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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
399077ba-8576-4317-99e1-d3cd0802b5ff | co-learning-planning-and-control-policies | 2303.01346 | null | https://arxiv.org/abs/2303.01346v1 | https://arxiv.org/pdf/2303.01346v1.pdf | Co-learning Planning and Control Policies Using Differentiable Formal Task Constraints | This paper presents a hierarchical reinforcement learning algorithm constrained by differentiable signal temporal logic. Previous work on logic-constrained reinforcement learning consider encoding these constraints with a reward function, constraining policy updates with a sample-based policy gradient. However, such techniques oftentimes tend to be inefficient because of the significant number of samples required to obtain accurate policy gradients. In this paper, instead of implicitly constraining policy search with sample-based policy gradients, we directly constrain policy search by backpropagating through formal constraints, enabling training hierarchical policies with substantially fewer training samples. The use of hierarchical policies is recognized as a crucial component of reinforcement learning with task constraints. We show that we can stably constrain policy updates, thus enabling different levels of the policy to be learned simultaneously, yielding superior performance compared with training them separately. Experiment results on several simulated high-dimensional robot dynamics and a real-world differential drive robot (TurtleBot3) demonstrate the effectiveness of our approach on five different types of task constraints. Demo videos, code, and models can be found at our project website: https://sites.google.com/view/dscrl | ['Suresh Jagannathan', 'Ahmed H. Qureshi', 'Daniel Lawson', 'Joe Eappen', 'Zikang Xiong'] | 2023-03-02 | null | null | null | null | ['hierarchical-reinforcement-learning'] | ['methodology'] | [ 8.51598307e-02 1.61631301e-01 -6.54953957e-01 -1.66552380e-01
-5.30813992e-01 -6.03860974e-01 7.18227446e-01 -4.87549528e-02
-7.47126937e-01 1.26634693e+00 -1.40847325e-01 -4.99813020e-01
-5.27948402e-02 -4.75490361e-01 -7.42671490e-01 -5.93366385e-01
-4.71363127e-01 5.42236686e-01 2.36670986e-01 -2.52466828e-01
5.27657568e-01 5.42288303e-01 -1.50067759e+00 -1.23242669e-01
7.36483276e-01 7.61196733e-01 4.24266875e-01 7.51351833e-01
1.63780659e-01 1.03498662e+00 -4.71063435e-01 3.98623258e-01
2.42162168e-01 -4.88050222e-01 -5.96037626e-01 1.97164938e-01
1.09836228e-01 -6.14089727e-01 -2.37755656e-01 1.10677564e+00
1.61744341e-01 3.36477816e-01 4.05786842e-01 -1.22976577e+00
-4.12253827e-01 7.47028172e-01 -1.41630948e-01 7.89701287e-03
2.44208902e-01 5.19935250e-01 9.69171286e-01 -3.38654846e-01
5.06639898e-01 1.38841248e+00 2.52828628e-01 8.86045039e-01
-1.51516140e+00 -6.35636866e-01 4.87664908e-01 3.96123417e-02
-9.13471341e-01 -3.80234033e-01 6.03542328e-01 -4.19050157e-01
1.13309598e+00 -5.16254455e-02 8.64722252e-01 1.07954586e+00
2.61211544e-01 7.77491868e-01 1.28495872e+00 -4.00670916e-01
6.18697047e-01 8.68156105e-02 -3.47536117e-01 9.44479346e-01
1.02345847e-01 7.68874168e-01 -3.77703875e-01 -2.87559509e-01
1.17039943e+00 -2.81299800e-01 -1.84052765e-01 -7.53149331e-01
-1.09732401e+00 9.12818313e-01 2.23262951e-01 4.27256711e-02
-3.65267903e-01 8.14980507e-01 3.60165089e-01 5.68169832e-01
3.05731986e-02 6.43210948e-01 -5.23460567e-01 -3.94732833e-01
-5.61722875e-01 7.87008166e-01 8.29730213e-01 9.59506631e-01
8.07895124e-01 4.58904147e-01 -8.16598162e-02 5.87347209e-01
3.21651518e-01 3.30753654e-01 3.83328885e-01 -1.56465101e+00
8.85892808e-02 1.20169930e-01 6.68643892e-01 -4.76551533e-01
-2.02242896e-01 -7.75782764e-02 -1.75158396e-01 9.15195644e-01
4.56116527e-01 -5.58524847e-01 -8.19998980e-01 2.04567003e+00
3.41943502e-01 4.77855429e-02 7.45305698e-03 9.76598859e-01
-3.80744070e-01 5.79057515e-01 3.05738277e-03 -4.52731043e-01
8.49363208e-01 -6.49881721e-01 -7.95226812e-01 -3.10306221e-01
6.13085985e-01 -2.03913674e-01 1.36575651e+00 6.38152182e-01
-1.16000843e+00 -3.07032377e-01 -1.26921427e+00 3.27244669e-01
-7.34501556e-02 -2.03998536e-01 8.14064503e-01 1.80276990e-01
-9.70493793e-01 9.61702287e-01 -1.31745434e+00 -1.59660369e-01
2.36595452e-01 5.94078541e-01 2.37834696e-02 3.62980753e-01
-1.15434742e+00 1.27309382e+00 5.03092170e-01 -1.65011629e-01
-1.33773589e+00 -4.39425796e-01 -7.54685223e-01 -1.50791109e-01
7.55557656e-01 -2.35420823e-01 1.97879231e+00 -6.66925073e-01
-2.08472872e+00 2.92971551e-01 2.14350581e-01 -6.28632188e-01
5.55622339e-01 -2.88678408e-01 1.84138611e-01 7.09474534e-02
-3.65429297e-02 7.70493090e-01 1.06221104e+00 -1.17301428e+00
-7.54773259e-01 8.83916393e-02 4.55515891e-01 3.67774636e-01
-2.63240695e-01 -2.33033597e-01 -1.08684555e-01 -2.57058531e-01
-3.17193747e-01 -1.10526431e+00 -5.36245465e-01 -7.44311232e-03
1.52859062e-01 -3.72131824e-01 9.17697430e-01 -1.41696990e-01
1.03620434e+00 -1.84860551e+00 5.55152714e-01 3.80882062e-02
-1.85418233e-01 2.67017968e-02 -2.36948542e-02 2.81775087e-01
2.92700976e-01 4.89617418e-03 -4.92803007e-01 -1.03144631e-01
3.86359900e-01 6.88537776e-01 -3.88643086e-01 5.80213308e-01
2.59044170e-01 6.56855106e-01 -1.24658287e+00 -4.35051471e-01
2.83336073e-01 2.90460661e-02 -9.70650136e-01 1.52231589e-01
-8.87062848e-01 6.65903687e-01 -8.16504836e-01 5.32942832e-01
3.35096009e-02 1.53743222e-01 6.39467657e-01 4.00993854e-01
-2.23343223e-01 3.85029942e-01 -1.21052432e+00 1.70313084e+00
-4.93358880e-01 3.54191482e-01 4.03509647e-01 -1.23564661e+00
7.44175196e-01 3.13282520e-01 5.51791012e-01 -6.52083278e-01
2.01433465e-01 2.27114320e-01 6.26726076e-02 -3.45970929e-01
4.36174691e-01 -2.79696912e-01 -2.26158440e-01 3.37355584e-01
5.69584146e-02 -7.95695543e-01 4.09869492e-01 4.55986224e-02
1.07754111e+00 9.75370824e-01 2.10623860e-01 -4.02727306e-01
4.25928563e-01 2.41816506e-01 8.10087085e-01 7.78245032e-01
-3.03485751e-01 -3.31115782e-01 8.46386611e-01 -8.57470185e-02
-1.06701684e+00 -8.85165513e-01 -1.22489527e-01 1.10720122e+00
2.43403707e-02 -3.08854520e-01 -4.90049928e-01 -7.42562830e-01
4.11987334e-01 8.29736948e-01 -4.99781221e-01 -4.84184222e-03
-8.91359270e-01 -3.03489715e-01 4.24827635e-01 4.19466496e-01
1.14939436e-01 -1.29100776e+00 -1.30197597e+00 2.98653036e-01
2.39582896e-01 -7.67612517e-01 -3.21383953e-01 6.81733429e-01
-1.08180726e+00 -1.02360547e+00 -3.58238727e-01 -5.92382848e-01
6.31555319e-01 -4.74594504e-01 7.34368265e-01 1.32152811e-01
-1.55708209e-01 5.82266033e-01 -1.40701622e-01 -1.85219660e-01
-5.12048781e-01 -2.00598493e-01 5.29527545e-01 -6.90765500e-01
-8.67464766e-02 -6.75667226e-01 -2.27479994e-01 1.40246481e-01
-7.12274492e-01 -9.37531292e-02 5.11745453e-01 1.21103811e+00
5.60602546e-01 -2.26709202e-01 5.95796824e-01 -5.68714082e-01
8.91597569e-01 -2.15527564e-01 -1.36519420e+00 -1.80675775e-01
-7.40040541e-01 4.21504855e-01 6.96141958e-01 -8.43207777e-01
-9.33863282e-01 1.70870379e-01 1.85727715e-01 -5.41180611e-01
-2.11813003e-01 3.47721994e-01 2.76808709e-01 6.95263818e-02
7.33009219e-01 1.90937743e-02 2.91290551e-01 -1.81132227e-01
5.19042015e-01 3.56652439e-01 3.03560555e-01 -1.23429108e+00
5.95056474e-01 1.36330903e-01 1.40643880e-01 -5.44011950e-01
-4.71419424e-01 5.16242050e-02 -3.00393671e-01 -3.67518097e-01
5.69579363e-01 -4.89034295e-01 -1.06233811e+00 -2.55710050e-03
-7.27259338e-01 -1.15057063e+00 -3.77677441e-01 6.65633142e-01
-1.22877979e+00 1.45930454e-01 -6.68953061e-01 -1.12069428e+00
2.39253670e-01 -1.32638896e+00 5.97968578e-01 4.02779840e-02
-3.17076445e-01 -8.63819718e-01 1.12745531e-01 -4.70442116e-01
3.87401193e-01 3.37656796e-01 8.63835335e-01 -2.10660562e-01
-4.79938865e-01 8.58413130e-02 4.16126221e-01 2.90765733e-01
1.75684348e-01 -2.18912423e-01 -5.54347336e-01 -5.78189135e-01
1.01500362e-01 -9.85004365e-01 4.98887837e-01 3.38552088e-01
9.71342444e-01 -3.72896552e-01 -2.98103333e-01 3.63076001e-01
1.39092863e+00 4.45832610e-01 1.27062917e-01 4.42424893e-01
3.19776475e-01 5.03915429e-01 1.19063258e+00 7.02897847e-01
4.10339748e-03 5.74865401e-01 6.88734531e-01 5.71578324e-01
2.59649515e-01 -2.96510071e-01 7.46744514e-01 2.67187417e-01
6.50167391e-02 2.68992633e-01 -7.28589773e-01 4.12854433e-01
-2.16498208e+00 -8.81063163e-01 4.57779258e-01 2.00522351e+00
1.22535503e+00 3.90333772e-01 2.72519767e-01 -6.34246543e-02
4.28482711e-01 -1.10667914e-01 -1.18781054e+00 -6.46305740e-01
5.03838897e-01 2.20579296e-01 6.26531661e-01 1.08609688e+00
-8.88564110e-01 1.17373002e+00 6.31002855e+00 4.38690692e-01
-1.19502056e+00 -1.87695250e-01 5.31568639e-02 -4.56973225e-01
-1.00393526e-01 2.65838683e-01 -7.66328096e-01 3.34991693e-01
8.23567629e-01 -1.71735451e-01 9.24818337e-01 9.52197552e-01
4.34944212e-01 -3.58327270e-01 -1.36417210e+00 5.68290174e-01
-5.83866596e-01 -1.15833139e+00 -3.25924754e-01 1.41925618e-01
5.03041565e-01 -7.99778998e-02 -2.06791097e-03 7.48594403e-01
8.12526107e-01 -7.14000642e-01 9.72017646e-01 3.13092887e-01
6.79505467e-01 -8.29390407e-01 2.75902115e-02 6.26525342e-01
-7.31140375e-01 -5.30959308e-01 -4.33761060e-01 -4.98220891e-01
4.53978218e-02 7.41672814e-02 -7.79463172e-01 4.78967838e-02
5.41035771e-01 7.18778253e-01 2.05610260e-01 6.43532515e-01
-5.59414148e-01 5.21550357e-01 -4.87891614e-01 -5.35527945e-01
4.85162288e-01 -2.96189755e-01 6.12648964e-01 1.06100810e+00
7.65050054e-02 7.14593530e-02 5.37500143e-01 1.04632664e+00
3.58086020e-01 -3.92483264e-01 -6.84519053e-01 -2.20383480e-01
4.89906639e-01 1.04735911e+00 -5.98977447e-01 -3.27569246e-01
-7.51664937e-02 7.30957031e-01 5.76554537e-01 3.60691339e-01
-1.04851818e+00 -3.53473872e-01 8.79811406e-01 -2.06125125e-01
6.10035300e-01 -8.13737214e-01 7.99395889e-02 -1.06476629e+00
-2.88004637e-01 -1.02117610e+00 3.86489816e-02 -2.69468993e-01
-8.18084002e-01 5.15339337e-02 3.80672663e-01 -1.28366399e+00
-7.55479932e-01 -6.47184968e-01 -1.81872502e-01 7.22120166e-01
-1.56029415e+00 -5.41630507e-01 3.26360881e-01 6.14882112e-01
5.64236939e-01 -4.90792729e-02 8.75186026e-01 -3.94244976e-02
-3.23363721e-01 1.79166853e-01 -9.29784775e-02 -2.58051097e-01
5.98703504e-01 -1.49839127e+00 -9.06536803e-02 5.98206341e-01
-3.54004353e-01 6.38028741e-01 1.02855992e+00 -6.64427102e-01
-1.60972524e+00 -6.57969236e-01 1.11219995e-01 -1.04379490e-01
9.67253864e-01 -3.80132586e-01 -7.28961527e-01 7.93369353e-01
2.54571736e-01 -2.03743920e-01 1.15880467e-01 -7.68554285e-02
1.03645176e-01 1.11937113e-01 -1.14508963e+00 8.55483890e-01
9.63147104e-01 -3.35417032e-01 -6.68523014e-01 2.34582618e-01
6.43481851e-01 -6.37914598e-01 -8.01111281e-01 3.39473128e-01
5.11005878e-01 -6.75620556e-01 5.08681774e-01 -6.05513036e-01
2.26155043e-01 -5.58658600e-01 -2.00454995e-01 -1.48994458e+00
-3.05416673e-01 -7.65494764e-01 -5.53999305e-01 7.23241150e-01
1.95740655e-01 -7.02313602e-01 8.54276419e-01 3.86270493e-01
-2.28314385e-01 -9.52390552e-01 -7.69607306e-01 -1.01088476e+00
3.93059075e-01 -4.19500858e-01 2.22786307e-01 9.04033542e-01
5.41834533e-01 2.12728027e-02 -3.94163638e-01 1.91940337e-01
6.18689179e-01 -1.92455873e-02 5.06864607e-01 -7.40960598e-01
-7.86602259e-01 -5.76020360e-01 4.51156348e-02 -1.28379965e+00
4.90864366e-01 -8.47047031e-01 5.45340240e-01 -1.30046380e+00
-3.65290761e-01 -6.51732087e-01 -2.39682928e-01 8.54787588e-01
2.01503754e-01 -2.72993028e-01 2.70031750e-01 1.25076413e-01
-3.72596681e-01 7.04147100e-01 1.39794242e+00 5.89931235e-02
-4.43321824e-01 -1.96618035e-01 -3.59795541e-01 7.43729293e-01
1.32659113e+00 -4.31735873e-01 -7.35785425e-01 -3.05376679e-01
-5.22327013e-02 4.38374370e-01 2.45491967e-01 -9.73067999e-01
2.20231444e-01 -7.89182007e-01 2.78736413e-01 -2.04231963e-01
4.59593862e-01 -7.48754442e-01 -4.31810021e-01 1.05644202e+00
-6.09467566e-01 1.44796401e-01 4.77131486e-01 5.14491081e-01
9.45343375e-02 -2.49636337e-01 8.58569920e-01 -2.43353710e-01
-8.16281438e-01 6.96853995e-02 -8.80130410e-01 -1.55313075e-01
1.09143066e+00 1.79924205e-01 1.05339577e-02 -2.78025240e-01
-8.43180060e-01 5.85797548e-01 4.51841891e-01 2.44030058e-01
7.46945858e-01 -1.15003002e+00 -3.43996346e-01 5.22065908e-02
-3.10860813e-01 -5.43016754e-02 -4.46545213e-01 6.90627038e-01
-2.28043094e-01 3.69558245e-01 -4.55756485e-01 -5.95937908e-01
-8.03090155e-01 4.91386145e-01 5.54495275e-01 -2.16791719e-01
-5.70495844e-01 6.72963202e-01 -3.60447764e-01 -5.51369786e-01
4.87848908e-01 -6.35392964e-01 1.01640411e-01 -1.61540821e-01
2.09615901e-01 8.66083279e-02 -5.09709477e-01 1.62922204e-01
-2.55444527e-01 3.20192814e-01 -3.86440456e-02 -7.52054632e-01
1.18140316e+00 1.57298639e-01 1.35703772e-01 5.25961220e-01
5.55582941e-01 -2.38240197e-01 -1.87124550e+00 -8.89521465e-02
2.27857962e-01 -3.26545715e-01 1.08176015e-01 -7.70469129e-01
-4.61586118e-01 4.86515284e-01 3.95136088e-01 1.29759476e-01
8.75717580e-01 -3.37973237e-01 2.03772500e-01 9.22422826e-01
8.24324667e-01 -1.32203591e+00 3.06983292e-01 8.03195059e-01
9.05638456e-01 -9.99919713e-01 2.10242584e-01 2.37267718e-01
-6.56771362e-01 1.03119957e+00 8.61511111e-01 -6.16281986e-01
4.33589607e-01 6.05688035e-01 -1.90555111e-01 2.99022585e-01
-1.08323884e+00 -2.50287175e-01 -3.78179550e-01 5.48939526e-01
2.67649531e-01 -6.72019273e-02 -4.64248002e-01 2.48047225e-02
-2.38103978e-02 3.12530458e-01 6.55469000e-01 1.57201457e+00
-8.98250103e-01 -1.54297972e+00 -3.44935119e-01 2.48075739e-01
-3.41042697e-01 1.67021573e-01 7.32419714e-02 9.30057883e-01
-1.69278800e-01 8.42743397e-01 3.99300307e-02 -1.86017960e-01
2.69777894e-01 2.65102461e-02 9.23962116e-01 -7.12578177e-01
-3.25144440e-01 1.44582033e-01 1.72440782e-01 -8.16080689e-01
-2.35731661e-01 -9.03807461e-01 -1.79437280e+00 -1.18585289e-01
2.73902547e-02 2.75799483e-01 6.12432241e-01 7.01933026e-01
-1.70330703e-02 5.68393528e-01 5.92915714e-01 -1.18740153e+00
-1.22670245e+00 -7.85211265e-01 -5.06850719e-01 4.24480215e-02
7.20923722e-01 -1.02013278e+00 -3.92340004e-01 -6.81650788e-02] | [4.274981498718262, 1.7170690298080444] |
1fed9d60-d8ef-4035-ae17-8b41c30dfd4d | geometric-visual-similarity-learning-in-3d | 2303.00874 | null | https://arxiv.org/abs/2303.00874v1 | https://arxiv.org/pdf/2303.00874v1.pdf | Geometric Visual Similarity Learning in 3D Medical Image Self-supervised Pre-training | Learning inter-image similarity is crucial for 3D medical images self-supervised pre-training, due to their sharing of numerous same semantic regions. However, the lack of the semantic prior in metrics and the semantic-independent variation in 3D medical images make it challenging to get a reliable measurement for the inter-image similarity, hindering the learning of consistent representation for same semantics. We investigate the challenging problem of this task, i.e., learning a consistent representation between images for a clustering effect of same semantic features. We propose a novel visual similarity learning paradigm, Geometric Visual Similarity Learning, which embeds the prior of topological invariance into the measurement of the inter-image similarity for consistent representation of semantic regions. To drive this paradigm, we further construct a novel geometric matching head, the Z-matching head, to collaboratively learn the global and local similarity of semantic regions, guiding the efficient representation learning for different scale-level inter-image semantic features. Our experiments demonstrate that the pre-training with our learning of inter-image similarity yields more powerful inner-scene, inter-scene, and global-local transferring ability on four challenging 3D medical image tasks. Our codes and pre-trained models will be publicly available on https://github.com/YutingHe-list/GVSL. | ['Shuo Li', 'Boyu Wang', 'Jean-Louis Coatrieux', 'Yang Chen', 'Rongjun Ge', 'Guanyu Yang', 'Yuting He'] | 2023-03-02 | null | http://openaccess.thecvf.com//content/CVPR2023/html/He_Geometric_Visual_Similarity_Learning_in_3D_Medical_Image_Self-Supervised_Pre-Training_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/He_Geometric_Visual_Similarity_Learning_in_3D_Medical_Image_Self-Supervised_Pre-Training_CVPR_2023_paper.pdf | cvpr-2023-1 | ['geometric-matching'] | ['computer-vision'] | [ 1.52776882e-01 -1.10048018e-01 -1.52776480e-01 -7.33080029e-01
-8.03937137e-01 -4.64913428e-01 4.61724430e-01 3.91103566e-01
-2.73825116e-02 -6.90969229e-02 2.78825969e-01 -4.14786935e-02
-2.89547414e-01 -5.96885085e-01 -4.75422293e-01 -6.53549373e-01
4.57138866e-02 1.72942057e-01 3.03016990e-01 -1.71950370e-01
2.43942320e-01 4.66266811e-01 -1.36208725e+00 3.88422698e-01
7.39326119e-01 1.00297248e+00 6.37602150e-01 2.83851236e-01
-2.24948809e-01 4.70847368e-01 -2.36439735e-01 1.77692741e-01
3.21171850e-01 -6.38258338e-01 -8.78675461e-01 1.58568665e-01
5.25715828e-01 1.20172478e-01 -2.38433585e-01 1.35961807e+00
6.36083841e-01 4.27589081e-02 8.04882467e-01 -1.27931333e+00
-9.51097369e-01 -9.32406411e-02 -5.80681264e-01 3.03087741e-01
4.70017463e-01 -4.56412882e-03 9.42114413e-01 -7.80639172e-01
8.29428136e-01 1.26938283e+00 6.05716527e-01 3.67465764e-01
-9.69010055e-01 -5.27294397e-01 -7.79752852e-03 3.06806058e-01
-1.65503991e+00 -1.32051185e-01 1.04882610e+00 -4.89917994e-01
5.39369226e-01 2.07287163e-01 4.86532599e-01 7.90997326e-01
4.20035213e-01 5.83885729e-01 1.24898040e+00 -2.67852217e-01
8.72070640e-02 -3.82709242e-02 -1.62512087e-03 1.02267075e+00
1.00570105e-01 6.31330442e-03 -2.78489500e-01 -1.27898082e-01
1.07603586e+00 7.34238148e-01 -2.58808672e-01 -9.23359931e-01
-1.25839281e+00 8.59435737e-01 9.66292143e-01 5.86814165e-01
-4.48162295e-02 -1.98323820e-02 4.78123307e-01 4.61812228e-01
4.18117166e-01 3.18009466e-01 -4.05869424e-01 4.25545096e-01
-5.19632578e-01 -4.59389463e-02 1.92047775e-01 1.08297861e+00
9.41077650e-01 -4.94853616e-01 -1.80910364e-01 8.52297008e-01
2.69482493e-01 6.05010271e-01 6.91256881e-01 -8.52099836e-01
1.85596019e-01 8.78953934e-01 -4.77880597e-01 -1.38976157e+00
-5.78091025e-01 -4.75064695e-01 -1.18985105e+00 1.56301349e-01
3.65752965e-01 5.84493101e-01 -1.06662965e+00 1.87306273e+00
6.44410789e-01 3.77231896e-01 2.79851053e-02 9.41604197e-01
1.07272804e+00 2.35425234e-01 -1.88880637e-02 2.28526499e-02
1.43957150e+00 -8.65269721e-01 -4.26285923e-01 -5.78323752e-02
8.97465289e-01 -9.25339341e-01 1.16145515e+00 -3.47714901e-01
-8.16193640e-01 -7.74561405e-01 -8.13526213e-01 -3.01664203e-01
-4.11389142e-01 -2.33069375e-01 5.57836592e-01 1.33981870e-03
-8.44690561e-01 4.83253807e-01 -8.39120269e-01 -6.22804761e-01
5.22808611e-01 5.75227328e-02 -7.21094012e-01 -3.89098883e-01
-8.42862606e-01 8.13291728e-01 2.76579976e-01 -4.18919832e-01
-6.89521313e-01 -9.37700689e-01 -1.06582808e+00 -2.84782171e-01
3.21608484e-02 -1.05782652e+00 6.08593047e-01 -9.24386919e-01
-9.99507606e-01 1.61809742e+00 -6.12891130e-02 3.03272516e-01
2.78378606e-01 2.42909640e-01 -2.70197630e-01 3.61280024e-01
5.41390359e-01 6.40161574e-01 7.67700970e-01 -1.37852681e+00
-2.24199936e-01 -7.12676704e-01 -2.37069383e-01 4.11702245e-01
-6.54072389e-02 -1.68819070e-01 -5.38424909e-01 -8.18349183e-01
6.61806464e-01 -9.59730029e-01 -2.85079330e-01 5.18029332e-01
-3.13890934e-01 -1.22399494e-01 5.79150617e-01 -4.84094113e-01
7.14348674e-01 -2.35973072e+00 2.01644674e-01 4.94649380e-01
2.61789888e-01 -1.90980434e-01 -4.82045949e-01 1.58937111e-01
-2.79110014e-01 -1.74480855e-01 -3.15390468e-01 -1.27898306e-01
-3.50614727e-01 2.92093903e-01 8.09862539e-02 6.19176626e-01
6.04025014e-02 1.14322019e+00 -1.25979400e+00 -8.66383433e-01
3.84626657e-01 3.30972135e-01 -4.44666773e-01 5.56758940e-01
3.58696401e-01 9.36810613e-01 -7.85948515e-01 6.91760004e-01
7.21295774e-01 -7.08718717e-01 -1.48887306e-01 -5.47165930e-01
4.20845300e-01 -3.19639221e-02 -9.74967360e-01 2.59374261e+00
-4.66806203e-01 -2.09797230e-02 -1.60656989e-01 -1.41099012e+00
1.08892512e+00 -3.53295058e-02 8.33950341e-01 -9.40260828e-01
2.27272332e-01 3.02333862e-01 -2.78201431e-01 -5.43790042e-01
-1.48134217e-01 -1.63696244e-01 -9.59440321e-02 5.33891797e-01
1.96547896e-01 -6.18111968e-01 -2.66220182e-01 2.50964105e-01
8.65672052e-01 -2.68216133e-01 5.08619189e-01 -4.56246763e-01
6.77679896e-01 -2.75895059e-01 4.61552650e-01 5.43408692e-01
-3.87847334e-01 1.08396685e+00 6.36712909e-02 -4.26802039e-01
-9.57858562e-01 -1.55680847e+00 -4.20939177e-01 7.14768112e-01
8.73060286e-01 -3.50687534e-01 -4.43733662e-01 -8.80430162e-01
1.88596398e-01 -2.47660163e-03 -8.52051198e-01 -5.04584849e-01
-3.45777303e-01 -4.29795802e-01 5.13062663e-02 3.76794457e-01
3.71502429e-01 -6.65427268e-01 -2.20187292e-01 -2.98914224e-01
-2.86081672e-01 -9.68393266e-01 -8.46804142e-01 4.96455990e-02
-1.01406062e+00 -1.22257578e+00 -8.78181458e-01 -1.14474404e+00
1.11355531e+00 7.33211100e-01 1.08540821e+00 5.18263698e-01
-5.43227434e-01 5.35201192e-01 -4.18466747e-01 -9.29406434e-02
-3.76347899e-01 -1.23271272e-01 -7.74758309e-02 -1.89174980e-01
2.10898563e-01 -6.68937266e-01 -9.30863559e-01 5.57640254e-01
-7.88312316e-01 3.10867101e-01 6.71755016e-01 9.53799486e-01
1.00888515e+00 -4.16201204e-01 3.19608599e-01 -6.06337547e-01
2.45938048e-01 -5.98799944e-01 1.14071578e-01 4.57156807e-01
-3.96431804e-01 1.23593502e-01 3.47678870e-01 -2.17458114e-01
-4.83464897e-01 -1.11962996e-01 -1.09511651e-01 -6.53154671e-01
-2.86742061e-01 2.19783977e-01 -1.01180427e-01 -3.56031626e-01
5.65218687e-01 3.86962265e-01 4.10026103e-01 -2.95541763e-01
4.92856085e-01 4.16199148e-01 5.41361570e-01 -4.83150989e-01
9.35115874e-01 8.04909348e-01 1.76268652e-01 -6.62109852e-01
-1.13155019e+00 -9.59735632e-01 -8.41174066e-01 7.89960101e-02
1.11317933e+00 -1.09045386e+00 -2.21873760e-01 3.71689647e-01
-9.87705886e-01 -1.25609815e-01 -3.26724231e-01 5.27598858e-01
-7.36565650e-01 7.63755679e-01 -4.08544987e-01 2.31023386e-01
-4.34214413e-01 -1.18350434e+00 1.36249375e+00 1.07493103e-01
-3.05015236e-01 -1.30134463e+00 2.86390781e-01 4.31836814e-01
3.31182688e-01 2.82515973e-01 9.79280651e-01 -5.12161434e-01
-3.49252641e-01 5.64335026e-02 -6.11417234e-01 3.32808763e-01
7.26976693e-01 -4.60244507e-01 -6.53400421e-01 -5.96084416e-01
8.15341249e-02 -3.33417535e-01 6.84923351e-01 5.12693107e-01
1.48534453e+00 1.28254578e-01 -4.76095498e-01 1.10341370e+00
1.35661519e+00 -2.82430083e-01 2.78341204e-01 1.24598257e-01
9.83300149e-01 6.51693285e-01 6.63285792e-01 2.91874170e-01
5.61384082e-01 7.83997774e-01 1.98669657e-01 -6.23643160e-01
-3.52392316e-01 -3.52611005e-01 -1.05952308e-01 1.27109754e+00
4.89861593e-02 3.32555056e-01 -7.59018958e-01 3.49267274e-01
-1.82892215e+00 -7.84158349e-01 2.13500574e-01 2.01336193e+00
9.09777343e-01 -3.28828514e-01 -1.99087113e-01 -2.43435755e-01
7.65786290e-01 1.20315321e-01 -4.96016532e-01 9.49584171e-02
-1.67768002e-02 1.10368885e-01 2.80645430e-01 4.64308619e-01
-1.11592662e+00 8.06888700e-01 5.36352444e+00 1.08905137e+00
-1.23550320e+00 2.80447364e-01 7.97340512e-01 2.49297217e-01
-5.54920912e-01 -1.58513397e-01 -3.68025750e-01 4.38755840e-01
1.54154539e-01 -1.29336804e-01 -5.91210723e-02 7.48056114e-01
1.22566834e-01 1.57377824e-01 -1.08514953e+00 1.46956432e+00
4.55863684e-01 -1.31361473e+00 4.91849065e-01 -1.51926562e-01
8.87196839e-01 -3.65401953e-02 2.61394978e-02 -1.95717156e-01
6.67626485e-02 -1.10360324e+00 4.91337806e-01 6.47354364e-01
9.15723383e-01 -4.96380687e-01 6.62958622e-01 -1.19974345e-01
-1.41982555e+00 2.19648749e-01 -5.79661310e-01 3.42128456e-01
5.16241342e-02 5.64339697e-01 -4.50517863e-01 7.96119034e-01
8.02710056e-01 1.30289721e+00 -7.69074976e-01 7.75059283e-01
-1.13604985e-01 -1.28739312e-01 -3.07322126e-02 3.65422517e-01
1.92223713e-01 -3.13792437e-01 2.86965936e-01 1.05299997e+00
4.16249484e-01 3.80227156e-02 5.04756272e-01 8.86418104e-01
1.41714305e-01 2.84581393e-01 -6.92884386e-01 4.73483533e-01
4.36162770e-01 1.23622298e+00 -8.47318470e-01 -2.69955963e-01
-4.32674944e-01 1.19628954e+00 2.84157842e-01 2.60173500e-01
-6.91005647e-01 -2.06323653e-01 8.42239320e-01 2.76735574e-02
3.62645611e-02 -3.02465588e-01 -4.43458408e-01 -1.18243206e+00
1.69193428e-02 -6.88693643e-01 7.03143835e-01 -7.08593786e-01
-1.70776212e+00 3.53074223e-01 -2.23947484e-02 -1.48931384e+00
5.94336316e-02 -5.50562143e-01 -8.29953730e-01 7.14983761e-01
-1.61944962e+00 -1.47719622e+00 -8.26238632e-01 1.24456358e+00
1.94956690e-01 -2.32875496e-01 7.68099546e-01 1.38235167e-01
-1.49900960e-02 7.34822452e-01 1.36485159e-01 1.39337450e-01
1.03895116e+00 -1.11067772e+00 1.24435224e-01 3.36373955e-01
1.54056340e-01 6.16570771e-01 3.22489589e-01 -4.13145840e-01
-1.28215647e+00 -1.10353935e+00 6.53216302e-01 -6.15565896e-01
5.92988253e-01 -3.08068395e-01 -1.06976438e+00 3.59381258e-01
-2.13340938e-01 5.20408273e-01 8.51431787e-01 -1.33097425e-01
-5.36190212e-01 -6.53454438e-02 -1.13898313e+00 4.29552794e-01
1.47764099e+00 -9.49728847e-01 -7.30232596e-01 4.54279870e-01
8.28326464e-01 -3.12996447e-01 -1.16729045e+00 4.91483361e-01
2.41811603e-01 -8.73362482e-01 1.48445129e+00 -4.47973490e-01
3.88822734e-01 -5.81347883e-01 -3.08218181e-01 -1.19643700e+00
-4.61572766e-01 -2.63040978e-02 5.50344527e-01 8.73773277e-01
-4.63271067e-02 -5.18654704e-01 4.99086887e-01 7.73729309e-02
-4.16506827e-01 -8.14370811e-01 -9.71953630e-01 -8.77646089e-01
3.44035923e-01 -1.67947084e-01 4.04977977e-01 1.44535184e+00
-9.92964879e-02 1.08719811e-01 2.94054039e-02 2.05868229e-01
7.67414808e-01 4.48902041e-01 7.66767621e-01 -9.86918151e-01
-1.60644084e-01 -4.93431181e-01 -1.08298409e+00 -8.67590606e-01
2.01915473e-01 -1.77380300e+00 -2.70931631e-01 -1.38054168e+00
6.86837971e-01 -6.60246789e-01 -5.96953332e-01 3.94829273e-01
-3.23918819e-01 3.42344970e-01 5.02645373e-02 4.81303751e-01
-8.60369742e-01 7.75188625e-01 1.80430448e+00 -2.74971157e-01
4.48580980e-02 -3.91487330e-01 -6.19477034e-01 6.02509797e-01
5.67022324e-01 -4.22883391e-01 -6.04667723e-01 -3.72125119e-01
-1.33926675e-01 -2.06637472e-01 6.43279672e-01 -9.92931366e-01
1.83673292e-01 -2.86324888e-01 5.06434858e-01 -3.02234292e-01
-1.95719860e-02 -7.12846041e-01 -3.88419889e-02 6.06153965e-01
-3.83196592e-01 5.77520728e-02 -5.07096201e-02 5.38403928e-01
-4.26515937e-01 6.73170462e-02 1.07906473e+00 -3.16907793e-01
-7.94770598e-01 7.48534620e-01 3.83991659e-01 4.56643283e-01
1.08463311e+00 -3.86061311e-01 -1.72310099e-01 -8.66517201e-02
-6.47495449e-01 2.17688173e-01 8.36241424e-01 5.74882090e-01
9.15081143e-01 -1.60036814e+00 -7.33934820e-01 2.91680902e-01
7.81650066e-01 6.91077858e-02 5.43541133e-01 9.55593526e-01
-5.14027297e-01 8.77575651e-02 -6.08685076e-01 -1.18624771e+00
-1.33836007e+00 7.23403692e-01 4.37394172e-01 -2.43965343e-01
-8.26383710e-01 7.43394852e-01 9.27096248e-01 -7.22834587e-01
-1.39729992e-01 -3.69384646e-01 6.96388409e-02 -3.35647434e-01
4.15621340e-01 -1.76678002e-02 6.93631638e-03 -7.70666599e-01
-6.14491403e-01 1.40950036e+00 -7.92974085e-02 4.49285120e-01
1.11506557e+00 -4.26050305e-01 -1.83281288e-01 3.63074034e-01
1.87941170e+00 -4.62248743e-01 -8.98235023e-01 -5.61031342e-01
-1.34975612e-01 -6.99438393e-01 -8.04910585e-02 -3.06480676e-01
-1.10983181e+00 8.50239635e-01 1.05078149e+00 -3.61684769e-01
1.00289905e+00 6.19240165e-01 7.45588779e-01 2.08331496e-01
4.12209064e-01 -9.07992661e-01 6.13584697e-01 4.23651606e-01
