-02 -3.46025974e-01 2.41253719e-01
2.81235963e-01 9.90311205e-01 -4.96864498e-01 -1.40154257e-01
2.03467473e-01 1.53794491e+00 3.15550506e-01 6.73195958e-01
-6.83594227e-01 1.90237403e-01 9.36533034e-01 3.87566924e-01
1.70325011e-01 -6.08455062e-01 -3.53357732e-01 -1.14878786e+00
7.01173097e-02 -9.13856208e-01 -2.01895699e-01 -1.10896647e+00
-9.79127467e-01 1.01824510e+00 -8.71826336e-02 -1.00275791e+00
-4.34435040e-01 -6.28106475e-01 -2.59621054e-01 1.08861566e+00
-1.03720081e+00 -1.55079067e+00 -7.62214065e-01 8.87510717e-01
5.45876205e-01 -1.67278796e-02 1.21623361e+00 -6.37525544e-02
-1.31901875e-01 6.55113637e-01 -3.53838772e-01 -8.67832154e-02
4.14170265e-01 -1.13394976e+00 6.83765531e-01 3.70955527e-01
1.74479246e-01 5.70226133e-01 7.56575108e-01 -5.50894618e-01
-1.93870211e+00 -5.79973459e-01 1.37514442e-01 -8.26325476e-01
1.82442684e-02 -1.77992523e-01 -8.45726311e-01 8.59344959e-01
-1.43593684e-01 2.33266447e-02 2.90820420e-01 -4.47612181e-02
-5.95544040e-01 1.93805262e-01 -1.26195109e+00 8.76256347e-01
1.59942591e+00 -5.76376319e-01 -3.21065903e-01 3.59434038e-01
5.18073440e-01 -1.18144286e+00 -1.03739834e+00 1.20930418e-01
1.47505367e+00 -9.25593317e-01 1.41345739e+00 -6.53507292e-01
1.70817733e-01 -1.09087661e-01 -3.02838475e-01 -1.14272547e+00
-4.61388469e-01 -1.16052270e+00 -5.81451893e-01 7.87789762e-01
-2.60292679e-01 -4.57071126e-01 9.63506877e-01 6.80858552e-01
6.89978302e-02 -9.36276317e-01 -8.98154914e-01 -7.55233943e-01
-6.63657561e-02 -4.81421590e-01 9.59034204e-01 7.47020006e-01
-4.04015630e-01 -1.26321033e-01 -6.83449626e-01 2.43418828e-01
8.40345621e-01 7.79720023e-02 1.27987182e+00 -1.30454016e+00
-5.86609960e-01 -2.87539184e-01 -1.26181930e-01 -1.24853384e+00
8.46011098e-03 -6.30577326e-01 -9.97998193e-02 -1.58436060e+00
-6.39073700e-02 -6.28644586e-01 4.35543567e-01 3.38626236e-01
1.00085951e-01 5.31270087e-01 5.18203676e-01 5.82392812e-01
2.84597307e-01 6.86135888e-01 1.87262177e+00 2.00917780e-01
-5.06084383e-01 8.87809768e-02 -9.39834118e-02 9.38945830e-01
4.43065315e-01 4.43962775e-02 -3.27903271e-01 -5.89309514e-01
1.93851158e-01 4.27129030e-01 8.12927663e-01 -5.73095798e-01
-2.32811198e-01 -2.78417081e-01 5.05854070e-01 -5.18743277e-01
9.29084241e-01 -9.25384641e-01 9.68661129e-01 3.11397314e-01
4.02723178e-02 3.02573621e-01 -7.66914710e-02 3.94646794e-01
5.38793266e-01 3.85457516e-01 5.70045769e-01 -3.51388186e-01
-4.98364151e-01 6.55979872e-01 1.29739806e-01 -2.57605910e-01
7.32503593e-01 -7.70728469e-01 2.57023007e-01 -6.52425051e-01
-9.81939614e-01 -3.53924185e-01 8.72795403e-01 4.44921374e-01
9.39462423e-01 -1.44615591e+00 -7.24202156e-01 4.34274048e-01
-1.65164575e-01 3.58299345e-01 3.80816638e-01 5.63062727e-01
-9.90009725e-01 -1.25536278e-01 -3.32861662e-01 -8.10321808e-01
-1.49576449e+00 3.13311309e-01 4.72553492e-01 1.68660089e-01
-1.38459218e+00 8.26957881e-01 1.10151082e-01 -5.60631931e-01
4.33524519e-01 -5.29608168e-02 1.78460151e-01 -3.64074826e-01
6.05882406e-01 7.06560552e-01 -1.30252093e-01 -9.08477306e-01
-2.04840615e-01 9.62495625e-01 5.80051839e-01 -3.53930175e-01
1.38086271e+00 3.08757033e-02 1.04360335e-01 3.22886020e-01
7.80391514e-01 3.37764293e-01 -1.92399859e+00 -5.39121032e-02
-9.14486349e-01 -7.52163351e-01 -3.25368851e-01 -9.36684847e-01
-1.20511150e+00 8.04736733e-01 5.28559089e-01 -4.66364413e-01
7.70047426e-01 3.79143320e-02 9.01868999e-01 2.10584253e-01
8.51994574e-01 -7.02332318e-01 3.32705587e-01 4.58512455e-01
1.68668568e+00 -5.50624967e-01 -1.63471401e-02 -5.11486351e-01
-6.65212929e-01 1.18953013e+00 3.88394207e-01 3.40187829e-03
3.65751714e-01 4.75303739e-01 2.70325661e-01 -4.28489074e-02
-9.94658470e-02 1.34836972e-01 4.80844021e-01 1.20571113e+00
1.05126150e-01 1.86687455e-01 1.97012857e-01 2.85988271e-01
-5.20104706e-01 -8.95218179e-02 2.31376827e-01 7.29910314e-01
1.78545073e-01 -9.94486332e-01 -6.35051787e-01 -2.28349686e-01
5.44288270e-02 1.83642805e-01 -2.87916750e-01 8.45455885e-01
-1.00059837e-01 2.18347892e-01 -4.91204411e-02 -5.61664343e-01
6.53152823e-01 -1.99760534e-02 1.27488089e+00 -3.74796420e-01
-7.71271646e-01 9.59256291e-02 -1.22775815e-01 -8.02962899e-01
-2.73854971e-01 -5.34376979e-01 -1.27271295e+00 -5.41849494e-01
4.12970074e-02 -4.80361700e-01 8.19978416e-01 5.63975275e-01
4.23894376e-01 6.05155408e-01 3.64435524e-01 -1.94326329e+00
-5.42221963e-01 -5.78260601e-01 -6.57515407e-01 4.50782359e-01
4.83862847e-01 -8.67606938e-01 3.91792273e-03 8.00317600e-02] | [7.169700622558594, -1.282175064086914] |
775da4f1-bbd3-4554-9f76-d8e1391086b2 | a-hybrid-data-association-framework-for | 1703.10764 | null | http://arxiv.org/abs/1703.10764v1 | http://arxiv.org/pdf/1703.10764v1.pdf | A Hybrid Data Association Framework for Robust Online Multi-Object Tracking | Global optimization algorithms have shown impressive performance in
data-association based multi-object tracking, but handling online data remains
a difficult hurdle to overcome. In this paper, we present a hybrid data
association framework with a min-cost multi-commodity network flow for robust
online multi-object tracking. We build local target-specific models interleaved
with global optimization of the optimal data association over multiple video
frames. More specifically, in the min-cost multi-commodity network flow, the
target-specific similarities are online learned to enforce the local
consistency for reducing the complexity of the global data association.
Meanwhile, the global data association taking multiple video frames into
account alleviates irrecoverable errors caused by the local data association
between adjacent frames. To ensure the efficiency of online tracking, we give
an efficient near-optimal solution to the proposed min-cost multi-commodity
flow problem, and provide the empirical proof of its sub-optimality. The
comprehensive experiments on real data demonstrate the superior tracking
performance of our approach in various challenging situations. | ['Yuwei Wu', 'Yunde Jia', 'Min Yang'] | 2017-03-31 | null | null | null | null | ['online-multi-object-tracking'] | ['computer-vision'] | [-4.68305200e-01 -4.42261219e-01 -5.93760192e-01 -1.71350449e-01
-6.46792471e-01 -4.50525910e-01 -1.49341613e-01 -1.85595065e-01
-4.02199805e-01 6.97879255e-01 -2.55971611e-01 -7.72636011e-02
-6.84550345e-01 -2.28977188e-01 -7.87398577e-01 -7.83506513e-01
-3.89444500e-01 5.10997176e-01 5.05715072e-01 2.63835579e-01
7.09649175e-03 6.15039051e-01 -1.23169553e+00 -2.83595622e-01
7.00851023e-01 1.26590288e+00 3.05300385e-01 6.18664980e-01
1.54238537e-01 5.64301431e-01 -4.03894782e-01 -2.97220051e-01
6.98656797e-01 -4.18029539e-02 -4.59015191e-01 5.92133224e-01
9.88463163e-01 -6.47082388e-01 -2.99388379e-01 1.22489250e+00
2.96425045e-01 2.17195019e-01 -1.29816964e-01 -1.94743729e+00
-3.37627590e-01 5.68728559e-02 -8.67833555e-01 5.16219020e-01
-6.27034232e-02 3.25010121e-01 1.05054212e+00 -6.63776577e-01
8.02510023e-01 1.31705618e+00 5.67754149e-01 3.38942528e-01
-1.02598405e+00 -8.92248213e-01 6.84250832e-01 3.45433086e-01
-1.33668196e+00 -2.36666381e-01 4.30168122e-01 -4.34064031e-01
1.74676508e-01 2.67202556e-01 9.05899882e-01 3.76514554e-01
1.72761485e-01 7.65343487e-01 4.94490832e-01 8.66225883e-02
-2.55512506e-01 -1.95919171e-01 1.17586069e-01 9.14808273e-01
7.60339200e-01 2.57435620e-01 -6.40396953e-01 -2.44127154e-01
9.92923796e-01 2.83619940e-01 -1.82069540e-01 -5.38493454e-01
-1.32322729e+00 6.77357435e-01 5.52435219e-01 7.56966835e-03
-2.73092270e-01 3.55410278e-01 3.63270521e-01 2.87287772e-01
2.28476301e-01 -2.14799568e-01 -4.35652137e-01 2.16351539e-01
-8.75830233e-01 4.96464431e-01 4.24270689e-01 1.52771223e+00
7.42187202e-01 1.64983362e-01 -3.18755090e-01 3.71860594e-01
6.89067245e-01 6.52814746e-01 -7.14895502e-02 -1.24788034e+00
7.28233695e-01 4.11904603e-01 4.27535355e-01 -1.36867642e+00
-2.94927686e-01 -5.76903641e-01 -6.98057175e-01 2.35361829e-01
7.27125049e-01 -3.64150643e-01 -3.72448593e-01 1.79048538e+00
1.03143013e+00 6.33186102e-01 -3.51891428e-01 1.41810417e+00
7.40272999e-02 5.29573262e-01 1.73617542e-01 -8.31404150e-01
1.34095693e+00 -1.10343373e+00 -9.16599333e-01 5.06938063e-03
4.76711333e-01 -7.37737060e-01 1.92047149e-01 -6.84131822e-03
-9.10377920e-01 -6.63041472e-01 -8.06065559e-01 2.79040307e-01
2.01416537e-01 -4.67650518e-02 4.76638794e-01 3.72521967e-01
-6.67167544e-01 2.04280615e-01 -6.64722264e-01 -1.43005267e-01
4.82393295e-01 6.69832468e-01 -2.25113615e-01 -5.44593781e-02
-7.62808025e-01 4.79419321e-01 7.21146166e-01 4.43456799e-01
-9.01608646e-01 -1.00904465e+00 -4.89079475e-01 -1.57030910e-01
1.11678362e+00 -7.14362264e-01 9.60501015e-01 -7.80776680e-01
-1.07288349e+00 4.24166024e-01 -1.06717393e-01 -3.65810275e-01
6.67955637e-01 -4.12489682e-01 -3.30251485e-01 3.48507091e-02
2.26242036e-01 4.20771599e-01 8.75830352e-01 -1.02527058e+00
-1.25068474e+00 -2.51129478e-01 1.65946141e-01 2.39305153e-01
-4.13191795e-01 2.07790017e-01 -8.71490538e-01 -8.19042623e-01
-5.59300669e-02 -1.15977907e+00 -2.97680080e-01 8.59075844e-01
-1.83446750e-01 -2.26539284e-01 1.26208031e+00 -4.81200665e-01
1.21970618e+00 -2.14851904e+00 2.65424728e-01 1.61679149e-01
3.15631092e-01 2.24571705e-01 -7.75875822e-02 -1.68297634e-01
3.95516366e-01 -3.52522045e-01 2.11508557e-01 -3.46830577e-01
-7.30773583e-02 3.58320653e-01 -3.92206013e-02 7.46948123e-01
-4.30992767e-02 6.93897843e-01 -9.99188900e-01 -1.05145967e+00
1.94308355e-01 1.35032147e-01 -5.56232452e-01 2.46265024e-01
-3.24794590e-01 5.83399236e-01 -8.35992157e-01 8.44822705e-01
8.70995343e-01 -2.89878577e-01 2.37362862e-01 -5.62120259e-01
-3.76056492e-01 -6.40465200e-01 -1.67646098e+00 1.61603439e+00
5.39153740e-02 4.34308618e-01 5.88697433e-01 -7.11549103e-01
4.89211291e-01 9.89404097e-02 1.09220517e+00 -4.63097870e-01
6.18294403e-02 7.72095025e-02 -6.18025381e-03 -2.73375690e-01
6.24865592e-01 1.25880614e-01 8.48909989e-02 2.16510311e-01
-1.58735812e-01 7.79610097e-01 3.25859338e-01 2.43669569e-01
7.13135481e-01 1.77261502e-01 1.42698377e-01 -3.41836065e-01
7.69362688e-01 2.53385812e-01 1.31525075e+00 6.52163327e-01
-6.79356396e-01 1.03500001e-01 -9.15730670e-02 -5.72160721e-01
-8.03312719e-01 -8.28320444e-01 1.67299375e-01 1.15063357e+00
8.08578730e-01 -3.47040087e-01 -2.16413856e-01 -6.83573365e-01
4.11295444e-01 -2.32870683e-01 -1.74439833e-01 2.18707427e-01
-8.66999686e-01 -6.49129450e-01 8.69522616e-02 3.60554248e-01
3.72734100e-01 -4.13445085e-01 -5.87683260e-01 4.47950780e-01
-2.11401671e-01 -1.45347393e+00 -1.19737482e+00 -5.57931125e-01
-9.29207921e-01 -1.27138257e+00 -5.08017123e-01 -6.02881730e-01
7.97502935e-01 7.41512418e-01 5.87101638e-01 6.10295534e-01
-3.95291656e-01 3.85383576e-01 -6.49420470e-02 -6.82991222e-02
2.30727479e-01 -2.48225465e-01 3.05087864e-01 3.26596290e-01
-2.15861887e-01 1.80907454e-02 -5.21131575e-01 8.86188507e-01
-6.61090970e-01 -1.53790355e-01 4.20513391e-01 6.52447343e-01
1.03206909e+00 4.72415797e-02 3.02060783e-01 -3.29514503e-01
-4.04110923e-02 -4.05653507e-01 -1.30870962e+00 4.88238513e-01
-3.65101367e-01 -7.76250809e-02 4.77749348e-01 -7.67777026e-01
-9.31339562e-01 2.54539639e-01 5.26007771e-01 -1.02193034e+00
4.15521413e-01 1.24506705e-01 -2.93471456e-01 -6.84446931e-01
-3.08272660e-01 -5.39795794e-02 6.68053851e-02 -3.97564560e-01
2.90474892e-01 5.28349653e-02 5.38136423e-01 -8.29958737e-01
1.20308590e+00 7.66426206e-01 3.44482124e-01 -4.44510549e-01
-1.07967126e+00 -7.20428050e-01 -4.19973016e-01 -8.39228928e-01
7.91593909e-01 -1.05802774e+00 -1.35893619e+00 3.37015212e-01
-1.10329342e+00 4.35885079e-02 -1.00837030e-01 6.37104034e-01
-4.76056218e-01 5.72272241e-01 -4.62909460e-01 -8.72651756e-01
-1.91261426e-01 -9.74578142e-01 7.84666657e-01 3.97001535e-01
3.44040424e-01 -9.61474001e-01 -1.09213157e-04 4.11452085e-01
1.89954579e-01 3.24289083e-01 1.40090093e-01 -1.85027525e-01
-1.43769026e+00 1.87131748e-01 -4.61983681e-01 -2.40637675e-01
7.35121295e-02 1.08365111e-01 -2.07161129e-01 -7.63612747e-01
-2.14242443e-01 2.13320702e-02 3.41335237e-01 4.41158682e-01
1.06105292e+00 -5.73985934e-01 -5.57060361e-01 7.47164845e-01
1.73906195e+00 3.87094729e-02 -3.47513035e-02 4.94560808e-01
1.01611531e+00 3.55506778e-01 1.51616073e+00 7.93552518e-01
4.43196058e-01 1.00271094e+00 7.44655907e-01 1.58962861e-01
-1.06250130e-01 1.20881855e-01 3.32156658e-01 7.23971009e-01
-9.85911041e-02 -2.44403020e-01 -3.53619099e-01 5.91810942e-01
-2.42214084e+00 -1.06563628e+00 -3.67871851e-01 2.26310897e+00
4.86260772e-01 2.09111115e-03 4.79902029e-01 -4.56206262e-01
1.08988130e+00 2.11259514e-01 -6.78267419e-01 4.02525514e-01
-5.64912334e-02 -6.74942553e-01 7.94999659e-01 5.28262794e-01
-1.22235763e+00 6.25725508e-01 5.87834358e+00 8.85320127e-01
-8.01575422e-01 2.83295125e-01 8.50049630e-02 -4.44370091e-01
2.34932423e-01 5.53035140e-02 -1.37423599e+00 8.00030947e-01
5.71247458e-01 -4.38977659e-01 2.81216890e-01 7.76887298e-01
2.56071150e-01 -3.48305106e-02 -8.41428220e-01 1.01437569e+00
-2.22790048e-01 -1.47135961e+00 -5.06325699e-02 3.35951716e-01
6.58587813e-01 -2.15712324e-01 -9.76409912e-02 -1.95454702e-01
2.50054359e-01 -5.33505790e-02 9.32063282e-01 5.07436752e-01
4.92066979e-01 -5.98222077e-01 3.07416260e-01 2.09460258e-01
-1.97544169e+00 -3.18316638e-01 -4.36288744e-01 3.00431639e-01
5.63185096e-01 4.58057910e-01 -1.08584493e-01 1.17025113e+00
8.09001207e-01 8.37346673e-01 -2.04349011e-01 1.35943627e+00
3.23747873e-01 -5.21580316e-02 -5.48054457e-01 3.60965818e-01
2.12833837e-01 -3.54345977e-01 9.62943316e-01 8.94496441e-01
2.40407825e-01 1.29061371e-01 9.03859138e-01 3.81489038e-01
7.29619013e-03 -9.91226360e-02 -1.31830350e-01 1.56655133e-01
6.03379488e-01 1.42886484e+00 -4.85149622e-01 -3.52712333e-01
-6.13640487e-01 4.64735478e-01 4.09140468e-01 1.42224789e-01
-1.31263387e+00 1.47130281e-01 1.08901453e+00 -1.79577038e-01
5.39926231e-01 -4.35419023e-01 9.77693945e-02 -1.12039077e+00
2.56046176e-01 -5.36992013e-01 9.98133838e-01 -3.94336820e-01
-1.22489381e+00 6.69269487e-02 -6.79030418e-02 -1.64692891e+00
1.11917093e-01 -3.94652307e-01 -3.81018221e-01 3.80752057e-01
-1.70611453e+00 -1.16380203e+00 -3.99326384e-01 9.29047823e-01
4.97540206e-01 3.61745134e-02 -4.49350961e-02 8.78684878e-01
-8.18861604e-01 6.30467176e-01 8.43456294e-03 1.06165767e-01
8.05039585e-01 -7.55690992e-01 -3.46343189e-01 1.17352927e+00
-1.93958580e-01 3.52881461e-01 6.82923555e-01 -8.84946048e-01
-1.87301743e+00 -1.46601617e+00 4.71361399e-01 -1.16136037e-02
1.03506446e+00 4.87002023e-02 -8.94520581e-01 9.60302114e-01
-1.19510651e-01 5.32996416e-01 3.29277784e-01 -1.71113268e-01
-1.01647869e-01 -5.80699742e-01 -1.02033317e+00 3.10634166e-01
1.26825213e+00 4.15520072e-02 -1.36301428e-01 4.93563473e-01
9.18451726e-01 -6.22317016e-01 -1.16730666e+00 2.22874612e-01
7.18763828e-01 -4.31196213e-01 1.21014988e+00 -6.84674203e-01
-3.95826161e-01 -9.22902822e-01 -7.77471885e-02 -6.56948566e-01
-6.09486341e-01 -8.30645025e-01 -6.27772868e-01 1.28613591e+00
-1.45534426e-01 -4.95769590e-01 8.41297209e-01 5.69696546e-01
-1.31155580e-01 -3.32082152e-01 -1.21176171e+00 -1.29577827e+00
-6.39808774e-01 -4.67788428e-02 2.36830667e-01 9.03568625e-01
-6.04080737e-01 -2.35485509e-01 -9.26025391e-01 7.57485867e-01
1.44300532e+00 5.57146251e-01 9.52320516e-01 -1.15636837e+00
-3.52417916e-01 -7.49399439e-02 -4.71345007e-01 -1.20540679e+00
1.12813532e-01 -4.98012334e-01 1.13690518e-01 -1.05716264e+00
4.42305923e-01 -7.73766816e-01 -4.38833475e-01 2.57624328e-01
-4.81319904e-01 1.37257138e-02 8.35668862e-01 5.56215584e-01
-1.43921077e+00 4.80724871e-01 1.56657493e+00 -1.42877981e-01
6.77555054e-02 -4.81343232e-02 -3.58365476e-01 3.23912859e-01
3.99780333e-01 -8.19438815e-01 -1.56225219e-01 -6.83042586e-01
-2.23557562e-01 3.18016678e-01 4.03234094e-01 -9.67976987e-01
7.52225757e-01 -6.81213915e-01 1.60253063e-01 -7.77114928e-01
4.29024458e-01 -1.48737550e+00 6.08789206e-01 8.52830529e-01
1.22021742e-01 1.35740578e-01 1.70849055e-01 1.09981680e+00
-1.30730256e-01 6.37238473e-02 8.49837720e-01 2.73595661e-01
-1.06287897e+00 1.12771058e+00 1.97091058e-01 2.88735963e-02
1.49434340e+00 -8.63279924e-02 -3.42753559e-01 4.18175757e-02
-7.24606872e-01 1.00559163e+00 3.48782748e-01 4.75307047e-01
3.50834548e-01 -1.74348593e+00 -5.51481903e-01 -7.21190050e-02
-1.56146824e-01 -6.91437125e-02 5.57687759e-01 1.27893186e+00
-1.97820529e-01 3.17762583e-01 -3.70731443e-01 -9.76471245e-01
-1.73412347e+00 7.72483110e-01 3.16142559e-01 -3.44793260e-01
-5.76887846e-01 5.55348516e-01 1.86738327e-01 1.24893352e-01
2.18962654e-01 4.06950153e-02 2.99209177e-01 -3.45244780e-02
6.18707299e-01 7.50203431e-01 -4.64541167e-01 -7.24307239e-01
-5.70556402e-01 8.89252901e-01 -1.99025303e-01 2.80863851e-01
1.10448253e+00 -6.10016704e-01 -3.73106264e-02 5.00445105e-02
9.09120142e-01 -3.76559198e-01 -1.84514344e+00 -5.54643571e-01
5.39873354e-02 -1.10605037e+00 -2.53308229e-02 -1.46106914e-01
-1.62958956e+00 1.58432633e-01 6.10007048e-01 -8.20107386e-03
1.02191865e+00 -1.33008227e-01 9.62142289e-01 2.83558279e-01
6.45470977e-01 -1.14286709e+00 3.81762125e-02 2.88262963e-01
5.00728607e-01 -1.26426375e+00 3.12076509e-01 -7.17310905e-01
-2.86307037e-01 9.76322055e-01 9.29792881e-01 -1.67939425e-01
5.70109665e-01 1.36657581e-01 -2.09254235e-01 -1.22766785e-01
-8.34273756e-01 -2.42955431e-01 1.98647201e-01 2.84063786e-01
-3.18932503e-01 -9.26769525e-02 -3.19696069e-01 8.64931792e-02
5.98807454e-01 3.81830111e-02 1.26907527e-01 1.07710660e+00
-5.36743104e-01 -9.73484814e-01 -7.61673570e-01 2.05026880e-01
-4.26562846e-01 5.30925453e-01 2.88572669e-01 8.94234419e-01
1.78312093e-01 7.92400181e-01 1.91281617e-01 2.06630174e-02
2.55002528e-01 -4.62472975e-01 5.86212099e-01 -1.36451513e-01
-4.51231271e-01 3.88512909e-01 -7.79094845e-02 -8.62875640e-01
-8.91306520e-01 -7.37182319e-01 -1.12697637e+00 -6.77774847e-01
-5.76399386e-01 1.11742727e-02 3.51691723e-01 9.42210138e-01
5.05562484e-01 5.21282673e-01 7.13298023e-01 -7.20510066e-01
-4.09084707e-01 -3.85584831e-01 -4.50509191e-01 4.60491389e-01
6.06469274e-01 -9.16465819e-01 -1.75969601e-01 2.34497458e-01] | [6.419556140899658, -2.077073574066162] |
359437f3-b66c-40a2-a41e-85a120af2503 | gcfsr-a-generative-and-controllable-face | 2203.07319 | null | https://arxiv.org/abs/2203.07319v1 | https://arxiv.org/pdf/2203.07319v1.pdf | GCFSR: a Generative and Controllable Face Super Resolution Method Without Facial and GAN Priors | Face image super resolution (face hallucination) usually relies on facial priors to restore realistic details and preserve identity information. Recent advances can achieve impressive results with the help of GAN prior. They either design complicated modules to modify the fixed GAN prior or adopt complex training strategies to finetune the generator. In this work, we propose a generative and controllable face SR framework, called GCFSR, which can reconstruct images with faithful identity information without any additional priors. Generally, GCFSR has an encoder-generator architecture. Two modules called style modulation and feature modulation are designed for the multi-factor SR task. The style modulation aims to generate realistic face details and the feature modulation dynamically fuses the multi-level encoded features and the generated ones conditioned on the upscaling factor. The simple and elegant architecture can be trained from scratch in an end-to-end manner. For small upscaling factors (<=8), GCFSR can produce surprisingly good results with only adversarial loss. After adding L1 and perceptual losses, GCFSR can outperform state-of-the-art methods for large upscaling factors (16, 32, 64). During the test phase, we can modulate the generative strength via feature modulation by changing the conditional upscaling factor continuously to achieve various generative effects. | ['Chao Dong', 'Lean Fu', 'Kai Chen', 'Wu Shi', 'Jingwen He'] | 2022-03-14 | null | http://openaccess.thecvf.com//content/CVPR2022/html/He_GCFSR_A_Generative_and_Controllable_Face_Super_Resolution_Method_Without_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/He_GCFSR_A_Generative_and_Controllable_Face_Super_Resolution_Method_Without_CVPR_2022_paper.pdf | cvpr-2022-1 | ['face-hallucination'] | ['computer-vision'] | [ 4.81148720e-01 1.88258678e-01 1.30200803e-01 -4.32838291e-01
-7.75954783e-01 -2.92462558e-01 6.69799864e-01 -8.43693793e-01
2.54320968e-02 7.83453524e-01 2.74233997e-01 1.42764002e-01
2.16401249e-01 -9.32138145e-01 -6.57493532e-01 -8.18982184e-01
2.81458110e-01 1.95230708e-01 -1.07423075e-01 -4.97709990e-01
-1.08007723e-02 5.09505391e-01 -1.55374002e+00 3.15178007e-01
9.27145541e-01 8.99134517e-01 3.14795792e-01 7.00880885e-01
-5.22927791e-02 6.25082314e-01 -7.24036038e-01 -6.84344471e-01
5.84849775e-01 -8.62484097e-01 -2.43519753e-01 1.23869762e-01
4.93341237e-01 -4.58207428e-01 -5.56095064e-01 1.03051043e+00
7.12546229e-01 -1.16541341e-01 4.40714538e-01 -1.07400823e+00
-9.62011874e-01 3.19649547e-01 -6.38478398e-01 -3.63251865e-02
3.40478331e-01 4.59789783e-01 6.14458382e-01 -9.06690478e-01
5.20737469e-01 1.70301187e+00 5.59701025e-01 1.01659513e+00
-1.52354610e+00 -8.68768990e-01 -6.29221424e-02 5.21473326e-02
-1.39797235e+00 -7.79600382e-01 9.01919961e-01 -1.92759618e-01
2.76272088e-01 2.60561824e-01 3.96512628e-01 1.30256915e+00
-4.15621363e-02 2.68509716e-01 1.33948028e+00 -2.49123827e-01
6.62988499e-02 -1.48667749e-02 -7.86400735e-01 7.26849020e-01
-1.70100659e-01 2.41137162e-01 -4.71693754e-01 -5.36515191e-02
1.40521979e+00 4.71019670e-02 -5.82719982e-01 4.96761315e-02
-9.76533830e-01 9.24670517e-01 5.14984190e-01 7.94763193e-02
-2.66927093e-01 2.28548139e-01 -2.16868389e-02 5.05249798e-01
3.14760029e-01 4.42573041e-01 -1.59387559e-01 2.53057808e-01
-9.71624494e-01 1.25807300e-01 3.49322379e-01 8.28377664e-01
8.15801263e-01 4.56415087e-01 -7.74081349e-01 1.10412180e+00
2.72109807e-02 6.65699422e-01 4.21777397e-01 -1.06044257e+00
2.43124232e-01 1.42548651e-01 6.68349713e-02 -8.49583089e-01
2.34010413e-01 -5.12518942e-01 -1.12185335e+00 3.85376960e-01
7.63030052e-02 -1.52418211e-01 -1.27746296e+00 2.11541891e+00
3.24165255e-01 4.03725177e-01 -7.02615380e-02 8.89564097e-01
8.66121769e-01 6.33701742e-01 -1.14712887e-01 -2.53557414e-01
1.38458526e+00 -8.50970089e-01 -9.08962786e-01 -2.11584479e-01
-2.31442288e-01 -8.87304544e-01 1.36106241e+00 2.51405537e-01
-1.36796153e+00 -7.67387033e-01 -1.12907279e+00 -5.29790223e-02
1.63187623e-01 2.02516794e-01 5.32294452e-01 5.73863506e-01
-1.34126937e+00 6.93741024e-01 -6.53233945e-01 2.50692070e-01
8.29261720e-01 3.59526902e-01 -5.06543934e-01 -3.69649440e-01
-1.14701331e+00 6.06177151e-01 -5.16539328e-02 1.04830265e-01
-1.16923237e+00 -8.40036213e-01 -8.07870805e-01 1.43861488e-01
2.01836571e-01 -1.14252043e+00 7.07628012e-01 -1.14468443e+00
-2.02845025e+00 7.13576078e-01 -1.21297628e-01 -7.29131922e-02
6.15291536e-01 9.23988074e-02 -4.04693753e-01 2.48405278e-01
-3.06689157e-03 6.95718586e-01 1.64577043e+00 -1.43692827e+00
-1.00015827e-01 -2.89962173e-01 2.38816403e-02 4.24896404e-02
-3.09100062e-01 5.82649149e-02 -4.13517475e-01 -1.16388941e+00
-1.79236710e-01 -7.89590836e-01 -1.87923610e-01 1.14695467e-01
-3.10210615e-01 4.01171416e-01 6.69637859e-01 -7.85835087e-01
7.68654585e-01 -2.22951078e+00 3.60384077e-01 -1.65368780e-01
2.10291654e-01 2.81842172e-01 -4.92882490e-01 1.71827331e-01
-1.53348520e-01 4.30555232e-02 -3.96394461e-01 -6.05604291e-01
-1.28101662e-01 1.24085248e-01 -3.35262954e-01 2.32747138e-01
4.29028809e-01 8.80723715e-01 -7.61172235e-01 -1.74119949e-01
1.93185676e-02 1.11883020e+00 -8.70289683e-01 5.31281769e-01
-9.99742076e-02 7.61461675e-01 -2.86782831e-01 4.96012241e-01
9.28943455e-01 -1.70736089e-01 5.41879088e-02 -3.06274593e-01
3.07697654e-01 -9.58926752e-02 -1.02977371e+00 1.67843127e+00
-8.31548214e-01 3.71075928e-01 3.32149595e-01 -5.01487672e-01
1.14575696e+00 3.05263847e-01 6.29740655e-02 -5.00841439e-01
2.28110366e-02 5.87506965e-02 -2.04496950e-01 -9.69167277e-02
2.33389094e-01 -6.82644665e-01 1.31776929e-01 8.56246948e-02
2.55757987e-01 -1.23041987e-01 -2.36810640e-01 3.83086060e-03
1.12166429e+00 6.77519217e-02 -3.54266241e-02 -5.60242608e-02
7.04040468e-01 -8.22277606e-01 6.30902708e-01 2.72601068e-01
3.60119134e-01 1.10746300e+00 4.67758328e-01 -2.72232056e-01
-1.06805468e+00 -1.17501724e+00 -9.05255675e-02 1.00075912e+00
-6.78559244e-02 -3.48369747e-01 -8.44397724e-01 -5.19551158e-01
-1.90929681e-01 5.14957964e-01 -8.08599234e-01 -3.36161911e-01
-5.74569643e-01 -7.30867445e-01 4.50816810e-01 4.39447522e-01
9.64624763e-01 -1.08276522e+00 -9.43072066e-02 -2.71658543e-02
-1.83452129e-01 -1.06014609e+00 -7.82602608e-01 -2.77761668e-01
-5.54205716e-01 -6.10534608e-01 -9.08517420e-01 -6.46634579e-01
9.01284993e-01 -2.42351852e-02 1.02793360e+00 2.23923638e-01
-2.96695709e-01 -7.88072199e-02 -1.53487131e-01 6.20272495e-02
-6.23053372e-01 -2.36864075e-01 -2.54500031e-01 3.24568957e-01
-5.00826061e-01 -1.13492465e+00 -8.74196351e-01 3.97068262e-01
-1.03693557e+00 1.76543668e-01 6.68194354e-01 1.11805689e+00
6.02046907e-01 8.35057627e-03 8.10773909e-01 -9.42855656e-01
3.85556430e-01 -2.33671814e-01 -3.85939956e-01 1.41689122e-01
-3.92841995e-01 2.03091785e-01 1.04872572e+00 -6.63881481e-01
-1.24212527e+00 -2.22734526e-01 -3.32790405e-01 -9.63096499e-01
1.06718339e-01 -1.74319461e-01 -7.16754258e-01 -1.91513866e-01
5.91090381e-01 3.19463879e-01 -4.82811294e-02 -5.57878911e-01
6.34011149e-01 4.33941871e-01 7.40099967e-01 -5.77050030e-01
1.04292405e+00 5.43676436e-01 4.19963561e-02 -4.08041120e-01
-7.13851273e-01 4.39335108e-01 -1.91307560e-01 -6.67669848e-02
6.28656387e-01 -1.18243659e+00 -3.90155852e-01 5.66433907e-01
-8.34592283e-01 -4.38096136e-01 -4.32195842e-01 6.31027296e-02
-6.18391156e-01 5.89443110e-02 -7.06818521e-01 -4.80268270e-01
-4.16453242e-01 -1.09051812e+00 1.28430259e+00 2.96675205e-01
2.79537976e-01 -5.66667378e-01 -1.97200835e-01 4.10023421e-01
9.29733276e-01 5.63399196e-01 7.57314503e-01 1.54443860e-01
-6.10118270e-01 8.27687159e-02 -1.18903443e-01 5.84846735e-01
4.36201990e-01 -3.11307251e-01 -1.02474868e+00 -6.70233727e-01
1.77571297e-01 -2.33531475e-01 1.02173829e+00 1.39317280e-02
1.44826376e+00 -7.62705743e-01 7.23028407e-02 1.07327509e+00
1.35055208e+00 7.27275088e-02 1.18928969e+00 -1.67947829e-01
9.42313910e-01 2.60410964e-01 2.36905634e-01 4.60532725e-01
1.94684714e-01 8.23412240e-01 3.50477576e-01 -9.98470858e-02
-6.34371758e-01 -5.83205938e-01 5.20359516e-01 3.53068471e-01
-2.95310318e-01 -8.88167098e-02 -1.55181125e-01 1.48237810e-01
-1.30949867e+00 -1.04927278e+00 5.86068273e-01 2.28518200e+00
1.20597601e+00 -6.67502061e-02 -8.86515677e-02 6.70069754e-02
9.45249975e-01 3.15573245e-01 -6.59304976e-01 -1.24449544e-01
-1.53689966e-01 6.60215616e-01 1.40432134e-01 5.49443185e-01
-6.82968259e-01 1.13855255e+00 5.53379679e+00 1.24367929e+00
-1.29954708e+00 2.61402786e-01 8.17341745e-01 -2.90977925e-01
-7.35868216e-01 -8.54012147e-02 -6.98423505e-01 6.97964489e-01
7.19212651e-01 1.95090383e-01 9.15815055e-01 6.34891868e-01
-1.17758242e-03 4.80409205e-01 -8.07976186e-01 1.18497038e+00
1.60086513e-01 -1.34617972e+00 3.61213267e-01 1.18079018e-02
8.67759585e-01 -5.60092330e-01 3.59328896e-01 2.81358898e-01
1.59132987e-01 -1.38602531e+00 5.56152582e-01 6.52370930e-01
1.55843925e+00 -9.47831035e-01 4.61930305e-01 -5.97562157e-02
-9.47968006e-01 -7.37210438e-02 -4.28034455e-01 3.14034879e-01
2.56334633e-01 5.81309795e-01 -4.21528608e-01 3.69946003e-01
4.93473917e-01 3.63275677e-01 -6.01390004e-01 3.34582001e-01
-5.51608086e-01 3.75955880e-01 -8.33751485e-02 6.86445475e-01
-2.48492554e-01 -2.93208420e-01 5.75221837e-01 7.76677489e-01
5.45574427e-01 1.52751252e-01 -1.25188217e-01 1.05765676e+00
-3.93988639e-01 -2.46042803e-01 -3.41982812e-01 1.21429160e-01
5.58543265e-01 1.44335842e+00 -3.77526104e-01 -1.53964281e-01
-2.32429896e-03 1.32128716e+00 2.00271383e-01 3.38564754e-01
-9.06588554e-01 -2.87479460e-01 8.18809032e-01 5.50707996e-01
6.61415160e-01 2.26693135e-02 -4.23690192e-02 -1.15334666e+00
-2.94460077e-02 -1.12367690e+00 6.71292990e-02 -7.90405631e-01
-1.30647051e+00 1.16951978e+00 -2.25532189e-01 -9.55632865e-01
-3.38672489e-01 -1.80873409e-01 -8.20114553e-01 9.66659725e-01
-1.49677765e+00 -1.38049519e+00 -5.08096814e-01 9.67788994e-01
3.13482612e-01 -3.99916053e-01 8.53009641e-01 3.03931266e-01
-5.39566636e-01 1.11223388e+00 -2.93559313e-01 -6.35337159e-02
7.41338968e-01 -9.95287478e-01 3.35269600e-01 8.31195593e-01
-1.43806145e-01 5.43033481e-01 7.58074820e-01 -4.48689520e-01
-1.30692494e+00 -1.30307364e+00 3.23279738e-01 -1.48828089e-01
1.40630841e-01 -5.14623284e-01 -8.36902678e-01 4.86522079e-01
2.42590338e-01 5.08186698e-01 4.93979931e-01 -3.69282007e-01
-6.45417035e-01 -3.38644266e-01 -1.71495664e+00 7.00793862e-01
1.39757955e+00 -4.24439102e-01 -1.03956774e-01 7.30575174e-02
9.07607377e-01 -3.76263440e-01 -1.03558505e+00 5.68884373e-01
2.41683811e-01 -1.10478067e+00 1.02606916e+00 -3.35089296e-01
7.05379665e-01 -5.14013410e-01 -2.29745448e-01 -1.44753397e+00
-4.29882377e-01 -1.19166422e+00 -1.16860949e-01 1.43895113e+00
4.69965078e-02 -7.15344906e-01 5.70703387e-01 6.87341020e-02
-8.42786208e-02 -8.08238328e-01 -8.05611253e-01 -6.23746932e-01
-6.87558800e-02 1.95987210e-01 1.08923090e+00 7.17836857e-01
-6.72737539e-01 4.56704676e-01 -7.38061368e-01 1.32248461e-01
7.37861812e-01 2.24038914e-01 8.02768528e-01 -8.06215286e-01
-7.15675175e-01 -2.42883205e-01 -2.84228086e-01 -9.20861125e-01
2.04660743e-01 -6.62486792e-01 -1.25694901e-01 -1.05370474e+00
1.48113668e-01 -3.88638258e-01 6.36775270e-02 5.17750144e-01
-3.83535832e-01 6.10910356e-01 3.63295436e-01 5.21164313e-02
-5.17017692e-02 1.07898426e+00 1.92231166e+00 1.70655563e-01
-4.67192521e-03 -1.79383874e-01 -1.13571072e+00 5.32069147e-01
7.71451473e-01 -1.81158915e-01 -5.69685459e-01 -3.41916561e-01
-4.67255563e-02 3.18103820e-01 4.60673600e-01 -8.90094221e-01
-2.82651007e-01 -1.46392062e-01 7.43528247e-01 2.16864999e-02
6.79269195e-01 -3.75544280e-01 6.26353204e-01 1.20167881e-01
-2.66329259e-01 -9.48760062e-02 -3.39263165e-03 5.76098919e-01
-2.02710405e-01 1.87224612e-01 1.37003517e+00 -1.69320613e-01
-4.97935303e-02 5.89771867e-01 2.27756009e-01 7.26649761e-02
7.83772230e-01 4.23644930e-02 -2.57812053e-01 -5.57564020e-01
-6.38721585e-01 -1.62488073e-01 7.44415343e-01 5.34595490e-01
7.80864060e-01 -1.65611374e+00 -1.03513455e+00 7.06164181e-01
-2.85216510e-01 1.08192086e-01 4.59497452e-01 2.72529066e-01
-3.04045647e-01 -2.74213761e-01 -4.61737305e-01 -3.31967175e-01
-9.90878761e-01 6.06237590e-01 3.87742609e-01 -2.09307998e-01
-5.57448924e-01 9.21489239e-01 5.06758213e-01 -1.58856705e-01
-1.54182658e-01 2.27306992e-01 -1.09224066e-01 -1.94238991e-01
8.69132757e-01 1.18005648e-01 -5.18861301e-02 -5.42322516e-01
-1.10884480e-01 6.22459292e-01 -9.05466266e-03 -3.31073292e-02
1.46227670e+00 -7.32415244e-02 -5.72954677e-02 -2.43385911e-01
1.12364399e+00 2.53737420e-01 -1.72226965e+00 -8.80434737e-02
-8.76929224e-01 -9.01015937e-01 1.49978817e-01 -8.01523924e-01
-1.78369939e+00 6.16683424e-01 6.06645048e-01 -1.64657444e-01
1.53651524e+00 -1.02336898e-01 8.76935005e-01 -3.57106060e-01
4.63400006e-01 -4.53002959e-01 4.05174553e-01 2.68556550e-02
1.33391964e+00 -9.02797163e-01 -2.46521622e-01 -5.64288080e-01
-7.03822017e-01 7.87211001e-01 6.16706908e-01 -4.30267930e-01
5.27136624e-01 4.54598188e-01 -1.62094578e-01 -4.15186174e-02
-7.28400648e-01 6.88342154e-02 2.49816149e-01 6.57925963e-01
1.91476107e-01 1.38581339e-02 -1.08865947e-01 7.11686850e-01
-5.80203235e-01 -5.59888184e-02 3.20462108e-01 2.25664794e-01
-8.04496035e-02 -1.24245727e+00 -4.09990400e-01 3.46476644e-01
-5.84247947e-01 -1.81733668e-01 6.44701440e-03 3.96956652e-01
3.09260875e-01 8.09408724e-01 -2.49506515e-02 -4.55730528e-01
2.07418233e-01 -2.77208716e-01 8.92582953e-01 -5.36777556e-01
-4.78282690e-01 1.47452354e-01 -7.54057243e-02 -6.50680423e-01
-2.06637010e-01 -4.77795631e-01 -9.32361186e-01 -5.16246676e-01
-1.06389910e-01 -5.87400943e-02 2.76380420e-01 6.12057209e-01
5.95760763e-01 6.52682900e-01 1.10370219e+00 -9.95639980e-01
-5.69661915e-01 -1.04352880e+00 -5.70248127e-01 3.12987268e-01
3.47562402e-01 -6.63521945e-01 -4.86890972e-01 9.59380269e-02] | [12.552632331848145, -0.2348136156797409] |
5eccdd01-1e0c-4baf-b31f-24a15d79da5c | exploiting-topic-information-for-joint-intent | null | null | https://openreview.net/forum?id=YXvbGWz1AGP | https://openreview.net/pdf?id=YXvbGWz1AGP | Exploiting Topic Information for Joint Intent Detection and Slot Filling | Intent detection and slot filling are two important basic tasks in natural language understanding. Actually, there are multiple intents in an utterance. How to map different intents to corresponding slot becomes a new challenge for recent research. Existing models solve this problem by using neural layers to adaptively capture related intent information for each slot, which the process of intent selection is not clear enough. It is observed that there is strong consistency between intents and topics of a sentence, thus we exploit topic information for joint intent detection and slot filling via a topic fusion mechanism, where token-level topic information take the place of intent information to guide slot prediction. In addition, sentence-level topic information is also utilized to enhance the intent detection. Experiment results show explicit improvements on two public datasets, where provide 4.8% improvement in sentence accuracy on MixATIS and 0.7% improvement in intent detection on MixSNIPS. | ['Anonymous'] | 2022-01-16 | null | null | null | acl-arr-january-2022-1 | ['slot-filling'] | ['natural-language-processing'] | [ 3.73931915e-01 3.38019788e-01 -6.09791577e-01 -8.78827512e-01
-6.07201993e-01 -2.03043923e-01 5.62913895e-01 4.56719339e-01
-5.49643755e-01 7.16437340e-01 7.05554962e-01 -2.11524114e-01
1.23858944e-01 -7.49770701e-01 -3.13745201e-01 -2.78066069e-01
1.44789740e-01 4.65364397e-01 3.07557911e-01 -5.60928106e-01
3.80505353e-01 -5.13124049e-01 -1.25623202e+00 6.55473113e-01
1.08906877e+00 1.05351555e+00 6.93496108e-01 3.81236792e-01
-1.01170933e+00 7.52260983e-01 -7.98491001e-01 8.96080583e-02
-1.69845864e-01 -1.76023498e-01 -1.01162159e+00 9.84738991e-02
-1.38668463e-01 -2.83657640e-01 6.29689768e-02 8.95615697e-01
1.90814108e-01 1.90303206e-01 3.92713726e-01 -1.23524010e+00
-1.53905332e-01 1.29227471e+00 -6.82937682e-01 2.98273802e-01
4.80265826e-01 -3.84317935e-01 1.23428786e+00 -6.88990712e-01
3.59410167e-01 1.36197412e+00 2.58092582e-01 3.67046326e-01
-7.21404374e-01 -6.58841014e-01 6.47535026e-01 1.53223962e-01
-1.16208899e+00 -3.15427780e-01 9.86699879e-01 -2.07515866e-01
9.60800648e-01 3.21130782e-01 4.07897115e-01 7.53910124e-01
3.05221379e-01 1.37137055e+00 7.82620609e-01 -4.26046669e-01
3.56799036e-01 3.77115816e-01 9.43969965e-01 2.87424654e-01
-1.50407523e-01 -6.51819885e-01 -8.36287022e-01 -1.93071395e-01
3.64183068e-01 2.86496550e-01 -4.07024994e-02 2.04661965e-01
-1.04716218e+00 1.15892851e+00 3.35633636e-01 2.67163306e-01
-5.47915816e-01 -3.53546530e-01 7.29781926e-01 2.54461139e-01
7.28828549e-01 4.33791310e-01 -7.21573651e-01 -1.95930019e-01
-7.48705268e-01 4.64721501e-01 7.68209457e-01 8.57459962e-01
1.08019531e+00 -3.48331451e-01 -6.04881465e-01 1.02630913e+00
4.94944543e-01 5.04070185e-02 8.91437173e-01 -5.99681437e-01
7.38816917e-01 1.10013211e+00 6.38208166e-02 -9.53962743e-01
-6.47994518e-01 -4.21357900e-01 -6.77774787e-01 -6.05493605e-01
-2.10121349e-01 -2.83681810e-01 -1.14928031e+00 1.85131907e+00
3.29989552e-01 1.76651374e-01 2.37012073e-01 8.04860055e-01
1.01562452e+00 9.06126261e-01 5.83841383e-01 -3.98873925e-01
2.02278543e+00 -8.96771193e-01 -1.10225165e+00 -7.72920549e-01
6.30062640e-01 -7.66896963e-01 8.88302267e-01 -1.46837970e-02
-6.57970190e-01 -4.92135912e-01 -9.59686100e-01 -1.33246213e-01
-3.24425995e-01 1.92306429e-01 1.15642536e+00 3.85878325e-01
-6.97295785e-01 -2.36951292e-01 -6.43510640e-01 -2.46699676e-01
1.48145586e-01 2.74718910e-01 1.72631778e-02 1.22612000e-01
-1.83024037e+00 3.19422096e-01 9.71547186e-01 -3.27604324e-01
-2.00283602e-01 -7.21906304e-01 -1.09479499e+00 1.47112429e-01
6.71584785e-01 -4.58501607e-01 1.61986530e+00 -4.36738998e-01
-1.20170081e+00 5.98780572e-01 -7.01758444e-01 -8.62070680e-01
-1.21512376e-01 -4.36967045e-01 -5.00040650e-01 -6.75489753e-02
7.54327595e-01 1.05238187e+00 7.03366280e-01 -7.83913314e-01
-1.32183397e+00 -2.09167480e-01 3.10058028e-01 8.22729886e-01
-3.96774918e-01 1.06818534e-01 -5.83758950e-01 -4.21015918e-01
6.60823166e-01 -4.68847454e-01 -2.71287322e-01 -5.49153864e-01
-6.47160172e-01 -9.12935853e-01 1.15778375e+00 -6.02025211e-01
1.51647544e+00 -1.93978465e+00 -3.61522317e-01 -3.68799716e-01
1.41276389e-01 -1.09001495e-01 2.99147308e-01 2.79139966e-01
1.18407778e-01 -2.18225494e-02 -2.08997041e-01 -5.77300251e-01
1.13392621e-01 2.06036776e-01 -9.16594446e-01 -2.21639946e-01
6.72882795e-02 7.27416933e-01 -6.64219201e-01 -6.14889979e-01
1.35052472e-01 9.24162716e-02 -5.65841913e-01 2.74371207e-01
-6.05339110e-01 2.14629501e-01 -8.15594912e-01 5.12458444e-01
6.81383073e-01 -4.24485594e-01 5.49430549e-02 -1.34650975e-01
-2.63832305e-02 1.23146248e+00 -1.16709292e+00 2.05315590e+00
-4.42068189e-01 4.62535679e-01 -8.28105733e-02 -1.15751898e+00
8.60133827e-01 5.25622487e-01 4.87051696e-01 -7.21873939e-01
-7.08079711e-02 -1.17695797e-03 -1.71749637e-01 -9.71305594e-02
1.11202908e+00 -2.03972489e-01 -7.45910645e-01 7.71050990e-01
4.95780669e-02 8.63899589e-02 3.85659158e-01 3.93500149e-01
8.27188075e-01 -4.26432997e-01 4.85539675e-01 -3.08825344e-01
4.80431020e-01 1.79100379e-01 8.16550434e-01 8.89003694e-01
-1.15446270e-01 3.38997513e-01 5.15422583e-01 -4.98759776e-01
-5.17386317e-01 -6.70057476e-01 -2.25764409e-01 1.54666424e+00
3.41328919e-01 -4.84924227e-01 -5.32625914e-01 -7.51591444e-01
-3.33127290e-01 9.14550602e-01 -6.18009031e-01 -4.24865931e-02
-6.02476120e-01 -8.38966429e-01 1.07130542e-01 3.46979499e-01
8.94617677e-01 -1.23869872e+00 -5.81330419e-01 5.37899256e-01
-7.04377294e-01 -1.14303911e+00 -4.01633501e-01 5.71428478e-01
-6.55746639e-01 -5.37837684e-01 -3.76282930e-01 -6.85540974e-01
4.46633011e-01 5.02405107e-01 1.25259805e+00 -8.89597759e-02
9.02498662e-02 -1.33160055e-01 -4.92968887e-01 -5.57663977e-01
-2.22715661e-01 5.73681891e-01 2.38705687e-02 -2.42552578e-01
8.40827882e-01 -2.37365261e-01 -6.31285131e-01 3.60358566e-01
-9.56237018e-01 4.78946924e-01 4.13273752e-01 8.62603486e-01
4.34515104e-02 3.25480670e-01 7.13009298e-01 -9.44607317e-01
8.09018135e-01 -7.64431834e-01 -1.68169439e-01 8.72079656e-02
-4.48280364e-01 -4.45002578e-02 2.73422062e-01 -7.89946169e-02
-1.41695201e+00 -1.79591686e-01 -3.97182077e-01 4.21259642e-01
-6.08091176e-01 7.49066174e-01 -1.00117058e-01 7.40707815e-01
1.89170718e-01 3.89600217e-01 -2.65459895e-01 -4.20179218e-01
1.05496183e-01 9.36638117e-01 1.16344303e-01 -3.89670938e-01
1.91782162e-01 3.52058321e-01 -7.69639432e-01 -9.15540993e-01
-1.33447146e+00 -1.03318512e+00 -1.96408272e-01 2.45527148e-01
7.97834933e-01 -1.24860334e+00 -3.88720721e-01 3.03943902e-01
-1.30623138e+00 6.54640794e-02 -1.39862731e-01 4.27136421e-01
-1.69799834e-01 2.97874928e-01 -7.19750404e-01 -9.09864724e-01
-5.45220375e-01 -1.11225069e+00 1.45184803e+00 6.70097232e-01
-5.64039588e-01 -1.04196608e+00 -1.64637998e-01 5.24469852e-01
3.13549191e-01 -6.98436141e-01 7.87677169e-01 -1.15829659e+00
-4.59638983e-01 -8.21491927e-02 -2.77039200e-01 -2.20710039e-01
4.25836563e-01 -6.93023205e-01 -9.91645515e-01 -1.05232680e-02
3.43480021e-01 -3.21354240e-01 1.03963649e+00 5.56967914e-01
7.64566600e-01 -3.25107932e-01 -6.44076824e-01 8.18546489e-02
1.04676759e+00 5.28365076e-01 4.06250477e-01 2.54564136e-01
4.40244645e-01 6.28193736e-01 1.04581773e+00 5.11025965e-01
7.77626693e-01 8.53026688e-01 1.07987016e-03 -8.39691162e-02
2.71364450e-01 -2.64311552e-01 2.16429785e-01 7.39277005e-01
7.80013084e-01 -4.47632223e-01 -7.74473250e-01 6.04099333e-01
-2.04669976e+00 -7.21578002e-01 1.88238993e-01 1.69402874e+00
1.03969073e+00 5.23557603e-01 -1.23779483e-01 1.23583093e-01
7.69445121e-01 3.87269318e-01 -4.75722581e-01 -1.51653677e-01
1.09534480e-01 -3.92363101e-01 4.82623167e-02 4.91853535e-01
-1.51030183e+00 1.24854279e+00 6.16009712e+00 1.01011348e+00
-1.10475349e+00 -5.66205382e-02 1.05422401e+00 4.94555026e-01
-4.04538751e-01 3.44775021e-02 -1.30834401e+00 5.72266698e-01
5.50746024e-01 -1.42814040e-01 -3.70705336e-01 1.09328651e+00
2.39988014e-01 -5.14174819e-01 -5.47500193e-01 6.87262177e-01
3.82236880e-03 -1.37042916e+00 -1.67850271e-01 -2.19339028e-01
6.10193908e-01 -1.55722186e-01 -2.89738420e-02 8.62403750e-01
3.42246771e-01 -6.33850574e-01 3.34291160e-01 1.27707869e-01
1.29320830e-01 -6.41781390e-01 8.75428259e-01 6.36669338e-01
-1.27519274e+00 2.12285072e-02 -5.03494799e-01 -4.78132546e-01
5.68259418e-01 8.87251437e-01 -1.41226375e+00 4.08403605e-01
5.28428674e-01 5.55593491e-01 -1.86890811e-01 7.62207866e-01
-1.24205500e-01 8.70068610e-01 -5.47396660e-01 -1.29252106e-01
3.47656131e-01 6.77125826e-02 4.90799874e-01 1.06566858e+00
-4.46058698e-02 -3.93914105e-03 7.18996167e-01 7.29872048e-01
7.07573369e-02 2.52277613e-01 -2.48862922e-01 2.79887207e-02
4.98927325e-01 1.27251661e+00 -1.16747403e+00 -5.85293174e-01
-3.56525481e-01 9.88770545e-01 4.10134532e-02 1.66324750e-01
-6.68624640e-01 -2.08329886e-01 8.63898993e-01 -3.29514086e-01
6.73692375e-02 -6.28399402e-02 -7.53093600e-01 -1.04753745e+00
-5.48032485e-02 -3.03053021e-01 5.55455089e-01 -5.00286043e-01
-8.78113389e-01 3.54910612e-01 2.16545627e-01 -9.25554037e-01
-5.52616358e-01 -2.05760106e-01 -8.94947290e-01 9.00980830e-01
-1.41686225e+00 -8.53733361e-01 -3.94369513e-02 -1.17922565e-02
1.37133253e+00 -1.97121128e-01 7.80731618e-01 6.58281446e-02
-4.04939711e-01 2.98263729e-01 -4.75086153e-01 1.41205713e-01
4.26135361e-01 -1.29855597e+00 7.43945360e-01 6.85450375e-01
8.30563530e-02 8.55894506e-01 9.54935133e-01 -8.78603101e-01
-8.33612859e-01 -8.37283731e-01 1.39119554e+00 -1.08513057e-01
3.19684356e-01 -4.85093981e-01 -1.08407176e+00 5.47671556e-01
4.02852505e-01 -6.49204969e-01 8.09914112e-01 6.11844838e-01
3.07581499e-02 2.80464351e-01 -9.13849235e-01 6.37642443e-01
6.23314261e-01 -4.98508602e-01 -1.05176032e+00 4.30611253e-01
1.60275054e+00 -5.72707653e-01 -3.33262295e-01 3.93998414e-01
3.11406180e-02 -8.32632482e-01 8.50555241e-01 -3.11808795e-01
2.58957297e-01 -3.81148160e-01 -8.38016570e-02 -1.02413297e+00
-1.04810307e-02 -4.36385423e-01 -6.08784780e-02 1.33426154e+00
5.83630323e-01 -4.11579758e-01 1.08011627e+00 9.05833542e-01
-2.95392334e-01 -7.04674661e-01 -8.12420905e-01 -7.07771108e-02
-2.87842035e-01 -7.43452907e-01 8.76644015e-01 8.59647274e-01
5.52552640e-01 9.98900771e-01 -2.96461135e-01 -3.43717122e-03
2.20756739e-01 4.58554596e-01 5.03586471e-01 -1.12427926e+00
-2.17597298e-02 -2.17530757e-01 -8.48451704e-02 -1.74185777e+00
2.56429344e-01 -4.63597685e-01 3.80340874e-01 -1.70097423e+00
2.81591743e-01 -6.69759333e-01 -2.95407385e-01 7.83496022e-01
-6.82385683e-01 -1.77362308e-01 -8.37805420e-02 1.22625165e-01
-1.10726428e+00 7.69521892e-01 7.50655532e-01 -3.22734088e-01
-4.77527887e-01 3.02582324e-01 -9.75873053e-01 7.49332666e-01
1.04055500e+00 -3.54822218e-01 -7.73958921e-01 -1.24816872e-01
1.04091957e-01 2.45871186e-01 -1.63790971e-01 -9.35434401e-01
4.22325104e-01 -2.58764625e-01 -3.27592785e-03 -1.19621897e+00
5.03167450e-01 -5.00140607e-01 -4.83559906e-01 3.40800345e-01
-7.08133698e-01 -3.72405320e-01 2.27686986e-01 5.33902228e-01
-4.95801061e-01 -3.14191878e-01 2.06192508e-01 -1.81172952e-01
-1.37245393e+00 1.61071658e-01 -5.22082388e-01 -1.23222759e-02
6.23694599e-01 -4.50616851e-02 -2.04830617e-01 -5.60993075e-01
-4.15000290e-01 5.04059792e-01 -1.00415908e-01 8.08432639e-01
5.10746717e-01 -1.17340267e+00 -4.79949266e-01 3.86061490e-01
2.21095115e-01 2.01384112e-01 6.34509921e-01 4.65317249e-01
2.29797617e-01 9.69201505e-01 1.54883832e-01 -7.85131693e-01
-1.06157768e+00 1.09222673e-01 -1.84762314e-01 -6.37448072e-01
-4.35987860e-01 1.14960563e+00 7.00306535e-01 -6.26636744e-01
2.59444088e-01 -4.26851124e-01 -6.90254927e-01 9.99106169e-02
8.73757064e-01 -2.76218206e-01 -5.24689490e-03 -5.37494779e-01
-2.96168178e-01 -3.52790987e-04 -7.76801169e-01 -2.16901585e-01
9.05806720e-01 -5.98752856e-01 -9.03073400e-02 5.24632037e-01
9.72197115e-01 -3.81222636e-01 -1.04678178e+00 -5.93837857e-01
2.33808324e-01 -7.72019029e-02 4.36205454e-02 -8.13549578e-01
-5.33088446e-01 5.80436349e-01 3.90156448e-01 6.17971838e-01
9.94532466e-01 9.65344310e-02 1.10336030e+00 3.66602093e-01
3.25900257e-01 -1.28161991e+00 -4.79793064e-02 1.01993477e+00
6.67524636e-01 -1.65732634e+00 -2.02489585e-01 -6.23265743e-01
-8.36666942e-01 7.83768475e-01 8.79292369e-01 3.51206690e-01
6.30477846e-01 1.44707739e-01 2.13277623e-01 -2.27367297e-01
-9.84432995e-01 -1.09261252e-01 3.12501013e-01 1.95343688e-01
7.86138296e-01 2.79445618e-01 -5.52463770e-01 9.11582410e-01
-3.51646096e-01 -3.45003635e-01 1.63112551e-01 1.10278022e+00
-1.08579826e+00 -1.10299695e+00 -3.22094083e-01 7.30332434e-01
-4.44889039e-01 -2.60951728e-01 9.34587270e-02 2.38012835e-01
-7.19540864e-02 1.16293931e+00 3.78050923e-01 -4.26257312e-01
-2.03033149e-01 4.12970960e-01 -5.99329829e-01 -1.06571913e+00
-3.00234079e-01 1.60532638e-01 1.57125279e-01 -3.89108717e-01
-1.99890763e-01 -6.07007980e-01 -1.61474073e+00 1.90669358e-01
-3.79466623e-01 6.74378991e-01 7.45856524e-01 1.44491982e+00
4.39619452e-01 6.44579232e-01 4.62004185e-01 -3.33221912e-01
-1.40707299e-01 -1.18268275e+00 -5.17781138e-01 8.76573995e-02
3.27449679e-01 -6.65631115e-01 -1.27102926e-01 -1.51507780e-01] | [12.569488525390625, 7.349836826324463] |
17bee82c-ac31-4a5a-bdcf-fd506d250187 | multiple-instance-neuroimage-transformer | 2208.09567 | null | https://arxiv.org/abs/2208.09567v1 | https://arxiv.org/pdf/2208.09567v1.pdf | Multiple Instance Neuroimage Transformer | For the first time, we propose using a multiple instance learning based convolution-free transformer model, called Multiple Instance Neuroimage Transformer (MINiT), for the classification of T1weighted (T1w) MRIs. We first present several variants of transformer models adopted for neuroimages. These models extract non-overlapping 3D blocks from the input volume and perform multi-headed self-attention on a sequence of their linear projections. MINiT, on the other hand, treats each of the non-overlapping 3D blocks of the input MRI as its own instance, splitting it further into non-overlapping 3D patches, on which multi-headed self-attention is computed. As a proof-of-concept, we evaluate the efficacy of our model by training it to identify sex from T1w-MRIs of two public datasets: Adolescent Brain Cognitive Development (ABCD) and the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA). The learned attention maps highlight voxels contributing to identifying sex differences in brain morphometry. The code is available at https://github.com/singlaayush/MINIT. | ['Ehsan Adeli', 'Kilian M. Pohl', 'Yuyin Zhou', 'Daniel K. Do', 'Qingyu Zhao', 'Ayush Singla'] | 2022-08-19 | null | null | null | null | ['brain-morphometry'] | ['medical'] | [ 1.95874542e-01 5.05100787e-01 -6.60137683e-02 -7.46229291e-01
-7.01463521e-01 -3.20322245e-01 6.37171209e-01 1.93584785e-01
-4.12228286e-01 3.41149181e-01 2.11537004e-01 -2.77070403e-01
-3.58765900e-01 -6.95938349e-01 -1.15468347e+00 -4.13687527e-01
-5.23866832e-01 8.77532601e-01 3.28397937e-02 1.50542825e-01
2.89507508e-01 2.66165197e-01 -1.33198059e+00 5.21315694e-01
1.13545108e+00 8.85129988e-01 2.42014378e-02 3.17448020e-01
-8.03891271e-02 3.68489534e-01 -4.73173382e-03 -7.89051056e-01
1.44058555e-01 -2.54410237e-01 -1.08636189e+00 -1.42077744e-01
9.51838493e-01 -4.30696249e-01 -1.34853348e-02 1.06875157e+00
5.85780263e-01 -1.06650367e-01 8.40223551e-01 -1.19709206e+00
-7.75221825e-01 9.43419158e-01 -8.36284399e-01 7.43615389e-01
-5.89120127e-02 4.78719585e-02 8.11602354e-01 -1.10224867e+00
7.30236411e-01 1.30526078e+00 8.64067137e-01 5.21400630e-01
-1.30900788e+00 -1.10761988e+00 7.50019997e-02 5.69114029e-01
-9.53521371e-01 -5.71609378e-01 3.68775457e-01 -6.00719452e-01
9.86763358e-01 -4.88780104e-02 9.72029507e-01 9.57920492e-01
3.43771964e-01 7.72628307e-01 1.50497079e+00 -8.37105513e-02
-2.86343531e-03 -4.10784513e-01 1.58802658e-01 6.40667200e-01
3.53103466e-02 -6.91429824e-02 -3.16055745e-01 -9.40458104e-02
6.81178987e-01 -1.14687026e-01 1.88872725e-01 -3.58400941e-01
-8.58792603e-01 7.21956193e-01 4.87645596e-01 4.13732380e-01
-5.19518256e-01 5.87132806e-03 3.46346349e-01 5.39401285e-02
9.50580835e-01 -9.37599763e-02 -2.81573921e-01 1.55874953e-01
-1.25264835e+00 1.80551201e-01 7.32565597e-02 3.75136167e-01
5.56111753e-01 -1.92925543e-01 -1.85338363e-01 1.04974806e+00
1.99266478e-01 2.99345881e-01 7.45269239e-01 -6.99258685e-01
4.37559932e-01 5.09541929e-01 -8.51994216e-01 -4.56752867e-01
-6.17765129e-01 -2.47385144e-01 -4.44179773e-01 1.85870007e-01
3.49024802e-01 3.18947136e-02 -1.11472082e+00 1.83164096e+00
5.20085156e-01 2.57526100e-01 -5.82686663e-01 8.00329626e-01
1.02023149e+00 -4.44948077e-02 3.91435772e-01 9.87788364e-02
1.43947101e+00 -7.03886807e-01 -6.38974085e-02 -1.05122782e-01
1.87862143e-01 -2.27819785e-01 3.90578508e-01 2.34240085e-01
-1.50183702e+00 -2.58479089e-01 -6.55876338e-01 -2.32461646e-01
-5.95303059e-01 -1.29413918e-01 4.99236226e-01 6.41570508e-01
-1.17671704e+00 6.94253683e-01 -1.16570210e+00 -3.66561323e-01
1.05383873e+00 5.12671351e-01 -7.24342108e-01 -1.54360965e-01
-1.00578856e+00 1.04218125e+00 -1.49405032e-01 -2.03100175e-01
-1.11509502e+00 -1.37999129e+00 -8.23099792e-01 2.08741501e-02
5.19248135e-02 -2.52247155e-01 1.01362622e+00 -1.01952696e+00
-8.51021767e-01 1.37814975e+00 -1.45568043e-01 -3.79532725e-01
2.36153528e-01 9.77708865e-03 -2.40187943e-01 2.94845313e-01
4.33261424e-01 8.73251617e-01 8.44308436e-01 -7.54321396e-01
-5.03747612e-02 -1.16285908e+00 -8.96494016e-02 -1.19446898e-02
-6.12180866e-03 4.58721489e-01 -7.04933777e-02 -5.10377169e-01
2.26726547e-01 -7.29367733e-01 -3.98297980e-02 6.22603949e-03
-5.15394449e-01 -1.28268480e-01 2.54882067e-01 -1.21740949e+00
5.88981569e-01 -1.84976101e+00 2.19062775e-01 1.33735135e-01
6.85989499e-01 2.33711712e-02 -2.85130084e-01 -7.55785480e-02
-8.28051627e-01 3.01803857e-01 -5.43954432e-01 -4.18381065e-01
-9.69624594e-02 -1.47812486e-01 3.23143542e-01 7.18798518e-01
4.35980082e-01 1.03807962e+00 -8.33121061e-01 -3.83589476e-01
-1.39972627e-01 5.95828950e-01 -6.58981919e-01 -9.00004730e-02
2.40503594e-01 6.03835583e-01 -2.07969606e-01 5.31418979e-01
9.14040148e-01 1.68263048e-01 1.94238275e-01 -1.94278285e-01
-3.84925723e-01 3.28470409e-01 -4.96985525e-01 1.50507855e+00
-1.47749171e-01 4.05348957e-01 2.25588024e-01 -1.29116023e+00
5.41097760e-01 2.86835492e-01 7.03974307e-01 -1.30389273e+00
2.87063092e-01 2.25796670e-01 5.40559709e-01 -5.51682472e-01
-1.97787806e-01 -2.07587942e-01 5.13786256e-01 8.06075335e-01
6.70516431e-01 2.37931877e-01 2.54863411e-01 2.35051587e-02
9.31604981e-01 2.57789850e-01 2.16874048e-01 -5.52146792e-01
2.82784373e-01 -4.91014242e-01 3.99940401e-01 2.64307201e-01
-1.70231670e-01 7.18772650e-01 7.34197795e-01 -5.10814786e-01
-1.14971030e+00 -1.05387378e+00 -6.05731308e-01 1.14154387e+00
-7.52630472e-01 -1.15368448e-01 -1.10665452e+00 -6.89008772e-01
1.91942006e-01 6.12950861e-01 -1.27480447e+00 -6.85961470e-02
-4.81865406e-01 -1.18808126e+00 5.63376665e-01 4.52794909e-01
2.42872313e-01 -8.94601822e-01 -5.88544607e-01 -6.93864450e-02
1.10459141e-01 -7.24096358e-01 -4.91479725e-01 3.11253726e-01
-8.76227915e-01 -1.39297664e+00 -1.11247456e+00 -4.06722963e-01
7.79816806e-01 -3.59539956e-01 1.19746268e+00 -3.47421914e-02
-5.42206049e-01 5.04857481e-01 -5.80767281e-02 -5.22116363e-01
-8.53407681e-02 1.53236300e-01 1.11508025e-02 1.29874796e-01
2.53284782e-01 -1.11707044e+00 -8.61977994e-01 2.80959960e-02
-6.33092582e-01 -4.84072277e-03 4.86308336e-01 5.65204620e-01
7.02253163e-01 -5.48158944e-01 4.49620038e-01 -8.79695952e-01
3.41910660e-01 -1.12337029e+00 -3.00446033e-01 2.32188359e-01
-5.65746367e-01 -4.33189794e-02 9.74084064e-02 -4.96736169e-01
-6.95866585e-01 -3.06575924e-01 -3.01464617e-01 -3.80846918e-01
-1.95820481e-01 2.27292791e-01 5.01195937e-02 -6.51339293e-02
3.40910733e-01 2.09359974e-01 2.15074003e-01 -7.51233399e-01
7.67953917e-02 2.72579074e-01 3.57463717e-01 -6.67558372e-01
3.43124509e-01 2.79748261e-01 -3.40598486e-02 -5.36947608e-01
-5.44379532e-01 4.62045781e-02 -1.06273484e+00 -2.76142210e-01
1.14219177e+00 -4.82303858e-01 -5.57078540e-01 6.53580010e-01
-6.16443336e-01 -6.26277149e-01 1.50230095e-01 3.59973758e-01
-3.35955918e-01 2.43591443e-02 -3.73832226e-01 -5.89022517e-01
-7.80480981e-01 -1.27334809e+00 1.07256758e+00 6.17491193e-02
-3.29938978e-02 -9.03466225e-01 3.38797301e-01 5.09360194e-01
5.32243073e-01 1.73059955e-01 1.49391663e+00 -1.04180527e+00
-3.41570117e-02 2.05258757e-01 -3.53489697e-01 1.22015208e-01
-3.29853803e-01 -2.65465617e-01 -9.18350518e-01 2.14691758e-02
-1.33631334e-01 -2.51670063e-01 1.05590963e+00 6.34972930e-01
1.33034182e+00 -4.71383743e-02 -7.01111704e-02 6.41195118e-01
1.01781833e+00 9.57734957e-02 5.54624259e-01 1.78498507e-01
5.74374557e-01 7.28944778e-01 -1.20437957e-01 2.14017987e-01
8.88747990e-01 6.19952857e-01 5.56119919e-01 -3.14811558e-01
-2.00741664e-01 2.54608714e-03 4.34531830e-02 3.77757013e-01
-4.62787569e-01 5.81012428e-01 -1.22167325e+00 8.27314019e-01
-1.10009897e+00 -1.06304646e+00 -1.49161622e-01 2.24122167e+00
8.53074074e-01 -1.33966997e-01 5.01984239e-01 -7.72171691e-02
7.31838465e-01 1.03394985e-02 -8.58066261e-01 -6.81594491e-01
1.92564398e-01 9.78830576e-01 2.79716879e-01 1.40097409e-01
-8.27323318e-01 4.26479220e-01 5.89112663e+00 5.17112255e-01
-1.23949289e+00 6.26694202e-01 1.15917361e+00 -4.62052882e-01
-3.01222086e-01 -2.96769977e-01 -2.79761136e-01 5.23184180e-01
1.13026011e+00 -1.11740008e-01 7.43155479e-01 5.94167054e-01
5.97487949e-02 -1.29204422e-01 -1.00738692e+00 3.52833420e-01
1.76708996e-01 -8.98775518e-01 -2.64556706e-01 1.34373650e-01
3.82777274e-01 4.35338646e-01 2.95270503e-01 3.94161880e-01
-1.21879466e-01 -1.38502741e+00 7.80690730e-01 5.15918195e-01
1.09465337e+00 -7.26447225e-01 3.03531051e-01 -6.81153983e-02
-9.76621151e-01 -3.45246121e-02 -2.71109287e-02 1.81614324e-01
-1.90251634e-01 5.87430418e-01 -8.27006757e-01 3.20740551e-01
1.01065636e+00 5.13540328e-01 -8.55234027e-01 1.13896644e+00
1.59635067e-01 6.04150832e-01 -1.86305791e-01 5.00969291e-01
1.70959607e-01 -1.41740710e-01 3.86754751e-01 1.04011905e+00
4.18109059e-01 2.69213468e-01 -2.38652349e-01 1.22107267e+00
-1.94334641e-01 2.80529946e-01 -2.23808929e-01 3.24446969e-02
1.36881828e-01 1.39845228e+00 -7.77677238e-01 -2.26710737e-01
-4.88108546e-01 5.10110021e-01 5.63422263e-01 4.57470082e-02
-6.95731163e-01 6.29236223e-03 5.57172537e-01 5.22321701e-01
5.37283897e-01 1.38382047e-01 -1.43017784e-01 -8.87188256e-01
-2.29273483e-01 -8.62966776e-01 3.64883602e-01 -7.33262479e-01
-1.29314697e+00 4.70524728e-01 2.98142821e-01 -4.38552499e-01
-1.69685725e-02 -4.81601387e-01 -9.47065830e-01 9.82919276e-01
-1.38333595e+00 -1.41630352e+00 3.09973937e-02 5.56599796e-01
3.38202626e-01 -1.74501408e-02 7.44646549e-01 5.60385644e-01
-8.08090985e-01 5.63744903e-01 -2.04431370e-01 2.84283221e-01
4.78196234e-01 -1.42289805e+00 4.46409851e-01 4.97146547e-01
-1.70711935e-01 7.79199839e-01 2.53981769e-01 -8.17382991e-01
-9.88373041e-01 -9.77447391e-01 8.61931384e-01 -2.46398121e-01
6.78079247e-01 -4.31472063e-01 -1.01449287e+00 8.89539123e-01
2.54790753e-01 -6.19864371e-03 7.99051046e-01 2.81222135e-01
-5.20893574e-01 1.15853474e-01 -1.42938077e+00 2.20987156e-01
1.17705870e+00 -3.45104337e-01 -6.77041352e-01 3.02681357e-01
1.70188710e-01 -4.45956886e-01 -1.35709512e+00 2.76186079e-01
7.95406580e-01 -1.15161395e+00 1.08856869e+00 -7.16549814e-01
1.02118587e+00 4.31044638e-01 1.81612790e-01 -1.34292972e+00
-2.96740443e-01 -2.60431841e-02 9.79145840e-02 1.14803898e+00
4.31610763e-01 -5.55061281e-01 5.46935678e-01 7.97736049e-01
-1.94553539e-01 -1.20351315e+00 -1.31914067e+00 -3.67369086e-01
7.97961533e-01 -4.81857032e-01 9.58644092e-01 9.01733279e-01
-6.33912487e-03 -5.82325086e-03 6.80216998e-02 -6.63947910e-02
8.85602593e-01 1.03832707e-02 -5.35280332e-02 -1.18637276e+00
-1.52744967e-02 -5.85950494e-01 -4.08814043e-01 -3.27223465e-02
4.12685424e-01 -1.43346310e+00 -4.77795303e-01 -1.39864719e+00
8.26237082e-01 -2.22920284e-01 -4.02490377e-01 1.05251420e+00
2.83374563e-02 4.80616361e-01 -5.43167405e-02 -3.63133818e-01
-3.45037431e-01 4.21996176e-01 1.27791846e+00 -2.48148948e-01
3.27943385e-01 -6.39369041e-02 -7.17082620e-01 5.74441135e-01
7.47039318e-01 -6.03477240e-01 -2.87712872e-01 -5.07165015e-01
-2.09394336e-01 -4.74001095e-02 5.01936316e-01 -8.38699996e-01
-1.12440698e-01 7.80825689e-02 8.64694238e-01 -4.95962828e-01
1.79958910e-01 -4.28881973e-01 5.18443249e-02 3.71921271e-01
-2.67561197e-01 2.14145824e-01 1.29900828e-01 -2.48165354e-01
2.57122576e-01 -2.54678309e-01 8.21476400e-01 -4.23798501e-01
-1.22814395e-01 7.57236421e-01 -3.29897016e-01 3.07649642e-01
5.97106814e-01 -1.38320297e-01 -3.49209577e-01 1.83043294e-02
-9.30475295e-01 1.99127585e-01 1.46049917e-01 3.55800152e-01
5.17014205e-01 -1.19027901e+00 -8.41057479e-01 3.95873547e-01
-1.75007194e-01 -3.82178277e-01 6.81039095e-01 1.54772437e+00
9.75174531e-02 3.51899534e-01 -8.58779311e-01 -4.04058039e-01
-1.33638096e+00 3.03865314e-01 6.34125769e-01 -1.97881997e-01
-6.13768637e-01 7.99322724e-01 3.54717880e-01 -7.12195098e-01
-1.39995754e-01 -3.46892923e-01 -5.32834947e-01 3.46388042e-01
6.97852135e-01 4.81522232e-01 3.53259712e-01 -8.99416745e-01
-3.72429281e-01 4.76785034e-01 -2.20701709e-01 -1.41526401e-01
2.09640169e+00 1.03312559e-01 -5.65442443e-01 1.09004267e-01
1.20100498e+00 -1.30825341e-01 -8.64196002e-01 -2.47874837e-02
-1.12504616e-01 -2.96662897e-01 1.98028520e-01 -8.44143152e-01
-1.62069809e+00 8.40408385e-01 8.59554827e-01 -9.42606404e-02
1.01195920e+00 1.84343591e-01 6.17786944e-01 -3.62256646e-01
1.28571749e-01 -6.85612798e-01 -1.87976226e-01 4.28369522e-01
1.07003999e+00 -8.69437158e-01 2.96069600e-04 2.63700206e-02
-4.68894094e-01 9.85766053e-01 6.73607707e-01 -1.71172544e-01
7.10398972e-01 1.32172689e-01 -2.53343940e-01 -7.05032408e-01
-5.43341875e-01 -2.66935140e-01 6.45281553e-01 7.95946836e-01
6.69393659e-01 -3.34508866e-02 -4.37080115e-01 1.13529694e+00
-3.28608572e-01 -2.05838516e-01 1.16499960e-01 4.57375795e-01
-1.09846801e-01 -1.19159472e+00 -4.45289850e-01 1.08917856e+00
-7.44509757e-01 -4.18921351e-01 -3.73169929e-01 5.83853662e-01
6.00995302e-01 2.53397822e-01 3.45478415e-01 -7.94427618e-02
2.41818935e-01 5.61354160e-01 8.24975312e-01 -6.33803785e-01
-8.52586567e-01 -1.86497524e-01 2.32893333e-01 -6.55133188e-01
-2.27085292e-01 -1.21730018e+00 -1.17582893e+00 -1.58722043e-01
4.12220240e-01 -3.41986775e-01 8.43993902e-01 9.92952943e-01
4.28212613e-01 5.45798898e-01 1.63907096e-01 -1.21789992e+00
-6.41699284e-02 -1.14857268e+00 -5.74999154e-01 3.65919083e-01
2.19793394e-01 -1.11132514e+00 8.73571932e-02 -1.19726725e-01] | [14.191361427307129, -1.9273695945739746] |
a458ea21-37b9-461e-b5c6-3d8d7f80884a | generating-varied-training-corpora-in | null | null | https://aclanthology.org/2020.inlg-1.34 | https://aclanthology.org/2020.inlg-1.34.pdf | Generating Varied Training Corpora in Runyankore Using a Combined Semantic and Syntactic, Pattern-Grammar-based Approach | Machine learning algorithms have been applied to achieve high levels of accuracy in tasks associated with the processing of natural language. However, these algorithms require large amounts of training data in order to perform efficiently. Since most Bantu languages lack the required training corpora because they are computationally under-resourced, we investigated how to generate a large varied training corpus in Runyankore, a Bantu language indigenous to Uganda. We found the use of a combined semantic and syntactic, pattern and grammar-based approach to be applicable to this purpose, and used it to generate one million sentences, both labelled and unlabelled, which can be applied as training data for machine learning algorithms. The generated text was evaluated in two ways: (1) assessing the semantics encoded in word embeddings obtained from the generated text, which showed correct word similarity; and (2) applying the labelled data to tasks such as sentiment analysis, which achieved satisfactory levels of accuracy. | ['Joan Byamugisha'] | null | null | null | null | inlg-acl-2020-12 | ['word-similarity'] | ['natural-language-processing'] | [ 5.71162224e-01 1.91509947e-01 2.53429145e-01 -4.86737132e-01
-7.26828814e-01 -7.59594202e-01 6.27842069e-01 5.74069440e-01
-8.19982529e-01 7.29853034e-01 4.20712471e-01 -7.52302706e-01
-2.82460731e-02 -8.13935876e-01 -1.37523100e-01 -5.74401200e-01
1.04505919e-01 5.54339945e-01 -1.28834724e-01 -4.01064217e-01
7.58440316e-01 3.37618709e-01 -1.57359946e+00 4.22795415e-01
9.29855943e-01 3.35791409e-01 6.17398679e-01 6.44440413e-01
-4.80135828e-01 7.49612749e-01 -6.19924068e-01 -3.24862182e-01
7.96113312e-02 -4.08838868e-01 -1.09356332e+00 -7.75470808e-02
-2.84235150e-01 -5.65442666e-02 4.07815784e-01 7.77846217e-01
4.90209132e-01 1.01339206e-01 7.87439525e-01 -6.64224982e-01
-4.57489312e-01 4.55200106e-01 -7.31480643e-02 1.13496155e-01
6.48544967e-01 -6.48297593e-02 9.99224186e-01 -8.49592745e-01
7.88627148e-01 1.30234110e+00 4.80508327e-01 4.98774201e-01
-1.00381017e+00 -5.29853940e-01 -4.64371800e-01 -8.98748785e-02
-8.12705815e-01 -4.67364669e-01 4.01474744e-01 -4.44210857e-01
1.47647274e+00 -1.04159482e-01 5.24146974e-01 7.21646547e-01
1.30849421e-01 2.51860440e-01 1.43430495e+00 -1.08552611e+00
3.46240729e-01 4.62323368e-01 -1.51088461e-01 4.69065368e-01
3.89691442e-01 -1.94728673e-01 -2.51602322e-01 -3.69470894e-01
1.77164435e-01 -4.90816325e-01 1.36068761e-01 3.25776696e-01
-1.07964897e+00 1.48752534e+00 1.29916444e-01 8.06909263e-01
-5.18066287e-01 -4.01454240e-01 7.52816856e-01 3.82043183e-01
8.47534835e-01 8.83319557e-01 -6.52005792e-01 -3.23721051e-01
-6.55929148e-01 8.17568451e-02 8.74926209e-01 3.82664233e-01
5.41026235e-01 6.35203645e-02 4.36758965e-01 1.00715029e+00
3.65479529e-01 5.47120214e-01 8.25654924e-01 -6.67893052e-01
6.76058590e-01 7.53406405e-01 -1.25696644e-01 -1.27460659e+00
-3.31628591e-01 3.53744030e-01 -2.63276786e-01 1.10407889e-01
3.67449254e-01 -3.85109752e-01 -1.10449100e+00 1.49587965e+00
4.31681097e-01 -5.79900146e-01 7.61660337e-01 4.02951032e-01
5.36152542e-01 9.66741860e-01 5.01292288e-01 -1.89553812e-01
1.41482234e+00 -4.10950273e-01 -5.71314931e-01 -4.58545446e-01
1.28075385e+00 -1.13629735e+00 1.13542032e+00 2.90743023e-01
-8.27682674e-01 -3.46029282e-01 -9.03531671e-01 5.60744591e-02
-8.64358485e-01 -3.28278273e-01 7.12876916e-01 9.17245865e-01
-9.63024914e-01 5.27087688e-01 -4.81554538e-01 -9.29842770e-01
2.48258948e-01 2.96129167e-01 -6.77544594e-01 -1.72830924e-01
-1.30467439e+00 1.39734519e+00 8.82097185e-01 3.61628868e-02
1.57135222e-02 -2.39445403e-01 -1.23444808e+00 -2.74225831e-01
-2.12842375e-02 -9.07750800e-03 1.01314747e+00 -1.15324998e+00
-1.03269041e+00 1.27292812e+00 -2.22769991e-01 -1.31116241e-01
-4.31764312e-02 1.83266863e-01 -4.61967468e-01 3.09135914e-01
6.12564445e-01 5.02846956e-01 3.94211024e-01 -9.14185226e-01
-9.98401344e-01 -5.71506619e-01 -1.17074542e-01 3.78501862e-01
-4.24098253e-01 6.27427459e-01 2.68478036e-01 -6.29516006e-01
-5.05746342e-02 -7.48923063e-01 -1.99327320e-01 -7.82554448e-01
3.30605090e-01 -3.84120435e-01 6.20766640e-01 -1.18112838e+00
1.05782783e+00 -2.03645945e+00 -2.66493887e-01 4.37285393e-01
-1.97635204e-01 6.23120785e-01 -2.46614143e-01 8.55442584e-01
8.34288970e-02 3.93804044e-01 -3.93058449e-01 2.96380967e-01
-1.44062534e-01 4.91893291e-01 -2.19301358e-01 1.42654598e-01
5.42366564e-01 4.51718569e-01 -1.29931295e+00 -4.75595891e-01
3.41264516e-01 4.34096545e-01 -1.81388199e-01 2.19604805e-01
5.06648347e-02 6.31876588e-02 -4.47333366e-01 4.68548447e-01
2.08148748e-01 4.11111921e-01 6.55587673e-01 2.94627666e-01
-1.50733888e-01 6.97408199e-01 -7.39665985e-01 1.08593667e+00
-8.08710039e-01 7.84434497e-01 -7.45154992e-02 -9.51245964e-01
1.17799211e+00 4.74251091e-01 6.22574538e-02 -8.44559133e-01
1.64675683e-01 6.91213191e-01 2.41037175e-01 -9.00623083e-01
7.58513093e-01 -5.09698570e-01 -3.44027609e-01 8.05544019e-01
6.87066317e-02 -4.57195342e-01 5.12309849e-01 9.06406343e-02
1.04304099e+00 5.79552986e-02 5.14130414e-01 -4.79951948e-01
4.17255878e-01 5.29689908e-01 3.39990020e-01 1.69387922e-01
1.32060632e-01 2.01350167e-01 4.08067912e-01 -4.64404464e-01
-1.34303653e+00 -3.44587535e-01 -1.72901466e-01 1.18227744e+00
-5.13114095e-01 -1.09479830e-01 -8.81641746e-01 -6.09474123e-01
-4.37713504e-01 1.17314422e+00 -5.08508027e-01 1.05967417e-01
-4.43570971e-01 -9.39873278e-01 5.63416243e-01 3.12282741e-01
4.74056462e-03 -1.70087981e+00 -8.88264477e-01 6.41444147e-01
-5.63107878e-02 -8.76399219e-01 1.31077781e-01 3.34414840e-01
-7.44983017e-01 -1.12187612e+00 -2.49480382e-01 -9.34616029e-01
8.15802038e-01 9.26078930e-02 7.85506725e-01 1.05017841e-01
-3.99048269e-01 -1.66978519e-02 -7.03934312e-01 -7.05147743e-01
-8.74145508e-01 1.17249429e-01 7.36188740e-02 -4.03290570e-01
9.19091702e-01 -1.24333322e-01 8.83872062e-02 -7.28183463e-02
-1.51996326e+00 -2.59931952e-01 6.64844394e-01 7.51741767e-01
-3.27391960e-02 2.64151961e-01 9.35426116e-01 -1.21433437e+00
1.10290515e+00 -6.65065110e-01 -8.97866711e-02 2.51713932e-01
-4.21199709e-01 -5.71517348e-02 6.62068903e-01 -2.72767901e-01
-1.04335809e+00 -3.48213226e-01 -4.75984544e-01 6.04492784e-01
-3.86206061e-01 8.59660566e-01 5.26578203e-02 9.53256711e-02
6.43423140e-01 -6.96056187e-02 2.20031068e-01 -7.30901435e-02
2.12411746e-01 1.37340200e+00 7.60699362e-02 -4.11354214e-01
5.91498077e-01 9.63073149e-02 -3.46273214e-01 -1.22919095e+00
-6.10166430e-01 -5.24520278e-01 -6.84785426e-01 -5.19522913e-02
7.90461242e-01 -6.16802871e-01 -3.88420597e-02 2.87854254e-01
-1.01454103e+00 -3.88854235e-01 2.37386320e-02 6.66704774e-01
-2.19764337e-01 2.02354923e-01 -1.31048054e-01 -9.16654527e-01
-5.64390361e-01 -7.20339358e-01 6.99436963e-01 1.95313379e-01
-7.86812007e-01 -1.44243193e+00 3.66356462e-01 4.55658436e-01
3.48975122e-01 2.91557103e-01 1.46331656e+00 -1.21775389e+00
5.97474039e-01 -3.27784032e-01 -2.20687330e-01 5.54544806e-01
6.20970786e-01 1.28143981e-01 -8.27841997e-01 -1.70258597e-01
2.05195233e-01 -6.30301356e-01 9.46751386e-02 -1.81015223e-01
2.83617318e-01 -3.47974479e-01 -3.84985702e-03 -1.08001523e-01
1.46217191e+00 5.10441899e-01 5.63738644e-01 6.47765398e-01
3.42912287e-01 1.32236719e+00 7.52605915e-01 1.05375990e-01
3.89993280e-01 3.57895106e-01 -3.07568520e-01 3.56127806e-02
3.59495997e-01 -2.17119351e-01 5.44596553e-01 1.07900167e+00
3.99727076e-02 -6.54762313e-02 -1.23227584e+00 9.80198026e-01
-1.60596907e+00 -9.11650479e-01 -1.58584043e-01 1.95677567e+00
8.64573419e-01 4.77256663e-02 2.93388567e-03 3.86539757e-01
6.33798003e-01 3.19070145e-02 8.51932243e-02 -1.40674365e+00
4.07766663e-02 7.03875721e-01 2.78444231e-01 3.07651192e-01
-7.84538865e-01 1.05500424e+00 6.25690269e+00 7.87463844e-01
-1.03589606e+00 -7.53994510e-02 4.72690880e-01 3.40237021e-01
-4.15311366e-01 6.14784770e-02 -4.88717645e-01 5.01612365e-01
1.48322630e+00 -7.71739781e-02 2.52803773e-01 5.16889691e-01
5.70931077e-01 -4.28489804e-01 -5.13727307e-01 3.50489974e-01
2.28968322e-01 -1.00726628e+00 7.96687081e-02 -1.73394959e-02
5.99571049e-01 -1.43463746e-01 -4.27810341e-01 2.44335443e-01
3.13958347e-01 -1.17976749e+00 3.81259382e-01 -1.46167919e-01
6.40680790e-01 -8.83979857e-01 1.28820503e+00 2.99572229e-01
-7.27020085e-01 1.60008460e-01 -4.88424778e-01 -5.29116273e-01
1.32113978e-01 3.35318327e-01 -1.14154816e+00 5.03802776e-01
5.09255886e-01 2.11118668e-01 -5.41230142e-01 4.87826347e-01
-3.41584384e-01 6.56890392e-01 -3.71951044e-01 -5.17401457e-01
8.40192437e-01 -2.83443809e-01 4.56249714e-02 1.29318583e+00
3.39079827e-01 5.72660714e-02 -1.03025772e-01 2.83362418e-01
2.41376460e-01 6.62571728e-01 -9.64318752e-01 -5.27368009e-01
4.57402587e-01 1.15200758e+00 -1.06689131e+00 -1.87481031e-01
-4.31383878e-01 7.46101677e-01 3.72797757e-01 9.88133177e-02
-1.16986319e-01 -9.30374086e-01 2.96872497e-01 1.52668860e-02
1.97886035e-01 -3.22284810e-02 -2.58365601e-01 -7.74687767e-01
3.87998833e-03 -1.05162823e+00 2.94052035e-01 -6.39713466e-01
-1.27440751e+00 8.46144974e-01 1.25007078e-01 -4.39077049e-01
-6.55139446e-01 -8.54570925e-01 -7.38015652e-01 1.25522661e+00
-1.08724391e+00 -1.22660673e+00 1.41302690e-01 6.47789985e-02
5.89746416e-01 -2.70795941e-01 1.20096612e+00 -5.18777631e-02
-1.29607260e-01 5.93378721e-03 1.12514518e-01 1.59625530e-01
3.79436404e-01 -1.13399529e+00 3.64305228e-01 7.40328312e-01
1.15476903e-02 6.96991026e-01 5.56311607e-01 -7.93702424e-01
-1.07895911e+00 -1.07392335e+00 1.89393890e+00 -2.77376950e-01
8.65980089e-01 -3.19541365e-01 -8.27579439e-01 3.21723461e-01
2.32054606e-01 -7.55060792e-01 1.20773244e+00 -6.06453232e-02
-4.14151661e-02 4.15653169e-01 -1.50067818e+00 5.67947388e-01
3.31846893e-01 -6.69427276e-01 -1.32503951e+00 2.81011790e-01
3.21561217e-01 2.10597366e-01 -7.54112422e-01 1.62854567e-01
4.88095462e-01 -4.25032049e-01 4.38170820e-01 -7.94880569e-01
6.02629721e-01 -7.77652636e-02 -2.33116657e-01 -1.36801672e+00
-7.02940673e-02 -4.33515936e-01 8.49924445e-01 1.53779817e+00
7.63336599e-01 -8.44578147e-01 4.37721491e-01 6.71998918e-01
1.82411969e-01 -6.58767462e-01 -7.07912505e-01 -5.02872467e-01
1.55230924e-01 -5.01332879e-01 3.90352756e-01 1.01420760e+00
2.16462195e-01 5.20036042e-01 1.07270040e-01 -2.64570445e-01
2.64145553e-01 -3.19716722e-01 4.65345621e-01 -1.01879227e+00
3.24475259e-01 -1.88435152e-01 -4.78088677e-01 8.75382423e-02
3.49941432e-01 -8.88477981e-01 1.48032427e-01 -1.75223315e+00
-9.16031227e-02 -6.95562541e-01 1.81939214e-01 4.54850674e-01
-3.09987754e-01 3.79403919e-01 4.42997627e-02 1.80995651e-03
1.56273827e-01 2.80784518e-02 3.94783944e-01 2.48914212e-01
-3.47537816e-01 -3.02912086e-01 -1.05232620e+00 7.89119780e-01
1.04029202e+00 -6.72012269e-01 -4.20564383e-01 -4.02660280e-01
3.89350533e-01 -5.96151769e-01 -1.20008200e-01 -6.30734265e-01
-2.54783154e-01 -3.62438828e-01 2.26529703e-01 -1.11230634e-01
-1.08489789e-01 -6.94975019e-01 1.17364883e-01 4.32361215e-01
-3.92313302e-01 2.68812448e-01 4.13112730e-01 -2.38988232e-02
-2.55235225e-01 -1.05116034e+00 6.63240075e-01 -3.84741306e-01
-8.22615206e-01 -3.42935055e-01 -7.67358482e-01 1.82959676e-01
1.21859467e+00 -4.56639588e-01 -2.14081276e-02 -7.00729191e-02
-2.82904655e-01 -2.71577883e-04 6.82827413e-01 2.57119805e-01
5.32592833e-01 -1.03418779e+00 -7.46625781e-01 2.45758832e-01
1.96814701e-01 -1.50126636e-01 -6.41243979e-02 4.04689938e-01
-1.23874331e+00 6.52033806e-01 -4.55233485e-01 9.37460288e-02
-1.16795778e+00 4.08984751e-01 -4.06043768e-01 -3.60998183e-01
-4.28916305e-01 4.26343054e-01 -2.50343382e-01 -8.80418718e-01
-3.07650268e-01 5.94490431e-02 -5.82709968e-01 4.41209644e-01
5.75372398e-01 6.03894517e-02 2.10751504e-01 -1.00838971e+00
-2.54246265e-01 3.74655336e-01 4.29429002e-02 -4.37605888e-01
1.61369908e+00 7.54364058e-02 -3.71799707e-01 3.79881829e-01
1.23535407e+00 -1.55677855e-01 -3.70678753e-01 9.95185748e-02
4.32282388e-01 -5.64328611e-01 5.91166578e-02 -6.18000865e-01
-3.61006945e-01 8.41240704e-01 1.23634070e-01 4.52012450e-01
9.62209463e-01 -3.00314635e-01 6.57944620e-01 4.84298736e-01
4.44803238e-01 -1.52975345e+00 -5.18717825e-01 5.27365804e-01
3.93321455e-01 -1.15830839e+00 -1.72211230e-01 -1.70871392e-01
-7.63245940e-01 1.21924102e+00 1.30805522e-01 1.35603711e-01
3.06237578e-01 4.33627740e-02 3.53711903e-01 -2.42464438e-01
-6.97378218e-01 -3.16493720e-01 -1.08949780e-01 9.70280766e-01
6.13198102e-01 4.87628877e-02 -9.25214708e-01 1.31325543e-01
-3.24510902e-01 -3.26386355e-02 6.83161497e-01 1.28981209e+00
-5.29022753e-01 -1.36280394e+00 -3.65153909e-01 7.92795539e-01
-9.19653475e-01 -3.27329963e-01 -5.26826143e-01 8.30975533e-01
5.03290668e-02 1.27207971e+00 7.59645998e-02 -1.96406201e-01
2.86493395e-02 1.98434189e-01 2.15980649e-01 -1.03779769e+00
-8.94249558e-01 -2.11805388e-01 7.17595398e-01 -1.18562067e-02
-6.24045730e-01 -6.37678504e-01 -1.14555550e+00 -4.68068868e-01
-2.87309647e-01 6.95805252e-01 1.05085802e+00 1.34588313e+00
6.81607127e-02 -5.79392761e-02 5.93665600e-01 -8.43264580e-01
-4.04295117e-01 -1.34063423e+00 -5.95851302e-01 5.89241147e-01
-3.50897402e-01 -1.82872728e-01 -4.58010912e-01 5.61185665e-02] | [10.370850563049316, 9.727494239807129] |
ee54b984-2cb8-4bef-b296-b2ca742ecc1c | thraws-a-novel-dataset-for-thermal-hotspots | 2305.11891 | null | https://arxiv.org/abs/2305.11891v1 | https://arxiv.org/pdf/2305.11891v1.pdf | THRawS: A Novel Dataset for Thermal Hotspots Detection in Raw Sentinel-2 Data | Nowadays, most of the datasets leveraging space-borne Earth Observation (EO) data are based on high-end levels products, which are ortho-rectified, coregistered, calibrated, and further processed to mitigate the impact of noise and distortions. Nevertheless, given the growing interest to apply Artificial Intelligence (AI) onboard satellites for time-critical applications, such as natural disaster response, providing raw satellite images could be useful to foster the research on energy-efficient pre-processing algorithms and AI models for onboard-satellite applications. In this framework, we present THRawS, the first dataset composed of Sentinel-2 (S-2) raw data containing warm temperature hotspots (wildfires and volcanic eruptions). To foster the realisation of robust AI architectures, the dataset gathers data from all over the globe. Furthermore, we designed a custom methodology to identify events in raw data starting from the corresponding Level-1C (L1C) products. Indeed, given the availability of state-of-the-art algorithms for thermal anomalies detection on the L1C tiles, we detect such events on these latter and we then re-project them on the corresponding raw images. Additionally, to deal with unprocessed data, we devise a lightweight coarse coregisteration and georeferencing strategy. The developed dataset is comprehensive of more than 100 samples containing wildfires, volcanic eruptions, and event-free volcanic areas to enable both warm-events detection and general classification applications. Finally, we compare performances between the proposed coarse spatial coregistration technique and the SuperGlue Deep Neural Network method to highlight the different constraints in terms of timing and quality of spatial registration to minimise the spatial displacement error for a specific scene. | ['Nicolas Longépé', 'Olivier Colin', 'Alix De Beussche', 'Federico Serva', 'Roberto Del Prete', 'Gabriele Meoni'] | 2023-05-12 | null | null | null | null | ['classification'] | ['methodology'] | [ 4.52838659e-01 3.96200605e-02 4.23650175e-01 -2.69680440e-01
-6.56724036e-01 -4.05081272e-01 9.94057417e-01 3.18132699e-01
-5.86708367e-01 5.73782563e-01 8.70732665e-02 -3.02582026e-01
-3.66206050e-01 -1.26421654e+00 -5.11074722e-01 -1.02501774e+00
-6.78918481e-01 2.80901402e-01 -5.79666980e-02 -6.45749390e-01
-1.25007674e-01 8.96623790e-01 -1.85741019e+00 9.13848430e-02
9.96764600e-01 1.20002711e+00 1.78681180e-01 3.86701733e-01
2.76587784e-01 1.14484780e-01 -4.46224838e-01 3.34890187e-01
5.04911482e-01 -1.94677234e-01 -5.09636521e-01 -2.31228203e-01
4.03302163e-01 -4.45422716e-02 1.63046673e-01 1.02974141e+00
3.39894056e-01 2.65007913e-01 6.08655095e-01 -9.81538057e-01
4.42535356e-02 4.19529602e-02 -2.84515828e-01 4.97682542e-01
-7.27547854e-02 1.85347006e-01 5.28669655e-01 -7.10812747e-01
4.86088932e-01 6.85977638e-01 7.90542960e-01 -3.06637764e-01
-1.05273449e+00 -3.51889372e-01 -5.91769367e-02 1.65940285e-01
-1.55689263e+00 -5.01115382e-01 5.17095089e-01 -7.03345954e-01
1.10434377e+00 6.04979575e-01 7.77314425e-01 9.35050607e-01
9.65392664e-02 1.96906999e-02 1.22845471e+00 -5.42447150e-01
5.02165556e-01 -3.24186742e-01 -1.98201433e-01 1.68269843e-01
3.64176959e-01 4.21650976e-01 -1.98233351e-01 -8.38012435e-04
3.14346135e-01 4.20147367e-02 -4.36923772e-01 3.02802712e-01
-1.03078461e+00 6.28799796e-01 9.47367847e-01 6.43137515e-01
-1.08617020e+00 -6.05030768e-02 1.55017287e-01 1.46655083e-01
9.51543093e-01 2.34418750e-01 -2.62055337e-01 1.98732704e-01
-1.41463888e+00 9.18897614e-02 2.50206023e-01 4.15778965e-01
9.52765167e-01 3.20784032e-01 1.89954296e-01 4.63086188e-01
4.33882564e-01 9.99054551e-01 1.99822828e-01 -2.79006571e-01