1.07511163e+00 -1.70283759e+00 1.29863396e-01 -5.59277058e-01
-5.25338531e-01 9.01230276e-01 4.81145173e-01 -2.78396249e-01
1.00933135e+00 -1.88387617e-01 2.48483643e-01 -5.82017303e-01
-2.96520680e-01 -2.48458579e-01 7.28850543e-01 7.94708252e-01
4.78652984e-01 1.03202775e-01 -4.06129546e-02 1.78119957e-01
-9.14032012e-02 -3.57776999e-01 -7.04116672e-02 7.64990211e-01
-2.44294181e-01 -9.67033923e-01 -1.25874951e-01 2.57136643e-01
-1.23920530e-01 3.24292779e-02 -1.20262548e-01 6.03444517e-01
2.38125920e-01 5.33681750e-01 2.85781145e-01 -3.86958331e-01
2.75126487e-01 -4.03419048e-01 5.91173768e-01 -7.51805067e-01
-1.37212202e-01 -2.33273417e-01 -5.35192072e-01 -9.19047892e-01
-4.47267383e-01 -4.84751850e-01 -1.42373347e+00 -1.92543373e-01
3.56379002e-02 -2.64422614e-02 3.63778442e-01 8.71908367e-01
4.79057997e-01 2.43664578e-01 1.01313102e+00 -6.64425910e-01
-3.98659706e-01 -5.90488434e-01 -7.56378353e-01 1.41801870e+00
1.34254590e-01 -8.93843055e-01 -4.14532304e-01 -9.17091966e-02] | [8.070387840270996, -3.1978418827056885] |
d1c0d478-0d26-4950-8e84-8fda801b021e | bangla-license-plate-recognition-using | 1809.00905 | null | http://arxiv.org/abs/1809.00905v1 | http://arxiv.org/pdf/1809.00905v1.pdf | Bangla License Plate Recognition Using Convolutional Neural Networks (CNN) | In the last few years, the deep learning technique in particular
Convolutional Neural Networks (CNNs) is using massively in the field of
computer vision and machine learning. This deep learning technique provides
state-of-the-art accuracy in different classification, segmentation, and
detection tasks on different benchmarks such as MNIST, CIFAR-10, CIFAR-100,
Microsoft COCO, and ImageNet. However, there are a lot of research has been
conducted for Bangla License plate recognition with traditional machine
learning approaches in last decade. None of them are used to deploy a physical
system for Bangla License Plate Recognition System (BLPRS) due to their poor
recognition accuracy. In this paper, we have implemented CNNs based Bangla
license plate recognition system with better accuracy that can be applied for
different purposes including roadside assistance, automatic parking lot
management system, vehicle license status detection and so on. Along with that,
we have also created and released a very first and standard database for BLPRS. | ['M M Shaifur Rahman', 'Mst Shamima Nasrin', 'Moin Mostakim', 'Md Zahangir Alom'] | 2018-09-04 | null | null | null | null | ['license-plate-recognition'] | ['computer-vision'] | [-5.53176522e-01 -6.92351520e-01 -1.39485031e-01 -4.34804827e-01
-4.78061944e-01 -3.07917327e-01 4.06686604e-01 -3.95482630e-01
-5.41815996e-01 6.05766475e-01 -4.10629243e-01 -5.16100228e-01
2.76387513e-01 -9.40329909e-01 -4.59201396e-01 -5.20287454e-01
4.81944472e-01 5.27961075e-01 7.59318888e-01 -4.19025600e-01
5.77514410e-01 9.01040494e-01 -1.26088524e+00 9.78976712e-02
6.02229834e-01 8.99626553e-01 2.41404306e-02 4.36351955e-01
-1.87636703e-01 8.06297362e-01 -3.41696173e-01 -4.00584280e-01
5.71366549e-01 3.24220359e-02 -4.67326254e-01 7.67489076e-02
2.53410578e-01 -4.62584585e-01 -9.13816869e-01 7.76406467e-01
5.48886776e-01 -6.07859157e-02 8.30938637e-01 -1.13531935e+00
-6.82292998e-01 -2.24769898e-02 -7.71717310e-01 5.18760085e-01
-7.10104227e-01 2.94318885e-01 2.47586027e-01 -1.11084926e+00
1.41947344e-01 8.56883347e-01 1.25649929e+00 4.26096708e-01
-2.25612521e-01 -8.54268909e-01 -8.07837307e-01 5.10393977e-01
-1.71383286e+00 -3.88099372e-01 6.02171957e-01 -4.99292821e-01
1.15056825e+00 3.91066372e-02 3.44390690e-01 2.86589921e-01
3.77202392e-01 1.02897525e+00 1.19634342e+00 -4.90635037e-01
1.07617863e-01 4.16832656e-01 5.48580289e-01 7.70744085e-01
2.99771965e-01 -3.25027704e-01 2.21348450e-01 6.06581271e-01
1.10068405e+00 3.13989818e-01 5.55268288e-01 3.60764444e-01
-6.79216802e-01 8.90335381e-01 1.00755863e-01 4.41330701e-01
-2.07358971e-01 5.74789708e-03 4.45951253e-01 5.76932281e-02
1.62914991e-01 -4.82283831e-02 -3.35344702e-01 -1.27625585e-01
-1.11798000e+00 1.86404124e-01 5.31625926e-01 8.88691366e-01
9.24808323e-01 5.40131688e-01 6.29698187e-02 1.30486727e+00
4.21487123e-01 5.50657153e-01 8.00532699e-01 -3.00211966e-01
5.69657981e-01 5.22903502e-01 -5.12795784e-02 -1.18684399e+00
-4.18964118e-01 -1.19416602e-01 -8.47117662e-01 5.26722908e-01
1.25197724e-01 -1.32508308e-01 -1.30727482e+00 5.43498158e-01
-3.46605122e-01 1.57802314e-01 1.38090372e-01 7.12800682e-01
9.59597170e-01 1.09637690e+00 -6.47379309e-02 5.70885837e-01
1.29636192e+00 -1.32630396e+00 -4.74646389e-01 -4.26936209e-01
5.54684162e-01 -9.82976317e-01 7.52183497e-01 1.86638802e-01
-8.77712011e-01 -8.99039567e-01 -1.30845654e+00 -6.61632419e-02
-8.72818172e-01 5.85548460e-01 7.99422622e-01 1.06055140e+00
-9.98291910e-01 1.48636982e-01 -6.67484760e-01 -7.80010343e-01
8.76256049e-01 7.96897054e-01 -3.25936526e-01 -1.95045531e-01
-9.64448452e-01 1.16316807e+00 2.42229953e-01 1.03294298e-01
-7.11595356e-01 5.66082746e-02 -3.74216884e-01 -5.38609400e-02
-6.67030811e-02 9.55325589e-02 9.44381773e-01 -9.50753868e-01
-1.38246024e+00 1.20437062e+00 2.98870236e-01 -7.04867065e-01
1.34935752e-01 2.18431965e-01 -9.92321968e-01 -2.99256027e-01
-8.59000161e-02 7.74864197e-01 2.91273534e-01 -7.20309556e-01
-8.19470406e-01 -7.01195002e-01 -1.82119414e-01 9.91956517e-02
-6.60856813e-02 2.69145399e-01 -6.19145215e-01 -3.52655083e-01
3.09191514e-02 -8.95080566e-01 -1.61195025e-02 -2.06411406e-01
-3.88271779e-01 -5.19768000e-01 1.38920486e+00 -6.63314760e-01
6.23659551e-01 -2.31686783e+00 -1.12016654e+00 1.67125359e-01
-2.00738817e-01 1.03611529e+00 7.26589710e-02 3.14725190e-01
1.45286784e-01 2.00468540e-01 3.54049094e-02 -9.26778242e-02
-1.50050640e-01 1.74434543e-01 1.54208159e-02 6.90850973e-01
2.36633167e-01 8.97046089e-01 1.93199232e-01 -5.66174865e-01
7.44846940e-01 3.70917439e-01 -7.36001926e-03 -2.41006792e-01
4.15946871e-01 -1.63015559e-01 -4.63012099e-01 1.12005746e+00
1.25817192e+00 8.07394683e-02 -5.22772551e-01 -4.37188298e-02
-4.49762881e-01 -4.50018466e-01 -8.64755154e-01 9.16430771e-01
-2.43334115e-01 1.32096517e+00 -7.59031102e-02 -1.55705023e+00
1.35639095e+00 8.88020843e-02 2.18336761e-01 -1.10374331e+00
4.08477545e-01 4.96402860e-01 -5.48673384e-02 -8.25785995e-01
4.92353469e-01 -7.17505533e-03 1.74765140e-01 -1.63558394e-01
6.38775080e-02 2.46801808e-01 -1.16565935e-02 -2.95472503e-01
6.30871236e-01 -3.58863443e-01 -9.33738500e-02 -1.48932368e-01
8.96894872e-01 5.43481767e-01 3.08610111e-01 6.83138311e-01
-6.16623282e-01 6.67434394e-01 -4.67062481e-02 -8.11799526e-01
-1.45676279e+00 -7.34887779e-01 -1.95499197e-01 7.62454629e-01
1.42707914e-01 6.82899237e-01 -6.84146464e-01 -1.63145795e-01
7.67795593e-02 3.08939159e-01 -2.01666772e-01 9.79885831e-02
-8.00002515e-01 -7.77365804e-01 1.39585614e+00 7.74792910e-01
1.51777351e+00 -1.18712735e+00 -1.21109559e-04 1.43426552e-01
4.74247694e-01 -1.31060636e+00 -1.73332363e-01 -3.07674333e-02
-9.62230802e-01 -7.96216607e-01 -9.07422483e-01 -1.52921259e+00
2.65238017e-01 4.83217359e-01 6.05961442e-01 -1.20197587e-01
-3.35075974e-01 -1.98256269e-01 5.01005054e-02 -6.81632638e-01
-3.16646755e-01 -7.39965513e-02 -6.76039681e-02 1.54856844e-02
1.00539160e+00 -2.72728205e-01 -3.97430837e-01 3.84001791e-01
-6.54743135e-01 -1.77381471e-01 1.06639516e+00 3.35618734e-01
2.67318547e-01 2.59807348e-01 8.39042187e-01 -7.26723552e-01
7.05860257e-01 -5.34003615e-01 -9.05533433e-01 -1.27766486e-02
-6.47606850e-01 -5.47458589e-01 6.29550874e-01 1.91732839e-01
-8.99236262e-01 -8.06305930e-03 -5.84642529e-01 -3.44893724e-01
-6.83602691e-01 4.50034410e-01 5.23712374e-02 -3.63492459e-01
4.39062983e-01 6.78591192e-01 1.36461198e-01 -4.02328968e-01
-2.76033640e-01 1.48193300e+00 6.77844405e-01 1.97931156e-01
4.79875654e-01 4.77425218e-01 -9.70301703e-02 -1.08078241e+00
1.20148330e-03 -1.04132462e+00 -5.38971901e-01 -3.70566875e-01
1.12836039e+00 -9.50887561e-01 -7.46015370e-01 1.49142361e+00
-9.08613920e-01 4.05400060e-02 4.07522082e-01 5.52911639e-01
-1.76308230e-01 2.84163982e-01 -6.30933166e-01 -8.46794665e-01
-2.73452938e-01 -1.18270826e+00 6.51871324e-01 8.59235883e-01
5.72404027e-01 -9.61456656e-01 1.15819193e-01 8.39218318e-01
9.32822943e-01 -1.36237159e-01 6.98002875e-01 -8.39220762e-01
-7.67122030e-01 -7.27803171e-01 -6.78821981e-01 7.17719972e-01
1.14900852e-02 1.51553661e-01 -9.60649192e-01 3.25470954e-01
-3.80702406e-01 -3.89779806e-01 1.06566632e+00 6.57194555e-01
1.13278043e+00 -1.45172417e-01 -3.36129814e-01 4.80304450e-01
1.75924301e+00 8.31941485e-01 1.42483914e+00 8.12389135e-01
6.94829226e-01 3.70778106e-02 3.42348129e-01 1.01866782e-01
3.16189498e-01 5.07422805e-01 2.54435539e-01 -2.53600329e-01
4.80946191e-02 1.61482513e-01 2.69508183e-01 6.75750375e-01
-3.49498153e-01 -1.32669315e-01 -1.35837829e+00 4.99951214e-01
-1.50044870e+00 -1.05006838e+00 -4.07725334e-01 1.87560070e+00
3.52703780e-01 2.91924655e-01 2.00442016e-01 2.16905430e-01
8.82388592e-01 -6.48338273e-02 -2.84943581e-01 -6.65954888e-01
-1.51677608e-01 1.88482627e-01 1.26778567e+00 -4.35204059e-03
-1.42444658e+00 1.35425770e+00 5.92246532e+00 1.13684344e+00
-1.78242683e+00 8.24477449e-02 8.67160320e-01 7.81341493e-01
5.06898165e-01 -5.79612017e-01 -9.53642607e-01 5.03534496e-01
1.08678079e+00 3.43802512e-01 1.56528145e-01 1.13826740e+00
2.52877504e-01 -1.13146603e-01 -2.23091274e-01 1.18186498e+00
5.01841418e-02 -1.79663587e+00 -1.10174172e-01 1.24332920e-01
5.84599435e-01 6.33537233e-01 3.30961257e-01 7.71563113e-01
-1.91561446e-01 -1.37622344e+00 1.71854213e-01 3.07564884e-01
6.80306315e-01 -1.18391442e+00 1.35815132e+00 4.49849963e-01
-6.31446600e-01 2.35612597e-02 -8.90751779e-01 2.90224124e-02
-3.84108812e-01 1.79960415e-01 -1.13188612e+00 -8.40157643e-02
5.74007750e-01 6.35200441e-01 -4.68059361e-01 1.31769466e+00
6.95095956e-01 1.00443423e+00 -1.80893987e-01 -3.98889244e-01
1.01906800e+00 -2.36115858e-01 -5.53155094e-02 1.48451817e+00
1.63878486e-01 3.37269530e-03 -1.80856362e-02 7.69618034e-01
-2.38711268e-01 1.23710044e-01 -6.07945383e-01 -1.36896476e-01
2.42616028e-01 1.37526214e+00 -1.09495306e+00 -3.03347826e-01
-8.49253297e-01 6.09049857e-01 -4.71421272e-01 1.58579722e-01
-9.62928534e-01 -6.25254095e-01 3.14616799e-01 2.76384056e-01
4.73846853e-01 -4.75820839e-01 -5.12951851e-01 -7.94534147e-01
-3.71517301e-01 -4.79744703e-01 -8.87788236e-02 -7.40194798e-01
-1.01715720e+00 3.37556958e-01 -3.64474028e-01 -1.14847183e+00
3.19168180e-01 -1.18964064e+00 -1.04055071e+00 9.00741398e-01
-1.58942294e+00 -1.20514655e+00 -3.70526403e-01 9.66826737e-01
1.05427539e+00 -1.03119481e+00 3.78714710e-01 7.36604214e-01
-8.17427278e-01 4.13211107e-01 6.97558701e-01 9.49138820e-01
4.74362314e-01 -7.37142444e-01 4.96556580e-01 6.50256753e-01
-2.56056607e-01 2.14383304e-01 1.04642414e-01 -3.32051635e-01
-1.32501280e+00 -1.41448808e+00 4.81946439e-01 3.28181721e-02
2.01544464e-01 -1.19006261e-01 -6.26554489e-01 7.56460786e-01
3.41069520e-01 1.60560444e-01 5.36708176e-01 -4.00784761e-01
-5.16064614e-02 -5.74064612e-01 -1.55700684e+00 2.25408137e-01
-1.86014976e-02 -2.08274126e-01 -2.74605781e-01 3.75604331e-01
-2.07028791e-01 -1.97606042e-01 -1.86130151e-01 1.29312769e-01
5.07989407e-01 -8.92046511e-01 8.26207995e-01 -3.75290811e-01
3.26393157e-01 -3.58759582e-01 -2.43013740e-01 -6.36999309e-01
-3.50621015e-01 2.84829080e-01 6.77032769e-01 1.18525314e+00
3.48228723e-01 -6.69823468e-01 1.24397898e+00 4.39620256e-01
-5.43476880e-01 -6.61748588e-01 -8.51472974e-01 -7.53823400e-01
2.57522851e-01 -2.31613040e-01 2.57944345e-01 1.01060820e+00
-7.38410413e-01 4.59904894e-02 -4.98114258e-01 4.25087735e-02
3.21048617e-01 -3.62745762e-01 6.79407656e-01 -1.14794266e+00
3.15058470e-01 -4.73816156e-01 -1.20071650e+00 -7.25936234e-01
-1.93996653e-01 -5.96073568e-01 -3.26571316e-02 -1.71329761e+00
2.12371215e-01 -5.11626542e-01 -2.48900011e-01 4.90421444e-01
6.61135912e-01 6.30142689e-01 1.08840510e-01 6.34044707e-01
-6.02004528e-01 -5.20299859e-02 9.46580470e-01 -5.80778003e-01
-6.11001300e-03 2.58575588e-01 -3.39293063e-01 7.91711986e-01
1.19002032e+00 -2.39257604e-01 -3.84515263e-02 -4.48700488e-01
-3.63126636e-01 -1.12185769e-01 3.47796977e-01 -1.52925789e+00
4.74256039e-01 3.37463059e-02 7.38050580e-01 -1.03708684e+00
2.78338909e-01 -6.03612661e-01 -1.96851641e-01 3.74357551e-01
2.23966196e-01 -1.54408887e-01 5.46862900e-01 1.02019832e-01
-4.48345095e-01 -3.01179707e-01 1.20873082e+00 -3.71638238e-01
-1.71529400e+00 3.85229230e-01 -1.02603722e+00 -3.14217031e-01
1.32466543e+00 -9.75252271e-01 -3.37018609e-01 -1.28195331e-01
-2.20787838e-01 -1.12361059e-01 4.45352495e-02 5.02960205e-01
6.92560911e-01 -1.26809907e+00 -7.91087449e-01 1.68629974e-01
5.63184731e-02 -2.77145863e-01 3.56864393e-01 7.15253472e-01
-1.58513176e+00 1.16522229e+00 -1.07587230e+00 -6.21657610e-01
-1.09982622e+00 5.44877388e-02 5.92105865e-01 -3.78178656e-02
-2.63745457e-01 6.16102040e-01 -3.13501328e-01 -5.64394295e-01
-5.81822991e-02 -3.45830433e-02 -6.84243858e-01 -3.91651869e-01
3.70237112e-01 6.47249222e-01 4.37682360e-01 -8.96452963e-01
-5.99843919e-01 6.60179615e-01 -2.96539605e-01 1.40744045e-01
1.44816482e+00 6.83997869e-02 7.07780272e-02 1.83500081e-01
1.24625432e+00 -1.74822301e-01 -7.82381892e-01 2.52607852e-01
-1.13823615e-01 -2.98974127e-01 2.93047458e-01 -8.93951237e-01
-1.44297957e+00 1.15118670e+00 1.01836646e+00 2.52726097e-02
7.10623384e-01 -4.33612674e-01 9.07685816e-01 8.05473864e-01
4.38645303e-01 -1.59283745e+00 -2.66010880e-01 7.52829194e-01
4.39285398e-01 -1.47006071e+00 -5.12823701e-01 2.08945394e-01
-7.57913172e-01 1.52869439e+00 8.24852645e-01 -6.70866549e-01
7.80692101e-01 4.95610565e-01 2.62668103e-01 -2.96761077e-02
2.57236753e-02 -1.45959444e-02 -1.50873229e-01 7.37888694e-01
3.80688518e-01 1.32137895e-01 -1.67804092e-01 5.49281418e-01
3.08767885e-01 3.50341409e-01 7.81580329e-01 8.95785034e-01
-1.09177852e+00 -6.51839972e-01 -5.25254250e-01 6.64361656e-01
-7.58127570e-01 -1.84537724e-01 -2.63933599e-01 1.32033658e+00
2.34499380e-01 7.26243734e-01 2.80033737e-01 -3.18224758e-01
1.57548249e-01 2.06735730e-02 -7.91356433e-03 -2.71363199e-01
-2.49792039e-01 -3.91051948e-01 -3.39819118e-02 4.01481502e-02
-3.77106994e-01 -3.96018505e-01 -1.44350851e+00 -8.42655540e-01
-3.28693062e-01 -2.35715546e-02 1.15557194e+00 1.10360718e+00
1.22847885e-01 1.70381755e-01 5.38468122e-01 -4.71163392e-01
-1.83347464e-01 -1.08552909e+00 -1.13110411e+00 -1.15083106e-01
-7.43566006e-02 -3.64313155e-01 1.97005466e-01 -3.28681059e-02] | [9.825091361999512, -4.957208156585693] |
b98aef9e-fe95-4cf5-a2f4-502d9ebdcb60 | learning-intrinsic-image-decomposition-from | 1804.00582 | null | http://arxiv.org/abs/1804.00582v1 | http://arxiv.org/pdf/1804.00582v1.pdf | Learning Intrinsic Image Decomposition from Watching the World | Single-view intrinsic image decomposition is a highly ill-posed problem, and
so a promising approach is to learn from large amounts of data. However, it is
difficult to collect ground truth training data at scale for intrinsic images.
In this paper, we explore a different approach to learning intrinsic images:
observing image sequences over time depicting the same scene under changing
illumination, and learning single-view decompositions that are consistent with
these changes. This approach allows us to learn without ground truth
decompositions, and to instead exploit information available from multiple
images when training. Our trained model can then be applied at test time to
single views. We describe a new learning framework based on this idea,
including new loss functions that can be efficiently evaluated over entire
sequences. While prior learning-based methods achieve good performance on
specific benchmarks, we show that our approach generalizes well to several
diverse datasets, including MIT intrinsic images, Intrinsic Images in the Wild
and Shading Annotations in the Wild. | ['Zhengqi Li', 'Noah Snavely'] | 2018-04-02 | learning-intrinsic-image-decomposition-from-1 | http://openaccess.thecvf.com/content_cvpr_2018/html/Li_Learning_Intrinsic_Image_CVPR_2018_paper.html | http://openaccess.thecvf.com/content_cvpr_2018/papers/Li_Learning_Intrinsic_Image_CVPR_2018_paper.pdf | cvpr-2018-6 | ['intrinsic-image-decomposition'] | ['computer-vision'] | [ 3.94685835e-01 -7.78855756e-02 8.91874880e-02 -3.29743385e-01
-9.98206079e-01 -7.65535712e-01 3.10248017e-01 -2.75134206e-01
-3.45213979e-01 5.94141006e-01 2.04576477e-01 2.39680514e-01
1.56545565e-01 -5.54647923e-01 -9.11121666e-01 -9.22759712e-01
2.27049544e-01 4.55458254e-01 3.07410836e-01 -6.46224320e-02
1.55806616e-01 3.21650088e-01 -1.45798075e+00 4.97738779e-01
3.67633462e-01 5.75191557e-01 3.77612174e-01 9.03937399e-01
3.05281132e-01 8.03936422e-01 -2.46282682e-01 -1.47088856e-01
4.64787424e-01 -3.34959686e-01 -9.22348499e-01 6.42180979e-01
6.27233088e-01 -6.04312360e-01 -2.76900619e-01 9.19061482e-01
2.91658193e-01 4.51382762e-03 4.96140540e-01 -1.01849544e+00
-2.83788204e-01 8.48015845e-02 -6.70391619e-01 2.62289289e-02
5.80048382e-01 2.58850724e-01 1.14671242e+00 -9.45195615e-01
8.88292909e-01 1.11330891e+00 5.69393218e-01 4.58829969e-01
-1.65576518e+00 -1.10683069e-01 2.96769261e-01 7.31609836e-02
-9.12027717e-01 -6.15367830e-01 7.96884060e-01 -6.49673760e-01
7.13965178e-01 6.08809218e-02 4.78390485e-01 1.28501558e+00
-2.13156268e-02 1.04275823e+00 1.34375060e+00 -4.49022114e-01
1.09471485e-01 3.05772461e-02 5.39810434e-02 7.13139474e-01
1.45044655e-01 5.21647856e-02 -6.94388449e-01 -8.27957690e-03
7.55966187e-01 2.01966256e-01 -6.48835897e-01 -9.05086696e-01
-1.54329872e+00 5.87211728e-01 1.67728260e-01 6.20568693e-02
-2.90107459e-01 1.62457183e-01 2.80781657e-01 3.59640688e-01
4.52326834e-01 2.58002043e-01 -6.57007456e-01 -1.27844304e-01
-8.25222611e-01 3.12040657e-01 7.66399503e-01 5.75568914e-01
1.13912976e+00 -1.13042235e-01 3.35556477e-01 6.12563193e-01
6.76900372e-02 4.40950543e-01 2.46468693e-01 -1.53946805e+00
4.70810026e-01 3.45637739e-01 1.96063921e-01 -8.29971373e-01
-2.66224861e-01 -3.58860463e-01 -7.63243675e-01 6.20519757e-01
7.42828369e-01 2.56102849e-02 -9.08772111e-01 1.99385309e+00
1.37583867e-01 7.57033974e-02 1.75124019e-01 8.85012686e-01
3.88531864e-01 4.58995223e-01 -7.66301572e-01 -1.59200191e-01
9.59233880e-01 -9.95204806e-01 -3.98562253e-01 -2.54097521e-01
5.29830515e-01 -6.44766688e-01 1.31284750e+00 8.74355018e-01
-1.09663725e+00 -5.89354217e-01 -1.00771213e+00 -2.72192538e-01
-4.49926145e-02 1.29271492e-01 3.26385587e-01 4.91436958e-01
-1.10545039e+00 6.01439357e-01 -1.10664475e+00 -6.01828098e-01
1.83672562e-01 2.22570896e-01 -6.41946614e-01 -5.20036340e-01
-5.35831451e-01 5.69942534e-01 1.86454162e-01 -1.02415204e-01
-1.40044534e+00 -5.83919466e-01 -8.76663268e-01 -1.26347989e-01
5.74369550e-01 -8.64739895e-01 1.02889335e+00 -1.08617806e+00
-1.39940619e+00 1.10459661e+00 -2.63511807e-01 -2.14355797e-01
5.84041059e-01 -2.62359649e-01 1.54993236e-01 1.91075668e-01
7.79480115e-02 5.12848198e-01 9.92147863e-01 -1.83071840e+00
-3.78839761e-01 -6.29201651e-01 5.02658784e-01 2.23400161e-01
-1.84774384e-01 -3.09628487e-01 -3.89223039e-01 -4.21102911e-01
3.64375085e-01 -1.12123823e+00 -2.33068526e-01 9.65043232e-02
-2.37253949e-01 4.77934778e-01 7.67973661e-01 -6.86254025e-01
5.43800235e-01 -2.06835365e+00 6.49582863e-01 -1.30471915e-01
1.59946606e-01 -4.94064987e-02 -3.35048705e-01 4.15743351e-01
-2.70906556e-02 -1.77178457e-01 -4.73322690e-01 -5.68732560e-01
-2.52009332e-01 6.44898534e-01 -3.44198644e-01 4.28774327e-01
-7.14115128e-02 7.07212687e-01 -1.09114659e+00 -1.82998821e-01
2.17201248e-01 3.98902893e-01 -8.51021230e-01 3.56073201e-01
-2.10574940e-01 9.10851717e-01 -1.84245437e-01 4.08897668e-01
5.52816272e-01 -5.00268877e-01 1.68478072e-01 -3.39544147e-01
1.31592795e-01 1.19505534e-02 -1.22429156e+00 2.16869521e+00
-6.70453012e-01 5.92468441e-01 3.34503949e-01 -1.37464380e+00
5.12532711e-01 2.83383787e-01 7.15917230e-01 -3.05842400e-01
-2.30349734e-01 1.45915478e-01 -2.62259841e-01 -6.83537662e-01
1.20443344e-01 -1.91118404e-01 1.69729397e-01 6.56708717e-01
3.42617154e-01 -1.99879944e-01 2.66880065e-01 2.83261746e-01
1.07330775e+00 4.97676879e-01 1.59457862e-01 -4.48863395e-02
6.61098838e-01 -2.17585370e-01 6.50409997e-01 5.86036623e-01
1.61953285e-01 1.06486201e+00 4.94985044e-01 -7.90683210e-01
-1.10411310e+00 -1.19697857e+00 6.17783256e-02 7.44762123e-01
8.17729831e-02 -4.99800920e-01 -6.27553403e-01 -7.03580558e-01
-3.41116250e-01 2.16702327e-01 -6.69386864e-01 1.75406352e-01
-4.98720646e-01 -5.24508357e-01 7.32846484e-02 4.47835892e-01
4.71641690e-01 -6.67119622e-01 -8.32768381e-01 -3.08821537e-02
-5.46984911e-01 -1.43942392e+00 -3.47339660e-01 2.67431408e-01
-1.13483119e+00 -1.10975313e+00 -7.60857165e-01 -4.44581509e-01
7.87265122e-01 6.01823449e-01 1.41581476e+00 -7.11243451e-02
-3.37830842e-01 8.42747152e-01 -2.12029919e-01 5.28764613e-02
-3.34698558e-01 -9.65075269e-02 9.01575258e-04 3.63869041e-01
-2.50804245e-01 -9.53704298e-01 -5.57290852e-01 3.43092650e-01
-1.02581060e+00 1.21605866e-01 4.06683922e-01 1.21635771e+00
7.76053369e-01 -8.55494067e-02 4.95064631e-02 -1.04146993e+00
-1.03660829e-01 -2.73365695e-02 -6.28532231e-01 3.29226494e-01
-3.49401385e-01 1.67272881e-01 7.36502409e-01 -5.98123111e-02
-1.06205094e+00 4.41246510e-01 5.81738465e-02 -7.96863496e-01
-3.23363811e-01 3.53350252e-01 -3.98974895e-01 -1.44629553e-01
5.93189836e-01 3.22470248e-01 8.84509310e-02 -6.66557372e-01
5.11738718e-01 8.49502534e-02 6.66988909e-01 -7.71754920e-01
7.99656212e-01 1.07449353e+00 1.57688223e-02 -9.12708402e-01
-1.14925373e+00 -5.59129775e-01 -1.05648720e+00 -1.17394321e-01
8.73433530e-01 -1.02034509e+00 -4.52333480e-01 3.63902837e-01
-1.00775504e+00 -5.20978570e-01 -4.48036671e-01 4.67044979e-01
-1.02526712e+00 6.97960079e-01 -4.15170342e-01 -4.55360055e-01
1.65815726e-01 -1.21772456e+00 1.35815823e+00 -2.15501904e-01
1.48689196e-01 -1.15720356e+00 3.59435409e-01 6.62986934e-01
-3.65112722e-02 3.88562322e-01 6.14637375e-01 -5.65145575e-02
-9.01603222e-01 1.64726842e-02 -7.76320398e-02 7.71896839e-01
3.29902798e-01 -1.04344591e-01 -1.15983653e+00 -6.52775884e-01
4.66401488e-01 -7.58996725e-01 1.13625324e+00 4.06279325e-01
1.24099648e+00 -1.66563451e-01 -1.21926084e-01 9.61885571e-01
1.67742932e+00 -2.87489593e-01 6.57557905e-01 1.76603019e-01
6.78459644e-01 6.87833786e-01 4.68797505e-01 2.94614911e-01
4.29977775e-01 8.50354612e-01 5.99535823e-01 -1.60765067e-01
-7.93642774e-02 -1.59714937e-01 4.68921542e-01 8.72822583e-01
-3.37745190e-01 -1.66377112e-01 -7.83789098e-01 6.23184741e-01
-1.83923936e+00 -1.16488111e+00 7.39315972e-02 2.37063694e+00
5.00338435e-01 5.15043400e-02 6.65859804e-02 1.76452994e-01
2.21132472e-01 3.64057243e-01 -5.82107246e-01 8.97293016e-02
-2.60941774e-01 -1.36875063e-02 3.24289739e-01 6.72726810e-01
-1.04430115e+00 5.57287633e-01 6.68421268e+00 2.57691264e-01
-1.13101661e+00 2.66120344e-01 7.71828592e-01 -8.44433159e-02
-4.06796932e-01 4.45525169e-01 -4.19225276e-01 9.84373689e-02
4.68033552e-01 1.44760519e-01 7.43543148e-01 5.90039372e-01
1.46636739e-01 -2.48673275e-01 -1.41553164e+00 1.20417869e+00
3.37645024e-01 -1.26200426e+00 -1.40249521e-01 1.51132435e-01
1.06890500e+00 2.93279856e-01 5.43875135e-02 -1.29718274e-01
3.33351165e-01 -7.23651528e-01 4.41885024e-01 4.35449630e-01
5.50582349e-01 -2.67057657e-01 4.34554279e-01 5.56509852e-01
-1.01621354e+00 -9.20635648e-03 -4.06604618e-01 -1.79506600e-01
2.78359592e-01 6.41667366e-01 -4.14305329e-01 7.71899343e-01
6.68871164e-01 1.20853829e+00 -6.22812569e-01 6.08624041e-01
-2.10878909e-01 4.58185136e-01 -3.01717788e-01 7.88187146e-01
-7.19603226e-02 -4.34969634e-01 5.41106761e-01 7.99086571e-01
2.66133606e-01 -5.65268332e-03 3.84343266e-01 7.97386348e-01
1.17811389e-01 -2.45844424e-01 -9.40216064e-01 2.87660539e-01
-3.21453243e-01 1.41269112e+00 -6.30381227e-01 -2.62071192e-01
-8.81585479e-01 1.26903141e+00 4.60853040e-01 6.96862698e-01
-5.00266850e-01 2.76012391e-01 4.90904421e-01 1.33873209e-01
4.26174313e-01 -5.87587595e-01 -1.55659877e-02 -1.92684877e+00
1.90192804e-01 -9.77476597e-01 3.79731804e-01 -9.39990461e-01
-1.15699732e+00 3.53607714e-01 3.59718762e-02 -1.41155601e+00
-3.83520603e-01 -7.36762345e-01 -4.35449123e-01 5.33038735e-01
-1.61617792e+00 -1.14470410e+00 -5.42713642e-01 7.91988075e-01
7.16196358e-01 8.03997219e-02 6.28313124e-01 1.28038272e-01
-3.36693913e-01 1.99292913e-01 3.25385660e-01 5.27924858e-02
7.31147826e-01 -1.39710069e+00 1.36574179e-01 9.58643496e-01
6.92846417e-01 4.21038061e-01 5.90811491e-01 -4.70569842e-02
-1.61773384e+00 -8.04075956e-01 2.76152432e-01 -5.62392116e-01
4.55644399e-01 -4.45498109e-01 -9.60566700e-01 9.04622018e-01
3.18894684e-01 4.32233334e-01 4.62653577e-01 1.18198749e-02
-5.11488974e-01 -2.48722553e-01 -7.85870254e-01 3.50947976e-01
1.16427696e+00 -7.05874205e-01 -1.89975500e-01 4.94140238e-01
4.47277308e-01 -4.01993811e-01 -7.75465846e-01 3.17036867e-01
5.17986357e-01 -1.38688457e+00 1.12547159e+00 -5.34174442e-01
5.69268465e-01 -4.23247099e-01 -4.71933395e-01 -1.35167241e+00
-6.32358110e-03 -5.64037263e-01 -6.42897040e-02 9.43622947e-01
2.27600217e-01 -6.05142832e-01 8.25538158e-01 3.51497501e-01
3.23156491e-02 -5.03096104e-01 -6.95725977e-01 -1.00053716e+00
-7.02272952e-02 -1.68607771e-01 4.82507683e-02 7.96887755e-01
-3.37835371e-01 4.41176891e-01 -5.34935117e-01 2.56405711e-01
1.09130549e+00 4.67580885e-01 1.24307179e+00 -1.06052506e+00
-8.11483026e-01 7.90472850e-02 -4.37555134e-01 -1.18466616e+00
2.27865607e-01 -7.70702541e-01 3.76842991e-02 -1.32629633e+00
5.19957602e-01 -8.01897123e-02 -1.28441602e-01 3.18068326e-01
-1.33202031e-01 3.88924181e-01 1.82460621e-01 3.57420117e-01
-5.08220255e-01 5.08436680e-01 1.31761158e+00 -1.86374828e-01
1.48025200e-01 -1.44938812e-01 -2.88595468e-01 8.66128504e-01
5.01339555e-01 -3.47057849e-01 -7.22955108e-01 -8.61078024e-01
3.01086843e-01 2.09868580e-01 6.73148334e-01 -9.38228011e-01
1.42335892e-01 -1.71083927e-01 3.68240803e-01 -5.06277978e-01
4.80992675e-01 -8.79950285e-01 2.80164480e-01 3.34944963e-01
-2.38776654e-01 2.53671966e-02 -1.19796179e-01 8.13672781e-01
-4.48555082e-01 -3.07630569e-01 8.52777779e-01 -5.43920755e-01
-5.50224543e-01 4.08933252e-01 4.98341583e-02 3.29936445e-01
6.87144697e-01 -2.36551389e-01 -1.35606565e-02 -5.58210254e-01
-9.99229014e-01 2.96451058e-02 1.11431682e+00 1.90351263e-01
6.19813204e-01 -1.13640678e+00 -6.98555052e-01 3.22435766e-01
2.18236864e-01 1.49453431e-01 2.08894953e-01 8.94569337e-01
-4.87805009e-01 1.33549601e-01 -3.26887727e-01 -1.11368883e+00
-1.21086156e+00 7.31394827e-01 3.95950466e-01 -4.62406129e-01
-7.78245211e-01 5.81006527e-01 7.91039526e-01 -5.22024274e-01
5.43880463e-02 -3.22513729e-01 1.69334218e-01 -1.32190540e-01
4.54554766e-01 4.73732837e-02 1.78287193e-01 -6.39940023e-01
-2.91088410e-02 9.71268475e-01 3.83759253e-02 -4.44024473e-01
1.54582071e+00 -4.31898832e-01 -1.45154998e-01 7.60949373e-01
1.50561678e+00 -6.48418292e-02 -1.72405112e+00 -4.02950317e-01
-1.56192094e-01 -9.47654963e-01 -8.97677913e-02 -3.84189606e-01
-1.32189167e+00 1.23834753e+00 5.05848944e-01 -9.51511133e-03
1.41281343e+00 -1.33174285e-01 5.55589020e-01 4.36123133e-01