4.07732725e-01 7.87000358e-01 -2.41558719e-03 -1.31766593e+00
-7.93063223e-01 -4.91172343e-01 -1.42540145e+00 2.91445524e-01
-5.97153530e-02 -9.63220298e-02 -1.08172488e+00 1.35498488e+00
4.01089013e-01 1.49799854e-01 6.95812181e-02 1.03800559e+00
2.04142675e-01 1.03332353e+00 2.06636935e-01 -1.41595438e-01
1.54221570e+00 -8.19986761e-02 -5.45358121e-01 -4.61553007e-01
6.58336163e-01 -2.49872401e-01 5.65478504e-01 2.23195270e-01
-2.23582357e-01 -4.37407494e-01 -1.13166285e+00 5.66353023e-01
-1.04914153e+00 8.97124782e-02 4.98098403e-01 4.94145870e-01
-1.18798876e+00 6.25817955e-01 -7.46669888e-01 -6.85675442e-01
2.17751041e-01 9.05474573e-02 -2.86386430e-01 3.54025275e-01
-1.56304502e+00 1.11137545e+00 8.13710451e-01 9.53693569e-01
-8.23543727e-01 -7.29382515e-01 -8.80014062e-01 1.11831091e-02
-1.86576846e-03 -2.58003891e-01 2.23049447e-01 -1.08581853e+00
-8.58437657e-01 9.38327849e-01 4.16942745e-01 -7.18040287e-01
2.58134872e-01 -7.78791085e-02 -9.40449357e-01 1.36134431e-01
1.15196221e-01 3.77963811e-01 8.18704426e-01 -1.07155859e+00
-7.91221142e-01 -5.62969685e-01 -4.79717255e-01 -1.05597861e-02
-3.22546422e-01 5.05515933e-02 -2.29868330e-02 -9.30604219e-01
1.39504716e-01 -7.68189192e-01 -2.66841203e-01 -1.81456774e-01
-8.38162675e-02 2.04456657e-01 6.96614802e-01 -1.04784536e+00
1.16519988e+00 -2.34336352e+00 9.06581432e-02 6.20622814e-01
-3.34622502e-01 2.08009794e-01 -1.10457711e-01 3.35993946e-01
-3.39656770e-01 8.27742815e-02 -8.33851635e-01 2.47237775e-02
-1.66210577e-01 3.74348372e-01 -5.42970479e-01 7.53468513e-01
4.70478594e-01 3.37646186e-01 -6.14821136e-01 -2.43406355e-01
5.77943623e-01 5.17037511e-01 -3.91973592e-02 1.33103833e-01
-3.06279033e-01 5.53052187e-01 -3.93529326e-01 5.96114099e-01
1.07510674e+00 5.19248426e-01 -9.97571275e-03 -2.67380327e-01
-6.85887277e-01 7.43053854e-02 -1.22061837e+00 1.52210844e+00
-5.73751867e-01 5.88476241e-01 2.69690901e-01 -8.77060056e-01
1.21804690e+00 3.79067093e-01 5.51465273e-01 -1.18712270e+00
7.85814449e-02 3.49390924e-01 -3.85160118e-01 -5.53827047e-01
6.71021223e-01 -2.00144038e-01 -5.51943593e-02 3.09833050e-01
-1.25156224e-01 -6.02601059e-02 7.68462569e-03 -2.79998571e-01
6.84633732e-01 1.81106523e-01 1.29027054e-01 -6.23658061e-01
6.47414446e-01 3.54514360e-01 4.02666450e-01 4.33370382e-01
3.76276672e-02 7.49980509e-01 7.35297725e-02 -9.14567828e-01
-1.16818964e+00 -7.47832119e-01 -4.22653466e-01 8.67369056e-01
-3.41731101e-01 2.16818273e-01 -5.75751007e-01 -2.71715462e-01
-2.67813206e-01 8.16572487e-01 -7.38592803e-01 -1.77842495e-03
-4.59301829e-01 -1.65642846e+00 8.78200650e-01 1.43544614e-01
7.96490848e-01 -1.00472808e+00 -1.17311132e+00 2.91779995e-01
-1.96412727e-01 -6.82311714e-01 5.64578295e-01 4.21010286e-01
-9.51188028e-01 -1.00983334e+00 -2.94888049e-01 -1.40348613e-01
5.00487864e-01 -7.93521851e-02 1.04559767e+00 2.34538708e-02
-3.06670874e-01 3.15944217e-02 -5.03459513e-01 -2.52801657e-01
-3.41655433e-01 8.01299810e-02 -4.02529957e-03 3.52338076e-01
1.83235213e-01 -5.89800179e-01 -6.49489641e-01 1.59847423e-01
-1.32806075e+00 -1.30635098e-01 5.45727253e-01 2.81526715e-01
3.53804469e-01 2.54622996e-01 2.69400954e-01 -9.77219939e-02
7.87289366e-02 -8.55168641e-01 -9.82769966e-01 2.57359922e-01
-4.07962173e-01 -2.42060591e-02 3.47943723e-01 3.27229470e-01
-1.20125091e+00 2.32361943e-01 -1.57951415e-01 5.86151108e-02
-7.24706709e-01 8.46298695e-01 4.50911634e-02 -2.71345805e-02
1.02225018e+00 3.60748887e-01 -2.62221187e-01 -5.84065676e-01
2.12688699e-01 9.54517186e-01 8.17386687e-01 -2.46676341e-01
9.95259047e-01 8.12063038e-01 1.53977117e-02 -1.19455290e+00
-4.52498049e-01 -4.40652668e-01 -7.47500420e-01 -6.77428246e-01
1.05664015e+00 -1.05420196e+00 -4.63914908e-02 5.87853253e-01
-1.05969286e+00 -2.50012606e-01 -1.19815558e-01 3.95315021e-01
-7.47552440e-02 1.51766926e-01 8.84894133e-02 -1.00720692e+00
-7.39524126e-01 -6.59331739e-01 1.08536732e+00 1.91401586e-01
1.04947649e-02 -9.75107312e-01 5.84458709e-01 -3.27647738e-02
7.94179976e-01 8.23904455e-01 6.40869141e-01 -4.35746670e-01
-4.13195461e-01 -2.23749250e-01 -2.38083135e-02 2.92229265e-01
1.86130345e-01 4.56365682e-02 -1.24647057e+00 -2.69915104e-01
1.69466108e-01 2.20578089e-01 1.01269853e+00 2.17645004e-01
6.31993115e-01 -2.60766000e-01 -1.81190863e-01 8.30759823e-01
1.68307436e+00 1.12493046e-01 9.61041451e-01 9.20616031e-01
2.73323506e-01 7.08325863e-01 6.07263446e-01 6.45361841e-01
7.16962963e-02 6.27557874e-01 1.02648449e+00 -4.72049505e-01
2.57372141e-01 4.40609127e-01 2.88819283e-01 1.47999898e-01
-5.10084748e-01 1.38675747e-02 -1.27847981e+00 9.16164458e-01
-1.61508036e+00 -9.71533477e-01 -6.12526119e-01 2.49769950e+00
3.15106720e-01 -6.50483444e-02 -3.07934105e-01 1.41180724e-01
6.87080026e-01 7.71612465e-01 -2.07807012e-02 -1.10155731e-01
-5.07052720e-01 3.64093661e-01 9.52412188e-01 1.78333148e-01
-1.56885493e+00 4.80789185e-01 5.25806713e+00 5.33900380e-01
-1.40251684e+00 1.96066841e-01 3.93293917e-01 4.73758765e-02
-3.11388433e-01 -1.62888393e-01 -3.90886635e-01 4.80893463e-01
1.39110887e+00 4.09617841e-01 4.17467743e-01 5.84740818e-01
6.42039835e-01 -5.15757978e-01 -4.26170200e-01 5.96110582e-01
-1.48919120e-01 -1.26789916e+00 -3.71430397e-01 1.01025820e-01
5.86989999e-01 6.26330972e-01 -2.10563168e-01 -7.11657107e-02
4.55327630e-02 -7.69041777e-01 7.64537692e-01 7.91394472e-01
6.56490386e-01 -6.24863744e-01 9.08393979e-01 9.31440517e-02
-1.26352894e+00 -1.26113102e-01 -1.92851961e-01 -9.64606032e-02
1.74816355e-01 1.14098454e+00 -5.51566124e-01 1.18378472e+00
1.11488926e+00 6.43160641e-01 -6.23673916e-01 8.61124218e-01
-2.39024073e-01 4.60433185e-01 -1.01130283e+00 6.15946174e-01
5.35529792e-01 -5.00459135e-01 4.63954628e-01 1.14388037e+00
8.76559258e-01 1.18197396e-01 -3.30397815e-01 7.83817112e-01
4.95229632e-01 -5.09832986e-02 -8.27679873e-01 3.50482553e-01
4.07222450e-01 1.50524402e+00 -6.34817123e-01 -3.57454598e-01
-7.28878602e-02 6.35255516e-01 -2.69019485e-01 2.68628180e-01
-7.72716582e-01 -1.85901046e-01 7.74215460e-01 9.97568071e-02
8.43914002e-02 -3.07327718e-01 -3.74349594e-01 -9.05681789e-01
1.53639868e-01 -5.71524382e-01 5.25855362e-01 -8.82254183e-01
-7.42175579e-01 8.36830139e-01 1.61605135e-01 -1.55556571e+00
-2.49157757e-01 -3.46961528e-01 -9.07845318e-01 9.26386774e-01
-1.74421716e+00 -1.11598814e+00 -7.52110422e-01 3.95774990e-01
-4.01955694e-02 9.04574096e-02 8.46568465e-01 3.93841892e-01
-6.41891599e-01 -4.73741651e-01 4.14133072e-01 -7.09245056e-02
3.92875135e-01 -9.74288702e-01 5.49617767e-01 1.43877208e+00
-2.70133346e-01 8.70612636e-02 6.84474826e-01 -7.19275892e-01
-1.08838356e+00 -1.51704586e+00 7.58368850e-01 8.73697326e-02
7.52380908e-01 -3.87470908e-02 -1.08458281e+00 6.09671772e-01
2.37858295e-01 2.09378153e-02 1.41142055e-01 -2.28758469e-01
4.90388200e-02 -5.60531139e-01 -1.26042330e+00 2.92455584e-01
4.25391734e-01 -5.04057229e-01 -6.80097461e-01 4.34883147e-01
-1.16312644e-02 1.14533398e-02 -1.08484316e+00 5.97093046e-01
6.44642785e-02 -1.15599310e+00 7.76382506e-01 -6.35189041e-02
1.93676636e-01 -6.29720867e-01 -1.65477753e-01 -1.46506500e+00
-1.66876048e-01 -9.30345431e-02 4.04380679e-01 1.11699939e+00
9.56484079e-02 -6.49235964e-01 3.41081738e-01 2.97002345e-01
-4.03653651e-01 2.31940299e-01 -1.48872519e+00 -7.62898028e-01
-1.67281285e-01 -5.75288415e-01 7.25535750e-01 9.83084321e-01
-4.36372608e-01 -4.87695277e-01 -1.71307236e-01 9.17149484e-01
5.98759413e-01 3.12068835e-02 3.63162845e-01 -1.55345273e+00
4.77430522e-01 -3.06801796e-01 -2.63017684e-01 1.00735575e-02
-3.98033381e-01 -6.44976258e-01 3.02536368e-01 -1.10330546e+00
-4.88296151e-01 -5.86202681e-01 -3.09849411e-01 7.63507068e-01
1.25582457e-01 3.99565876e-01 -1.37292057e-01 1.88994423e-01
1.32364050e-01 7.87162483e-01 2.23244384e-01 -6.48008734e-02
-2.29155365e-02 -3.91409129e-01 2.11284801e-01 5.53467989e-01
7.71526814e-01 -6.73506796e-01 2.84486532e-01 -5.72778583e-01
7.16813147e-01 -2.06137896e-01 7.77990282e-01 -1.52807891e+00
1.15953766e-01 -1.92956537e-01 1.52897313e-01 -9.14491475e-01
-1.23445928e-01 -1.17457056e+00 8.46930206e-01 4.78527546e-01
3.54163140e-01 -1.49674013e-01 3.20788562e-01 2.44107351e-01
-4.69827056e-01 -1.54807121e-01 8.83409083e-01 9.55414474e-02
-1.07863820e+00 1.25381753e-01 -8.07306945e-01 -4.61586177e-01
1.10465586e+00 -1.60291716e-01 -3.80981445e-01 1.70810372e-01
-4.51534301e-01 2.16680437e-01 5.67148447e-01 3.18643481e-01
1.80143058e-01 -1.10395813e+00 -8.36372554e-01 5.95319629e-01
3.37304115e-01 3.28505278e-01 7.58330464e-01 8.85498881e-01
-9.47486222e-01 4.02234584e-01 -4.26928699e-01 -7.07047164e-01
-7.45015681e-01 3.68718356e-01 6.16061509e-01 -1.14430711e-01
-5.57887673e-01 3.80032063e-01 -3.60067844e-01 -4.94363457e-01
-3.89515847e-01 -5.69196105e-01 -3.41959625e-01 7.01053500e-01
5.39116859e-01 2.66620606e-01 6.82582378e-01 -8.31658185e-01
-6.64686143e-01 5.48238635e-01 7.04672098e-01 -7.91065469e-02
1.79521692e+00 -1.34475246e-01 -3.67785990e-01 7.78162330e-02
8.52540851e-01 -5.62514007e-01 -1.09281826e+00 -1.48863755e-02
3.88171226e-01 -2.26184174e-01 7.00489163e-01 -5.43960452e-01
-1.45396006e+00 6.65641606e-01 1.10866618e+00 3.78913432e-01
1.59813237e+00 -3.06568265e-01 1.96570426e-01 3.62948895e-01
1.78672940e-01 -1.26598358e+00 -7.61771381e-01 3.32710415e-01
1.08987713e+00 -1.10849071e+00 1.80370271e-01 9.15600434e-02
-2.31149718e-01 1.22422373e+00 4.52406444e-02 1.27042243e-02
5.64284563e-01 3.70352641e-02 5.29503524e-02 -5.39444029e-01
-2.53019691e-01 -5.30363679e-01 1.03147283e-01 6.65414393e-01
-7.39364699e-02 1.78627387e-01 8.13249424e-02 1.96296498e-01
3.43373939e-02 1.66530445e-01 4.52650994e-01 8.86987627e-01
-6.05459273e-01 -4.02630776e-01 -1.27674186e+00 2.90697932e-01
4.25927825e-02 -2.65145391e-01 2.31504198e-02 7.51031995e-01
2.17062294e-01 6.87342763e-01 5.57212412e-01 -2.03230441e-01
3.36430669e-01 8.82480219e-02 -1.69625565e-01 -8.37437361e-02
-8.77616405e-01 -6.70196041e-02 1.44305199e-01 -5.34336030e-01
-7.01238513e-01 -7.21173942e-01 -7.44083881e-01 -2.46106297e-01
-2.01334152e-02 -2.14289650e-02 1.21147156e+00 1.11445987e+00
4.77170318e-01 4.84140754e-01 7.49958158e-01 -1.54217136e+00
-8.58397037e-02 -1.31125045e+00 -6.53181553e-01 1.65583074e-01
4.29198533e-01 -6.75752878e-01 -4.49513793e-01 -1.25069797e-01] | [9.744267463684082, -1.6957331895828247] |
fda7adb8-f4b0-4bf9-adfd-b087c6caf859 | guiding-visual-attention-in-deep | 2206.10587 | null | https://arxiv.org/abs/2206.10587v2 | https://arxiv.org/pdf/2206.10587v2.pdf | Guiding Visual Attention in Deep Convolutional Neural Networks Based on Human Eye Movements | Deep Convolutional Neural Networks (DCNNs) were originally inspired by principles of biological vision, have evolved into best current computational models of object recognition, and consequently indicate strong architectural and functional parallelism with the ventral visual pathway throughout comparisons with neuroimaging and neural time series data. As recent advances in deep learning seem to decrease this similarity, computational neuroscience is challenged to reverse-engineer the biological plausibility to obtain useful models. While previous studies have shown that biologically inspired architectures are able to amplify the human-likeness of the models, in this study, we investigate a purely data-driven approach. We use human eye tracking data to directly modify training examples and thereby guide the models' visual attention during object recognition in natural images either towards or away from the focus of human fixations. We compare and validate different manipulation types (i.e., standard, human-like, and non-human-like attention) through GradCAM saliency maps against human participant eye tracking data. Our results demonstrate that the proposed guided focus manipulation works as intended in the negative direction and non-human-like models focus on significantly dissimilar image parts compared to humans. The observed effects were highly category-specific, enhanced by animacy and face presence, developed only after feedforward processing was completed, and indicated a strong influence on face detection. With this approach, however, no significantly increased human-likeness was found. Possible applications of overt visual attention in DCNNs and further implications for theories of face detection are discussed. | ['Walter R. Gruber', 'Sebastian J. Denzler', 'Leonard E. van Dyck'] | 2022-06-21 | null | null | null | null | ['face-detection'] | ['computer-vision'] | [ 4.25308049e-01 1.78797930e-01 2.90993661e-01 -2.92807877e-01
5.57787776e-01 -4.05473888e-01 9.89459574e-01 -9.25687999e-02
-7.58380175e-01 3.99966240e-01 1.00534439e-01 -1.85819924e-01
-3.50611299e-01 -4.13400322e-01 -6.44203544e-01 -7.72503972e-01
1.25234842e-01 -1.54508531e-01 2.82512069e-01 -4.71855313e-01
6.61067843e-01 1.06194758e+00 -2.20673561e+00 2.79463917e-01
3.76456469e-01 8.06429088e-01 3.60228091e-01 4.42602843e-01
1.26337618e-01 3.90134275e-01 -6.08347356e-01 -1.87554419e-01
3.06122631e-01 -8.42879534e-01 -3.74715447e-01 -3.98852900e-02
7.01470792e-01 -9.48958918e-02 -3.92326787e-02 1.00835538e+00
7.62095332e-01 1.07336886e-01 5.80501854e-01 -1.29550290e+00
-1.12176335e+00 2.24264741e-01 -4.57488924e-01 7.89254308e-01
3.70301634e-01 7.21198022e-01 4.71514285e-01 -1.01758873e+00
7.32337058e-01 1.43386519e+00 4.71707553e-01 9.50956821e-01
-1.54890394e+00 -4.46706474e-01 5.88115081e-02 2.88956493e-01
-1.11336446e+00 -7.99189329e-01 7.57924497e-01 -4.83133554e-01
1.21277297e+00 3.17482382e-01 1.14561999e+00 1.27968097e+00
2.64793336e-01 8.87221470e-02 1.52748728e+00 -6.79736674e-01
1.67925045e-01 3.84131730e-01 -6.12806007e-02 2.50891089e-01
4.74257082e-01 8.70781004e-01 -7.72793710e-01 1.90963492e-01
9.84349608e-01 -1.78649768e-01 -5.19182801e-01 -3.80095273e-01
-1.36116135e+00 6.36857629e-01 8.12516451e-01 8.13440979e-01
-7.52742112e-01 -3.43362652e-02 1.32240683e-01 -1.77100301e-02
1.13775909e-01 7.49239445e-01 -3.03432107e-01 3.47462982e-01
-9.20219064e-01 2.29094326e-01 1.63156494e-01 3.21536481e-01
4.64723945e-01 4.18445647e-01 -5.01878083e-01 5.52777529e-01
3.07279587e-01 4.06658828e-01 8.98139060e-01 -8.15045357e-01
-4.23515171e-01 5.43736815e-01 -4.37711626e-02 -1.38616252e+00
-7.80041397e-01 -4.01406616e-01 -4.41246033e-01 8.12478602e-01
4.80230808e-01 1.57118499e-01 -8.74777377e-01 1.99707150e+00
3.03182155e-01 -4.38650906e-01 -2.91572988e-01 1.37226939e+00
6.28524959e-01 1.27452195e-01 4.36483294e-01 -3.91918957e-01
1.50773585e+00 -4.46080983e-01 -7.96161652e-01 -4.23159808e-01
3.02514404e-01 -4.28067476e-01 1.15397286e+00 5.46273552e-02
-1.13133073e+00 -8.96717072e-01 -9.68554795e-01 -1.75685380e-02
-5.55199325e-01 -1.70987442e-01 6.08972251e-01 7.35426188e-01
-1.59622669e+00 7.44742155e-01 -5.81853449e-01 -8.76802146e-01
7.61207640e-01 2.96882421e-01 -3.24707329e-01 3.98853123e-01
-1.09808958e+00 1.44643188e+00 4.17935044e-01 4.86927629e-01
-7.60402679e-01 -7.27543950e-01 -6.50513113e-01 9.70968828e-02
-6.54151812e-02 -7.35683322e-01 9.71283555e-01 -1.71973145e+00
-1.22057617e+00 1.28438425e+00 -2.50439078e-01 -4.42626655e-01
7.68592805e-02 1.58328161e-01 -5.63923836e-01 4.30257052e-01
-1.20488696e-01 1.45104253e+00 1.06605697e+00 -1.12945879e+00
-2.25449324e-01 -4.90740061e-01 -4.33636419e-02 -4.06225696e-02
-3.29846174e-01 3.56944174e-01 4.69726056e-01 -6.44773304e-01
-1.15501443e-02 -7.66594887e-01 6.84322864e-02 5.34940243e-01
1.40611306e-01 -3.14414263e-01 6.91256106e-01 -3.45742077e-01
9.61843014e-01 -2.03145409e+00 -6.73205033e-03 6.12361841e-02
3.69501203e-01 6.09121561e-01 -3.21109802e-01 2.00217068e-01
-6.93621874e-01 1.30255893e-01 -2.16308638e-01 3.23573261e-01
-9.76209417e-02 -9.25993174e-02 -2.11332235e-02 7.24972188e-01
5.47465086e-01 1.17086291e+00 -8.60601306e-01 -1.64688513e-01
2.69972771e-01 6.20684743e-01 -5.73693097e-01 -1.22266285e-01
-2.77306121e-02 5.05038977e-01 3.55868369e-01 4.91531640e-01
6.13807201e-01 1.06441371e-01 -5.77164162e-03 -4.35223430e-01
-5.56985557e-01 1.66509986e-01 -5.07840693e-01 1.35728037e+00
3.39031406e-02 1.12367523e+00 -1.00807667e-01 -9.74879324e-01
7.70487309e-01 2.34193251e-01 -1.67569835e-02 -1.36497247e+00
4.62263823e-01 1.13018893e-01 1.09958827e+00 -6.01419032e-01
-2.53617833e-03 -4.57815528e-01 7.57275283e-01 3.78597140e-01
2.10526541e-01 1.41456187e-01 1.59954533e-01 -1.71834603e-01
5.97778678e-01 3.25751722e-01 3.54329318e-01 -6.71897948e-01
4.06926483e-01 -1.89374760e-01 1.04063325e-01 5.09254515e-01
-5.98339200e-01 5.00999689e-01 5.29372633e-01 -4.96620893e-01
-8.29611480e-01 -7.44964480e-01 -1.94125026e-01 1.27590513e+00
8.02330077e-02 1.53353542e-01 -1.01116264e+00 -2.61115134e-01
-2.76551485e-01 1.08362246e+00 -1.15739262e+00 -7.41037726e-01
-3.16291392e-01 -4.79232430e-01 3.26865882e-01 3.16410899e-01
3.29349995e-01 -1.89277685e+00 -1.66840565e+00 -3.13935317e-02
4.31807458e-01 -7.00251341e-01 -9.87796262e-02 -3.16055417e-02
-7.52310455e-01 -1.08739090e+00 -8.28785598e-01 -6.97425485e-01
7.99920380e-01 4.00611728e-01 9.45496500e-01 2.35990837e-01
-6.88863754e-01 4.69105393e-01 -3.71163338e-02 -6.29488826e-01
-1.70409843e-01 -4.99065638e-01 1.65814385e-01 7.27435946e-03
6.16067052e-01 -5.95186353e-01 -8.32278550e-01 3.32638174e-01
-9.87697303e-01 -2.87676826e-02 6.14234805e-01 6.57499194e-01
-1.84883550e-02 -4.86326694e-01 6.02644980e-01 -3.29468131e-01
6.51784897e-01 -1.72935471e-01 -4.50488806e-01 -9.53021869e-02
-7.34513938e-01 -8.74735266e-02 1.38193518e-01 -7.48372853e-01
-1.15543079e+00 -3.04347038e-01 1.92792609e-01 -5.50919950e-01
-6.19285345e-01 2.97547698e-01 1.54481575e-01 -4.04247254e-01
1.25815058e+00 1.00668482e-01 1.68296918e-01 2.10417286e-02
9.57511589e-02 2.05861911e-01 3.17688406e-01 -5.25200628e-02
6.22118533e-01 5.78313589e-01 -5.65027185e-02 -9.05005217e-01
-4.78886187e-01 7.24787489e-02 -8.02176476e-01 -5.44735909e-01
9.36353028e-01 -4.04818475e-01 -9.05964911e-01 4.58871931e-01
-1.29384065e+00 -4.12043363e-01 -3.75275940e-01 4.95382994e-01
-3.94651085e-01 -1.81965515e-01 -4.12332267e-02 -8.21239471e-01
-2.03610614e-01 -8.54956090e-01 7.69350111e-01 5.61854601e-01
-5.01798570e-01 -9.13765848e-01 1.25282556e-02 -2.14470938e-01
7.25065768e-01 2.37615600e-01 9.41239297e-01 -6.24451280e-01
-1.48471326e-01 1.60216406e-01 -3.73027354e-01 1.33290276e-01
2.23689988e-01 1.89375192e-01 -1.37428880e+00 -1.23853311e-01
3.18002611e-01 -1.49069875e-01 7.17316568e-01 5.10461509e-01
7.87468433e-01 -2.45571092e-01 -2.90853530e-01 2.26529524e-01
1.16010237e+00 3.60196859e-01 9.21999216e-01 1.47688091e-01
2.31041417e-01 1.21972430e+00 3.02132457e-01 1.78636499e-02
-2.64005452e-01 5.45168817e-01 5.46187043e-01 -2.40112707e-01
-4.80398655e-01 -4.36333008e-02 3.34122956e-01 4.71193641e-02
-2.41359860e-01 -3.89057882e-02 -8.33360016e-01 6.97175741e-01
-1.22877657e+00 -1.15514112e+00 -3.25372934e-01 2.25479436e+00
5.95506907e-01 2.49787301e-01 1.24089234e-01 7.02953860e-02
8.78105819e-01 -2.70966627e-02 -6.04339182e-01 -6.06426716e-01
-3.82889152e-01 3.89004290e-01 2.76470669e-02 -4.14327048e-02
-6.17776871e-01 8.82914305e-01 6.58271027e+00 2.85561383e-01
-1.60297132e+00 1.05829395e-01 3.41981590e-01 -4.18117583e-01
-2.10212499e-01 -1.24607176e-01 -4.42281336e-01 3.47286344e-01
1.23385262e+00 -9.55590084e-02 5.82520664e-01 4.79121149e-01
5.08503675e-01 -3.15332413e-01 -1.26540053e+00 9.41333115e-01
2.52368599e-01 -1.20217705e+00 -2.65407208e-02 1.04117617e-01
3.36192310e-01 -2.18460128e-01 3.30857337e-01 1.73169285e-01
-5.43172240e-01 -1.14052916e+00 1.03602386e+00 7.14821219e-01
6.91384852e-01 -2.49875575e-01 3.21227670e-01 1.71609998e-01
-5.30577660e-01 -1.08630687e-01 -4.98239815e-01 -5.29888391e-01
-1.17319655e-02 8.89542922e-02 -5.04909456e-01 -3.31087619e-01
8.11594903e-01 3.86130750e-01 -1.00117397e+00 1.14917648e+00
-8.99394155e-02 3.48341525e-01 9.62070189e-04 -2.33840972e-01
4.33694609e-02 2.08594248e-01 6.47452414e-01 9.31001365e-01
1.31905884e-01 1.33409292e-01 -7.77485490e-01 1.42817271e+00
3.79990280e-01 -1.92523077e-02 -7.50589073e-01 -2.47351706e-01
2.12764874e-01 1.39244747e+00 -1.07303083e+00 -1.50474101e-01
-3.55036944e-01 9.23371613e-01 1.88096404e-01 5.10493159e-01
-8.80391598e-01 -3.66787493e-01 5.15150011e-01 3.95299762e-01
3.57944697e-01 9.80150476e-02 -3.01031590e-01 -6.28749669e-01
-3.64612415e-02 -6.95380092e-01 -2.19661430e-01 -1.33667314e+00
-8.91988456e-01 6.53802216e-01 8.36394057e-02 -7.07245409e-01
-8.72600749e-02 -8.41148198e-01 -5.61069548e-01 1.09688795e+00
-1.28910267e+00 -8.87882113e-01 -3.11436713e-01 5.61153173e-01
3.89105916e-01 1.17184386e-01 7.24828243e-01 -8.89048874e-02
-3.27286631e-01 3.70370924e-01 -6.16577923e-01 -2.58013844e-01
6.23438716e-01 -7.33411431e-01 7.12960884e-02 8.57980549e-01
1.35737881e-01 9.25319076e-01 8.41157675e-01 -4.96923268e-01
-8.12583566e-01 -5.49924731e-01 7.95883596e-01 -4.67518330e-01
3.44280541e-01 -4.28248286e-01 -1.09199393e+00 1.82866335e-01
6.17193639e-01 1.54172122e-01 4.20593500e-01 -2.89492905e-01
-3.55472118e-01 7.18947873e-02 -1.32730973e+00 8.48446488e-01
1.33214700e+00 -4.75509375e-01 -9.12620306e-01 1.44378170e-01
4.30089235e-01 1.75919190e-01 -2.98535138e-01 3.84683490e-01
6.76121593e-01 -1.47931170e+00 8.34151685e-01 -8.56231391e-01
2.03428254e-01 -2.04094157e-01 2.58934170e-01 -1.22312617e+00
-7.14482427e-01 -2.81351209e-01 7.73586184e-02 9.87335503e-01
2.12606698e-01 -7.56506443e-01 3.82208884e-01 5.77060759e-01
-2.03705966e-01 -2.81565994e-01 -9.05019879e-01 -5.41901529e-01
-1.05728835e-01 -7.31659830e-02 1.91580087e-01 8.17511916e-01
-7.32957125e-02 2.36271232e-01 2.85397798e-01 -2.97163837e-02
2.43448198e-01 -2.57308125e-01 2.51484871e-01 -1.43978441e+00
1.69627070e-01 -9.48541343e-01 -6.40838146e-01 -8.27703699e-02
1.39802337e-01 -7.44720817e-01 -2.35361643e-02 -1.12842882e+00
-7.95329139e-02 2.60545284e-01 -3.75132382e-01 6.49168968e-01
-1.70197934e-02 3.76036793e-01 4.24992055e-01 8.42563584e-02
3.91639881e-02 4.25920963e-01 1.23936760e+00 2.48017922e-01
-1.95898995e-01 -3.61780018e-01 -1.06669092e+00 6.19014382e-01
8.22557271e-01 -3.38595659e-01 -3.85154217e-01 -2.51819104e-01
1.58294499e-01 -4.75551724e-01 9.66570020e-01 -1.21093190e+00
1.11342423e-01 -8.54346529e-02 8.14095080e-01 1.56056415e-02
1.05563745e-01 -8.90505373e-01 -1.84404925e-02 8.24507833e-01
-4.06109154e-01 -1.01503395e-01 7.58429587e-01 2.75277019e-01
2.30229422e-01 -2.36311153e-01 1.18383181e+00 -1.50880009e-01
-8.07510257e-01 -1.50591061e-01 -6.08919084e-01 -2.39977747e-01
1.06469381e+00 -5.93540490e-01 -3.17770690e-01 -5.78841157e-02
-8.04399192e-01 -4.28041130e-01 3.81142586e-01 5.84815800e-01
6.01476192e-01 -1.29524744e+00 -4.75620300e-01 4.78593409e-01
2.18816171e-03 -7.74281800e-01 4.38026994e-01 1.37388563e+00
-2.17305899e-01 6.86895669e-01 -8.36159468e-01 -7.33676374e-01
-1.00428355e+00 1.10610557e+00 5.77283323e-01 6.97755575e-01
-1.29597396e-01 8.94608140e-01 6.43919230e-01 -1.11246165e-02
-2.14746520e-02 -2.96978116e-01 -4.90762293e-01 3.20717871e-01
6.42679632e-01 1.82528973e-01 1.25938445e-01 -7.56871045e-01
-4.73079592e-01 3.69253397e-01 1.78478166e-01 -8.61677080e-02
1.13044214e+00 -1.13558017e-01 -2.15956420e-01 4.88564402e-01
7.25940883e-01 -2.55740285e-01 -1.29222465e+00 9.81411263e-02
-6.98938146e-02 -4.51293230e-01 -1.32099958e-02 -1.11264789e+00
-1.12379861e+00 1.16219008e+00 1.17669797e+00 2.50595003e-01
1.24888992e+00 1.56066895e-01 -1.04247026e-01 1.31071776e-01
1.56584591e-01 -9.42632914e-01 1.14842743e-01 -4.97976784e-04
1.49588323e+00 -1.12216282e+00 -2.30435982e-01 6.08252510e-02
-5.38897812e-01 9.33359861e-01 1.02834880e+00 -1.79196194e-01
4.68687475e-01 -3.52747262e-01 -1.28476620e-01 -5.01176357e-01
-7.51795173e-01 -5.55301547e-01 3.62853110e-01 1.11180735e+00
5.39443076e-01 -3.72453302e-01 -2.96577364e-01 3.86594296e-01
-1.55350536e-01 9.79687199e-02 3.53776306e-01 8.82857263e-01
-5.57573140e-01 -4.02791023e-01 -6.41308188e-01 3.43417853e-01
-1.71796963e-01 -1.71672776e-01 -5.12330949e-01 1.08072734e+00
4.30233747e-01 6.35574162e-01 4.11347568e-01 1.38767874e-02
4.07373607e-01 3.45177710e-01 7.54877925e-01 -6.38741076e-01
-7.02008545e-01 1.14425428e-01 -3.07560802e-01 -5.69447935e-01
-7.82526851e-01 -7.91441321e-01 -1.09536457e+00 2.70262118e-02
-3.09314281e-01 -3.04251879e-01 8.16829681e-01 8.37944388e-01
4.70777869e-01 5.62898099e-01 1.11322515e-01 -1.46099234e+00
-5.49117401e-02 -1.23632646e+00 -3.28997880e-01 3.64235073e-01
3.64364713e-01 -1.04592395e+00 -6.11078560e-01 2.86556423e-01] | [10.176288604736328, 2.2160637378692627] |
fc423369-1a6c-4abe-b783-0e8e6c280336 | covlr-coordinating-cross-modal-consistency | 2304.07567 | null | https://arxiv.org/abs/2304.07567v1 | https://arxiv.org/pdf/2304.07567v1.pdf | CoVLR: Coordinating Cross-Modal Consistency and Intra-Modal Structure for Vision-Language Retrieval | Current vision-language retrieval aims to perform cross-modal instance search, in which the core idea is to learn the consistent visionlanguage representations. Although the performance of cross-modal retrieval has greatly improved with the development of deep models, we unfortunately find that traditional hard consistency may destroy the original relationships among single-modal instances, leading the performance degradation for single-modal retrieval. To address this challenge, in this paper, we experimentally observe that the vision-language divergence may cause the existence of strong and weak modalities, and the hard cross-modal consistency cannot guarantee that strong modal instances' relationships are not affected by weak modality, resulting in the strong modal instances' relationships perturbed despite learned consistent representations.To this end, we propose a novel and directly Coordinated VisionLanguage Retrieval method (dubbed CoVLR), which aims to study and alleviate the desynchrony problem between the cross-modal alignment and single-modal cluster-preserving tasks. CoVLR addresses this challenge by developing an effective meta-optimization based strategy, in which the cross-modal consistency objective and the intra-modal relation preserving objective are acted as the meta-train and meta-test tasks, thereby CoVLR encourages both tasks to be optimized in a coordinated way. Consequently, we can simultaneously insure cross-modal consistency and intra-modal structure. Experiments on different datasets validate CoVLR can improve single-modal retrieval accuracy whilst preserving crossmodal retrieval capacity compared with the baselines. | ['Wenjie Li', 'Xiangyu Wu', 'Zhongtian Fu', 'Yang Yang'] | 2023-04-15 | null | null | null | null | ['instance-search'] | ['computer-vision'] | [-1.26929745e-01 -3.80550563e-01 -3.66427302e-01 -1.30894050e-01
-1.30596685e+00 -5.38741112e-01 9.27429855e-01 -2.48998955e-01
-2.49706447e-01 4.28219020e-01 1.01315089e-01 1.78813383e-01
-4.59175378e-01 -3.17163438e-01 -7.09181786e-01 -1.02867794e+00
4.98715192e-01 4.98950332e-01 1.00055598e-02 -1.11985050e-01
2.11835206e-01 9.27887335e-02 -1.80444586e+00 5.04704237e-01
8.79379094e-01 7.88912296e-01 3.59785855e-01 2.30447710e-01
-2.38878891e-01 5.90180695e-01 -3.68364066e-01 -4.22986299e-01
2.29512990e-01 -3.52531672e-01 -8.74845028e-01 -1.11624703e-01
8.07553232e-01 -1.08679853e-01 -3.36699963e-01 1.42019522e+00
6.30312979e-01 1.53764457e-01 5.97685993e-01 -1.46996236e+00
-1.10390687e+00 3.32137316e-01 -8.74194980e-01 -5.93033843e-02
2.61742562e-01 1.89438850e-01 1.25123250e+00 -9.26700890e-01
5.29258192e-01 1.47178113e+00 3.95315588e-01 7.00146317e-01
-1.18393970e+00 -6.21199608e-01 2.68684417e-01 4.14982826e-01
-1.70472574e+00 -5.42033970e-01 8.02558362e-01 -1.76010504e-01
7.91638374e-01 3.38597894e-01 4.07691836e-01 9.80870128e-01
-1.94917396e-02 8.89619708e-01 9.82418299e-01 -5.09582162e-01
-1.64620757e-01 2.35623494e-01 2.37641498e-01 6.77126229e-01
8.62404406e-02 1.60157993e-01 -6.33948982e-01 -1.28561184e-01
4.18341994e-01 4.38354723e-03 -4.68225598e-01 -3.97777915e-01
-1.57020164e+00 5.50293624e-01 5.48784912e-01 4.37306523e-01
-2.18440637e-01 -1.56334543e-03 5.52286804e-01 4.28189427e-01
9.77305248e-02 2.09596992e-01 -1.56170130e-01 3.27605367e-01
-8.10272634e-01 1.85733348e-01 2.94284582e-01 9.45337057e-01
8.81983697e-01 -2.84062684e-01 -4.67684865e-01 1.14247334e+00
6.53113067e-01 6.99674904e-01 7.07150877e-01 -9.23224330e-01
3.93785655e-01 7.38519132e-01 -2.41598070e-01 -1.18303108e+00
-1.82349235e-02 -2.86825687e-01 -1.13680768e+00 -2.21407667e-01
8.09933618e-02 5.49724042e-01 -8.99301946e-01 2.23416185e+00
2.56543905e-01 1.20461598e-01 3.96110803e-01 1.13381410e+00
1.05422103e+00 6.35489583e-01 2.56912448e-02 -4.85608786e-01
1.33102524e+00 -1.18531561e+00 -8.08328509e-01 -1.89863518e-01
3.00349534e-01 -1.02993476e+00 1.15683174e+00 -7.64278648e-03
-1.23286688e+00 -7.05782652e-01 -8.13741922e-01 -1.77284181e-01
-3.13823551e-01 -1.39836311e-01 5.37200212e-01 6.32119328e-02
-1.11260211e+00 -1.64413705e-01 -4.08712804e-01 -2.72453040e-01
1.55346796e-01 1.66382611e-01 -5.65521419e-01 -3.81024629e-01
-1.21130133e+00 8.98690760e-01 6.52205765e-01 3.10107559e-01
-8.35538745e-01 -6.20285988e-01 -6.27252817e-01 -9.27974731e-02
3.60179067e-01 -1.01818657e+00 8.14039171e-01 -1.03931785e+00
-9.61335063e-01 1.26772141e+00 -2.96274066e-01 5.59456274e-02
2.37624958e-01 -6.84615672e-02 -4.76867825e-01 9.09452066e-02
3.63446355e-01 9.02424097e-01 9.86130595e-01 -1.77151000e+00
-5.25623977e-01 -4.60956305e-01 6.86143264e-02 5.51523626e-01
-3.19934577e-01 -2.56220669e-01 -1.26835775e+00 -4.57249045e-01
3.16903234e-01 -9.89906967e-01 3.73972684e-01 -1.38880804e-01
-4.32403713e-01 -4.06894177e-01 6.78069949e-01 -3.15381378e-01
1.07655036e+00 -2.19619298e+00 6.62298441e-01 1.36284068e-01
5.06802574e-02 1.61836579e-01 -5.26238441e-01 3.13273698e-01
-1.23233914e-01 -1.44584179e-01 -1.25527397e-01 -7.21040308e-01
-1.87492892e-02 6.40564978e-01 -6.02754712e-01 4.21001881e-01
8.80147591e-02 1.14513862e+00 -7.59707510e-01 -9.25742269e-01
1.12944499e-01 4.69209462e-01 -3.26702982e-01 3.83701026e-01
-2.19950601e-01 3.56569886e-01 -2.79388756e-01 1.09681463e+00
9.23983455e-01 -4.50108200e-01 1.63930357e-01 -6.75397933e-01
1.24559797e-01 -2.74446994e-01 -1.02259612e+00 2.13633156e+00
-2.01193705e-01 3.45989287e-01 2.94177327e-02 -1.19084036e+00
5.13901055e-01 5.76237440e-02 3.38589549e-01 -1.28884757e+00
-2.07109153e-01 2.15154335e-01 -2.54242241e-01 -5.28151095e-01
5.53759396e-01 -2.54767627e-01 -2.72384030e-03 3.64919722e-01
1.76425353e-02 5.85323721e-02 -4.37361747e-03 4.42372471e-01
3.95407140e-01 -4.74259257e-04 -1.40871227e-01 -1.33340046e-01
6.83013737e-01 1.40908942e-01 4.01462674e-01 8.07113290e-01
-1.82757109e-01 6.45625353e-01 1.16622090e-01 -6.05757907e-02
-5.92247903e-01 -1.21675277e+00 -1.95540383e-01 1.09484375e+00
8.41857195e-01 -2.28634655e-01 -3.64430726e-01 -5.07929146e-01
-3.30736078e-02 2.53548890e-01 -5.85982561e-01 -6.25428259e-01
-3.18392336e-01 -7.14572906e-01 5.94379306e-01 1.17210202e-01
6.84233248e-01 -1.00462127e+00 -5.83625250e-02 -3.48935366e-01
-5.63022852e-01 -9.89337087e-01 -5.76236427e-01 -4.27678553e-03
-4.91221756e-01 -1.10456705e+00 -7.45748699e-01 -1.14568770e+00
5.38040161e-01 7.90385962e-01 1.30210638e+00 4.15072292e-01
-2.19263047e-01 8.11264694e-01 -2.17733070e-01 -1.72421671e-02
-1.02981977e-01 -9.52605382e-02 8.55021030e-02 9.78740677e-03
3.42922479e-01 -3.75137419e-01 -5.93902707e-01 2.06266031e-01
-1.27769327e+00 -3.97139862e-02 6.70801461e-01 1.34070480e+00
9.86290932e-01 2.43691746e-02 4.52068448e-01 -3.31252426e-01
6.84931040e-01 -3.94065022e-01 -4.76244807e-01 1.00717998e+00
-8.07724714e-01 1.92539304e-01 2.68169880e-01 -6.14307821e-01
-9.04686391e-01 -2.90229797e-01 3.27355176e-01 -9.90692437e-01
1.38070405e-01 7.75611281e-01 -4.05619383e-01 -4.75038439e-02
2.51184255e-01 7.42403686e-01 1.50668725e-01 -1.44010440e-01
5.28458714e-01 3.71283799e-01 7.80855417e-01 -7.99823046e-01
8.57433259e-01 6.18122816e-01 -1.45927161e-01 -4.89833951e-01
-9.62476730e-01 -6.35052145e-01 -3.46890122e-01 -1.98447704e-01
8.75914812e-01 -1.15759957e+00 -6.84850574e-01 5.32029152e-01
-1.22984481e+00 5.72537333e-02 -3.25350314e-02 2.69867778e-01
-2.61645049e-01 6.66196108e-01 -2.44068727e-01 -6.45802140e-01
-4.58059311e-01 -1.28107440e+00 1.46793771e+00 3.32164407e-01
2.36581564e-01 -9.74840701e-01 3.49457055e-01 5.88735998e-01
2.58659959e-01 -2.56809443e-01 1.15022421e+00 -3.46372008e-01
-8.02216530e-01 6.20042831e-02 -5.28035581e-01 2.62303948e-01
4.16820571e-02 -2.01670397e-02 -1.06388807e+00 -5.64247370e-01
-1.83497116e-01 -7.09368169e-01 1.04653478e+00 2.40496695e-01
8.49241614e-01 -9.32121575e-02 -2.03114539e-01 5.91925502e-01
1.49666381e+00 -1.08369343e-01 7.28722692e-01 5.00200510e-01
8.43034685e-01 6.23193562e-01 6.91015482e-01 7.14030191e-02
6.53075635e-01 7.69743800e-01 4.67949867e-01 -8.07701796e-02
-2.06488550e-01 -1.69757545e-01 2.35888898e-01 1.11313868e+00
1.53813258e-01 -1.29423484e-01 -7.00437367e-01 6.80425763e-01
-2.30593491e+00 -1.16079307e+00 1.66014314e-01 2.22070527e+00
1.07595575e+00 -3.03873539e-01 -1.62753507e-01 -3.12221676e-01
6.42611742e-01 2.11927816e-01 -5.12328982e-01 7.68986577e-03
-4.80757505e-01 -3.13217521e-01 -1.17914997e-01 5.17312706e-01
-8.48877907e-01 8.88105392e-01 5.72181797e+00 1.00499344e+00
-1.15107727e+00 1.11467004e-01 1.91839904e-01 -1.40645802e-01
-6.34532094e-01 5.71272895e-02 -6.77286863e-01 2.75055289e-01
1.35575563e-01 3.09539046e-02 5.87620199e-01 5.92390478e-01
-2.86420465e-01 -7.54118264e-02 -1.18404198e+00 1.52233148e+00
4.72909451e-01 -1.13948500e+00 5.68272591e-01 -1.30430665e-02
7.51866341e-01 -2.15601120e-02 3.26804757e-01 5.94464839e-01
-9.11750644e-02 -9.46601033e-01 6.25269890e-01 8.91879618e-01
7.38173664e-01 -7.86622047e-01 5.92473269e-01 4.44257170e-01
-1.13685739e+00 1.99245036e-01 -5.26615024e-01 6.81408644e-01
-1.46381056e-03 3.61570865e-01 -1.06861077e-01 9.44490016e-01
8.75340700e-01 7.91840613e-01 -6.64326727e-01 7.12979972e-01
2.96057314e-01 -1.27317339e-01 -9.86162797e-02 5.16545236e-01
1.90862119e-01 -2.76096702e-01 6.37533665e-01 1.10337138e+00
-3.75666469e-03 -1.77227169e-01 2.43259564e-01 1.02258217e+00
2.12628935e-02 -1.53166473e-01 -5.80307961e-01 2.47886288e-03
4.99972463e-01 1.09479713e+00 -7.78714567e-02 -2.47856557e-01
-3.53904188e-01 1.09675479e+00 5.02219200e-01 5.59297383e-01
-8.90230060e-01 -6.50761044e-03 7.43400156e-01 -5.81416547e-01
3.24019529e-02 9.04030800e-02 -1.21856809e-01 -1.58122551e+00
2.80198783e-01 -1.13834906e+00 6.67214930e-01 -9.17552710e-01
-1.79137146e+00 4.60245550e-01 6.68703392e-02 -1.32390976e+00
-1.45159751e-01 -1.94012195e-01 -3.38213503e-01 9.76602435e-01
-2.03030777e+00 -1.58086610e+00 -3.83443713e-01 1.21802318e+00
5.24090767e-01 -3.47150177e-01 8.07293534e-01 4.19322073e-01
-6.41130030e-01 9.59426343e-01 3.66595052e-02 -2.20545113e-01
1.16357088e+00 -8.81785333e-01 -6.28937542e-01 7.50643075e-01
3.15786749e-01 1.06428611e+00 4.46381360e-01 -5.34126699e-01
-1.85613930e+00 -8.69919777e-01 6.48310542e-01 -3.75264108e-01
6.31150484e-01 9.26932469e-02 -1.21440041e+00 4.04522121e-01
3.97984415e-01 4.32947129e-02 4.78674263e-01 1.45337433e-01
-8.28233719e-01 -3.08088958e-01 -7.56053388e-01 6.85892761e-01
9.52703714e-01 -1.12679231e+00 -8.87291729e-01 4.51864570e-01
7.63821125e-01 -2.27405235e-01 -6.47326708e-01 6.48126602e-01
6.10484540e-01 -9.37047720e-01 1.22087717e+00 -6.15042984e-01
4.56198096e-01 -5.02214849e-01 -5.48737764e-01 -9.26859736e-01
-3.50976616e-01 -1.39255792e-01 -1.69477284e-01 1.48477983e+00
1.74129322e-01 -4.03095663e-01 1.86759293e-01 5.12475669e-01
-8.26410204e-02 -3.75854850e-01 -1.09819877e+00 -6.99085474e-01
5.89924641e-02 -2.55045921e-01 3.02588642e-01 1.28079045e+00
-1.61318645e-01 4.60155159e-01 -5.28383911e-01 4.87314552e-01
7.59916008e-01 5.81316173e-01 7.09581614e-01 -8.27438474e-01
-3.70694429e-01 -6.33360207e-01 -9.64022502e-02 -1.01341581e+00
5.87117195e-01 -9.69254255e-01 1.77987680e-01 -1.33974504e+00
8.13969433e-01 -4.67345983e-01 -6.61201358e-01 5.27155757e-01
-3.45301747e-01 2.35062659e-01 2.07426846e-01 8.54873240e-01
-1.19602180e+00 7.86669612e-01 1.32030308e+00 -5.11438131e-01
-6.54375479e-02 -5.28983176e-01 -7.79758215e-01 3.37201655e-01
2.78179973e-01 -2.25813329e-01 -6.50229573e-01 -7.39646435e-01
4.11898285e-01 -6.34499043e-02 6.91881537e-01 -6.23108208e-01
5.71106195e-01 -2.18418777e-01 9.31409225e-02 -7.61691391e-01
4.53909338e-01 -1.03425133e+00 1.56377152e-01 1.63478181e-01
-4.12112981e-01 1.53045729e-01 2.31690511e-01 8.27839613e-01
-6.32483482e-01 4.39187475e-02 6.29920304e-01 -1.08420879e-01
-9.51844871e-01 2.33611375e-01 2.24164084e-01 1.49421990e-01
8.36829066e-01 -8.95868242e-02 -7.48776197e-01 -9.82782245e-02
-4.65539336e-01 5.74137926e-01 5.39320409e-01 6.32322788e-01
7.78455615e-01 -1.65838790e+00 -6.53758347e-01 6.24504350e-02
6.00717485e-01 5.66331595e-02 6.30651295e-01 1.01813829e+00
-5.59654785e-03 4.24130112e-01 -4.38487977e-02 -1.12899208e+00
-1.35272348e+00 8.06433916e-01 4.88383353e-01 -2.97902256e-01
-3.19509834e-01 8.83570135e-01 4.72622871e-01 -4.03760284e-01
4.29382384e-01 2.59529650e-01 -1.30305693e-01 1.78737372e-01
5.57727456e-01 -1.00234032e-01 -2.27549989e-02 -8.38910937e-01
-6.62200570e-01 9.27799881e-01 -3.06262553e-01 -6.11656755e-02
1.01508415e+00 -4.66175079e-01 -6.62646651e-01 5.50617754e-01
1.40985250e+00 -1.95875749e-01 -7.54701257e-01 -5.91249228e-01
-3.55914325e-01 -5.13367057e-01 1.60614960e-02 -6.99702561e-01
-1.26728642e+00 6.24344528e-01 8.72661829e-01 8.61113593e-02
1.36628520e+00 4.32615399e-01 7.18950152e-01 4.89189267e-01
2.92238116e-01 -9.88335192e-01 3.26293409e-01 4.53907132e-01
9.77502823e-01 -1.65678084e+00 -9.74003375e-02 -2.86405589e-02
-7.92403936e-01 6.23693466e-01 8.41684103e-01 1.00401655e-01
4.44499552e-01 -2.06602946e-01 8.86464119e-02 -4.93178248e-01
-9.04432952e-01 -2.79004812e-01 7.98172653e-01 5.79949379e-01
5.37847728e-02 -1.88416034e-01 -3.90880629e-02 1.99346498e-01
3.36679012e-01 -3.22343469e-01 -3.54319513e-01 6.54267073e-01
-9.93540883e-02 -1.08167028e+00 -5.06289005e-01 -7.61811621e-03
-6.04467578e-02 -2.93560714e-01 -5.43498278e-01 7.64769316e-01
1.56474724e-01 8.18499804e-01 2.60816440e-02 -4.48956370e-01
2.86095262e-01 3.88149358e-02 4.84702647e-01 -1.86073527e-01
-4.22713220e-01 3.13556045e-01 -2.72861123e-01 -5.17168283e-01
-9.74167705e-01 -3.58820170e-01 -1.08432341e+00 -2.16796339e-01