5.61152458e-01 -7.86916435e-01 6.24632835e-01 3.83064628e-01
8.55615914e-01 -1.44954503e+00 8.66017491e-02 -2.63732880e-01
-4.12999958e-01 1.23161507e+00 3.93900305e-01 -2.52651453e-01
5.19657671e-01 3.57352734e-01 8.11953843e-03 -2.36827850e-01
-1.00263298e+00 -1.87184736e-01 -1.64593030e-02 6.30463243e-01
3.00706416e-01 -2.98131347e-01 2.16764003e-01 -7.94234797e-02
2.28074342e-01 -5.97474584e-03 8.49365950e-01 8.72299135e-01
-1.22274913e-01 -1.44498336e+00 -2.99268335e-01 2.64721751e-01
-3.22145998e-01 1.81033969e-01 -3.08667362e-01 5.06613314e-01
-9.17509049e-02 6.37579739e-01 -3.36554646e-01 -1.77161545e-01
1.71838984e-01 -2.03533527e-02 8.82895947e-01 -7.88108587e-01
-3.19590010e-02 1.65539846e-01 -1.63597181e-01 -9.21725154e-01
-1.02844131e+00 -1.04377615e+00 -7.27012098e-01 4.32661772e-02
-1.47810057e-01 -2.13781103e-01 4.34750050e-01 9.11539197e-01
2.89728623e-02 3.49759758e-01 8.89128089e-01 -1.21028340e+00
-4.33945179e-01 -4.62883025e-01 -6.00416183e-01 8.26881647e-01
7.51236141e-01 -5.19778550e-01 -5.18701673e-01 7.24529564e-01] | [9.22297191619873, -2.6729371547698975] |
08e7d76a-9e19-4e65-9549-8f4fb7928ec6 | tvsum-summarizing-web-videos-using-titles | null | null | http://openaccess.thecvf.com/content_cvpr_2015/html/Song_TVSum_Summarizing_Web_2015_CVPR_paper.html | http://openaccess.thecvf.com/content_cvpr_2015/papers/Song_TVSum_Summarizing_Web_2015_CVPR_paper.pdf | TVSum: Summarizing Web Videos Using Titles | Video summarization is a challenging problem in part because knowing which part of a video is important requires prior knowledge about its main topic. We present TVSum, an unsupervised video summarization framework that uses title-based image search results to find visually important shots. We observe that a video title is often carefully chosen to be maximally descriptive of its main topic, and hence images related to the title can serve as a proxy for important visual concepts of the main topic. However, because titles are free-formed, unconstrained, and often written ambiguously, images searched using the title can contain noise (images irrelevant to video content) and variance (images of different topics). To deal with this challenge, we developed a novel co-archetypal analysis technique that learns canonical visual concepts shared between video and images, but not in either alone, by finding a joint-factorial representation of two data sets. We introduce a new benchmark dataset, TVSum50, that contains 50 videos and their shot-level importance scores annotated via crowdsourcing. Experimental results on two datasets, SumMe and TVSum50, suggest our approach produces superior quality summaries compared to several recently proposed approaches. | ['Jordi Vallmitjana', 'Amanda Stent', 'Alejandro Jaimes', 'Yale Song'] | 2015-06-01 | null | null | null | cvpr-2015-6 | ['unsupervised-video-summarization'] | ['computer-vision'] | [ 4.34132725e-01 -3.62800658e-02 -2.49727696e-01 3.61835919e-02
-1.17068875e+00 -6.92444265e-01 4.29733962e-01 3.19682837e-01
-3.00906032e-01 7.26403058e-01 9.71136630e-01 4.70368266e-01
3.04914918e-02 -2.96745241e-01 -1.03783834e+00 -7.86714852e-01
-1.21107325e-01 4.89798710e-02 4.01780605e-01 1.52637407e-01
4.83795166e-01 -1.09658934e-01 -1.77522230e+00 5.17914832e-01
8.38978112e-01 9.41017151e-01 5.12844622e-01 8.15130770e-01
-1.80316985e-01 1.04919636e+00 -1.07694650e+00 -3.96933228e-01
1.75332189e-01 -8.58513057e-01 -7.06406534e-01 6.65400565e-01
1.09355414e+00 -4.08623844e-01 -5.33512175e-01 1.15764427e+00
3.62652600e-01 5.78556657e-01 7.72547066e-01 -1.37739623e+00
-6.71707928e-01 5.65505743e-01 -8.19732785e-01 4.43768203e-01
6.97911263e-01 1.36193514e-01 1.33543348e+00 -8.59652042e-01
1.20136690e+00 1.26015460e+00 2.89062202e-01 3.09479535e-01
-1.23164618e+00 -3.61707687e-01 4.34731573e-01 4.16342080e-01
-1.40753925e+00 -5.84538341e-01 8.00300717e-01 -5.62867701e-01
5.47453046e-01 5.39659441e-01 8.10472131e-01 1.09633672e+00
1.51809290e-01 1.18213928e+00 6.10238135e-01 -1.80133522e-01
5.19668758e-01 -4.64622304e-02 1.31847203e-01 4.88687932e-01
3.85133117e-01 -6.89982772e-01 -8.80511045e-01 -3.29992443e-01
2.99467117e-01 1.25206172e-01 -7.23107457e-01 -6.63391173e-01
-1.37982666e+00 6.43177152e-01 1.25826284e-01 1.29106075e-01
-6.93244636e-01 1.43150121e-01 5.17680645e-01 1.77181870e-01
3.63556653e-01 5.19918084e-01 -4.66295145e-02 -2.13085517e-01
-1.31813991e+00 5.28764665e-01 7.71596611e-01 1.14746201e+00
7.53592134e-01 5.44649027e-02 -7.80783951e-01 7.13787258e-01
-2.33808160e-01 4.17245328e-01 4.92547095e-01 -1.36362100e+00
4.72784817e-01 4.98912424e-01 1.07519396e-01 -1.23231709e+00
1.96408898e-01 1.30465976e-03 -7.37166941e-01 -2.23655358e-01
8.41701329e-02 1.38040513e-01 -8.71401727e-01 1.62291443e+00
1.05496250e-01 7.41520897e-02 1.39744580e-01 1.02594697e+00
1.36427057e+00 9.85041976e-01 5.16158249e-03 -7.40029573e-01
1.50900936e+00 -1.08240283e+00 -1.00989592e+00 -5.03615290e-02
-5.19608818e-02 -6.49321139e-01 1.02634561e+00 4.10767078e-01
-1.27712035e+00 -4.05847162e-01 -9.77540493e-01 -1.55994266e-01
-1.31726965e-01 -3.86308506e-02 8.36123750e-02 1.66055456e-01
-1.12368131e+00 3.86906147e-01 -1.88010707e-01 -5.86640418e-01
6.10468984e-01 -1.90078184e-01 -4.95547563e-01 -3.38584512e-01
-8.93781364e-01 3.25356245e-01 5.74866354e-01 -6.16925776e-01
-1.10669219e+00 -6.81737483e-01 -9.35250819e-01 2.53859669e-01
1.00950229e+00 -7.06995308e-01 1.14358950e+00 -1.13182306e+00
-8.27717364e-01 7.14221358e-01 -5.03785431e-01 -4.77811038e-01
3.56794268e-01 -2.98372179e-01 -1.26851514e-01 9.64370012e-01
4.72592175e-01 8.20143044e-01 1.30527580e+00 -1.37750435e+00
-7.56762385e-01 -2.02034518e-01 1.44717142e-01 4.38067794e-01
-4.56552267e-01 3.23994681e-02 -1.19057918e+00 -1.12224662e+00
-1.79805964e-01 -6.60007536e-01 -1.25511348e-01 -1.05254576e-01
-4.83006030e-01 -1.08207837e-01 8.08045924e-01 -9.00429010e-01
1.53271353e+00 -2.05582094e+00 5.08184671e-01 -9.39061120e-02
4.72435653e-01 -1.26547664e-01 -1.08331457e-01 5.33681571e-01
1.17462806e-01 1.44838557e-01 -3.00237417e-01 -1.27565593e-01
-2.35262126e-01 4.19511534e-02 -4.24211115e-01 2.66781688e-01
2.41385251e-02 8.33897233e-01 -1.03159773e+00 -9.94163513e-01
-2.16967799e-02 1.27150461e-01 -4.54400182e-01 2.34796077e-01
-3.06856692e-01 1.59236923e-01 -3.39933693e-01 7.83596575e-01
5.27844489e-01 -3.34323674e-01 -3.26241404e-02 -4.58431751e-01
2.13809416e-01 -4.70185310e-01 -1.12617898e+00 1.68411493e+00
3.13263535e-01 1.08578539e+00 -4.81825396e-02 -8.75045776e-01
4.17018950e-01 1.73751816e-01 7.22470164e-01 -5.12563825e-01
-7.27458820e-02 -2.63368249e-01 -6.08346343e-01 -8.62900853e-01
1.09514868e+00 3.05193782e-01 -1.64328963e-01 3.09306979e-01
1.63850054e-01 -3.50730009e-02 7.73905456e-01 8.69950533e-01
1.10115552e+00 -1.49514750e-01 5.62618434e-01 -1.01815015e-01
2.86497593e-01 2.74915308e-01 6.26442969e-01 9.94113445e-01
-2.71258861e-01 1.14537621e+00 8.87786269e-01 -1.50211841e-01
-9.33029175e-01 -9.34137404e-01 4.42986965e-01 1.23705304e+00
5.22320271e-01 -9.57111478e-01 -8.80350649e-01 -6.22088969e-01
-1.30736634e-01 5.26539564e-01 -6.88639164e-01 5.09417504e-02
-4.54895467e-01 -2.64496148e-01 1.44002855e-01 3.75861138e-01
2.96684265e-01 -8.84518743e-01 -8.06517780e-01 1.11621823e-02
-7.31595814e-01 -1.26773155e+00 -1.07727146e+00 -2.97241002e-01
-5.48644364e-01 -1.43186498e+00 -1.45916438e+00 -6.83541954e-01
7.62170494e-01 9.52871442e-01 1.36736822e+00 -1.01986594e-01
-3.06417167e-01 9.48433518e-01 -5.90420008e-01 -9.89819393e-02
-2.74918914e-01 -3.36823225e-01 -8.90775546e-02 2.07580924e-01
1.09389581e-01 -2.24723637e-01 -7.18485057e-01 9.34070274e-02
-1.23220801e+00 3.65386643e-02 5.21410167e-01 5.98595977e-01
7.97469854e-01 1.93728983e-01 2.04678431e-01 -6.44593298e-01
7.46616423e-01 -5.69588900e-01 -1.24062039e-01 5.06748736e-01
-2.38150451e-02 -3.15908119e-02 3.25732887e-01 -5.07878602e-01
-8.73835981e-01 1.34250401e-02 5.91967821e-01 -9.81275141e-01
2.44739335e-02 4.38135147e-01 -1.58558443e-01 3.62166166e-01
5.01673639e-01 6.43585384e-01 -8.13292488e-02 -3.62539440e-01
3.70555788e-01 3.59299064e-01 7.49905348e-01 -3.74536157e-01
4.15789455e-01 5.21648049e-01 -2.95783609e-01 -1.41167903e+00
-6.07668221e-01 -8.86381984e-01 -3.42191964e-01 -5.79909682e-01
9.32345450e-01 -1.23134398e+00 -3.50357383e-01 1.44675791e-01
-1.13180709e+00 1.90736830e-01 -4.53095257e-01 2.20192015e-01
-5.32169580e-01 9.81971443e-01 -2.26181090e-01 -5.48506439e-01
-3.62904400e-01 -1.20926058e+00 1.23964036e+00 3.44702125e-01
-4.46224660e-01 -4.32114303e-01 -1.38294324e-01 3.25979084e-01
5.07355726e-04 4.07854736e-01 6.47534668e-01 -6.53608799e-01
-7.75121272e-01 -1.06056966e-01 -1.96715474e-01 2.92364806e-01
-1.41761377e-02 1.91880077e-01 -6.76895618e-01 -3.70563507e-01
-2.17858881e-01 -3.57685804e-01 1.32667112e+00 6.85799062e-01
1.15986037e+00 -7.84549952e-01 -2.75571615e-01 1.59578323e-01
1.26759982e+00 6.75675422e-02 6.21797144e-01 1.20276473e-01
7.04175532e-01 6.60357058e-01 7.30374336e-01 6.70285642e-01
4.25977349e-01 7.44027853e-01 3.33583951e-01 2.28881419e-01
-2.11282119e-01 -2.55932420e-01 6.38442039e-01 6.40865445e-01
-1.41858652e-01 -5.69800019e-01 -4.63273317e-01 7.19490230e-01
-1.97387850e+00 -1.42653322e+00 -6.72625750e-02 2.09753537e+00
7.06152678e-01 -2.11092338e-01 3.95029426e-01 -1.23019814e-01
8.61921787e-01 5.17559528e-01 -4.94000852e-01 8.87765735e-02
-5.38760722e-01 -4.72652406e-01 3.43269616e-01 -1.39652006e-02
-1.30280399e+00 6.75105989e-01 6.40928936e+00 1.18917668e+00
-5.75919151e-01 7.83070643e-03 5.68406999e-01 -4.30918813e-01
-3.29780847e-01 -1.69407770e-01 -4.20884073e-01 6.34152055e-01
3.80916923e-01 -4.94075060e-01 2.43282970e-02 8.95302415e-01
2.81099856e-01 -6.53308868e-01 -1.04117525e+00 1.40086985e+00
7.32783318e-01 -1.55536044e+00 5.91848969e-01 -6.44944087e-02
1.16256249e+00 -3.99961084e-01 5.38305193e-02 1.40650585e-01
-4.45324481e-02 -5.32396615e-01 9.81082320e-01 5.54472446e-01
6.57413661e-01 -6.98788285e-01 4.54983741e-01 4.75457869e-02
-1.25569510e+00 6.12055324e-02 -4.91704524e-01 3.17542046e-01
6.87246621e-02 3.15279603e-01 -4.00959998e-01 4.73994762e-01
1.02345943e+00 9.62067306e-01 -7.87706435e-01 1.18141997e+00
2.20838655e-02 4.52685028e-01 1.14913560e-01 -3.14037129e-02
1.89606875e-01 4.87165060e-03 1.18428195e+00 1.36092544e+00
3.78467232e-01 3.34162295e-01 3.68650675e-01 5.95572650e-01
-3.41142863e-01 2.19193563e-01 -7.16257155e-01 -2.49511883e-01
2.52284080e-01 1.09657156e+00 -1.01133633e+00 -8.12294662e-01
-3.58540207e-01 1.32170463e+00 -1.80889860e-01 6.36012733e-01
-8.11782718e-01 -3.66710722e-01 6.79733098e-01 7.42689371e-02
6.94401741e-01 6.64414093e-02 2.65190184e-01 -1.52421689e+00
2.58014172e-01 -1.01352060e+00 6.27000570e-01 -1.08807433e+00
-9.10919785e-01 5.09781361e-01 3.17214191e-01 -1.61590338e+00
-1.96019724e-01 -7.31003806e-02 -5.60284257e-01 2.30803326e-01
-1.08172846e+00 -7.35356987e-01 -5.00344515e-01 5.98625064e-01
1.33073926e+00 -2.04200864e-01 2.88583070e-01 -1.65550217e-01
-4.12893683e-01 3.33594412e-01 3.74517083e-01 -9.43183750e-02
8.73856306e-01 -1.12838316e+00 -4.34291400e-02 1.05668509e+00
2.67636806e-01 3.19688827e-01 1.10308647e+00 -8.23683977e-01
-1.57517123e+00 -1.06670439e+00 5.35400510e-01 -2.86862075e-01
4.02320743e-01 -1.50400549e-01 -9.43739057e-01 3.10559452e-01
6.71042562e-01 -3.57885450e-01 6.57000005e-01 -3.65200430e-01
-2.22443074e-01 4.10925001e-02 -7.85572469e-01 7.63760567e-01
1.00167418e+00 -2.55151600e-01 -8.90572786e-01 4.09509927e-01
9.69214976e-01 -1.99185297e-01 -4.58637744e-01 -5.81107102e-03
4.50590402e-01 -9.01595950e-01 1.04623282e+00 -5.36230147e-01
8.54017615e-01 -2.97638535e-01 -3.16318214e-01 -1.40134525e+00
-8.94005522e-02 -7.43374586e-01 -4.38322425e-01 1.36083686e+00
-1.11474693e-01 2.93693900e-01 6.03914678e-01 4.70993876e-01
-4.62796837e-02 -2.92674243e-01 -7.33857214e-01 -6.91238344e-01
-6.46876991e-01 -1.84473842e-01 1.55982897e-01 6.54574335e-01
1.93266589e-02 4.24565673e-01 -7.60615885e-01 -1.32494539e-01
7.55894244e-01 2.54513770e-01 9.51780617e-01 -9.93825436e-01
1.04122989e-01 -5.82017660e-01 -5.82253158e-01 -9.88944411e-01
-6.57233372e-02 -4.68879968e-01 2.37685889e-01 -1.71909785e+00
8.26521635e-01 6.42296433e-01 -5.95926084e-02 1.66602775e-01
-4.67845678e-01 3.78829837e-01 4.66403693e-01 5.77970564e-01
-1.39428627e+00 5.74176311e-01 1.15372252e+00 -5.37605941e-01
-2.71602929e-01 -3.69277567e-01 -8.51046860e-01 7.59302616e-01
2.55817682e-01 -2.98161864e-01 -5.21275818e-01 -2.27525979e-01
1.93661705e-01 1.72191501e-01 3.42740238e-01 -1.03698647e+00
3.31157625e-01 -3.12018842e-01 4.90059406e-01 -9.61996019e-01
2.79799879e-01 -6.83704317e-01 1.44018590e-01 1.06672689e-01
-6.53684199e-01 -5.91521747e-02 6.52015656e-02 1.17845941e+00
-5.56928992e-01 -2.26829559e-01 4.35479045e-01 -3.94604564e-01
-1.27294242e+00 2.45016322e-01 -6.63974166e-01 4.45042044e-01
1.20080066e+00 -4.81530756e-01 -1.77530974e-01 -8.11805606e-01
-5.44244587e-01 4.11156774e-01 7.44435608e-01 5.10857701e-01
9.99018013e-01 -1.31647933e+00 -8.93631697e-01 -2.19772473e-01
5.45336008e-01 -1.81739062e-01 6.69902325e-01 6.31311655e-01
-3.97367209e-01 1.97126389e-01 -1.58093274e-01 -7.02872336e-01
-1.66092384e+00 9.29810345e-01 -4.71286029e-01 1.07181385e-01
-7.48928368e-01 7.06297398e-01 6.93256259e-01 6.43568754e-01
5.27326822e-01 -2.65564442e-01 -5.58271646e-01 7.32563138e-01
1.03383648e+00 4.54022527e-01 -3.70559752e-01 -1.06197846e+00
-3.07824731e-01 6.08179748e-01 -1.08815439e-01 3.72129306e-02
1.23352444e+00 -6.46702528e-01 -7.24703586e-03 4.20015752e-01
1.34307349e+00 -2.37852596e-02 -1.26898432e+00 -4.40296650e-01
-1.53458446e-01 -7.02095628e-01 -1.74854577e-01 -3.85351330e-01
-1.08324170e+00 4.59647834e-01 1.73325270e-01 2.44827375e-01
1.35618997e+00 3.13073903e-01 6.93679035e-01 5.58427513e-01
-5.68882562e-03 -1.24469852e+00 5.54727912e-01 1.05793484e-01
1.31128228e+00 -1.34268367e+00 4.13971871e-01 -2.48355448e-01
-1.33078015e+00 1.02845001e+00 4.21967745e-01 1.85639411e-02
8.16020593e-02 -3.10796291e-01 -4.21252757e-01 -3.14621955e-01
-7.64008045e-01 -3.64898890e-01 7.14593530e-01 4.81018394e-01
-7.98867643e-02 -1.86168402e-01 -1.37887388e-01 7.28554428e-01
2.68280685e-01 -1.98544770e-01 8.34435523e-01 1.04248118e+00
-7.67620683e-01 -2.31781140e-01 -7.33331978e-01 5.69510996e-01
-5.87612092e-01 -1.37654627e-02 -6.34892523e-01 5.37051678e-01
-1.97007447e-01 8.84773493e-01 1.40444964e-01 -3.17706794e-01
3.09358269e-01 -4.60743159e-02 1.89001963e-01 -4.47216690e-01
-2.58995980e-01 4.03369904e-01 -1.15177616e-01 -6.29679263e-01
-8.35725188e-01 -6.92802429e-01 -8.63044798e-01 -1.24341480e-01
1.57322977e-02 2.67864317e-01 2.15993360e-01 6.48477376e-01
4.15052950e-01 6.12197101e-01 4.98183310e-01 -1.09218538e+00
-1.00811698e-01 -7.15991378e-01 -5.50247669e-01 7.23519981e-01
5.01147509e-01 -5.77856660e-01 -3.05572510e-01 7.47978389e-01] | [10.434208869934082, 0.5209068655967712] |
0eb33340-420d-4418-9407-c0fe35e71d5b | relationship-quantification-of-image | 2212.04148 | null | https://arxiv.org/abs/2212.04148v2 | https://arxiv.org/pdf/2212.04148v2.pdf | Relationship Quantification of Image Degradations | In this paper, we study two challenging but less-touched problems in image restoration, namely, i) how to quantify the relationship between different image degradations and ii) how to improve the performance on a specific degradation using the quantified relationship. To tackle the first challenge, Degradation Relationship Index (DRI) is proposed to measure the degradation relationship, which is defined as the mean drop rate difference in the validation loss between two models, i.e., one is trained using the anchor degradation only and another is trained based on both the anchor and the auxiliary degradations. Through quantifying the relationship between different degradations using DRI, we empirically observe that i) the degradation combination proportion is crucial to the image restoration performance. In other words, the combinations with only appropriate degradation proportions could improve the performance of the anchor restoration; ii) a positive DRI always predicts the performance improvement of image restoration. Based on the observations, we propose an adaptive Degradation Proportion Determination strategy (DPD) which could improve the performance on the anchor degradation with the assist of another auxiliary degradation. Extensive experimental results verify the effective of our method by taking haze as the anchor degradation and noise, rain streak, and snow as the auxiliary degradations. The code will be released after acceptance. | ['Xi Peng', 'Peng Hu', 'Yuanbiao Gou', 'Boyun Li', 'Wenxin Wang'] | 2022-12-08 | null | null | null | null | ['image-dehazing'] | ['computer-vision'] | [ 2.18993410e-01 -5.53509057e-01 1.21131755e-01 -2.04336330e-01
-3.14649135e-01 -3.99168432e-02 2.58597255e-01 1.49162160e-02
-7.39543438e-02 6.17228389e-01 1.38346151e-01 -2.03921184e-01
-2.70043164e-01 -6.14777267e-01 -4.74334478e-01 -1.37199891e+00
2.25197613e-01 -2.82726794e-01 3.09541196e-01 -3.99952054e-01
1.66678816e-01 3.01579446e-01 -1.62395680e+00 -4.32569571e-02
1.53237975e+00 1.19196260e+00 4.96869981e-01 5.63820302e-01
1.53801724e-01 6.14750445e-01 -6.75043523e-01 3.19177075e-03
4.04084355e-01 -5.59385777e-01 -2.15109006e-01 3.19604576e-01
1.56370718e-02 -2.40604088e-01 -1.50620654e-01 1.25409293e+00
5.95048130e-01 2.70680971e-02 4.74439949e-01 -1.18375850e+00
-4.00789887e-01 9.17041600e-02 -7.53258884e-01 5.48167288e-01
-1.81607511e-02 4.09872651e-01 5.80746770e-01 -5.07705390e-01
2.28740618e-01 1.23479021e+00 4.18659568e-01 1.89206108e-01
-1.07427788e+00 -4.12611902e-01 2.04015121e-01 4.89918649e-01
-1.19500673e+00 -4.32241678e-01 8.13409030e-01 -4.80210453e-01
2.29141593e-01 1.60687864e-01 4.25775230e-01 5.42151988e-01
4.33934808e-01 4.79960263e-01 1.49161553e+00 -4.77975607e-01
8.20337161e-02 1.89626411e-01 2.08711401e-01 2.43788645e-01
1.60133570e-01 2.74678290e-01 -1.13182493e-01 1.19765289e-01
5.25094450e-01 -3.22607279e-01 -8.49401057e-01 -1.31968170e-01
-9.41895783e-01 4.86637056e-01 4.46431398e-01 1.76290452e-01
-3.89818490e-01 -3.20643008e-01 2.15371743e-01 5.26592612e-01
5.34567356e-01 1.44736201e-01 -3.07178855e-01 1.39355108e-01
-7.45691717e-01 -1.23086937e-01 4.06038284e-01 2.70256341e-01
7.85187960e-01 4.76195477e-02 -3.45055372e-01 9.67882931e-01
4.09764856e-01 7.22325385e-01 2.89479047e-01 -7.66953945e-01
5.05074501e-01 4.71696347e-01 4.21690643e-01 -9.87421751e-01
-5.44678457e-02 -7.10514307e-01 -9.63490605e-01 4.74784285e-01
4.39035863e-01 -3.93114947e-02 -8.46815765e-01 1.65717542e+00
4.06896830e-01 1.75925404e-01 1.75521851e-01 1.15416622e+00
6.33002996e-01 9.05022204e-01 1.82292126e-02 -7.50029087e-01
1.22470760e+00 -8.50591660e-01 -9.35799479e-01 -2.08036795e-01
2.43094519e-01 -8.45614076e-01 1.27650154e+00 3.53174925e-01
-8.74750912e-01 -7.27910221e-01 -1.29159844e+00 4.39029396e-01
1.70688048e-01 3.72463286e-01 -1.08092409e-02 6.00758553e-01
-1.01413429e+00 6.03217900e-01 -6.54162169e-01 -2.52622038e-01
-4.48133498e-02 -3.20144743e-02 1.48010915e-02 -4.27832484e-01
-1.24776506e+00 8.78236711e-01 9.74947140e-02 4.31098014e-01
-8.98768842e-01 -6.01842463e-01 -3.29878718e-01 -1.05387799e-01
1.52859122e-01 -6.05200112e-01 6.24248505e-01 -1.31929302e+00
-1.39021206e+00 4.99442190e-01 -6.54812157e-02 -2.90269703e-01
6.41592205e-01 -3.26058805e-01 -5.92248619e-01 1.90507114e-01
1.02656521e-02 1.16661809e-01 9.30846393e-01 -1.63413608e+00
-6.74441397e-01 -3.23304355e-01 1.16950087e-01 4.67768133e-01
-3.83252233e-01 -9.24155861e-02 -4.15271997e-01 -5.21657348e-01
2.34140173e-01 -7.90186882e-01 -4.09461297e-02 2.22561747e-01
-4.58233327e-01 1.46849975e-01 8.71702790e-01 -1.12309289e+00
1.28297222e+00 -2.48067427e+00 2.76327133e-01 1.98178351e-01
-6.61770552e-02 4.58046287e-01 -1.09447621e-01 1.73092052e-01
3.20691653e-02 3.48540209e-02 -4.78612781e-01 -2.92945020e-02
-3.02851737e-01 3.46882373e-01 -2.09112138e-01 5.22733092e-01
1.78892240e-01 1.14411637e-01 -8.39309096e-01 -4.08647090e-01
2.03447059e-01 4.64924932e-01 3.04854307e-02 5.60114563e-01
1.27304122e-01 7.70536363e-01 -2.09219798e-01 6.02977276e-01
9.69344258e-01 7.90675171e-03 8.52025449e-02 -6.45856440e-01
-1.71271890e-01 -6.12750314e-02 -1.31678963e+00 7.88762152e-01
-4.38482195e-01 5.32215893e-01 2.83120215e-01 -9.13212299e-01
1.14699578e+00 1.90612227e-01 1.19675331e-01 -8.11090291e-01
-3.78451906e-02 3.38673770e-01 1.19722724e-01 -7.23060966e-01
1.23830535e-01 -2.52601266e-01 5.03690779e-01 3.73086408e-02
-4.55316663e-01 1.36367261e-01 1.76364958e-01 -4.29754741e-02
9.06122386e-01 -5.40966727e-02 2.26772264e-01 -3.71790230e-01
8.32939565e-01 -3.79912615e-01 8.08997452e-01 5.11220813e-01
-4.63619798e-01 3.97759706e-01 5.03498673e-01 -1.51487604e-01
-9.06604111e-01 -1.03427982e+00 -2.53404975e-01 6.13269687e-01
8.06637108e-01 2.56431013e-01 -5.49186885e-01 -3.66244286e-01
-2.89700776e-01 5.99549174e-01 -3.85430753e-01 -2.97888845e-01
-5.03414989e-01 -1.13506925e+00 1.37877733e-01 9.63324681e-02
9.19629753e-01 -7.58603811e-01 -2.49969617e-01 -1.30777717e-01
-5.27698994e-01 -9.25665677e-01 -2.91615993e-01 -5.73365465e-02
-9.11191165e-01 -1.09555578e+00 -7.24200964e-01 -6.54444516e-01
7.92718470e-01 6.53576970e-01 8.18262756e-01 4.81716722e-01
2.45846897e-01 2.35224575e-01 -5.32395124e-01 3.26180877e-03
-6.43297255e-01 -4.49969143e-01 -4.86188248e-04 4.08548802e-01
-3.40939790e-01 -7.41991639e-01 -8.94519866e-01 6.94501281e-01
-1.00353622e+00 3.40294577e-02 5.99560261e-01 7.81436026e-01
5.71241736e-01 5.61105371e-01 4.60369676e-01 -3.57793719e-01
6.24734938e-01 -5.77063203e-01 -4.05978769e-01 3.73257548e-01
-8.40360343e-01 6.00814037e-02 5.94501138e-01 -5.75649917e-01
-1.10640812e+00 -3.92491311e-01 -4.42969659e-03 -3.71688724e-01
-9.02960151e-02 5.31986654e-01 -6.13524914e-01 -1.96455251e-02
5.02116740e-01 4.26472545e-01 -2.33194269e-02 -6.05292678e-01
8.36241171e-02 6.85200751e-01 6.76036417e-01 -4.99827772e-01
9.32943463e-01 4.91244853e-01 1.11859534e-02 -8.50729227e-01
-5.62157333e-01 -4.51186895e-01 -2.27571577e-01 -5.34629464e-01
7.36147881e-01 -8.51939321e-01 -3.29628021e-01 8.70272517e-01
-9.44057405e-01 -3.50159734e-01 1.74312256e-02 4.69442368e-01
-1.56463414e-01 8.21284771e-01 -4.06852812e-01 -1.03948522e+00
-3.46688241e-01 -1.21928442e+00 6.79394126e-01 4.70541567e-01
5.59411108e-01 -9.23999608e-01 -1.16170524e-02 3.14470798e-01
2.72911578e-01 4.77105409e-01 9.76262271e-01 4.33823802e-02
-4.63212818e-01 4.77891117e-02 -4.49689716e-01 8.64769697e-01
3.27335417e-01 1.60898909e-01 -6.30548656e-01 -5.00697196e-01
2.98890352e-01 3.03214580e-01 9.16133225e-01 4.85969514e-01
7.18995810e-01 -2.52887309e-01 -1.03457361e-01 5.82980216e-01
1.62226486e+00 4.59627181e-01 1.28145552e+00 7.29969144e-01
4.18816596e-01 5.13305664e-01 9.45255578e-01 3.10169995e-01
2.46367604e-01 7.91936934e-01 7.55320787e-01 -2.48385832e-01
-4.54732776e-01 1.29201546e-01 6.75083399e-01 9.07983720e-01
-1.43337905e-01 -5.45983672e-01 -6.58429921e-01 5.04527152e-01
-1.62665856e+00 -5.13353884e-01 -4.22752261e-01 2.36189246e+00
8.18113923e-01 2.61918426e-01 -2.59099543e-01 4.20263290e-01
1.02319479e+00 2.20850959e-01 -5.57923377e-01 -7.45910108e-02
-4.64778841e-01 -3.06889713e-01 3.57498765e-01 5.90289235e-01
-8.65550816e-01 3.58433425e-01 5.51950645e+00 8.52144003e-01
-1.16637897e+00 2.22307201e-02 7.14000106e-01 3.43211979e-01
-1.83757886e-01 2.08848879e-01 -5.57077527e-01 1.01081657e+00
6.66171908e-01 5.97835593e-02 4.33334589e-01 3.77011508e-01
7.34872103e-01 -4.33701158e-01 -5.16318798e-01 5.67270756e-01
-1.19325586e-01 -5.51859021e-01 8.12986642e-02 -1.18032746e-01
6.01038635e-01 -4.49201316e-01 1.27008399e-02 -1.32102698e-01
6.33351654e-02 -5.06161749e-01 6.36875331e-01 9.69461501e-01
4.67750430e-01 -3.97689253e-01 1.03391075e+00 3.83651346e-01
-1.09773040e+00 -2.64921784e-01 -2.14447081e-01 9.37068015e-02
2.33813047e-01 1.09992278e+00 -2.29851559e-01 8.82508457e-01
7.79366553e-01 6.45254552e-01 -5.55904210e-01 1.40970445e+00
-7.19611883e-01 8.48235250e-01 -7.59111717e-02 5.05143046e-01
-2.75266856e-01 -5.22271931e-01 9.12485540e-01 9.20449078e-01
4.71420258e-01 6.65896982e-02 1.39735073e-01 4.71910030e-01
3.96314234e-01 4.45317924e-02 2.90575605e-02 3.32996070e-01
4.59426492e-01 1.01468897e+00 -5.49386978e-01 -2.12014347e-01
-9.29366872e-02 8.09394360e-01 -1.37237579e-01 6.30845189e-01
-8.37071955e-01 -2.84714967e-01 8.09772134e-01 1.94187909e-01
3.37010026e-01 -1.98282018e-01 -3.01199317e-01 -8.74757946e-01
3.16349804e-01 -7.72341907e-01 1.53988138e-01 -1.05106020e+00
-1.27741992e+00 6.44972742e-01 -1.41811416e-01 -1.54530215e+00
4.07195836e-01 -1.95453167e-01 -7.69864440e-01 1.04924142e+00
-2.05677724e+00 -7.37973392e-01 -6.95378959e-01 4.32742178e-01
2.94966996e-01 1.81749299e-01 2.01635540e-01 5.29992759e-01
-1.01038861e+00 4.31909502e-01 3.55548650e-01 -3.01619023e-01
7.55252421e-01 -9.00643766e-01 -3.93098831e-01 1.35341549e+00
-4.57885563e-01 1.29609570e-01 1.26118338e+00 -6.25389278e-01
-8.84280920e-01 -8.47089827e-01 6.48108661e-01 1.53898001e-01
5.30716836e-01 2.65303165e-01 -1.12376297e+00 -8.50125477e-02
-1.52588356e-03 -7.95725361e-02 2.80170679e-01 -3.92123848e-01
-2.96438754e-01 -7.03882635e-01 -1.19673967e+00 4.76377934e-01
6.47147119e-01 -2.08526939e-01 -2.16950998e-01 3.14518034e-01
8.63418043e-01 -2.93673843e-01 -9.19624627e-01 6.20526075e-01
3.14340323e-01 -1.07951546e+00 1.00113881e+00 -7.00053945e-02
6.20047450e-01 -8.81513894e-01 -4.19435352e-01 -1.48187745e+00
-2.63945639e-01 -1.06498353e-01 -1.73669875e-01 1.46334481e+00
3.45147029e-02 -7.36134648e-01 -3.97308636e-03 1.83528647e-01
-2.49078631e-01 -8.83934796e-01 -8.36010039e-01 -9.99312282e-01
-1.78364411e-01 1.15891844e-01 4.66997534e-01 6.47705257e-01
-6.99916780e-01 6.25675078e-03 -6.31969571e-01 7.69264996e-01
6.16054237e-01 3.48305218e-02 7.08543003e-01 -1.04443514e+00
-3.23574930e-01 -9.28156376e-02 -4.15919900e-01 -1.05116510e+00
-2.96469122e-01 -2.40212888e-01 2.95184344e-01 -1.69215679e+00
1.89334080e-01 -5.63810527e-01 -5.22042692e-01 1.92767352e-01
-7.46580482e-01 1.07370935e-01 2.33443975e-01 7.45394707e-01
-2.40236610e-01 6.95212722e-01 1.43236470e+00 -2.99526781e-01
-2.68475920e-01 1.42645463e-01 -7.03523159e-01 3.26230377e-01
7.47345388e-01 -4.19189423e-01 -4.75190341e-01 -4.17357057e-01
4.50529158e-02 1.57183707e-01 5.04622400e-01 -1.22042608e+00
-5.60457222e-02 -3.58917683e-01 9.83161703e-02 -2.25875944e-01
2.27910593e-01 -7.33603776e-01 1.87781468e-01 5.17813623e-01
-5.76451514e-03 -2.11136460e-01 7.53628612e-02 7.86088586e-01
-2.64125764e-01 -1.02639958e-01 1.10737467e+00 1.31746799e-01
-7.17251241e-01 6.26032427e-03 -1.96990728e-01 -3.18981916e-01
7.83442199e-01 -3.58182937e-01 -6.04873955e-01 -5.58850706e-01
-6.81670189e-01 2.27440000e-01 5.50238788e-01 2.36774042e-01
4.28995311e-01 -1.16796827e+00 -9.98729706e-01 -6.73839524e-02
-3.50190029e-02 -5.40627122e-01 5.21533608e-01 1.19208324e+00
-2.68360734e-01 -2.44196773e-01 -1.82490170e-01 -7.17051804e-01
-1.46072662e+00 4.52196091e-01 5.40740848e-01 -5.91959476e-01