-4.58088547e-01 -5.59205338e-02 3.03636760e-01 1.07901716e+00
4.66524243e-01 3.60635608e-01 5.09489298e-01 -6.46299124e-01
-5.46708882e-01 -5.51391780e-01 -4.35018569e-01 7.77874649e-01
5.23666263e-01 -7.28897929e-01 -5.07461190e-01 1.92949418e-02] | [10.87809944152832, 1.3771897554397583] |
9581bd1d-d666-434c-aeee-88f8d1133ab1 | segment-anything-is-a-good-pseudo-label | 2305.01275 | null | https://arxiv.org/abs/2305.01275v1 | https://arxiv.org/pdf/2305.01275v1.pdf | Segment Anything is A Good Pseudo-label Generator for Weakly Supervised Semantic Segmentation | Weakly supervised semantic segmentation with weak labels is a long-lived ill-posed problem. Mainstream methods mainly focus on improving the quality of pseudo labels. In this report, we attempt to explore the potential of 'prompt to masks' from the powerful class-agnostic large segmentation model, segment-anything. Specifically, different weak labels are used as prompts to the segment-anything model, generating precise class masks. The class masks are utilized to generate pseudo labels to train the segmentation networks. We have conducted extensive experiments on PASCAL VOC 2012 dataset. Experiments demonstrate that segment-anything can serve as a good pseudo-label generator. The code will be made publicly available. | ['YuQi Yang', 'Peng-Tao Jiang'] | 2023-05-02 | null | null | null | null | ['pseudo-label'] | ['miscellaneous'] | [ 5.49587250e-01 5.52589178e-01 -3.96305442e-01 -9.20136273e-01
-1.18156600e+00 -9.56509709e-01 4.80568945e-01 -2.45430946e-01
-4.94404882e-01 8.42055738e-01 -6.27481844e-03 -2.20024362e-01
5.98003507e-01 -5.47822773e-01 -8.85330200e-01 -5.18106639e-01
5.26071250e-01 5.82724333e-01 4.24679607e-01 1.70862302e-02
1.99713349e-01 7.26922527e-02 -1.24893296e+00 6.62554562e-01
7.84780502e-01 6.48873627e-01 1.32681161e-01 4.66320068e-01
-3.61915261e-01 7.74371207e-01 -9.68732834e-01 -3.95828754e-01
2.35574931e-01 -5.33647239e-01 -1.18094420e+00 1.54410347e-01
4.33043659e-01 -1.73407152e-01 3.17442328e-01 1.26248467e+00
2.84144729e-01 -7.13125989e-02 5.93833208e-01 -1.27988482e+00
-3.88821214e-01 1.02594173e+00 -3.34269345e-01 -1.54412761e-02
2.80224025e-01 3.67191583e-01 1.00865090e+00 -7.16461241e-01
8.05857897e-01 1.30877972e+00 7.44558871e-01 1.13276994e+00
-1.26816213e+00 -6.84183538e-01 1.29077241e-01 -1.85003519e-01
-1.32747173e+00 -3.74983072e-01 7.89774776e-01 -4.03445512e-01
4.57959384e-01 3.40769112e-01 2.88283408e-01 1.34579206e+00
-4.33314115e-01 1.14640558e+00 1.38816726e+00 -5.32149374e-01
2.44475186e-01 3.68479520e-01 4.46016699e-01 6.63256705e-01
5.90949692e-02 1.19938165e-01 -1.92063957e-01 -1.15754046e-02
6.01503789e-01 -4.84698653e-01 1.76845479e-03 7.90876802e-04
-1.08353138e+00 9.30807054e-01 4.50405359e-01 2.47081950e-01
7.76815936e-02 2.08301857e-01 2.63352841e-01 1.01802997e-01
7.11809397e-01 8.71305704e-01 -5.99562585e-01 1.26126613e-02
-8.07931840e-01 3.81698996e-01 8.89101624e-01 1.14934170e+00
9.62206602e-01 3.38830985e-02 -6.77253664e-01 9.65884984e-01
1.86485797e-01 3.08606654e-01 3.81579578e-01 -1.18533719e+00
3.33880156e-01 5.26677728e-01 1.07431427e-01 -4.41225290e-01
-4.10029322e-01 -3.98289829e-01 -1.98361486e-01 5.54046333e-02
5.84738493e-01 -5.74166596e-01 -1.28660595e+00 1.89816666e+00
3.18650275e-01 1.64162189e-01 4.87286262e-02 8.32556486e-01
1.14393604e+00 6.24182343e-01 4.84655082e-01 2.23537222e-01
1.12276244e+00 -1.34764147e+00 -6.07959092e-01 -4.48390573e-01
8.83036911e-01 -7.22238362e-01 1.41212142e+00 4.57171798e-02
-8.58539402e-01 -7.27020144e-01 -7.26760626e-01 4.71137054e-02
-4.51437831e-01 3.05569291e-01 5.71480751e-01 6.09375894e-01
-1.10226059e+00 3.79778445e-01 -6.51738167e-01 -1.94071636e-01
6.65978193e-01 8.71868730e-02 -5.47926426e-02 -1.14867412e-01
-1.07414067e+00 5.08354247e-01 7.60910809e-01 -4.10638638e-02
-1.07906675e+00 -4.31358844e-01 -8.76596093e-01 -1.79409519e-01
6.24254823e-01 -3.43671769e-01 1.77263594e+00 -1.36029732e+00
-1.35568106e+00 1.39202428e+00 -4.92968671e-02 -2.53735125e-01
5.97223222e-01 -6.94786012e-03 -9.60947871e-02 1.57471597e-01
5.89608550e-01 1.33808076e+00 8.32980990e-01 -1.69981730e+00
-6.13439858e-01 4.90181670e-02 1.96273834e-01 1.02920569e-01
1.00929111e-01 1.23675950e-02 -3.50078791e-01 -8.08219373e-01
3.05950232e-02 -1.14757264e+00 -4.59738642e-01 -4.86293495e-01
-9.25381660e-01 -5.19752681e-01 7.38950729e-01 -2.50681281e-01
8.58714581e-01 -2.16451335e+00 -4.55651224e-01 -1.18686251e-01
-3.56886201e-02 4.75914270e-01 -4.03465092e-01 4.79967333e-03
-1.41340837e-01 3.51479739e-01 -6.53172910e-01 -2.31351942e-01
3.22907828e-02 2.25339010e-01 -2.17691466e-01 8.66929442e-02
4.34455872e-01 1.26935565e+00 -1.09485924e+00 -5.79360008e-01
4.94965501e-02 1.38854664e-02 -4.29429263e-01 5.32372594e-01
-7.08910942e-01 5.21888256e-01 -6.19133592e-01 7.41051137e-01
4.66220796e-01 -3.77407491e-01 -2.69693851e-01 -8.30549672e-02
1.60698146e-01 1.16749696e-01 -6.85162663e-01 1.72079718e+00
-5.50293028e-02 4.25529361e-01 1.03694210e-02 -1.01907599e+00
8.27411950e-01 1.38886377e-01 1.78704113e-01 -5.69301486e-01
3.09524238e-01 3.53546470e-01 -2.66861588e-01 -5.47502220e-01
2.22083986e-01 -4.65285808e-01 -4.24378633e-01 4.53205734e-01
7.87519366e-02 -7.12799370e-01 4.50161695e-01 2.48719141e-01
9.51964855e-01 5.05288303e-01 -2.07120538e-01 -3.32848281e-01
4.78287131e-01 5.56461453e-01 5.09630680e-01 7.75622129e-01
-4.03191864e-01 8.87632668e-01 4.82448280e-01 -1.66307151e-01
-7.92506576e-01 -1.08561814e+00 -1.42289177e-01 1.19963217e+00
1.81435093e-01 -2.12431356e-01 -1.32303786e+00 -1.29621375e+00
-2.50170410e-01 9.97841954e-01 -5.48130572e-01 -5.61309457e-02
-3.69093657e-01 -6.31323576e-01 7.88740456e-01 6.18914723e-01
5.44771552e-01 -1.51506257e+00 -4.04898793e-01 1.22970521e-01
-3.21504593e-01 -1.29174960e+00 -4.79938775e-01 4.94750738e-01
-5.97641349e-01 -1.07498693e+00 -8.51894796e-01 -1.20922267e+00
1.00196004e+00 5.67835830e-02 1.20874393e+00 -1.03811966e-02
-5.05979508e-02 2.86612004e-01 -5.74152470e-01 -4.94889379e-01
-6.59694672e-01 2.72332996e-01 -4.24550653e-01 -1.29973143e-01
3.83767813e-01 3.98833305e-02 -3.29001367e-01 4.00352567e-01
-1.00598300e+00 3.79223347e-01 4.41837698e-01 6.25858903e-01
6.66636169e-01 -2.59105474e-01 7.73342013e-01 -1.83853173e+00
5.82012713e-01 -3.72185081e-01 -5.42808414e-01 -2.76045632e-02
-4.26077008e-01 1.80530861e-01 6.96908951e-01 -4.76072758e-01
-1.19999552e+00 4.43050474e-01 -6.27565801e-01 -6.11695461e-02
-6.52703166e-01 3.35398227e-01 -3.03764671e-01 3.57218944e-02
7.13578045e-01 -1.66514128e-01 -4.50291783e-01 -4.58307505e-01
6.92010164e-01 6.98525012e-01 7.65922725e-01 -9.24675524e-01
7.93303907e-01 2.94122100e-01 -4.26445633e-01 -3.91979188e-01
-1.47242641e+00 -5.39749146e-01 -4.91686225e-01 -1.00611560e-01
1.04046214e+00 -8.36603343e-01 -4.37954590e-02 5.48343778e-01
-1.10154843e+00 -1.09427226e+00 -5.50802231e-01 -1.00333631e-01
-5.22296488e-01 -1.18467852e-01 -7.29435503e-01 -2.29331374e-01
-1.03284970e-01 -1.17523527e+00 1.14154482e+00 5.60737550e-01
-3.05280715e-01 -1.06391311e+00 -5.52278431e-03 6.60463572e-01
2.24989608e-01 3.63250911e-01 5.86266875e-01 -1.07774949e+00
-4.61158127e-01 -2.42025867e-01 -5.71018994e-01 5.81226587e-01
1.80407062e-01 -1.68969065e-01 -1.33582377e+00 -6.58181161e-02
-6.94061592e-02 -7.91406691e-01 7.74780273e-01 2.85210878e-01
1.51564312e+00 -3.72866929e-01 -3.64330888e-01 6.26003325e-01
1.21271586e+00 1.23223580e-01 1.78734452e-01 2.06406787e-01
1.09308803e+00 8.68241608e-01 7.98658013e-01 2.84703039e-02
4.57206219e-01 2.60545880e-01 1.14174142e-01 -3.61488163e-01
-4.65608478e-01 -5.00907123e-01 3.43387872e-02 7.78329074e-01
4.86093700e-01 -2.59102046e-01 -9.79843497e-01 5.99427760e-01
-1.65829611e+00 -5.26859999e-01 -3.81809205e-01 1.62114465e+00
1.35061324e+00 4.78629887e-01 -2.04832464e-01 -2.19217420e-01
8.23646605e-01 2.57268667e-01 -4.64706898e-01 -3.87989700e-01
-1.31790414e-01 3.14707249e-01 4.79024827e-01 5.36536634e-01
-1.30084479e+00 1.60080326e+00 6.60565948e+00 8.89164805e-01
-1.05297244e+00 2.68140167e-01 9.60743785e-01 2.18198478e-01
-4.26016927e-01 2.34013319e-01 -9.17583108e-01 5.43309510e-01
9.71860766e-01 3.34812216e-02 1.09286912e-01 9.68155921e-01
8.58064666e-02 -1.43551573e-01 -1.22664857e+00 7.49398887e-01
4.15499769e-02 -1.10513091e+00 1.23583049e-01 -2.99842566e-01
1.17304683e+00 8.08128063e-03 -2.64549553e-01 5.07240832e-01
7.30182588e-01 -9.31660712e-01 8.58515263e-01 1.66843534e-01
1.11663055e+00 -4.88527030e-01 7.22899854e-01 3.83168072e-01
-8.70190203e-01 2.02242926e-01 -2.01200500e-01 3.96922231e-02
3.30360234e-01 6.15580440e-01 -9.71659541e-01 8.85887370e-02
4.29456353e-01 5.71891427e-01 -8.89716268e-01 9.31856751e-01
-8.06114733e-01 1.10206366e+00 -2.74007887e-01 8.04966018e-02
5.33810735e-01 -9.63410661e-02 1.58905342e-01 1.51315987e+00
-2.97976404e-01 2.16725841e-01 4.63768154e-01 1.13709331e+00
-4.16184187e-01 -1.10535011e-01 -5.81809044e-01 -2.78089970e-01
1.84547633e-01 1.34677327e+00 -1.29506135e+00 -7.05977976e-01
-2.11254969e-01 9.86493826e-01 1.99318692e-01 5.36343932e-01
-8.59666526e-01 -3.61927092e-01 2.84461887e-03 -9.08412784e-02
-6.61953464e-02 1.95041731e-01 -5.91397524e-01 -9.75228786e-01
-3.24062139e-01 -9.07548428e-01 3.26318204e-01 -9.71933663e-01
-1.28306091e+00 5.64411163e-01 9.00619924e-02 -7.25029945e-01
-1.00674972e-01 -4.49526012e-01 -6.50371850e-01 6.63478613e-01
-1.46548963e+00 -1.20325291e+00 -3.88000906e-01 3.68833750e-01
9.62041199e-01 1.90090895e-01 6.70863986e-01 1.66297182e-01
-6.15766764e-01 5.79653382e-01 -3.69626760e-01 3.87724876e-01
6.78665757e-01 -1.56879342e+00 5.25408387e-01 8.04347634e-01
3.60193104e-01 3.22410345e-01 6.98069155e-01 -8.12177479e-01
-5.94346106e-01 -1.36444223e+00 7.89943695e-01 -6.79944277e-01
4.02373999e-01 -5.47582448e-01 -9.76747751e-01 7.40825713e-01
1.30355656e-01 1.25064269e-01 5.12078464e-01 -2.16749325e-01
-1.75711766e-01 1.47365376e-01 -1.25162470e+00 5.24550855e-01
9.41193879e-01 -5.37122488e-01 -6.60203755e-01 5.97973585e-01
1.18071830e+00 -4.00719136e-01 -2.44540378e-01 6.13346457e-01
8.59553143e-02 -6.85051620e-01 7.14658618e-01 -4.30022389e-01
3.60618860e-01 -2.06197515e-01 1.57487631e-01 -1.25129056e+00
2.49608755e-02 -4.84450161e-01 4.79765087e-01 1.68664253e+00
7.30502546e-01 -5.60430706e-01 1.06137407e+00 6.81121945e-01
-3.03800195e-01 -5.50458431e-01 -5.30825138e-01 -6.86355114e-01
5.33242188e-02 -4.87421155e-01 3.89456004e-01 1.11976635e+00
-3.86352181e-01 5.77727675e-01 1.47376031e-01 -1.04031458e-01
6.32303536e-01 -3.61197963e-02 6.44015849e-01 -9.57228720e-01
-1.81233987e-01 -3.40167671e-01 3.47531676e-01 -1.02331674e+00
5.13240874e-01 -1.13952339e+00 8.04488242e-01 -1.55522943e+00
2.34715536e-01 -9.04859781e-01 -2.25413799e-01 9.00479555e-01
-5.22707224e-01 7.05670476e-01 1.68521896e-01 1.23551108e-01
-8.12770963e-01 3.68489146e-01 1.46634221e+00 -3.33000004e-01
-1.20956339e-01 2.44589314e-01 -7.78327882e-01 9.68729436e-01
1.10139060e+00 -7.69745767e-01 -5.28273761e-01 -4.26146179e-01
-1.73041318e-02 -2.25535259e-01 1.78411096e-01 -9.22813773e-01
-2.04781681e-01 -1.69491932e-01 2.99278758e-02 -4.74098772e-01
9.15096328e-02 -4.90905523e-01 -2.60205686e-01 1.36101842e-01
-6.96771741e-01 -2.98201561e-01 2.42093205e-01 2.15724364e-01
-2.68951148e-01 -7.19245374e-01 8.44280958e-01 -5.18496275e-01
-8.12952697e-01 1.61138311e-01 -6.56050667e-02 6.32274508e-01
1.04959548e+00 -3.05670351e-02 -3.15275788e-01 -1.36184216e-01
-6.11376524e-01 4.14660543e-01 5.67050457e-01 3.38894695e-01
3.66930932e-01 -1.02340293e+00 -6.71726167e-01 2.40845587e-02
2.69746989e-01 4.14808184e-01 -2.54612625e-01 2.36869395e-01
-5.78963399e-01 3.94306958e-01 2.48583760e-02 -6.49152935e-01
-9.54740167e-01 5.35577476e-01 1.40923917e-01 -6.01024702e-02
-3.52288812e-01 1.30389440e+00 3.62966418e-01 -8.49143863e-01
2.64353424e-01 -2.61737913e-01 -1.18977472e-01 -1.35745546e-02
2.67247409e-01 -1.83928423e-02 -4.27556932e-01 -5.98327160e-01
-1.45639047e-01 3.74691993e-01 2.57786121e-02 -2.50108123e-01
1.12702835e+00 -5.94354756e-02 2.14248407e-03 5.45401216e-01
1.19471025e+00 -7.09813312e-02 -1.60078084e+00 -1.20025747e-01
3.46691608e-01 -1.22284852e-01 -3.32363605e-01 -1.16243052e+00
-1.07811773e+00 8.19066405e-01 4.77764279e-01 2.69455433e-01
8.75138342e-01 4.20778215e-01 8.91644180e-01 1.31249458e-01
4.89180684e-01 -1.03063571e+00 2.46289060e-01 4.40277278e-01
5.61049700e-01 -1.50676835e+00 -4.94708925e-01 -6.31801963e-01
-8.36337924e-01 6.56812668e-01 1.08213747e+00 -1.25614971e-01
2.54483134e-01 2.70814061e-01 6.63468659e-01 -2.14584649e-01
-3.25139463e-01 -4.85680521e-01 1.58440873e-01 7.26365268e-01
4.69476432e-01 2.20300213e-01 -4.41952497e-01 7.19028533e-01
-2.79440552e-01 -2.02942304e-02 4.91763860e-01 9.22766328e-01
-4.35618937e-01 -1.37610614e+00 -3.40619177e-01 4.12441164e-01
-5.19634724e-01 -1.24110386e-01 -7.07760513e-01 4.31155890e-01
4.16427553e-01 1.05305982e+00 -3.18848580e-01 -1.46644175e-01
1.93392023e-01 2.56369084e-01 2.10830644e-01 -1.26235199e+00
-5.62557459e-01 -3.83302420e-02 2.69878358e-01 -5.67360222e-01
-3.93469512e-01 -4.00023729e-01 -1.64649475e+00 3.46602470e-01
-2.48943135e-01 3.57116431e-01 6.13872468e-01 1.09249425e+00
-5.02809621e-02 5.87354422e-01 5.19710600e-01 -7.16951966e-01
-5.03899336e-01 -1.08515942e+00 -2.03994378e-01 8.75372112e-01
1.39560506e-01 -4.47460741e-01 -3.86031479e-01 4.39037740e-01] | [9.581050872802734, 0.6445053815841675] |
229b078c-591a-49dd-918c-7cfc69b72fce | material-recognition-via-heat-transfer-given | 2012.02176 | null | https://arxiv.org/abs/2012.02176v1 | https://arxiv.org/pdf/2012.02176v1.pdf | Material Recognition via Heat Transfer Given Ambiguous Initial Conditions | Humans and robots can recognize materials with distinct thermal effusivities by making physical contact and observing temperatures during heat transfer. This works well with room temperature materials and humans and robots at human body temperatures. Past research has shown that cooling or heating a material can result in temperatures that are similar to contact with another material. To thoroughly investigate this perceptual ambiguity, we designed a psychophysical experiment in which a participant discriminates between two materials given ambiguous initial conditions. We conducted a study with 32 human participants and a robot. Humans and the robot confused the materials. We also found that robots can overcome this ambiguity using two temperature sensors with different temperatures prior to contact. We support this conclusion based on a mathematical proof using a heat transfer model and empirical results in which a robot achieved 100% accuracy compared to 5% human accuracy. Our results also indicate that robots can use subtle cues to distinguish thermally ambiguous materials with a single temperature sensor. Overall, our work provides insights into challenging conditions for material recognition via heat transfer, and suggests methods by which robots can overcome these challenges to outperform humans. | ['Charles C. Kemp', 'Joshua Wade', 'Henry M. Clever', 'Tapomayukh Bhattacharjee'] | 2020-12-03 | null | null | null | null | ['material-recognition'] | ['computer-vision'] | [ 5.67030013e-01 2.71107882e-01 4.15494978e-01 -3.92100781e-01
-2.49886736e-01 -6.10423744e-01 3.94410670e-01 3.85494493e-02
-5.28779984e-01 4.44588989e-01 -4.46037620e-01 8.81155133e-02
3.47656220e-01 -2.92672813e-01 -5.50593138e-01 -7.20565915e-01
-1.24616034e-01 6.32862568e-01 -1.30457699e-01 -1.05651416e-01
4.77181047e-01 2.01315165e-01 -1.73899925e+00 6.69585615e-02
7.05918372e-01 1.06495607e+00 5.24717093e-01 7.24045217e-01
6.86513603e-01 1.84667304e-01 -6.34450853e-01 2.01690495e-01
6.05141521e-01 -6.39057979e-02 -1.18177652e+00 -1.22578166e-01
3.11361074e-01 -5.18043160e-01 3.16823721e-01 6.23488545e-01
2.95347124e-01 2.78544247e-01 1.08754909e+00 -1.40102041e+00
-8.88262928e-01 3.09193522e-01 -3.29549402e-01 -6.83399260e-01
9.69787478e-01 4.72743869e-01 5.49172819e-01 -3.66375893e-01
1.48306027e-01 1.52023458e+00 9.13123727e-01 8.43265295e-01
-1.29322064e+00 -3.18429112e-01 -1.20636486e-01 -4.71504740e-02
-1.19306672e+00 -4.58587438e-01 6.29657507e-01 -6.26548111e-01
1.14474380e+00 3.58130157e-01 6.41008914e-01 1.23252964e+00
7.58469224e-01 2.81627476e-01 1.70056665e+00 -5.03725946e-01
4.27543074e-01 2.64226168e-01 -6.21843785e-02 3.33008289e-01
4.39591706e-01 3.59793931e-01 -1.92672443e-02 -2.10232839e-01
9.04707491e-01 -5.19794226e-01 -8.90810788e-02 -1.26852140e-01
-1.35295153e+00 -2.44667307e-02 6.13486171e-01 3.83603811e-01
-5.58607996e-01 3.51406455e-01 2.81635839e-02 2.89391786e-01
-2.97453720e-02 1.33278823e+00 -4.10912424e-01 -3.02119970e-01
1.39057368e-01 2.75487313e-03 1.16218936e+00 8.29238296e-01
6.69503033e-01 -4.12714094e-01 3.99213016e-01 8.65158081e-01
3.12455803e-01 8.93557549e-01 3.60930085e-01 -1.14696980e+00
1.96641296e-01 1.91139325e-01 9.82100725e-01 -9.10044134e-01
-7.59046495e-01 6.38090909e-01 -4.05819178e-01 6.53977633e-01
6.07060015e-01 -3.50108355e-01 -1.06796956e+00 1.47287655e+00
2.85249263e-01 -6.39904439e-01 2.92631656e-01 1.57811832e+00
3.75905223e-02 5.86626112e-01 3.86366814e-01 2.08547339e-01
1.72375548e+00 -5.46321034e-01 -5.74521959e-01 -7.71879971e-01
6.66212857e-01 -7.63638735e-01 1.27873874e+00 5.75594604e-01
-6.56675398e-01 -7.08583534e-01 -1.46356297e+00 -1.11241758e-01
-3.39708239e-01 -6.10788651e-02 8.36044252e-01 8.77468228e-01
-1.04156566e+00 8.30401540e-01 -9.49257195e-01 -9.95333374e-01
-4.05898333e-01 4.58528996e-01 -1.38716042e-01 3.92643437e-02
-1.03175354e+00 1.53782129e+00 3.08021545e-01 6.80834889e-01
-4.46262598e-01 -2.65290946e-01 -9.17686224e-01 -5.99023700e-01
-4.03045565e-02 -6.46815956e-01 1.58956575e+00 -1.11937606e+00
-1.95795274e+00 9.82438207e-01 1.64715633e-01 -4.68054190e-02
4.38823164e-01 -7.97219694e-01 -5.75468130e-02 3.72844338e-01
2.82034520e-02 9.70190108e-01 8.84392440e-01 -1.63438797e+00
2.18906313e-01 -2.52429456e-01 -2.53492445e-01 4.16945755e-01
2.14422435e-01 -7.06005394e-02 6.68132842e-01 -9.97618735e-02
4.17977095e-01 -1.41255605e+00 -1.06541499e-01 -7.91297574e-03
-4.35921222e-01 -6.67142943e-02 5.84429860e-01 -5.10664880e-01
-2.06644475e-01 -1.90126741e+00 -1.19247027e-01 3.50119382e-01
-6.47694152e-03 -3.81838292e-01 3.77493277e-02 4.70658779e-01
1.28755331e-01 1.61419794e-01 -2.82614738e-01 -8.96706507e-02
5.05043328e-01 -3.98029983e-02 -1.61322728e-02 7.35415995e-01
2.09954277e-01 7.66965628e-01 -9.95741785e-01 -2.68315434e-01
3.75614285e-01 4.20821935e-01 1.36102796e-01 3.21556866e-01
2.30012447e-01 2.55415142e-01 -2.38255531e-01 7.51021922e-01
7.87518501e-01 2.16067791e-01 3.61607909e-01 -2.43309006e-01
-5.34234978e-02 2.32273430e-01 -8.50407302e-01 1.14329720e+00
-4.33897436e-01 4.84913528e-01 7.17284560e-01 -4.30474818e-01
1.23118162e+00 4.53314811e-01 3.98340821e-01 -8.73393059e-01
4.86648142e-01 5.36428988e-01 -5.58650168e-03 -7.05541193e-01
8.73607039e-01 -6.44829273e-01 -3.94983083e-01 6.73469186e-01
-6.93417847e-01 -1.11005783e+00 -6.73493266e-01 -4.90867913e-01
9.16175067e-01 4.13237393e-01 -1.70419946e-01 -5.91225088e-01
-1.63240343e-01 3.30382362e-02 1.28704518e-01 9.65784252e-01
-7.58551717e-01 4.86381590e-01 -2.74175912e-01 -2.90567696e-01
-1.15279555e+00 -1.13992405e+00 -4.33633942e-03 1.22285771e+00
9.10182774e-01 3.56313616e-01 -8.57457817e-01 1.88051715e-01
1.07911244e-01 6.39216363e-01 -7.36552715e-01 -4.23574924e-01
-4.72254395e-01 -5.09338081e-01 4.31588799e-01 6.23466432e-01
6.03316069e-01 -1.23451805e+00 -1.52786350e+00 -8.24584514e-02
-3.67594630e-01 -1.08332765e+00 -5.59456684e-02 3.80300939e-01
-7.59319127e-01 -1.01608396e+00 -6.40277445e-01 -9.94072616e-01
1.06210864e+00 4.06045467e-01 6.99753106e-01 2.57052761e-02
-5.03539920e-01 8.59453261e-01 -6.02147341e-01 -3.72383475e-01
-6.13029838e-01 -2.12977722e-01 6.91223741e-01 -6.85745537e-01
2.64077544e-01 -3.84453237e-01 -7.36651897e-01 9.49823439e-01
-6.09559178e-01 4.22116369e-02 8.24834049e-01 2.72243142e-01
-2.72487879e-01 -1.75339356e-01 1.07455671e-01 -7.18009174e-02
8.08949351e-01 -2.21998319e-01 -1.02820762e-01 2.89737850e-01
-3.87499124e-01 2.85552740e-01 3.39037269e-01 -7.65491664e-01
-1.20523119e+00 2.77126163e-01 6.07168436e-01 2.16846347e-01
-7.08453894e-01 -1.69355065e-01 5.36325686e-02 -2.26424247e-01
1.10237646e+00 -4.26683813e-01 1.00516200e-01 1.13788461e-02
2.49147162e-01 8.79141808e-01 5.90138376e-01 -1.34563041e+00
6.86278105e-01 2.68962830e-01 -2.73866747e-02 -8.97273958e-01
6.58135116e-02 -1.10131681e-01 -1.05199671e+00 -6.14682734e-01
8.88623297e-01 -4.30344909e-01 -1.29229867e+00 9.80933547e-01
-9.69883978e-01 -9.03005302e-01 1.79320276e-01 7.28597403e-01
-7.52682388e-01 3.62931073e-01 -9.86966431e-01 -1.40196300e+00
-1.56953797e-01 -1.08865881e+00 1.38536859e+00 2.25436449e-01
-9.82068479e-01 -6.63300753e-01 -9.40478295e-02 3.56095344e-01
5.10571182e-01 2.81462908e-01 5.05269587e-01 -4.07550251e-03
-5.03024161e-02 9.98482853e-02 -6.51076809e-03 -1.59030527e-01
6.25737906e-01 -1.18753709e-01 -1.41407025e+00 -4.83336270e-01
2.59201556e-01 -9.94615197e-01 4.27329957e-01 1.42340273e-01
1.91445217e-01 -7.55689740e-02 -5.03741682e-01 -9.31019261e-02
1.01495957e+00 5.26983082e-01 5.92970788e-01 4.83116895e-01
6.77563310e-01 1.25060439e+00 7.33314157e-01 7.15288892e-02
1.08264729e-01 3.56275618e-01 6.77660927e-02 4.75258157e-02
5.88518798e-01 9.56669599e-02 5.64224541e-01 4.01585519e-01
-2.57256895e-01 8.54856744e-02 -9.79554355e-01 1.58708557e-01
-1.39599895e+00 -5.02769887e-01 8.49255472e-02 2.27504444e+00
6.92471206e-01 -4.08924036e-02 -4.09125611e-02 2.04032794e-01
9.28778231e-01 -2.74669111e-01 -6.34334505e-01 -8.48749161e-01
3.19119185e-01 -2.15451255e-01 7.40667343e-01 5.52823365e-01
-1.04515409e+00 7.12470770e-01 7.87707853e+00 -3.52031589e-01
-1.15821886e+00 -4.47652310e-01 6.20329082e-01 3.15389901e-01
4.91820797e-02 -2.35791102e-01 1.06203325e-01 1.20813772e-02
8.04629743e-01 4.00269777e-01 5.34667611e-01 7.02432454e-01
4.80057187e-02 -9.12460208e-01 -1.83564603e+00 9.30087209e-01
-5.57829104e-02 2.86421567e-01 -6.08251572e-01 -2.46277034e-01
1.15138203e-01 -5.53698421e-01 3.90422136e-01 -2.56958697e-02
4.37430471e-01 -1.17863560e+00 9.09044802e-01 2.90548623e-01
5.91671050e-01 9.37711447e-02 4.79782611e-01 1.33742867e-02
-1.06264699e+00 5.73197119e-02 -4.17117476e-01 -6.92862451e-01
1.75970998e-02 2.69374251e-01 -1.61213303e+00 2.62306184e-01
8.77106309e-01 -3.16128880e-02 -2.42278323e-01 6.48529708e-01
-1.14139058e-01 9.55463052e-02 -7.91846752e-01 -4.54679102e-01
-1.88435316e-01 -1.52867690e-01 1.76277906e-01 1.03467119e+00
2.58359432e-01 4.07233417e-01 -8.59708935e-02 1.24363410e+00
6.58467948e-01 -3.10575902e-01 -5.97285926e-01 1.37774751e-01
7.74329722e-01 1.05812085e+00 -1.24725986e+00 -1.31330207e-01
3.50372493e-01 1.60550368e+00 1.35681452e-02 4.89389300e-01
-5.50206780e-01 -6.12740755e-01 4.91713703e-01 -4.97651368e-01
-3.38154733e-01 -8.20228100e-01 -6.28443003e-01 -5.94423056e-01
3.95053387e-01 -1.71227172e-01 -3.56435686e-01 -1.35953414e+00
-1.40532708e+00 2.34038591e-01 2.00040907e-01 -1.05168831e+00
-3.33394483e-02 -9.34946477e-01 -1.94877729e-01 7.79928744e-01
-6.63858175e-01 -7.32550204e-01 -5.68657339e-01 -1.71406761e-01
7.92645365e-02 8.69053662e-01 8.20480347e-01 -4.59179699e-01
-2.14370579e-01 4.11091685e-01 -1.70464054e-01 -2.06901938e-01
1.18420720e+00 -1.30893743e+00 5.61571717e-01 1.66848019e-01
-9.64478612e-01 8.51151228e-01 1.09988213e+00 -9.78578031e-01
-1.78106427e+00 -2.94577718e-01 7.76882917e-02 -6.36028349e-01
2.64021605e-01 -7.02749491e-01 -9.23979223e-01 4.37512606e-01
3.91510010e-01 -5.87000310e-01 3.87783229e-01 1.25194281e-01
-3.36563975e-01 3.09834421e-01 -1.69343925e+00 4.71305817e-01
8.54153931e-01 -7.06233144e-01 -1.09224319e+00 5.13612628e-02
5.33461511e-01 -2.98127264e-01 -9.58802462e-01 2.08935797e-01
1.20214629e+00 -7.15028048e-01 5.20185053e-01 2.22736701e-01
1.83239281e-02 -2.03280970e-01 -2.24736147e-03 -1.63309205e+00
-4.19859916e-01 -6.72589958e-01 7.94348717e-01 7.50293612e-01
2.97467619e-01 -1.08983517e+00 3.79675537e-01 1.88210547e+00
-1.28598318e-01 -2.08931230e-02 -6.08218491e-01 -1.06324267e+00
1.44163132e-01 -2.82043695e-01 3.00523639e-01 9.18263495e-01
9.72917497e-01 -5.72013063e-03 -4.44183275e-02 3.91700596e-01
5.34924448e-01 -1.48258179e-01 3.29146832e-01 -1.23295319e+00
1.03523791e-01 -2.99623221e-01 -2.60349363e-01 -8.47806394e-01
7.28660449e-03 -1.68316424e-01 1.24684155e+00 -1.30952978e+00
8.52647126e-02 -7.93223083e-01 1.49202570e-01 5.47041833e-01
-2.33083144e-01 -8.78560320e-02 1.47094518e-01 4.43083227e-01
-3.09020281e-01 3.77831250e-01 1.03389144e+00 -1.25173181e-01
-3.66634160e-01 -7.28011131e-01 -4.55501944e-01 3.93122315e-01
1.01862156e+00 -4.12833504e-02 -1.28694192e-01 -3.51077497e-01
2.20461443e-01 -3.02270085e-01 5.61729372e-01 -1.37513065e+00
7.40820095e-02 -2.49341890e-01 6.22299314e-01 1.73364937e-01
6.70997977e-01 -1.20247757e+00 2.39344805e-01 7.34837890e-01
-3.52343470e-01 -5.13005517e-02 4.05762643e-01 2.14854985e-01
6.11905396e-01 -5.52055426e-02 5.53084373e-01 -2.42153659e-01
-6.55093610e-01 -7.88305044e-01 -1.07360911e+00 -5.79974532e-01
1.02856934e+00 -4.95445758e-01 -7.58795202e-01 -2.54002780e-01
-4.35736746e-01 2.81778872e-01 1.13485885e+00 4.46898520e-01
5.37230253e-01 -1.23324502e+00 -1.14147246e-01 3.51343825e-02
-4.37499434e-02 -1.75695896e-01 3.48154157e-02 6.63319528e-01
-7.36631155e-01 1.87058210e-01 -5.74779391e-01 -6.73964143e-01
-9.04499114e-01 5.91716111e-01 6.43608034e-01 8.23924065e-01
-1.66007608e-01 8.12274873e-01 1.75390616e-01 -6.08362377e-01
-2.91255340e-02 -1.08419645e+00 5.29588699e-01 -5.28283775e-01
5.14276624e-02 5.54029346e-01 -1.83729827e-01 -5.32833636e-01
-5.96933782e-01 1.00354886e+00 2.04811469e-01 -6.45218372e-01
5.71390927e-01 -2.35116377e-01 -4.63889204e-02 9.76707757e-01
1.04680085e+00 -5.76915622e-01 -1.43601263e+00 2.47491390e-01
-1.80783421e-01 -2.59105176e-01 -7.09983170e-01 -1.03187346e+00
-1.73123524e-01 4.81751770e-01 6.81860685e-01 4.49179739e-01
8.53353858e-01 1.34429932e-01 6.95866525e-01 1.04334927e+00
7.16230035e-01 -1.63582945e+00 3.19888145e-01 2.94563890e-01
1.01043737e+00 -1.25042355e+00 9.00144503e-02 -7.74711967e-01
-6.62251651e-01 1.08828092e+00 1.06285071e+00 -6.40723631e-02
8.27828348e-02 3.01887453e-01 3.98072958e-01 -1.66554287e-01
-3.92359942e-01 2.79566258e-01 -8.09493586e-02 9.40432250e-01
4.52455908e-01 6.49953544e-01 2.94380218e-01 5.17277494e-02
-7.56216466e-01 -3.29803973e-01 5.16430974e-01 1.64249909e+00
-8.49918962e-01 -6.28836393e-01 -1.27870560e+00 1.85776845e-01
1.42718032e-01 4.81456280e-01 -1.03114045e+00 5.74730217e-01
-3.37039828e-01 1.49307811e+00 1.75878763e-01 -9.34220016e-01
2.06638813e-01 3.50393891e-01 1.00696409e+00 -1.62113577e-01
-4.17409152e-01 -3.39749128e-01 2.29997441e-01 -7.16580331e-01
-4.98094082e-01 -8.37375402e-01 -1.45661318e+00 1.23991007e-02
-2.28605717e-01 2.50355810e-01 8.24091971e-01 7.87974119e-01
2.41138563e-01 2.47257594e-02 6.27593338e-01 -1.81477940e+00
-3.66299421e-01 -1.35268879e+00 -6.85631454e-01 5.10291338e-01
4.14920151e-01 -8.25335920e-01 -7.80092299e-01 6.83304593e-02] | [5.753418922424316, -0.6731293797492981] |
5c6e8219-cdf9-47d5-a6be-677489440ae3 | membership-inference-attacks-against-1 | 2302.03262 | null | https://arxiv.org/abs/2302.03262v2 | https://arxiv.org/pdf/2302.03262v2.pdf | Membership Inference Attacks against Diffusion Models | Diffusion models have attracted attention in recent years as innovative generative models. In this paper, we investigate whether a diffusion model is resistant to a membership inference attack, which evaluates the privacy leakage of a machine learning model. We primarily discuss the diffusion model from the standpoints of comparison with a generative adversarial network (GAN) as conventional models and hyperparameters unique to the diffusion model, i.e., time steps, sampling steps, and sampling variances. We conduct extensive experiments with DDIM as a diffusion model and DCGAN as a GAN on the CelebA and CIFAR-10 datasets in both white-box and black-box settings and then confirm if the diffusion model is comparably resistant to a membership inference attack as GAN. Next, we demonstrate that the impact of time steps is significant and intermediate steps in a noise schedule are the most vulnerable to the attack. We also found two key insights through further analysis. First, we identify that DDIM is vulnerable to the attack for small sample sizes instead of achieving a lower FID. Second, sampling steps in hyperparameters are important for resistance to the attack, whereas the impact of sampling variances is quite limited. | ['Naoto Yanai', 'Takayuki Miura', 'Tomoya Matsumoto'] | 2023-02-07 | null | null | null | null | ['inference-attack', 'membership-inference-attack'] | ['adversarial', 'computer-vision'] | [ 1.65925190e-01 2.08092541e-01 -6.99698180e-02 -6.29438534e-02
-6.69471860e-01 -1.17190969e+00 1.00251913e+00 -2.33966112e-01
-2.40314558e-01 6.59897804e-01 2.01774433e-01 -6.78627551e-01
-5.06047271e-02 -1.02203953e+00 -8.39135230e-01 -7.92369723e-01
-1.96850121e-01 1.71472192e-01 -1.95695445e-01 1.13626979e-02
-1.14717409e-01 4.21569884e-01 -7.10683525e-01 -1.51623800e-01
7.58269310e-01 7.51771569e-01 -8.91938448e-01 6.13609195e-01
4.91301954e-01 8.02663684e-01 -1.24923587e+00 -8.85655999e-01
8.46425891e-01 -7.46807158e-01 -5.60746968e-01 -2.54529953e-01
4.43407476e-01 -7.11234272e-01 -4.96600181e-01 1.30874908e+00
7.10292459e-01 -1.79331303e-01 5.07489324e-01 -1.67889082e+00
-8.00132811e-01 9.12764251e-01 -3.58002692e-01 3.05020083e-02
5.10113388e-02 5.01764059e-01 7.25023568e-01 6.08487017e-02
6.30729318e-01 1.13337123e+00 7.30317056e-01 8.07518363e-01
-1.45830846e+00 -1.03411365e+00 -3.24685574e-02 -3.90274972e-01
-1.36083436e+00 -3.13903123e-01 7.14481890e-01 -3.64748001e-01
5.51038444e-01 4.40395743e-01 4.50625330e-01 1.73832238e+00
2.55557895e-01 4.22594428e-01 1.27375901e+00 -4.97267358e-02
7.72281170e-01 2.32119828e-01 5.60331233e-02 3.00981462e-01
5.57244062e-01 4.88817155e-01 -2.24450201e-01 -7.98428297e-01
7.97569156e-01 -1.57963499e-01 -3.91043335e-01 -2.43859887e-01
-6.80532873e-01 1.20036554e+00 2.44947344e-01 1.04411162e-01
-1.56005144e-01 5.14444232e-01 3.19835693e-01 5.71630895e-01
3.49458814e-01 5.49481094e-01 -2.02807814e-01 -1.34282991e-01
-7.44590402e-01 2.06613261e-02 1.06677186e+00 9.62733984e-01
4.59200680e-01 3.48662466e-01 -1.62974522e-01 4.78100441e-02
1.60925210e-01 6.16686225e-01 2.12185875e-01 -7.83203065e-01
5.08890033e-01 9.21045244e-02 -1.22942604e-01 -7.81054318e-01
6.77431189e-03 -4.65075284e-01 -1.13513923e+00 2.83918589e-01
6.63191855e-01 -9.37945247e-01 -7.77873337e-01 2.14685225e+00
2.12761760e-01 2.24138379e-01 2.44086117e-01 5.05657971e-01
4.23155218e-01 4.55638111e-01 5.57548031e-02 9.88692418e-03
1.21940303e+00 -4.40405905e-01 -5.25014162e-01 -1.29780725e-01
6.31121457e-01 -3.48040789e-01 9.51529324e-01 2.21019104e-01
-9.94908094e-01 -1.87085226e-01 -1.12174678e+00 2.65459776e-01
-3.31226975e-01 -3.89355302e-01 6.79050326e-01 1.59186530e+00
-1.07720315e+00 4.50858623e-01 -7.77800858e-01 -2.08912402e-01
7.15147078e-01 3.22234988e-01 -1.88995257e-01 1.74690843e-01
-1.46418715e+00 5.67212284e-01 -1.26846526e-02 -2.52792597e-01
-1.17387033e+00 -9.99393344e-01 -5.81211090e-01 2.19139397e-01
2.16806456e-01 -8.06444347e-01 8.00885260e-01 -9.86257970e-01
-1.52627110e+00 4.51896578e-01 2.03517020e-01 -1.03623068e+00
1.05108750e+00 3.80730443e-02 -3.39439869e-01 -1.64642818e-02
-3.36197704e-01 1.90271795e-01 9.89393771e-01 -1.24733305e+00
-1.31076157e-01 -3.37089509e-01 1.82519674e-01 -2.35445991e-01
-4.27421868e-01 2.49586254e-02 -1.65037122e-02 -8.07760596e-01
-5.59418201e-01 -1.20927072e+00 -1.99350730e-01 -2.49639899e-01
-8.80398989e-01 3.10329169e-01 9.73943770e-01 -3.94231260e-01
1.19040954e+00 -2.30945897e+00 -3.25359434e-01 7.86528170e-01
3.71979356e-01 3.26280743e-01 2.80999858e-02 4.42917764e-01
-9.92243364e-03 8.40529442e-01 -5.47992326e-02 -2.43037522e-01
2.19832659e-01 -1.08089419e-02 -6.09697342e-01 5.73215723e-01
-5.35207354e-02 1.01356018e+00 -3.79080743e-01 1.65844768e-01
-2.68402100e-01 6.26986504e-01 -6.14059508e-01 9.54151899e-02
-2.31262743e-02 4.79763091e-01 -3.07334989e-01 4.39168721e-01
8.27784419e-01 -2.22358093e-01 1.96636438e-01 -6.37275651e-02
4.15092170e-01 6.19069207e-03 -8.66817772e-01 8.98953736e-01
-1.00188486e-01 6.80842817e-01 2.40110997e-02 -4.43971753e-01
7.87643015e-01 2.81620651e-01 -8.38300958e-02 -5.62904000e-01
3.25768918e-01 4.46108617e-02 3.07795435e-01 1.12077326e-01
2.02301234e-01 -1.74057201e-01 -2.68652886e-01 9.65033472e-01
-1.09753296e-01 -4.13918309e-02 -1.61231935e-01 4.13517177e-01
1.32354736e+00 -6.18348598e-01 9.62100178e-02 -3.34428430e-01
-2.75576301e-02 -4.15991336e-01 4.02153671e-01 1.22914076e+00
-2.06475586e-01 4.17059422e-01 9.11800265e-01 -1.13498636e-01
-9.93968129e-01 -1.12347937e+00 7.65684247e-02 5.14738500e-01
2.19424441e-02 -5.14210165e-01 -1.20399284e+00 -1.02115226e+00
1.02155901e-01 9.08583224e-01 -9.69732046e-01 -4.25200045e-01
-2.57043064e-01 -1.10811675e+00 1.28036678e+00 5.59438348e-01
7.61653960e-01 -5.06857872e-01 -3.50169510e-01 -3.52932662e-01
2.45576128e-01 -8.62914860e-01 -7.26975441e-01 -4.54544723e-02
-6.22011304e-01 -9.10032272e-01 -4.63323534e-01 -1.35602251e-01
7.76634097e-01 -2.08351523e-01 9.43883300e-01 -6.73576966e-02
7.83807561e-02 6.18893206e-01 -4.25223224e-02 -6.75282300e-01
-8.42677653e-01 2.37682268e-01 -8.98437351e-02 8.90766829e-02
2.53737122e-01 -6.74156308e-01 -7.11742699e-01 3.77613515e-01
-1.07081628e+00 -5.41551113e-01 2.68373013e-01 6.18676722e-01
1.94828242e-01 2.93070197e-01 5.87435722e-01 -1.33964205e+00
1.06420600e+00 -6.87637150e-01 -7.96757519e-01 1.31182775e-01
-7.72646487e-01 -1.43502980e-01 7.19065547e-01 -7.34230280e-01
-8.26871991e-01 -3.76236171e-01 -2.54641175e-01 -4.04858679e-01
-2.99039446e-02 1.88941002e-01 -5.70527077e-01 -2.75643647e-01
6.50236189e-01 8.34745318e-02 1.10354073e-01 -1.50210470e-01
3.98238271e-01 4.06113982e-01 9.96156186e-02 -6.81982100e-01
1.12007833e+00 6.09920323e-01 3.43343765e-02 -7.03029871e-01
-2.85351068e-01 4.61089224e-01 -2.84635555e-02 1.12880506e-01
5.96463442e-01 -8.51710618e-01 -6.90946281e-01 8.30079675e-01
-6.80628955e-01 -6.25689387e-01 -6.20469332e-01 2.61098385e-01
-4.24426883e-01 2.53465533e-01 -7.74138927e-01 -6.60577297e-01
-5.14943182e-01 -1.09121275e+00 3.69343489e-01 2.59287298e-01
-3.65598321e-01 -1.38761830e+00 -8.94149244e-02 2.06623718e-01
7.56921709e-01 5.55606365e-01 9.88999724e-01 -1.18252933e+00
-6.94779694e-01 -3.94033283e-01 1.63182124e-01 4.40721780e-01
5.79902977e-02 1.01102412e-01 -1.00068510e+00 -6.72741950e-01
3.29126388e-01 -2.81818211e-02 5.82150578e-01 2.56179243e-01
1.15765119e+00 -8.93184662e-01 -8.17102119e-02 1.14441729e+00
1.40973318e+00 4.43562716e-01 9.74310040e-01 3.30110610e-01
5.02121389e-01 1.62502930e-01 -4.22868356e-02 3.34706068e-01
1.06997333e-01 2.94472426e-01 3.95804733e-01 -3.61382306e-01
2.38032907e-01 -6.47366464e-01 4.14121032e-01 1.10449828e-01
3.18367690e-01 -6.65239811e-01 -7.48331487e-01 1.84081271e-01
-1.25434244e+00 -1.03847253e+00 2.14661136e-01 2.35638189e+00
8.72097671e-01 3.05785179e-01 3.60153913e-01 -1.81627218e-02
5.32528162e-01 2.68426061e-01 -6.73755288e-01 -5.55044174e-01
-3.59754175e-01 3.56404066e-01 8.45817864e-01 3.40149909e-01
-9.49941397e-01 7.49080002e-01 7.26025867e+00 8.67848694e-01
-1.10586274e+00 2.23385096e-01 1.24245429e+00 -1.65073603e-01
-5.41802406e-01 1.45799533e-01 -7.61858821e-01 8.56511116e-01
1.09159946e+00 -4.74317163e-01 5.14367282e-01 7.99129725e-01
-2.71053344e-01 2.62061179e-01 -1.24372268e+00 5.67557931e-01
-3.98016311e-02 -1.31335938e+00 1.33519605e-01 5.46270132e-01
9.40254390e-01 -8.67013335e-02 6.20368838e-01 8.34373012e-02
9.71590638e-01 -1.24555540e+00 6.00686073e-01 1.95574492e-01
7.79483140e-01 -1.02520788e+00 7.24596679e-01 2.94437528e-01
-3.89519364e-01 -1.24753483e-01 -1.76665112e-01 2.40298778e-01
-4.95879538e-03 5.52080154e-01 -6.46931946e-01 2.71391124e-01
4.02201414e-01 3.04171722e-02 -5.80590487e-01 5.56182027e-01
-2.79156774e-01 1.24535620e+00 -3.88646960e-01 1.98341399e-01
1.09105535e-01 -1.62817523e-01 4.91493076e-01 9.27872539e-01
4.21888083e-01 -1.16382696e-01 -3.00982177e-01 1.09131062e+00
-4.05732512e-01 -4.02027160e-01 -5.87769330e-01 -3.30585450e-01
7.02042103e-01 8.60389650e-01 -4.47280139e-01 -1.30014956e-01
-1.45309612e-01 9.00193214e-01 -5.21600805e-02 6.62685692e-01
-9.61883366e-01 -2.64250338e-01 8.65733683e-01 1.00344583e-01
2.61980563e-01 -2.36614905e-02 -3.15025687e-01 -1.03721654e+00
-3.58794391e-01 -1.25530922e+00 3.45461935e-01 -3.22300673e-01
-1.53305006e+00 5.28345406e-01 -2.03768760e-01 -9.30214524e-01
-2.18403131e-01 -1.08789057e-01 -9.59226549e-01 1.03320575e+00
-9.96988893e-01 -1.02788639e+00 -3.51925194e-02 7.37213731e-01
-3.55893284e-01 -1.89795092e-01 7.58566797e-01 2.84970980e-02
-7.65642166e-01 1.36182952e+00 4.05423820e-01 6.10504150e-01
5.13427615e-01 -1.07442963e+00 8.26735854e-01 1.16219008e+00
1.38227521e-02 1.21419168e+00 5.30399621e-01 -7.25885868e-01
-1.37183428e+00 -1.18835318e+00 1.73195317e-01 -5.75604677e-01
6.37231290e-01 -6.57067299e-01 -7.46668458e-01 9.85170007e-01
2.01480418e-01 -5.55910021e-02 8.91298413e-01 -1.19488060e-01
-6.03784263e-01 7.87375402e-03 -1.71284533e+00 7.79206753e-01
8.86600196e-01 -7.17556715e-01 1.29899621e-01 2.39055436e-02
7.38716722e-01 -3.39948863e-01 -1.13926017e+00 1.69787593e-02
4.57613587e-01 -9.77900684e-01 8.94696534e-01 -5.36715269e-01
9.99284983e-02 3.13087367e-02 2.42289416e-02 -1.41706216e+00
-8.54659528e-02 -1.30324495e+00 -2.93290555e-01 1.71284473e+00
4.60144132e-01 -1.23915696e+00 7.97291756e-01 9.45710540e-01
8.22476625e-01 -3.39369804e-01 -8.83701682e-01 -9.34494972e-01
5.97863197e-01 -1.29107371e-01 1.10543573e+00 1.09099925e+00
-4.43393558e-01 9.52401385e-02 -4.73498046e-01 2.38410518e-01
6.72191560e-01 -2.63807595e-01 1.12424004e+00 -8.88914526e-01
-4.24335599e-01 -4.17219341e-01 -2.96731174e-01 -8.02007973e-01
-5.56591377e-02 -6.18308902e-01 -5.12558937e-01 -9.06116605e-01
-2.96709072e-02 -5.54638386e-01 -8.31194222e-02 2.15226606e-01
-1.40450120e-01 1.35380283e-01 4.25793290e-01 1.97386220e-01