-4.81532395e-01 5.38134038e-01 2.80448526e-01 4.31920029e-02
3.90869081e-02 -6.69826269e-02 -5.13330400e-01 -1.37324005e-01
5.04129589e-01 5.42719722e-01 1.05059065e-01 -7.17941463e-01
-1.95466802e-01 5.77671826e-01 2.49352559e-01 1.27773747e-01
1.18807042e+00 -7.75368154e-01 -3.52996945e-01 5.25562286e-01
8.74249279e-01 -1.66127861e-01 -1.59474540e+00 -2.32050329e-01
-3.33679765e-01 -8.27532649e-01 2.95853108e-01 -9.91622865e-01
-1.37746406e+00 6.98798120e-01 1.20458508e+00 3.77374947e-01
1.85095203e+00 -3.60070705e-01 7.30009735e-01 -3.01281303e-01
1.54035702e-01 -7.22377598e-01 8.79167691e-02 7.59690478e-02
9.78349745e-01 -1.08775818e+00 1.88341856e-01 -5.58591127e-01
-5.27675688e-01 1.07798207e+00 5.35115004e-01 6.48550019e-02
6.35694087e-01 -1.29727021e-01 2.18578264e-01 -2.13397816e-02
-4.06476498e-01 -1.67657867e-01 2.26689488e-01 5.86622298e-01
2.95672305e-02 8.80536884e-02 -6.89361155e-01 3.27921629e-01
2.78293282e-01 -1.39853746e-01 4.30126637e-01 5.99212587e-01
-5.80250382e-01 -9.74078774e-01 -6.35757029e-01 4.07106429e-01
-3.50259505e-02 2.89237592e-02 1.73048880e-02 3.84151340e-01
3.47475916e-01 1.42609966e+00 -2.05563277e-01 -4.38219070e-01
3.89232874e-01 -3.83867115e-01 7.58744180e-02 -8.13204795e-02
-1.47465065e-01 -5.49005046e-02 -6.85805781e-03 -3.10597420e-01
-6.57054782e-01 -4.79535550e-01 -7.65694022e-01 -3.26843441e-01
-6.23327017e-01 2.15523526e-01 6.63721442e-01 1.11602545e+00
2.38177255e-01 7.19294250e-01 1.09475458e+00 -6.36823475e-01
-5.20039320e-01 -9.85514879e-01 -8.77482951e-01 3.11220437e-01
5.44694841e-01 -9.00763929e-01 -9.28002954e-01 9.57493708e-02] | [11.618261337280273, -2.0913193225860596] |
de6f5a64-c9d1-48fd-84e1-f2b8efb247b6 | on-defending-against-label-flipping-attacks | 1908.04473 | null | https://arxiv.org/abs/1908.04473v3 | https://arxiv.org/pdf/1908.04473v3.pdf | On Defending Against Label Flipping Attacks on Malware Detection Systems | Label manipulation attacks are a subclass of data poisoning attacks in adversarial machine learning used against different applications, such as malware detection. These types of attacks represent a serious threat to detection systems in environments having high noise rate or uncertainty, such as complex networks and Internet of Thing (IoT). Recent work in the literature has suggested using the $K$-Nearest Neighboring (KNN) algorithm to defend against such attacks. However, such an approach can suffer from low to wrong detection accuracy. In this paper, we design an architecture to tackle the Android malware detection problem in IoT systems. We develop an attack mechanism based on Silhouette clustering method, modified for mobile Android platforms. We proposed two Convolutional Neural Network (CNN)-type deep learning algorithms against this \emph{Silhouette Clustering-based Label Flipping Attack (SCLFA)}. We show the effectiveness of these two defense algorithms - \emph{Label-based Semi-supervised Defense (LSD)} and \emph{clustering-based Semi-supervised Defense (CSD)} - in correcting labels being attacked. We evaluate the performance of the proposed algorithms by varying the various machine learning parameters on three Android datasets: Drebin, Contagio, and Genome and three types of features: API, intent, and permission. Our evaluation shows that using random forest feature selection and varying ratios of features can result in an improvement of up to 19\% accuracy when compared with the state-of-the-art method in the literature. | ['Mohammad Shojafar', 'Reza Javidan', 'Rahim Taheri', 'Zahra Pooranian', 'Mauro Conti', 'Ali Miri'] | 2019-08-13 | null | null | null | null | ['android-malware-detection'] | ['miscellaneous'] | [ 4.13655102e-01 -2.12346658e-01 -1.61284506e-01 -2.15305865e-01
-5.63048482e-01 -9.79701936e-01 6.99528635e-01 1.55051306e-01
-3.62775326e-01 6.46489561e-01 -3.69323373e-01 -7.80159771e-01
-2.57186979e-01 -7.55304456e-01 -7.56303549e-01 -7.73590803e-01
-3.45786572e-01 2.57756561e-01 5.48825264e-01 -2.52592087e-01
5.56989014e-01 6.03723049e-01 -1.44847417e+00 4.61032569e-01
5.28768659e-01 1.25339568e+00 -6.26353204e-01 8.84975672e-01
2.89588511e-01 7.93013990e-01 -9.71773922e-01 -5.72134733e-01
6.06310010e-01 -1.71235763e-02 -7.37285078e-01 -6.51973128e-01
1.86232418e-01 -1.26809791e-01 -4.52201031e-02 1.36437666e+00
6.59416437e-01 -4.43028897e-01 7.65199959e-01 -1.69014728e+00
-4.52252805e-01 5.81744492e-01 -7.83461988e-01 4.50610757e-01
4.21053380e-01 1.10863492e-01 3.26118082e-01 -2.86739498e-01
1.97894782e-01 1.18762529e+00 1.12437463e+00 8.73597085e-01
-8.25110435e-01 -1.24552000e+00 -3.38465810e-01 1.49367779e-01
-1.29175997e+00 -2.41268471e-01 7.25761652e-01 -3.55236650e-01
8.04122806e-01 4.91216540e-01 5.67164235e-02 1.43252456e+00
5.62595427e-01 2.49585092e-01 1.43142235e+00 -2.64874607e-01
3.86499971e-01 2.39739820e-01 3.23577940e-01 7.19892085e-01
5.18513322e-01 4.77077335e-01 -4.83704284e-02 -6.64992511e-01
-1.21091321e-01 1.85510397e-01 1.91313133e-01 8.47244039e-02
-4.55233663e-01 1.14263570e+00 4.23808753e-01 3.40222418e-02
-1.23818114e-03 2.93835104e-01 8.28460276e-01 3.39064211e-01
3.14212859e-01 4.72223014e-01 -7.68646002e-01 -2.84422506e-02
-7.73700178e-01 -1.44294769e-01 8.92012715e-01 4.71496642e-01
2.89662510e-01 1.87578127e-01 1.03478238e-01 2.98488736e-01
5.14303923e-01 6.24475002e-01 5.70539057e-01 -6.84177399e-01
1.04865111e-01 5.49286366e-01 -1.52069733e-01 -1.24691653e+00
-5.64172447e-01 -2.84092724e-01 -7.84727573e-01 4.60654497e-01
3.41236889e-01 -5.09393632e-01 -1.02784085e+00 1.74069047e+00
5.02852917e-01 5.05886912e-01 6.85443804e-02 1.14136696e-01
5.17822742e-01 4.25584823e-01 2.30562404e-01 -2.43391693e-01
1.32826447e+00 -5.94642043e-01 -4.41534579e-01 1.02523200e-01
8.34707737e-01 -6.20242417e-01 8.22183490e-01 6.52796566e-01
-4.31588501e-01 -2.44786054e-01 -1.22212410e+00 4.82492208e-01
-1.00785530e+00 -2.98481107e-01 3.66099834e-01 1.70351446e+00
-7.80090928e-01 4.24900740e-01 -6.88165963e-01 -2.46800959e-01
8.32780063e-01 1.03419471e+00 -2.31658340e-01 2.59185731e-01
-1.18487835e+00 5.65050840e-01 9.60720107e-02 -3.56294572e-01
-1.41055131e+00 -3.28034490e-01 -4.70126003e-01 -2.66913533e-01
3.66942286e-01 4.26301174e-02 1.01437163e+00 -8.47985446e-01
-1.25841761e+00 7.19633102e-01 3.41726273e-01 -9.42920744e-01
2.16062501e-01 -9.75863859e-02 -7.44091868e-01 -1.40718161e-03
-6.61852509e-02 3.95775527e-01 1.23017514e+00 -1.40194416e+00
-4.61690277e-01 -5.35255134e-01 2.97625631e-01 -4.94835466e-01
-4.99807596e-01 2.82244295e-01 4.52100396e-01 -5.25344610e-01
-4.28249866e-01 -1.31827664e+00 -8.02095234e-02 -4.53543246e-01
-6.73945725e-01 -1.76421523e-01 1.79422784e+00 -6.09929800e-01
1.29146385e+00 -2.12724257e+00 -3.28945369e-01 4.22646046e-01
8.97329897e-02 9.70338941e-01 1.39546856e-01 1.93738699e-01
-1.21419348e-01 6.15383685e-01 -5.11245489e-01 -1.29381448e-01
-2.11839423e-01 1.51902065e-01 -4.49310541e-01 6.37425840e-01
-3.89750265e-02 5.61159015e-01 -7.87727416e-01 -2.49329284e-01
3.83804669e-03 5.43218434e-01 -4.60140377e-01 -8.91032144e-02
-4.28306848e-01 4.90558773e-01 -4.43183333e-01 8.65446508e-01
6.62685335e-01 1.96032673e-01 5.97652495e-02 8.70521665e-02
2.99902081e-01 7.08056018e-02 -9.23333168e-01 8.79966080e-01
-4.11814272e-01 2.97802895e-01 -6.05605692e-02 -7.81129360e-01
8.39338839e-01 1.66595936e-01 3.33503693e-01 -3.82754683e-01
7.01636434e-01 3.24466139e-01 3.26683670e-01 -3.84865373e-01
-5.95619529e-02 3.24041307e-01 -3.59438241e-01 6.94653511e-01
-3.11014652e-01 3.33137095e-01 -3.43734443e-01 1.64319530e-01
1.57641673e+00 -3.86865348e-01 3.05134058e-01 -2.33143836e-01
9.52631891e-01 -1.26958832e-01 3.27763677e-01 8.30127478e-01
-6.33069873e-01 -2.59215441e-02 7.26825476e-01 -5.25572002e-01
-7.17635036e-01 -7.85637856e-01 4.06651832e-02 1.10536587e+00
-1.47217950e-02 -2.42221966e-01 -1.25588357e+00 -1.59957600e+00
-1.23907469e-01 5.62420726e-01 -8.18073034e-01 -5.94861031e-01
-6.70475602e-01 -9.61376071e-01 1.50664461e+00 1.94508940e-01
9.84825253e-01 -1.18162155e+00 -6.36714220e-01 -3.51917863e-01
4.02785212e-01 -9.72509265e-01 -1.70927927e-01 4.99432772e-01
-3.15701783e-01 -1.55682850e+00 9.84692201e-02 -6.69107318e-01
3.88073266e-01 4.22243401e-02 5.98885179e-01 4.60202426e-01
-2.82225668e-01 3.15998971e-01 -5.69895148e-01 -6.93892181e-01
-8.21824193e-01 2.62486339e-01 4.21798944e-01 1.08485129e-02
5.56378663e-01 -5.33156991e-01 -4.25446510e-01 4.95721132e-01
-1.09841764e+00 -1.17957044e+00 4.12657797e-01 6.25163317e-01
2.50712037e-01 6.70228481e-01 7.89841354e-01 -1.37883711e+00
8.13736737e-01 -9.84632969e-01 -5.35672069e-01 3.67124341e-02
-8.48367035e-01 -1.09542593e-01 1.03679144e+00 -7.84146786e-01
-5.73361516e-01 2.69374609e-01 -5.38849950e-01 -3.18827271e-01
-3.54381710e-01 -1.07195422e-01 -4.40969586e-01 -6.66148424e-01
1.04833603e+00 -7.13705178e-03 -1.74617901e-01 -3.51943038e-02
1.95253536e-01 1.13579607e+00 2.64617532e-01 -2.88551569e-01
1.02099323e+00 6.43446386e-01 3.54056835e-01 -6.63541615e-01
-7.73109317e-01 -1.43028408e-01 -3.60738933e-01 6.95168898e-02
1.02815580e+00 -3.68144721e-01 -1.04042029e+00 8.48908365e-01
-8.15162599e-01 -2.87251145e-01 5.56681193e-02 -1.17913395e-01
-8.29156488e-02 5.88546276e-01 -5.01771212e-01 -8.59143674e-01
-5.57451189e-01 -1.48865151e+00 8.28201473e-01 1.82861909e-01
8.34193677e-02 -9.48302448e-01 1.04807224e-02 3.96353334e-01
4.49233502e-01 7.24745929e-01 9.30742979e-01 -1.50321186e+00
1.37092814e-01 -5.10490596e-01 -1.04097016e-02 5.43775439e-01
1.36707246e-01 1.21516772e-01 -1.14398789e+00 -4.26951081e-01
3.89311433e-01 -4.40152317e-01 6.51558518e-01 2.40124278e-02
1.44114745e+00 -6.63771331e-01 -5.12532592e-01 5.88818431e-01
1.17149329e+00 6.19691849e-01 7.09007502e-01 3.48140717e-01
8.49786997e-01 3.48731905e-01 4.75688487e-01 2.91851670e-01
5.61077856e-02 6.02862060e-01 1.05825198e+00 3.12694609e-01
9.74349603e-02 -6.00107852e-03 5.67150950e-01 8.64771605e-02
4.25832391e-01 -3.34871948e-01 -9.34422433e-01 1.52258575e-01
-1.43969297e+00 -9.71937060e-01 -3.04527044e-01 2.16361475e+00
6.88344359e-01 5.09829819e-01 4.12934870e-01 8.25246513e-01
7.54306495e-01 8.81182402e-02 -6.27104998e-01 -1.06108594e+00
1.43041670e-01 4.93391901e-01 8.63978267e-01 4.52831537e-01
-1.62052810e+00 9.45526004e-01 5.25289536e+00 1.16808724e+00
-1.23960829e+00 4.88721490e-01 6.26269639e-01 3.93866807e-01
2.64799237e-01 -1.83750644e-01 -9.17766631e-01 7.71163106e-01
1.41523373e+00 7.24494994e-01 5.18986762e-01 1.06034803e+00
-2.16633976e-01 5.46370037e-02 -6.33230627e-01 7.22447753e-01
2.98113257e-01 -1.05368257e+00 -2.61678901e-02 2.88241088e-01
7.21195519e-01 -8.58869329e-02 5.79648376e-01 2.61570364e-01
6.67388737e-01 -1.20940840e+00 2.68799841e-01 -1.42514080e-01
8.64140868e-01 -1.09262037e+00 8.53145599e-01 3.80257726e-01
-1.01156771e+00 -7.91443586e-01 -4.23214072e-03 1.16595984e-01
-3.52569073e-01 2.37115175e-01 -8.70221555e-01 1.59016714e-01
9.33436453e-01 3.11500072e-01 -8.81683350e-01 5.61135530e-01
-2.49425575e-01 1.01147926e+00 -3.09081495e-01 -1.86089694e-01
3.17865938e-01 5.50441325e-01 5.32358527e-01 1.05693460e+00
-7.25569716e-03 7.68318353e-03 9.91481990e-02 1.11589625e-01
-1.79311633e-01 -2.94765145e-01 -8.32760572e-01 1.13756709e-01
7.68442690e-01 1.22906351e+00 -9.68037605e-01 -9.19016376e-02
5.59285656e-02 7.22506166e-01 -3.36628743e-02 -2.78002530e-01
-1.08695424e+00 -3.97918820e-01 5.30129373e-01 2.82927126e-01
2.41923943e-01 -1.98058009e-01 -3.14978480e-01 -4.70968932e-01
-4.41377789e-01 -1.33033442e+00 6.18163168e-01 -1.60959020e-01
-1.25723362e+00 6.97991550e-01 9.70446691e-02 -1.03630590e+00
-1.38029130e-02 -7.96331763e-01 -6.66640460e-01 1.18880674e-01
-1.14458275e+00 -1.20353460e+00 -1.03466868e-01 9.57276881e-01
4.78502274e-01 -6.05611920e-01 8.13375592e-01 3.95949006e-01
-5.92224300e-01 1.02614045e+00 -2.31006779e-02 3.07044357e-01
4.42782640e-01 -1.08316875e+00 2.10257530e-01 8.50190103e-01
8.02682117e-02 5.58974266e-01 4.56628412e-01 -1.01360321e+00
-1.19646478e+00 -1.44875562e+00 2.61934251e-01 -7.23263025e-01
5.72350502e-01 -4.82035190e-01 -5.19041657e-01 6.80746317e-01
1.06156975e-01 2.18084574e-01 8.86082172e-01 -6.29517555e-01
-8.25721145e-01 -1.21920213e-01 -2.08253288e+00 3.83310705e-01
7.76992738e-01 -3.84015858e-01 7.49950390e-03 5.39760172e-01
9.04833555e-01 -8.18666443e-02 -5.90163827e-01 6.33413851e-01
4.57045704e-01 -1.09266543e+00 1.06541562e+00 -5.38217127e-01
-5.98382428e-02 -4.43217784e-01 -4.44128066e-01 -8.22526753e-01
7.95516074e-02 -8.64517987e-01 -4.26461130e-01 1.30017054e+00
4.22052801e-01 -7.46415138e-01 8.00911248e-01 -1.78691208e-01
1.31271586e-01 -8.18034291e-01 -1.07626498e+00 -7.21281171e-01
1.98711440e-01 -3.92883062e-01 6.56747758e-01 9.99613702e-01
-5.14122009e-01 1.37836576e-01 -4.69776601e-01 5.24597704e-01
6.58801913e-01 -8.57522845e-01 7.43724763e-01 -1.37534833e+00
-9.04799923e-02 -1.17660426e-01 -6.11695349e-01 -4.55307662e-02
3.33206415e-01 -6.88872933e-01 -3.65325868e-01 -6.44208908e-01
-1.57257617e-01 -6.85730457e-01 -3.01104754e-01 7.41903245e-01
2.12535813e-01 8.24182212e-01 -4.65265028e-02 1.63315296e-01
-5.44919252e-01 -9.27710906e-02 3.54836136e-01 -2.45574504e-01
-1.41471267e-01 3.41371655e-01 -6.72809660e-01 1.02058911e+00
1.27643871e+00 -1.10702789e+00 -4.03254271e-01 3.25906239e-02
1.71297774e-01 -5.49184144e-01 4.06193793e-01 -1.22683787e+00
1.34281972e-02 2.52272010e-01 3.12316149e-01 -2.38458201e-01
2.21891508e-01 -1.14649212e+00 -9.94402170e-03 1.11729884e+00
-2.02424511e-01 3.77986372e-01 1.30532905e-01 7.62700677e-01
4.43524182e-01 -2.36270770e-01 1.19477618e+00 -1.68444529e-01
-3.70990962e-01 4.36870940e-02 -3.77259523e-01 -5.88354655e-02
1.59357607e+00 -2.40815178e-01 -8.89957845e-01 -2.54304767e-01
-3.73093128e-01 -2.43257716e-01 3.61025602e-01 3.85705531e-01
5.57514906e-01 -9.48765099e-01 -2.58715779e-01 3.07332039e-01
-2.14858264e-01 -5.46971142e-01 -3.10296476e-01 7.04528570e-01
-5.85335195e-01 9.99967381e-02 -2.92692691e-01 -4.13378477e-01
-1.69576001e+00 1.01698923e+00 3.42166483e-01 -5.02558351e-01
1.47753775e-01 7.68480062e-01 -5.84008336e-01 -7.18111038e-01
3.99851620e-01 2.16149807e-01 -4.93386745e-01 -2.23053873e-01
4.26625580e-01 7.45279253e-01 4.99740280e-02 -9.71339405e-01
-7.65121639e-01 6.47270203e-01 -1.73663661e-01 2.69374311e-01
8.22710276e-01 2.66861558e-01 -2.54236549e-01 -6.17929921e-02
1.14065075e+00 3.90752167e-01 -7.27483928e-01 3.16042185e-01
3.12548041e-01 -2.09071457e-01 -2.47407451e-01 -1.06379545e+00
-1.04464197e+00 7.81960607e-01 1.25079095e+00 8.32599819e-01
1.02529621e+00 -3.18780959e-01 9.57101464e-01 3.56258899e-01
3.77689987e-01 -6.01370037e-01 3.38849217e-01 3.86293143e-01
1.86515257e-01 -1.34082139e+00 1.33284898e-02 -3.91190231e-01
-3.69460464e-01 8.38158071e-01 7.28417575e-01 -4.33592349e-01
1.29559672e+00 4.69020814e-01 7.60478526e-02 -2.86877036e-01
-2.25107893e-01 1.57981902e-01 3.61149898e-03 1.08110213e+00
-1.48230493e-01 1.82439224e-03 -2.87964821e-01 5.14207780e-01
-2.06666246e-01 -4.89708483e-01 5.24864316e-01 1.10788679e+00
-4.57316071e-01 -1.29631543e+00 -5.24539590e-01 6.13316417e-01
-1.09208274e+00 1.93768237e-02 -9.11335111e-01 6.36752844e-01
8.32368433e-01 1.57342410e+00 -4.12090063e-01 -1.16315675e+00
-9.97116119e-02 -1.68339193e-01 -2.13675275e-02 -5.05794644e-01
-1.27277207e+00 -4.72478300e-01 -9.73568186e-02 -4.41386878e-01
-3.41423661e-01 -3.29392493e-01 -1.26177216e+00 -3.89607161e-01
-4.32621002e-01 1.41774759e-01 8.13579381e-01 9.37175810e-01
3.98935348e-01 3.85939687e-01 9.46684062e-01 -5.68519354e-01
-6.07219994e-01 -8.80301178e-01 -1.49487361e-01 2.34064996e-01
2.56102651e-01 -9.37794983e-01 -6.10260963e-01 -3.23151797e-01] | [14.394573211669922, 9.654362678527832] |
9d7a7668-5946-45da-b047-81cbbe41a4e3 | contributions-of-shape-texture-and-color-in | 2207.09510 | null | https://arxiv.org/abs/2207.09510v1 | https://arxiv.org/pdf/2207.09510v1.pdf | Contributions of Shape, Texture, and Color in Visual Recognition | We investigate the contributions of three important features of the human visual system (HVS)~ -- ~shape, texture, and color ~ -- ~to object classification. We build a humanoid vision engine (HVE) that explicitly and separately computes shape, texture, and color features from images. The resulting feature vectors are then concatenated to support the final classification. We show that HVE can summarize and rank-order the contributions of the three features to object recognition. We use human experiments to confirm that both HVE and humans predominantly use some specific features to support the classification of specific classes (e.g., texture is the dominant feature to distinguish a zebra from other quadrupeds, both for humans and HVE). With the help of HVE, given any environment (dataset), we can summarize the most important features for the whole task (task-specific; e.g., color is the most important feature overall for classification with the CUB dataset), and for each class (class-specific; e.g., shape is the most important feature to recognize boats in the iLab-20M dataset). To demonstrate more usefulness of HVE, we use it to simulate the open-world zero-shot learning ability of humans with no attribute labeling. Finally, we show that HVE can also simulate human imagination ability with the combination of different features. We will open-source the HVE engine and corresponding datasets. | ['Laurent Itti', 'Xingrui Wang', 'Zhi Xu', 'Yao Xiao', 'Yunhao Ge'] | 2022-07-19 | null | null | null | null | ['classification'] | ['methodology'] | [-2.22747192e-01 -3.29647124e-01 3.01606447e-01 -3.96487117e-01
-3.56064774e-02 -5.05954087e-01 4.68893021e-01 -1.95229836e-02
-5.65843523e-01 4.06176418e-01 -3.13440889e-01 7.69253746e-02
-1.23556145e-01 -8.49897027e-01 -4.64919955e-01 -5.45526564e-01
-1.67758420e-01 4.36464667e-01 3.98802400e-01 -3.80070657e-01
3.18941146e-01 6.16906643e-01 -2.23730135e+00 3.51505369e-01
5.25637209e-01 1.16795874e+00 3.12862009e-01 7.93859661e-01
1.52914643e-01 7.75992453e-01 -5.58779299e-01 -1.65680349e-01
3.58957767e-01 -3.23437929e-01 -8.02922308e-01 1.82819486e-01
4.82282132e-01 -2.29817554e-01 -3.44031751e-01 9.47866321e-01
4.88557816e-01 3.63559753e-01 1.24633455e+00 -1.55425835e+00
-5.70271313e-01 -3.64252739e-02 -3.65304142e-01 1.75023109e-01
4.56289470e-01 5.62925100e-01 8.31780076e-01 -9.53660011e-01
5.05079746e-01 1.22805011e+00 3.32004011e-01 4.70283478e-01
-1.04180121e+00 -4.59649712e-01 -2.55589038e-01 5.29114187e-01
-1.50016272e+00 -3.29443097e-01 6.71494424e-01 -7.33956575e-01
7.46039808e-01 4.35920596e-01 9.05347645e-01 1.00501645e+00
1.79438457e-01 8.17017019e-01 1.27387750e+00 -3.11634243e-01
3.85944933e-01 2.21065685e-01 5.61115623e-01 1.07851136e+00
1.98070481e-01 3.04819196e-01 -5.08691967e-01 1.65567830e-01
6.39662743e-01 -2.76617080e-01 -1.49352983e-01 -4.59262520e-01
-1.38804388e+00 5.72067916e-01 5.86475194e-01 1.35542843e-02
-5.86968251e-02 1.01701014e-01 1.68552414e-01 2.70033032e-01
-1.70068055e-01 5.22114515e-01 -1.25151649e-01 2.52185448e-04
-4.15328324e-01 2.22076848e-01 7.78458953e-01 8.75551343e-01
9.26639020e-01 -8.51311162e-03 -2.38376036e-01 9.59202409e-01
1.18606120e-01 5.97892582e-01 4.29898649e-01 -1.11798441e+00
-1.63102552e-01 6.58243299e-01 2.08785683e-01 -1.00623941e+00
-6.09129429e-01 -6.03662357e-02 -7.72919536e-01 6.14015639e-01
4.34267551e-01 3.27866115e-02 -1.04145539e+00 1.62115240e+00
1.70940951e-01 -1.40420273e-01 9.70348343e-02 1.19293070e+00
1.19403052e+00 4.18287843e-01 -1.52465001e-01 1.77348629e-01
1.51546860e+00 -6.59114242e-01 -2.18593821e-01 -3.32082540e-01
3.29396635e-01 -3.22365075e-01 1.23424959e+00 4.16367203e-01
-5.57410479e-01 -8.34209204e-01 -1.24927688e+00 5.28631695e-02
-7.13012099e-01 2.50848442e-01 6.86881602e-01 4.65283364e-01
-7.00495362e-01 4.22109216e-01 -3.83770734e-01 -7.18512416e-01
1.71540864e-02 1.73227504e-01 -5.61268747e-01 -7.09186075e-03
-1.10418272e+00 1.08275318e+00 5.61299980e-01 -3.20409417e-01
-1.12794781e+00 -1.83195248e-01 -9.61370289e-01 -9.72086638e-02
2.31571496e-01 -6.12815857e-01 9.64070737e-01 -7.33792901e-01
-1.22976494e+00 1.16488147e+00 1.06660590e-01 -2.03940243e-01
4.32572633e-01 1.18355192e-01 -2.37870857e-01 3.33563387e-01
1.56626239e-01 1.00851381e+00 6.99128151e-01 -1.38680947e+00
-7.94554710e-01 -2.93066472e-01 1.17839903e-01 2.78278083e-01
-1.30758137e-01 7.34201148e-02 -5.25886893e-01 -4.86866176e-01
-4.32542562e-02 -1.00673234e+00 7.41055831e-02 3.94285202e-01
-3.59540552e-01 -2.13913396e-01 5.85244656e-01 -3.57088834e-01
7.51185536e-01 -2.39128327e+00 2.23311916e-01 2.29889065e-01
2.58570284e-01 1.44168124e-01 -2.09141210e-01 1.12518184e-02
9.89929885e-02 -9.23728719e-02 -2.63013691e-01 1.79455385e-01
-1.91093013e-02 4.28038985e-01 1.09232098e-01 5.27972996e-01
7.55971894e-02 9.03865218e-01 -8.42262805e-01 -6.71745837e-01
4.48498964e-01 9.33393165e-02 -3.08817178e-01 2.75843501e-01
4.30541746e-02 3.05979729e-01 -2.85395235e-01 6.56019449e-01
2.85915196e-01 -9.10414010e-02 -1.50998309e-01 -1.86275393e-01
-2.48649586e-02 -5.47418535e-01 -1.23470879e+00 1.14637613e+00
-2.11402893e-01 6.84721708e-01 -1.22960292e-01 -7.91263342e-01
9.80057955e-01 -2.45576333e-02 1.55230597e-01 -6.59452200e-01
3.22073907e-01 -2.10366137e-02 1.90791354e-01 -6.47637546e-01
3.88569593e-01 7.12334961e-02 -2.48632610e-01 2.90154546e-01
2.96880394e-01 -2.64335394e-01 1.52192622e-01 1.84898376e-01
7.85384595e-01 1.46042407e-02 4.91854161e-01 -4.13451135e-01
3.02801788e-01 8.76703709e-02 5.24619579e-01 8.76103997e-01
-4.39690232e-01 6.58834040e-01 3.21587503e-01 -5.78221798e-01
-1.00964129e+00 -1.11103785e+00 1.87787656e-02 1.32594752e+00
5.72772503e-01 -8.76110271e-02 -6.41930044e-01 -3.87262017e-01
1.90655097e-01 7.92914391e-01 -7.80775309e-01 -3.81649941e-01
3.28441411e-02 -6.02584898e-01 5.67591786e-01 5.39948940e-01
6.58700526e-01 -1.24003506e+00 -1.26583719e+00 -1.39282987e-01
-2.62072623e-01 -8.74412894e-01 -8.25964808e-02 2.65368104e-01
-2.39366189e-01 -1.24456668e+00 -5.75943649e-01 -8.02965760e-01
5.81440210e-01 4.28154260e-01 1.00946236e+00 2.60053486e-01
-7.74634719e-01 7.22107410e-01 -4.81763601e-01 -4.49948847e-01
-5.84747121e-02 -4.68909055e-01 3.95756096e-01 8.13336000e-02
8.72735977e-02 -2.54096895e-01 -5.13037205e-01 6.76892042e-01
-4.59117383e-01 3.04875463e-01 2.81895041e-01 7.05968976e-01
3.95541310e-01 9.04702172e-02 9.38282236e-02 -4.85431254e-01
3.85957956e-01 -2.70714402e-01 -3.93627018e-01 6.05227411e-01
-4.10798900e-02 1.92599148e-02 3.76988053e-01 -5.27465701e-01
-8.08197439e-01 1.68535143e-01 1.86424926e-01 -3.63964826e-01
-5.70808887e-01 2.90724039e-01 -2.12977767e-01 -3.18145901e-01
8.45618308e-01 4.13368315e-01 -1.93999454e-01 -3.25997710e-01
3.49092692e-01 7.44858861e-01 8.56477916e-01 -7.80909479e-01
6.71426475e-01 3.38616103e-01 1.84409589e-01 -1.07750142e+00
-6.49675071e-01 -5.01659811e-01 -7.28452504e-01 -5.88677168e-01
9.93730009e-01 -6.78520024e-01 -1.05940902e+00 7.96131074e-01
-9.23091769e-01 -1.75515816e-01 -1.57923400e-01 4.88687277e-01
-8.27376366e-01 3.71407032e-01 -5.28920412e-01 -1.00477743e+00
-9.75167975e-02 -1.06970191e+00 8.12837124e-01 3.79444689e-01
-1.09391041e-01 -3.85926664e-01 -2.37592518e-01 1.72565192e-01
-2.69652084e-02 3.02601099e-01 1.04484844e+00 -6.37041807e-01
-2.01485708e-01 6.44130260e-03 -4.46032166e-01 1.85208082e-01
-2.14533120e-01 2.60821521e-01 -1.02716541e+00 -2.21912906e-01
-1.14923544e-01 -6.22622311e-01 9.74074781e-01 1.24039836e-01
1.06967437e+00 8.31291601e-02 -1.49995685e-01 6.89555049e-01
1.18785250e+00 2.17903063e-01 5.64212382e-01 3.13528210e-01
6.55698657e-01 8.31465900e-01 9.14837062e-01 5.35091102e-01
3.89013439e-01 7.40072370e-01 3.67405444e-01 -4.73138280e-02
-1.40333772e-01 9.82561707e-02 3.68884981e-01 5.09747863e-01
-4.00378764e-01 -2.62968466e-02 -1.05386603e+00 4.70184028e-01
-1.81929910e+00 -8.56042385e-01 -9.84536484e-02 2.26353335e+00
5.27161956e-01 1.89437612e-03 1.52731597e-01 8.97509828e-02
7.22480237e-01 -1.91302672e-01 -4.17242765e-01 -2.16665968e-01
-1.90548763e-01 5.18484227e-02 2.39537537e-01 1.28062814e-01
-1.13750935e+00 8.67281914e-01 6.78254938e+00 8.43312979e-01
-9.00303721e-01 -3.86699706e-01 4.06207770e-01 1.18340872e-01
2.24917948e-01 -1.86562479e-01 -6.52891397e-01 3.61974865e-01
4.79887426e-01 -3.44825089e-01 6.96113944e-01 1.05236185e+00
-2.69300163e-01 -4.01712596e-01 -1.28589952e+00 1.27194762e+00
2.75214732e-01 -7.86068916e-01 1.03982575e-01 -1.44432068e-01
2.42873505e-01 -1.69066846e-01 -2.34984253e-02 5.23434639e-01
4.93296772e-01 -1.01264906e+00 1.00241184e+00 6.62624478e-01
9.39041674e-01 -6.57739580e-01 6.00968719e-01 5.61964631e-01
-1.30580962e+00 -1.84554800e-01 -7.16082215e-01 -1.01991057e-01
-2.79793471e-01 1.12995699e-01 -5.62543035e-01 4.62035626e-01
9.18998361e-01 4.41503972e-01 -9.98612523e-01 1.21342170e+00
-3.11355025e-01 1.32042184e-01 -3.35895628e-01 -9.39904153e-02
-6.39833286e-02 -2.11218558e-02 5.76344550e-01 1.13080895e+00
3.05918366e-01 4.63616371e-01 4.72725600e-01 7.30637908e-01
3.76648426e-01 1.70183480e-02 -6.08503163e-01 1.85148090e-01
2.99459517e-01 1.33427405e+00 -7.28107810e-01 -5.72160542e-01
-3.03049266e-01 1.08470953e+00 4.03300643e-01 3.80557239e-01
-7.06610799e-01 -5.98301411e-01 7.40148246e-01 -2.26592660e-01
2.22702533e-01 -9.31514278e-02 -1.43750131e-01 -1.16303289e+00
-3.88693511e-01 -7.98063934e-01 4.12558109e-01 -1.19151950e+00
-1.42685485e+00 6.81213617e-01 7.64033869e-02 -1.34326398e+00
-1.70312777e-01 -9.94547606e-01 -6.09681785e-01 6.30718768e-01
-9.56066787e-01 -1.11935592e+00 -7.29398668e-01 9.21232164e-01
3.78952146e-01 -3.68656814e-01 8.21855724e-01 -1.03512712e-01
-5.48752010e-01 3.59753907e-01 -3.09306413e-01 4.39039290e-01
6.75085127e-01 -1.26679635e+00 3.22038502e-01 6.38710022e-01
2.12002188e-01 4.89681989e-01 6.82820082e-01 -5.62263846e-01
-1.25252748e+00 -8.98528337e-01 3.35503072e-01 -5.57055533e-01
3.10915142e-01 -2.80698955e-01 -7.32554317e-01 3.10125381e-01
-1.76092431e-01 2.01317787e-01 6.96782470e-01 3.33913118e-02
-4.86196399e-01 3.33306082e-02 -1.14773774e+00 4.62621421e-01
1.00560498e+00 -5.63505650e-01 -8.91084313e-01 1.67627744e-02
4.42812055e-01 -1.58415765e-01 -6.52373135e-01 5.07062793e-01
9.17098463e-01 -1.09944904e+00 9.35330391e-01 -6.48451805e-01
2.69352496e-01 -6.56981230e-01 -4.96516258e-01 -1.35862470e+00
-6.72582984e-01 2.44780362e-01 1.16030030e-01 9.05275285e-01
3.08272727e-02 -4.46160018e-01 2.29540557e-01 4.51031476e-01
9.61022303e-02 -3.60190094e-01 -7.04514802e-01 -9.50171173e-01
-3.01161438e-01 -4.11647290e-01 3.35823625e-01 8.98919821e-01
-2.79221078e-03 3.13445479e-01 -6.83808267e-01 1.97488200e-02
8.03018570e-01 3.25341612e-01 9.95126426e-01 -1.52640867e+00
-3.56216371e-01 -2.38188982e-01 -7.88953424e-01 -6.44713104e-01
-5.05363308e-02 -7.81172216e-01 2.66158581e-01 -1.40098226e+00
4.69575405e-01 -2.14682549e-01 -4.58500206e-01 6.41929269e-01
-1.89192668e-01 4.91378725e-01 4.88265276e-01 1.31176993e-01
-5.92759490e-01 3.65536183e-01 1.13204849e+00 -2.70222157e-01
1.13980033e-01 -1.74552560e-01 -4.20700580e-01 8.95319164e-01
5.36008716e-01 -1.57023191e-01 -9.02008042e-02 -9.27372649e-02
-1.00667521e-01 1.37698829e-01 7.25080013e-01 -1.36772645e+00