-4.05926071e-03 2.26250410e-01 8.64135981e-01 -1.77684709e-01
-1.11906089e-01 3.86329442e-01 -1.17887819e+00 4.95178938e-01
1.06367528e+00 -5.57163119e-01 -8.03888440e-01 -2.20418796e-01
3.69092733e-01 -2.02417031e-01 4.44117546e-01 -7.65412927e-01
1.64877519e-01 1.64796039e-02 2.55405396e-01 -3.75883728e-02
4.79932204e-02 -9.01094496e-01 6.75207794e-01 5.31299293e-01
-4.42602605e-01 -4.79040435e-03 2.33403027e-01 6.94431841e-01
2.05338046e-01 4.46076319e-02 7.54132092e-01 1.27691878e-02
-3.83026525e-02 4.65456277e-01 -3.77740204e-01 4.59820390e-01
1.16334260e+00 -1.48487985e-01 -8.91385019e-01 -7.89161980e-01
-5.14692068e-01 -8.69533494e-02 8.73199999e-01 2.02196211e-01
1.72002330e-01 -1.10060561e+00 -6.84509754e-01 6.21977448e-01
-2.89824665e-01 -2.88900405e-01 2.08304599e-01 6.42214298e-01
-2.95547158e-01 -1.05046056e-01 -3.20094191e-02 -1.35744959e-01
-1.20692515e+00 6.41556025e-01 5.82629681e-01 -2.47952372e-01
-3.86796683e-01 9.96494293e-01 3.40241998e-01 -1.15960017e-01
3.64522845e-01 -1.55688763e-01 5.20015717e-01 -1.29160658e-01
3.54698211e-01 3.67921948e-01 -2.38400653e-01 -4.27128285e-01
-2.41718769e-01 2.00306743e-01 -5.19748181e-02 -1.49113923e-01
8.95580113e-01 5.73528670e-02 8.81059766e-02 -3.88685055e-02
1.18154287e+00 4.73527700e-01 -1.24941480e+00 -4.54089269e-02
-5.00890255e-01 -4.81994301e-01 -1.64544970e-01 -9.82711911e-01
-1.36338651e+00 7.08179116e-01 5.28576612e-01 7.29400575e-01
1.09332645e+00 -2.64858842e-01 6.91728294e-01 -6.75302893e-02
2.45116308e-01 -7.73441374e-01 -2.54811525e-01 1.41893283e-01
4.78995562e-01 -7.23338664e-01 -1.64109200e-01 -2.16201276e-01
-7.44064093e-01 5.15232801e-01 3.47758323e-01 -6.82147443e-02
8.51059616e-01 6.95318520e-01 4.24165390e-02 3.51016410e-03
-6.61542714e-01 3.88342917e-01 -4.83877026e-02 8.51310015e-01
-1.71435088e-01 4.38888520e-02 6.08323850e-02 7.92162061e-01
-5.80630839e-01 -2.60297328e-01 7.43547499e-01 8.06242764e-01
2.90957928e-01 -9.72258091e-01 -3.71061653e-01 4.05220121e-01
-8.03685129e-01 -1.47657275e-01 -7.93660104e-01 8.22144985e-01
-5.86339012e-02 9.70552742e-01 1.11032605e-01 -6.00690186e-01
1.06333099e-01 -8.88221897e-03 8.04066658e-02 -9.99025479e-02
-1.04768705e+00 -2.62354702e-01 -3.09203826e-02 -3.64556104e-01
7.56465411e-03 -6.31438792e-01 -6.39400780e-01 -9.04649913e-01
-4.74706441e-01 2.38019586e-01 5.55691540e-01 5.27982414e-01
6.81768477e-01 3.72577459e-01 7.75359869e-01 -3.37897148e-03
-1.08096504e+00 -6.49191380e-01 -6.54038548e-01 3.84454876e-01
1.84074536e-01 -6.39675781e-02 -9.67716098e-01 -2.95391232e-01] | [5.923239231109619, 7.231140613555908] |
e3554d5e-5544-4e6a-8a9f-491e21ae89eb | improved-topic-representations-of-medical | null | null | https://aclanthology.org/2020.nlpcovid19-2.12 | https://aclanthology.org/2020.nlpcovid19-2.12.pdf | Improved Topic Representations of Medical Documents to Assist COVID-19 Literature Exploration | Efficient discovery and exploration of biomedical literature has grown in importance in the context of the COVID-19 pandemic, and topic-based methods such as latent Dirichlet allocation (LDA) are a useful tool for this purpose. In this study we compare traditional topic models based on word tokens with topic models based on medical concepts, and propose several ways to improve topic coherence and specificity. | ['Simon Šuster', 'Timothy Baldwin', 'Karin Verspoor', 'Yulia Otmakhova'] | null | null | null | null | emnlp-nlp-covid19-2020-12 | ['topic-models'] | ['natural-language-processing'] | [-4.08887178e-01 8.40159133e-03 -4.76282060e-01 -2.43458122e-01
-4.61632937e-01 -1.82749312e-02 7.17202365e-01 8.72245967e-01
-5.60903728e-01 9.40031350e-01 7.59446800e-01 -2.69042611e-01
-2.79405892e-01 -8.79179716e-01 2.04837993e-01 -8.66795003e-01
-3.86285365e-01 9.83507276e-01 2.21783087e-01 1.17956184e-01
5.13102353e-01 3.02645527e-02 -1.03008330e+00 -3.26152705e-02
8.97187769e-01 5.69122098e-02 3.73075485e-01 6.17744923e-02
-7.17742026e-01 -2.22120360e-01 -7.14698553e-01 -1.16202131e-01
-3.65883857e-01 -1.42508790e-01 -6.32190883e-01 -3.36266160e-01
-8.64572525e-01 -1.15204282e-01 1.91191196e-01 8.21798861e-01
6.05832636e-01 2.60020703e-01 9.14313257e-01 -1.12361348e+00
-1.12427548e-01 5.89427769e-01 -7.38920987e-01 3.52333397e-01
2.52576143e-01 -3.57840538e-01 7.16018498e-01 -8.28215718e-01
9.48280156e-01 1.39120245e+00 5.97784579e-01 3.75220388e-01
-1.11881280e+00 -8.12514186e-01 6.20401166e-02 3.59243155e-01
-1.45882320e+00 1.18571185e-01 7.74627566e-01 -8.10610592e-01
8.37434292e-01 -2.44114324e-01 9.51474667e-01 1.08492720e+00
6.92911983e-01 6.18312418e-01 9.43084002e-01 -5.18591523e-01
6.11504257e-01 3.01260203e-01 2.66805530e-01 -2.39642952e-02
8.57163072e-01 -5.11186659e-01 -3.44302684e-01 -1.04463685e+00
5.53230464e-01 4.15792733e-01 1.35009870e-01 -2.58940250e-01
-1.21788704e+00 1.64299965e+00 -3.54109824e-01 7.64397442e-01
-7.31821477e-01 -1.29019007e-01 7.63342321e-01 -4.23211992e-01
1.30399787e+00 5.23801744e-01 -1.48647353e-01 -5.08180141e-01
-1.30767500e+00 3.50441515e-01 6.84049785e-01 3.52621436e-01
2.97243983e-01 -2.01354265e-01 3.80277522e-02 8.27281058e-01
6.68397129e-01 2.94335425e-01 8.54335010e-01 -4.01643664e-01
-9.83331129e-02 3.28142226e-01 -8.17333236e-02 -1.04447591e+00
-3.59001398e-01 -1.94507271e-01 -6.62577152e-01 -4.47277158e-01
-9.32210088e-02 -1.64939180e-01 -9.05070484e-01 1.38516021e+00
5.10531664e-01 1.53988808e-01 -3.91033068e-02 4.27004665e-01
6.40957415e-01 8.26048076e-01 8.71975660e-01 -5.80139697e-01
1.69470537e+00 -1.63712144e-01 -1.28625631e+00 3.79930288e-01
6.18092000e-01 -6.85361266e-01 3.65767926e-01 5.31778753e-01
-6.77928209e-01 2.02332154e-01 -4.60803479e-01 2.32243434e-01
-7.08702445e-01 -6.35951519e-01 8.07591796e-01 9.53453839e-01
-1.04562247e+00 -3.96714918e-02 -9.85038936e-01 -7.58980513e-01
4.28479195e-01 3.82656492e-02 4.12053103e-03 -6.75583109e-02
-1.65597081e+00 9.53599811e-01 5.88126361e-01 -4.78184640e-01
-8.16252053e-01 -7.10244358e-01 -6.10417187e-01 2.01405853e-01
9.27969441e-02 -6.50910437e-01 6.08930588e-01 2.50808477e-01
-1.10682511e+00 7.87286758e-01 -4.89618152e-01 -4.84289557e-01
-1.82861656e-01 -3.06522157e-02 -2.20221296e-01 2.65615255e-01
3.20424318e-01 6.73547506e-01 4.54260111e-01 -9.12196696e-01
-5.63968956e-01 -4.25398111e-01 -4.37213510e-01 3.47741023e-02
-6.11220896e-01 6.27876699e-01 -1.52893901e-01 -7.06588984e-01
7.74927214e-02 -7.92598069e-01 -6.07934773e-01 -6.84230387e-01
-4.17841196e-01 -9.82145309e-01 6.41615391e-01 -6.01720214e-01
1.30897331e+00 -1.72462106e+00 -1.20695211e-01 1.47669256e-01
5.26510954e-01 6.22969940e-02 6.32605433e-01 6.97257042e-01
6.70825988e-02 3.09354454e-01 -1.36123151e-01 -1.56107232e-01
-3.49351346e-01 1.82248950e-01 -4.68342483e-01 5.14101982e-01
-2.17356533e-01 4.50688273e-01 -9.93770778e-01 -8.26569974e-01
3.48497517e-02 7.45214522e-01 -5.22690415e-01 -2.33051181e-01
-2.39335507e-01 2.20103741e-01 -7.56788731e-01 4.10926759e-01
4.07956421e-01 -3.91918898e-01 5.45578897e-01 4.68183905e-01
-4.45156008e-01 3.66001695e-01 -4.05440122e-01 1.53116059e+00
-3.30760069e-02 7.08542764e-01 -4.38128769e-01 -9.67966378e-01
9.32375669e-01 9.41012561e-01 1.20264614e+00 -1.54532888e-03
3.86686288e-02 -2.75447190e-01 -7.47930259e-02 -3.57344240e-01
6.13919318e-01 -4.54531014e-01 3.98049206e-02 8.17568243e-01
-1.13484479e-01 3.65971997e-02 3.22244056e-02 3.35717291e-01
5.46541691e-01 -3.63972932e-01 7.51262605e-01 -6.94796801e-01
-2.41185233e-01 2.62402505e-01 4.19611067e-01 2.91472584e-01
5.07861190e-02 1.66121200e-01 6.44684017e-01 -3.94188732e-01
-9.08058345e-01 -7.07728565e-01 -7.12022960e-01 7.87286341e-01
-2.20264077e-01 -6.34086728e-01 -6.35348141e-01 -2.08122134e-01
-1.24649711e-01 8.85929108e-01 -7.60906994e-01 1.41768560e-01
-1.59153387e-01 -1.42022038e+00 2.79580325e-01 1.96106836e-01
1.88149273e-01 -7.88583577e-01 -9.34000731e-01 5.56101322e-01
-4.55166876e-01 -4.83360887e-01 2.07327083e-01 5.97974882e-02
-1.19339597e+00 -7.01562226e-01 -1.49877632e+00 -1.84540942e-01
6.13302112e-01 4.44841646e-02 8.37445617e-01 -3.59176219e-01
-5.64332783e-01 3.52854282e-01 -5.28460085e-01 -9.10750151e-01
-3.09150722e-02 1.62996724e-01 3.12621415e-01 -3.46807271e-01
9.56898332e-01 -4.37139213e-01 -7.24071026e-01 -2.52172761e-02
-9.54080760e-01 -1.87728360e-01 1.92459598e-01 6.58369005e-01
2.49258712e-01 1.06987052e-01 7.36626863e-01 -1.01004755e+00
1.02569735e+00 -1.18328381e+00 -4.34611410e-01 -2.85519306e-02
-1.09446037e+00 -3.94518286e-01 -6.45512342e-01 -3.97738844e-01
-1.05238843e+00 -5.96658945e-01 6.82993932e-03 -3.41258645e-02
-1.80843845e-01 9.18005705e-01 2.94995666e-01 5.68788230e-01
5.37668109e-01 2.07651202e-02 -7.24394619e-02 -6.72052085e-01
2.54456341e-01 7.49815226e-01 -3.94272715e-01 -2.47267097e-01
1.32574454e-01 6.54409885e-01 -1.18580945e-01 -1.19494712e+00
-2.83057123e-01 -1.04992902e+00 -2.38207787e-01 -8.66939425e-02
1.24106634e+00 -9.16344702e-01 -4.85695899e-01 6.95458949e-02
-1.29073155e+00 2.35034838e-01 -1.38129309e-01 1.01918316e+00
-2.08395153e-01 2.94735014e-01 -2.96964884e-01 -9.68829989e-01
-3.12659830e-01 -1.15200138e+00 8.72006774e-01 1.53339192e-01
-6.72762513e-01 -1.59742212e+00 8.08533490e-01 -8.93352851e-02
7.39008844e-01 2.53341645e-01 1.01692545e+00 -1.02623141e+00
3.38639840e-02 -1.75934121e-01 -1.88623875e-01 -4.26166683e-01
4.79858398e-01 -3.03318679e-01 -7.69727707e-01 -5.61283343e-02
2.27107406e-01 3.65672320e-01 9.31657612e-01 1.06650627e+00
6.89864695e-01 -1.57788828e-01 -1.23339462e+00 -2.21983552e-01
1.26371276e+00 5.19019783e-01 6.27598107e-01 2.27813840e-01
1.49340808e-01 8.08435738e-01 5.38775086e-01 8.56113553e-01
4.65599358e-01 6.44790947e-01 -2.12196391e-02 -2.90051103e-02
4.04344171e-01 1.19733058e-01 -2.56816804e-01 7.79350936e-01
-1.26032531e-01 -2.00490668e-01 -1.43310273e+00 1.13312054e+00
-1.51714790e+00 -6.86291754e-01 -4.38307039e-02 1.84352136e+00
1.08168852e+00 -5.37421145e-02 3.83922517e-01 -1.43430859e-01
8.10895264e-01 -1.44116404e-02 -2.50491612e-02 -3.50198150e-01
6.52166530e-02 -8.28029811e-02 1.00918718e-01 3.43712270e-01
-7.79540002e-01 7.89310157e-01 7.68079901e+00 7.25051582e-01
-8.91983926e-01 4.73724157e-01 6.04728103e-01 2.72783190e-01
-5.97740710e-01 4.34614420e-02 -1.01312506e+00 6.17107153e-01
9.22565281e-01 -6.44335985e-01 -7.49659717e-01 9.83274162e-01
4.42137659e-01 -6.67829752e-01 -4.84878242e-01 6.58870518e-01
2.29700893e-01 -1.14925289e+00 1.58459410e-01 5.12469053e-01
8.25153053e-01 -2.55981237e-01 1.02134526e-01 -3.04779131e-02
3.47004235e-01 -5.74775577e-01 -2.80590057e-01 5.23632050e-01
5.79205692e-01 -4.92797911e-01 1.06123447e+00 -2.31076758e-02
-6.35232508e-01 5.42381883e-01 -5.16407490e-01 1.22035503e-01
6.78946018e-01 1.42721713e+00 -1.62113869e+00 2.87808090e-01
6.19100332e-01 3.84628296e-01 9.90478098e-02 1.51809132e+00
-4.24463563e-02 1.03644264e+00 -2.68401086e-01 -1.17644161e-01
2.11964354e-01 -3.55972983e-02 9.17979002e-01 1.24881971e+00
4.20240968e-01 1.63585573e-01 9.41288099e-02 6.89935982e-01
4.72059101e-01 5.37184358e-01 -7.02987075e-01 -3.37022305e-01
3.22696596e-01 8.39983463e-01 -1.41951060e+00 -8.01627517e-01
-1.89672172e-01 3.08794111e-01 -5.68927467e-01 2.93035418e-01
-4.31096286e-01 -1.80969998e-01 5.51505566e-01 2.26787761e-01
1.32172108e-01 -4.14030820e-01 -2.86394626e-01 -8.51697326e-01
-7.82940924e-01 -2.95083404e-01 5.58543205e-01 -3.43834788e-01
-1.01927912e+00 4.40036714e-01 9.05302167e-01 -8.96323740e-01
-4.30962414e-01 8.35566502e-03 -4.29379761e-01 9.27374184e-01
-1.23318326e+00 -8.29760015e-01 1.23635642e-01 2.46016860e-01
6.92167878e-01 -9.85691771e-02 1.18224561e+00 -4.19494957e-02
-3.79766934e-02 -1.64051458e-01 4.58960086e-01 -3.04328203e-01
7.63747871e-01 -1.07415664e+00 2.04009905e-01 1.12144321e-01
-2.28096157e-01 1.16448653e+00 1.06990838e+00 -1.27974963e+00
-5.77804685e-01 -4.32598859e-01 1.36461270e+00 -1.39410853e-01
4.97662157e-01 -2.06531242e-01 -9.20219302e-01 2.85729378e-01
2.55179167e-01 -9.52022672e-01 1.25624084e+00 7.00167835e-01
-1.58183917e-01 4.91576344e-01 -8.97966385e-01 4.22847956e-01
1.71871379e-01 2.06853673e-02 -8.67986798e-01 6.98061585e-01
8.57232869e-01 1.65197700e-01 -8.36231589e-01 1.50258854e-01
4.86937374e-01 -2.14286298e-01 9.64683950e-01 -6.15104079e-01
-3.74716967e-02 2.11233869e-01 1.90172121e-01 -1.29369557e+00
-2.84740835e-01 -4.17047769e-01 3.68408442e-01 8.47212970e-01
1.96897358e-01 -7.32302308e-01 7.65218198e-01 6.59445584e-01
9.11430568e-02 -6.03316188e-01 -9.77589428e-01 -4.84800339e-01
2.27424517e-01 -1.94035634e-01 4.00076360e-01 1.28235245e+00
6.78377926e-01 -6.40394492e-03 -2.54662603e-01 -3.55609775e-01
4.77005392e-01 -3.95621769e-02 2.82182842e-01 -1.78536582e+00
2.24656865e-01 -3.57235372e-01 -3.51274878e-01 -3.50153953e-01
-1.42334431e-01 -3.04427832e-01 9.15056793e-04 -1.75035834e+00
6.41738296e-01 -5.36961436e-01 -2.21660465e-01 3.91844332e-01
-2.29342312e-01 -8.93726479e-03 -3.65357757e-01 5.07788301e-01
-6.23254240e-01 4.24111962e-01 5.10096133e-01 -5.35618933e-03
-4.74395603e-01 -7.36039281e-02 -4.76510286e-01 6.27671182e-01
8.61067712e-01 -9.51123178e-01 -2.86046386e-01 3.83657840e-04
6.17465138e-01 1.99080608e-03 -2.50692010e-01 -4.46019679e-01
1.98520988e-01 -7.76781663e-02 -1.75185874e-02 -7.93817222e-01
3.74906570e-01 -3.73380154e-01 2.69124269e-01 6.12622976e-01
-3.53183657e-01 -7.90784582e-02 2.72439271e-01 6.58016801e-01
-3.24727297e-01 -2.13477835e-01 2.62577921e-01 -1.49241596e-01
-2.12001607e-01 -7.37880766e-02 -1.09766877e+00 -2.42744744e-01
1.03564870e+00 -8.05135369e-02 -9.15938541e-02 -4.87501860e-01
-9.02180493e-01 1.27596229e-01 1.44929782e-01 1.97974637e-01
6.79713488e-01 -9.66565788e-01 -9.17526245e-01 -3.75302702e-01
3.48714739e-02 -1.90679520e-01 4.02940392e-01 1.05977976e+00
-2.10568070e-01 1.26097310e+00 -1.74278185e-01 -5.52199304e-01
-1.14354861e+00 6.55575216e-01 -4.55721289e-01 -6.66715860e-01
-6.39714360e-01 6.42353952e-01 6.68434858e-01 2.91875929e-01
6.76769018e-02 1.13651976e-01 -9.50122118e-01 6.96075201e-01
6.21479154e-01 6.33480906e-01 -2.44975388e-01 -3.82000208e-01
-6.09335244e-01 2.64246285e-01 -2.08344504e-01 -5.28271019e-01
1.36046231e+00 -1.47155300e-01 -3.40863943e-01 7.27887988e-01
7.10347116e-01 -6.36023432e-02 -1.93487912e-01 -2.70177335e-01
4.61637408e-01 -1.41074479e-01 2.45156884e-01 -4.78832394e-01
-2.91423410e-01 9.45783198e-01 4.76237059e-01 5.75211048e-01
7.41371572e-01 2.56733596e-01 4.80180830e-01 -1.08377121e-01
2.86412805e-01 -8.65627289e-01 -2.33046010e-01 2.44499609e-01
6.11473501e-01 -9.04731274e-01 3.66330951e-01 -4.50778365e-01
-4.85235602e-01 8.60591710e-01 -3.00412655e-01 2.00314641e-01
1.22178876e+00 1.73871908e-02 -3.05458009e-01 -7.66860664e-01
-7.80975044e-01 -1.16149738e-01 1.85490027e-01 5.95253468e-01
6.63258135e-01 1.18940793e-01 -1.09906805e+00 5.50399840e-01
1.59214646e-01 4.62480411e-02 2.29192734e-01 8.71968329e-01
-6.17921829e-01 -1.21376610e+00 -6.47575498e-01 5.32052815e-01
-1.00415146e+00 -2.85456777e-01 -6.07143380e-02 5.88827908e-01
-5.00567406e-02 7.77983844e-01 2.01677069e-01 1.46348998e-01
-5.57548702e-01 4.45007265e-01 -1.66065171e-01 -1.02231383e+00
-1.45300195e-01 7.89470971e-01 6.67678714e-02 -1.38484702e-01
-6.63003623e-01 -1.07281756e+00 -9.59358633e-01 1.42001480e-01
-7.69457281e-01 9.05988216e-01 1.28578138e+00 9.30974841e-01
2.98367441e-01 3.26703489e-01 1.91145703e-01 -3.20808828e-01
8.46911073e-02 -1.29372370e+00 -6.73343658e-01 -1.93898231e-01
-2.46111616e-01 -9.17353094e-01 -2.24030599e-01 2.06289336e-01] | [10.112195014953613, 7.285647392272949] |
c77b4966-e17a-4852-a9db-b4811bbdbbf8 | led-a-dataset-for-life-event-extraction-from | 2304.08327 | null | https://arxiv.org/abs/2304.08327v1 | https://arxiv.org/pdf/2304.08327v1.pdf | LED: A Dataset for Life Event Extraction from Dialogs | Lifelogging has gained more attention due to its wide applications, such as personalized recommendations or memory assistance. The issues of collecting and extracting personal life events have emerged. People often share their life experiences with others through conversations. However, extracting life events from conversations is rarely explored. In this paper, we present Life Event Dialog, a dataset containing fine-grained life event annotations on conversational data. In addition, we initiate a novel conversational life event extraction task and differentiate the task from the public event extraction or the life event extraction from other sources like microblogs. We explore three information extraction (IE) frameworks to address the conversational life event extraction task: OpenIE, relation extraction, and event extraction. A comprehensive empirical analysis of the three baselines is established. The results suggest that the current event extraction model still struggles with extracting life events from human daily conversations. Our proposed life event dialog dataset and in-depth analysis of IE frameworks will facilitate future research on life event extraction from conversations. | ['Hsin-Hsi Chen', 'Hideki Nakayama', 'Hen-Hsen Huang', 'An-Zi Yen', 'Yi-Pei Chen'] | 2023-04-17 | null | null | null | null | ['event-extraction'] | ['natural-language-processing'] | [ 6.19507208e-02 4.07188296e-01 -4.05647367e-01 -6.14349067e-01
-6.20024979e-01 -6.19225919e-01 1.26949465e+00 3.06447208e-01
-4.42241818e-01 1.11330962e+00 1.24792016e+00 -1.11823604e-02
2.64709204e-01 -7.23437786e-01 5.64180352e-02 -2.03938946e-01
1.41454756e-01 3.44535142e-01 1.84019178e-01 -4.09408122e-01
-7.62008457e-03 1.56355590e-01 -1.48507297e+00 4.46769118e-01
3.72542351e-01 6.29826546e-01 -1.92863882e-01 7.77017832e-01
-5.26093602e-01 1.16822863e+00 -8.08104992e-01 -7.84239352e-01
-3.46874475e-01 -6.98240161e-01 -1.28167057e+00 -1.35052890e-01
-4.16087687e-01 -4.69093472e-01 -4.15126204e-01 3.66087526e-01
4.72653866e-01 5.35116553e-01 5.89954138e-01 -1.72562683e+00
-2.25602344e-01 9.48977649e-01 1.02967240e-01 4.78025317e-01
1.03135669e+00 -1.09652914e-01 1.13150907e+00 -6.23652756e-01
9.73484099e-01 1.24427307e+00 5.57825923e-01 5.28569400e-01
-6.74698472e-01 -6.20962262e-01 8.41857716e-02 2.36142531e-01
-9.09583271e-01 -6.72452092e-01 6.68165982e-01 -4.55732673e-01
1.53856361e+00 3.23280245e-01 7.98719645e-01 1.79520428e+00
2.78321236e-01 9.12094116e-01 6.67489588e-01 -3.49230975e-01
-3.22278254e-02 3.17552954e-01 8.60110581e-01 1.65068999e-01
-3.45725901e-02 -6.47259802e-02 -9.55359936e-01 -6.01791203e-01
2.28736207e-01 3.00046653e-01 -1.24215387e-01 5.93394578e-01
-1.27948248e+00 7.87470818e-01 -2.43527249e-01 3.54899913e-01
-4.52679813e-01 -2.01378658e-01 7.04389095e-01 3.85968894e-01
8.17654550e-01 3.07044953e-01 -5.53785145e-01 -7.05434144e-01
-3.99812520e-01 5.36424875e-01 1.89341891e+00 1.04791653e+00
5.24824083e-01 -8.23510885e-01 -2.72378534e-01 8.38617265e-01
2.66990781e-01 1.70126650e-03 3.54718745e-01 -7.74052620e-01
4.65222776e-01 7.06379592e-01 4.52825695e-01 -1.02033114e+00
-7.79145181e-01 1.27076894e-01 -3.43681186e-01 -7.60221303e-01
5.51306367e-01 -6.64641440e-01 5.15821576e-02 1.66420317e+00
6.42005086e-01 3.58967662e-01 2.26121858e-01 3.65735143e-01
1.49706280e+00 7.02730954e-01 2.98121184e-01 -7.77409852e-01
1.89112544e+00 -8.97188246e-01 -1.28861403e+00 -3.10084939e-01
4.32996869e-01 -7.43968129e-01 7.74881721e-01 9.31726694e-02
-8.75270605e-01 -8.18281993e-02 -8.20349574e-01 -3.43028784e-01
-5.65687120e-01 -3.48802596e-01 9.25807416e-01 3.80001605e-01
-3.59962612e-01 2.47978628e-01 -7.48117805e-01 -1.05865753e+00
-5.71535528e-02 -9.10131633e-02 -1.64334729e-01 4.84771192e-01
-1.72149146e+00 8.23997796e-01 3.85442585e-01 -5.83753705e-01
-3.95054430e-01 -5.36172152e-01 -9.44494903e-01 -4.61196788e-02
6.23896241e-01 -6.33070529e-01 1.73328674e+00 -3.65387857e-01
-1.41267037e+00 1.12133253e+00 -3.76115322e-01 -5.80393493e-01
1.48527250e-01 -3.40917170e-01 -8.85704815e-01 -9.25075710e-02
1.13991454e-01 2.23399356e-01 3.74418169e-01 -4.49325562e-01
-9.34808195e-01 -2.69534379e-01 2.83529162e-01 2.44138733e-01
-6.34707630e-01 5.54341197e-01 -4.30063099e-01 -7.37989366e-01
-5.88045001e-01 -1.07650709e+00 9.82539952e-02 -5.27207017e-01
-3.18942606e-01 -8.97144675e-01 7.05677509e-01 -5.99243462e-01
1.68243349e+00 -2.11459422e+00 -1.55613020e-01 -5.50728500e-01
2.95812994e-01 -3.49160314e-01 4.55054551e-01 1.14708316e+00
2.58291811e-01 3.61251421e-02 2.65600234e-01 -7.65379012e-01
1.95257753e-01 2.92819947e-01 -5.53223312e-01 5.78731708e-02
-9.82806087e-02 1.05983162e+00 -1.15122283e+00 -8.93981934e-01
1.43000931e-01 4.42267358e-01 -3.46762180e-01 4.76161540e-01
-4.46950734e-01 5.68471432e-01 -4.46275383e-01 2.49812216e-01
-1.67289093e-01 -6.52374864e-01 3.30908835e-01 -2.54148930e-01
-7.69269615e-02 1.09324634e+00 -7.12267101e-01 1.72940576e+00
-4.05693144e-01 9.01787996e-01 -2.07501128e-01 -3.21138501e-01
4.28439647e-01 9.24148262e-01 6.66300178e-01 -4.74493466e-02
3.37934822e-01 -4.05133754e-01 -4.78290856e-01 -7.05856681e-01
8.68273497e-01 -8.76874924e-02 -8.06111932e-01 9.97793913e-01
3.20355803e-01 3.50147128e-01 3.51621032e-01 4.56758946e-01
1.21498454e+00 -1.16656341e-01 1.13404572e+00 1.63176119e-01
4.41798121e-01 7.32800085e-03 5.65729618e-01 5.49385846e-01
-6.61093593e-01 2.37513080e-01 3.07946652e-01 -2.59709507e-01
-3.23476940e-01 -8.18877041e-01 -9.57655832e-02 1.45841312e+00
2.18190461e-01 -9.13875818e-01 -5.02865374e-01 -7.70015955e-01
-2.21711308e-01 8.75072420e-01 -3.84936869e-01 1.46125913e-01
-7.94348180e-01 -1.01926792e+00 7.03089833e-01 3.33443999e-01
5.69929898e-01 -1.37792492e+00 -4.42306399e-01 6.77268028e-01
-9.82555866e-01 -1.59102857e+00 -5.82380474e-01 -1.67728081e-01
-2.24867910e-01 -1.04035592e+00 -1.71146482e-01 -5.76876044e-01
-2.15674981e-01 1.80776402e-01 1.51765907e+00 -3.63581121e-01
1.08107794e-02 5.97757459e-01 -5.92778623e-01 -5.23872733e-01
-4.97383416e-01 3.28242779e-01 -7.33559057e-02 -1.69198275e-01
1.06952095e+00 -8.73668015e-01 -5.72056651e-01 3.79510581e-01
-5.68524241e-01 1.15334697e-01 -1.66386694e-01 5.66434085e-01
-1.69592395e-01 -1.35129616e-01 1.16766405e+00 -1.13369787e+00
9.85507011e-01 -1.11979234e+00 2.93456495e-01 7.28842989e-02
-4.00079161e-01 -4.85192239e-01 2.50914991e-01 -4.75216299e-01
-1.71353805e+00 -2.60285050e-01 -3.51530105e-01 5.10880351e-01
-3.87797087e-01 5.40662229e-01 -2.52113372e-01 7.37036765e-01
6.03632987e-01 -2.35809058e-01 -4.97417361e-01 -4.37687337e-01
5.40072203e-01 1.00276518e+00 6.21618390e-01 -3.45558256e-01
1.10587232e-01 5.65589905e-01 -7.54145503e-01 -9.62099254e-01
-1.22980356e+00 -8.45817626e-01 -2.25797877e-01 -3.66062313e-01
1.04386425e+00 -8.72963309e-01 -9.73709643e-01 4.19998854e-01
-1.40439415e+00 -7.32207969e-02 -4.52579498e-01 3.64583969e-01
-3.50379169e-01 2.77383387e-01 -1.14136028e+00 -1.00923002e+00
-5.63789010e-01 -3.77507895e-01 9.05182421e-01 4.63559836e-01
-1.26826298e+00 -1.13105738e+00 3.02900612e-01 5.26611626e-01
-4.21611480e-02 4.59627867e-01 3.83950621e-01 -1.25186121e+00
-1.39926478e-01 -2.22330615e-01 1.11015879e-01 -5.41448712e-01
4.36899811e-01 -3.36369991e-01 -1.17234266e+00 -5.54399602e-02
5.94767071e-02 -2.96209514e-01 6.14000857e-01 -2.09924998e-03
3.70282412e-01 -6.32772863e-01 -7.07316995e-01 1.27309486e-01
5.71868956e-01 4.78343695e-01 4.12820429e-01 1.58992037e-01
2.88067430e-01 9.51774597e-01 6.03191674e-01 9.03377831e-01
9.39608693e-01 5.17961383e-01 -1.83970138e-01 4.78745550e-01
-7.65350834e-02 -4.46188897e-01 4.92497355e-01 8.61527026e-01
3.19994867e-01 -1.03141952e+00 -7.59935915e-01 8.28038931e-01
-2.04024935e+00 -1.13703907e+00 4.89493832e-02 1.49792302e+00
1.13225365e+00 1.59957826e-01 3.95827770e-01 1.49367884e-01
6.88123345e-01 4.30660278e-01 -3.66414696e-01 4.98822294e-02
1.59140721e-01 -2.19481260e-01 -1.37897357e-01 3.80713671e-01
-1.24392402e+00 7.65266716e-01 5.83539438e+00 3.83132637e-01
-5.75727046e-01 5.03078043e-01 4.31016892e-01 -1.26537710e-01
4.15725745e-02 1.21540517e-01 -1.43119884e+00 5.62358439e-01
1.30871582e+00 -6.95506871e-01 1.20592862e-01 6.62656009e-01
1.44818544e-01 -3.69685590e-02 -1.57698596e+00 1.18726122e+00
1.54453544e-02 -1.29096937e+00 -2.85966575e-01 -9.19998214e-02
4.16321397e-01 -1.32387146e-01 -7.74203777e-01 5.72149158e-01
6.80674970e-01 -5.60310245e-01 2.38978893e-01 5.83432317e-01
4.44752008e-01 -5.95750451e-01 5.00093699e-01 5.79291821e-01
-1.33817005e+00 8.38789493e-02 4.87320244e-01 -4.13392693e-01
1.01287711e+00 7.19903708e-01 -1.14811158e+00 2.66766489e-01
6.60978258e-01 1.10768199e+00 -1.43325120e-01 3.67984772e-01
-8.73290598e-02 8.63259315e-01 -3.51902723e-01 -3.44481677e-01
-4.12348181e-01 1.77047607e-02 8.52179885e-01 1.52382600e+00
-6.08861968e-02 5.51637530e-01 3.63517880e-01 7.11589754e-01
-2.73139238e-01 -6.34199008e-02 -6.55140281e-01 -6.11199498e-01
7.06153870e-01 1.36175811e+00 -7.63247490e-01 -4.38128799e-01
-6.94031954e-01 1.12125623e+00 2.14014918e-01 7.59796053e-02
-6.62179828e-01 -3.85906219e-01 9.57996905e-01 -4.33420576e-02
-3.53819966e-01 -5.21309562e-02 5.77240484e-03 -1.40651035e+00
-1.72796667e-01 -5.14196873e-01 8.87065828e-01 -4.67397422e-01
-1.72072256e+00 6.40316546e-01 4.40768480e-01 -7.08732784e-01
-1.11523771e+00 9.88993421e-02 -9.73041773e-01 3.26308161e-01
-9.28894222e-01 -1.06555378e+00 -4.86899942e-01 5.24216592e-01
9.10032332e-01 1.30451918e-01 9.06431258e-01 5.10725200e-01
-6.07679129e-01 4.56956953e-01 -5.83911717e-01 3.17860663e-01
9.90156949e-01 -1.04930079e+00 7.98454762e-01 4.37735915e-01
-3.54579762e-02 7.43928790e-01 8.07571232e-01 -9.07590032e-01
-1.21757889e+00 -1.03482163e+00 1.69919300e+00 -7.67931938e-01
7.62416601e-01 -5.02312005e-01 -7.45325148e-01 1.12173760e+00
5.64946890e-01 -5.96424639e-01 1.12360787e+00 3.51870418e-01
1.24683440e-01 3.12102348e-01 -1.20655751e+00 7.89275646e-01
1.43503141e+00 -7.85963118e-01 -1.01787519e+00 4.20231998e-01
1.16714108e+00 -1.12199776e-01 -1.10402167e+00 1.77873269e-01
6.56695366e-01 -7.23782420e-01 1.04826403e+00 -6.32025898e-01
2.25540578e-01 3.32494646e-01 1.91357613e-01 -9.47905958e-01
2.42942348e-01 -1.14894485e+00 -6.85069740e-01 2.14290357e+00
2.83098489e-01 -7.38096774e-01 4.81707633e-01 9.87422287e-01
1.56009734e-01 -9.65229198e-02 -7.34227359e-01 -3.42386216e-01
-5.07374346e-01 -6.12960458e-01 6.12515271e-01 1.23079455e+00
6.40027165e-01 1.23850095e+00 -4.98759687e-01 -2.26654276e-01
1.93709537e-01 1.59983903e-01 9.56909537e-01 -1.60332131e+00
-3.37722689e-01 6.46523759e-02 1.73796490e-01 -1.16060364e+00
4.06764716e-01 -6.15606785e-01 -2.58373935e-02 -1.50228596e+00
3.97215068e-01 2.75401417e-02 2.71723717e-01 2.98320502e-01
-2.61619598e-01 -8.33317563e-02 -1.29441246e-01 3.27653915e-01
-9.08574283e-01 4.54096138e-01 7.73067236e-01 2.40834549e-01
-5.74856341e-01 1.14705466e-01 -6.51853800e-01 1.10551953e+00
6.36696458e-01 -3.08168441e-01 -5.27745962e-01 4.56663311e-01
6.93002880e-01 7.43075669e-01 2.27935776e-01 -6.13321066e-01
3.86506468e-01 -1.63134441e-01 -2.07767278e-01 -8.60705674e-01
4.85537469e-01 -5.97196162e-01 2.79752493e-01 -2.20086113e-01
-5.03636003e-01 -1.77008942e-01 -1.73001215e-02 8.62324953e-01
-1.05819270e-01 5.53463101e-02 4.39088345e-01 -1.40080437e-01
-6.90836906e-01 2.44696718e-02 -8.29897881e-01 4.21064496e-01
9.36208189e-01 8.09435099e-02 -5.13895810e-01 -9.12818909e-01
-1.27039337e+00 3.06227565e-01 2.12804213e-01 8.27355325e-01
1.74555928e-01 -1.21623766e+00 -6.26160681e-01 -2.21940026e-01
2.08412260e-01 -3.33128542e-01 -5.44083454e-02 6.04895890e-01
2.25256264e-01 2.84873039e-01 2.66038895e-01 -1.63908392e-01
-1.50058496e+00 4.27999228e-01 -4.62944731e-02 -6.67186677e-01
-8.44049513e-01 6.82233036e-01 1.42804191e-01 -3.26552957e-01
4.46644604e-01 -3.10080171e-01 -7.39334106e-01 6.26410723e-01
9.70188916e-01 5.56496501e-01 -2.15818703e-01 -4.98191148e-01
-3.39121282e-01 -4.39949125e-01 -1.05619982e-01 -3.85216743e-01
1.28904939e+00 -7.92372465e-01 -2.25711867e-01 9.91878569e-01
1.17576063e+00 -9.08492208e-02 -8.30380321e-01 -3.99050772e-01
5.05865812e-01 -1.04089633e-01 -2.90970832e-01 -6.59909844e-01
-3.01986068e-01 1.60480723e-01 7.86373094e-02 7.71239340e-01
9.03295279e-01 2.88471133e-01 1.74782193e+00 3.82351071e-01
4.73955333e-01 -8.81803453e-01 3.72974016e-02 7.51973152e-01
1.02793193e+00 -1.39002419e+00 -1.71639219e-01 -7.30999887e-01
-4.62014586e-01 8.33514035e-01 4.86617416e-01 2.12029159e-01
1.06801593e+00 4.99068975e-01 -2.92153478e-01 -4.08646464e-01
-1.08324015e+00 -3.47256929e-01 2.52677590e-01 1.29139081e-01
7.34174252e-01 -1.67121172e-01 -6.95872307e-01 1.30706275e+00
-3.97951692e-01 1.63963631e-01 2.87866116e-01 1.06825745e+00
-1.14219122e-01 -1.09896779e+00 -6.42016977e-02 3.86267602e-01
-8.32105517e-01 -7.74432272e-02 -9.41590428e-01 5.66295087e-01
-9.91357043e-02 1.65648723e+00 2.03354999e-01 -3.15912068e-01
1.54302076e-01 5.66937625e-01 8.28904063e-02 -7.32722998e-01
-1.07905054e+00 -1.51192024e-01 1.07332575e+00 -4.40915674e-01
-8.29312027e-01 -8.45191479e-01 -1.34702575e+00 -5.45902610e-01
-3.48316193e-01 2.99309611e-01 2.02339664e-01 1.22935891e+00
4.52951252e-01 4.76040721e-01 3.43362838e-01 -6.57628775e-01
1.86714515e-01 -1.12751722e+00 -2.08862096e-01 5.64289033e-01
2.08158568e-01 -5.88426530e-01 -5.45964360e-01 3.19831580e-01] | [9.188844680786133, 9.263701438903809] |
4c9523e0-66bd-4419-a835-5c08127dcf6e | measurement-of-individual-alteration-in | 2302.05177 | null | https://arxiv.org/abs/2302.05177v1 | https://arxiv.org/pdf/2302.05177v1.pdf | Measurement of Individual Alteration in Perioperative ECGs During Elective Percutaneous Coronary Intervention | The increasing availability of wearable electrocardiography (ECG) devices enables the continuous monitoring of individual ECG alterations. This could be beneficial for patients suffering from acute ischemia but with non-standard ECG findings that do not fit to the subject-independent and absolute thresholds defined in clinical guidelines. In this work, we evaluate the inter-patient magnitude of individual ECG alterations during ischemia. The freely available STAFF III database provides 12-lead ECG recordings of patients before, during, and after elective percutaneous coronary intervention(PCI), where a coronary vessel is widened with a balloon inflation. We compute individual alterations of ST-interval and T-wave amplitudes w.r.t. QRS amplitude over time for each patient and lead. We demonstrate that determining relative ST-interval/T-wave amplitudes and deriving individual alterations over time is feasible in standard and non-standard ECG recordings. To demonstrate clinical relevance, we use the features for differentiating N=54 STAFF III patients with atherosclerotic plaque in either the right coronary artery (RCA) or left ascending artery (LAD). Results show significant differences in 5 leads for ST-interval alterations and 3 leads for T-wave alterations, which are also suggested by clinical guidelines for ischemia detection. Clinical relevance- Assessing individual alterations over time could eventually close the gap in ECG evaluation of patients presenting with pre-existing heart conditions and non-standard ECGs. | ['Nicolai Spicher', 'Tim Kacprowski', 'Dagmar Krefting', 'Henning Dathe', 'Ennio Idrobo-Avila', 'Theresa Bender', 'Philip Gemke'] | 2023-02-10 | null | null | null | null | ['electrocardiography-ecg'] | ['methodology'] | [ 5.16197979e-01 -5.20801544e-01 -1.12615280e-01 -1.29207909e-01
-7.78430223e-01 -1.30415344e+00 -3.48286539e-01 6.93561137e-01
-4.49033707e-01 7.11716294e-01 -8.19519013e-02 -1.03620493e+00
-5.83209157e-01 -3.33621055e-01 -8.05640966e-02 -4.17701960e-01
-8.08450937e-01 4.64529842e-01 -7.75029976e-03 3.29991847e-01
1.95980947e-02 9.62496758e-01 -6.89013422e-01 2.70046026e-01
6.62764788e-01 9.64846194e-01 -2.04120904e-01 1.21491432e+00
5.53275466e-01 -1.84740573e-02 -7.20381856e-01 1.14840001e-01
2.79665083e-01 -1.21726477e+00 -3.18651140e-01 -1.10195227e-01
3.95999849e-02 -2.70885617e-01 2.65651077e-01 4.20582235e-01
1.18352342e+00 -5.76715231e-01 6.34777784e-01 -4.95115072e-01
4.07409936e-01 5.46587050e-01 -4.57742840e-01 9.77871537e-01
1.05822600e-01 3.48586261e-01 3.92162919e-01 -5.93311667e-01
4.63103384e-01 2.27380320e-01 1.19789064e+00 -2.88314968e-01
-1.37934566e+00 -5.96983060e-02 -5.02187014e-01 2.24146828e-01
-1.60171831e+00 -3.93331572e-02 8.02283287e-01 -2.60132939e-01
5.59816718e-01 7.73573339e-01 1.22117305e+00 4.79700595e-01
5.11467576e-01 -3.17787647e-01 1.00486040e+00 -5.56018651e-01
-2.97857635e-02 -3.32927316e-01 2.49738932e-01 -1.46042686e-02
8.76862347e-01 1.50857106e-01 2.45544463e-02 -4.93468225e-01
9.79464650e-01 1.61986444e-02 -3.91480267e-01 -2.02721104e-01
-1.60967660e+00 3.04030031e-01 -1.38254970e-01 5.59383750e-01
-6.51276171e-01 -7.17453733e-02 8.46686780e-01 4.69141781e-01
-8.23531747e-02 3.91700655e-01 -6.51606083e-01 -5.81017792e-01
-8.54830682e-01 -1.42852798e-01 4.25066054e-01 2.70973831e-01
2.84953624e-01 -6.26295581e-02 -2.67220348e-01 3.87619108e-01
2.07547575e-01 4.83067304e-01 6.18112326e-01 -1.10944009e+00
2.13436425e-01 6.02015376e-01 4.06444252e-01 -7.93064177e-01
-6.35535359e-01 -9.58459496e-01 -1.00096548e+00 2.21679285e-01
7.26879120e-01 -2.66912460e-01 -3.78413588e-01 1.20002532e+00
2.68485039e-01 -3.17744836e-02 9.30856764e-02 9.86246228e-01
3.53522688e-01 -2.02904910e-01 2.21592247e-01 -9.51236010e-01
1.61124539e+00 2.20727727e-01 -3.67993236e-01 -1.17640018e-01
1.13142574e+00 -5.60054958e-01 7.36905396e-01 5.80261350e-01
-1.10271525e+00 -5.29241860e-01 -1.08373785e+00 4.39675778e-01
3.20288450e-01 3.73066068e-01 -4.81683575e-02 1.26161110e+00
-6.75970137e-01 9.66434658e-01 -8.70310843e-01 -2.69007027e-01
2.22923517e-01 2.58157581e-01 -2.26441458e-01 3.50030065e-01
-1.18914592e+00 8.86530638e-01 7.79419914e-02 3.53394061e-01
-1.21526472e-01 -9.10528660e-01 -4.92035091e-01 -2.52817214e-01
1.21730432e-01 -8.80815327e-01 5.81299484e-01 -6.36224389e-01
-1.00790334e+00 9.97837365e-01 -2.24530697e-01 -3.91915053e-01
6.26395941e-01 -5.93156926e-02 -4.10198092e-01 4.70382184e-01
1.25263289e-01 -6.47534788e-01 4.12505448e-01 -6.84651911e-01
-2.58315772e-01 -1.07636225e+00 -2.44729668e-01 6.30247518e-02
5.03950834e-01 2.11403698e-01 2.54183471e-01 -6.13416255e-01
5.50061345e-01 -8.06163728e-01 -5.67043364e-01 -3.17066580e-01
-2.40813762e-01 6.21143460e-01 4.01345164e-01 -7.11695135e-01
1.71258855e+00 -2.18535042e+00 -2.37076193e-01 6.95925653e-01
4.25885111e-01 4.29020524e-01 3.68143857e-01 2.91477531e-01
-2.46231824e-01 6.81856334e-01 -3.66557956e-01 7.32279658e-01
-3.33002985e-01 2.29657531e-01 -1.57100242e-02 8.31174672e-01
-1.82102084e-01 1.10532129e+00 -5.27096272e-01 -5.15522242e-01
1.16641015e-01 3.79665256e-01 -7.54301576e-03 -1.36218071e-01
6.05351329e-01 1.00493526e+00 -2.56806016e-01 4.10091579e-01
5.29864192e-01 -3.86345275e-02 9.18045104e-01 -3.77894789e-01
-2.54562289e-01 3.31592947e-01 -1.26605380e+00 1.48184109e+00
-6.45217905e-03 4.04513270e-01 -2.53249794e-01 -9.88542259e-01
1.19301295e+00 8.42829823e-01 4.98885930e-01 -5.21160901e-01
2.33005986e-01 5.68934798e-01 9.48268831e-01 -4.79133964e-01
-5.44792175e-01 -4.33769524e-01 2.17432022e-01 5.86102724e-01
-7.16661155e-01 1.37691712e-02 2.98700899e-01 -3.43849778e-01
9.42442536e-01 -9.15405061e-03 1.05228961e+00 -5.42735696e-01
6.05855048e-01 -2.93579638e-01 6.35598004e-01 9.07861233e-01
-4.15854096e-01 9.22885835e-01 1.04252088e+00 -8.59884560e-01
-6.79371536e-01 -1.27320063e+00 -5.21661162e-01 1.43965781e-01
-1.78324874e-03 -2.06992164e-01 -1.20007701e-01 -2.97378331e-01
-8.69518593e-02 7.51302093e-02 -3.60320598e-01 -2.80148406e-02
-1.05045033e+00 -1.00069046e+00 8.58876050e-01 6.59372866e-01
-1.05637766e-01 -6.03742242e-01 -1.73364329e+00 8.15982938e-01
-2.59384215e-01 -6.37842596e-01 4.23187949e-02 1.13427632e-01
-1.53200638e+00 -1.26560974e+00 -9.00851846e-01 -3.18120450e-01
3.23224366e-01 -3.78975123e-01 1.23993802e+00 2.29690388e-01
-9.53857183e-01 2.15166822e-01 -2.21636191e-01 -4.91450518e-01
-4.18009937e-01 -2.19066709e-01 -9.50877070e-02 -8.70155618e-02
-6.59391806e-02 -9.73971605e-01 -1.21422541e+00 6.05064631e-01
-2.60553539e-01 -4.19865549e-01 5.38862944e-01 3.26049805e-01
8.24456155e-01 -5.64246058e-01 8.48795772e-01 -5.46550870e-01
6.39546275e-01 -3.01974773e-01 -1.34452760e-01 7.90764540e-02
-1.05871725e+00 -6.99403465e-01 2.00082570e-01 -3.06023568e-01
-3.39306593e-01 -2.44142458e-01 -7.38437995e-02 1.46547824e-01
-5.61941981e-01 5.54227650e-01 2.14236379e-02 3.34366381e-01
1.06818032e+00 -7.48053119e-02 4.78918217e-02 -6.26394302e-02
-2.69195586e-01 2.95349002e-01 7.65803814e-01 -6.19364798e-01
3.96391004e-01 3.71680975e-01 4.83576328e-01 -6.22010708e-01
-2.17360243e-01 -7.04112053e-01 -1.16004014e+00 -1.63401589e-01
6.85110390e-01 -4.51724887e-01 -6.98039889e-01 -1.40294388e-01
-1.07999444e+00 -1.96559057e-01 -8.34835529e-01 8.21785688e-01
-3.22786629e-01 8.06181967e-01 -3.61931622e-01 -7.43380904e-01
-9.53343868e-01 -9.56701458e-01 5.84684849e-01 -2.79691339e-01
-8.84167850e-01 -9.11584139e-01 3.36619049e-01 -4.53927159e-01
3.02893519e-01 1.27557671e+00 1.06046736e+00 -5.32149851e-01
2.61023827e-02 -5.59988856e-01 2.05661595e-01 3.04942131e-01
3.29671144e-01 3.82606536e-02 -2.13371336e-01 -4.82633710e-02
-4.55537848e-02 6.40174091e-01 1.00288846e-01 8.38986695e-01
6.02324605e-01 3.95450205e-01 -2.38816023e-01 6.28088117e-01
1.28439760e+00 6.51513398e-01 9.22802031e-01 3.51054132e-01
3.55607629e-01 2.07215071e-01 2.63682723e-01 6.95407689e-01
-4.72285226e-02 3.93846244e-01 -4.52160463e-02 -3.19000006e-01
1.39713794e-01 6.35093749e-01 8.57911259e-02 2.81165242e-01
-7.67161071e-01 1.79477230e-01 -1.20119917e+00 5.26081443e-01
-1.31117427e+00 -7.70556092e-01 -1.30527711e+00 2.55202103e+00
6.93606615e-01 1.88683018e-01 3.85575622e-01 6.36467278e-01
7.14212120e-01 -3.62713814e-01 -6.06577933e-01 -6.23778522e-01
-2.75481999e-01 5.43843150e-01 1.92903087e-01 3.04676056e-01
-4.68531370e-01 -2.31973469e-01 6.76630402e+00 -4.82265413e-01
-1.46338201e+00 8.05697367e-02 9.40075696e-01 2.07473934e-01
9.34915021e-02 3.23639065e-01 -3.16281348e-01 1.82188794e-01
1.24851096e+00 -4.71330732e-01 -4.43945915e-01 2.97560096e-01
5.89486539e-01 -1.49059698e-01 -1.11955702e+00 1.17209589e+00
-3.99626374e-01 -1.34058321e+00 -6.93277597e-01 -7.89186060e-02
1.05689861e-01 -1.71907708e-01 -2.57753313e-01 -2.60385960e-01
-1.13381255e+00 -4.88940984e-01 5.09767830e-01 5.56155741e-01
1.49062371e+00 -3.85964096e-01 9.14841890e-01 -1.71305239e-02