1.99702278e-01 -4.40391988e-01 7.73975849e-01 -3.79228503e-01
3.89692664e-01 -7.28844047e-01 1.00979611e-01 6.43022120e-01
-1.42666742e-01 -3.17451179e-01 5.68811633e-02 4.84829992e-01
-9.37651396e-02 -2.99998164e-01 9.92130756e-01 -3.17831963e-01
-1.54063463e+00 7.29889944e-02 -4.49120939e-01 -1.12310208e-01
1.06886613e+00 -3.55966628e-01 -3.86767417e-01 -2.78896391e-01
-7.91080356e-01 2.70426959e-01 7.14909196e-01 4.73737657e-01
7.25365460e-01 -1.21757925e+00 -7.36858010e-01 4.34814960e-01
8.08380306e-01 -4.57305700e-01 8.72997046e-02 4.36228633e-01
-6.38705075e-01 -4.85718995e-02 -9.79251564e-01 -5.97528219e-01
-1.38744783e+00 7.59946167e-01 2.97287494e-01 3.37932229e-01
-5.19569457e-01 7.67003417e-01 6.26907408e-01 -2.79866785e-01
3.50234032e-01 -9.88467336e-02 -4.51577038e-01 8.00936371e-02
5.57185054e-01 4.41192001e-01 -2.75649041e-01 -8.27874064e-01
-5.37080884e-01 7.61738956e-01 3.68467540e-01 -2.11578846e-01
1.08678520e+00 4.48082127e-02 -5.05882278e-02 6.22227550e-01
7.79233992e-01 -3.53368074e-01 -9.88434434e-01 -1.29729941e-01
-3.40721041e-01 -4.39426512e-01 -2.10027412e-01 -8.46890032e-01
-9.00196850e-01 1.04054701e+00 7.89661705e-01 -1.30197182e-01
9.69775259e-01 -3.42766233e-02 1.49217635e-01 6.30912721e-01
8.36170673e-01 -1.07147479e+00 1.65558904e-01 5.87473631e-01
8.63749981e-01 -1.28008091e+00 -1.34339891e-02 -3.93947512e-01
-1.12171674e+00 1.07399249e+00 9.75556135e-01 1.46460137e-03
4.14463043e-01 3.58150303e-02 -6.25374615e-02 -3.13261956e-01
-7.12523699e-01 -5.88051081e-01 6.14313424e-01 7.90297985e-01
7.41055831e-02 4.00330156e-01 -8.60635266e-02 7.10160732e-01
-2.43103057e-01 -1.18567161e-01 4.61172909e-01 7.27946877e-01
-9.12948668e-01 -5.88817358e-01 -3.97999734e-01 5.22659838e-01
2.79363692e-01 1.76635027e-01 -5.96649766e-01 5.93744159e-01
4.26980883e-01 9.35454488e-01 1.13116823e-01 -8.62368703e-01
2.87886769e-01 -8.09663236e-02 6.34437561e-01 -6.54679656e-01
-3.86926621e-01 -4.42170948e-01 6.54351115e-02 -5.48392892e-01
-1.60838008e-01 -2.43337169e-01 -1.14298534e+00 -3.08868825e-01
-5.57878017e-02 -1.31883070e-01 4.25422072e-01 7.52293229e-01
-1.56781524e-01 2.89733380e-01 4.93715227e-01 -8.67802978e-01
-3.61042887e-01 -8.19258690e-01 -9.30074394e-01 6.99048102e-01
-6.32944554e-02 -1.07814419e+00 -4.58562613e-01 3.54517736e-02] | [10.118318557739258, 2.1580114364624023] |
da117d17-bd99-4fa6-8e71-945e8c801b40 | variational-inference-for-neyman-scott | 2303.03701 | null | https://arxiv.org/abs/2303.03701v1 | https://arxiv.org/pdf/2303.03701v1.pdf | Variational Inference for Neyman-Scott Processes | Neyman-Scott processes (NSPs) have been applied across a range of fields to model points or temporal events with a hierarchy of clusters. Markov chain Monte Carlo (MCMC) is typically used for posterior sampling in the model. However, MCMC's mixing time can cause the resulting inference to be slow, and thereby slow down model learning and prediction. We develop the first variational inference (VI) algorithm for NSPs, and give two examples of suitable variational posterior point process distributions. Our method minimizes the inclusive Kullback-Leibler (KL) divergence for VI to obtain the variational parameters. We generate samples from the approximate posterior point processes much faster than MCMC, as we can directly estimate the approximate posterior point processes without any MCMC steps or gradient descent. We include synthetic and real-world data experiments that demonstrate our VI algorithm achieves better prediction performance than MCMC when computational time is limited. | ['Christian R. Shelton', 'Chengkuan Hong'] | 2023-03-07 | null | null | null | null | ['point-processes'] | ['methodology'] | [-1.21516529e-02 -3.80599648e-01 -1.42314121e-01 -4.14951324e-01
-1.14986575e+00 -3.95069778e-01 9.47579563e-01 9.14275199e-02
-3.06414038e-01 9.61298764e-01 -1.42048165e-01 -6.58460319e-01
-1.02597095e-01 -7.36820877e-01 -4.73775476e-01 -6.91265583e-01
-1.17851816e-01 1.11191213e+00 4.86648679e-01 5.84805429e-01
3.04124326e-01 4.70625579e-01 -1.14907658e+00 -8.79602656e-02
6.88001633e-01 6.27626836e-01 1.05856903e-01 9.90850389e-01
3.46904173e-02 6.94899917e-01 -5.31780720e-01 -4.44281608e-01
1.31819351e-02 -4.55974668e-01 -3.41189891e-01 -2.23380521e-01
-4.69576508e-01 -3.17480743e-01 1.85744032e-01 9.21099603e-01
1.89297259e-01 2.13570878e-01 1.51972365e+00 -1.29011190e+00
-1.49330750e-01 3.45733672e-01 -8.10716867e-01 1.33942664e-01
2.60961466e-02 2.10917756e-01 8.02946329e-01 -8.77183318e-01
2.25603148e-01 1.58287263e+00 9.15637493e-01 3.94125700e-01
-1.79627597e+00 -6.31237924e-01 -2.32282847e-01 -1.64717451e-01
-1.67913008e+00 -7.58537352e-02 3.43470573e-01 -5.90349734e-01
6.42494500e-01 2.33145971e-02 8.89182985e-01 1.28567719e+00
5.79527080e-01 9.89297867e-01 9.66606319e-01 -2.45551810e-01
1.05072832e+00 -1.49996594e-01 -1.07315488e-01 2.32475325e-01
2.40998894e-01 2.07282484e-01 -2.63037145e-01 -9.63110566e-01
1.09658563e+00 3.52154940e-01 2.59937704e-01 -4.91678156e-02
-9.57900643e-01 1.19308734e+00 -4.05950040e-01 -5.04917026e-01
-5.39549053e-01 7.77273595e-01 -6.22651121e-03 -1.22139692e-01
4.79607135e-01 -2.22124219e-01 -2.52993077e-01 -3.37090582e-01
-1.38214743e+00 6.88370585e-01 1.13757074e+00 9.69588578e-01
5.70422471e-01 -1.18628420e-01 -4.98408139e-01 5.81163704e-01
1.02354121e+00 1.11805058e+00 -2.37000585e-01 -1.46814632e+00
2.69949615e-01 -3.73634309e-01 7.81172395e-01 -4.47650880e-01
1.33943364e-01 4.42065708e-02 -8.99112523e-01 2.04270482e-01
5.99386275e-01 -4.46036726e-01 -9.64352310e-01 1.43547606e+00
4.04098719e-01 4.95246649e-01 -2.88386494e-01 3.64289254e-01
-2.81742215e-01 1.12115788e+00 4.06639069e-01 -6.13378763e-01
1.09737599e+00 -3.79416168e-01 -5.81436038e-01 1.11000828e-01
8.56768116e-02 -9.51169729e-01 6.43868148e-01 1.55786976e-01
-1.18907642e+00 -2.34375954e-01 -6.26461208e-01 4.97168541e-01
1.88483626e-01 -4.69445616e-01 7.48761535e-01 6.16236866e-01
-7.97096074e-01 9.05840099e-01 -1.68072379e+00 -1.93345070e-01
6.40288174e-01 1.19036725e-02 5.67232132e-01 1.60081580e-01
-8.54080856e-01 6.47234440e-01 4.68454994e-02 2.57388558e-02
-1.39634669e+00 -7.88388371e-01 -3.07818502e-01 1.64828792e-01
1.05646633e-01 -7.37141907e-01 1.43667531e+00 -1.42621875e-01
-1.51589489e+00 3.09321642e-01 -6.43356740e-01 -6.60847068e-01
8.92722905e-01 5.59140481e-02 -2.51172394e-01 5.51515147e-02
1.01550080e-01 4.75933909e-01 1.09732831e+00 -1.31747127e+00
-7.15536058e-01 2.55784709e-02 -6.89724445e-01 -6.97121993e-02
5.23676693e-01 1.68134332e-01 -5.66036999e-01 -4.05303538e-01
2.34189257e-01 -9.32268381e-01 -6.84083879e-01 -2.87878998e-02
-3.72932196e-01 -4.06361192e-01 4.78661776e-01 -1.94028333e-01
1.06194115e+00 -1.78424108e+00 -5.44442773e-01 5.06466687e-01
4.26076874e-02 -2.44418889e-01 1.80171132e-01 7.78246343e-01
5.82173824e-01 1.99277416e-01 -4.52743977e-01 -8.31797004e-01
1.46976963e-01 4.84369487e-01 -5.21927774e-01 6.50917232e-01
-4.15694490e-02 7.10964620e-01 -9.01930749e-01 -8.87690365e-01
2.86745727e-01 3.11802387e-01 -6.09997571e-01 -5.12612276e-02
-4.12912607e-01 5.68402886e-01 -4.92642939e-01 5.47904730e-01
8.18757772e-01 -6.53049946e-01 2.30024293e-01 4.53934371e-01
1.19801417e-01 -6.42073825e-02 -1.31562936e+00 8.78742695e-01
-1.43836305e-01 4.33176547e-01 -1.53486833e-01 -6.51030362e-01
6.89676046e-01 2.70387083e-01 4.05383646e-01 2.85885394e-01
-4.33745086e-02 1.91982850e-01 -3.42058569e-01 7.87415430e-02
4.37721491e-01 -6.57715738e-01 -1.23009622e-01 5.52216709e-01
-2.38694236e-01 -3.74452353e-01 1.50592715e-01 1.53513759e-01
8.63512933e-01 2.21802115e-01 3.03787261e-01 -3.17560703e-01
1.45540237e-02 6.09498508e-02 8.02989304e-01 1.51754975e+00
-3.39595109e-01 6.12115860e-01 5.50621152e-01 -2.32042477e-01
-1.19730186e+00 -2.02039528e+00 -2.72721112e-01 5.73627412e-01
1.14359483e-01 -1.49310201e-01 -3.65277857e-01 -2.33865425e-01
1.99804559e-01 9.33509886e-01 -4.25708771e-01 2.23123387e-01
-3.37269545e-01 -1.25146556e+00 4.63639677e-01 5.67390621e-01
2.77148373e-02 -7.62758315e-01 -3.94447863e-01 5.68920195e-01
-4.69214618e-02 -6.94654644e-01 -2.41294160e-01 4.81412821e-02
-1.16805017e+00 -7.54836023e-01 -7.58540273e-01 -1.91112086e-01
4.53748137e-01 -2.93906629e-01 1.04770947e+00 -4.36602354e-01
7.74796605e-02 1.47433951e-01 1.14489533e-01 -5.74833512e-01
-6.82580471e-01 -3.00556213e-01 1.33672133e-01 -1.38392672e-01
6.95535660e-01 -6.77998900e-01 -6.09411120e-01 3.12430561e-01
-5.05235910e-01 -7.41849914e-02 3.56979698e-01 8.12060297e-01
8.86027157e-01 1.37343556e-01 3.67995292e-01 -8.17355871e-01
7.28941560e-01 -5.19283950e-01 -1.15561211e+00 1.95269629e-01
-5.42627633e-01 -2.32933778e-02 1.62670523e-01 -5.02768874e-01
-1.40320516e+00 -1.75104350e-01 3.97498123e-02 -6.60188854e-01
-1.41315773e-01 3.92464399e-01 4.60715860e-01 5.37835002e-01
3.28004837e-01 1.50124922e-01 -2.99152751e-02 -5.01161516e-01
-1.29350219e-02 5.58707535e-01 3.84181261e-01 -7.77133346e-01
6.48018420e-01 8.63819480e-01 4.60323155e-01 -6.82567954e-01
-8.56797278e-01 -3.71458024e-01 -2.94028521e-01 -1.42430574e-01
1.03944933e+00 -9.96946275e-01 -1.22311926e+00 3.41010273e-01
-1.27408552e+00 -4.30952072e-01 -3.27619165e-01 1.08166993e+00
-9.07053590e-01 1.90034240e-01 -8.71817112e-01 -1.55331421e+00
2.79149339e-02 -8.86943042e-01 1.15827453e+00 1.96691319e-01
-4.01238829e-01 -1.33594418e+00 3.65825027e-01 -2.22441610e-02
2.88681258e-02 2.48823449e-01 8.85895610e-01 -3.99389505e-01
-7.25617468e-01 -3.93355995e-01 -4.17446457e-02 3.55734453e-02
-2.71047592e-01 4.67735410e-01 -6.95818841e-01 -1.74223959e-01
-3.19437613e-03 3.01575452e-01 7.20733464e-01 1.05633724e+00
1.07521677e+00 -9.15259048e-02 -8.20385635e-01 3.55132520e-01
1.61340618e+00 3.45163167e-01 6.51986301e-01 -2.06473231e-01
3.16285998e-01 2.70875424e-01 5.57254195e-01 8.45963418e-01
2.20535919e-01 1.34237573e-01 -6.11811504e-02 3.12165290e-01
5.17976880e-01 -7.68373549e-01 3.62553895e-01 7.28009284e-01
-2.37058744e-01 -2.98644215e-01 -1.17383850e+00 5.95130444e-01
-1.98324168e+00 -1.38220286e+00 -4.99928504e-01 2.31611204e+00
8.65137160e-01 2.92257220e-01 5.49323857e-02 -2.62594908e-01
1.01530588e+00 -1.52238861e-01 -5.91007411e-01 -7.56065845e-02
3.59150499e-01 1.77947313e-01 7.94995844e-01 7.57634342e-01
-9.74346817e-01 8.15463185e-01 7.93499804e+00 1.21171784e+00
-2.84944773e-01 2.16936022e-01 4.96900797e-01 -3.43234509e-01
-1.56725764e-01 3.67477715e-01 -1.15947640e+00 9.54127431e-01
1.28468764e+00 -8.42489749e-02 2.14776307e-01 7.19643116e-01
6.57127440e-01 -4.40985262e-01 -1.16611993e+00 1.00868738e+00
-7.88429737e-01 -1.39148033e+00 2.86907870e-02 3.39506865e-01
1.01066411e+00 1.84807718e-01 -1.16050161e-01 3.21216822e-01
1.29572988e+00 -9.24312830e-01 7.10145712e-01 1.03292286e+00
4.11495447e-01 -1.04167020e+00 7.45628417e-01 5.59480011e-01
-9.14866328e-01 1.59188494e-01 -7.20244706e-01 -8.30613822e-02
9.56515491e-01 1.01026917e+00 -8.79042566e-01 -5.68376929e-02
6.09894812e-01 3.76104772e-01 -6.07290817e-03 1.08524013e+00
-1.79306284e-01 1.24981177e+00 -8.11898530e-01 -4.81039822e-01
2.29183957e-01 -5.46038687e-01 6.37696326e-01 9.04641509e-01
5.10388196e-01 -9.38407257e-02 1.44268498e-01 1.21065366e+00
3.85382265e-01 -4.92382258e-01 -1.61364168e-01 -4.61669676e-02
1.09079134e+00 6.45760894e-01 -1.03112209e+00 -5.60228169e-01
-1.46834269e-01 7.62754261e-01 -1.77457124e-01 7.66701400e-01
-1.06436956e+00 -1.32371679e-01 6.89184487e-01 4.16888259e-02
5.83004236e-01 -3.41558069e-01 -4.80108082e-01 -1.02908480e+00
-4.98249620e-01 -2.80812740e-01 1.73471808e-01 -6.91673100e-01
-1.82840729e+00 -8.91472250e-02 4.29002851e-01 -1.03218901e+00
-7.22367644e-01 -2.30718732e-01 -8.41388881e-01 1.21887505e+00
-9.17635143e-01 -7.29103565e-01 3.97924423e-01 4.20154482e-01
3.63233477e-01 6.39091954e-02 5.89270413e-01 -2.25247130e-01
-2.78755724e-01 -1.79262571e-02 8.55090261e-01 -2.21457541e-01
2.75063485e-01 -1.26646459e+00 7.01514542e-01 5.85716963e-01
-1.65250316e-01 7.18754292e-01 1.19762981e+00 -1.07400298e+00
-9.95147705e-01 -9.14270341e-01 7.13131011e-01 -7.78933704e-01
7.96249330e-01 -3.55262667e-01 -6.57789171e-01 7.85804391e-01
-1.98414475e-01 -3.33392262e-01 7.30803311e-01 2.69069970e-01
1.64835885e-01 2.38066003e-01 -1.18263793e+00 8.23385656e-01
5.08026183e-01 -4.42374557e-01 -3.53722125e-01 1.92087844e-01
3.31949174e-01 1.80406556e-01 -9.64249015e-01 1.34321049e-01
6.57211781e-01 -7.14749515e-01 9.54804361e-01 -4.16316271e-01
1.91793084e-01 -5.73981345e-01 -3.18101555e-01 -9.57975805e-01
-2.92348266e-01 -8.91238987e-01 -4.39825237e-01 1.20499432e+00
5.12250125e-01 -6.78362668e-01 8.05492401e-01 6.78973675e-01
5.60207009e-01 -5.34763515e-01 -9.21915531e-01 -8.94193053e-01
2.93328643e-01 -9.12601292e-01 6.76174462e-01 3.07439268e-01
-4.86579448e-01 2.09496394e-01 -4.84641790e-01 3.40133131e-01
1.50395274e+00 1.83432162e-01 6.77692056e-01 -1.59258366e+00
-5.06047845e-01 -1.32843450e-01 1.16987219e-02 -1.29257464e+00
-1.22824386e-01 -3.90901953e-01 4.16976810e-01 -1.54019332e+00
4.44054067e-01 -4.17040199e-01 1.41865537e-01 -1.00751869e-01
-2.76226074e-01 -4.92050648e-02 1.33296978e-02 7.00919330e-01
-5.84640920e-01 6.77106440e-01 1.01528287e+00 1.89125970e-01
-1.07586101e-01 5.46145260e-01 -1.67317152e-01 1.11252344e+00
7.83065796e-01 -8.57531190e-01 -4.05292481e-01 3.17008123e-02
1.94836140e-01 5.27399540e-01 4.67169553e-01 -8.41993511e-01
3.02377999e-01 -6.03823602e-01 7.05020905e-01 -1.32619560e+00
5.09349465e-01 -4.70887899e-01 5.67763150e-01 5.51616013e-01
-3.27847093e-01 -2.16921017e-01 -1.70001969e-01 1.17803109e+00
4.20875289e-02 -5.16491115e-01 1.06238270e+00 -2.02095985e-01
-2.37704664e-02 3.76996785e-01 -1.06545103e+00 -1.01522148e-01
8.90908122e-01 4.34373058e-02 1.92344829e-01 -7.64973581e-01
-8.09092164e-01 4.48069364e-01 3.76693398e-01 -3.97022009e-01
4.25750315e-01 -1.27728748e+00 -5.72809041e-01 -9.04563218e-02
-3.62426728e-01 -2.59747641e-04 1.44732833e-01 8.16114843e-01
-6.48857713e-01 1.25992253e-01 4.97874051e-01 -9.48347151e-01
-8.54363620e-01 4.38415915e-01 1.71053663e-01 -5.44530332e-01
-4.82229799e-01 6.33266687e-01 4.31769304e-02 -3.30919147e-01
-7.06152320e-02 -1.36392206e-01 4.10792649e-01 -9.74529013e-02
5.99791050e-01 8.98675263e-01 -8.49593639e-01 1.26571329e-02
-1.97244763e-01 3.29958051e-01 -5.86802438e-02 -9.92231309e-01
1.11130071e+00 -4.60635215e-01 1.14851505e-01 9.49296296e-01
9.14476752e-01 -4.24796402e-01 -1.83880579e+00 1.15074158e-01
1.47474751e-01 -4.74304765e-01 -2.12982878e-01 -3.14037859e-01
-1.97821930e-01 1.04862309e+00 3.83080363e-01 4.03679639e-01
3.23487252e-01 8.39773268e-02 6.60506368e-01 1.41086861e-01
5.98229945e-01 -1.24313617e+00 -4.22283500e-01 4.50270683e-01
1.45029411e-01 -9.47794378e-01 8.99196938e-02 -2.47112155e-01
-6.08925223e-01 5.40074885e-01 -1.07159078e-01 -1.83527693e-01
1.22094846e+00 3.39093685e-01 -2.81102210e-01 -8.59813094e-02
-1.07470465e+00 6.73478991e-02 -1.21609263e-01 6.56891465e-01
-4.60993201e-02 2.74192929e-01 -2.01272577e-01 3.64614934e-01
-1.23733580e-01 2.34439641e-01 4.52500314e-01 8.51909459e-01
-4.80819792e-01 -8.15313101e-01 -4.90586072e-01 7.25462556e-01
-8.46384108e-01 -1.53029114e-01 4.89911109e-01 4.95840937e-01
-3.30355525e-01 1.17879188e+00 5.55120409e-01 2.78858423e-01
-2.45344684e-01 1.54319093e-01 4.53856260e-01 -2.66680926e-01
1.90247148e-01 6.54313684e-01 -1.51348203e-01 -3.15187335e-01
-3.40260923e-01 -1.16882038e+00 -1.09071469e+00 -8.78151596e-01
-2.23292544e-01 3.49686742e-01 6.35511875e-01 9.84604359e-01
1.72920935e-02 4.21893001e-02 5.57131648e-01 -8.10672522e-01
-9.50474799e-01 -1.02038074e+00 -1.06790721e+00 1.08260401e-01
4.87404950e-02 -6.15431011e-01 -6.10525310e-01 2.55456895e-01] | [6.789247512817383, 3.9176416397094727] |
219ebabe-d914-4862-a480-1d5813e232d9 | designing-an-illumination-aware-network-for | 2207.10582 | null | https://arxiv.org/abs/2207.10582v1 | https://arxiv.org/pdf/2207.10582v1.pdf | Designing An Illumination-Aware Network for Deep Image Relighting | Lighting is a determining factor in photography that affects the style, expression of emotion, and even quality of images. Creating or finding satisfying lighting conditions, in reality, is laborious and time-consuming, so it is of great value to develop a technology to manipulate illumination in an image as post-processing. Although previous works have explored techniques based on the physical viewpoint for relighting images, extensive supervisions and prior knowledge are necessary to generate reasonable images, restricting the generalization ability of these works. In contrast, we take the viewpoint of image-to-image translation and implicitly merge ideas of the conventional physical viewpoint. In this paper, we present an Illumination-Aware Network (IAN) which follows the guidance from hierarchical sampling to progressively relight a scene from a single image with high efficiency. In addition, an Illumination-Aware Residual Block (IARB) is designed to approximate the physical rendering process and to extract precise descriptors of light sources for further manipulations. We also introduce a depth-guided geometry encoder for acquiring valuable geometry- and structure-related representations once the depth information is available. Experimental results show that our proposed method produces better quantitative and qualitative relighting results than previous state-of-the-art methods. The code and models are publicly available on https://github.com/NK-CS-ZZL/IAN. | ['Ming-Ming Cheng', 'Chun-Le Guo', 'Rui-Xun Zhang', 'Zhen Li', 'Zuo-Liang Zhu'] | 2022-07-21 | null | null | null | null | ['image-relighting'] | ['computer-vision'] | [ 3.63254398e-01 -2.45361641e-01 1.06358089e-01 -6.21271968e-01
-2.45102242e-01 -3.14900488e-01 3.55954409e-01 -5.31731129e-01
-1.36311188e-01 5.54270267e-01 4.89969701e-02 -1.43783823e-01
2.54288346e-01 -1.04893839e+00 -7.17822194e-01 -8.21225524e-01
4.36303705e-01 -3.43293577e-01 5.06779477e-02 -3.87837321e-01
5.15528798e-01 8.43846560e-01 -1.65702188e+00 1.73965901e-01
8.35959554e-01 1.08910036e+00 4.98214394e-01 4.69637454e-01
-1.38067335e-01 7.06732333e-01 -3.53216857e-01 -4.34990615e-01
4.40757036e-01 -5.06607294e-01 -5.29500425e-01 2.61432856e-01
3.93062294e-01 -7.27631211e-01 -5.38666427e-01 1.21590531e+00
5.88479340e-01 2.64265120e-01 3.37328762e-01 -1.05583954e+00
-8.98200750e-01 1.68053005e-02 -8.08548391e-01 -1.71017662e-01
4.16836262e-01 1.88536033e-01 4.39503878e-01 -9.15277362e-01
6.06118083e-01 1.28395784e+00 2.57367045e-01 3.43261153e-01
-9.43106115e-01 -7.33108878e-01 1.56997725e-01 2.58615375e-01
-1.50160813e+00 -5.27388453e-01 1.37065136e+00 -8.24481100e-02
4.39283460e-01 3.40776861e-01 7.59197116e-01 8.03490937e-01
2.44785398e-01 6.02514625e-01 1.23865747e+00 -5.04006386e-01
9.90301967e-02 1.66446909e-01 -3.82643163e-01 7.98659384e-01
-6.69589546e-03 2.63769954e-01 -6.25288367e-01 3.28763753e-01
1.33328128e+00 1.80349559e-01 -7.37342477e-01 -3.61389160e-01
-9.82857823e-01 5.72614908e-01 7.22685516e-01 1.07975202e-02
-2.32155010e-01 1.98802069e-01 1.44103440e-02 1.99860394e-01
6.19621694e-01 4.78356361e-01 -1.01572983e-01 1.35882087e-02
-8.71177197e-01 -7.30712116e-02 3.71628523e-01 1.16910720e+00
1.12347913e+00 1.06453523e-01 9.11561027e-02 8.30356658e-01
7.42978677e-02 5.05719125e-01 6.36963844e-02 -1.36482692e+00
2.26698041e-01 5.60845077e-01 1.89199328e-01 -1.28221154e+00
-1.10243276e-01 -2.59312749e-01 -1.06254017e+00 4.98058170e-01
1.11619502e-01 1.15060218e-01 -7.51712680e-01 1.33715141e+00
5.27060091e-01 2.97028315e-03 -2.53957778e-01 1.22087491e+00
8.57039154e-01 7.76699960e-01 -3.42359006e-01 -3.12277734e-01
1.17194653e+00 -9.98840451e-01 -8.94395709e-01 3.82060595e-02
2.41748273e-01 -1.06099367e+00 1.34483588e+00 6.67450845e-01
-1.36775482e+00 -6.69392347e-01 -1.00109041e+00 -7.21682489e-01
-1.96612194e-01 1.95518181e-01 7.38106728e-01 3.92011493e-01
-1.08900130e+00 5.50360322e-01 -5.98111093e-01 -1.11947775e-01
4.20345157e-01 1.00812644e-01 -1.11700185e-01 -3.42034996e-01
-1.07926452e+00 7.23118126e-01 1.69388708e-02 4.32723045e-01
-6.75146103e-01 -5.45977712e-01 -8.69649410e-01 -1.57807276e-01
3.05815041e-01 -6.26674950e-01 9.69913006e-01 -9.56186533e-01
-1.96411395e+00 7.46851027e-01 -3.07194293e-01 2.18260035e-01
4.53203231e-01 -1.09912604e-01 -2.36839518e-01 4.94003028e-01
-1.88635811e-01 8.01916182e-01 1.00120997e+00 -1.67796361e+00
-3.35104406e-01 -1.72675475e-01 4.22956944e-01 4.68927532e-01
-2.82940447e-01 -5.71961440e-02 -8.02681923e-01 -5.17400503e-01
2.75695115e-01 -5.50762117e-01 -1.26283273e-01 5.86656868e-01
-3.87546688e-01 2.38816947e-01 7.66015351e-01 -7.10446656e-01
1.04957259e+00 -2.22152853e+00 1.48052529e-01 -8.69088620e-02
8.79578963e-02 1.58564746e-01 -1.08655863e-01 4.93604332e-01
-6.43831119e-02 -5.23387939e-02 -5.05578704e-02 -4.57964957e-01
-2.28293240e-01 1.83430031e-01 -4.33160365e-01 6.92525506e-01
2.63902154e-02 8.08741093e-01 -7.96523273e-01 -7.26212800e-01
7.69990027e-01 9.04480696e-01 -7.24189937e-01 3.75634730e-01
-7.38554522e-02 6.28865421e-01 -3.81864846e-01 7.40799308e-01
1.01260817e+00 -9.77265537e-02 -2.08520144e-01 -7.27210641e-01
-3.69925678e-01 7.37664700e-02 -1.08193719e+00 1.97612751e+00
-8.01604748e-01 5.71759641e-01 1.29681379e-01 -6.38209939e-01
1.06123292e+00 -3.52725647e-02 2.58213550e-01 -8.57371628e-01
4.82753664e-01 1.92217063e-03 -5.45221686e-01 -5.47261238e-01
6.65771127e-01 -1.67913333e-01 2.25214899e-01 1.72595367e-01
-4.43410277e-01 -6.53377175e-01 -6.44720420e-02 9.47795659e-02
4.76800412e-01 5.38161099e-01 1.39922485e-01 7.18450397e-02
5.13327181e-01 -2.73428589e-01 6.17163658e-01 2.46936247e-01
8.03318843e-02 9.50849235e-01 2.50822902e-01 -4.98393327e-01
-9.31889236e-01 -9.44424391e-01 -1.46967679e-01 6.95576012e-01
7.69616783e-01 -1.79193839e-01 -6.57265842e-01 -2.53114309e-02
-5.03730059e-01 9.50854242e-01 -3.97354156e-01 -9.07660425e-02
-6.06419265e-01 -3.46720576e-01 3.45126465e-02 2.14646742e-01
9.39504921e-01 -1.01270604e+00 -6.57996893e-01 -1.69809490e-01
-4.15440798e-01 -1.10739851e+00 -5.35555243e-01 -2.08366185e-01
-7.59725153e-01 -8.44851315e-01 -5.46762705e-01 -6.87327921e-01
9.75766599e-01 7.06682980e-01 1.00849843e+00 2.80677915e-01
-4.03131813e-01 4.23143283e-02 -3.83983642e-01 -2.77091146e-01
-3.24079283e-02 -1.74598530e-01 -3.04716587e-01 1.10323608e-01
-1.48791596e-01 -7.75717080e-01 -1.15679038e+00 4.29073930e-01
-1.21704018e+00 7.60249138e-01 5.90652883e-01 5.47563553e-01
6.41537011e-01 3.08463842e-01 -3.19019929e-02 -5.50436676e-01
2.07946464e-01 8.74839276e-02 -7.30509043e-01 4.85865325e-02
-3.09858769e-01 -2.43542418e-01 7.78065920e-01 -2.31286258e-01
-1.54100764e+00 9.30787921e-02 1.64060947e-02 -6.58316076e-01
-1.28945708e-01 4.80801538e-02 -4.63901371e-01 -3.68933320e-01
2.98032165e-01 3.46852481e-01 -2.44567115e-02 -3.02650392e-01
5.46903014e-01 6.03120506e-01 4.39443350e-01 -5.57368994e-01
9.84256744e-01 9.25617576e-01 1.87442034e-01 -9.46896732e-01
-9.53109622e-01 -2.98847973e-01 -6.96319163e-01 -4.13050979e-01
6.95623338e-01 -9.17114198e-01 -8.38135183e-01 5.04195154e-01
-1.22688794e+00 -4.10192013e-01 -2.36056998e-01 3.24332982e-01
-4.82418299e-01 5.18447995e-01 -6.16927564e-01 -6.15828812e-01
-2.15562940e-01 -1.23163068e+00 1.34443521e+00 3.32635611e-01
1.25558183e-01 -8.23733211e-01 -3.03674519e-01 5.22040963e-01
4.01293904e-01 1.42672464e-01 6.28316164e-01 8.14800262e-01
-1.13300896e+00 1.50571421e-01 -5.88470459e-01 4.74762678e-01
3.12990606e-01 1.16975382e-01 -1.10258663e+00 -1.83643699e-01
2.58501321e-01 -3.03339213e-01 5.28558910e-01 3.26167136e-01
1.67777932e+00 -1.34119242e-01 -7.77709670e-03 1.25563478e+00
1.71686900e+00 2.52676070e-01 1.12072289e+00 2.81540573e-01
9.52384353e-01 7.10022330e-01 7.65282989e-01 4.27610397e-01
3.84268910e-01 7.68198490e-01 5.34399688e-01 -6.10554576e-01
-3.08513701e-01 -3.01877320e-01 2.99573481e-01 7.69254208e-01
-2.81231284e-01 -4.07267451e-01 -3.85971397e-01 2.17982292e-01
-1.38090456e+00 -9.08991218e-01 -1.39769137e-01 2.02885008e+00
9.81200933e-01 -3.42485011e-01 -5.06221592e-01 4.23925854e-02
5.66004038e-01 3.76464278e-01 -5.20978272e-01 -4.78495568e-01
-7.70293698e-02 1.11926891e-01 4.92619812e-01 6.72893465e-01
-6.74412251e-01 1.07109344e+00 5.52292824e+00 1.01424921e+00
-1.43923020e+00 -9.93846729e-02 7.79145300e-01 -2.24806041e-01
-5.08535862e-01 1.00674061e-02 -5.28183162e-01 3.01841855e-01
2.87271172e-01 1.02552928e-01 7.68658817e-01 6.44084811e-01
6.85492873e-01 -3.73749852e-01 -8.43691885e-01 1.34364533e+00
2.75295436e-01 -1.05777025e+00 2.38597229e-01 7.60255530e-02
7.80984461e-01 -3.92368644e-01 1.75291762e-01 -2.77796894e-01
5.22341058e-02 -8.34669113e-01 6.58970833e-01 6.52907073e-01
1.18106842e+00 -7.92050302e-01 3.80988210e-01 1.61067128e-01
-1.06324971e+00 3.13154846e-01 -6.45083487e-01 -1.59405336e-01
3.44090849e-01 7.96681583e-01 -3.00715119e-01 7.03102708e-01
6.61091745e-01 8.12892735e-01 -4.09846932e-01 6.71153188e-01
-5.67801058e-01 2.63224870e-01 -9.69159603e-02 2.25071177e-01
3.22610810e-02 -5.95753610e-01 1.36590853e-01 7.62551308e-01
4.38416243e-01 4.91717935e-01 -1.80438571e-02 9.40051138e-01
-1.73026919e-01 1.39434293e-01 -6.14680111e-01 3.30728352e-01
2.01863736e-01 1.57424080e+00 -8.17020237e-01 -2.29530975e-01
-3.36621821e-01 1.36347997e+00 2.06920519e-01 5.50577402e-01
-9.76693451e-01 -4.95027632e-01 6.12201095e-01 2.51897693e-01
1.68498099e-01 -3.83040369e-01 -3.60851049e-01 -1.37331259e+00
1.18303142e-01 -7.06943154e-01 -2.58874148e-01 -1.40990984e+00
-1.02940357e+00 5.41751266e-01 -3.25801857e-02 -1.49659503e+00
3.21623802e-01 -5.63485265e-01 -5.22571623e-01 9.39480543e-01
-2.06799793e+00 -1.22264290e+00 -8.38750601e-01 7.09801316e-01
5.29854715e-01 4.28847343e-01 4.79647279e-01 3.26567769e-01
-4.31879252e-01 2.36488760e-01 -9.97810513e-02 4.80328575e-02
8.02638173e-01 -8.52038622e-01 1.13190703e-01 9.27840531e-01
-1.63034737e-01 6.80890799e-01 6.41486704e-01 -3.94359499e-01
-1.62294102e+00 -1.12478888e+00 5.48405409e-01 -1.02677114e-01
1.38997689e-01 -3.71972263e-01 -7.97213316e-01 4.14493501e-01
3.62794340e-01 5.56594273e-03 3.48278284e-01 -3.81031275e-01
-1.26615122e-01 -3.64407092e-01 -1.05894709e+00 9.53692794e-01
1.32788098e+00 -5.88489950e-01 -1.22473389e-01 2.78437018e-01
8.33587289e-01 -5.52232146e-01 -6.79080367e-01 1.24430716e-01
5.33147097e-01 -1.47607195e+00 1.15645850e+00 2.85791218e-01
8.79067838e-01 -6.27377033e-01 -1.15984775e-01 -1.26608753e+00
-2.58575499e-01 -7.51837850e-01 3.78228843e-01 1.18959296e+00
-2.18707502e-01 -6.41959310e-01 6.30080819e-01 5.86486697e-01
-2.99941540e-01 -8.32819760e-01 -4.90834594e-01 -5.29353321e-01
-1.97616249e-01 -4.79323268e-01 8.19957137e-01 8.91398072e-01
-3.35987180e-01 9.16500762e-02 -5.56108236e-01 3.16513836e-01
7.38421500e-01 4.95149374e-01 1.00154972e+00 -6.75340056e-01
-1.28564999e-01 -2.55599320e-01 -2.43114829e-01 -1.33156252e+00
-7.41922557e-02 -5.47935665e-01 7.88942948e-02 -1.58037126e+00
3.86132188e-02 -4.43277419e-01 4.96972911e-02 1.11869328e-01
-5.21757826e-02 5.00875235e-01 5.73434979e-02 2.03751981e-01
-3.66435885e-01 9.55062330e-01 2.02736855e+00 2.33019292e-01
-1.87095523e-01 -3.99205506e-01 -7.72280037e-01 8.18545043e-01
9.41634715e-01 -1.03073604e-01 -7.92532504e-01 -6.74592733e-01
2.88123637e-01 1.18694916e-01 5.11873543e-01 -7.16973603e-01
1.61357030e-01 -4.54484522e-01 5.30541956e-01 -7.51556754e-01
7.45072067e-01 -7.63960004e-01 1.90541446e-01 1.56915486e-01
-2.39498556e-01 -1.57797650e-01 -1.82999238e-01 3.19871098e-01
-3.07302684e-01 -1.32092074e-01 8.93651128e-01 -1.84587196e-01
-5.80183029e-01 4.71041441e-01 5.76710217e-02 -1.92107022e-01
8.85846734e-01 -3.50286812e-01 -2.34259173e-01 -5.85361719e-01
-3.24138403e-02 -9.69188064e-02 1.02491462e+00 1.44207552e-01
9.74489748e-01 -1.25465894e+00 -3.22686076e-01 4.51948017e-01
2.18821019e-02 4.31290746e-01 4.47359920e-01 5.18123746e-01
-1.01432359e+00 5.07907048e-02 -2.25550324e-01 -4.46186006e-01
-1.19291747e+00 5.68654716e-01 2.19375327e-01 1.93152249e-01
-7.19018817e-01 7.65287101e-01 6.88348830e-01 -2.13715285e-01
4.14662287e-02 -4.57566172e-01 5.34341112e-02 -2.73161978e-01
5.85094035e-01 2.41659820e-01 -1.83305413e-01 -4.55829948e-01
4.45208400e-02 9.29638028e-01 1.21026628e-01 -8.00499395e-02
1.36035621e+00 -5.67782044e-01 -3.76631975e-01 2.66297787e-01
1.29073906e+00 -3.18307243e-02 -1.40104997e+00 -2.06685528e-01
-9.17680383e-01 -1.15152407e+00 4.70124513e-01 -4.81081337e-01
-1.41979241e+00 1.19211257e+00 3.37288022e-01 -2.26677284e-01
1.63772297e+00 -3.48199546e-01 9.70172763e-01 2.03118518e-01
6.18144691e-01 -1.02707171e+00 1.82945877e-01 3.34827267e-02
1.13447642e+00 -1.20011759e+00 3.04431438e-01 -9.61069942e-01
-4.31080461e-01 1.16688287e+00 6.52050614e-01 2.74954550e-02
5.89222252e-01 2.15553939e-01 1.85379669e-01 -2.80952692e-01
-4.18469459e-01 9.26617086e-02 1.81555837e-01 4.34526771e-01
4.11128253e-01 -1.19302318e-01 5.74507518e-03 -2.79831905e-02
-3.02476883e-01 1.85667023e-01 6.52908802e-01 8.09856594e-01
-2.35602722e-01 -9.88822103e-01 -4.03341532e-01 -8.23748186e-02
-1.08679071e-01 -2.44074628e-01 -6.86996728e-02 5.79222381e-01
2.78058201e-01 9.39566076e-01 -8.39092657e-02 -2.09215701e-01
2.14492321e-01 -4.98547852e-01 6.52622461e-01 -5.20840943e-01
-1.15524210e-01 1.82034060e-01 -9.43581238e-02 -9.23814118e-01
-6.55367553e-01 -3.63766640e-01 -1.05445158e+00 -5.22010267e-01
-3.22812259e-01 -3.44327509e-01 7.29220688e-01 4.57439482e-01
3.03781509e-01 4.97127235e-01 9.89184082e-01 -1.31517911e+00
3.73276100e-02 -6.00177288e-01 -6.93167686e-01 4.19620484e-01
1.51938632e-01 -6.94424927e-01 -5.46635091e-01 2.22097546e-01] | [10.059147834777832, -2.6702287197113037] |
21712fd8-b058-473c-b7df-337ac0fc5efb | grounded-language-image-pre-training | 2112.03857 | null | https://arxiv.org/abs/2112.03857v2 | https://arxiv.org/pdf/2112.03857v2.pdf | Grounded Language-Image Pre-training | This paper presents a grounded language-image pre-training (GLIP) model for learning object-level, language-aware, and semantic-rich visual representations. GLIP unifies object detection and phrase grounding for pre-training. The unification brings two benefits: 1) it allows GLIP to learn from both detection and grounding data to improve both tasks and bootstrap a good grounding model; 2) GLIP can leverage massive image-text pairs by generating grounding boxes in a self-training fashion, making the learned representation semantic-rich. In our experiments, we pre-train GLIP on 27M grounding data, including 3M human-annotated and 24M web-crawled image-text pairs. The learned representations demonstrate strong zero-shot and few-shot transferability to various object-level recognition tasks. 1) When directly evaluated on COCO and LVIS (without seeing any images in COCO during pre-training), GLIP achieves 49.8 AP and 26.9 AP, respectively, surpassing many supervised baselines. 2) After fine-tuned on COCO, GLIP achieves 60.8 AP on val and 61.5 AP on test-dev, surpassing prior SoTA. 3) When transferred to 13 downstream object detection tasks, a 1-shot GLIP rivals with a fully-supervised Dynamic Head. Code is released at https://github.com/microsoft/GLIP. | ['Jianfeng Gao', 'Kai-Wei Chang', 'Jenq-Neng Hwang', 'Lei Zhang', 'Lu Yuan', 'Lijuan Wang', 'Yiwu Zhong', 'Chunyuan Li', 'Jianwei Yang', 'Haotian Zhang', 'Pengchuan Zhang', 'Liunian Harold Li'] | 2021-12-07 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Li_Grounded_Language-Image_Pre-Training_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Li_Grounded_Language-Image_Pre-Training_CVPR_2022_paper.pdf | cvpr-2022-1 | ['phrase-grounding'] | ['natural-language-processing'] | [ 2.78306723e-01 7.05641583e-02 -2.87821800e-01 -1.87389314e-01
-1.38134289e+00 -6.11987352e-01 8.80148709e-01 1.36464983e-02
-4.11027551e-01 3.08958948e-01 1.46188542e-01 -3.09108019e-01
4.60522145e-01 -5.74026465e-01 -1.17359507e+00 -3.45937997e-01
-2.95588356e-02 3.62501442e-01 3.91477376e-01 -2.28979334e-01
-9.10871662e-03 -1.23344041e-01 -1.64224541e+00 9.04922128e-01
6.01670384e-01 1.07172298e+00 4.82313097e-01 7.74318337e-01
-1.92823410e-01 9.75546658e-01 -4.10857379e-01 -5.34422815e-01
3.63941014e-01 -3.99882704e-01 -8.45363796e-01 4.09626484e-01
1.01754928e+00 -2.31462717e-01 -2.38787830e-01 8.97704065e-01
3.80879074e-01 4.55017351e-02 7.64630735e-01 -1.48347664e+00
-1.18287468e+00 5.78656495e-01 -8.07608843e-01 2.61401445e-01
1.88843220e-01 6.20423794e-01 1.41077960e+00 -1.51680744e+00
7.58861065e-01 1.34519124e+00 6.94763362e-01 5.80633402e-01
-1.33802927e+00 -7.40025163e-01 7.98519850e-02 -2.11297609e-02
-1.42119336e+00 -5.02019882e-01 2.12766066e-01 -5.86110711e-01
1.16453385e+00 -1.86843261e-01 4.62648153e-01 1.47061741e+00
-8.59930813e-02 1.18647373e+00 1.07529533e+00 -5.31156361e-01
-9.76355672e-02 1.99406985e-02 2.06563264e-01 9.93124306e-01
2.70016670e-01 5.02778143e-02 -7.86967635e-01 2.08114952e-01
5.40624201e-01 -1.41376942e-01 -1.11543722e-01 -3.06625515e-01
-1.33474660e+00 7.34836280e-01 8.44799340e-01 1.85836941e-01
-8.17433819e-02 4.84252334e-01 2.88856506e-01 1.16366386e-01
4.26420718e-01 5.01323462e-01 -2.11257532e-01 3.90396863e-02
-1.02772987e+00 1.45654222e-02 4.53096181e-01 1.34244287e+00
8.33450556e-01 1.81375667e-01 -5.25657117e-01 7.70445466e-01
2.87091702e-01 8.10370147e-01 4.34368789e-01 -5.83596528e-01
7.91655838e-01 6.20205641e-01 -1.16484240e-02 -6.12943828e-01
-2.32775450e-01 -4.64376867e-01 -6.84274912e-01 2.58801877e-01
5.00055850e-01 -2.15078369e-02 -1.38060915e+00 1.63193262e+00
-1.61674261e-01 2.31365159e-01 8.76859128e-02 7.43601561e-01
1.29338253e+00 9.92674351e-01 4.25070882e-01 3.37475806e-01
1.72797418e+00 -1.36891758e+00 -2.31620088e-01 -7.27945328e-01
6.47426009e-01 -6.02695882e-01 1.70312321e+00 3.25419098e-01
-1.08442712e+00 -8.67608905e-01 -1.11874557e+00 -3.82382125e-01
-4.65333343e-01 2.04266071e-01 3.75670403e-01 2.14984655e-01
-1.19177866e+00 2.11139157e-01 -6.17247343e-01 -6.12082481e-01
7.33088851e-01 -1.81595266e-01 -3.99909884e-01 -3.77552360e-01
-9.45365369e-01 7.29036450e-01 6.08870089e-01 -3.39023709e-01
-1.51799393e+00 -8.19443285e-01 -1.02611744e+00 -4.31168042e-02
4.07511204e-01 -6.86699927e-01 1.31900322e+00 -9.12827671e-01
-9.43296015e-01 1.52119386e+00 9.45064127e-02 -7.55368769e-01
4.28560466e-01 -4.01345223e-01 -3.92114431e-01 3.52340102e-01
5.20575404e-01 1.44862187e+00 9.24308777e-01 -1.38861120e+00
-7.88394332e-01 1.59629304e-02 7.95432106e-02 2.09199399e-01
-2.00679019e-01 -1.43671287e-02 -8.81263077e-01 -6.99438751e-01
-2.36619294e-01 -7.77369261e-01 -3.08767203e-02 1.75466165e-01
-5.04743755e-01 -1.15969241e-01 5.78467548e-01 -5.07057309e-01
5.30813575e-01 -2.37978721e+00 -1.12719595e-01 -3.97342741e-01
2.64256269e-01 2.64438629e-01 -8.39217305e-01 4.51215386e-01
6.64911643e-02 1.36393115e-01 -1.77177638e-01 -6.43388331e-01
5.05784191e-02 1.39700279e-01 -5.53770781e-01 2.92588145e-01
6.65298104e-01 1.50617588e+00 -1.04036319e+00 -6.32725477e-01
1.65194526e-01 3.32045108e-01 -5.61958790e-01 3.44343632e-01
-4.83787358e-01 1.78510696e-01 7.62768611e-02 7.04557538e-01
2.64683992e-01 -7.32904136e-01 -9.10095274e-02 -1.52655751e-01
1.41362295e-01 1.52077660e-01 -8.46729457e-01 1.92746603e+00
-5.58018446e-01 9.25570548e-01 -1.90068960e-01 -8.20251942e-01
8.57746482e-01 4.17757332e-02 7.94536099e-02 -1.01048017e+00
6.24389481e-03 -4.89939339e-02 -3.12006623e-01 -3.23793262e-01
5.51394224e-01 9.03230626e-03 -1.71502143e-01 2.62590677e-01
7.67530978e-01 -2.09447250e-01 3.45348060e-01 6.25138760e-01
9.92564380e-01 3.08677733e-01 2.23788530e-01 -3.30202162e-01
2.15226576e-01 1.22076824e-01 2.20535487e-01 1.09497750e+00
-1.09331064e-01 9.79529858e-01 4.73971218e-01 -1.63430363e-01
-9.38970506e-01 -1.41652632e+00 -1.61137700e-01 1.64780915e+00
3.14084888e-01 -5.98859549e-01 -5.62935770e-01 -6.56020164e-01
-2.81492546e-02 7.43823946e-01 -7.40104020e-01 -5.46659529e-02
-1.97076872e-01 -4.71953124e-01 6.97710991e-01 7.69577861e-01
6.24476552e-01 -1.21773422e+00 -3.56991142e-01 -4.81211580e-02
-2.90553123e-01 -1.41254795e+00 -4.46913511e-01 2.03477919e-01
-6.04305685e-01 -9.92487431e-01 -7.39721239e-01 -9.51774299e-01
4.57508355e-01 7.56829798e-01 1.73172331e+00 1.94251046e-01
-5.96387804e-01 5.76650679e-01 -5.16953886e-01 -4.81329530e-01
-3.09448332e-01 1.36309490e-01 -2.53495097e-01 -1.09451845e-01
3.63065869e-01 -1.17207490e-01 -4.98633146e-01 2.33861998e-01
-6.57716453e-01 4.06472296e-01 7.70700693e-01 8.61443758e-01
8.44813645e-01 -4.78470802e-01 4.63888884e-01 -6.57037616e-01
1.45173445e-01 -5.36345840e-01 -6.57778800e-01 3.18168670e-01
-4.34459418e-01 -1.15857430e-01 2.00637460e-01 -3.47059250e-01
-8.73775661e-01 -1.12134635e-01 -2.28598397e-02 -3.88680100e-01
-1.96644306e-01 2.01388747e-01 -1.16244117e-02 2.25638896e-01
1.06439424e+00 2.37294286e-01 -1.37954026e-01 -3.71514738e-01
9.44793999e-01 5.03645778e-01 9.17290330e-01 -7.73339927e-01
9.79258835e-01 4.54646438e-01 -5.67143500e-01 -7.79515445e-01
-1.39304614e+00 -7.08895922e-01 -4.09865767e-01 -2.66127884e-02
1.22199190e+00 -1.41508555e+00 -2.86851853e-01 2.19002351e-01
-9.90539372e-01 -8.30980361e-01 -4.73949671e-01 2.07185015e-01
-6.50070429e-01 1.32660106e-01 -5.46955287e-01 -5.21213591e-01
-4.66729760e-01 -9.69464481e-01 1.38342619e+00 4.39518690e-02
-1.10759203e-04 -6.04468703e-01 -1.75739646e-01 5.45545757e-01
7.13161156e-02 8.90385434e-02 4.71176803e-01 -5.65510035e-01
-6.67416811e-01 -1.09037213e-01 -7.93386817e-01 4.91922259e-01
-2.45689869e-01 -1.77251190e-01 -1.14299762e+00 -4.21885222e-01
-6.36811256e-01 -9.79883671e-01 1.19072294e+00 2.25450873e-01
9.68982637e-01 -1.36678383e-01 -2.14366645e-01 7.94736564e-01
1.42593753e+00 -3.12817991e-01 6.19142890e-01 3.20794046e-01
7.34182596e-01 3.97754818e-01 6.40900850e-01 1.36670515e-01
3.25377464e-01 6.64194345e-01 6.05812430e-01 -4.31059748e-01
-7.26174355e-01 -6.13236725e-01 7.10405231e-01 4.15784568e-01
9.74509865e-02 -4.14690316e-01 -1.24059963e+00 7.48372972e-01
-1.82218707e+00 -9.98762965e-01 -1.48173571e-01 1.86217880e+00
9.29663718e-01 3.02549362e-01 2.41989791e-01 -4.31835562e-01
6.33969069e-01 3.06236923e-01 -3.03709149e-01 6.66384175e-02
-3.26985836e-01 2.22219884e-01 4.00137842e-01 5.47694325e-01
-1.22810888e+00 1.29549766e+00 5.83894444e+00 7.98232794e-01
-8.41385424e-01 4.90555614e-01 6.96453214e-01 -1.56311408e-01
-1.40613794e-01 -1.68559089e-01 -1.06321204e+00 2.52807677e-01
7.88852930e-01 -7.92041421e-02 1.84277475e-01 1.16092932e+00
-3.00238580e-01 1.48503870e-01 -1.12700105e+00 1.19436336e+00
2.78195411e-01 -1.63075483e+00 2.30440900e-01 -9.83309820e-02
9.84628797e-01 7.28025615e-01 2.15762153e-01 7.25237429e-01
6.61007524e-01 -1.25004303e+00 1.10549831e+00 -3.08401673e-03
1.11158955e+00 -3.11409205e-01 5.62964201e-01 1.69027522e-01
-1.24528897e+00 6.97465464e-02 -4.05086398e-01 -2.67318189e-02
1.56550184e-01 3.36165488e-01 -9.13916647e-01 4.40890282e-01
8.77988696e-01 8.01482975e-01 -9.73635912e-01 8.12597394e-01
-6.68905437e-01 7.93679297e-01 -1.11050822e-01 1.97458968e-01
6.81320608e-01 1.97657421e-01 3.75622094e-01 1.71857178e+00
-1.47063972e-03 -1.91476986e-01 4.60433096e-01 9.72220540e-01
-3.66867512e-01 1.11502998e-01 -6.14257812e-01 -1.46253094e-01
2.81727552e-01 1.33969951e+00 -7.49691784e-01 -6.65834010e-01
-6.30914271e-01 1.02495575e+00 3.27064544e-01 6.00255787e-01
-9.99709606e-01 -2.49473602e-01 5.49463689e-01 1.19507037e-01
4.31414217e-01 -2.62811184e-01 -1.51649177e-01 -1.37808478e+00
-1.76512390e-01 -9.27362442e-01 5.94183266e-01 -1.18499851e+00
-1.45050347e+00 7.60610640e-01 8.26454759e-02 -1.13798738e+00
-1.55147061e-01 -9.71743345e-01 -4.88379896e-01 6.58644497e-01
-1.40443265e+00 -1.62863970e+00 -6.04045272e-01 5.65290391e-01
8.57976735e-01 -1.42658398e-01 6.10014081e-01 1.04873903e-01
-3.89268428e-01 6.63110793e-01 -3.14413637e-01 5.13898790e-01
7.15172887e-01 -1.26726711e+00 7.72141933e-01 1.19443750e+00
9.47543025e-01 3.67616564e-01 3.98387432e-01 -3.93128693e-01
-1.19999766e+00 -1.46474457e+00 6.38953090e-01 -9.66019213e-01
7.84642696e-01 -8.46865416e-01 -9.80146646e-01 8.72653365e-01
5.40738225e-01 4.24696833e-01 5.55782795e-01 4.08030540e-01
-1.03022635e+00 8.29714760e-02 -6.70883000e-01 6.85376465e-01
1.25188434e+00 -6.81014955e-01 -8.92531991e-01 6.18582129e-01
8.50408494e-01 -4.16930676e-01 -5.12977362e-01 1.84328526e-01
3.27528119e-01 -8.07048261e-01 1.21478045e+00 -7.34026015e-01
6.52682483e-01 -3.63359153e-01 -5.01439571e-01 -8.95917535e-01
-5.19036472e-01 -4.95406151e-01 -7.04684481e-02 1.28613508e+00
5.64377427e-01 -3.83010179e-01 4.96424705e-01 -5.82125187e-02
-2.89536476e-01 -6.61624491e-01 -6.00313842e-01 -1.17578745e+00
1.17481522e-01 -7.91106343e-01 2.14460209e-01 7.50630736e-01
-2.42509246e-01 7.45467305e-01 -3.47935051e-01 1.17830284e-01
6.81196451e-01 7.60156140e-02 1.05635047e+00 -8.36580098e-01
-5.23880482e-01 -4.04973865e-01 -4.40353960e-01 -1.12800229e+00
5.59435375e-02 -1.41490352e+00 3.31661314e-01 -1.74139547e+00
5.38206398e-01 -2.27341071e-01 -4.61331606e-01 1.00042343e+00
-2.17978418e-01 9.07299101e-01 5.98154724e-01 2.85516143e-01
-9.91839826e-01 4.31855857e-01 9.56294417e-01 -4.67425317e-01
5.43225110e-02 -5.75000823e-01 -7.90396988e-01 6.59552574e-01
7.13737428e-01 -3.97202879e-01 -5.29050112e-01 -6.58088684e-01
1.92895964e-01 -3.88656676e-01 7.85627425e-01 -1.08944929e+00
-1.25580043e-01 2.72373552e-03 3.55515748e-01 -6.40896916e-01
1.90129831e-01 -2.65285283e-01 -3.55036676e-01 3.78992051e-01
-3.47280592e-01 -1.60649687e-01 4.66642380e-01 6.31806910e-01
-2.15450272e-01 -1.97643161e-01 9.18006539e-01 -1.90107584e-01
-1.38696647e+00 4.22287852e-01 -4.06253561e-02 6.64081097e-01
8.56914878e-01 -5.48508912e-02 -6.12407804e-01 -2.64767915e-01
-5.13131857e-01 3.47715050e-01 4.86945122e-01 8.06473255e-01
5.59624314e-01 -1.31285310e+00 -9.74217653e-01 8.95977095e-02
7.95188189e-01 -2.91359182e-02 7.75419027e-02 6.27642155e-01
-4.35125679e-01 3.02549034e-01 -1.83480024e-01 -1.09028542e+00
-9.54726398e-01 7.40418851e-01 5.79544120e-02 -1.08035475e-01
-1.06746173e+00 1.28622162e+00 8.17621291e-01 -1.49890989e-01
2.84853905e-01 -3.54871631e-01 3.00478727e-01 1.26408964e-01
7.63940871e-01 -1.64789334e-01 -1.45172879e-01 -4.67042923e-01
-4.63440031e-01 5.78100979e-01 -8.58342871e-02 -3.33271682e-01
1.09204102e+00 3.75901647e-02 2.57735610e-01 4.54017520e-01
1.16522050e+00 -2.72354424e-01 -1.55687988e+00 -3.89922649e-01
7.67530920e-03 -2.86456972e-01 -9.47407857e-02 -9.21515226e-01
-8.36192191e-01 1.18946195e+00 4.68969822e-01 -1.62995622e-01
8.60019386e-01 5.75354636e-01 5.03874421e-01 5.14994144e-01
3.50505769e-01 -8.44303727e-01 8.23329389e-01 5.40536940e-01
1.18237019e+00 -1.55173516e+00 -1.35805085e-01 -2.46342555e-01
-9.09907997e-01 6.74871087e-01 8.28830540e-01 -1.72081068e-01
1.78788170e-01 1.73006058e-01 1.14931665e-01 -1.50416315e-01
-8.65022063e-01 -7.41547525e-01 5.80342889e-01 7.66041756e-01
3.89632672e-01 -1.67612676e-02 3.10457140e-01 5.42876303e-01
3.75034660e-02 -1.05184071e-01 2.52809674e-01 6.48346603e-01
-6.03836417e-01 -5.92585683e-01 -1.63283944e-01 1.63972512e-01
-1.36541307e-01 -4.84747767e-01 -2.55437076e-01 1.00187802e+00
1.41973332e-01 8.79873037e-01 2.58694589e-01 -1.72479510e-01
1.95287272e-01 1.31692290e-01 4.46711987e-01 -1.06383920e+00
-2.77824700e-01 -1.21162958e-01 1.45540729e-01 -7.30823398e-01
-8.13232288e-02 -2.95564294e-01 -1.26458323e+00 1.17055722e-01
-2.11336091e-01 -2.42175788e-01 4.06449258e-01 7.07188368e-01
5.15510321e-01 4.66833025e-01 4.04405892e-02 -8.40075493e-01
-4.25036550e-01 -1.05204833e+00 -2.70152241e-01 7.59582102e-01
2.32729614e-01 -6.97530091e-01 -1.63804665e-01 4.10150409e-01] | [10.381685256958008, 1.6775542497634888] |
d72c96b6-1c92-4899-a509-304ce7832336 | iterative-scene-graph-generation-with | 2211.16636 | null | https://arxiv.org/abs/2211.16636v1 | https://arxiv.org/pdf/2211.16636v1.pdf | Iterative Scene Graph Generation with Generative Transformers | Scene graphs provide a rich, structured representation of a scene by encoding the entities (objects) and their spatial relationships in a graphical format. This representation has proven useful in several tasks, such as question answering, captioning, and even object detection, to name a few. Current approaches take a generation-by-classification approach where the scene graph is generated through labeling of all possible edges between objects in a scene, which adds computational overhead to the approach. This work introduces a generative transformer-based approach to generating scene graphs beyond link prediction. Using two transformer-based components, we first sample a possible scene graph structure from detected objects and their visual features. We then perform predicate classification on the sampled edges to generate the final scene graph. This approach allows us to efficiently generate scene graphs from images with minimal inference overhead. Extensive experiments on the Visual Genome dataset demonstrate the efficiency of the proposed approach. Without bells and whistles, we obtain, on average, 20.7% mean recall (mR@100) across different settings for scene graph generation (SGG), outperforming state-of-the-art SGG approaches while offering competitive performance to unbiased SGG approaches. | ['Sathyanarayanan N. Aakur', 'Sanjoy Kundu'] | 2022-11-30 | null | null | null | null | ['scene-graph-generation'] | ['computer-vision'] | [ 6.17430389e-01 4.16257262e-01 7.58473203e-02 -4.63864118e-01
-6.91477180e-01 -6.50392532e-01 9.05740499e-01 5.88488340e-01
1.41470125e-02 4.84353632e-01 1.40320882e-01 -3.71919513e-01
2.68511958e-02 -1.11797690e+00 -9.93949175e-01 -4.05639142e-01
-1.80345789e-01 6.01151466e-01 6.31337702e-01 1.77129999e-01
2.18546465e-01 3.95723879e-01 -1.77161407e+00 5.11145711e-01
8.57330084e-01 9.05148387e-01 5.27016699e-01 8.08246732e-01
-5.14388859e-01 1.13155651e+00 -6.15016997e-01 -6.70139194e-01
-5.01622148e-02 -4.29809242e-01 -8.55429828e-01 5.89471281e-01
7.36403525e-01 -1.59079835e-01 -3.12219650e-01 8.79874468e-01
7.19045997e-02 1.60518467e-01 5.53015411e-01 -1.34292436e+00
-5.62620759e-01 6.28457427e-01 -5.04302740e-01 -7.86460787e-02
7.22133338e-01 -1.09162465e-01 1.37659359e+00 -1.05787361e+00
9.23012614e-01 1.54154229e+00 1.36997491e-01 2.16843411e-01
-1.33994544e+00 -2.24936739e-01 4.19434756e-01 2.63309240e-01
-1.35140479e+00 -1.80422395e-01 8.92973185e-01 -4.50768173e-01
7.64622748e-01 3.82796794e-01 7.60293067e-01 5.54418862e-01
-2.85203636e-01 9.92245197e-01 9.17820752e-01 -6.00480914e-01
3.22859317e-01 2.25324780e-01 3.68780196e-02 9.74765182e-01
3.80670518e-01 -5.47580004e-01 -5.00095308e-01 -2.50895292e-01
6.43969893e-01 -1.14843421e-01 -1.96988598e-01 -7.00873077e-01
-1.25968993e+00 7.39659250e-01 8.80016863e-01 -1.83686957e-01
-3.92245173e-01 2.30931684e-01 1.72310770e-02 -2.19135508e-01
4.03345197e-01 2.69728839e-01 1.87505528e-01 3.66924465e-01
-6.84368491e-01 4.95775878e-01 7.34692216e-01 1.31231475e+00
8.52965951e-01 -8.32829252e-02 -5.48136830e-01 7.52717972e-01
3.07998151e-01 3.84200454e-01 -1.60345614e-01 -6.51685238e-01
6.91784918e-01 9.09436464e-01 2.18317285e-02 -1.33124006e+00
-1.34407207e-01 -1.59607857e-01 -5.64440429e-01 -2.34580234e-01
3.04799736e-01 1.65551513e-01 -1.20094728e+00 1.48539710e+00
7.57863164e-01 2.06218511e-01 -1.44453451e-01 5.95454276e-01
1.01832807e+00 8.23778391e-01 2.11469665e-01 2.13659942e-01
1.56284606e+00 -1.03368974e+00 -4.25500333e-01 -5.31921566e-01
4.90175635e-01 -6.60901785e-01 1.10515058e+00 1.42320141e-01
-7.76587427e-01 -3.65638882e-01 -8.93716931e-01 -5.91321215e-02
-4.61858839e-01 1.77831352e-01 8.51776659e-01 4.79990184e-01
-1.00830925e+00 1.80676937e-01 -5.11749923e-01 -5.87494254e-01
6.44790947e-01 7.15037249e-03 -2.71699846e-01 -3.06521565e-01
-5.88948727e-01 4.35127109e-01 7.80501604e-01 -1.07468583e-01
-8.85710835e-01 -4.22387242e-01 -1.19688404e+00 1.99160259e-02
6.53268874e-01 -7.71807551e-01 1.13446629e+00 -4.26680684e-01
-8.33041072e-01 7.86056936e-01 -4.91576821e-01 -5.42255461e-01
2.23091900e-01 -1.29323244e-01 -2.10266978e-01 4.15099889e-01
2.74468064e-01 8.84647369e-01 6.36033714e-01 -1.58529401e+00
-6.67440832e-01 -9.66485366e-02 3.66806746e-01 4.60144162e-01
1.73639506e-01 -6.17771558e-02 -8.15175712e-01 -4.01529729e-01
3.97276282e-01 -8.94137621e-01 -3.37115377e-01 1.34945154e-01
-9.62731421e-01 -2.48251051e-01 9.17206466e-01 -4.71542478e-01
1.07725465e+00 -1.92762494e+00 -2.76570529e-01 3.50568146e-01
2.85077840e-01 3.67199369e-02 -1.47688702e-01 6.58333182e-01
3.38655114e-02 4.28144559e-02 -2.73580968e-01 -4.69081610e-01
-4.13738303e-02 6.97674677e-02 -4.76222903e-01 1.90858215e-01
3.54788482e-01 1.06948590e+00 -1.13030565e+00 -7.56860912e-01
5.51630914e-01 2.83594310e-01 -6.67090535e-01 3.24367374e-01
-7.86859810e-01 1.37405381e-01 -5.70224166e-01 6.19604528e-01
5.31883240e-01 -6.50314629e-01 3.61252993e-01 -2.44117931e-01
3.42710435e-01 4.67026979e-01 -1.26542914e+00 1.45062923e+00
-3.05636674e-01 7.50873148e-01 -4.94005561e-01 -8.89738321e-01
1.00670159e+00 -1.21851675e-01 2.43658628e-02 -5.22016525e-01
-1.59701675e-01 -1.73271284e-01 -2.04035670e-01 -2.15066031e-01
6.86147749e-01 2.25165442e-01 -2.29757890e-01 2.91613221e-01
2.31605582e-02 -2.27833852e-01 4.89706248e-01 8.68398309e-01
9.98266041e-01 2.20812231e-01 4.89355505e-01 1.37486150e-02
2.54133940e-01 2.80686885e-01 2.43132794e-03 8.87963474e-01
4.54503149e-01 6.96865797e-01 6.70150757e-01 -2.21522734e-01
-9.00060177e-01 -1.19791615e+00 3.20389748e-01 7.69413650e-01
3.66580725e-01 -7.90663540e-01 -7.53973305e-01 -6.17632508e-01
-2.15009734e-01 9.60461736e-01 -5.68730235e-01 1.27255425e-01
-4.17732447e-01 -4.88388658e-01 1.82480440e-01 4.14934665e-01
4.62460399e-01 -1.18324161e+00 -7.02254653e-01 4.69945818e-02
-3.69243860e-01 -1.55539668e+00 -2.36800671e-01 -3.11038345e-01
-6.36049688e-01 -1.10850489e+00 -3.00928056e-01 -8.56665552e-01
1.07581544e+00 5.08602977e-01 1.35333514e+00 1.06681503e-01
-5.78568220e-01 4.60705072e-01 -3.55487257e-01 -2.85269678e-01
-3.01190674e-01 -2.21364498e-01 -5.16490221e-01 2.66327888e-01
-5.75977229e-02 -2.41833776e-01 -5.40724218e-01 6.48507103e-02
-8.12811613e-01 5.81135452e-01 4.84007984e-01 5.54248571e-01
9.54021275e-01 -7.22350436e-04 2.06841722e-01 -1.36399686e+00
4.12848949e-01 -4.15771365e-01 -7.51595199e-01 5.21191657e-01
-2.44990349e-01 1.57905638e-01 5.68874955e-01 -2.47539774e-01
-1.15934563e+00 2.10576966e-01 2.88796037e-01 -3.32880080e-01
-2.59861380e-01 4.37776744e-01 -3.20940465e-01 1.45630881e-01
6.00962877e-01 4.37080681e-01 -4.56936896e-01 -2.80841142e-01
8.46547246e-01 2.02815458e-01 6.96476340e-01 -5.51914096e-01
9.14754212e-01 5.54341078e-01 1.97624147e-01 -9.27895546e-01
-1.07093406e+00 -5.44649065e-01 -4.72787291e-01 -3.41472805e-01
8.38395238e-01 -8.77629757e-01 -4.77270573e-01 1.00224800e-01
-1.24000883e+00 -2.63645351e-01 -2.59744972e-01 1.12470537e-01
-4.21879917e-01 2.76590824e-01 -1.57786995e-01 -9.06947672e-01
-7.04099536e-02 -7.44817376e-01 1.53300416e+00 2.40949675e-01
-5.11809550e-02 -8.30952227e-01 -3.25477391e-01 3.37730438e-01
1.07439449e-02 6.55807912e-01 1.07988656e+00 -5.27598023e-01
-1.24521661e+00 -2.15007886e-01 -5.94194233e-01 -1.96757749e-01
1.68374881e-01 -1.15040056e-01 -8.94711435e-01 2.94190552e-02
-6.98503435e-01 -6.33006096e-02 6.45680964e-01 2.36013532e-01
1.42011404e+00 -5.56480825e-01 -6.54457390e-01 3.80524457e-01
1.61766613e+00 2.17892632e-01 6.83794498e-01 7.46334791e-02
1.10296559e+00 7.41301894e-01 7.62068093e-01 4.50108767e-01
6.70125663e-01 7.71761417e-01 5.93883157e-01 -2.24357009e-01
-4.13721323e-01 -8.60748649e-01 -1.06318884e-01 2.32384935e-01
3.63525212e-01 -6.75638020e-01 -8.95053387e-01 8.19005668e-01
-1.81967604e+00 -9.60319817e-01 -5.29626966e-01 2.10867763e+00
3.82804841e-01 8.20869505e-02 7.83623233e-02 3.07181291e-02
8.61449182e-01 2.87896454e-01 -3.26577216e-01 -2.48574749e-01
1.52723957e-02 9.00907442e-02 3.73106241e-01 4.76590782e-01
-1.10027277e+00 1.18469214e+00 5.89401484e+00 7.00137496e-01
-6.67326450e-01 -3.04284632e-01 7.10822284e-01 2.87001729e-01
-4.69773233e-01 3.32213402e-01 -8.56554508e-01 2.06962749e-01
4.58789080e-01 -3.70510072e-01 3.79011959e-01 1.05522561e+00
2.15676408e-02 -3.36774319e-01 -9.95116413e-01 1.05070353e+00
3.29678208e-01 -1.46207464e+00 5.69296062e-01 -1.03930116e-01
6.99409366e-01 -2.94289500e-01 -2.63255656e-01 -2.38703359e-02
5.29989660e-01 -8.49894285e-01 1.00188196e+00 2.72252530e-01
6.78733110e-01 -5.14373422e-01 2.76111454e-01 3.39057326e-01
-1.56404126e+00 2.25964785e-01 -3.26139122e-01 7.91320428e-02
4.15560305e-01 6.83422446e-01 -1.62443268e+00 7.73102045e-01
4.85485494e-01 3.97014946e-01 -9.71415520e-01 1.34751904e+00
-4.81235713e-01 6.60628021e-01 -3.28566462e-01 -2.97194242e-01
1.19767517e-01 -1.50391743e-01 4.89091933e-01 1.32559121e+00
3.39493960e-01 -4.77761403e-02 3.62347126e-01 1.00541794e+00
-1.50085539e-01 8.81911293e-02 -9.13677394e-01 -1.09451478e-02
7.09691048e-01 1.37918997e+00 -1.31176865e+00 -5.82181692e-01
-2.31294498e-01 1.04479396e+00 3.55481148e-01 3.28680307e-01
-8.81602705e-01 -4.74181294e-01 1.35150477e-01 3.75066489e-01
4.80301499e-01 -1.68575943e-01 -5.84799424e-02 -8.80256414e-01
2.29154333e-01 -6.24489367e-01 4.93611485e-01 -1.15656495e+00
-8.99802327e-01 6.73643231e-01 4.67685491e-01 -1.19498944e+00
-4.17394131e-01 -3.68069381e-01 -4.37269330e-01 6.11957490e-01
-1.17850792e+00 -1.31952488e+00 -7.72965133e-01 4.15426075e-01
5.89259148e-01 3.90609741e-01 5.91683805e-01 -2.21642219e-02
-2.27670491e-01 1.24862045e-01 -3.67887884e-01 2.24911049e-01
1.57522514e-01 -1.47818363e+00 8.80211532e-01 1.17120433e+00
7.87468016e-01 4.71338660e-01 6.61877990e-01 -7.14967251e-01
-1.23867559e+00 -1.54180980e+00 1.03822029e+00 -4.18558210e-01
4.26976353e-01 -8.61293495e-01 -7.18657732e-01 7.41968691e-01
9.37967971e-02 7.80659169e-02 2.60952264e-01 -7.96141848e-02
-4.64774549e-01 4.96956147e-02 -9.65610981e-01 9.81997848e-01
1.29672682e+00 -3.96924078e-01 -3.16965014e-01 5.53984284e-01
8.10507953e-01 -5.54431975e-01 -3.15019965e-01 2.07538515e-01
1.92470342e-01 -9.30262864e-01 1.04377615e+00 -4.15201217e-01
3.81518483e-01 -7.05422044e-01 -1.73916459e-01 -1.10772860e+00
-1.84609741e-01 -4.94614989e-01 -1.67987451e-01 1.28365970e+00
4.85180974e-01 -4.26039785e-01 7.92920530e-01 4.85293776e-01