-1.24574804e+00 6.18156232e-02 1.33856758e-01 -2.84283072e-01
4.13818300e-01 8.16092551e-01 -7.00516224e-01 5.65473080e-01
3.47268552e-01 6.54061735e-01 -4.33743775e-01 1.12118673e+00
-2.05402538e-01 1.05391777e+00 -4.59582686e-01 7.12356150e-01
-4.49232399e-01 -2.64054000e-01 7.99153268e-01 9.62824941e-01
3.68507296e-01 4.84197825e-01 3.38436733e-03 6.62188947e-01
5.78377485e-01 5.49047768e-01 -4.21844631e-01 5.50706267e-01
3.71517777e-01 1.02798498e+00 -1.05831659e+00 -6.17612422e-01
-2.48388946e-01 4.26452994e-01 -6.33216441e-01 5.27754426e-01
-5.80703676e-01 -5.91096401e-01 1.31031543e-01 7.80441463e-01
-2.84712732e-01 -2.52701771e-02 -1.20001733e+00 -6.21384442e-01
4.36918348e-01 -8.24788570e-01 5.52343667e-01 -1.83073968e-01
-5.22698045e-01 6.33324444e-01 4.79254760e-02 -1.48505867e+00
-6.86130464e-01 -2.68159688e-01 -1.04119980e+00 1.43584561e+00
-1.12264585e+00 -5.47182262e-01 -1.42662957e-01 2.45751739e-01
-1.32567719e-01 3.76407593e-01 1.26888740e+00 2.68458635e-01
-2.44920760e-01 4.73111004e-01 -5.69947958e-01 -1.41610252e-02
6.31040096e-01 -1.36162198e+00 1.61619738e-01 8.52158844e-01
-3.63987237e-01 8.38556588e-01 6.54998660e-01 -6.15586817e-01
-8.71065378e-01 -5.54705977e-01 1.08219075e+00 -4.10466194e-01
1.60835207e-01 4.22723621e-01 -6.68341279e-01 3.54981273e-01
-1.82068482e-01 4.72548872e-01 1.19329846e+00 -8.14861804e-02
6.71791509e-02 -2.72545397e-01 -8.95073533e-01 3.32582831e-01
5.88980258e-01 -2.56581962e-01 -5.78108549e-01 -1.21274263e-01
-3.62955064e-01 -3.33073229e-01 -1.67037034e+00 7.09535658e-01
1.29981399e+00 -1.13721812e+00 1.13309622e+00 -5.91958761e-01
-1.87540397e-01 -2.72578120e-01 3.63766044e-01 -8.90715420e-01
-3.99086416e-01 -1.35592079e+00 3.48537594e-01 7.12532938e-01
3.71348798e-01 -1.00749922e+00 4.37128067e-01 5.37431180e-01
-2.33058020e-01 -7.32031643e-01 -7.27020621e-01 -5.91459572e-01
2.65669245e-02 -4.14092481e-01 1.11623742e-01 5.64686418e-01
2.49470100e-01 7.60546606e-03 1.62741423e-01 7.92121664e-02
4.96620089e-01 2.93032050e-01 3.92183959e-01 -1.48420048e+00
-3.60583872e-01 -4.04081136e-01 -5.64663708e-01 -6.77201226e-02
-5.88981926e-01 -8.71994674e-01 -6.21317446e-01 -1.65440547e+00
-1.16018511e-01 -5.34910798e-01 -6.68966651e-01 4.59296405e-02
-5.25216758e-01 5.20576060e-01 -7.57997259e-02 1.59137487e-01
2.12086603e-01 -6.20575905e-01 1.00146079e+00 4.27967012e-01
-8.73060524e-01 3.46489698e-01 -5.31549692e-01 8.46137702e-01
9.50148284e-01 -8.45422387e-01 -5.77339530e-01 3.59654874e-01
3.74255449e-01 7.23382771e-01 4.77030903e-01 -9.45838749e-01
-3.63197565e-01 1.83802396e-01 5.93534946e-01 -8.11528504e-01
-2.68060148e-01 -5.32567501e-01 5.77155828e-01 9.64307547e-01
-2.09738258e-02 7.22217441e-01 2.29854241e-01 4.39996757e-02
-8.05275068e-02 -7.17606992e-02 6.88178003e-01 -3.60528052e-01
2.38089985e-03 1.39294505e-01 -9.70725894e-01 5.78586519e-01
1.01237130e+00 -8.13567579e-01 1.07347906e-01 4.45858128e-02
-1.51841843e+00 -5.44509768e-01 -6.81653444e-04 -2.34447777e-01
5.70513546e-01 -1.05622292e+00 -1.12179208e+00 1.28601417e-01
1.98134869e-01 -2.83366084e-01 6.12956107e-01 2.08292961e+00
-1.13721299e+00 4.35886420e-02 -3.08450699e-01 -9.65535641e-01
-1.42993462e+00 1.50501132e-01 7.41121292e-01 -3.21571767e-01
-1.01935542e+00 4.17368680e-01 -1.57502651e-01 7.69655526e-01
-8.52879509e-02 -4.00832981e-01 -8.14490765e-02 2.21684322e-01
6.57308578e-01 5.19183695e-01 3.34250718e-01 -3.46983641e-01
-4.93770599e-01 8.29290211e-01 3.60709131e-01 2.17273459e-02
5.40177584e-01 -5.91666579e-01 -2.03258425e-01 6.71947718e-01
9.24667954e-01 1.13072336e-01 -5.51418662e-01 2.49979079e-01
-1.22518629e-01 -2.04628602e-01 -1.16386496e-01 -9.69107091e-01
-7.92612612e-01 1.05472541e+00 1.26188803e+00 2.03935325e-01
1.17466974e+00 -3.24778438e-01 5.46096027e-01 -5.56676695e-03
3.77109557e-01 -4.81540859e-01 -3.88989687e-01 -4.21142370e-01
6.76550210e-01 -4.38936979e-01 2.76605606e-01 -2.09570527e-01
-4.63371366e-01 1.39853561e+00 -4.07984734e-01 -1.73182592e-01
7.90111721e-01 1.94233030e-01 6.51496112e-01 -1.24320433e-01
-2.66648442e-01 -8.49123970e-02 2.09186859e-02 7.09393322e-01
8.61823499e-01 3.53242576e-01 -1.16280162e+00 4.91328239e-01
2.17717811e-01 1.67557925e-01 6.48450553e-01 1.02590621e+00
-2.50016481e-01 -1.41228974e+00 -6.13426387e-01 5.94436824e-01
-8.70405734e-01 -2.88971588e-02 -1.69832602e-01 6.72351182e-01
3.08808148e-01 9.02343333e-01 -3.52406383e-01 1.99404016e-01
7.69837677e-01 4.75004673e-01 6.49433374e-01 -3.67973566e-01
-9.26637948e-01 6.16530597e-01 1.94789186e-01 -1.08457506e-01
-2.46428728e-01 -1.05752623e+00 -1.32382011e+00 4.54935849e-01
-6.21118583e-02 7.54512474e-02 5.35826027e-01 6.44045830e-01
5.11945069e-01 7.14533925e-01 2.66314596e-01 -5.07701337e-01
-5.13265193e-01 -9.97022808e-01 -1.01248562e+00 3.12607020e-01
4.83497024e-01 -2.48807728e-01 -3.13936174e-01 6.33661747e-01] | [14.244291305541992, 3.224187135696411] |
9a5b1d3b-a4a4-42c0-80cd-ca46b91dc62d | neaf-learning-neural-angle-fields-for-point | 2211.16869 | null | https://arxiv.org/abs/2211.16869v1 | https://arxiv.org/pdf/2211.16869v1.pdf | NeAF: Learning Neural Angle Fields for Point Normal Estimation | Normal estimation for unstructured point clouds is an important task in 3D computer vision. Current methods achieve encouraging results by mapping local patches to normal vectors or learning local surface fitting using neural networks. However, these methods are not generalized well to unseen scenarios and are sensitive to parameter settings. To resolve these issues, we propose an implicit function to learn an angle field around the normal of each point in the spherical coordinate system, which is dubbed as Neural Angle Fields (NeAF). Instead of directly predicting the normal of an input point, we predict the angle offset between the ground truth normal and a randomly sampled query normal. This strategy pushes the network to observe more diverse samples, which leads to higher prediction accuracy in a more robust manner. To predict normals from the learned angle fields at inference time, we randomly sample query vectors in a unit spherical space and take the vectors with minimal angle values as the predicted normals. To further leverage the prior learned by NeAF, we propose to refine the predicted normal vectors by minimizing the angle offsets. The experimental results with synthetic data and real scans show significant improvements over the state-of-the-art under widely used benchmarks. | ['Zhizhong Han', 'Yu-Shen Liu', 'Baorui Ma', 'Junsheng Zhou', 'Shujuan Li'] | 2022-11-30 | null | null | null | null | ['surface-normals-estimation'] | ['computer-vision'] | [ 3.37281376e-01 3.17794271e-02 -1.54440418e-01 -6.72640264e-01
-7.23694623e-01 -2.93694288e-01 4.65541601e-01 -5.16631529e-02
-2.17571646e-01 1.26545921e-01 -1.43892318e-01 1.80575863e-01
-8.86563584e-03 -9.33192253e-01 -1.07768750e+00 -6.39291525e-01
2.06184626e-01 8.01160157e-01 3.66166592e-01 -1.15192896e-02
5.67231238e-01 8.80911112e-01 -1.38327610e+00 -1.74039289e-01
8.44957292e-01 1.12776768e+00 1.34198308e-01 1.72457680e-01
-1.28153667e-01 -5.46367140e-03 -3.83211493e-01 -3.13182175e-02
5.65093160e-01 2.01851889e-01 -3.98829967e-01 3.68599072e-02
7.50877500e-01 -3.44507545e-01 -8.69278237e-02 1.13723612e+00
2.52417088e-01 1.31170645e-01 7.34040678e-01 -9.18852091e-01
-5.96240103e-01 -2.09422737e-01 -7.62521803e-01 -3.20286155e-01
3.10771018e-01 -1.21653482e-01 9.74541724e-01 -1.17886341e+00
6.84733987e-01 1.06012738e+00 8.62418711e-01 4.07544941e-01
-1.20429933e+00 -6.61872685e-01 2.07172766e-01 5.59603870e-02
-1.70613766e+00 -1.58319771e-01 1.17882919e+00 -4.67866480e-01
5.99511385e-01 2.43180990e-01 6.52326107e-01 6.26811743e-01
3.53052951e-02 5.37819028e-01 6.93462253e-01 -3.41286093e-01
3.74140710e-01 -2.97199469e-02 -2.58964181e-01 5.58234096e-01
1.35465339e-01 -7.25943148e-02 -2.87963927e-01 -3.63234311e-01
1.23458827e+00 3.95372212e-01 -6.25876665e-01 -1.00091469e+00
-1.36476696e+00 9.39846992e-01 8.17173421e-01 -2.88627982e-01
-7.23730624e-01 9.85520426e-03 -8.19411799e-02 -1.20610796e-01
6.30982995e-01 5.29370546e-01 -5.90477526e-01 1.19876564e-01
-6.71760380e-01 5.22660077e-01 6.16570532e-01 8.14494729e-01
1.27899301e+00 -2.70969123e-01 1.80522501e-02 1.06930816e+00
6.05136991e-01 7.63700306e-01 3.28575186e-02 -1.15542233e+00
5.24673283e-01 8.29419732e-01 2.03613833e-01 -1.40101361e+00
-2.20039800e-01 -4.02786523e-01 -6.97155297e-01 2.63160765e-01
2.86425054e-01 9.14199427e-02 -1.06826925e+00 1.35777450e+00
8.34174573e-01 4.32269573e-01 -2.85732418e-01 1.14939463e+00
3.77547890e-01 6.57776654e-01 -7.08206296e-01 6.84438124e-02
7.91561067e-01 -7.61509240e-01 -4.08995971e-02 -4.31131482e-01
2.00910032e-01 -8.02223682e-01 1.11507320e+00 4.99780297e-01
-7.12149084e-01 -3.28077286e-01 -1.00314295e+00 1.07645378e-01
1.77300543e-01 -1.21862078e-02 2.83938438e-01 5.73076010e-02
-7.56436765e-01 6.65208638e-01 -1.20715058e+00 -1.82596873e-02
4.64446038e-01 3.32580745e-01 -4.82020229e-02 -1.92346990e-01
-6.81140602e-01 4.57981527e-01 -1.54073149e-01 2.15224668e-01
-4.25455511e-01 -9.60806429e-01 -8.45356047e-01 -2.38176838e-01
5.58571577e-01 -6.27025127e-01 1.17441070e+00 -5.19112170e-01
-1.45842171e+00 7.39619017e-01 -3.71433556e-01 -1.15205750e-01
3.58702868e-01 -4.81718630e-01 6.32911101e-02 4.56365012e-02
3.03470969e-01 7.18255162e-01 8.29780817e-01 -1.64239144e+00
-4.61750478e-01 -6.30385637e-01 -1.08227037e-01 3.09671879e-01
-3.50163542e-02 -5.88347495e-01 -7.20805228e-01 -4.41886425e-01
8.60005677e-01 -1.03432715e+00 -4.95432109e-01 5.27205825e-01
-5.77489793e-01 -4.25716102e-01 8.27919483e-01 -2.41435543e-01
7.68486857e-01 -2.03922868e+00 -6.63840845e-02 7.08917558e-01
3.41884077e-01 -2.32040994e-02 9.19292793e-02 -1.53866075e-02
1.09849133e-01 -1.91529453e-01 -2.61676133e-01 -2.09168136e-01
-1.31982267e-01 4.02677715e-01 -2.87788093e-01 6.48620725e-01
3.01254839e-01 7.22850502e-01 -8.17968309e-01 -1.35252744e-01
2.81692326e-01 6.22442245e-01 -7.92916059e-01 3.66140723e-01
-4.68736261e-01 4.05323535e-01 -9.07474637e-01 6.04431629e-01
8.75429749e-01 -4.28261548e-01 -4.56382126e-01 -4.06017214e-01
2.58370936e-02 1.06842548e-01 -1.38885319e+00 1.73401272e+00
-2.61246353e-01 2.77207732e-01 -1.30351543e-01 -1.08943868e+00
1.60009420e+00 -2.50980183e-02 6.30522549e-01 -4.58969504e-01
-4.85251993e-02 9.97248739e-02 -2.66731739e-01 -3.10789466e-01
1.91677526e-01 -8.62031505e-02 2.75458544e-01 8.16665068e-02
-4.28122789e-01 -5.80048561e-01 -5.23466587e-01 -3.20731252e-01
6.08969629e-01 2.81270474e-01 2.03173503e-01 1.05754837e-01
4.62730706e-01 2.01232098e-02 8.27782512e-01 5.12127578e-01
1.80466369e-01 9.71118927e-01 2.66411573e-01 -7.05849111e-01
-1.15662909e+00 -1.15308309e+00 -3.48279059e-01 5.68717122e-01
4.20902163e-01 -3.61181170e-01 -6.35128736e-01 -5.03758729e-01
1.21243410e-01 4.08029348e-01 -5.16267061e-01 2.04568077e-02
-8.24974298e-01 -2.47505113e-01 -1.03248596e-01 6.27368510e-01
2.43525133e-01 -8.18052411e-01 -5.35962462e-01 -5.19666635e-02
5.84786423e-02 -9.97990489e-01 -5.10824740e-01 -5.43636829e-02
-9.49092329e-01 -9.75260973e-01 -6.47898376e-01 -6.43264055e-01
1.01356542e+00 2.45731816e-01 9.65187967e-01 4.21845494e-03
-8.84724632e-02 1.87512621e-01 -1.22943975e-01 -5.29373825e-01
1.29368201e-01 1.47199472e-02 -6.87147230e-02 1.00012608e-01
4.57762748e-01 -5.94452679e-01 -8.86824429e-01 5.58048666e-01
-5.20118535e-01 5.67936003e-02 4.69634771e-01 7.66567826e-01
1.40272903e+00 -4.27332938e-01 1.16955005e-01 -8.85220885e-01
1.38163596e-01 -4.92003113e-01 -8.02412629e-01 -1.15701288e-01
-7.42472351e-01 1.53194591e-01 5.50696254e-01 -4.42144513e-01
-6.03521526e-01 3.62808347e-01 -1.49465904e-01 -8.46603096e-01
-3.65384132e-01 4.43307579e-01 2.66741663e-02 -2.50406832e-01
6.88101470e-01 1.69989616e-01 -2.43307860e-03 -5.60764253e-01
2.33427182e-01 4.50240135e-01 5.40195704e-01 -6.74747586e-01
1.06552756e+00 6.64421916e-01 1.99375927e-01 -8.99218321e-01
-1.02865756e+00 -6.24448359e-01 -7.02139497e-01 -8.36328641e-02
7.19107449e-01 -6.90990925e-01 -4.94595736e-01 2.80651808e-01
-1.25335658e+00 -1.56301141e-01 -2.77114093e-01 5.04962265e-01
-5.14795840e-01 3.75628352e-01 -7.63457119e-02 -4.85074520e-01
-4.69506592e-01 -1.24945188e+00 1.51732457e+00 3.09829712e-01
-1.51360780e-02 -7.72836149e-01 2.11188465e-01 1.56167522e-01
1.63592204e-01 4.91056025e-01 8.85965347e-01 -4.61258411e-01
-8.95141482e-01 -4.56904471e-01 -2.47929171e-01 3.69571716e-01
1.74605295e-01 5.20429090e-02 -6.90873981e-01 -1.67306140e-01
1.43926188e-01 -8.88441578e-02 2.94475496e-01 5.33684969e-01
1.46589208e+00 -2.04721928e-01 -5.80036163e-01 1.13157320e+00
1.42410231e+00 -2.45457282e-03 3.53300214e-01 4.50164914e-01
9.25745249e-01 3.60190421e-01 7.91705549e-01 5.08133650e-01
5.02848744e-01 8.08569431e-01 7.07708240e-01 5.90079576e-02
3.51216048e-01 -3.56837153e-01 -1.18557133e-01 7.49066591e-01
-1.83899581e-01 1.83154568e-01 -1.10708654e+00 2.82430232e-01
-1.61523175e+00 -3.87615412e-01 -4.06540036e-02 2.39800644e+00
7.32083142e-01 3.18918616e-01 -2.79638559e-01 -1.13394707e-01
5.89697540e-01 1.52713597e-01 -9.70834553e-01 3.92887108e-02
2.96405435e-01 2.48621076e-01 3.92787814e-01 4.88022625e-01
-8.24680328e-01 8.33548069e-01 5.83182478e+00 4.85301077e-01
-1.60305536e+00 -3.72275054e-01 3.19048941e-01 -3.34639587e-02
-4.31792289e-01 3.15064713e-02 -9.03472185e-01 2.86557406e-01
3.25468332e-01 2.26644222e-02 5.11357963e-01 1.13429058e+00
1.34175956e-01 2.10909650e-01 -1.07060683e+00 1.06204975e+00
1.91555515e-01 -1.34783363e+00 8.44234899e-02 6.22838885e-02
8.31604123e-01 5.68619847e-01 -1.98801771e-01 -1.95602514e-02
5.12638409e-03 -9.73701715e-01 5.61118066e-01 8.21458280e-01
8.53093386e-01 -5.01705408e-01 5.24293840e-01 6.70507610e-01
-1.06283867e+00 3.91487420e-01 -6.89756870e-01 3.43916006e-02
1.07571125e-01 6.40643299e-01 -1.32739127e+00 2.46334702e-01
7.68492520e-01 9.07807529e-01 -4.32257533e-01 1.20696092e+00
-2.62878835e-01 4.64023888e-01 -7.98146069e-01 -4.42336611e-02
1.05594359e-01 -5.27909160e-01 5.72892070e-01 6.13837242e-01
3.82111520e-01 -7.05261007e-02 4.82716769e-01 1.08625424e+00
7.35925511e-02 1.71091542e-01 -6.00755513e-01 4.18407977e-01
8.39939833e-01 1.21332824e+00 -5.78133762e-01 3.08129061e-02
-2.98579723e-01 6.68593884e-01 4.65662569e-01 4.64145362e-01
-4.21236932e-01 -3.61352324e-01 7.33437300e-01 5.14982700e-01
4.48939055e-01 -3.91609192e-01 -4.47279364e-01 -9.59560871e-01
4.31027442e-01 -4.56757009e-01 -1.58949450e-01 -8.54934871e-01
-1.35382211e+00 4.44717139e-01 -2.93120518e-02 -1.48102379e+00
-2.93978214e-01 -5.85201740e-01 -7.83465147e-01 8.85361433e-01
-1.45027530e+00 -9.14184630e-01 -6.86228573e-01 4.70365107e-01
5.32121301e-01 1.30144238e-01 5.66676378e-01 -1.07331723e-01
-2.69380398e-02 3.81978154e-01 4.36063930e-02 1.46567464e-01
6.50540531e-01 -1.30698776e+00 6.20861948e-01 3.49499077e-01
4.84562255e-02 7.67813206e-01 5.53385377e-01 -6.38861299e-01
-1.36597598e+00 -1.01671386e+00 3.32341701e-01 -3.61769140e-01
2.82560378e-01 -4.02753055e-01 -1.29435742e+00 6.72958016e-01
-4.27983612e-01 5.60248315e-01 2.00466692e-01 -2.95377076e-02
-2.77595192e-01 -2.18721479e-01 -1.03692806e+00 5.15301824e-01
1.01102769e+00 -2.84877241e-01 -5.14373660e-01 3.26410145e-01
8.03904593e-01 -8.45344365e-01 -1.00597131e+00 7.63167441e-01
4.82109547e-01 -8.37278962e-01 1.26187563e+00 -1.98115811e-01
3.32112253e-01 -5.31745017e-01 -2.32443973e-01 -1.42331302e+00
-1.00476354e-01 -2.52259433e-01 -9.36722010e-02 8.31914961e-01
1.98866278e-01 -6.92704201e-01 1.33642721e+00 6.14475548e-01
-2.73332328e-01 -1.60175586e+00 -8.77936006e-01 -3.47861320e-01
5.66271394e-02 -5.68162560e-01 8.61163318e-01 8.80502582e-01
-6.45020545e-01 3.62313390e-01 1.43717369e-02 5.99983156e-01
8.22156072e-01 2.50792474e-01 1.21938324e+00 -1.61750197e+00
-3.43907475e-02 -2.52554625e-01 -5.06667852e-01 -1.71903217e+00
8.10385197e-02 -7.75033295e-01 3.73785943e-01 -1.52958310e+00
-2.36328453e-01 -8.97504210e-01 1.24583870e-01 2.64042199e-01
-1.22703493e-01 1.59331217e-01 -1.86054379e-01 4.25819635e-01
-2.57216185e-01 7.51584530e-01 1.56586528e+00 9.57893953e-02
-2.49886885e-01 3.11425805e-01 -4.68524665e-01 1.11947298e+00
5.13368249e-01 -4.95866030e-01 -3.83459002e-01 -6.10160112e-01
2.49194056e-01 -2.71800384e-02 2.79817790e-01 -7.86524534e-01
4.21296775e-01 -5.94351470e-01 5.40992558e-01 -1.11975515e+00
4.92154658e-01 -8.87848139e-01 -3.47269922e-02 -3.46678495e-02
-1.33132741e-01 1.84464362e-02 -3.45759064e-01 6.66620970e-01
-3.38464417e-02 -1.89594716e-01 6.81056023e-01 5.36147319e-02
-3.26619714e-01 7.56096244e-01 4.02957886e-01 -3.16017903e-02
8.85043144e-01 -5.80155790e-01 -4.00712527e-02 -2.15706229e-01
-4.37889963e-01 3.07443947e-01 8.65867794e-01 4.36427116e-01
9.49483335e-01 -1.35648620e+00 -5.14843762e-01 7.51663923e-01
3.73147637e-01 1.18960774e+00 -1.63264185e-01 6.09215319e-01
-7.04834044e-01 1.96696311e-01 2.12763473e-01 -1.28109765e+00
-7.69265175e-01 1.74791381e-01 5.34261167e-01 2.63296455e-01
-9.24153924e-01 8.50860119e-01 2.47067705e-01 -8.57527852e-01
2.69420147e-01 -6.07706726e-01 -8.59289244e-02 -5.80001354e-01
1.35473639e-01 1.97592303e-01 1.73584983e-01 -5.65345109e-01
-3.52363735e-01 1.22653592e+00 -1.06120288e-01 1.86055481e-01
1.53078914e+00 7.58048519e-02 2.64193658e-02 3.93656820e-01
1.38144100e+00 1.88744873e-01 -1.69262588e+00 -5.67457139e-01
5.36075868e-02 -8.26620996e-01 4.77381684e-02 -3.08656216e-01
-1.19579685e+00 6.62086010e-01 5.61504781e-01 -1.41781336e-02
7.07468033e-01 2.02213168e-01 8.57328653e-01 5.82948983e-01
2.71805584e-01 -7.74937391e-01 9.39912647e-02 4.77816314e-01
9.51812506e-01 -1.20615983e+00 5.53286113e-02 -5.93536615e-01
-4.62193936e-01 1.26872849e+00 7.94913590e-01 -6.23412311e-01
7.80822694e-01 1.78377852e-01 3.57177347e-01 -4.90013987e-01
-4.08381641e-01 4.47015762e-01 5.18189788e-01 6.89813673e-01
2.12007046e-01 -2.80757658e-02 1.63093716e-01 2.21268222e-01
-4.36218143e-01 -2.67967954e-02 1.69087112e-01 7.18033016e-01
-5.61160445e-01 -9.19479549e-01 -6.31455779e-01 6.43197834e-01
-8.95976126e-02 3.37836415e-01 -8.60339031e-02 5.13344407e-01
1.98808480e-02 2.55274475e-01 3.53851676e-01 -3.03593546e-01
6.23389661e-01 -3.18984568e-01 2.69944698e-01 -7.89836884e-01
1.91887900e-01 1.16599761e-01 -4.53954220e-01 -6.13733888e-01
-2.32974857e-01 -7.59468019e-01 -1.49411929e+00 1.84100449e-01
-4.63418752e-01 1.19569369e-01 8.04292440e-01 9.11613345e-01
3.93509358e-01 -1.67204440e-02 9.91385520e-01 -1.29528379e+00
-7.61084676e-01 -7.77105749e-01 -2.20785812e-01 5.04258692e-01
3.96638393e-01 -8.17484200e-01 -3.97736162e-01 -1.49378687e-01] | [8.152814865112305, -3.4314024448394775] |
cb4483d5-8066-4920-b751-244c84135824 | stc-ids-spatial-temporal-correlation-feature | 2204.10990 | null | https://arxiv.org/abs/2204.10990v2 | https://arxiv.org/pdf/2204.10990v2.pdf | STC-IDS: Spatial-Temporal Correlation Feature Analyzing based Intrusion Detection System for Intelligent Connected Vehicles | Intrusion detection is an important defensive measure for automotive communications security. Accurate frame detection models assist vehicles to avoid malicious attacks. Uncertainty and diversity regarding attack methods make this task challenging. However, the existing works have the limitation of only considering local features or the weak feature mapping of multi-features. To address these limitations, we present a novel model for automotive intrusion detection by spatial-temporal correlation features of in-vehicle communication traffic (STC-IDS). Specifically, the proposed model exploits an encoding-detection architecture. In the encoder part, spatial and temporal relations are encoded simultaneously. To strengthen the relationship between features, the attention-based convolutional network still captures spatial and channel features to increase the receptive field, while attention-LSTM builds meaningful relationships from previous time series or crucial bytes. The encoded information is then passed to detector for generating forceful spatial-temporal attention features and enabling anomaly classification. In particular, single-frame and multi-frame models are constructed to present different advantages respectively. Under automatic hyper-parameter selection based on Bayesian optimization, the model is trained to attain the best performance. Extensive empirical studies based on a real-world vehicle attack dataset demonstrate that STC-IDS has outperformed baseline methods and obtains fewer false-alarm rates while maintaining efficiency. | ['Aoxue Li', 'Mu Han', 'Pengzhou Cheng', 'Fengwei Zhang'] | 2022-04-23 | null | null | null | null | ['anomaly-classification'] | ['computer-vision'] | [ 5.04419841e-02 -5.40847182e-01 -4.08236295e-01 -3.14432472e-01
-5.63259244e-01 -1.19266853e-01 5.95283687e-01 4.77375686e-02
-4.26871121e-01 4.62377727e-01 -1.39328599e-01 -4.63435113e-01
-3.04456595e-02 -7.88583875e-01 -5.63393772e-01 -7.12117374e-01
-3.19034368e-01 -2.41518050e-01 6.74832404e-01 -2.90314108e-01
5.78315675e-01 8.88335466e-01 -1.49211526e+00 4.74740297e-01
4.73903775e-01 1.45903718e+00 -8.65167454e-02 6.04458213e-01
-2.40055978e-01 5.64608335e-01 -7.30397284e-01 -3.56601536e-01
1.83543712e-01 -8.81665349e-02 -2.52802193e-01 -1.11054167e-01
8.48840270e-03 -4.12821651e-01 -6.35027885e-01 1.21643865e+00
1.92451030e-01 1.59499377e-01 3.51457089e-01 -1.54981673e+00
-4.17823166e-01 9.81327593e-02 -5.05479038e-01 9.59296048e-01
1.80309862e-01 3.97384286e-01 6.48937285e-01 -8.00617754e-01
1.34477183e-01 1.44578087e+00 3.84861201e-01 5.02680838e-01
-6.57384634e-01 -1.01481688e+00 4.39788550e-01 1.03335226e+00
-1.24954224e+00 -3.80532712e-01 1.09565771e+00 -1.79740459e-01
1.18633211e+00 1.95441395e-01 5.58949471e-01 1.33729076e+00
8.50299358e-01 5.98754048e-01 6.06211841e-01 -1.34544119e-01
-5.71288925e-04 4.48372178e-02 3.43434781e-01 6.78964853e-01
1.65795401e-01 5.22409856e-01 -4.98209089e-01 -1.67697221e-01
3.48779649e-01 3.04492921e-01 -4.45634201e-02 2.66852945e-01
-9.00369704e-01 9.09242034e-01 3.04009736e-01 3.05523485e-01
-4.99224007e-01 5.36931530e-02 8.29007447e-01 4.82782006e-01
2.00504720e-01 3.44484374e-02 -3.16394955e-01 -2.73432225e-01
-5.72319031e-01 -1.92354422e-03 3.20317239e-01 6.94754064e-01
6.41675413e-01 6.43362939e-01 -1.28346607e-01 3.97728622e-01
5.13452947e-01 5.53383529e-01 6.15897715e-01 -5.05142748e-01
4.72318500e-01 5.44585764e-01 -4.57149595e-01 -1.49649620e+00
-3.35355818e-01 -5.48348248e-01 -7.93494523e-01 3.58542919e-01
-1.26119629e-01 1.04598984e-01 -8.00473094e-01 1.52981257e+00
2.92494774e-01 7.89569795e-01 1.48108646e-01 7.28714764e-01
5.13623238e-01 7.70526111e-01 2.82268375e-01 -4.87338781e-01
1.52912736e+00 -5.49567997e-01 -1.09439337e+00 -3.74430120e-02
6.25564396e-01 -8.48445594e-01 3.81761014e-01 2.13893443e-01
-6.83272839e-01 -8.87673140e-01 -1.35602820e+00 5.39870262e-01
-6.42991602e-01 -2.65998960e-01 1.38424441e-01 8.18843186e-01
-5.04274726e-01 2.17462823e-01 -5.92545033e-01 -3.55967395e-02
4.97754276e-01 4.71668571e-01 -2.63493836e-01 2.03476310e-01
-1.71269858e+00 7.73227274e-01 4.93688405e-01 2.50471890e-01
-8.77832830e-01 -4.01853412e-01 -7.81489909e-01 -1.05458617e-01
4.19642180e-01 -3.09137404e-02 9.31140959e-01 -8.23105931e-01
-1.46243739e+00 1.82791516e-01 -2.17787370e-01 -9.75018382e-01
-5.38041629e-02 -1.99895367e-01 -1.09839392e+00 2.19698057e-01
-8.72413740e-02 3.88105154e-01 1.27679181e+00 -9.21914399e-01
-9.28163528e-01 -2.35281363e-01 -6.62578493e-02 -1.26184955e-01
-5.21264970e-01 3.41263294e-01 -2.74850696e-01 -8.04350853e-01
1.76971499e-02 -7.09650755e-01 -1.87294185e-02 -3.61532241e-01
-3.06551307e-01 -2.00583726e-01 1.86366856e+00 -4.55031812e-01
1.58795381e+00 -2.35803080e+00 -3.93244535e-01 4.20311451e-01
1.43616110e-01 8.45078528e-01 -7.63277784e-02 1.61310270e-01
-2.07645074e-01 3.59132551e-02 1.87847868e-01 1.99264027e-02
-2.45363921e-01 2.10784644e-01 -5.76631010e-01 5.80208421e-01
5.97792327e-01 6.52534723e-01 -5.28064907e-01 -6.11304343e-01
5.20182967e-01 3.79570633e-01 -5.45370877e-01 8.31324533e-02
5.05431816e-02 4.03375387e-01 -8.22293520e-01 7.40067244e-01
7.59835720e-01 3.25056911e-01 -4.79371607e-01 -3.54956985e-01
-1.42276943e-01 -2.76843016e-03 -9.15997386e-01 1.02697182e+00
-3.20968866e-01 7.88096130e-01 -1.57242224e-01 -1.32951343e+00
1.12269855e+00 4.92784888e-01 5.45019329e-01 -1.18193877e+00
5.59405982e-01 1.47257268e-01 3.65839034e-01 -6.61382139e-01
2.62708217e-01 1.93161607e-01 -4.89493385e-02 6.93373159e-02
-9.96461362e-02 5.93521416e-01 -1.30997077e-01 2.46367641e-02
1.05136299e+00 -4.27479029e-01 1.75299093e-01 -1.76065508e-02
1.35959196e+00 -3.63508344e-01 7.57643700e-01 5.04405737e-01
-6.33244216e-01 -1.72021184e-02 4.61003631e-01 -7.09128678e-01
-7.21654594e-01 -8.64097238e-01 -2.15602160e-01 8.28149796e-01
3.12250286e-01 -4.00279582e-01 -5.89230597e-01 -9.27530348e-01
-5.08117750e-02 7.63751745e-01 -5.95182061e-01 -7.73741305e-01
-8.03226590e-01 -4.54988271e-01 7.23414242e-01 5.48202693e-01
6.70284927e-01 -1.01059687e+00 -5.78929245e-01 2.87186146e-01
7.56315514e-02 -1.43818605e+00 -2.51849204e-01 -6.96947202e-02
-6.14321768e-01 -1.04199624e+00 6.23060800e-02 -5.09855509e-01
3.26590478e-01 3.37491661e-01 5.13148546e-01 3.20154965e-01
-4.12887812e-01 2.04837710e-01 -4.21409965e-01 -6.10251963e-01
-2.12184057e-01 -2.10406318e-01 2.47793674e-01 5.92402935e-01
8.35699975e-01 -4.11302775e-01 -3.46167833e-01 3.39140505e-01
-7.04690635e-01 -5.37255585e-01 7.35134006e-01 9.79113519e-01
2.83479035e-01 2.72079229e-01 8.15292537e-01 -4.50650901e-01
4.41440552e-01 -7.08313584e-01 -5.81118345e-01 -2.37563089e-01
-3.83299857e-01 -1.41357556e-01 7.48004198e-01 -5.16914427e-01
-1.01026225e+00 -4.14267778e-01 -3.58810246e-01 -7.73407876e-01
-2.70981163e-01 1.34677514e-01 -3.38983744e-01 -1.94362819e-01
3.13765317e-01 3.94714206e-01 1.24779746e-01 7.13235186e-03
-1.55816182e-01 5.65068007e-01 3.82830471e-01 -4.36368197e-01
8.74498844e-01 4.25550342e-01 1.10037588e-01 -1.09663665e+00
-3.83832991e-01 -3.69493037e-01 -5.73209524e-01 -5.03200173e-01
1.04198945e+00 -6.78433001e-01 -9.39224124e-01 3.57740134e-01
-1.36920965e+00 4.90577638e-01 4.57375534e-02 7.73370147e-01
-2.13127494e-01 5.26366234e-01 -6.19054258e-01 -1.06576037e+00
-8.84653851e-02 -1.43408334e+00 9.92346585e-01 7.28004426e-02
1.04679786e-01 -7.79594839e-01 -3.70488435e-01 -5.52336965e-03
5.67523301e-01 1.56340688e-01 1.00780702e+00 -1.05081892e+00
-7.81538844e-01 -5.30867100e-01 -2.83895463e-01 3.47683042e-01
-5.55063970e-02 1.58748582e-01 -9.37887013e-01 -2.33210906e-01
3.28402013e-01 1.48126092e-02 8.67129028e-01 3.09467942e-01
1.60918009e+00 -2.83826262e-01 -4.24430251e-01 5.66255629e-01
9.99939382e-01 8.73089433e-01 7.74360061e-01 4.90232170e-01
5.41345835e-01 4.65574443e-01 8.31877232e-01 5.12575030e-01
-4.36415486e-02 7.12236285e-01 6.93562984e-01 1.69454664e-01
1.04452476e-01 1.92013443e-01 5.38482189e-01 7.68067956e-01
1.00430630e-01 -2.68251002e-01 -5.89505196e-01 1.25523642e-01
-1.69193065e+00 -1.29371643e+00 -5.95647320e-02 1.97978997e+00
2.49533996e-01 8.54933023e-01 -5.35627417e-02 4.62668389e-01
7.83591092e-01 3.75954330e-01 -3.68393272e-01 -6.43929422e-01
-6.24267980e-02 -2.08326709e-02 4.26573247e-01 3.82175118e-01
-1.28118634e+00 7.85137415e-01 5.74251604e+00 1.18153441e+00
-1.32246745e+00 -9.88812000e-03 6.58622682e-01 -5.61188981e-02
2.49355417e-02 -2.07802325e-01 -1.05867791e+00 7.39860058e-01
1.29216993e+00 8.11068416e-02 -1.05756328e-01 7.11576164e-01
2.34538525e-01 2.26716593e-01 -6.55677855e-01 9.38598037e-01
1.20163135e-01 -1.22265625e+00 2.74396330e-01 2.49475524e-01
1.68446675e-01 -2.75796741e-01 3.87709439e-01 4.73402321e-01
-2.61769146e-01 -8.59028757e-01 4.81288046e-01 6.26610994e-01
4.74757880e-01 -1.20774972e+00 9.92705226e-01 5.89130148e-02
-1.51164424e+00 -5.50323248e-01 -3.16191763e-01 2.11198151e-01
2.52762914e-01 3.16471010e-01 -5.82737386e-01 6.82831824e-01
5.54219484e-01 7.20103681e-01 -5.18133223e-01 5.85806668e-01
2.89320290e-01 7.35867500e-01 -2.63627052e-01 -2.11626552e-02
4.82348949e-01 1.40329763e-01 8.95209253e-01 1.27986920e+00
3.80646825e-01 5.54677993e-02 3.93074155e-01 4.41623777e-01
3.79544139e-01 -2.29143444e-02 -7.99768806e-01 1.74356386e-01
6.44313633e-01 1.14066148e+00 -4.92248476e-01 -2.94814616e-01
-7.44411647e-01 4.78438795e-01 -6.55328035e-02 3.93388450e-01
-1.36878538e+00 -4.44534063e-01 1.00682771e+00 1.45362001e-02
3.35520327e-01 -3.05057079e-01 -1.24817677e-01 -6.26178324e-01
-1.16994061e-01 -9.47950542e-01 3.81693155e-01 -5.83685413e-02
-1.02548575e+00 9.01873827e-01 5.16806096e-02 -1.49805105e+00
-2.16512755e-01 -8.56408536e-01 -9.10754740e-01 5.88599384e-01
-1.58852983e+00 -1.06533170e+00 -1.41217768e-01 8.61330688e-01
9.21282768e-01 -8.82655561e-01 5.62677562e-01 5.20727813e-01
-1.03716707e+00 9.36640024e-01 -3.15509111e-01 1.33554354e-01
4.81312484e-01 -5.62256336e-01 3.21904898e-01 1.09790015e+00
-1.97039962e-01 4.78041381e-01 5.06810486e-01 -8.49345744e-01
-1.32882035e+00 -1.19419467e+00 4.97212499e-01 -3.99400145e-02
7.74019957e-01 -5.74912913e-02 -1.10099328e+00 4.19905633e-01
2.24354997e-01 2.94846386e-01 6.52665615e-01 -3.33130896e-01
-4.23878640e-01 -3.77707809e-01 -9.97231066e-01 4.85651702e-01
6.91282213e-01 -6.71966195e-01 -4.19779688e-01 -9.42123458e-02
6.69055641e-01 -1.28524885e-01 -5.20998061e-01 6.18374348e-01
4.21496570e-01 -9.27059114e-01 1.01628900e+00 -6.94089413e-01
-9.99613181e-02 -4.73526388e-01 -3.71615320e-01 -6.55962229e-01
-4.21296388e-01 -6.09109879e-01 -4.54465628e-01 1.07169580e+00
2.53083944e-01 -8.85745525e-01 5.76643467e-01 -1.93046022e-03
-4.64964509e-01 -9.60270882e-01 -1.24009979e+00 -7.08734751e-01
-3.72806042e-01 -8.27649534e-01 6.93034053e-01 6.54373944e-01
-2.72484064e-01 5.29478043e-02 -4.18197334e-01 6.22612119e-01
5.17425179e-01 -3.89441490e-01 5.35605133e-01 -1.18170321e+00
1.22351862e-01 -5.39597213e-01 -1.04206300e+00 -8.91360939e-01
4.84783202e-01 -4.74345922e-01 -2.24552602e-01 -4.91033792e-01
-2.13910237e-01 -2.03586578e-01 -6.65309429e-01 2.23040357e-01
-5.72359599e-02 2.89048791e-01 -6.09737188e-02 9.72524658e-02
-6.27803266e-01 6.20730937e-01 8.61714661e-01 -1.82614833e-01
1.97737351e-01 1.40709028e-01 -1.76858455e-01 7.42741585e-01
9.11213696e-01 -2.99205750e-01 -4.51565415e-01 -3.57431024e-02
-4.21111226e-01 -7.41045747e-04 5.11006176e-01 -1.29746449e+00
5.06126225e-01 -9.69080403e-02 6.29487753e-01 -8.48407269e-01
5.38664460e-01 -1.05289495e+00 -3.20871025e-01 7.64418721e-01
-1.68580770e-01 5.47854424e-01 3.81603032e-01 9.62905645e-01
-5.29636145e-01 -2.45436519e-01 8.17270458e-01 2.31496632e-01
-1.05193353e+00 5.35316288e-01 -8.81558776e-01 -3.07862252e-01
1.45663011e+00 -4.68560100e-01 -1.11584835e-01 -1.56500146e-01
-3.93645346e-01 7.33747259e-02 -2.02360407e-01 7.67710745e-01
1.03089631e+00 -1.72376966e+00 -5.47575593e-01 8.44787121e-01
1.11877397e-01 -5.61259449e-01 4.52017903e-01 8.28064024e-01
-2.27916300e-01 7.77703881e-01 -4.04626101e-01 -7.68290102e-01
-1.37285733e+00 7.92016923e-01 2.29810849e-01 -1.00812018e-01
-5.33339381e-01 5.63889384e-01 2.46652901e-01 3.48766237e-01
3.11202049e-01 -9.91246849e-02 -6.20787978e-01 -8.11069757e-02
8.01517546e-01 2.86557585e-01 7.42026940e-02 -1.10057580e+00
-4.94891971e-01 4.54334527e-01 -4.67950076e-01 1.40541255e-01
6.48488581e-01 -1.59513965e-01 1.97963357e-01 8.72694775e-02
1.39672542e+00 -1.42282531e-01 -1.07002890e+00 -2.90435076e-01
1.91563703e-02 -6.57491565e-01 1.95803776e-01 -1.17944717e-01
-1.20712972e+00 9.86527920e-01 8.47561061e-01 3.47409695e-01
1.24139380e+00 -5.30377209e-01 1.11233461e+00 3.84839892e-01
8.37801769e-02 -9.27355587e-01 4.79052484e-01 8.27703655e-01
6.37797892e-01 -1.17900646e+00 -2.75306404e-01 -2.87820965e-01
-4.75798011e-01 1.38989806e+00 1.01595116e+00 -2.13962540e-01
1.02441883e+00 4.63429093e-01 -1.16207659e-01 -2.33550847e-01
-8.49109292e-01 -1.55857831e-01 3.44297826e-01 6.05909228e-01
1.15444086e-01 -3.94188285e-01 -1.20252989e-01 5.13380885e-01
5.43588810e-02 -6.61071002e-01 4.87775058e-02 8.30108225e-01
-7.50977457e-01 -9.52401698e-01 -5.59018373e-01 3.18882406e-01
-6.59774959e-01 1.92095026e-01 9.33229625e-02 7.37530649e-01
3.47222984e-01 1.04711306e+00 3.41477692e-01 -9.25676882e-01
2.38446042e-01 7.62681141e-02 -1.19545192e-01 -9.12596509e-02
-3.01386297e-01 4.79697101e-02 -8.98279622e-02 -8.22027981e-01
-2.20637232e-01 -4.83190566e-01 -1.24190402e+00 -3.54405463e-01
-4.51352805e-01 2.66923130e-01 4.93847638e-01 1.17434132e+00
4.34701025e-01 1.04464376e+00 1.04835927e+00 -7.52764881e-01
-4.21774626e-01 -8.26753020e-01 -1.95210502e-01 2.60318071e-01
6.48862720e-01 -1.02515149e+00 -3.39215010e-01 -1.87972888e-01] | [7.770087718963623, 1.7503784894943237] |
c564306e-2cc5-4413-8ac3-9aeb88e73dfa | transforming-worlds-automated-involutive-mcmc | null | null | https://openreview.net/forum?id=8Itm8dQnJRc | https://openreview.net/pdf?id=8Itm8dQnJRc | Transforming Worlds: Automated Involutive MCMC for Open-Universe Probabilistic Models | Open-universe probabilistic models enable Bayesian inference about how many objects underlie data, and how they are related. Effective inference in OUPMs remains a challenge, however, often requiring the use of custom, trans-dimensional MCMC kernels, based on heuristics, deep learning, or domain knowledge, that can be difficult to derive and to implement correctly. This paper adapts the recently introduced involutive MCMC framework to the open-universe setting, and shows how error-prone aspects of kernel design and implementation (e.g., the computation of valid accept/reject probabilities) can be automated, using techniques from probabilistic and differentiable programming. The result is an intuitive design space for MCMC kernels for OUPMs: users write programs that propose incremental changes to possible worlds, creating, deleting, or modifying objects according to arbitrary application-specific logic, and their proposals are automatically converted into stationary MCMC kernels. We demonstrate in preliminary experiments that data-driven involutive MCMC kernels outperform generic probabilistic programming language inference, as well as generic birth/death reversible-jump kernels without application-specific logic. | ['Vikash Mansinghka', 'Marco Cusumano-Towner', 'Stuart Russell', 'Matin Ghavamizadeh', 'Alexander K. Lew', 'George Matheos'] | 2020-11-23 | null | null | null | pproximateinference-aabi-symposium-2021-1 | ['probabilistic-programming'] | ['methodology'] | [ 1.13631219e-01 6.90100417e-02 -7.79774413e-02 -6.39329433e-01
-6.73979461e-01 -8.74085188e-01 9.65812147e-01 3.44391584e-01
-4.33746606e-01 7.16467142e-01 -3.48435231e-02 -7.80195296e-01
-4.66218829e-01 -1.30712366e+00 -8.81027997e-01 -4.27311510e-01
-2.04512358e-01 1.12939727e+00 5.42335987e-01 2.60845721e-01
3.23234558e-01 3.85539323e-01 -1.85643554e+00 1.80576101e-01
7.10574746e-01 3.21994066e-01 -1.58474922e-01 1.11582696e+00
-2.04665035e-01 6.80158913e-01 -1.60503402e-01 -7.59655714e-01
-3.17174196e-01 1.26751453e-01 -6.93421543e-01 -6.30472839e-01
4.62637320e-02 -4.53414202e-01 1.83077589e-01 8.89384270e-01
2.23554224e-01 1.21179156e-01 1.17380106e+00 -1.25682032e+00
-8.21322680e-01 1.07781398e+00 3.44808921e-02 -2.01909959e-01
3.18996280e-01 3.15878451e-01 9.06784952e-01 -7.08171606e-01
4.50059056e-01 1.62631047e+00 6.94030404e-01 5.45598328e-01
-1.88619459e+00 -1.72019616e-01 -1.55267194e-01 3.46243940e-02
-1.50890422e+00 -4.94456500e-01 1.70052186e-01 -6.18521273e-01
1.10857534e+00 7.44692564e-01 2.54304290e-01 1.06424356e+00
2.56364822e-01 8.19447100e-01 8.51796985e-01 -6.75101459e-01
9.16272998e-01 4.49812889e-01 5.27389586e-01 6.27516508e-01
3.46163660e-01 6.98000863e-02 -2.56360233e-01 -8.23172748e-01
6.22257888e-01 -9.93772820e-02 2.22210810e-01 -6.62495077e-01
-1.00579596e+00 9.26937044e-01 -2.97206759e-01 -1.86605394e-01
-7.53054675e-03 6.38872087e-01 3.04481804e-01 -4.10171188e-02
3.36111873e-01 1.14567742e-01 -7.28476465e-01 -4.79297101e-01
-9.02079880e-01 7.82575786e-01 1.24346173e+00 1.09819520e+00
8.70061755e-01 -4.15518701e-01 -2.74676800e-01 5.33689141e-01
7.45511591e-01 5.17702460e-01 -7.55558386e-02 -1.06383824e+00
-3.86268250e-03 4.01386440e-01 3.65562528e-01 -5.13265550e-01
-1.10915437e-01 2.37014011e-01 -3.47816199e-01 1.98153123e-01
3.45855236e-01 -1.92056924e-01 -7.14579284e-01 1.76380754e+00
5.05549252e-01 1.13104239e-01 -2.42583156e-02 2.58285582e-01
2.32425123e-01 6.13583624e-01 3.45100492e-01 9.87426788e-02
1.34746540e+00 -7.33155981e-02 -2.75323153e-01 -3.89930024e-03
6.70687020e-01 -4.65246320e-01 1.15947795e+00 2.97476858e-01
-1.19917011e+00 -1.05950475e-01 -7.24796295e-01 -2.92360187e-01
-4.97572511e-01 -1.68140277e-01 1.25916803e+00 1.17284632e+00
-1.11977172e+00 5.56545198e-01 -1.16764331e+00 -3.39294195e-01
2.17645958e-01 2.64579475e-01 1.22531183e-01 -2.06654705e-02
-1.08580768e+00 1.00125706e+00 8.76912117e-01 -4.73512746e-02
-9.25572991e-01 -8.04259777e-01 -8.95052373e-01 3.59010279e-01
3.64371628e-01 -8.28345001e-01 1.32791650e+00 -3.27284396e-01
-1.56576693e+00 5.66575468e-01 -5.02778916e-03 -5.69674313e-01
5.56006670e-01 4.40270528e-02 -7.08410591e-02 -3.57476354e-01
-3.82847376e-02 6.63780808e-01 6.93180144e-01 -8.64662647e-01
-6.95659041e-01 -2.29906052e-01 3.51116568e-01 -4.74248417e-02
1.56958237e-01 8.90815556e-02 -4.91688371e-01 -2.64730781e-01
-1.47583142e-01 -8.40781391e-01 -2.17533261e-01 4.72089462e-02
-3.61941993e-01 -5.76879263e-01 4.13693815e-01 -2.13427722e-01
1.16595924e+00 -1.97754884e+00 8.26757774e-02 4.14848924e-01
-1.30079657e-01 -3.13164860e-01 4.13818359e-01 2.60049015e-01
2.87590951e-01 1.14821419e-01 -5.12421787e-01 -1.28650010e-01
7.94811606e-01 5.59119284e-01 -5.05937159e-01 3.44145060e-01
2.31077969e-01 7.39056826e-01 -9.51429546e-01 -6.91454589e-01
4.49101865e-01 2.37759754e-01 -8.97934973e-01 -1.23851538e-01
-8.75037193e-01 -3.77050519e-01 -2.62438267e-01 6.29535377e-01
6.22824490e-01 -2.11626403e-02 4.10616368e-01 1.82570398e-01
-1.56167969e-01 2.19779551e-01 -1.42172658e+00 1.36337626e+00
-3.40105057e-01 3.90350819e-01 -3.73067021e-01 -6.41340375e-01
6.16994262e-01 2.50823647e-02 -2.54061580e-01 3.30313712e-01
-1.73249155e-01 3.90427895e-02 -3.58715534e-01 -1.73860922e-01
7.38398790e-01 -1.32744804e-01 -4.46724027e-01 8.03554475e-01