-4.69391160e-02 -6.29765987e-01 -3.56674463e-01 -6.29013598e-01
-8.50210130e-01 -5.77143013e-01 6.66066229e-01 5.83684385e-01
-3.76973301e-01 6.41346455e-01 -2.37580210e-01 2.88020253e-01
8.56403232e-01 5.73248684e-01 1.05025017e+00 -1.14504755e+00
-3.80422056e-01 -1.09665297e-01 -6.13524854e-01 -1.28920484e+00
-1.45437330e-01 -9.52034056e-01 1.35993421e-01 -2.28410387e+00
4.04736370e-01 -4.53830749e-01 3.04963768e-01 5.12801945e-01
-4.16524231e-01 2.62128681e-01 3.71446162e-01 -1.41210347e-01
-1.01995480e+00 2.97154158e-01 1.21181417e+00 -1.02245584e-01
-1.01187654e-01 -3.01621705e-01 -7.98945427e-01 7.28553236e-01
5.86775124e-01 -3.43201518e-01 -7.97452390e-01 -3.62502217e-01
9.15474519e-02 3.81842740e-02 7.08276570e-01 -9.08951998e-01
1.55191764e-01 -6.70386702e-02 2.21019760e-01 -8.34347844e-01
5.47790408e-01 -5.51069617e-01 4.11151320e-01 1.90653399e-01
-1.89076900e-01 -1.40316063e-03 1.48810059e-01 7.99487591e-01
-2.28034362e-01 -5.00960015e-02 4.66216415e-01 -1.65282518e-01
-9.09813285e-01 2.34265238e-01 1.49373813e-02 3.83486077e-02
1.14996517e+00 -3.14122468e-01 -5.33209741e-01 -6.16828442e-01
-5.53963900e-01 8.55175480e-02 4.16235358e-01 3.92776936e-01
8.66033852e-01 -1.29773808e+00 -5.59496760e-01 -2.17855182e-02
4.65585887e-01 3.38077813e-01 1.10769086e-01 4.05059308e-01
-7.21129477e-01 4.64827299e-01 3.02478492e-01 -8.61176193e-01
-1.52010596e+00 5.37972689e-01 -1.01663671e-01 -2.45040804e-01
-8.16973448e-01 8.71121585e-01 7.39007950e-01 -1.22608826e-01
5.70255704e-02 -3.64457190e-01 -1.92659795e-01 -1.12937607e-01
3.61379892e-01 1.85927451e-01 -7.19114323e-04 -6.22747421e-01
-3.41306031e-01 3.98277879e-01 -1.18313869e-02 -1.09866343e-01
9.13947880e-01 4.29880545e-02 2.30547450e-02 3.36224645e-01
8.99765849e-01 7.69920722e-02 -9.91806388e-01 -1.77067772e-01
8.69890973e-02 -6.82031333e-01 -3.07160616e-01 -6.50775075e-01
-7.15596199e-01 7.07961082e-01 1.89564973e-01 5.75537980e-01
1.10598278e+00 5.86121619e-01 3.90456170e-01 3.47073078e-01
5.75522661e-01 -3.76907170e-01 2.61337787e-01 9.53661874e-02
8.29513967e-01 -1.00379229e+00 5.00929123e-03 -1.23006356e+00
-8.20381641e-01 6.29886627e-01 5.28686345e-01 -1.65525556e-01
2.88126737e-01 -4.73353751e-02 -3.29349488e-01 -4.44538951e-01
-8.77864838e-01 -5.39894640e-01 6.10927045e-01 6.47083938e-01
2.44138360e-01 3.02600235e-01 -4.73039038e-02 -2.81287637e-02
-3.00783157e-01 -4.12680387e-01 4.16246176e-01 7.76374340e-01
-4.81310338e-01 -1.00614786e+00 -2.93619633e-01 7.35092342e-01
-1.90861598e-01 -3.15041631e-01 -6.09359145e-01 8.19792867e-01
-1.29455939e-01 1.06116557e+00 1.98468223e-01 -1.62516221e-01
3.77005488e-01 -1.58918023e-01 4.87162530e-01 -9.28686917e-01
-3.09130494e-02 -2.92746536e-02 4.72498119e-01 -6.59851670e-01
-4.15210873e-01 -6.23737752e-01 -1.19222879e+00 1.88887656e-01
-3.65787685e-01 3.50161344e-02 5.64011574e-01 6.54000401e-01
5.38863480e-01 4.98163998e-01 5.36468387e-01 -6.84134305e-01
2.23640710e-01 -6.85688972e-01 -3.40916067e-01 6.59392774e-01
4.48749103e-02 -7.46154428e-01 -3.90565619e-02 3.79690140e-01] | [10.378231048583984, 1.6078689098358154] |
7d450006-152c-4426-a86e-63ce780aaf57 | on-the-impact-of-object-and-sub-component | 1911.08216 | null | https://arxiv.org/abs/1911.08216v1 | https://arxiv.org/pdf/1911.08216v1.pdf | On the Impact of Object and Sub-component Level Segmentation Strategies for Supervised Anomaly Detection within X-ray Security Imagery | X-ray security screening is in widespread use to maintain transportation security against a wide range of potential threat profiles. Of particular interest is the recent focus on the use of automated screening approaches, including the potential anomaly detection as a methodology for concealment detection within complex electronic items. Here we address this problem considering varying segmentation strategies to enable the use of both object level and sub-component level anomaly detection via the use of secondary convolutional neural network (CNN) architectures. Relative performance is evaluated over an extensive dataset of exemplar cluttered X-ray imagery, with a focus on consumer electronics items. We find that sub-component level segmentation produces marginally superior performance in the secondary anomaly detection via classification stage, with true positive of ~98% of anomalies, with a ~3% false positive. | ['Yona Falinie A. Gaus', 'Neelanjan Bhowmik', 'Toby P. Breckon', 'Samet Akcay', 'Jack W. Barker'] | 2019-11-19 | null | null | null | null | ['supervised-anomaly-detection'] | ['computer-vision'] | [ 7.95043826e-01 -6.54122531e-02 1.04460798e-01 4.22719568e-02
-1.12513661e+00 -6.50017440e-01 5.84382832e-01 9.31358814e-01
-3.51490557e-01 1.95750371e-01 -3.41391772e-01 -8.90237033e-01
-2.22859859e-01 -7.42841184e-01 -8.27341855e-01 -5.16259015e-01
-2.94482589e-01 1.07582221e-02 3.07604074e-01 -1.94548398e-01
4.81658459e-01 1.02434409e+00 -1.08186615e+00 6.12275779e-01
2.56840229e-01 1.66571259e+00 -5.88696837e-01 8.19707096e-01
3.43550384e-01 1.96013361e-01 -9.06582594e-01 -7.09125936e-01
4.42663997e-01 -1.41261831e-01 -8.37410271e-01 2.06272274e-01
4.90571320e-01 -4.62033182e-01 5.20050079e-02 1.11969769e+00
3.86678040e-01 -1.18817441e-01 6.06138170e-01 -8.75337362e-01
-1.75676450e-01 2.28936803e-02 -6.15917444e-01 7.19104052e-01
7.00889230e-01 5.35606503e-01 6.57415569e-01 -3.83192807e-01
2.85687894e-01 7.67517209e-01 1.00297904e+00 3.24033260e-01
-1.01289952e+00 -5.70661306e-01 -1.57871827e-01 -2.14037165e-01
-1.17585933e+00 2.25734204e-01 5.48858762e-01 -4.72385347e-01
1.08380961e+00 6.05303228e-01 5.15240371e-01 1.03582990e+00
8.91627431e-01 5.70396245e-01 1.02999771e+00 -1.24857552e-01
3.08234453e-01 -5.90053871e-02 5.13236403e-01 7.45749533e-01
7.86880136e-01 1.00473613e-01 -2.13028923e-01 -6.35217905e-01
4.81694221e-01 -6.25468716e-02 5.79181612e-02 7.97931179e-02
-6.11809134e-01 8.58398139e-01 3.37227643e-01 6.52863085e-02
-4.49931264e-01 2.25483049e-02 5.60167909e-01 7.54846856e-02
5.13338983e-01 1.01495242e+00 -3.30938309e-01 9.78958160e-02
-5.99918127e-01 3.35986912e-01 3.98649812e-01 6.21303320e-01
1.58405274e-01 2.57268906e-01 -2.19680205e-01 4.00743514e-01
1.83089346e-01 -3.39727625e-02 -1.88962743e-02 -1.20822147e-01
4.77967948e-01 6.69058144e-01 -2.40009185e-02 -8.44678640e-01
-5.20322502e-01 -6.18010700e-01 -6.06206894e-01 4.63530809e-01
4.19557631e-01 -3.55767995e-01 -1.45039535e+00 9.13388431e-01
3.34414184e-01 -2.00795501e-01 -3.10207158e-01 4.38742906e-01
2.99315125e-01 4.09871072e-01 3.64229321e-01 1.16603754e-01
1.58748770e+00 -4.97175418e-02 -4.37480301e-01 -1.79790237e-04
6.12227380e-01 -6.65065110e-01 5.76872408e-01 9.07315493e-01
-1.17671645e+00 -3.21287483e-01 -1.59893310e+00 5.95301628e-01
-7.61506140e-01 -9.00987685e-02 5.59935272e-01 1.40374124e+00
-4.42842990e-01 5.16749263e-01 -8.85519505e-01 -2.89904535e-01
8.90131772e-01 8.49668860e-01 -2.55083919e-01 1.41135752e-01
-8.24243844e-01 8.93822074e-01 6.20177537e-02 8.67205709e-02
-7.56455481e-01 -8.47305477e-01 -9.41689193e-01 -8.39329511e-02
2.60149211e-01 8.26919302e-02 1.16225791e+00 -4.39570516e-01
-9.76949275e-01 5.84630489e-01 5.47749639e-01 -6.67006314e-01
5.42106748e-01 -5.51849008e-01 -6.24058604e-01 3.16661775e-01
-1.39070630e-01 2.32393533e-01 4.28976476e-01 -1.23067296e+00
-6.36683881e-01 -3.98079872e-01 -3.49052474e-02 -2.60451972e-01
-2.00862214e-01 5.47970653e-01 -1.21697851e-01 -7.50952423e-01
1.02909766e-01 -6.88587427e-01 -2.24513665e-01 -6.81952015e-02
-6.28220558e-01 -2.86650416e-02 1.04894996e+00 -1.03940010e+00
1.06116784e+00 -1.80775201e+00 -6.55871451e-01 5.85184336e-01
-7.87897557e-02 4.24979597e-01 2.87783772e-01 1.73050120e-01
-2.72348672e-01 1.50884539e-01 -7.28858888e-01 1.94682494e-01
-4.19134408e-01 -4.26827490e-01 -2.17153504e-01 8.43207061e-01
8.68369579e-01 6.48437381e-01 -5.92252314e-01 1.58098802e-01
2.71280468e-01 1.68296427e-01 -3.21719974e-01 -1.68524012e-02
-1.38247043e-01 1.59871578e-01 -5.12256145e-01 1.48335814e+00
7.55017221e-01 3.01324409e-02 -1.98637456e-01 -5.61427400e-02
6.49698777e-03 5.01806363e-02 -1.12119246e+00 1.08824444e+00
2.24537879e-01 5.53360581e-01 3.51361901e-01 -5.16561389e-01
4.37256485e-01 3.44103426e-02 5.78087211e-01 -8.41961622e-01
4.57524925e-01 2.79315040e-02 2.94142067e-01 -2.64476627e-01
7.20100403e-01 -1.99484721e-01 -3.60922426e-01 3.37702632e-01
-6.41066954e-02 1.03319943e-01 -3.43317389e-01 1.59041494e-01
1.53058803e+00 -2.58344561e-01 1.29009709e-01 -4.05808508e-01
2.78464437e-01 5.96414387e-01 7.22962152e-03 8.05755258e-01
-2.81098336e-01 9.49858963e-01 5.33524096e-01 -3.63308519e-01
-9.03481781e-01 -9.85760629e-01 -7.08906651e-01 6.11366272e-01
3.27610701e-01 9.61638913e-02 -9.32382047e-01 -9.98941183e-01
2.63911746e-02 6.62962019e-01 -6.89670444e-01 -4.25337970e-01
-6.59521461e-01 -1.42383957e+00 8.48399460e-01 7.45553553e-01
5.46181083e-01 -9.29524779e-01 -8.85527790e-01 1.86834499e-01
6.42848849e-01 -9.90842283e-01 2.06686869e-01 5.70261121e-01
-6.18749201e-01 -1.42210519e+00 -3.58736157e-01 -1.13364473e-01
5.33856928e-01 -2.60116130e-01 8.45741630e-01 3.85351419e-01
-9.90818083e-01 6.74867630e-01 -2.13307038e-01 -6.11317933e-01
-6.79478347e-01 -2.52402216e-01 -7.00371340e-02 -1.47654250e-01
4.25549090e-01 4.61555302e-01 -7.69830227e-01 2.78288692e-01
-1.20896399e+00 -7.61920691e-01 3.53699893e-01 5.10266483e-01
3.86755347e-01 1.75501406e-01 1.85558155e-01 -9.59492624e-01
7.54875362e-01 -8.72553289e-01 -6.94546580e-01 -2.30947584e-02
-3.86764795e-01 -5.40904880e-01 3.41339320e-01 -3.28974605e-01
-9.49863672e-01 -5.21600284e-02 -6.65126741e-01 -2.75582243e-02
-6.60949230e-01 2.02204615e-01 -1.04111962e-01 -5.52320480e-01
7.25480795e-01 -1.69007197e-01 -1.54637754e-01 -4.61914986e-01
-3.86103302e-01 3.97277117e-01 4.40978318e-01 -4.08907771e-01
5.67223191e-01 5.04119754e-01 3.25395912e-01 -9.57622766e-01
-4.28715050e-01 -4.15458590e-01 -3.57294619e-01 -1.30361393e-01
1.18016648e+00 -5.18869638e-01 -4.74615604e-01 6.75791562e-01
-8.62092197e-01 -1.21384799e-01 8.67065117e-02 3.57728302e-01
3.19741555e-02 5.85614622e-01 -8.74573410e-01 -8.91736567e-01
-3.19050282e-01 -1.27983558e+00 1.31464016e+00 -1.04309782e-01
-4.70923901e-01 -5.25566280e-01 -3.80448431e-01 1.79413497e-01
2.41708681e-01 7.61729658e-01 1.25189841e+00 -1.27284253e+00
-5.79840481e-01 -8.65159988e-01 -8.58602896e-02 3.87157023e-01
8.91195238e-02 -1.25693798e-01 -9.54085529e-01 -5.27377546e-01
2.27089077e-01 -4.15770784e-02 5.13066649e-01 3.45882475e-01
1.47479427e+00 1.07022814e-01 -3.85619015e-01 3.97111595e-01
1.58690166e+00 7.50092924e-01 8.19219291e-01 4.91557747e-01
6.55781806e-01 5.64746141e-01 7.23201513e-01 2.09612444e-01
-6.39132380e-01 4.29655254e-01 6.37546599e-01 -2.04484209e-01
9.99103189e-02 3.74469697e-01 -9.73554105e-02 -3.77407908e-01
2.72553712e-01 -4.44979519e-01 -1.28085315e+00 5.22991300e-01
-1.03293753e+00 -6.63876951e-01 -3.18690926e-01 2.18271375e+00
1.19117349e-01 6.01797163e-01 3.04358482e-01 3.86918336e-01
8.49455416e-01 -3.25628459e-01 -5.36218464e-01 -5.04578292e-01
1.77253708e-01 5.22806883e-01 8.08121204e-01 3.30588296e-02
-1.42418504e+00 1.18181594e-01 7.27119398e+00 7.92936683e-01
-9.75079894e-01 -1.55023411e-01 1.24235320e+00 -1.04584031e-01
1.66475445e-01 -5.60266495e-01 -7.79400468e-01 6.00109458e-01
8.98919225e-01 6.80809855e-01 -4.70854700e-01 5.72508216e-01
-4.30202007e-01 -4.21394527e-01 -9.80773151e-01 3.39585871e-01
2.06679821e-01 -1.50327885e+00 1.05316743e-01 1.33809507e-01
4.71745551e-01 -8.50585327e-02 3.04554343e-01 -1.96074203e-01
2.22946908e-02 -1.02051640e+00 7.61350334e-01 -1.03583429e-02
7.54888296e-01 -1.14162838e+00 7.23847985e-01 -1.74433619e-01
-9.63541329e-01 -2.92459130e-01 2.16633663e-01 2.13151470e-01
1.87968537e-01 -4.82107438e-02 -8.13408792e-01 5.30093074e-01
1.01492941e+00 -2.67828435e-01 -6.25063539e-01 1.21493554e+00
5.18914998e-01 7.68114328e-01 -5.39126515e-01 -9.48354602e-02
6.38278663e-01 1.54825300e-01 6.53873086e-01 1.58832812e+00
2.42815092e-01 2.81057328e-01 -1.10021986e-01 5.74238479e-01
1.56290993e-01 -3.47491264e-01 -6.43911064e-01 4.30399030e-02
-4.83372472e-02 1.05252850e+00 -1.36297297e+00 -4.70883492e-03
-6.72575593e-01 6.27539515e-01 -4.59460288e-01 6.14654794e-02
-6.35874331e-01 -5.09313822e-01 6.58176064e-01 6.22293234e-01
3.49769086e-01 1.06403455e-02 -7.00414836e-01 -3.26601505e-01
5.13043180e-02 -9.88658071e-01 7.72051930e-01 -4.85342592e-01
-1.31370103e+00 5.42290688e-01 3.61864030e-01 -1.22433054e+00
1.65095609e-02 -1.38643026e+00 -8.46315980e-01 7.89539933e-01
-9.15112674e-01 -1.08791244e+00 -5.31652607e-02 2.42560402e-01
6.54526234e-01 -4.07244414e-01 4.04156387e-01 1.54257536e-01
-5.68032265e-01 7.43412852e-01 -4.65732776e-02 3.96025896e-01
-1.97876260e-01 -1.24604952e+00 7.55820930e-01 1.17922449e+00
-4.47925895e-01 5.39963722e-01 6.40536189e-01 -1.21615851e+00
-1.59110987e+00 -1.25230539e+00 -1.35713339e-01 -9.16247010e-01
5.69421232e-01 -5.69328368e-01 -1.14021897e+00 5.37475705e-01
9.70140547e-02 -1.75805703e-01 7.79353976e-01 -1.88293517e-01
1.20038576e-01 4.46203560e-01 -1.85269725e+00 5.14309943e-01
5.62453449e-01 -4.25627977e-01 -2.47727886e-01 1.83891773e-01
4.06127930e-01 -6.08309448e-01 -7.83895254e-01 9.15478647e-01
1.98581606e-01 -7.46006310e-01 9.93906796e-01 -4.77411091e-01
3.00169557e-01 -2.99949497e-02 -4.35254648e-02 -4.42203283e-01
-2.99744189e-01 -5.38546383e-01 4.04264852e-02 9.50915158e-01
5.10542572e-01 -3.25665683e-01 1.21591878e+00 7.94564545e-01
-3.18873405e-01 -1.08693302e+00 -1.04885030e+00 -4.96705174e-01
9.58034843e-02 -7.62245953e-01 4.88667458e-01 6.77144289e-01
-3.88378859e-01 -1.20433688e-01 6.95366710e-02 8.01977217e-01
4.41172063e-01 -3.88687581e-01 2.18145952e-01 -1.04445016e+00
-1.33158550e-01 -3.44181627e-01 -8.08542967e-01 -2.52934545e-01
-5.38447082e-01 -4.12298143e-01 -5.55431806e-02 -9.34689105e-01
3.13541889e-02 -1.77809134e-01 -3.72463226e-01 1.63053229e-01
-1.57580942e-01 6.80559099e-01 -2.37360150e-01 -3.68823171e-01
-2.04283029e-01 -1.77449644e-01 5.25510550e-01 -2.30195269e-01
1.00183502e-01 3.37032169e-01 -7.13444054e-01 8.69923294e-01
7.14010715e-01 -5.22683561e-01 -5.17641427e-04 -2.71688819e-01
1.83959380e-01 -9.36059654e-02 6.27444208e-01 -1.37952399e+00
-6.61017299e-02 3.18006068e-01 9.82368350e-01 -1.00780439e+00
2.24794313e-01 -9.49398577e-01 1.82448002e-03 8.61078501e-01
-1.47890747e-01 5.82518637e-01 1.01240695e+00 8.01191390e-01
2.01161623e-01 -4.17240679e-01 8.89625072e-01 -1.98716566e-01
-5.13739169e-01 2.18137279e-01 -7.32503355e-01 -6.32953525e-01
1.27197945e+00 -7.88867414e-01 -3.13219950e-02 1.92484483e-01
-4.50308502e-01 -7.33522102e-02 2.19127119e-01 2.39828810e-01
7.08637118e-01 -1.04910088e+00 -2.68905759e-01 4.21488792e-01
2.15019703e-01 -2.24979490e-01 2.51452718e-02 2.65569121e-01
-8.37452590e-01 3.03949326e-01 -3.30883116e-01 -7.67834842e-01
-1.53435302e+00 5.20223320e-01 4.62587535e-01 -1.19297668e-01
-5.66700876e-01 9.18574631e-01 -1.92870535e-02 3.64337415e-02
1.21635087e-01 -3.42918783e-01 -3.20088193e-02 -1.96108252e-01
4.12615895e-01 7.15246499e-01 6.93559766e-01 -4.45696294e-01
-4.13577974e-01 3.09124053e-01 -4.05006319e-01 -4.05726843e-02
1.09033513e+00 4.42799002e-01 1.91230834e-01 -1.57585457e-01
1.10279560e+00 -5.21601290e-02 -1.10443556e+00 3.04240733e-01
3.31897736e-01 -2.85403997e-01 1.36028558e-01 -1.11950243e+00
-7.56878912e-01 7.31117547e-01 1.03498471e+00 6.16589248e-01
1.11504054e+00 6.27968907e-02 7.41918743e-01 4.45804268e-01
1.63775817e-01 -8.85717094e-01 1.86667275e-02 8.44517816e-03
5.47875226e-01 -1.06004596e+00 5.84753454e-01 -5.11534035e-01
-4.88805652e-01 9.21866775e-01 8.73824596e-01 -2.26951361e-01
5.05061984e-01 3.83332312e-01 -1.91731781e-01 -9.76214826e-01
-6.65656328e-02 2.54563212e-01 4.02737617e-01 5.70994437e-01
1.66560590e-01 -1.86750442e-01 3.95492315e-02 5.84973752e-01
3.80808443e-01 -7.46268868e-01 2.87727922e-01 1.47821259e+00
-3.64168286e-01 -8.71473074e-01 -6.42897129e-01 1.00388920e+00
-1.33779919e+00 1.06531411e-01 -6.29005611e-01 1.26210332e+00
2.02909112e-01 7.74925351e-01 3.54361862e-01 -1.89789623e-01
4.58892912e-01 -2.03411564e-01 3.61407638e-01 -6.37742758e-01
-1.40499449e+00 6.12514876e-02 2.63674676e-01 -5.91983378e-01
1.75300524e-01 -7.52329946e-01 -9.78567362e-01 2.09127516e-02
-6.51096404e-01 -1.21168330e-01 6.69884980e-01 9.54669833e-01
3.19314227e-02 6.54842019e-01 2.40133509e-01 -6.14822865e-01
-4.86326098e-01 -6.86606228e-01 -4.94303703e-01 4.57442135e-01
5.05822062e-01 -5.32742858e-01 -1.36689633e-01 -3.52535933e-01] | [7.499978065490723, 1.90800940990448] |
3672a046-8f73-4ab3-86ec-ba3434d6caa9 | gpt-finre-in-context-learning-for-financial | 2306.17519 | null | https://arxiv.org/abs/2306.17519v1 | https://arxiv.org/pdf/2306.17519v1.pdf | GPT-FinRE: In-context Learning for Financial Relation Extraction using Large Language Models | Relation extraction (RE) is a crucial task in natural language processing (NLP) that aims to identify and classify relationships between entities mentioned in text. In the financial domain, relation extraction plays a vital role in extracting valuable information from financial documents, such as news articles, earnings reports, and company filings. This paper describes our solution to relation extraction on one such dataset REFinD. The dataset was released along with shared task as a part of the Fourth Workshop on Knowledge Discovery from Unstructured Data in Financial Services, co-located with SIGIR 2023. In this paper, we employed OpenAI models under the framework of in-context learning (ICL). We utilized two retrieval strategies to find top K relevant in-context learning demonstrations / examples from training data for a given test example. The first retrieval mechanism, we employed, is a learning-free dense retriever and the other system is a learning-based retriever. We were able to achieve 4th rank on the leaderboard. Our best F1-score is 0.718. | ['Ankur Parikh', 'Pawan Kumar Rajpoot'] | 2023-06-30 | null | null | null | null | ['retrieval', 'relation-extraction'] | ['methodology', 'natural-language-processing'] | [ 9.56033766e-02 4.96817559e-01 -3.23530167e-01 -2.55812436e-01
-1.12653649e+00 -6.38125420e-01 7.20855296e-01 6.53925061e-01
-6.53948963e-01 1.18365741e+00 5.22144377e-01 -4.16375875e-01
-5.85082531e-01 -7.36855209e-01 -6.78578794e-01 -3.11432034e-01
-3.00691783e-01 9.38571990e-01 1.72665507e-01 -2.74834633e-01
3.78810227e-01 6.02270961e-01 -1.32723308e+00 7.55321443e-01
5.77472746e-01 9.01779771e-01 1.01527601e-01 4.87762988e-01
-5.76044917e-01 1.31421709e+00 -5.55379272e-01 -4.86338198e-01
6.73481748e-02 -4.88175154e-02 -1.35023367e+00 -4.36085761e-01
1.02775097e-01 2.25200221e-01 -2.02784017e-01 6.43135250e-01
3.75127882e-01 7.41183758e-02 8.18982422e-01 -7.76453912e-01
-3.59398037e-01 1.02110541e+00 -4.31939870e-01 6.09771073e-01
6.20352924e-01 -7.90272534e-01 1.44830322e+00 -1.17840338e+00
1.10447621e+00 9.01082814e-01 3.23134392e-01 1.22616731e-01
-9.41250682e-01 -6.68898344e-01 -9.53743011e-02 3.29320401e-01
-1.61902726e+00 -5.17456889e-01 4.94899064e-01 -3.49893004e-01
1.65879953e+00 3.30869585e-01 6.04236126e-01 7.62976229e-01
2.46878475e-01 1.03669488e+00 7.05523968e-01 -1.00989914e+00
9.72699095e-03 5.06255329e-01 5.57995260e-01 3.67865831e-01
2.86204964e-01 5.76592200e-02 -8.77650619e-01 -3.00968438e-01
4.03332591e-01 -2.79469162e-01 -4.78604063e-02 -2.77818330e-02
-8.56782138e-01 8.09130430e-01 3.75949472e-01 6.90897584e-01
-3.49992186e-01 -4.48767304e-01 1.97027594e-01 6.37872040e-01
7.13728964e-01 8.66012871e-01 -8.33385587e-01 9.05682668e-02
-6.86722696e-01 4.79992867e-01 1.40971649e+00 9.38875437e-01
5.85131288e-01 -7.81336844e-01 -5.88758439e-02 7.99418032e-01
9.66316387e-02 -2.15227045e-02 6.28395975e-01 -2.68725663e-01
7.56806016e-01 7.52490759e-01 -6.47435263e-02 -8.58317673e-01
-4.08611596e-01 -4.16332394e-01 -4.54957336e-01 -4.78886843e-01
9.44302753e-02 -4.00578827e-01 -4.35600728e-01 1.05443668e+00
2.66613990e-01 -3.81299928e-02 4.62347895e-01 4.89476204e-01
1.29878891e+00 5.70744336e-01 9.25942734e-02 -6.53099239e-01
1.38180077e+00 -7.19962597e-01 -8.40187013e-01 -1.45434424e-01
9.52773631e-01 -9.95368540e-01 2.52999336e-01 4.95014906e-01
-7.50179410e-01 -1.45106703e-01 -9.83262300e-01 -2.00195462e-01
-8.48978817e-01 2.39475854e-02 8.17561924e-01 4.38817739e-02
-7.89992392e-01 5.67558885e-01 -2.47152388e-01 -4.44186479e-01
1.66425198e-01 3.56266975e-01 -5.49159050e-01 -6.80343285e-02
-1.43276799e+00 1.17549169e+00 7.24870205e-01 -1.88225612e-01
-3.60867590e-01 -4.38198209e-01 -7.00648963e-01 1.08780034e-01
8.90558422e-01 -3.82151514e-01 1.14618051e+00 -5.33703208e-01
-8.25052798e-01 1.01051950e+00 -1.33737057e-01 -7.70090103e-01
-7.11286766e-03 -8.13885450e-01 -7.02129483e-01 -8.66906270e-02
1.06205575e-01 2.78783083e-01 3.15527827e-01 -9.46216226e-01
-8.25388610e-01 -4.42692190e-01 -1.60122439e-01 3.62471461e-01
-2.83603787e-01 3.96586388e-01 -4.32715327e-01 -4.79427010e-01
1.36888549e-01 -5.59440315e-01 2.04590429e-02 -1.07907784e+00
-5.79711795e-01 -7.84635305e-01 7.06125081e-01 -6.30885780e-01
1.54520035e+00 -1.75242138e+00 3.34659591e-02 5.43792486e-01
2.70316303e-01 2.21446157e-01 2.39988029e-01 7.39790678e-01
-3.48903328e-01 1.67368576e-01 3.06418926e-01 1.75020292e-01
-3.12237829e-01 3.40969749e-02 -5.15856445e-01 -1.68306515e-01
1.62396893e-01 8.35065424e-01 -8.41117024e-01 -7.81820595e-01
-3.27705592e-01 2.19338298e-01 -6.05884604e-02 2.14007646e-01
-2.04892069e-01 -4.09081616e-02 -7.16941297e-01 6.48087919e-01
-1.36889011e-01 -2.60084420e-01 4.63553280e-01 -2.16688111e-01
-1.29048914e-01 7.25627482e-01 -1.14898705e+00 1.46386945e+00
-3.10432792e-01 6.43012047e-01 -4.87232596e-01 -9.22014177e-01
1.06187820e+00 6.75008118e-01 5.79708993e-01 -6.79928660e-01
2.26106757e-04 2.13794976e-01 -9.46955830e-02 -6.98304772e-01
8.62397790e-01 -6.73294812e-02 2.15162951e-02 3.97634774e-01
1.66267291e-01 1.01331420e-01 6.48189723e-01 5.37990332e-01
1.34106648e+00 1.88669279e-01 9.78092194e-01 -1.61057949e-01
4.75398242e-01 1.96592346e-01 3.59671354e-01 5.68189263e-01
3.96612972e-01 1.64130986e-01 5.77751577e-01 -6.67833507e-01
-5.58600605e-01 -3.80231142e-01 -3.32426906e-01 9.08883214e-01
-3.41199547e-01 -7.42864847e-01 -2.85592735e-01 -7.82453597e-01
2.68324260e-02 5.65793216e-01 -4.51683074e-01 2.22586006e-01
-4.93089259e-01 -5.02146244e-01 3.26678902e-01 2.50809669e-01
1.96583703e-01 -1.41873896e+00 -2.77596980e-01 3.77001733e-01
-2.30727792e-01 -1.11123431e+00 -7.06556514e-02 8.04969907e-01
-8.03314507e-01 -1.45343077e+00 -4.43374455e-01 -1.06518865e+00
2.88664907e-01 -2.09039211e-01 1.38322902e+00 -2.19842926e-01
-4.44963068e-01 -4.81008692e-03 -5.65512002e-01 -5.59732020e-01
-9.67478901e-02 6.26238048e-01 -1.55589834e-01 -3.76411766e-01
9.96573389e-01 -2.74349153e-01 -5.86246662e-02 -9.26196426e-02
-5.97792745e-01 -1.06799208e-01 8.34343076e-01 5.82339764e-01
7.27210760e-01 2.54449099e-01 7.44532824e-01 -1.45996857e+00
1.02599871e+00 -5.59114754e-01 -4.30421472e-01 5.72951496e-01
-9.75146174e-01 1.97305828e-01 5.14359847e-02 -8.88023674e-02
-1.04297948e+00 1.20708063e-01 1.57594487e-01 8.45891461e-02
-2.69487262e-01 1.19198036e+00 -5.65477870e-02 2.68443614e-01
8.83054376e-01 -1.98117152e-01 -6.10282838e-01 -5.38659036e-01
1.95762619e-01 8.02090287e-01 3.02344918e-01 -6.32678509e-01
4.01146233e-01 -2.75271237e-01 3.96168679e-02 -8.43911111e-01
-1.27686763e+00 -9.82624054e-01 -5.61050475e-01 2.41059065e-03
5.41938007e-01 -9.43941414e-01 -4.11045223e-01 -9.45171788e-02
-1.04250586e+00 5.09080663e-02 -4.24043238e-01 5.15064776e-01
-5.28932810e-02 -9.75023285e-02 -7.74955571e-01 -7.92157829e-01
-5.86274683e-01 -3.77245456e-01 8.49670768e-01 3.43683451e-01
-5.33733308e-01 -8.96268427e-01 5.24605572e-01 4.47227120e-01
-6.24204464e-02 1.52172640e-01 9.95000482e-01 -1.57235074e+00
-3.82999659e-01 -4.89845812e-01 -1.08462498e-01 -1.87490992e-02
2.41551176e-01 -3.00826341e-01 -9.31742549e-01 3.60002508e-03
-3.24042022e-01 -6.18850589e-01 8.92553508e-01 7.64589086e-02
7.82856226e-01 -2.68180668e-01 -7.18369007e-01 -2.61460319e-02
1.16425502e+00 4.18637782e-01 4.66985881e-01 5.27213931e-01
6.11647308e-01 8.95938039e-01 9.24921393e-01 2.02948019e-01
2.65040964e-01 5.33373713e-01 -2.84764975e-01 2.20063314e-01
1.97239652e-01 -1.99920774e-01 2.28348523e-02 6.20141089e-01
-2.75192350e-01 -6.57520369e-02 -1.21928191e+00 5.19225955e-01
-1.85129881e+00 -9.52887058e-01 -1.39767081e-01 2.01147294e+00
1.28631282e+00 4.93722051e-01 -1.04158618e-01 9.65322778e-02
2.94162601e-01 -1.97816908e-01 -2.28551194e-01 -1.47288516e-01
-1.47493243e-01 5.67031980e-01 2.17713922e-01 6.34672701e-01
-1.26679218e+00 1.12538695e+00 5.75188446e+00 8.46397221e-01
-4.68495578e-01 -2.18473420e-01 5.49373388e-01 -3.72874998e-02
-1.82609081e-01 5.77697530e-02 -1.27920747e+00 -1.54495880e-01
1.15640497e+00 -4.20738429e-01 -4.24053892e-02 8.32640409e-01
-2.80687898e-01 -4.35853034e-01 -1.45181251e+00 6.51656449e-01
2.02605814e-01 -1.46923280e+00 6.85353950e-02 2.24114493e-01
6.58064425e-01 2.64061280e-02 -5.13517320e-01 6.57913089e-01
3.05378497e-01 -1.09022427e+00 1.42472237e-01 6.85349882e-01
5.30285239e-01 -9.93696749e-01 1.11517584e+00 4.58266914e-01
-1.13158309e+00 5.38562052e-02 -1.38969794e-01 -1.24162361e-02
-1.71076715e-01 8.18989098e-01 -1.34261131e+00 8.01200926e-01
6.65287256e-01 8.35863948e-01 -3.71898919e-01 8.50278080e-01
-3.46598357e-01 3.96253645e-01 -1.57902211e-01 -2.53916919e-01
-5.92661835e-02 6.80355951e-02 3.47868562e-01 1.35512781e+00
-9.47178677e-02 5.22002280e-01 1.96146637e-01 1.39513180e-01
-3.44154894e-01 4.64157730e-01 -8.14568639e-01 -2.07278758e-01
2.62695640e-01 1.27733409e+00 -7.03805923e-01 -4.60185856e-01
-1.57555073e-01 4.13392812e-01 5.60969114e-01 2.93234974e-01
-1.15541674e-01 -6.11886501e-01 1.49451634e-02 2.80238856e-02
2.06580684e-01 1.91392124e-01 3.06391530e-02 -1.03106308e+00
2.04015560e-02 -8.57928514e-01 8.37359488e-01 -5.68888009e-01
-1.25792313e+00 7.90671527e-01 3.46159220e-01 -9.17724371e-01
-7.42567122e-01 -5.19222498e-01 -9.56466049e-02 9.05383646e-01
-1.47399986e+00 -8.51379633e-01 2.06278890e-01 5.81421137e-01
5.74309111e-01 -5.68260491e-01 1.13271224e+00 4.02158737e-01
-2.88867474e-01 1.86459735e-01 -5.33722639e-02 4.41681325e-01
8.48001719e-01 -1.33978546e+00 1.02788009e-01 5.54649651e-01
8.62234414e-01 9.15451109e-01 4.57087606e-01 -8.70890975e-01
-9.94496703e-01 -6.52198911e-01 1.95710564e+00 -5.12331724e-01
7.56573200e-01 -1.58473384e-02 -9.16684926e-01 8.04889619e-01
2.47196749e-01 -5.62402494e-02 1.18894374e+00 7.93956518e-01
-2.97039360e-01 -2.24624738e-01 -8.51025105e-01 1.57606080e-01
7.35966861e-01 -6.09475493e-01 -1.18289769e+00 5.83773434e-01
7.30634689e-01 -6.05727494e-01 -9.52182770e-01 3.55449080e-01
3.91680986e-01 -4.06328738e-01 9.04835045e-01 -9.78857696e-01
5.38131833e-01 3.02373488e-02 -1.22835450e-01 -1.03141582e+00
-6.72215223e-02 -4.93429154e-01 -6.83149874e-01 1.39597166e+00
9.85036314e-01 -3.65745872e-01 6.13259137e-01 5.99421918e-01
2.38895804e-01 -9.06437993e-01 -6.66379988e-01 -5.23369133e-01
-1.15528524e-01 -3.41630459e-01 2.43575826e-01 1.02821457e+00