2.04526931e-01 -4.55624253e-01 3.69482696e-01 3.12375724e-01
1.12965274e+00 5.10521889e-01 2.47440115e-01 -4.77177501e-01
2.32809812e-01 -2.66673360e-02 4.85313803e-01 1.39518678e+00
1.69109017e-01 3.06351125e-01 9.84470427e-01 -3.22133958e-01
-9.91915584e-01 -1.71745408e+00 -6.67451262e-01 1.41171908e+00
-3.16788822e-01 -3.42977107e-01 -7.91949868e-01 -4.46418464e-01
1.48863941e-01 1.54276276e+00 -6.29231870e-01 -2.38445401e-01
-1.64421424e-01 -1.13799763e+00 7.13484824e-01 5.00948966e-01
-7.31106550e-02 -8.82134736e-01 -7.66326606e-01 3.50157291e-01
1.45461887e-01 -5.21694541e-01 -1.33348405e-01 3.29643846e-01
-9.46373343e-01 -8.64388704e-01 -3.25804561e-01 -3.27892780e-01
6.68842912e-01 -5.25870979e-01 1.13905501e+00 -3.74383926e-01
-3.29089940e-01 6.33783221e-01 -5.15812030e-03 -4.82663929e-01
-5.86985230e-01 -7.83544183e-02 -5.69609832e-03 -3.02591652e-01
6.19395792e-01 -5.89733124e-01 -8.33455771e-02 1.33878946e-01
-9.77610588e-01 1.69122860e-01 2.81611562e-01 5.94665587e-01
3.52173507e-01 2.29832381e-01 9.67530459e-02 -1.16675854e+00
7.02921152e-01 -4.02947575e-01 -9.89387333e-01 6.83332920e-01
-4.44616944e-01 6.31292641e-01 1.10835508e-01 -5.07525623e-01
-1.33468175e+00 1.24740653e-01 1.32836014e-01 7.91842937e-02
-2.54878759e-01 5.96612692e-01 -2.73626506e-01 3.75878870e-01
9.77815628e-01 2.77069539e-01 -3.04990083e-01 -2.20708489e-01
9.23402965e-01 5.34758925e-01 5.17122388e-01 -1.42149746e+00
4.10421968e-01 5.23660004e-01 -1.77399233e-01 -6.72519207e-01
-3.57640535e-01 8.32232833e-02 -4.44358736e-01 -1.88724436e-02
7.57010758e-01 -5.13157427e-01 -9.34531212e-01 3.82897109e-01
-1.22091913e+00 -6.10754967e-01 -4.34642524e-01 3.86792362e-01
-8.48238528e-01 2.44564697e-01 -6.27915204e-01 -1.20014119e+00
-8.26809406e-02 -1.20971656e+00 1.04480398e+00 -1.73102226e-02
-5.11730075e-01 -1.05662870e+00 1.61834747e-01 -1.66180730e-01
2.94919610e-01 -9.77827460e-02 1.51004136e+00 -5.56527257e-01
-8.20440710e-01 -5.34470320e-01 -2.23061368e-01 9.44051817e-02
-3.85677457e-01 7.25463867e-01 -7.72766650e-01 2.27615476e-01
-3.03930253e-01 -1.91677868e-01 6.18977070e-01 3.65995049e-01
1.19616759e+00 -2.96368837e-01 -4.66118813e-01 2.56001204e-01
1.20072556e+00 -2.16757685e-01 7.55454183e-01 -1.09083980e-01
2.52430648e-01 4.39371377e-01 1.67767212e-01 6.04486942e-01
7.05604076e-01 3.86055022e-01 1.06542632e-01 8.01234722e-01
6.21307552e-01 -1.36061355e-01 4.54667956e-01 1.54353946e-01
-6.63981661e-02 4.93190028e-02 -1.01012671e+00 3.91775042e-01
-2.20326519e+00 -1.11267424e+00 -2.54437923e-01 2.44004035e+00
1.35970044e+00 4.97444063e-01 -1.10055849e-01 -3.34373266e-01
7.39305019e-01 -3.78302604e-01 -5.15257895e-01 -5.13138115e-01
1.82621092e-01 3.14014554e-01 5.81850469e-01 6.87948883e-01
-9.66216266e-01 8.11216772e-01 7.21852112e+00 9.39191997e-01
-4.41960663e-01 1.05454221e-01 5.03854811e-01 -8.52840990e-02
-8.82030785e-01 5.63830554e-01 -8.40114474e-01 5.00272632e-01
1.25718021e+00 -2.25888461e-01 6.05052769e-01 1.09715772e+00
-2.68249869e-01 -4.61097926e-01 -1.71209359e+00 7.17322469e-01
-3.50137711e-01 -1.56343591e+00 3.59800570e-02 -5.06678782e-02
4.39925939e-01 -1.96526840e-01 -1.32351235e-01 6.76041007e-01
1.30543971e+00 -5.98611116e-01 1.17228663e+00 1.03719854e+00
4.70328242e-01 -9.25085366e-01 3.95353973e-01 4.63706940e-01
-7.09134877e-01 1.99824065e-01 -6.29425347e-01 -3.96155380e-02
1.59437675e-02 7.96807468e-01 -8.43333483e-01 1.95373461e-01
5.38234353e-01 1.49440616e-01 -4.20481354e-01 7.38656580e-01
-3.31422001e-01 7.76010156e-01 -7.84249723e-01 -4.89538431e-01
-8.39745849e-02 -1.29711822e-01 3.03716987e-01 1.44476259e+00
2.14018971e-01 -4.24361750e-02 -1.23220831e-01 1.54143596e+00
3.36895406e-01 -5.17684758e-01 -3.63986254e-01 8.56948420e-02
5.94311059e-01 9.60564792e-01 -7.40746021e-01 -6.57687485e-01
-1.01279832e-01 5.82742214e-01 3.30588669e-01 4.32609648e-01
-1.03378487e+00 -3.44541430e-01 6.33369803e-01 -1.43606484e-01
2.08939791e-01 -1.70346692e-01 -3.60323399e-01 -1.18655932e+00
-2.58715749e-01 -4.76873070e-01 5.59171021e-01 -7.90470302e-01
-1.37354743e+00 -3.13621223e-01 6.48369312e-01 -3.29655170e-01
-5.12013733e-01 -6.66829944e-01 -6.89459085e-01 8.27657461e-01
-6.61499500e-01 -1.00832772e+00 5.07200897e-01 3.10205281e-01
1.03869684e-01 3.29649657e-01 1.11354184e+00 -9.00437385e-02
-4.08354551e-01 3.77658099e-01 2.74085134e-01 -1.19224101e-01
3.29377115e-01 -1.65035903e+00 5.82855165e-01 7.58867204e-01
-7.37442225e-02 1.15682983e+00 8.89399946e-01 -7.23004699e-01
-1.69033802e+00 -9.71794426e-01 6.79970920e-01 -1.05050433e+00
8.91493559e-01 -7.70899653e-01 -7.72387445e-01 9.62043285e-01
-3.34658861e-01 -8.07556063e-02 7.06506312e-01 7.40521848e-01
-5.30080616e-01 3.60397585e-02 -1.34297431e+00 1.03497994e+00
7.06174970e-01 -7.49943912e-01 -6.95562363e-01 2.69195497e-01
6.30392611e-01 -2.36728057e-01 -6.86536014e-01 8.25928897e-02
7.86571264e-01 -7.67700911e-01 1.02748072e+00 -7.99562037e-01
1.01823278e-01 -5.17190218e-01 -4.23546910e-01 -6.10451281e-01
-2.67880768e-01 -5.25922894e-01 -4.61033285e-01 1.18297040e+00
5.51556349e-01 -7.05475867e-01 5.42325854e-01 1.47019482e+00
1.15858249e-01 -1.74772471e-01 -8.83522749e-01 -5.35273433e-01
1.27136588e-01 -1.30201924e+00 8.05412889e-01 7.22028017e-01
9.78491455e-02 -4.19760309e-02 8.37086514e-03 3.76739353e-01
9.93511915e-01 4.31155376e-02 8.34031880e-01 -1.35876715e+00
-6.79000258e-01 -4.23611671e-01 -3.92780215e-01 -6.09678090e-01
9.51969773e-02 -8.32034290e-01 3.01732898e-01 -1.31405222e+00
3.00897658e-01 -7.36328542e-01 8.52034390e-02 6.17245972e-01
-1.26775622e-01 -3.92458230e-01 -2.95695841e-01 -6.67889342e-02
-7.79685318e-01 2.63597161e-01 1.64246932e-01 -5.55219352e-02
-1.56417325e-01 1.23401172e-01 -4.99652952e-01 1.03103340e+00
6.02844536e-01 -5.05018651e-01 -4.46227878e-01 -3.29772681e-01
7.37816811e-01 6.97560757e-02 7.43780673e-01 -7.43619621e-01
4.15237010e-01 -3.48507851e-01 3.42166543e-01 -7.49220848e-01
1.63293764e-01 -4.24646795e-01 5.17729044e-01 2.76439697e-01
-5.85792840e-01 -3.27323467e-01 2.37327591e-01 7.38642693e-01
5.59423387e-01 -6.72521114e-01 4.67785776e-01 -3.06838512e-01
-6.03520453e-01 -6.74457662e-03 -7.37869561e-01 -1.14021473e-01
9.28602040e-01 -2.07993612e-02 -5.98349422e-02 7.66422749e-02
-7.70132601e-01 9.67321694e-02 5.54970980e-01 3.67503799e-02
4.86102968e-01 -1.02253795e+00 -6.14690244e-01 -1.30584771e-02
1.46423548e-01 1.87606409e-01 2.23461837e-01 4.64656442e-01
-6.42641246e-01 2.40052909e-01 2.40997329e-01 -6.39295816e-01
-8.51372063e-01 4.91150469e-01 1.40687481e-01 -9.74235777e-03
-3.09947431e-01 7.13158906e-01 6.03193380e-02 -1.08123887e+00
1.26000807e-01 -7.12609053e-01 5.45709252e-01 -1.20285966e-01
6.30563259e-01 4.30034012e-01 -1.16981417e-01 3.66709888e-01
-4.83269602e-01 -9.11081955e-02 -2.35240296e-01 -5.33427715e-01
1.15984964e+00 -3.28690279e-03 -5.32013416e-01 7.98886061e-01
4.45768535e-01 -3.75576854e-01 -1.24110794e+00 -9.11195502e-02
1.83069050e-01 -2.68612385e-01 -1.72555193e-01 -8.37346554e-01
-7.40989074e-02 7.73274839e-01 2.52025872e-01 1.58100277e-01
4.26512927e-01 2.10675448e-01 -1.01377882e-01 8.03861082e-01
6.87033832e-01 -1.09031856e+00 -8.01708162e-01 4.80865896e-01
4.81031626e-01 -7.72075236e-01 9.53478515e-02 -3.33153009e-02
-1.66725010e-01 1.15205622e+00 3.10895681e-01 1.27305910e-01
8.07145655e-01 3.60262245e-01 -8.14045906e-01 8.02020426e-04
-1.08580995e+00 3.85189988e-02 -8.87053311e-02 6.97776973e-01
2.12784767e-01 5.63552260e-01 2.89547294e-02 4.87244874e-01
-8.61784667e-02 2.49697536e-01 5.47809601e-01 1.14170706e+00
-4.65932757e-01 -1.08706725e+00 -5.09865582e-01 6.90557003e-01
-4.31030802e-02 -4.10367221e-01 3.97571623e-02 6.56695187e-01
9.68265906e-02 5.09908020e-01 3.14984620e-01 1.56788096e-01
-4.61086631e-02 3.17208081e-01 7.58329272e-01 -7.25269020e-01
2.23782901e-02 -3.38603020e-01 2.05846772e-01 -1.11285843e-01
9.15877968e-02 -9.98387814e-01 -1.18804240e+00 -6.17922127e-01
-2.62342095e-01 8.07451978e-02 7.41405666e-01 9.67961013e-01
3.60505044e-01 1.25911504e-01 -1.13597386e-01 -6.23289645e-01
-8.51980925e-01 -6.97270870e-01 -5.40011764e-01 -1.71625271e-01
-1.49253935e-01 -5.04699588e-01 -2.04748824e-01 2.68743843e-01] | [8.432159423828125, 6.4592814445495605] |
d1a90fc3-4413-41e6-af07-fed9a3bbbad3 | object-proposal-generation-using-two-stage | 1407.5242 | null | http://arxiv.org/abs/1407.5242v1 | http://arxiv.org/pdf/1407.5242v1.pdf | Object Proposal Generation using Two-Stage Cascade SVMs | Object proposal algorithms have shown great promise as a first step for
object recognition and detection. Good object proposal generation algorithms
require high object recall rate as well as low computational cost, because
generating object proposals is usually utilized as a preprocessing step. The
problem of how to accelerate the object proposal generation and evaluation
process without decreasing recall is thus of great interest. In this paper, we
propose a new object proposal generation method using two-stage cascade SVMs,
where in the first stage linear filters are learned for predefined quantized
scales/aspect-ratios independently, and in the second stage a global linear
classifier is learned across all the quantized scales/aspect-ratios for
calibration, so that all the proposals can be compared properly. The proposals
with highest scores are our final output. Specifically, we explain our
scale/aspect-ratio quantization scheme, and investigate the effects of
combinations of $\ell_1$ and $\ell_2$ regularizers in cascade SVMs with/without
ranking constraints in learning. Comprehensive experiments on VOC2007 dataset
are conducted, and our results achieve the state-of-the-art performance with
high object recall rate and high computational efficiency. Besides, our method
has been demonstrated to be suitable for not only class-specific but also
generic object proposal generation. | ['Ziming Zhang', 'Philip H. S. Torr'] | 2014-07-20 | null | null | null | null | ['object-proposal-generation'] | ['computer-vision'] | [ 9.01865661e-02 -4.27344352e-01 -4.59574610e-01 -5.67944944e-01
-7.53475130e-01 -1.48396239e-01 3.61425966e-01 3.00747901e-01
-7.03839421e-01 5.14258504e-01 -4.36231762e-01 -3.07242908e-02
3.98289599e-02 -9.42831516e-01 -6.15230441e-01 -8.28169525e-01
3.15808952e-01 4.69335377e-01 8.81348491e-01 6.91627152e-03
2.46580482e-01 7.45087743e-01 -1.61556649e+00 2.40579084e-01
6.95567131e-01 1.45505083e+00 3.24099600e-01 4.11327124e-01
-1.83054402e-01 2.31305078e-01 -7.41156936e-01 -6.84188426e-01
2.79663295e-01 -2.26584285e-01 -3.99293661e-01 2.69859403e-01
7.86899090e-01 -2.66577750e-01 -8.20329785e-02 1.14873397e+00
5.47438800e-01 3.89205366e-02 6.87229812e-01 -1.15550458e+00
-5.20746291e-01 4.27508861e-01 -6.47933006e-01 1.77220210e-01
-3.22707444e-01 2.69663811e-01 1.12445760e+00 -1.23022568e+00
2.14470416e-01 1.08532965e+00 3.08964700e-01 4.84420508e-01
-1.25019968e+00 -1.00303638e+00 3.56693804e-01 3.61944765e-01
-1.52522349e+00 -1.98062748e-01 7.84435093e-01 -4.22796577e-01
7.38501072e-01 1.62124753e-01 7.92507410e-01 5.66736281e-01
3.78587395e-02 1.03618765e+00 8.29878628e-01 -3.64243329e-01
3.54612052e-01 5.31963944e-01 1.79213375e-01 7.88318992e-01
4.72562909e-01 -1.72184050e-01 -5.04776537e-01 1.46232873e-01
9.84639049e-01 8.88602808e-02 -3.85804884e-02 -5.22867322e-01
-9.59086061e-01 1.05789959e+00 9.60085034e-01 6.61824420e-02
-1.27019837e-01 7.91963339e-02 2.14133218e-01 -9.54642892e-02
2.39976823e-01 3.81772935e-01 -2.96397179e-01 5.31413257e-01
-9.67529058e-01 2.68215954e-01 2.35291153e-01 1.06146789e+00
7.82802403e-01 2.51221880e-02 -4.06910688e-01 1.08486700e+00
3.85782957e-01 3.67908746e-01 5.03201365e-01 -4.36878800e-01
6.06663942e-01 9.20221567e-01 1.09188631e-01 -8.52228284e-01
-2.28315234e-01 -8.23524117e-01 -7.20543802e-01 2.28205428e-01
3.75221997e-01 2.70014375e-01 -9.99783456e-01 1.39354074e+00
4.65476304e-01 5.66963963e-02 -2.06753910e-01 1.05767763e+00
9.98426974e-01 5.34689248e-01 2.14784756e-01 -1.67360261e-01
1.46815550e+00 -1.20119107e+00 -2.94192344e-01 -3.46739978e-01
3.65438670e-01 -9.18298841e-01 1.18181133e+00 3.38649452e-01
-1.01428032e+00 -9.66708779e-01 -1.27250314e+00 8.64830916e-04
-2.25659862e-01 9.34717715e-01 6.59837365e-01 7.81148732e-01
-5.27995706e-01 3.96115661e-01 -8.22587192e-01 4.49519418e-03
6.89627945e-01 5.82787097e-01 -2.55722227e-03 1.23174869e-01
-8.12677801e-01 7.62997150e-01 6.52063310e-01 2.65317708e-01
-5.43029308e-01 -3.03025365e-01 -6.85494602e-01 3.94712426e-02
2.73581564e-01 -2.90675312e-01 1.18845701e+00 -8.21649075e-01
-1.30996132e+00 7.14028716e-01 7.16639403e-03 -4.44293320e-01
4.65776950e-01 -1.03393108e-01 -4.20142353e-01 -1.96816269e-02
1.06876910e-01 1.08897638e+00 9.19546902e-01 -1.04627788e+00
-1.10393775e+00 -2.51062959e-01 -6.33073449e-02 1.45123988e-01
-6.09888971e-01 -3.65180708e-02 -7.53497839e-01 -6.00195169e-01
6.55436337e-01 -7.84448504e-01 -2.17691496e-01 3.26747298e-01
-1.76458955e-01 -4.12278265e-01 7.97811031e-01 -3.31379622e-01
1.19904983e+00 -1.93180501e+00 -1.07475087e-01 1.91231489e-01
-1.32010683e-01 4.72843945e-01 -8.63983929e-02 -2.68455207e-01
2.20008865e-01 -1.28105223e-01 -1.21957183e-01 -2.97217757e-01
-3.21263582e-01 -1.22651331e-01 -3.45571458e-01 2.26130858e-01
7.73033738e-01 7.77426183e-01 -6.69206083e-01 -6.28645360e-01
4.00626749e-01 3.26315075e-01 -6.10167265e-01 2.31517136e-01
-2.11212277e-01 2.49976269e-03 -5.11989355e-01 8.42424512e-01
5.83007634e-01 -3.60786140e-01 8.17157794e-03 -7.09099650e-01
4.07635756e-02 1.69416443e-01 -1.51384020e+00 1.06477511e+00
-2.28283077e-01 4.65007335e-01 -2.85584569e-01 -9.61959302e-01
1.30943120e+00 -6.85634091e-02 1.25209153e-01 -6.21498346e-01
8.19319263e-02 3.34209114e-01 1.28668517e-01 -1.14070959e-01
6.24549985e-01 -1.63353384e-01 -5.71176149e-02 -8.51024538e-02
1.74780786e-01 -3.80705297e-01 3.43858004e-01 -1.04214154e-01
3.90715897e-01 -2.18936950e-02 3.21616530e-01 -1.35385673e-02
7.12191224e-01 1.00530557e-01 5.86839020e-01 6.61910892e-01
-1.48656875e-01 5.64528823e-01 3.34246606e-01 -4.37732190e-01
-9.94168341e-01 -1.04763246e+00 -3.81172419e-01 1.00049686e+00
5.09072483e-01 -9.48944837e-02 -5.59977531e-01 -8.10681641e-01
-4.47311997e-03 5.04860997e-01 -3.23794007e-01 -1.68992266e-01
-8.17112446e-01 -1.05764925e+00 1.83234826e-01 9.12811756e-01
7.46325314e-01 -1.03522158e+00 -6.08723581e-01 3.49945992e-01
1.28030851e-01 -1.10767591e+00 -4.16379929e-01 2.19430655e-01
-1.17930186e+00 -9.94760871e-01 -7.91058004e-01 -9.71316695e-01
1.01216185e+00 3.47033918e-01 7.49419689e-01 2.25432552e-02
-4.82108146e-01 -2.41647765e-01 -2.34910160e-01 -3.79787922e-01
-3.27747576e-02 1.54287845e-01 -9.42422636e-03 1.92286313e-01
1.93580553e-01 4.17036489e-02 -8.05492103e-01 8.10831010e-01
-7.48309076e-01 1.22132182e-01 9.10167098e-01 1.01050973e+00
8.86141062e-01 -8.85963291e-02 5.72047114e-01 -6.19902194e-01
1.52276874e-01 1.64554358e-01 -1.03558195e+00 3.53272319e-01
-4.79671657e-01 2.32096948e-02 5.42493761e-01 -7.00405955e-01
-9.56107318e-01 3.28526348e-01 -2.40810007e-01 -2.57220566e-01
1.41848624e-01 1.03640012e-01 -3.87619466e-01 -2.45483801e-01
7.73698449e-01 1.24114215e-01 -3.38412046e-01 -3.65091652e-01
2.91795313e-01 5.84153295e-01 2.92145431e-01 -4.18334126e-01
8.98206651e-01 3.25122237e-01 -1.01384737e-01 -8.15579414e-01
-8.31049204e-01 -5.26390612e-01 -7.22521901e-01 -1.00753337e-01
5.74521542e-01 -1.05746448e+00 -4.61710274e-01 4.08569008e-01
-9.54955637e-01 -1.52420238e-01 -3.16338629e-01 6.86732590e-01
-2.03141659e-01 -4.35196459e-02 -4.33215499e-01 -6.19815946e-01
-3.25796813e-01 -1.44008553e+00 9.85110462e-01 6.06694698e-01
9.13851112e-02 -3.71027887e-01 -5.73205948e-01 3.91858935e-01
3.74867111e-01 -3.65673661e-01 8.01937103e-01 -5.78871667e-01
-9.92827654e-01 -3.48018408e-01 -7.21668720e-01 5.75908124e-01
1.20372690e-01 -8.62340257e-02 -8.41264546e-01 -4.81365919e-01
-1.94653898e-01 -4.14156437e-01 1.06266928e+00 3.46992880e-01
1.29232132e+00 -1.46515757e-01 -3.74173462e-01 5.42809725e-01
1.39151001e+00 2.29632065e-01 3.88913453e-01 2.99538165e-01
6.42925441e-01 2.67459363e-01 8.53580594e-01 2.90656567e-01
-4.30409387e-02 9.05149460e-01 1.50693730e-01 -8.47832486e-02
-2.99491197e-01 -1.48117125e-01 1.09657511e-01 4.82972980e-01
8.08791071e-02 -1.35001183e-01 -5.87352455e-01 4.26180065e-01
-1.55381262e+00 -5.79873621e-01 -8.50209668e-02 2.35866976e+00
7.88827837e-01 5.59189379e-01 2.59195060e-01 2.71818399e-01
7.29797482e-01 2.52067000e-02 -5.90062737e-01 1.50183365e-01
-4.06044237e-02 3.51638496e-01 5.31994164e-01 2.87047565e-01
-1.35238481e+00 1.03331959e+00 5.33042002e+00 1.17784572e+00
-1.55863416e+00 3.69219072e-02 6.71498716e-01 -1.20059654e-01
1.64577439e-01 -9.97835770e-02 -1.60415089e+00 3.66268396e-01
3.20010841e-01 3.15369368e-01 -2.71135062e-01 1.09741819e+00
-2.25152552e-01 -2.52634317e-01 -9.59793866e-01 9.72034395e-01
-1.86242573e-02 -1.32917523e+00 2.54983962e-01 -3.52752715e-01
6.82608068e-01 -2.79933661e-01 1.47271618e-01 3.09282213e-01
-1.18016466e-01 -6.02988601e-01 9.80260253e-01 1.31198540e-01
7.16597736e-01 -6.70814693e-01 7.19381928e-01 2.25735918e-01
-1.45025587e+00 -2.39536270e-01 -6.40322804e-01 4.14119840e-01
1.52499294e-02 6.47133470e-01 -1.08261859e+00 1.92092970e-01
6.46964371e-01 3.73032629e-01 -8.20803404e-01 1.50817299e+00
-4.75150675e-01 6.01714015e-01 -4.13482726e-01 -5.44463813e-01
-3.73272062e-03 -4.64037657e-02 1.72847033e-01 1.30048704e+00
2.85843194e-01 1.59464106e-01 3.31108004e-01 7.61540651e-01
-1.66065022e-01 4.07457173e-01 2.21876189e-01 2.28252411e-01
5.06206751e-01 1.29774904e+00 -1.28595686e+00 -4.70028430e-01
-2.52990365e-01 7.28339970e-01 3.09871733e-01 2.41769813e-02
-9.50792730e-01 -4.10133839e-01 3.06550264e-01 2.65369803e-01
5.57559550e-01 -3.12957346e-01 -3.20209235e-01 -1.04590380e+00
2.15549394e-01 -4.84430462e-01 4.26971018e-01 -3.98886651e-01
-1.06945431e+00 4.37829942e-01 -5.17502576e-02 -1.45625508e+00
2.63733119e-01 -8.42513859e-01 -4.71301407e-01 6.45994723e-01
-1.47886097e+00 -1.26434875e+00 -4.62518334e-01 2.82991648e-01
8.54241014e-01 -2.01928705e-01 4.33767051e-01 4.22802091e-01
-6.98355794e-01 8.82104278e-01 -5.74744716e-02 2.56168246e-01
5.90508819e-01 -9.45040941e-01 3.18926334e-01 6.99023783e-01
4.82556254e-01 5.21798909e-01 3.36499870e-01 -5.98493636e-01
-1.05635870e+00 -1.26039922e+00 5.87716520e-01 -1.40037686e-01
3.34724545e-01 -3.90249044e-01 -9.42909777e-01 2.99822450e-01
-5.85916817e-01 2.44400948e-01 3.71328771e-01 -1.15121819e-01
-3.52324098e-01 -6.31016731e-01 -1.04125106e+00 4.53893393e-01
7.07656741e-01 -1.50241494e-01 -3.77700359e-01 4.11424875e-01
5.33544242e-01 -5.01104116e-01 -5.62567830e-01 5.73594511e-01
7.29204059e-01 -7.98502207e-01 1.11448467e+00 -2.67668545e-01
1.51372924e-01 -5.63503861e-01 -1.80438831e-01 -7.58535326e-01
-3.98720264e-01 7.52047673e-02 5.39966859e-02 1.11798584e+00
6.63613856e-01 -4.82324332e-01 1.08448195e+00 2.19530120e-01
-4.72126752e-02 -8.95133674e-01 -8.83893728e-01 -9.57318604e-01
-2.79753417e-01 -3.75903308e-01 2.40259081e-01 3.73637617e-01
-6.14501715e-01 3.18650663e-01 -1.81352928e-01 1.89744443e-01
6.63302064e-01 2.94120252e-01 7.80291975e-01 -1.06062937e+00
-3.79768401e-01 -6.10435724e-01 -5.25253117e-01 -1.37761390e+00
-3.91617715e-01 -7.66162097e-01 1.69036523e-01 -1.29913414e+00
1.86248392e-01 -8.59886885e-01 -2.98276842e-01 4.15450335e-01
-3.71570617e-01 6.80316329e-01 2.32855633e-01 4.30439383e-01
-4.67851788e-01 6.22599900e-01 1.14224494e+00 -4.08111870e-01
-4.48996693e-01 4.59284097e-01 -5.29680669e-01 7.39520431e-01
5.91618717e-01 -4.19241697e-01 -2.73481846e-01 -3.03455830e-01
-2.10089445e-01 -2.68869072e-01 4.83720005e-01 -1.29759085e+00
2.37215474e-01 -2.57236242e-01 8.55264843e-01 -8.42786610e-01
6.34863019e-01 -5.72846413e-01 -1.56240821e-01 6.35309100e-01
-3.52068156e-01 -3.10334891e-01 1.73434302e-01 4.78592873e-01
-2.87747175e-01 -3.89992416e-01 1.25538516e+00 1.21727340e-01
-8.02005291e-01 4.01469588e-01 1.20881207e-01 -2.57414103e-01
1.04156566e+00 -3.58237892e-01 1.49516342e-02 2.14770988e-01
-5.56942940e-01 1.09657072e-01 2.17312291e-01 4.92940038e-01
8.24430823e-01 -1.35305083e+00 -6.15987659e-01 3.76367182e-01
3.41656923e-01 1.63542718e-01 1.06413802e-02 4.73691106e-01
-3.10364723e-01 3.27451140e-01 -2.09918454e-01 -9.55058217e-01
-1.39929557e+00 3.90300810e-01 2.29908392e-01 -3.65196541e-02
-3.11057061e-01 1.40891695e+00 1.98985457e-01 -1.03383489e-01
3.93990904e-01 -5.23921251e-01 -1.48252293e-01 2.03247219e-01
4.95734841e-01 1.81508511e-01 3.07100803e-01 -6.03893578e-01
-3.21465254e-01 8.80893946e-01 -3.96672308e-01 5.81029095e-02
1.04972827e+00 2.93420702e-01 2.53114343e-01 1.35061502e-01
1.07907784e+00 -2.59158939e-01 -1.23894119e+00 -4.28777695e-01
7.81660154e-02 -6.24233246e-01 -1.55743230e-02 -5.04880905e-01
-1.14942050e+00 1.04063249e+00 9.77566779e-01 -1.21236809e-01
1.01929915e+00 8.84047821e-02 6.47467256e-01 4.67657655e-01
4.20779943e-01 -1.09171128e+00 4.92162615e-01 2.28230581e-01
8.97173405e-01 -1.41476500e+00 3.83855373e-01 -6.95181966e-01
-5.70209444e-01 1.06204188e+00 9.99795377e-01 -2.09975287e-01
4.60942179e-01 5.67557709e-03 -1.16557397e-01 1.29451796e-01
-3.54842782e-01 -1.97219104e-01 6.98139846e-01 2.59591371e-01
3.86728793e-01 1.48557946e-01 -4.32041496e-01 6.10121965e-01
-1.23502634e-01 -1.63083538e-01 9.86480638e-02 7.54153490e-01
-8.94322932e-01 -1.22182631e+00 -4.16694552e-01 7.36012101e-01
-1.54250354e-01 1.54960036e-01 -6.79242238e-03 7.21739769e-01
5.04400790e-01 6.26889348e-01 5.40124997e-02 -3.28558594e-01
5.16203046e-01 -7.08660111e-02 5.04961014e-01 -7.46693075e-01
-4.87525254e-01 2.35453531e-01 -1.73984095e-01 -2.70192415e-01
-2.31767669e-01 -5.21707594e-01 -1.35250700e+00 4.45573449e-01
-1.02850354e+00 7.46829882e-02 8.72992992e-01 7.94084609e-01
-8.05619173e-04 4.00739282e-01 5.90927720e-01 -8.69349480e-01
-7.25950241e-01 -9.37778413e-01 -3.78835529e-01 2.74030238e-01
6.80425912e-02 -7.95650542e-01 -5.75975589e-02 9.66039672e-02] | [9.228340148925781, 0.86128169298172] |
595112b8-bd95-4b32-be1d-fda38a5feb7d | on-the-relationship-between-rnn-hidden-state | 2306.16854 | null | https://arxiv.org/abs/2306.16854v1 | https://arxiv.org/pdf/2306.16854v1.pdf | On the Relationship Between RNN Hidden State Vectors and Semantic Ground Truth | We examine the assumption that the hidden-state vectors of recurrent neural networks (RNNs) tend to form clusters of semantically similar vectors, which we dub the clustering hypothesis. While this hypothesis has been assumed in the analysis of RNNs in recent years, its validity has not been studied thoroughly on modern neural network architectures. We examine the clustering hypothesis in the context of RNNs that were trained to recognize regular languages. This enables us to draw on perfect ground-truth automata in our evaluation, against which we can compare the RNN's accuracy and the distribution of the hidden-state vectors. We start with examining the (piecewise linear) separability of an RNN's hidden-state vectors into semantically different classes. We continue the analysis by computing clusters over the hidden-state vector space with multiple state-of-the-art unsupervised clustering approaches. We formally analyze the accuracy of computed clustering functions and the validity of the clustering hypothesis by determining whether clusters group semantically similar vectors to the same state in the ground-truth model. Our evaluation supports the validity of the clustering hypothesis in the majority of examined cases. We observed that the hidden-state vectors of well-trained RNNs are separable, and that the unsupervised clustering techniques succeed in finding clusters of similar state vectors. | ['Thomas Pock', 'Bernhard K. Aichernig', 'Ingo Pill', 'Martin Tappler', 'Edi Muškardin'] | 2023-06-29 | null | null | null | null | ['clustering'] | ['methodology'] | [ 1.23931989e-01 1.37420148e-01 -1.71766013e-01 -3.41662526e-01
-2.65442491e-01 -7.42258847e-01 8.18279088e-01 8.39400291e-03
-3.26469690e-01 2.46329576e-01 2.66717285e-01 -6.72333241e-01
-1.05315901e-01 -5.08732259e-01 -3.92200142e-01 -9.24014926e-01
-1.00963131e-01 6.87463284e-01 3.15180987e-01 -1.56016588e-01
9.62284803e-02 5.93695104e-01 -1.98855174e+00 1.46694213e-01
4.63612437e-01 6.17646337e-01 6.82881698e-02 6.14061832e-01
-2.71435887e-01 1.00788260e+00 -7.45927811e-01 1.64781272e-01
-2.47941129e-02 -8.52940440e-01 -1.13769090e+00 1.92028522e-01
2.11985707e-01 3.90190214e-01 -3.26685399e-01 1.29222023e+00
-8.65424126e-02 2.87269562e-01 7.98508108e-01 -1.23871970e+00
-2.30903283e-01 7.93195128e-01 2.30815098e-01 1.79055303e-01
3.01556252e-02 -3.69493663e-01 1.17257571e+00 -3.63425016e-01
8.80652845e-01 9.62128520e-01 5.94701886e-01 8.50899220e-01
-1.37293243e+00 -4.39897180e-01 1.67977780e-01 4.58146557e-02
-1.51400244e+00 -7.53378749e-01 7.36266971e-01 -6.26265764e-01
1.13428962e+00 2.49640614e-01 7.37176895e-01 8.24533582e-01
-7.60121197e-02 7.85876632e-01 9.82286215e-01 -7.54626632e-01
4.04490590e-01 1.67498842e-01 7.58608699e-01 7.77720332e-01
1.54713109e-01 1.16775157e-02 -5.49287163e-02 5.45605198e-02
6.53844953e-01 9.46739614e-02 -4.27824482e-02 -5.48164427e-01
-1.14309645e+00 7.60893404e-01 1.56949088e-01 1.07831037e+00
-1.58336505e-01 2.04274848e-01 3.30707103e-01 2.85089403e-01
2.45470926e-01 2.24689990e-01 -1.35176331e-01 -5.97400554e-02
-1.18444645e+00 -2.84963548e-01 7.81043828e-01 9.24050331e-01
8.63175392e-01 3.75344992e-01 3.90142202e-01 5.44853032e-01
3.47465903e-01 2.20185459e-01 8.55817735e-01 -8.62728953e-01
7.94476643e-02 8.15680981e-01 -3.18480909e-01 -9.18804288e-01
-3.23152781e-01 -5.30833006e-01 -9.57807779e-01 -4.50709835e-02
3.84310365e-01 -6.36963546e-03 -9.62874770e-01 1.94390452e+00
-2.32467413e-01 3.10345143e-01 6.12421215e-01 3.99974197e-01
3.99349213e-01 8.09053898e-01 -2.16216415e-01 -6.16277575e-01
8.50430787e-01 -5.45991421e-01 -8.71890187e-01 2.52361298e-01
9.97227192e-01 -3.22964609e-01 8.51442575e-01 1.06394261e-01
-8.73723209e-01 -6.25275612e-01 -9.76368666e-01 3.95999610e-01
-4.06285018e-01 2.70914435e-01 4.39160317e-01 7.24794626e-01
-1.43595564e+00 7.39088774e-01 -1.44281924e+00 -6.91437602e-01
-1.46539778e-01 5.31316757e-01 -2.08075494e-01 4.50736284e-01
-9.87374306e-01 1.05958843e+00 7.31809378e-01 4.26250920e-02
-9.93653476e-01 2.33541951e-01 -6.22974873e-01 -1.04661435e-01
1.15331694e-01 -9.90805104e-02 1.11008263e+00 -1.37609541e+00
-1.32108903e+00 7.78607607e-01 -5.08835077e-01 -4.25123572e-01
-1.54910922e-01 4.20772374e-01 -7.69015551e-01 1.58129439e-01
-2.86758095e-01 3.60694766e-01 5.62859774e-01 -1.62305474e+00
-5.49673200e-01 -3.77159774e-01 -4.12338048e-01 -3.77107486e-02
-2.23538592e-01 8.92680883e-02 -3.27603102e-01 -4.16117787e-01
7.82805681e-01 -1.33270264e+00 -2.57195950e-01 -9.28274572e-01
-7.03181624e-01 -4.56328213e-01 7.02558160e-01 -9.41193700e-02
1.49767661e+00 -2.24071646e+00 2.67673373e-01 6.95455849e-01
2.73943484e-01 1.13331862e-01 1.87664047e-01 4.32834804e-01
-6.20437384e-01 1.31411180e-01 -3.55016798e-01 -6.70685589e-01
2.06742845e-02 7.31932640e-01 -6.15850210e-01 7.82835126e-01
-1.11661464e-01 7.18634129e-01 -7.54794002e-01 -6.01436615e-01
3.11091654e-02 3.07567656e-01 -2.76610255e-01 1.02367021e-01
-4.23817560e-02 6.43528551e-02 1.31763332e-03 1.69890165e-01
-1.78432196e-01 -2.65313089e-01 5.58657169e-01 -8.41569304e-02
-4.14080322e-01 6.54287577e-01 -1.00824976e+00 1.22530055e+00
1.50148749e-01 1.01927531e+00 -4.37646329e-01 -1.24851716e+00
1.04267085e+00 5.06806731e-01 5.56236148e-01 -1.06457584e-01
3.23833048e-01 9.38254446e-02 4.03111994e-01 8.33248720e-02
5.40860355e-01 -5.50843894e-01 -2.25728482e-01 8.57963026e-01
5.71853757e-01 4.09069240e-01 -3.22865136e-02 2.01245323e-01
9.51430559e-01 -2.46004254e-01 6.38027117e-02 -5.98651826e-01
2.27788329e-01 1.74669832e-01 3.42523068e-01 5.25553405e-01
-1.27983227e-01 4.78366226e-01 5.32156825e-01 -4.13935110e-02
-1.12571168e+00 -1.15755200e+00 -1.24525011e-01 9.58937585e-01
-1.66280657e-01 -4.94847536e-01 -9.18982625e-01 -3.58328998e-01
-5.95101535e-01 6.55989170e-01 -7.25261509e-01 -2.91923046e-01
-3.17093343e-01 -5.16813636e-01 9.95844960e-01 4.18165833e-01
4.13668193e-02 -1.15442610e+00 -6.27511501e-01 -3.66278030e-02
-1.52924746e-01 -7.66290247e-01 1.42019153e-01 8.49428594e-01
-1.06891561e+00 -7.91090429e-01 -5.10574698e-01 -1.11373663e+00
8.95881891e-01 -1.08465098e-01 1.01617765e+00 1.36735067e-01
2.20006019e-01 4.75834429e-01 -3.55341494e-01 -1.95940230e-02
-9.54900384e-01 3.15417439e-01 5.72855592e-01 7.87032396e-02
4.28769588e-01 -5.09767950e-01 3.60357873e-02 4.91067708e-01
-1.12988067e+00 -2.03058034e-01 2.07003683e-01 1.94821805e-01
6.70218647e-01 6.20093703e-01 9.13441926e-02 -8.33973825e-01
4.35951740e-01 -4.27145064e-01 -5.19847393e-01 3.28095496e-01
-5.19094825e-01 5.48308492e-01 7.32571602e-01 -4.70844984e-01
-5.55523634e-01 3.01899374e-01 -3.35315503e-02 -8.67723227e-01
-5.15336156e-01 5.32636285e-01 -1.30322456e-01 2.73470700e-01
5.49924493e-01 7.10547686e-01 9.98833627e-02 -4.01919395e-01
4.13989544e-01 5.82395732e-01 8.12341213e-01 -4.44754869e-01
7.12898254e-01 4.87088233e-01 -1.72184035e-01 -1.14447904e+00
-5.32216489e-01 -7.67574549e-01 -9.17790174e-01 -2.57595420e-01
1.00920010e+00 -5.87654412e-01 -5.52977085e-01 3.07674080e-01
-1.13658142e+00 -3.18919152e-01 -3.64763469e-01 4.17612165e-01
-7.49896407e-01 5.60318351e-01 -5.59670329e-01 -1.15017760e+00
1.98270738e-01 -1.04836524e+00 4.39887643e-01 -2.64078081e-01
-5.83608031e-01 -1.31737399e+00 5.45947134e-01 -2.17595518e-01
-1.88064963e-01 2.55291769e-03 1.09837818e+00 -1.25068986e+00
-3.66165526e-02 -4.42879647e-02 4.51366425e-01 4.24590528e-01
2.31536180e-01 4.01030242e-01 -9.89503145e-01 -2.18957379e-01
3.54421318e-01 1.58388898e-01 9.91712093e-01 2.90938735e-01
6.28236175e-01 -3.03244472e-01 -5.20675421e-01 1.02145210e-01
1.39835978e+00 4.98799741e-01 6.13848686e-01 -5.58462143e-02
7.44319022e-01 9.06714916e-01 4.60027792e-02 -1.78056732e-01
-3.42387408e-02 5.54780364e-01 1.93269521e-01 1.00564800e-01
2.63075590e-01 -2.61084914e-01 6.69302464e-01 1.67364812e+00
-2.20586926e-01 -2.03186318e-01 -1.38251400e+00 7.38140523e-01
-2.07117653e+00 -1.22348738e+00 -9.41592827e-02 2.05321455e+00
7.65776396e-01 3.78721744e-01 4.66255963e-01 6.75106049e-01
9.61006105e-01 7.55119231e-03 -1.25017032e-01 -4.35080707e-01
-2.04688594e-01 -1.23430137e-02 2.97691286e-01 5.15505314e-01
-8.75895858e-01 1.04850614e+00 7.38367748e+00 6.61170363e-01
-1.02804732e+00 -6.87712878e-02 3.50910425e-01 2.78545916e-03
-3.75556022e-01 1.87476993e-01 -9.02092695e-01 2.91133106e-01
1.57284451e+00 2.21968502e-01 3.29975605e-01 7.05408335e-01
-5.26934266e-02 2.79901564e-01 -1.26130676e+00 6.49929225e-01
1.28380448e-01 -1.32486987e+00 1.62603438e-01 1.88091725e-01
8.20035696e-01 2.14821339e-01 5.86693436e-02 1.99296385e-01
6.83523178e-01 -1.11403358e+00 8.03140223e-01 6.28126085e-01
6.31041706e-01 -9.23943520e-01 5.00329018e-01 4.05453563e-01
-1.15167904e+00 1.69455335e-01 -3.27289909e-01 3.44302133e-02
-1.62239790e-01 2.99399227e-01 -7.70024896e-01 2.26931483e-01
3.01443934e-01 8.14893246e-01 -4.29558963e-01 4.26208735e-01
1.29498363e-01 9.92249727e-01 -3.44971955e-01 -1.43796712e-01
6.74565136e-01 -4.61511254e-01 2.98818052e-01 1.23769045e+00
-3.30372453e-02 1.02299601e-01 -3.94780114e-02 8.94380748e-01
3.65683883e-01 -8.36342722e-02 -7.21228004e-01 -4.29974318e-01
2.85005391e-01 8.08463991e-01 -1.32858932e+00 -6.48990452e-01
-1.14881404e-01 6.11060560e-01 3.29790175e-01 3.45419765e-01
-5.62640190e-01 -2.96643794e-01 7.30426788e-01 -6.72208965e-02
3.59095812e-01 -5.49796164e-01 -5.60452417e-02 -1.13156438e+00
-3.52210850e-01 -7.27535009e-01 1.20101303e-01 -5.28002977e-01
-9.75861073e-01 1.16170955e+00 1.53094053e-01 -1.19870639e+00
-6.22716784e-01 -5.11910558e-01 -5.76701403e-01 4.30924535e-01
-7.67996192e-01 -6.38450861e-01 4.54193294e-01 9.94033217e-01
2.77495027e-01 -4.84059691e-01 9.64310944e-01 -1.59672707e-01
-7.93068409e-01 4.02727008e-01 4.35904503e-01 5.95772505e-01
-4.36727777e-02 -1.28650141e+00 3.34112942e-01 9.80193496e-01
6.86495662e-01 1.23414958e+00 8.94941330e-01 -6.37874901e-01
-1.09985042e+00 -1.11311698e+00 1.04232550e+00 -5.73014021e-01
8.81056726e-01 -3.72543752e-01 -1.04327488e+00 9.58311975e-01
1.89672224e-03 -2.52262205e-01 8.84587169e-01 2.98180580e-01
-3.32195640e-01 3.33417952e-01 -4.04453754e-01 7.28550315e-01
8.87356400e-01 -1.13900614e+00 -1.05564797e+00 -1.69373024e-02
5.10300696e-01 3.87117982e-01 -8.50018382e-01 3.17071490e-02
2.93302387e-01 -1.07626998e+00 5.70619702e-01 -1.00623858e+00
2.32566714e-01 -5.43332815e-01 -4.55986410e-01 -1.06998181e+00
-3.92325044e-01 -4.35786486e-01 -4.61318269e-02 1.11208296e+00
5.78165293e-01 -4.28105950e-01 9.17824507e-01 3.07967663e-01
-2.63928056e-01 -5.21311045e-01 -9.21908855e-01 -9.67646480e-01
7.66047463e-02 -6.96724594e-01 1.02690734e-01 1.26621628e+00
4.71898973e-01 4.94975030e-01 1.16574898e-01 1.98133752e-01
4.01621222e-01 1.51197702e-01 3.20173562e-01 -1.36983085e+00
-1.91727698e-01 -7.69134104e-01 -7.74518847e-01 -8.80478203e-01
7.54968226e-01 -1.01313829e+00 3.75126898e-01 -1.30554295e+00
-9.27548110e-03 -3.09109539e-01 -5.62850237e-01 6.11032367e-01
3.62737060e-01 1.02346927e-01 1.82129651e-01 9.07761037e-01
-9.12454963e-01 3.03844810e-01 3.42787057e-01 1.54667869e-01
-3.04622531e-01 1.42811015e-01 -2.35426992e-01 8.44789326e-01
8.81789684e-01 -6.10429466e-01 -5.51251113e-01 4.18806672e-02
3.50614339e-01 -1.01885363e-01 1.60704061e-01 -1.33474159e+00
3.12667578e-01 1.97958380e-01 -7.95018673e-02 -7.01290965e-01
1.24318749e-01 -1.07863164e+00 4.19651747e-01 5.86898744e-01
-5.99115670e-01 1.57976776e-01 3.95897068e-02 4.49252248e-01
-3.11787963e-01 -2.52907366e-01 7.72509813e-01 -9.84518826e-02
-5.68148196e-01 -1.27076656e-01 -1.10027242e+00 -1.23278260e-01
8.42079818e-01 -5.78204215e-01 1.70104250e-01 -4.47750270e-01
-9.61512923e-01 -3.47395360e-01 6.17317438e-01 2.14233771e-01
4.64205444e-01 -1.26516545e+00 -1.36617765e-01 1.88244730e-01
1.22991741e-01 -3.77356172e-01 -4.49301571e-01 6.99019432e-01
-3.73211026e-01 5.63646317e-01 -1.06365234e-01 -9.05825913e-01
-1.30350745e+00 7.97955215e-01 4.95877951e-01 -2.88301967e-02
-3.83150399e-01 2.61590302e-01 -1.05575606e-01 -2.96452761e-01
3.90234023e-01 -8.57316256e-01 -4.54659522e-01 2.39840567e-01
9.21961069e-02 2.33504206e-01 -4.58166897e-02 -1.06364048e+00
-2.83251017e-01 4.28518891e-01 2.68346220e-02 -3.91289294e-01
1.08222532e+00 -6.62740171e-02 -2.68680513e-01 1.44550061e+00
1.36103868e+00 -3.35481644e-01 -6.90901816e-01 -2.04308540e-01
4.01021630e-01 5.38746595e-01 -5.94744161e-02 -1.38703734e-01
-9.95252490e-01 8.52585495e-01 5.18343449e-01 8.75189304e-01
8.08601975e-01 2.81343549e-01 4.06473517e-01 6.43431067e-01
1.85337022e-01 -8.42428565e-01 -1.76143080e-01 9.49356735e-01
6.47725835e-02 -6.42581165e-01 -3.46343994e-01 1.01685226e-01
-7.38950133e-01 9.05491710e-01 1.40454015e-02 -3.58486295e-01
8.05823803e-01 2.57306308e-01 2.53918976e-01 -3.31006765e-01
-8.67640138e-01 -4.53985989e-01 3.16281676e-01 3.48252714e-01
4.08133626e-01 8.10988396e-02 2.67053932e-01 3.45842063e-01
-5.46459675e-01 -6.42023683e-01 4.36467856e-01 9.40207481e-01
-8.05566311e-01 -7.90120840e-01 -2.04258680e-01 2.30183527e-01
-2.38696858e-01 -9.88001823e-02 -7.53363311e-01 7.33249903e-01
-5.31606227e-02 9.13173974e-01 4.73431945e-01 -8.18785965e-01
2.89905220e-02 6.04889631e-01 2.51289576e-01 -7.81557262e-01
-7.07626343e-01 2.21439958e-01 -1.83649823e-01 -2.76247859e-01
-7.86255717e-01 -6.79817677e-01 -1.58112168e+00 -1.82442248e-01
-4.55973059e-01 6.66716814e-01 7.01060891e-01 1.27699006e+00
2.92687863e-02 4.03141379e-01 4.76528287e-01 -5.53282142e-01
-3.10056418e-01 -7.69515514e-01 -7.02185273e-01 4.56348777e-01
4.61127639e-01 -4.04035360e-01 -4.84807849e-01 3.43060136e-01] | [14.384725570678711, 6.549631595611572] |
698d8ef5-95cf-4559-a5b6-7ae7f2a854a6 | icdar2019-robust-reading-challenge-on | 1909.07145 | null | https://arxiv.org/abs/1909.07145v1 | https://arxiv.org/pdf/1909.07145v1.pdf | ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (RRC-ArT) | This paper reports the ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (RRC-ArT) that consists of three major challenges: i) scene text detection, ii) scene text recognition, and iii) scene text spotting. A total of 78 submissions from 46 unique teams/individuals were received for this competition. The top performing score of each challenge is as follows: i) T1 - 82.65%, ii) T2.1 - 74.3%, iii) T2.2 - 85.32%, iv) T3.1 - 53.86%, and v) T3.2 - 54.91%. Apart from the results, this paper also details the ArT dataset, tasks description, evaluation metrics and participants methods. The dataset, the evaluation kit as well as the results are publicly available at https://rrc.cvc.uab.es/?ch=14 | ['ChuanMing Fang', 'Chee-Kheng Chng', 'Junyu Han', 'Errui Ding', 'Yuliang Liu', 'Lianwen Jin', 'Jingtuo Liu', 'Chun Chet Ng', 'Chee Seng Chan', 'Zihan Ni', 'Shuaitao Zhang', 'Yipeng Sun', 'Dimosthenis Karatzas', 'Canjie Luo'] | 2019-09-16 | null | null | null | null | ['text-spotting'] | ['computer-vision'] | [ 2.74333835e-01 -2.69347042e-01 3.55290353e-01 -1.25556767e-01
-1.10356700e+00 -6.36573255e-01 1.10218823e+00 3.03699195e-01
-5.92800379e-01 2.69066304e-01 4.17077869e-01 -1.08456850e-01
3.89801025e-01 -4.20983374e-01 -8.24527860e-01 -3.70434493e-01
2.92744100e-01 5.84032476e-01 2.67467767e-01 -7.46128634e-02
7.75150895e-01 -8.44759420e-02 -1.49368584e+00 7.76172519e-01
8.61229002e-01 1.05170596e+00 1.08175114e-01 1.03438163e+00
-2.85539389e-01 7.82108963e-01 -9.59817410e-01 -5.02945364e-01
3.31659019e-02 -2.83802003e-01 -1.01991165e+00 -3.57042952e-03
8.16949785e-01 -7.31472075e-02 -3.13555956e-01 7.23128796e-01
8.90706480e-01 1.84442490e-01 8.81268859e-01 -9.10340071e-01
-7.29512751e-01 7.44859815e-01 -8.84376585e-01 1.37961894e-01
6.78356051e-01 3.76661122e-02 6.14532053e-01 -1.29153705e+00
7.18822598e-01 1.18927693e+00 5.48918128e-01 5.46633244e-01
-6.53007805e-01 -3.68778914e-01 2.06140339e-01 1.78061739e-01
-1.64684021e+00 -5.41232586e-01 2.76998222e-01 -7.18318522e-01
1.03149080e+00 4.34637308e-01 1.99499920e-01 1.67235577e+00
-4.61031906e-02 1.17014015e+00 1.44776952e+00 -4.81466204e-01
2.85107315e-01 1.67785883e-01 5.10269225e-01 4.59206760e-01
3.50350559e-01 -7.44721770e-01 -7.22750008e-01 3.69160920e-01
3.44212145e-01 -4.89672691e-01 -1.44978777e-01 3.73775303e-01
-1.48814523e+00 4.74698186e-01 5.95961399e-02 3.83059353e-01
-1.56922594e-01 -1.71490893e-01 8.46797347e-01 1.05826985e-02
5.64467907e-01 3.74054343e-01 -1.61139861e-01 -2.86823452e-01
-8.42074215e-01 4.32110101e-01 6.26166403e-01 1.31976366e+00
5.83524890e-02 1.39955878e-01 -6.86247289e-01 1.23524296e+00
1.21061191e-01 1.00964797e+00 4.59604502e-01 -2.34960198e-01
1.10942209e+00 4.53734934e-01 7.37313777e-02 -8.92660916e-01
-5.65300047e-01 -2.67134011e-01 -1.01578891e+00 -3.66379410e-01
2.27362126e-01 -1.37753785e-01 -1.11445820e+00 1.15021360e+00
-6.72537163e-02 -2.62781262e-01 -3.58191580e-02 7.91029692e-01
1.67845643e+00 8.40412140e-01 -1.00873426e-01 2.47792020e-01
1.53410172e+00 -1.11926389e+00 -8.97654474e-01 -3.06547344e-01
4.78682727e-01 -1.25944459e+00 1.37618411e+00 5.53190231e-01
-1.03475142e+00 -4.66280460e-01 -8.75681877e-01 -3.59124064e-01
-7.33465254e-01 8.02037477e-01 -3.99816558e-02 5.40295482e-01
-1.11176491e+00 -1.33849159e-01 -4.23578739e-01 -8.72067809e-01
4.27259237e-01 -8.23254511e-02 -5.29595688e-02 -1.47712380e-01
-7.64638662e-01 7.79938042e-01 1.14533916e-01 -8.43088776e-02
-9.41317022e-01 -4.49910492e-01 -5.40842235e-01 -4.58433896e-01
2.74250299e-01 -3.32733929e-01 1.32926357e+00 -4.83174056e-01
-1.35299575e+00 1.35173213e+00 -2.22426981e-01 -4.50136542e-01
1.16731322e+00 -7.42529273e-01 -5.53900301e-01 1.18387662e-01
4.29919213e-01 6.26987934e-01 5.73221803e-01 -1.07578850e+00
-4.62704688e-01 -4.19633120e-01 -4.56955314e-01 4.96789545e-01
-3.94346714e-02 5.10610700e-01 -8.00580442e-01 -1.00096846e+00
-1.65448725e-01 -7.40518868e-01 2.55728573e-01 -5.09414375e-01
-1.10255897e+00 -4.13682133e-01 7.81936944e-01 -1.03618729e+00
1.09273052e+00 -1.95115805e+00 -2.15555280e-01 -2.55215168e-01
1.96195826e-01 2.71694154e-01 -8.30880776e-02 5.76945007e-01
8.46456513e-02 2.68748969e-01 -8.53293911e-02 -9.18204129e-01
1.95494369e-01 -5.85843384e-01 -5.38354158e-01 4.10828114e-01
-6.67538345e-02 9.38105643e-01 -3.97392392e-01 -4.55899119e-01
4.98918235e-01 4.63776112e-01 1.97157398e-01 -4.01740037e-02
-4.11225446e-02 3.65402579e-01 -3.77019942e-01 7.66808689e-01
8.67343366e-01 -2.60030597e-01 -4.18216735e-01 -1.62144154e-02
-4.29213226e-01 1.70867875e-01 -1.28122759e+00 1.37741423e+00
-4.04606201e-02 1.26284027e+00 -4.04848084e-02 -5.90056300e-01
1.06499088e+00 1.16181672e-01 2.79160500e-01 -9.38497365e-01
3.79395008e-01 9.61599126e-02 -7.66078949e-01 -4.89221632e-01
1.15047920e+00 4.37377274e-01 -3.07331592e-01 1.80250660e-01
-1.03943333e-01 -1.19537063e-01 4.85233873e-01 4.01374787e-01
1.10054708e+00 1.97832230e-02 7.88570046e-02 -4.42353249e-01
4.82282758e-01 -7.33858272e-02 -9.95105505e-03 9.94348705e-01
-2.50892550e-01 1.16092050e+00 3.81469756e-01 -4.09593284e-01
-8.01718831e-01 -9.07431602e-01 -2.50031471e-01 1.02595639e+00
3.10515881e-01 -5.83176732e-01 -1.04569638e+00 -4.39576834e-01
-3.77094805e-01 8.63935709e-01 -8.20583642e-01 3.49804580e-01
-3.97917986e-01 -4.45760220e-01 9.77579594e-01 6.52702391e-01
1.10085118e+00 -1.06992817e+00 -4.68706012e-01 -3.70685905e-01
-7.72008896e-01 -1.46589768e+00 -6.20209217e-01 -1.97415546e-01
-4.39205706e-01 -9.10773396e-01 -1.11779869e+00 -7.03600585e-01
3.43503714e-01 5.29582381e-01 1.06855059e+00 -2.66495585e-01
-5.43481112e-01 6.55012310e-01 -7.21910298e-01 -8.13917100e-01
1.62465200e-02 2.07182035e-01 -3.40739940e-03 1.41929969e-01
2.22110540e-01 4.02210802e-01 -2.80697793e-01 4.02975857e-01
-5.57020366e-01 3.25995147e-01 4.81266111e-01 3.01605314e-01
7.36457884e-01 -2.72004217e-01 8.22912753e-02 -7.03186333e-01
4.72970575e-01 -1.55078486e-01 -5.74033201e-01 5.82487047e-01
-2.56703407e-01 -5.34856319e-01 5.49559832e-01 -3.00905198e-01
-1.05461085e+00 -3.43353748e-02 -5.31158783e-02 -9.48750749e-02
-5.84775507e-01 3.46741229e-02 -5.73786795e-02 4.89545971e-01
1.08512998e+00 3.95266175e-01 -8.03415775e-01 -5.31507730e-01
1.02623828e-01 1.04284132e+00 7.77952373e-01 -4.60674435e-01
5.93137860e-01 4.44669157e-01 -5.54089963e-01 -1.55021632e+00
-9.49678600e-01 -8.24930727e-01 -5.98410606e-01 -3.81146163e-01
1.45020413e+00 -1.19803190e+00 -2.69372642e-01 1.20109773e+00
-1.18810153e+00 -5.94350815e-01 -6.54037297e-02 4.24134552e-01
-2.93110877e-01 3.55534822e-01 -4.57264900e-01 -8.61196339e-01
-8.03292274e-01 -1.00132596e+00 1.68147314e+00 3.37807238e-01
-1.29254848e-01 -7.73898184e-01 8.57788995e-02 9.37061965e-01
4.01760548e-01 3.50360900e-01 -8.80898610e-02 -6.82336569e-01
-2.45521620e-01 -2.93973655e-01 -4.33905959e-01 -1.24440052e-01
-3.96840185e-01 1.77513510e-01 -1.24091041e+00 -2.95452088e-01
-6.26487255e-01 -6.73834801e-01 9.25205231e-01 4.11232412e-01
1.16771185e+00 -1.29606068e-01 -2.19975084e-01 3.37771028e-01
1.11332977e+00 -2.42493525e-02 9.68361259e-01 5.63912868e-01
8.16076398e-01 5.08759797e-01 5.51181614e-01 6.11604273e-01
5.76805830e-01 8.41452599e-01 1.53908327e-01 3.98445018e-02
-3.54106218e-01 -2.65210658e-01 5.84308386e-01 8.28903973e-01
-9.18578077e-03 -7.05074191e-01 -1.58191168e+00 5.68426907e-01
-1.63399887e+00 -8.31711054e-01 -1.11870372e+00 2.00040102e+00
4.31974679e-01 -4.26066965e-02 3.47914875e-01 1.76582813e-01
1.01628697e+00 2.29357675e-01 -3.21811587e-01 -4.44339097e-01
-6.38922870e-01 1.29478285e-03 3.58222485e-01 1.93224087e-01
-1.53142917e+00 1.11234534e+00 5.34530354e+00 7.74052083e-01
-8.74018133e-01 -2.72043087e-02 5.71285307e-01 -4.06624526e-02
1.94401354e-01 -4.54965472e-01 -1.01764500e+00 5.63198209e-01
8.98759604e-01 1.89042017e-02 3.35257471e-01 5.35810649e-01
-9.97297317e-02 -2.27936268e-01 -6.62149072e-01 1.34334588e+00
6.56167030e-01 -1.12633789e+00 3.06272686e-01 -1.93578795e-01
9.01285946e-01 5.32476485e-01 1.89803883e-01 5.77593207e-01
1.37601405e-01 -1.24889994e+00 1.27402949e+00 4.57150698e-01
1.12064624e+00 -3.46087754e-01 7.58423269e-01 1.91489860e-01
-1.30807734e+00 2.49537736e-01 -3.10910046e-01 3.94851565e-01
-9.92431641e-02 5.55070877e-01 -5.74989855e-01 7.54412770e-01
1.19813585e+00 7.97013521e-01 -1.19878316e+00 1.07703078e+00
-3.36012423e-01 7.73913860e-01 -2.21578866e-01 -4.67125356e-01
2.75338050e-02 3.65774706e-02 5.61334372e-01 1.98375905e+00
1.29317671e-01 -2.09327191e-01 6.55280501e-02 4.54144746e-01
-3.59122515e-01 5.20827413e-01 -4.95338470e-01 -5.11082709e-02
4.70319688e-01 1.16954076e+00 -1.01037395e+00 -4.11134034e-01
-6.90441057e-02 1.10742354e+00 6.09451793e-02 4.86570925e-01
-9.37371731e-01 -6.98721170e-01 -9.48910639e-02 -2.18535557e-01
1.90111235e-01 1.20627724e-01 -7.17609406e-01 -1.36876643e+00
3.16685319e-01 -9.02237952e-01 5.02454877e-01 -1.05533803e+00
-1.09168386e+00 7.00548351e-01 -3.31087619e-01 -1.07714927e+00
3.98933291e-01 -6.96331322e-01 -6.30360961e-01 7.52674878e-01
-1.02875626e+00 -1.22844195e+00 -8.88400376e-01 6.01977646e-01
1.06859326e+00 -4.53826308e-01 5.83520532e-01 1.98854104e-01
-9.50811625e-01 9.77001965e-01 2.78486282e-01 3.74339163e-01
9.60844398e-01 -1.45627034e+00 7.00461388e-01 1.05830812e+00
4.79972437e-02 1.06784470e-01 5.69401503e-01 -6.37477398e-01
-1.37582207e+00 -1.40868735e+00 1.03677428e+00 -1.03341997e+00
5.45587540e-01 -7.57604897e-01 -7.40565538e-01 6.03173912e-01
4.03029054e-01 -3.14173996e-01 3.71155649e-01 -3.75788845e-02
-5.67750156e-01 1.09138548e-01 -9.12968457e-01 7.99421668e-01
7.89481521e-01 -4.77439612e-01 -3.25565606e-01 6.58480048e-01
3.76126051e-01 -8.95459712e-01 -7.58956194e-01 -3.81334908e-02
2.88846940e-01 -5.33166766e-01 6.18755162e-01 -1.00895166e-01
7.86386490e-01 -2.92416066e-01 -5.12756050e-01 -5.60756266e-01
-1.38043046e-01 -5.42968154e-01 5.91639988e-02 1.44380951e+00
5.26635408e-01 -3.64675909e-01 3.14798415e-01 2.14044467e-01
-3.94361287e-01 -3.34763587e-01 -8.61642241e-01 -7.30048239e-01
3.17722350e-01 -7.12640405e-01 -2.41074041e-02 1.08132446e+00
-3.77510458e-01 6.51910245e-01 -4.26085889e-01 -2.43960898e-02
6.37957513e-01 -1.10156931e-01 1.19354224e+00 -9.94116187e-01
3.62292044e-02 -7.84126341e-01 -7.96931311e-02 -1.01135314e+00
-4.10942227e-01 -8.65871072e-01 1.45177677e-01 -1.96729147e+00
4.08968538e-01 1.11856177e-01 2.10243270e-01 5.97883940e-01
-1.79511264e-01 2.91355222e-01 2.64456123e-01 3.62646401e-01
-1.37717223e+00 3.70606214e-01 7.38434851e-01 -3.26339036e-01
-2.94193067e-02 -1.85534164e-01 -5.90900600e-01 3.87335449e-01
1.41244018e+00 -3.59131582e-02 -8.13946128e-02 -7.91970611e-01
1.26378760e-01 -4.49881852e-01 4.38298821e-01 -9.91659880e-01
3.72232825e-01 2.09607556e-01 5.78317702e-01 -1.18416250e+00
3.19343925e-01 -1.09972559e-01 -3.40912730e-01 2.39488214e-01
-4.99412954e-01 8.93684104e-02 3.33295137e-01 2.86665916e-01
1.01433918e-01 -1.80831328e-01 7.85281599e-01 2.49560893e-01
-7.13832617e-01 -1.46986023e-01 -6.56719804e-01 6.91078126e-01
9.18434322e-01 -1.71503589e-01 -9.20479655e-01 -3.41197282e-01
-2.23911166e-01 4.36750203e-01 3.85312200e-01 8.81960690e-01
6.80633843e-01 -9.38111424e-01 -1.32025301e+00 -3.47613752e-01
5.81868768e-01 3.19974427e-03 5.92890203e-01 1.05391550e+00
-5.89017570e-01 7.03246653e-01 1.65940076e-01 -8.94303858e-01
-1.52691293e+00 2.13936180e-01 1.42449602e-01 -3.60308200e-01
-8.28015387e-01 7.88561046e-01 3.40541340e-02 -5.34969330e-01
5.98047972e-01 -2.98233945e-02 -4.59866494e-01 6.92001581e-02
9.04564321e-01 8.49962115e-01 3.42718035e-01 -8.53574932e-01
-4.80716109e-01 9.57556844e-01 -1.96502268e-01 -2.38635331e-01
1.05982745e+00 -1.63103268e-01 1.46555334e-01 7.27496326e-01
9.88818288e-01 9.65727493e-03 -6.21005058e-01 -2.02714771e-01
9.48512182e-02 -2.13983491e-01 -1.13236219e-01 -1.46830118e+00
-5.49734116e-01 9.71833825e-01 5.35391748e-01 6.86786249e-02
1.08927417e+00 -2.79104505e-02 4.56759244e-01 4.02722001e-01
6.08984940e-02 -1.49191999e+00 1.65713966e-01 1.07584286e+00
1.28736711e+00 -1.20991409e+00 5.38131036e-02 -1.19716898e-01
-1.04989278e+00 9.57816720e-01 5.30482948e-01 2.38858625e-01
1.11755364e-01 2.18519166e-01 9.62033197e-02 -2.72899777e-01
-6.93631351e-01 -1.22857332e-01 6.26805782e-01 5.03564596e-01
5.90850532e-01 1.60601780e-01 -8.78796130e-02 5.09534240e-01
-5.04929364e-01 -4.11228448e-01 7.00676322e-01 8.78623366e-01
-2.21186638e-01 -3.31242442e-01 -5.12608469e-01 4.18896258e-01
-5.52234054e-01 -8.14680383e-02 -1.36649454e+00 8.28477442e-01
-3.60281020e-01 1.38355124e+00 6.91974387e-02 -6.26140893e-01
6.79548085e-01 -1.39726192e-01 1.84251919e-01 -3.85503531e-01
-6.76844656e-01 1.33140355e-01 2.04108328e-01 -3.67937446e-01
8.05555284e-03 -9.04208362e-01 -1.16859400e+00 -5.98463774e-01
-1.28014117e-01 -5.85854128e-02 1.01244855e+00 4.07314628e-01
3.98955435e-01 5.77822387e-01 4.52447891e-01 -5.39177239e-01
-2.31791466e-01 -1.30404902e+00 -3.19056541e-01 1.99314684e-01
1.59478337e-01 -3.13365340e-01 -2.76156247e-01 2.87533969e-01] | [11.94224739074707, 2.3065192699432373] |
8037b51c-7c09-4246-ae94-ff7763ba96f9 | semantic-image-cropping | 2107.07153 | null | https://arxiv.org/abs/2107.07153v1 | https://arxiv.org/pdf/2107.07153v1.pdf | Semantic Image Cropping | Automatic image cropping techniques are commonly used to enhance the aesthetic quality of an image; they do it by detecting the most beautiful or the most salient parts of the image and removing the unwanted content to have a smaller image that is more visually pleasing. In this thesis, I introduce an additional dimension to the problem of cropping, semantics. I argue that image cropping can also enhance the image's relevancy for a given entity by using the semantic information contained in the image. I call this problem, Semantic Image Cropping. To support my argument, I provide a new dataset containing 100 images with at least two different entities per image and four ground truth croppings collected using Amazon Mechanical Turk. I use this dataset to show that state-of-the-art cropping algorithms that only take into account aesthetics do not perform well in the problem of semantic image cropping. Additionally, I provide a new deep learning system that takes not just aesthetics but also semantics into account to generate image croppings, and I evaluate its performance using my new semantic cropping dataset, showing that using the semantic information of an image can help to produce better croppings. | ['Oriol Corcoll'] | 2021-07-15 | null | null | null | null | ['image-cropping'] | ['computer-vision'] | [ 4.74859685e-01 3.90710592e-01 1.53256685e-01 -2.05651909e-01
-2.15714455e-01 -6.61292076e-01 3.38398278e-01 9.79356691e-02
-1.98246449e-01 3.81197274e-01 1.61640495e-01 -1.26745373e-01
2.10212708e-01 -1.16247201e+00 -9.66773808e-01 -3.88844103e-01
5.96478224e-01 -6.56535253e-02 1.63245760e-02 -5.46130240e-01
3.79930794e-01 2.55431980e-01 -1.80877686e+00 4.99308050e-01
6.87800109e-01 7.95980871e-01 3.56105745e-01 3.15957993e-01
-3.26603740e-01 6.71986282e-01 -8.01254511e-01 -6.63098395e-01
4.80735123e-01 -5.43976307e-01 -9.31314051e-01 4.91917282e-01
4.47523057e-01 -3.99694517e-02 1.79340765e-01 1.20069146e+00
2.36921340e-01 -7.78056085e-02 5.46195865e-01 -1.72604716e+00
-1.15455770e+00 6.78659379e-01 -1.00365174e+00 -3.90205860e-01
3.62100512e-01 4.28140722e-02 1.06364012e+00 -1.00556207e+00
1.14720595e+00 1.38639712e+00 5.77671051e-01 4.71985012e-01
-1.25268149e+00 -7.02276826e-01 -1.49262905e-01 -3.72948907e-02
-1.13384759e+00 2.78424192e-02 1.00663686e+00 -3.17828983e-01
3.92224401e-01 3.04556191e-01 1.00146675e+00 9.11269844e-01
-1.37299104e-02 8.59173954e-01 1.46002257e+00 -8.01861584e-01
2.72638649e-01 4.30895329e-01 -2.54210919e-01 4.81509238e-01
3.33247781e-01 -1.16278760e-01 -4.96099323e-01 3.29998046e-01
6.39649034e-01 -5.65296523e-02 4.21352126e-02 -5.26409328e-01
-1.22707093e+00 1.07763863e+00 9.58730102e-01 5.22565365e-01
-4.02132899e-01 2.33650446e-01 1.14057280e-01 1.69419765e-01
4.04437989e-01 9.64885354e-01 -1.04857370e-01 2.84188151e-01
-1.17428243e+00 4.73964423e-01 3.76851618e-01 7.59068966e-01
1.10843980e+00 -5.39336614e-02 4.62845759e-03 6.74903333e-01
2.14947686e-01 6.84668183e-01 2.60338545e-01 -1.11028445e+00
2.74590343e-01 8.87409210e-01 1.00778684e-01 -1.37494385e+00
-2.94753134e-01 -1.13438874e-01 -5.58391273e-01 8.62259209e-01
3.93243968e-01 -1.57246694e-01 -1.12204981e+00 1.67877734e+00
1.52608946e-01 -4.54982847e-01 -9.15687252e-03 9.60425019e-01
9.72900152e-01 6.42475963e-01 4.36629504e-01 3.48393589e-01
1.61618030e+00 -1.00266731e+00 -7.54790604e-01 -5.80487013e-01
4.35274392e-01 -9.44444776e-01 1.47789133e+00 2.75987774e-01
-1.17678905e+00 -4.96275097e-01 -1.29634714e+00 -2.72934973e-01
-1.02786660e+00 1.72760099e-01 7.00240850e-01 7.46495545e-01
-1.19179261e+00 5.34046233e-01 2.15464935e-01 -4.39753801e-01
4.10611928e-01 -1.87743664e-01 -5.70459664e-01 1.42957121e-01
-1.31551635e+00 9.64037359e-01 6.23131454e-01 -3.87775034e-01
-3.39792758e-01 -6.53484702e-01 -9.85977232e-01 9.34961960e-02
1.06842861e-01 -6.14742160e-01 8.03396761e-01 -1.97904670e+00
-7.01745510e-01 1.35762942e+00 3.80178839e-02 -3.65178078e-01
4.78423625e-01 -9.02558342e-02 -6.51605651e-02 4.16708022e-01
5.15610158e-01 1.55775607e+00 1.18103254e+00 -1.86882460e+00
-4.37113225e-01 -2.61559755e-01 1.98018402e-01 3.81430686e-01
-3.88215363e-01 -8.81272182e-02 -8.83842707e-02 -1.07607186e+00
-3.33827287e-02 -9.40277457e-01 -5.84294498e-02 3.79001684e-02
-4.98919398e-01 2.22043827e-01 8.72463346e-01 -8.40548456e-01
1.03196025e+00 -2.22719169e+00 -1.51245892e-02 3.33139569e-01
1.24339551e-01 3.28068659e-02 -3.97204489e-01 4.14532304e-01
-4.64217752e-01 5.93518138e-01 -2.91517854e-01 -1.37394950e-01
-1.03398412e-02 -7.85824955e-02 -1.54948696e-01 -1.59471348e-01
4.26424146e-01 1.43840289e+00 -8.68711472e-01 -5.73254466e-01
4.58433807e-01 4.82938588e-01 -3.04939330e-01 -3.30836952e-01
-1.50436491e-01 1.17986225e-01 -2.32600514e-02 5.88701069e-01
9.31563914e-01 -1.22025125e-01 -1.02342948e-01 -2.97172338e-01
2.39803240e-01 -4.92620528e-01 -1.13170230e+00 1.66788495e+00
-3.91471475e-01 1.05472529e+00 -2.81593204e-01 -6.76209986e-01
9.55447972e-01 -2.95529626e-02 5.18178999e-01 -1.09211874e+00
1.04417637e-01 1.46864220e-01 -4.59661394e-01 -5.71577251e-01
9.87475276e-01 -4.68268424e-01 -2.25823075e-01 7.21071362e-01
-3.37230563e-01 -6.97001278e-01 1.50384381e-01 4.59989727e-01
4.81718838e-01 2.70848662e-01 9.54399109e-02 -2.23492891e-01
1.92590415e-01 4.34955269e-01 -3.88640910e-02 5.78992128e-01
-1.68516263e-02 1.14211941e+00 3.53324383e-01 -5.32317638e-01
-1.63782322e+00 -1.00396514e+00 2.06234410e-01 1.18005228e+00
4.42176968e-01 -4.44685370e-02 -1.54517746e+00 -4.49385554e-01
-6.53348267e-02 9.69997108e-01 -1.16171408e+00 -5.77828586e-02
-2.25329489e-01 -6.04047954e-01 2.45673984e-01 2.55474567e-01
6.73110068e-01 -1.81410539e+00 -1.02163351e+00 -6.70589283e-02
-7.53295481e-01 -1.05217171e+00 -2.47868523e-01 -1.10159658e-01
-4.26279217e-01 -1.15259349e+00 -9.96269286e-01 -8.54518950e-01
9.35903370e-01 7.72336721e-01 1.44743228e+00 3.29765111e-01
-6.23393059e-01 3.17691654e-01 -8.57170522e-01 -7.75362253e-01
-6.42149627e-01 3.05085778e-02 -6.09589398e-01 -1.18049279e-01
3.85082573e-01 -1.54785961e-01 -4.51561481e-01 8.05684552e-02
-1.61575854e+00 5.07715821e-01 6.92320108e-01 6.53848231e-01
2.71706998e-01 3.19122195e-01 4.53699768e-01 -8.35354924e-01
8.29932511e-01 -2.55517274e-01 8.96937326e-02 4.10922408e-01
-5.46763182e-01 1.42743200e-01 4.25588116e-02 -2.34164000e-01
-1.09411395e+00 2.62775570e-01 3.03264052e-01 -2.06844598e-01
-3.48741822e-02 1.30921900e-01 5.53101748e-02 3.02502755e-02
8.28497291e-01 6.77297413e-02 2.18678206e-01 -3.26044485e-02
9.22571778e-01 4.04853702e-01 4.29602712e-01 -1.78063169e-01
9.19553578e-01 6.94670379e-01 1.11261427e-01 -5.78117490e-01
-7.99875498e-01 -1.54702723e-01 -7.18544126e-01 -5.48178256e-01
1.09112823e+00 -7.60895908e-01 -3.30958188e-01 4.14147943e-01
-1.10979569e+00 -1.77116111e-01 -7.05866754e-01 -2.13884890e-01
-6.51720822e-01 3.84678602e-01 -5.58768958e-02 -5.99874556e-01
-4.58268851e-01 -1.00924385e+00 1.30910158e+00 3.95026237e-01
-3.31342369e-01 -6.59692466e-01 -4.07194495e-01 5.63633978e-01
4.18436974e-01 7.08750486e-01 8.52255523e-01 8.38563070e-02
-2.37882480e-01 -9.26666483e-02 -8.03180218e-01 2.50851303e-01
2.43333634e-02 1.20571755e-01 -1.09467840e+00 1.79424524e-01
-3.69900644e-01 -2.57830709e-01 8.99573863e-01 4.04660374e-01
8.47901821e-01 -3.17650437e-01 -1.01183750e-01 1.66772768e-01
1.75910008e+00 -6.40517026e-02 1.21637332e+00 7.71228671e-01
5.93002975e-01 1.13634264e+00 9.49537456e-01 2.35067770e-01
2.20738515e-01 4.17476863e-01 7.58124352e-01 -6.64983213e-01
-4.76851761e-01 -6.33988261e-01 1.12866387e-01 -1.60772011e-01
1.92035854e-01 -1.39197126e-01 -5.31054139e-01 8.06524396e-01
-1.65786552e+00 -9.17658627e-01 -2.55573541e-01 1.96775627e+00
7.16387212e-01 -1.43131360e-01 2.13919982e-01 3.51943433e-01
8.87085617e-01 -8.04537460e-02 -2.13161111e-01 -7.45134771e-01
-5.30225575e-01 3.38463545e-01 6.09796703e-01 3.11338663e-01
-1.09785891e+00 1.18963981e+00 6.41521931e+00 8.68393600e-01
-9.05106723e-01 -2.89025195e-02 1.15476525e+00 2.19448388e-01
-7.13027537e-01 -4.10899110e-02 -1.58813670e-01 4.17875350e-01
3.61955911e-01 5.03889099e-02 5.27717650e-01 9.76151705e-01
2.65455335e-01 -6.93341076e-01 -4.78655159e-01 1.12467527e+00
2.28021577e-01 -1.29344881e+00 2.64569461e-01 3.03545338e-03
9.00847495e-01 -7.34877050e-01 3.04597646e-01 -2.87640184e-01
2.13527426e-01 -1.35554194e+00 1.16195309e+00 4.33383405e-01
6.82195783e-01 -6.73146963e-01 7.45187759e-01 -2.14902282e-01
-9.94129598e-01 -9.85257402e-02 -2.81744868e-01 -1.20593840e-02
6.78023472e-02 5.14851093e-01 -7.66126275e-01 2.19037920e-01
9.20522451e-01 3.98174763e-01 -1.15892565e+00 7.48339117e-01
-4.03210729e-01 -1.12465270e-01 1.76854022e-02 1.58227570e-02
3.44399959e-01 -1.46287858e-01 1.89056218e-01 1.06339097e+00
4.39966112e-01 -2.93815702e-01 -4.55477536e-01 1.23744512e+00
-5.72316907e-02 4.32459295e-01 -8.60403180e-01 -1.49501979e-01
3.09687972e-01 1.47261202e+00 -1.26557100e+00 -3.90591711e-01
-9.46935862e-02 1.33407354e+00 -9.63829309e-02 1.72016561e-01
-8.84227693e-01 -4.80713636e-01 1.84622869e-01 3.87250841e-01
7.91034568e-03 2.28596598e-01 -8.00313234e-01 -7.73736596e-01
9.11327004e-02 -7.43660748e-01 6.13095760e-02 -1.84970534e+00
-1.01144648e+00 4.09748822e-01 -9.86735746e-02 -1.27801466e+00
5.35172522e-02 -4.95123208e-01 -3.50251049e-01 7.81178355e-01
-1.37393034e+00 -1.61420238e+00 -8.70970011e-01 2.45767176e-01
5.23259044e-01 4.15404469e-01 6.71930254e-01 -8.15247744e-02
2.49192417e-01 9.55639407e-02 -3.28227371e-01 2.56165385e-01
8.43129098e-01 -1.44804692e+00 6.35186553e-01 7.24694371e-01
2.90169805e-01 2.56092608e-01 9.37324584e-01 -6.26806259e-01
-8.33190262e-01 -8.40903997e-01 9.77505088e-01 -3.75009447e-01
2.88853854e-01 -1.60175651e-01 -4.11642462e-01 2.50531286e-01
6.31549835e-01 -5.71205497e-01 3.60487074e-01 -5.13023853e-01
-5.04978597e-01 3.10990751e-01 -1.51887786e+00 7.46117175e-01
8.34142268e-01 -8.23885351e-02 -6.30725384e-01 2.61621505e-01
8.59782517e-01 -8.53998736e-02 -4.43214118e-01 8.08922946e-02
7.04754233e-01 -1.26579952e+00 1.30426407e+00 -3.12334299e-01
1.00030446e+00 -3.20431739e-01 8.94263461e-02 -1.37060547e+00
-2.00734779e-01 -2.43197367e-01 7.07112670e-01 1.16069794e+00
5.61565399e-01 -1.90898731e-01 8.41708839e-01 8.22796822e-01
4.07414466e-01 -1.68683916e-01 -4.24814403e-01 -5.26841938e-01
1.73771456e-01 -3.83453935e-01 7.22224832e-01 1.04554915e+00
5.04816659e-02 1.84171081e-01 -4.67052728e-01 -5.43554425e-01
5.28843045e-01 2.02266008e-01 9.08752501e-01 -1.19334447e+00
2.69168764e-01 -5.02027512e-01 -3.25258672e-01 5.23318313e-02
-2.11326554e-01 -6.56253517e-01 -1.38874426e-01 -1.81511962e+00
4.49275166e-01 -3.46374065e-01 6.29712939e-02 7.07820833e-01
-3.13783199e-01 1.07114697e+00 7.74683774e-01 8.75194818e-02
-3.60577434e-01 5.92973642e-02 1.58145022e+00 -2.76399732e-01
-6.60732985e-02 -7.57691324e-01 -1.25826097e+00 5.44396877e-01
9.43898201e-01 -2.76810199e-01 -2.62796611e-01 -1.75001070e-01
7.14821637e-01 -5.01005650e-01 6.04955018e-01 -9.93516326e-01
-3.82920504e-01 -1.88140661e-01 8.18024278e-01 -4.35214937e-01
4.83873934e-01 -9.66272652e-01 3.13876480e-01 5.35167038e-01
-4.69848782e-01 1.19769968e-01 2.57080108e-01 2.46189445e-01
-1.43241942e-01 -5.86247206e-01 9.00385737e-01 -5.47363639e-01
-1.14171278e+00 -4.24768597e-01 -2.09244907e-01 2.61904001e-02
1.29601777e+00 -5.28043330e-01 -1.86849490e-01 -7.91642666e-01
-7.90277004e-01 -2.27072567e-01 9.64269221e-01 8.70887816e-01
5.60883462e-01 -1.47775865e+00 -5.57697356e-01 5.65248616e-02
3.48300993e-01 -4.87893313e-01 3.45611334e-01 2.57682770e-01
-8.36600661e-01 1.08888775e-01 -6.89281166e-01 -2.35008076e-01
-1.50129724e+00 9.10781860e-01 7.00238794e-02 2.43628770e-02
-5.43063700e-01 6.70602620e-01 3.75127852e-01 -1.96944684e-01
-2.92781770e-01 -1.05977125e-01 -2.38100857e-01 5.13648152e-01
5.28526723e-01 7.46444911e-02 -2.07159936e-01 -9.29199576e-01
-7.20452443e-02 7.91496158e-01 4.14762825e-01 -4.40619260e-01
1.16622758e+00 -4.45555061e-01 -1.56808883e-01 -1.50108635e-02
9.49133396e-01 -1.93712022e-02 -9.22227263e-01 -2.97042355e-02
-3.29829566e-02 -7.93335736e-01 -1.11208327e-01 -1.04756904e+00
-1.32651639e+00 8.71963084e-01 6.38705194e-01 6.45083964e-01
1.40230024e+00 -9.82352644e-02 6.69732332e-01 -1.34083629e-01
3.58717412e-01 -1.53347135e+00 5.81491828e-01 -1.40282631e-01
1.11034369e+00 -1.53064644e+00 1.44235060e-01 -5.93107522e-01
-1.34871352e+00 1.19136500e+00 4.33609247e-01 -1.99977070e-01
2.37313420e-01 -3.39401178e-02 3.21298301e-01 -3.97076875e-01
1.33518264e-01 -4.80517566e-01 2.88086146e-01 9.39915717e-01
3.12818825e-01 3.33570629e-01 -3.14411908e-01 3.85047883e-01
-5.02171755e-01 5.84850386e-02 8.41759503e-01 6.82725966e-01
-6.38330340e-01 -1.04768038e+00 -7.16897368e-01 2.04871759e-01
-3.66110265e-01 -2.64753819e-01 -1.05860662e+00 9.63206410e-01
3.71954888e-01 1.13714159e+00 -8.05599988e-02 -4.25595999e-01
-1.42804589e-02 1.33197561e-01 3.54104966e-01 -3.61353964e-01
-6.42870963e-01 -1.02421053e-01 -1.78239167e-01 -5.18191695e-01
-6.50991857e-01 -4.06212687e-01 -1.10915387e+00 -4.58846033e-01
-1.58903077e-01 -2.10519016e-01 1.14978504e+00 5.75331926e-01
3.99202973e-01 5.00485659e-01 4.94393319e-01 -8.95281971e-01
2.85801888e-01 -8.24560881e-01 -6.37593448e-01 1.12803793e+00
-1.67638436e-01 -5.46299517e-01 -2.71061897e-01 4.55199301e-01] | [11.474088668823242, -0.9405665397644043] |
ba08b636-c8f9-49b7-a641-2f00aca844ae | learning-to-count-grave-sites-for-cemetery | null | null | https://ieeexplore.ieee.org/abstract/document/9203976 | https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9203976 | Learning to Count Grave Sites for Cemetery Observation Models With Satellite Imagery | Understanding how people occupy open spaces is important for research in support of population modeling, policy, national security, emergency response, and sustainability. For the past decade, there has been an increase in research toward capturing and reporting population dynamics and patterns of life at the building level and in some open public spaces such as cemeteries and parks. This is done through observation models developed from local sociocultural information acquired at various spatiotemporal scales to inform night, day, and episodic population occupancy estimates (people/1000 sq ft). Sociocultural information for cemeteries and parks is scarcely available and often collected manually. The process is not only marred by inconsistencies but is laborious and time consuming. In this study, we leverage convolutional neural networks (CNNs) and satellite imagery to derive grave site counts as proxy variables to support scalable and accurate sociocultural data required in a population observation model. Through a hybrid workflow (weak localization plus regression model), we characterize a large scale automation process to counting of grave sites. We evaluate and demonstrate the efficacy of proposed workflow using out-of-data set large satellite imagery and establish its broader impact on cemetery observation models. | ['Marie Urban', 'Nikhil Makkar', 'Lauryn Bragg', 'Sarah Walters', 'Rohan Dhamdhere', 'Dalton Lunga'] | 2020-09-22 | null | null | null | ieee-geoscience-and-remote-sensing-letters-4 | ['object-counting'] | ['computer-vision'] | [ 3.18640471e-02 -4.07620996e-01 5.13575040e-02 -2.64862031e-01
-2.63992429e-01 -5.60471177e-01 8.30260932e-01 6.28264427e-01
-8.26476038e-01 1.06209302e+00 7.37338185e-01 -4.19049799e-01
-1.12075031e-01 -1.54339623e+00 -4.71851647e-01 -6.20839059e-01
-3.45522255e-01 4.24861759e-01 -2.01162100e-01 -5.60874343e-01
1.33125618e-01 6.77324533e-01 -1.78815997e+00 -1.63634852e-01
7.00650752e-01 7.53859222e-01 1.13644466e-01 6.95599794e-01
-8.50418136e-02 2.53857404e-01 -3.41454238e-01 1.95426106e-01
2.75128067e-01 3.23828816e-01 -4.57252085e-01 -1.46569327e-01
6.38349652e-01 -7.91220129e-01 -1.18929729e-01 5.68121612e-01
6.46643043e-01 4.63768356e-02 5.11367202e-01 -1.33826816e+00
-8.07912588e-01 4.06904191e-01 -5.07121205e-01 5.45942366e-01
1.63038746e-01 3.53864580e-01 5.48257649e-01 -4.20803428e-01
-3.91914099e-02 1.04991221e+00 1.01764417e+00 -2.76547700e-01
-1.08451152e+00 -8.17528188e-01 1.64579317e-01 -4.22238320e-01
-1.99398506e+00 -5.37214756e-01 6.54594600e-02 -5.15513718e-01
1.14697802e+00 5.27411699e-01 1.10271263e+00 9.57308590e-01
4.11104411e-02 8.70918259e-02 1.18601274e+00 -1.13450892e-01
2.33346596e-01 -8.28498825e-02 -9.52341780e-02 5.92984736e-01
4.80917424e-01 5.83428480e-02 -4.04862374e-01 -2.45889321e-01
9.21871126e-01 7.32293010e-01 2.18029305e-01 4.83865350e-01
-1.09349692e+00 6.83409214e-01 6.81014180e-01 3.01333874e-01
-5.44515252e-01 2.35246450e-01 4.04206552e-02 -1.36637241e-01
7.64946520e-01 -9.42728072e-02 -1.43731922e-01 -3.55366588e-01
-1.38996792e+00 4.81874615e-01 6.14013791e-01 8.31017196e-01
8.02450538e-01 -7.38430917e-02 -7.16108903e-02 7.42181599e-01
1.00202344e-01 1.28137922e+00 -4.88073491e-02 -4.99846190e-01
6.40867412e-01 8.59879613e-01 5.64668953e-01 -1.28041518e+00
-9.11042094e-01 -4.22409832e-01 -1.14229608e+00 -2.51316726e-01
1.29155010e-01 -1.73709884e-01 -8.34656775e-01 1.53527379e+00
4.67675984e-01 2.77395070e-01 -3.99141520e-01 8.58410597e-01
5.70946813e-01 6.28849328e-01 4.17066485e-01 1.85865015e-01
1.26897931e+00 -3.88138324e-01 -4.91687924e-01 -3.43613237e-01
5.23137748e-01 -2.06625566e-01 9.12640035e-01 -3.10829401e-01
-7.10764945e-01 -3.70664239e-01 -8.04977357e-01 -3.05454955e-02
-9.69680071e-01 1.75944805e-01 7.01827526e-01 7.19317734e-01
-1.15132999e+00 1.93925351e-01 -1.06447041e+00 -9.94834185e-01
7.57307589e-01 3.82817179e-01 -2.94486552e-01 3.03506702e-01
-1.01024818e+00 7.25097537e-01 -2.04124048e-01 5.04768431e-01
-8.05642426e-01 -6.44355357e-01 -8.21800828e-01 1.55366391e-01
-1.19737327e-01 -5.66891968e-01 4.92989123e-01 -2.95486629e-01
-6.30903244e-01 5.41201115e-01 -8.57195929e-02 -4.28168267e-01
3.65234911e-01 -4.14353386e-02 -6.73864841e-01 -5.40523231e-01
7.07122684e-01 7.20824957e-01 1.93616241e-01 -9.93868053e-01
-1.15309525e+00 -7.99939930e-01 8.51625353e-02 1.07927240e-01
-5.43153524e-01 -3.73961516e-02 1.84917942e-01 -3.93013626e-01
-9.87809384e-04 -9.60716605e-01 -2.38217875e-01 -1.67562589e-01
-6.49567991e-02 2.89082032e-04 5.46516120e-01 -8.70150089e-01
1.66272151e+00 -1.89117908e+00 -5.14932275e-01 2.32280880e-01
1.09796710e-01 -2.47932803e-02 3.37065123e-02 9.62420344e-01
6.91526234e-01 4.04992908e-01 -3.78106296e-01 -4.22586977e-01
1.25031158e-01 3.65479946e-01 -3.36084634e-01 4.25263882e-01
1.23132899e-01 7.96607018e-01 -8.84052813e-01 -5.97020268e-01
6.89393878e-01 7.33646452e-01 -4.47945058e-01 -1.56507000e-01
2.83536166e-01 6.06044054e-01 -3.90553385e-01 1.19093704e+00
8.61380935e-01 -1.49596715e-02 -8.53215307e-02 1.51139572e-01
-9.32783544e-01 1.67933792e-01 -1.06879497e+00 1.25027907e+00
-7.69447803e-01 5.52031040e-01 3.33265029e-02 -5.11392415e-01
8.25069785e-01 7.92955048e-03 3.07010770e-01 -7.99951613e-01
3.79559323e-02 -4.74962406e-03 -6.51635766e-01 -6.89119160e-01
9.21254337e-01 -1.13749504e-02 -3.76009464e-01 6.01530969e-01
-4.49429929e-01 3.25757205e-01 -2.79982239e-02 -2.07134172e-01
8.99540484e-01 -1.14723161e-01 4.72716719e-01 -3.74959767e-01
1.74147710e-01 2.24764764e-01 1.50520414e-01 9.52590466e-01
-4.16483492e-01 4.26017761e-01 -1.15940601e-01 -1.04967391e+00
-1.16044879e+00 -1.16838431e+00 -3.02920789e-01 1.37534738e+00
-8.27676281e-02 -9.32591632e-02 -5.02803504e-01 -5.76703576e-03
2.17986330e-01 2.85111636e-01 -8.57482672e-01 4.14527833e-01
-4.08734798e-01 -1.19170403e+00 8.54847074e-01 7.71755695e-01
9.09506738e-01 -8.14255297e-01 -5.94669282e-01 3.04489970e-01
-4.56970274e-01 -9.58114743e-01 -2.23562002e-01 -1.55082181e-01
-7.46602476e-01 -7.38124788e-01 -3.96433264e-01 -5.10323048e-01
6.38066351e-01 5.13723254e-01 9.35153067e-01 2.51977473e-01
-2.66064405e-01 1.62155494e-01 -1.11665025e-01 -3.82330984e-01
2.99645692e-01 5.73768795e-01 2.61568666e-01 -3.20246480e-02
8.15800548e-01 -1.26956928e+00 -9.12487447e-01 3.70057583e-01
-8.07318687e-01 1.08494431e-01 4.85657573e-01 2.58372277e-01
3.86017323e-01 -6.07141256e-02 6.27618432e-01 -2.02335209e-01
4.52025533e-01 -1.10577106e+00 -5.69653988e-01 3.34451459e-02
-1.84693858e-01 -3.65617275e-01 3.98691118e-01 -1.63537234e-01
-6.07509673e-01 -1.30950110e-02 -1.42691419e-01 3.87627572e-01
-6.60866380e-01 5.32898307e-01 2.28281543e-01 2.69223273e-01
5.65807283e-01 3.03542525e-01 -3.45285237e-01 -3.54253829e-01
-3.01302057e-02 1.43276393e+00 4.83447194e-01 -4.90247607e-01
6.87165856e-01 1.19728935e+00 -1.49224997e-01 -1.32363033e+00
-7.59443939e-01 -6.72607303e-01 -8.07621300e-01 -1.63838089e-01
8.08302581e-01 -1.20620883e+00 -1.01601303e+00 2.55780071e-01
-6.11818194e-01 -3.74912798e-01 1.85559750e-01 1.93151742e-01
7.10509866e-02 -6.95799366e-02 -3.87213156e-02 -1.34530914e+00
-4.99868780e-01 -5.66949844e-01 1.52925098e+00 3.07257116e-01
-1.44764140e-01 -1.00419605e+00 2.84446269e-01 4.71588463e-01
9.01430607e-01 6.44729018e-01 2.64258087e-01 1.25003569e-02
-6.97407424e-01 -3.56614292e-01 -4.50959474e-01 -2.14479208e-01
3.76965493e-01 -3.01009357e-01 -9.20048952e-01 -2.95240402e-01
-7.10614800e-01 -5.81074208e-02 7.34947622e-01 5.53974688e-01
8.66289020e-01 -5.95875800e-01 -4.17040646e-01 6.19932652e-01
1.50264978e+00 -3.94741297e-01 6.33131623e-01 4.98228043e-01
6.95787609e-01 5.11576712e-01 2.19623730e-01 9.01239991e-01
1.08518350e+00 3.84497285e-01 5.02585530e-01 -3.22985083e-01
3.19454014e-01 -3.51232916e-01 1.28064409e-01 3.46954018e-01
-7.21158445e-01 1.46961119e-02 -1.17267370e+00 1.17130339e+00
-1.80217600e+00 -1.21810889e+00 -8.70638564e-02 2.25247812e+00
3.78319919e-01 -3.54227036e-01 3.03423464e-01 -1.91869393e-01
5.83842754e-01 3.69056314e-01 -1.30041197e-01 1.15297493e-02
-1.63890600e-01 -3.33867706e-02 1.21837962e+00 5.48588157e-01
-1.23256540e+00 8.81483376e-01 6.02861595e+00 4.12973642e-01
-1.08263183e+00 2.85597593e-01 4.43694502e-01 -5.21777511e-01
-1.20868320e-02 -1.49744004e-01 -1.04539287e+00 5.67756355e-01
9.58767772e-01 5.56413889e-01 7.86351025e-01 5.01987875e-01
1.01750529e+00 -4.67527658e-01 -4.14382100e-01 7.80002773e-01
-2.21304357e-01 -1.46649230e+00 -2.11724997e-01 5.67676127e-01
7.44791448e-01 4.76110280e-01 -5.08484505e-02 2.62833476e-01
5.01678348e-01 -1.22146893e+00 4.96423304e-01 5.74318945e-01
7.44718015e-01 -7.00232744e-01 7.40712523e-01 4.61152732e-01
-1.70704222e+00 -3.58575195e-01 -2.20338613e-01 -7.35618770e-01
4.57126200e-01 2.46589690e-01 -7.65788674e-01 -4.67927344e-02
1.19026482e+00 4.32871819e-01 -5.47618747e-01 7.36474514e-01
1.82632402e-01 5.75921357e-01 -9.79096115e-01 -1.90058127e-01
5.85248709e-01 -2.88858622e-01 1.24091255e-02 1.19144571e+00
6.38752401e-01 2.34608725e-01 3.22151750e-01 8.82659137e-01
1.45698488e-01 -1.36393011e-01 -8.68985891e-01 6.54806271e-02
9.49709356e-01 1.38101804e+00 -6.01007104e-01 -1.72172219e-01
-3.42047840e-01 4.08053488e-01 3.70852590e-01 2.13055402e-01
-8.33745539e-01 1.29570886e-01 7.31290221e-01 6.27900958e-01
5.11515960e-02 -7.64965951e-01 -9.99266878e-02 -9.07931626e-01
-1.13323852e-01 -2.72436678e-01 1.56124681e-01 -6.18907571e-01
-1.05560124e+00 1.53749451e-01 2.13212922e-01 -9.61552203e-01
2.07927898e-01 -1.44976139e-01 -5.67060053e-01 9.24965382e-01
-1.64355111e+00 -1.84714079e+00 -7.91374087e-01 4.70377564e-01
1.91610768e-01 2.04760984e-01 9.02013719e-01 5.56081116e-01
-5.37027836e-01 9.58286896e-02 2.93421686e-01 2.58852363e-01
3.65236811e-02 -9.80647564e-01 5.97672522e-01 5.03653169e-01
-2.05629826e-01 8.12809289e-01 4.44409490e-01 -8.67828012e-01
-1.17967725e+00 -1.54167461e+00 1.17624688e+00 -5.31528115e-01
6.51555598e-01 -6.12434626e-01 -3.00846159e-01 9.73536193e-01
-2.33724147e-01 -1.40018417e-02 8.72440279e-01 4.27856922e-01
2.09761709e-01 -3.24701667e-01 -1.26137161e+00 6.28231108e-01
1.15932190e+00 -5.50010800e-01 -1.93758115e-01 2.65659958e-01
2.03940213e-01 8.70186463e-02 -1.00662208e+00 1.82598624e-02
9.80598688e-01 -7.65034795e-01 9.63102043e-01 -7.25821406e-02
8.96791816e-02 -3.20953339e-01 -4.56857681e-01 -1.05475140e+00
-1.96359754e-01 -1.31688327e-01 4.76140857e-01 1.20577085e+00
1.55246854e-01 -7.54577756e-01 7.06907213e-01 5.95282972e-01
-8.10245574e-02 -1.51751786e-01 -9.40112829e-01 -4.83862847e-01
-1.19645849e-01 -4.51997340e-01 1.34222233e+00 8.16378415e-01
-3.35096329e-01 4.60934937e-02 -5.22756338e-01 9.59717691e-01
4.21026945e-01 -5.99992424e-02 1.05475128e+00 -1.26330638e+00
4.13434178e-01 -1.16199851e-01 -3.85871053e-01 -6.12536073e-01
-4.95352328e-01 -4.77900028e-01 -3.03054303e-01 -1.73475242e+00
2.52634138e-01 -8.78731132e-01 -3.03788856e-02 7.43257403e-01
1.76934123e-01 5.81318974e-01 -2.65709013e-01 3.01785529e-01
-5.17619967e-01 6.29532099e-01 5.56386232e-01 -2.12775648e-01
-4.71367955e-01 -2.52000660e-01 -5.30787349e-01 4.06253725e-01
9.45664942e-01 -4.53636914e-01 1.25976875e-01 -8.21549714e-01
6.48018599e-01 -2.28172272e-01 8.32666337e-01 -1.42344391e+00
2.26594225e-01 -4.77240682e-01 3.07988107e-01 -1.02904510e+00
4.14263189e-01 -1.01378846e+00 3.09004009e-01 2.79068232e-01
1.80386990e-01 2.29776472e-01 1.58010006e-01 4.33854043e-01
9.33734477e-02 4.92674738e-01 2.81676203e-01 -2.12017626e-01
-3.31103295e-01 6.45155132e-01 -4.75408733e-01 -4.69052017e-01
7.32973099e-01 -5.35631537e-01 -5.31933188e-01 -1.79574639e-01
-6.73910141e-01 5.33796489e-01 6.42872274e-01 2.82608926e-01
2.00445697e-01 -1.25157249e+00 -8.01315427e-01 3.84264350e-01
3.08796585e-01 -3.73571143e-02 4.46646839e-01 9.90512133e-01
-7.91174650e-01 5.76532483e-01 -3.74478132e-01 -5.25305271e-01
-9.60772634e-01 -5.26216961e-02 3.19743365e-01 -2.23839264e-02
-3.10662925e-01 2.26070195e-01 -3.77961040e-01 -8.21983099e-01
-6.97929412e-02 -6.25506282e-01 -3.54904622e-01 7.02781230e-02
7.53656209e-01 6.92265451e-01 -1.07330017e-01 -9.20543492e-01
-2.64734834e-01 3.96158606e-01 5.05917847e-01 -1.33749872e-01
1.89091671e+00 -7.47475028e-01 -1.42511740e-01 4.79783356e-01
8.13484132e-01 -5.49712703e-02 -1.24917400e+00 4.13655080e-02
-3.06569636e-01 -5.08893073e-01 4.53710437e-01 -6.41214788e-01
-7.00419009e-01 7.50056922e-01 1.10099363e+00 5.42721510e-01
7.08936393e-01 -1.76019922e-01 8.54187548e-01 4.85207528e-01
5.80882788e-01 -1.31385064e+00 -7.08495319e-01 2.72948682e-01
7.10166752e-01 -1.50864196e+00 1.40453517e-01 1.68855429e-01
-2.01394930e-01 5.85425019e-01 2.99542516e-01 8.41829088e-03
8.09015989e-01 2.87276566e-01 -2.48005792e-01 -3.56811672e-01
-2.23622784e-01 -6.60986006e-01 -4.14589912e-01 8.06602538e-01
2.99578309e-01 5.89906991e-01 1.66297644e-01 5.05129695e-01
-4.61123675e-01 2.79997557e-01 2.40166232e-01 1.21446049e+00
-8.25303018e-01 -3.98223490e-01 -7.86059797e-01 7.34972298e-01
-8.29721615e-02 -4.81557906e-01 -5.30250482e-02 8.50585699e-01
4.89499778e-01 1.05654919e+00 6.05286181e-01 1.39907021e-02
7.50144273e-02 -4.24363017e-01 1.02724634e-01 -3.11681420e-01
-6.66231811e-01 -3.38287473e-01 1.33971319e-01 -3.24414611e-01
-5.24647057e-01 -7.13476777e-01 -8.36670220e-01 -1.03002489e+00
-1.13921359e-01 -2.21697360e-01 1.00871837e+00 1.10055435e+00
3.65002155e-01 7.72842467e-02 5.73184431e-01 -1.14791882e+00
2.27592625e-02 -1.37691188e+00 -7.46494472e-01 6.74991459e-02
4.55328941e-01 -6.83084846e-01 -1.80664957e-01 -1.19821705e-01] | [9.34260368347168, -1.2613507509231567] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.