4.70131159e-01 1.07940304e+00 -8.05959031e-02 -2.43657362e-02
2.99707681e-01 3.33787352e-01 5.25958002e-01 -1.50278401e+00
-3.33503962e-01 -1.57994896e-01 -1.03541613e-01 -6.18183792e-01
2.17608348e-01 -9.11281288e-01 -1.65711656e-01 -1.54872537e+00
2.96636105e-01 -5.04953742e-01 -5.47609031e-01 6.38875544e-01
-3.28495875e-02 -3.38889390e-01 -5.36649041e-02 4.34558809e-01
-7.11254179e-01 -7.02052861e-02 8.31632197e-01 -1.44614235e-01
-3.16208303e-01 1.94942206e-01 -8.25310290e-01 5.09092689e-01
7.16555417e-01 -7.19161272e-01 -4.50744838e-01 1.64724842e-01
6.54262900e-01 2.66728878e-01 -1.61476344e-01 -6.22199535e-01
5.67396760e-01 -2.96124513e-03 4.54924852e-01 -8.21658611e-01
4.12851684e-02 -8.36196482e-01 -5.35017997e-03 9.78119820e-02
-6.63830578e-01 -1.40162274e-01 9.81356353e-02 3.70692551e-01
-5.62197685e-01 -4.36717540e-01 -1.30833820e-01 -2.79283971e-01
-8.75528455e-01 -5.56603223e-02 -8.92748982e-02 2.76052922e-01
8.01796198e-01 3.64828348e-01 -4.13616985e-01 -8.18872079e-02
-8.94034564e-01 2.80478626e-01 -3.80026817e-01 4.46950734e-01
6.96363866e-01 -9.81151819e-01 -8.19047809e-01 -5.58576025e-02
3.25565279e-01 2.50953615e-01 -2.91253030e-01 5.11395395e-01
-2.55042434e-01 9.18062508e-01 3.79695028e-01 -9.08084139e-02
-1.54001808e+00 3.94502014e-01 -9.16709304e-02 -1.11793673e+00
-4.69261795e-01 9.93813217e-01 -3.68364275e-01 -2.33363882e-01
4.83403236e-01 -2.02519953e-01 -9.02217567e-01 5.33332050e-01
8.60308707e-01 -6.10390387e-04 5.63808262e-01 -2.65148610e-01
-4.27529991e-01 2.08670095e-01 -7.49544501e-01 -2.24905372e-01
1.58540225e+00 3.33070189e-01 -2.01660469e-01 6.43612444e-01
9.80598390e-01 1.92380641e-02 -4.73065466e-01 -5.78548074e-01
9.14786100e-01 -8.06813464e-02 1.27832949e-01 -1.03275728e+00
-7.08188951e-01 7.57174641e-02 9.20886919e-02 5.06652474e-01
8.06407154e-01 3.48019063e-01 4.58265662e-01 1.12847352e+00
5.25994539e-01 -1.14272702e+00 -1.93743110e-01 6.86592400e-01
9.61323559e-01 -1.34214282e+00 6.14340842e-01 -4.70905811e-01
-6.88063860e-01 1.10102129e+00 4.81697589e-01 1.63905650e-01
9.32536006e-01 4.77591753e-01 -5.70218079e-03 -6.53983593e-01
-1.05671775e+00 -4.05250669e-01 6.89068913e-01 3.52781534e-01
1.15496814e+00 -6.85037002e-02 -5.23574889e-01 6.01663291e-01
-2.46975526e-01 8.62886831e-02 -1.72657631e-02 1.23660207e+00
-4.67286974e-01 -1.31672728e+00 -8.16403851e-02 6.50699079e-01
-6.68660283e-01 -4.35618401e-01 -8.62325788e-01 8.87547970e-01
-5.51171862e-02 9.55162704e-01 -8.58163908e-02 -2.40159959e-01
4.80133265e-01 4.66254592e-01 2.96902686e-01 -9.47485209e-01
-9.16179299e-01 2.41090506e-02 7.70657361e-01 -4.12446916e-01
-6.55829430e-01 -7.56398737e-01 -1.26889014e+00 3.32429409e-02
-5.19557178e-01 7.16699660e-01 4.64763254e-01 1.25427985e+00
1.19766183e-01 6.77186787e-01 3.38392824e-01 -2.48742059e-01
-1.07665591e-01 -1.09739041e+00 -7.24140584e-01 2.74783261e-02
1.79167777e-01 -3.72923672e-01 -1.68374375e-01 1.44992098e-01] | [9.37785530090332, 8.66728687286377] |
90da262c-eadd-4ec0-a737-fb1ab1f00bf6 | local-explanation-of-dialogue-response | 2106.06528 | null | https://arxiv.org/abs/2106.06528v2 | https://arxiv.org/pdf/2106.06528v2.pdf | Local Explanation of Dialogue Response Generation | In comparison to the interpretation of classification models, the explanation of sequence generation models is also an important problem, however it has seen little attention. In this work, we study model-agnostic explanations of a representative text generation task -- dialogue response generation. Dialog response generation is challenging with its open-ended sentences and multiple acceptable responses. To gain insights into the reasoning process of a generation model, we propose a new method, local explanation of response generation (LERG) that regards the explanations as the mutual interaction of segments in input and output sentences. LERG views the sequence prediction as uncertainty estimation of a human response and then creates explanations by perturbing the input and calculating the certainty change over the human response. We show that LERG adheres to desired properties of explanations for text generation including unbiased approximation, consistency and cause identification. Empirically, our results show that our method consistently improves other widely used methods on proposed automatic- and human- evaluation metrics for this new task by 4.4-12.8%. Our analysis demonstrates that LERG can extract both explicit and implicit relations between input and output segments. | ['William Yang Wang', 'Lise Getoor', 'Wenhu Chen', 'Connor Pryor', 'Yi-Lin Tuan'] | 2021-06-11 | null | http://proceedings.neurips.cc/paper/2021/hash/03b92cd507ff5870df0db7f074728830-Abstract.html | http://proceedings.neurips.cc/paper/2021/file/03b92cd507ff5870df0db7f074728830-Paper.pdf | neurips-2021-12 | ['implicit-relations'] | ['natural-language-processing'] | [ 6.36560380e-01 1.11543715e+00 -4.70430776e-02 -6.64938986e-01
-7.76511192e-01 -6.68694913e-01 1.01953292e+00 7.47695491e-02
2.38439828e-01 1.05271924e+00 6.57220185e-01 -5.44251382e-01
6.75362125e-02 -5.10929644e-01 -2.76052266e-01 -1.95619375e-01
3.73870701e-01 7.88336873e-01 6.94457348e-03 -4.27064657e-01
6.43850327e-01 -3.05421185e-02 -1.20316410e+00 7.56422818e-01
1.12207639e+00 6.26924396e-01 -1.36806652e-01 1.15001357e+00
-3.34117651e-01 9.69143987e-01 -9.16925967e-01 -6.45809054e-01
-8.99721161e-02 -1.08825803e+00 -1.49483764e+00 3.58122424e-03
-1.09283559e-01 -2.57884294e-01 1.65016517e-01 8.26384664e-01
4.17377740e-01 2.58154571e-01 1.08034432e+00 -1.58922744e+00
-9.51628804e-01 1.26905286e+00 -5.82254529e-02 -1.54871628e-01
9.59025085e-01 1.83041379e-01 1.11963964e+00 -7.97778964e-01
5.24572968e-01 1.60217631e+00 4.08307493e-01 1.13863552e+00
-1.22245288e+00 -2.73520440e-01 2.22382054e-01 6.03689924e-02
-7.32747853e-01 -1.94123000e-01 3.52311879e-01 -5.41519761e-01
1.10374975e+00 8.31993520e-01 1.30375296e-01 1.24937952e+00
2.65439808e-01 7.02980995e-01 9.67743278e-01 -6.14139557e-01
2.10416630e-01 5.28561234e-01 6.02598011e-01 4.12350923e-01
-1.23141721e-01 5.55225536e-02 -5.61378539e-01 -5.20773232e-01
2.35322595e-01 -5.19789934e-01 -4.76966321e-01 3.37858468e-01
-1.17741489e+00 1.10642469e+00 -1.29533093e-02 -7.86200352e-03
-3.10912073e-01 8.37925673e-02 2.58633882e-01 2.94445217e-01
5.67451894e-01 6.12502933e-01 -4.55183089e-01 -2.79012829e-01
-4.88489896e-01 5.40219784e-01 1.31716287e+00 1.22255814e+00
3.95469844e-01 4.03253473e-02 -6.75938606e-01 5.36798954e-01
4.77023095e-01 3.04791927e-01 6.69482768e-01 -7.47924387e-01
4.57816452e-01 8.31651270e-01 4.31607306e-01 -8.28651726e-01
-2.54957438e-01 4.89377901e-02 -7.55474746e-01 -2.59358343e-02
5.74732661e-01 -3.97901475e-01 -5.44354677e-01 1.81617463e+00
2.71342278e-01 -2.90225416e-01 4.27656204e-01 9.31160092e-01
9.19460833e-01 6.82729304e-01 4.64322381e-02 -4.81524318e-01
1.21733034e+00 -1.11216378e+00 -8.88336062e-01 -2.16483682e-01
4.25986767e-01 -8.49942684e-01 1.26887035e+00 1.94126204e-01
-1.07777941e+00 -6.11929893e-01 -7.79235184e-01 1.29269315e-02
2.56992448e-02 9.18331742e-02 4.69832093e-01 4.98738259e-01
-8.64126623e-01 4.34137911e-01 -1.07827559e-01 -3.43157977e-01
-4.48462963e-01 3.82675797e-01 -2.43861470e-02 4.34729487e-01
-1.40265167e+00 9.43552554e-01 3.35446090e-01 -1.21594984e-02
-4.55078393e-01 -4.38479364e-01 -6.59541667e-01 1.14500389e-01
3.25373858e-01 -9.32849884e-01 1.90516543e+00 -1.02757525e+00
-1.80155969e+00 3.07387769e-01 -3.95769924e-01 -5.72010696e-01
7.32836425e-01 -3.83834928e-01 -2.53039896e-01 -2.00591162e-01
4.39695455e-02 6.67929947e-01 7.66924739e-01 -1.34625316e+00
-4.17043895e-01 2.74722967e-02 -2.57540736e-02 1.02591239e-01
2.24257708e-01 -7.27612972e-02 3.07270944e-01 -4.65359420e-01
5.42254336e-02 -9.02288496e-01 -2.83200562e-01 -7.47762024e-01
-8.65038574e-01 -5.30367792e-01 5.09299874e-01 -5.83461165e-01
1.33611786e+00 -1.51653838e+00 7.70757021e-03 1.67279616e-01
2.52801836e-01 6.29430935e-02 -8.59408528e-02 7.59920120e-01
-1.34464413e-01 5.29017031e-01 -2.17806444e-01 -2.29786754e-01
4.85668749e-01 6.19417019e-02 -1.02161753e+00 -2.62524277e-01
1.78140059e-01 1.01503468e+00 -9.56195652e-01 -3.45782071e-01
-4.47232369e-03 -1.77935306e-02 -4.49473560e-01 7.74039686e-01
-6.84731603e-01 4.16087955e-01 -4.29141492e-01 1.70899972e-01
3.30947727e-01 -3.30696017e-01 2.65845925e-01 2.36982584e-01
7.28859529e-02 5.46884835e-01 -8.51970792e-01 1.15945566e+00
-2.49157667e-01 5.51040113e-01 -6.04442060e-01 -5.12183785e-01
1.11878991e+00 6.32235825e-01 -3.21864605e-01 -4.99656983e-02
1.26385927e-01 1.90960616e-01 2.64547497e-01 -6.73828661e-01
8.15309942e-01 -1.35194823e-01 -2.85505146e-01 1.01153171e+00
-1.20692886e-01 -3.55512679e-01 1.73934754e-02 5.97294509e-01
9.26194012e-01 -2.12672856e-02 8.80368650e-01 -1.44548342e-01
7.41962433e-01 7.30833784e-02 2.14965120e-01 1.16546309e+00
5.08027598e-02 7.72991061e-01 7.74030268e-01 -2.18188599e-01
-8.42212975e-01 -8.91363680e-01 3.95870388e-01 8.41475368e-01
1.50889307e-01 -4.66705084e-01 -1.09744895e+00 -9.88185942e-01
-6.44773617e-02 1.47958374e+00 -6.66690946e-01 -1.29674897e-01
-3.05460721e-01 -3.91927391e-01 6.01580679e-01 4.91531312e-01
2.24570304e-01 -1.22001970e+00 -3.53967041e-01 2.47984856e-01
-6.95135474e-01 -8.38505447e-01 -7.90221751e-01 -6.88567609e-02
-6.12493992e-01 -1.03613150e+00 -2.96160161e-01 -3.28406662e-01
7.29974627e-01 -6.12450950e-02 1.17720032e+00 3.10722768e-01
8.23205560e-02 3.68859679e-01 -5.20392358e-01 -4.45536345e-01
-1.23828781e+00 1.22396024e-02 -1.30290433e-03 -2.88411882e-02
3.52486670e-01 -1.75013676e-01 -3.81467670e-01 4.45695877e-01
-6.59422278e-01 3.58677804e-01 3.38750094e-01 1.12750435e+00
8.57231021e-02 -7.41845429e-01 1.08611131e+00 -1.20436072e+00
1.48453319e+00 -4.52984750e-01 -1.35516331e-01 5.38022041e-01
-1.06378341e+00 5.32316267e-01 7.67410278e-01 -5.11906028e-01
-1.49950564e+00 -1.61843285e-01 -3.13058980e-02 3.57639730e-01
-3.67161721e-01 2.75955141e-01 -8.36442932e-02 5.74285924e-01
9.42827821e-01 2.17450887e-01 -4.89363726e-03 -5.61512299e-02
6.90332294e-01 9.69775319e-01 4.94672865e-01 -5.72729230e-01
5.71299970e-01 -1.65961400e-01 -3.56794149e-01 -4.40773427e-01
-7.84500659e-01 -2.21543655e-01 -2.22735554e-01 -3.23056072e-01
7.76537597e-01 -2.79811859e-01 -9.23801422e-01 -3.93369980e-02
-1.78245354e+00 -2.52967417e-01 -2.88646489e-01 3.69838178e-01
-8.29138875e-01 5.65376103e-01 -5.52292526e-01 -1.42154312e+00
-6.66965902e-01 -9.77289855e-01 9.56244409e-01 1.53029129e-01
-1.30562282e+00 -9.59854245e-01 3.79719399e-02 3.81123692e-01
3.36384714e-01 1.46620110e-01 1.19667375e+00 -1.39064026e+00
-4.05678570e-01 -4.96652983e-02 1.97318615e-03 1.43982157e-01
1.28130794e-01 -4.63014729e-02 -1.00561273e+00 3.66920203e-01
2.13992596e-01 -3.82315814e-01 5.00105679e-01 5.79353571e-02
8.17932546e-01 -9.92761791e-01 -1.16331942e-01 -2.65435874e-01
7.74733007e-01 3.18618357e-01 5.50931871e-01 -1.49192452e-01
2.00765610e-01 1.11662316e+00 8.85949314e-01 5.73086083e-01
2.02589139e-01 5.40248156e-01 3.25601727e-01 2.35878065e-01
1.52232006e-01 -6.06928706e-01 3.88799667e-01 6.15662217e-01
8.32762718e-02 -9.44795132e-01 -6.25386357e-01 2.58554518e-01
-2.44575667e+00 -1.21243560e+00 -7.51797140e-01 1.97876847e+00
8.93448889e-01 -2.41502225e-02 -7.36688972e-02 1.18388077e-02
7.46680915e-01 -1.67320609e-01 -4.00644124e-01 -1.00195730e+00
1.41817242e-01 -2.82827951e-02 -6.68263435e-02 1.06458890e+00
-3.67811948e-01 8.13895166e-01 6.60830593e+00 5.10739446e-01
-6.38663352e-01 -8.96599293e-02 7.31071055e-01 4.20287400e-01
-8.21972370e-01 2.18048409e-01 -7.55299985e-01 3.33442837e-01
9.19971764e-01 -6.24949396e-01 2.76323020e-01 8.68586600e-01
4.67466205e-01 -1.86553691e-02 -1.58812451e+00 4.91033524e-01
6.55464530e-02 -1.14032578e+00 4.50261056e-01 -1.36641279e-01
6.95951343e-01 -9.39804971e-01 -2.08437108e-02 2.47144938e-01
5.19341111e-01 -1.28210545e+00 8.10099304e-01 7.10997283e-01
3.92809123e-01 -4.72595453e-01 1.09326458e+00 7.64156342e-01
-6.24246240e-01 5.10549508e-02 -2.51063049e-01 -3.86185348e-01
3.83165985e-01 3.21734756e-01 -1.72545969e+00 6.03129387e-01
-1.17367931e-01 1.61719266e-02 -2.88325787e-01 3.24311823e-01
-7.21045196e-01 7.72888601e-01 1.92427903e-01 -5.58670163e-01
-3.24880145e-02 -7.16491714e-02 5.34835875e-01 1.26895440e+00
3.89923841e-01 3.96812111e-01 -5.94610386e-02 1.39281809e+00
1.55752033e-01 1.13066636e-01 -5.24716556e-01 -8.53910819e-02
3.94833505e-01 1.13607502e+00 -3.89790386e-01 -6.06750488e-01
2.02864215e-01 1.09032905e+00 1.34335741e-01 2.29770586e-01
-9.85556722e-01 -1.49229839e-01 3.96615088e-01 -2.27381676e-01
-3.20131630e-01 3.64574432e-01 -6.31034851e-01 -9.24121022e-01
-1.02851383e-01 -1.15889478e+00 2.05875978e-01 -8.95931244e-01
-1.26530349e+00 1.07906163e+00 1.13388367e-01 -1.10738385e+00
-1.28585017e+00 -3.57003748e-01 -9.33771431e-01 1.25283277e+00
-7.59335101e-01 -8.08260083e-01 -3.40324074e-01 7.23059848e-02
1.05432260e+00 -2.01225609e-01 1.08167124e+00 -5.95796049e-01
-1.72615081e-01 6.57119453e-01 -5.64039886e-01 -2.41498113e-01
6.97599769e-01 -1.56667244e+00 7.63069212e-01 7.96375453e-01
2.15483569e-02 9.21267509e-01 1.21679592e+00 -7.51261055e-01
-8.33136022e-01 -8.64909828e-01 1.52362335e+00 -7.46431768e-01
4.10894245e-01 -2.15661600e-01 -8.97149980e-01 7.06336141e-01
6.07059777e-01 -8.13102067e-01 7.15673268e-01 -1.04983382e-01
-1.35708377e-01 5.29994667e-01 -1.06548619e+00 8.26358438e-01
9.45200086e-01 -3.02099526e-01 -9.90539849e-01 4.14409637e-01
1.02689731e+00 -4.65514123e-01 -3.50254118e-01 1.48436576e-01
4.93624657e-01 -1.11947560e+00 4.48944986e-01 -9.81835902e-01
7.26864517e-01 -3.05386454e-01 -5.80103472e-02 -1.51521993e+00
-1.46409959e-01 -1.07615411e+00 -1.35392815e-01 1.35768723e+00
9.30108488e-01 -6.70658231e-01 4.46632683e-01 1.19518936e+00
-2.79675484e-01 -5.52631915e-01 -4.23228204e-01 -5.19991696e-01
-2.00419471e-01 -3.72005463e-01 7.88986087e-01 6.13843799e-01
7.46550024e-01 8.82888675e-01 -5.86957574e-01 8.18507895e-02
2.48719022e-01 2.86942333e-01 1.03396928e+00 -1.01742327e+00
-5.16583741e-01 -2.44471341e-01 1.51503772e-01 -1.23972225e+00
4.09644455e-01 -7.05595434e-01 4.60004717e-01 -1.57422125e+00
1.29220888e-01 7.27178305e-02 5.50884604e-01 2.33926430e-01
-5.14451861e-01 -3.15774173e-01 1.33853197e-01 5.27708754e-02
-2.94548720e-01 5.44254005e-01 1.01267707e+00 -5.68838306e-02
-2.88256973e-01 3.91700000e-01 -1.06549430e+00 7.23590732e-01
9.99648035e-01 -4.83339250e-01 -7.96799362e-01 -3.03541478e-02
1.19121276e-01 4.55871105e-01 2.38583565e-01 -3.18098933e-01
1.76124692e-01 -3.68968338e-01 -2.18569972e-02 -4.23705906e-01
8.59649777e-02 -3.43993396e-01 2.74733275e-01 5.21888673e-01
-1.20799363e+00 2.65884966e-01 -2.25555494e-01 7.19707549e-01
-1.55303031e-01 -6.76474571e-01 2.63480097e-01 -1.44328833e-01
-3.34605247e-01 -3.17822903e-01 -5.79183161e-01 1.61532417e-01
7.84404099e-01 -1.91236928e-01 -5.51339209e-01 -9.77836728e-01
-6.57243431e-01 1.34434044e-01 6.17904551e-02 4.94244725e-01
8.50126803e-01 -1.11335957e+00 -9.04842675e-01 -7.59211630e-02
9.11808535e-02 -3.94560844e-01 1.17846495e-02 6.14446223e-01
-7.92853013e-02 5.69794297e-01 2.57082880e-01 -4.21073496e-01
-1.49581254e+00 3.61822993e-01 2.70431906e-01 -4.01314676e-01
-2.37685516e-01 7.57816017e-01 4.24780428e-01 -5.74217677e-01
1.38744846e-01 -5.43143451e-01 -2.86470741e-01 -3.98570120e-01
5.56151509e-01 3.52291048e-01 -2.98131853e-01 -2.73306400e-01
-8.01752955e-02 -1.86060324e-01 -7.65912980e-02 -5.68292320e-01
5.11171162e-01 -2.16158792e-01 -7.58601055e-02 5.95366776e-01
6.00440979e-01 7.54484162e-02 -8.96709621e-01 1.51983023e-01
2.26325303e-01 -2.49406964e-01 -8.86611342e-01 -1.22298360e+00
-1.20312586e-01 7.81468689e-01 8.24255794e-02 6.94059014e-01
6.65526807e-01 -1.62779968e-02 6.28108025e-01 3.82159919e-01
8.42214599e-02 -9.25356925e-01 2.82273859e-01 6.82817340e-01
1.38932621e+00 -1.18850553e+00 -2.74771482e-01 -7.14094341e-01
-1.16029918e+00 1.23874629e+00 9.31082189e-01 3.74460429e-01
-1.14247836e-01 1.22441044e-02 2.03625128e-01 7.61614218e-02
-1.36121213e+00 2.58347452e-01 5.10067165e-01 6.25837624e-01
8.46171796e-01 1.76624924e-01 -6.86653972e-01 1.06710434e+00
-6.08746648e-01 -2.57358998e-01 8.60453725e-01 3.02628726e-01
-4.75434244e-01 -1.25162411e+00 -2.98198998e-01 2.76824147e-01
-2.73302704e-01 -2.59518743e-01 -1.31614661e+00 7.34267592e-01
-3.42898786e-01 1.63854110e+00 -3.47954810e-01 -6.21062577e-01
2.17136517e-01 5.52855909e-01 7.23428130e-02 -8.12939227e-01
-9.37389553e-01 -3.57882828e-01 5.79654753e-01 -2.36101180e-01
-1.84940889e-01 -4.54885960e-01 -1.54169714e+00 -2.75758237e-01
-5.98003685e-01 7.23008752e-01 3.84854048e-01 1.17523658e+00
3.91028464e-01 3.99952352e-01 6.41921937e-01 -3.18517923e-01
-1.34008956e+00 -1.34710634e+00 -1.85498953e-01 5.86910367e-01
8.31684843e-02 -1.73788041e-01 -7.08588302e-01 1.80465356e-01] | [12.5418119430542, 8.314375877380371] |
54077676-1ad2-45e6-8543-57cfc7242894 | magnetic-resonance-fingerprinting-with | 2209.08734 | null | https://arxiv.org/abs/2209.08734v1 | https://arxiv.org/pdf/2209.08734v1.pdf | Magnetic Resonance Fingerprinting with compressed sensing and distance metric learning | Magnetic Resonance Fingerprinting (MRF) is a novel technique that simultaneously estimates multiple tissue-related parameters, such as the longitudinal relaxation time T1, the transverse relaxation time T2, off resonance frequency B0 and proton density, from a scanned object in just tens of seconds. However, the MRF method suffers from aliasing artifacts because it significantly undersamples the k-space data. In this work, we propose a compressed sensing (CS) framework for simultaneously estimating multiple tissue-related parameters based on the MRF method. It is more robust to low sampling ratio and is therefore more efficient in estimating MR parameters for all voxels of an object. Furthermore, the MRF method requires identifying the nearest atoms of the query fingerprints from the MR-signal-evolution dictionary with the L2 distance. However, we observed that the L2 distance is not always a proper metric to measure the similarities between MR Fingerprints. Adaptively learning a distance metric from the undersampled training data can significantly improve the matching accuracy of the query fingerprints. Numerical results on extensive simulated cases show that our method substantially outperforms stateof-the-art methods in terms of accuracy of parameter estimation. | ['Xiaogang Wang', 'Jing Yuan', 'Qinwei Zhang', 'Hongsheng Li', 'Zhe Wang'] | 2022-09-19 | null | null | null | null | ['magnetic-resonance-fingerprinting'] | ['medical'] | [ 5.43800294e-01 -2.92482167e-01 -3.91005754e-01 -3.25807691e-01
-1.14567113e+00 -1.43446937e-01 -1.38160456e-02 6.94004372e-02
-3.54308277e-01 6.05938613e-01 2.20977560e-01 1.38693154e-01
-5.84215760e-01 -2.79315054e-01 -5.26947856e-01 -1.01948547e+00
-4.09749359e-01 3.79025578e-01 3.63194197e-01 2.90367365e-01
4.67346609e-01 5.72668076e-01 -1.07110429e+00 -3.81889790e-02
9.03846025e-01 1.25372171e+00 6.65866375e-01 1.46119669e-01
9.70553532e-02 8.22943807e-01 -9.63716805e-02 2.74704903e-01
1.85699478e-01 -2.19002515e-01 -7.88374424e-01 -2.59677649e-01
5.21554649e-01 -3.34312230e-01 -6.30900979e-01 1.20461047e+00
6.51831210e-01 2.98143148e-01 6.20533407e-01 -5.68285286e-01
-2.50866503e-01 7.47639596e-01 -7.45129108e-01 5.26631296e-01
2.75828004e-01 -3.97303969e-01 3.32279861e-01 -9.71888363e-01
6.79004014e-01 6.57644153e-01 8.84700358e-01 2.47019291e-01
-1.08149695e+00 -7.08708167e-01 -4.46580440e-01 3.56478035e-01
-1.74004197e+00 -4.40560043e-01 9.03715730e-01 -5.10272682e-01
3.13568860e-01 4.19888496e-01 3.64258230e-01 5.29048741e-01
4.86235410e-01 3.93560380e-01 1.22017205e+00 -3.75462651e-01
1.66211620e-01 -1.94581971e-01 8.97088498e-02 7.31392086e-01
1.13149293e-01 1.33088484e-01 -5.42844832e-01 -6.98488533e-01
1.01279473e+00 -4.62883227e-02 -7.46010244e-01 -3.87574226e-01
-1.69553256e+00 5.64919472e-01 1.39355674e-01 6.12036049e-01
-5.37994802e-01 1.84463993e-01 3.05691957e-01 -1.07257314e-01
-1.69367455e-02 3.51479232e-01 1.26988932e-01 6.49444237e-02
-1.11404538e+00 4.49584238e-02 3.67615461e-01 6.57736778e-01
4.71095502e-01 -2.08385333e-01 -2.47866377e-01 9.72399890e-01
1.30275562e-01 6.62715852e-01 8.05548787e-01 -1.25202250e+00
1.55077547e-01 -2.02381149e-01 1.09461688e-01 -1.22393882e+00
-4.60112751e-01 -4.62375730e-01 -9.75252748e-01 -6.38107300e-01
3.94406378e-01 2.13196129e-01 -3.84419471e-01 1.57561529e+00
6.43027008e-01 6.47645175e-01 -3.83684278e-01 1.35148382e+00
5.98534882e-01 3.93275082e-01 -2.13905171e-01 -6.59035146e-01
1.13807535e+00 -7.38091052e-01 -7.59733915e-01 -5.81490323e-02
3.51631790e-01 -8.47405791e-01 6.58542991e-01 3.41293097e-01
-1.05173206e+00 -4.98175830e-01 -9.93014395e-01 4.56643134e-01
4.50555772e-01 6.27945065e-02 6.06207550e-01 5.31892598e-01
-5.02828956e-01 8.47446799e-01 -9.35271382e-01 1.59553379e-01
6.31301701e-02 3.41598243e-01 -4.25917000e-01 -4.93825734e-01
-1.18527651e+00 7.71026611e-01 2.39516050e-02 1.07770562e-01
-8.19801867e-01 -1.08406234e+00 -5.12955964e-01 -4.42262322e-01
3.09421659e-01 -2.04164773e-01 8.43766630e-01 -1.57345891e-01
-1.33504987e+00 5.59552193e-01 -1.44478396e-01 -2.62901604e-01
3.41609120e-01 2.96880454e-01 -6.84078634e-01 8.53268087e-01
2.35381588e-01 2.21345842e-01 8.48931491e-01 -8.55129302e-01
1.59190595e-01 -6.39894128e-01 -6.14908218e-01 7.18349172e-03
-1.61524601e-02 -1.60661176e-01 -5.27367555e-02 -7.53766954e-01
8.36305022e-01 -9.01175976e-01 -3.92380893e-01 8.03069249e-02
-2.29066625e-01 3.95117462e-01 3.33912313e-01 -8.30262423e-01
1.19335270e+00 -2.30622697e+00 -1.70885801e-01 4.93031800e-01
2.41381720e-01 -1.45976797e-01 -7.48137906e-02 2.09848970e-01
-1.57561898e-01 -4.63929325e-01 -4.74705905e-01 5.36361933e-01
-4.68435496e-01 -1.63403049e-01 -9.77518782e-02 1.19488823e+00
-5.64664364e-01 5.61078072e-01 -1.10013366e+00 -8.24961364e-01
-2.81588640e-02 5.60295284e-01 -2.53889620e-01 -1.26844302e-01
5.30901670e-01 8.70772839e-01 -7.71773100e-01 5.56113660e-01
9.21118975e-01 -3.05933952e-01 2.92799115e-01 -8.27671349e-01
-4.22338545e-02 -8.35368186e-02 -1.24164832e+00 1.90819883e+00
-1.82026461e-01 2.59708285e-01 2.09930927e-01 -1.38085210e+00
1.06154168e+00 4.68892157e-01 1.41347873e+00 -9.48354781e-01
-7.31826574e-02 5.71063876e-01 -3.10074333e-02 -8.49949300e-01
1.33616731e-01 -4.28733796e-01 2.06168577e-01 5.04681230e-01
-2.44750798e-01 4.54404205e-03 -1.53941020e-01 -5.47753349e-02
1.05111754e+00 -4.56800848e-01 1.76147163e-01 -7.01575816e-01
6.86102867e-01 -4.35105801e-01 6.02861285e-01 7.23066986e-01
-4.15275812e-01 4.84927177e-01 -1.82797059e-01 -3.09915394e-01
-9.51920748e-01 -9.93096530e-01 -9.04309928e-01 3.01719457e-01
4.00192499e-01 -3.14092338e-02 -7.47584403e-01 -2.51936972e-01
2.34593719e-01 2.25611180e-01 -5.12520909e-01 -1.81008130e-01
-7.61175811e-01 -8.08530390e-01 5.19541323e-01 2.95136929e-01
3.47289830e-01 -4.30265278e-01 -7.23111928e-01 4.48041409e-01
-6.78670228e-01 -1.34083438e+00 -1.00469685e+00 -1.12630732e-01
-1.20124233e+00 -9.74437594e-01 -9.69858170e-01 -5.94473660e-01
6.54756963e-01 4.39342201e-01 6.88223898e-01 -7.18237832e-02
-4.34406489e-01 3.02130014e-01 -1.33898586e-01 2.31348112e-01
-2.88920164e-01 -3.35373998e-01 1.25299767e-01 2.85851061e-01
-6.99658245e-02 -6.68620586e-01 -8.69341612e-01 7.88895369e-01
-8.40566158e-01 -2.70921648e-01 4.34399188e-01 7.19647110e-01
1.25128150e+00 3.98918837e-02 5.77118397e-01 -6.89777613e-01
1.45710215e-01 -4.96166974e-01 -3.28647852e-01 3.12256604e-01
-5.40062606e-01 1.35500565e-01 4.10281241e-01 -7.93662965e-01
-5.92998922e-01 1.01338185e-01 2.10264683e-01 -3.73136610e-01
2.56938666e-01 5.97695053e-01 2.72625744e-01 -6.88373685e-01
6.26554012e-01 7.53047168e-01 1.79605514e-01 -5.01433432e-01
-1.29556969e-01 5.94326079e-01 1.00239396e+00 -8.50091636e-01
2.57285118e-01 7.44692445e-01 4.45349008e-01 -9.93754327e-01
-6.29973829e-01 -7.38281012e-01 -4.78758454e-01 -3.70893776e-01
3.21734816e-01 -6.14892662e-01 -6.10424578e-01 3.34233582e-01
-6.25080347e-01 9.41644236e-02 -4.96327057e-02 1.21516716e+00
-7.10275114e-01 1.03307104e+00 -6.63726568e-01 -5.09502113e-01
-5.54158807e-01 -1.24459314e+00 7.62458265e-01 -1.48109630e-01
-7.97821283e-02 -6.82451367e-01 1.53496638e-01 5.06392121e-01
6.00911796e-01 4.76443619e-01 9.96866405e-01 -1.92172825e-01
-2.61720061e-01 -2.91761994e-01 1.86175942e-01 3.33609283e-02
8.98320153e-02 -6.80172026e-01 -4.58867788e-01 -4.10917610e-01
6.29725814e-01 5.37901074e-02 4.31486338e-01 8.37472677e-01
1.54985881e+00 -1.51707768e-01 -4.10131037e-01 7.33719587e-01
1.40853798e+00 3.13722312e-01 6.96696341e-01 1.72856506e-02
4.29097801e-01 3.23499471e-01 8.63225043e-01 5.71336269e-01
9.59795564e-02 8.69169712e-01 3.90192233e-02 3.52937609e-01
-1.14208654e-01 -9.92702767e-02 1.01295717e-01 1.19461501e+00
2.59549674e-02 5.64140618e-01 -8.03950369e-01 4.31171000e-01
-1.46597004e+00 -1.04164100e+00 -1.97292224e-01 2.51383114e+00
9.66512799e-01 -4.42546695e-01 5.48448227e-03 1.70967907e-01
9.81494725e-01 -4.55982611e-03 -9.16460931e-01 4.34037417e-01
1.05570570e-01 2.17334062e-01 8.72941256e-01 4.69713241e-01
-5.58244109e-01 6.64298907e-02 6.93006325e+00 8.63143504e-01
-1.30048966e+00 3.62660974e-01 3.65004182e-01 1.00663248e-02
-3.98646295e-01 -2.03228906e-01 -4.23972368e-01 7.50072777e-01
8.60525489e-01 -1.41842872e-01 7.08293140e-01 4.75371063e-01
2.08283156e-01 -3.23906004e-01 -7.70054877e-01 1.31268311e+00
1.72154307e-02 -1.31357789e+00 -2.56762713e-01 1.50036976e-01
4.71059173e-01 9.46183503e-02 -6.16206527e-02 -3.98934305e-01
-7.60409534e-01 -9.43242967e-01 6.50885284e-01 8.91452730e-01
1.15585268e+00 -5.70174515e-01 4.54471469e-01 2.83738881e-01
-1.20127904e+00 1.08279087e-01 -5.98610401e-01 6.63917005e-01
8.92882571e-02 1.20686793e+00 -6.72324300e-01 4.38118756e-01
5.30100763e-01 4.41577554e-01 -8.92905593e-02 1.17483950e+00
5.38325608e-01 4.70007360e-01 -3.40614766e-01 3.58227164e-01
-2.94821393e-02 -3.18577081e-01 5.34394741e-01 8.12914491e-01
5.99152207e-01 4.88547236e-01 1.68426007e-01 6.41949236e-01
3.45719963e-01 1.79801121e-01 -1.49910673e-01 -1.00704404e-02
8.55404437e-01 1.10106790e+00 -6.40879333e-01 -2.06145391e-01
-1.11721151e-01 6.73058331e-01 -1.40051112e-01 1.67936176e-01
-7.83295870e-01 -3.18605691e-01 1.24500327e-01 6.17956161e-01
2.02669933e-01 -2.24825129e-01 2.12280899e-01 -1.05706000e+00
1.37571827e-01 -9.38361526e-01 2.56878316e-01 -5.53545356e-01
-1.16927159e+00 4.35412675e-01 5.38890399e-02 -1.17305684e+00
1.58838574e-02 -2.05082625e-01 -8.84431005e-02 7.89034069e-01
-1.35681152e+00 -4.39158291e-01 -3.52122158e-01 8.35588872e-01
-5.37169464e-02 8.85481462e-02 6.70798838e-01 6.58634245e-01
-2.31375664e-01 4.75377262e-01 4.87012327e-01 -7.23116472e-02
6.58955932e-01 -7.65211642e-01 -2.16075718e-01 3.10598761e-01
-2.52148002e-01 6.94224417e-01 6.14960372e-01 -6.67458534e-01
-1.85594630e+00 -5.82674444e-01 4.21131492e-01 2.20978409e-01
5.44837534e-01 2.08614394e-01 -1.06683600e+00 1.56026125e-01
-5.61554253e-01 6.43137038e-01 8.35285306e-01 -6.29432857e-01
-4.73465137e-02 -3.63389045e-01 -1.57368386e+00 -1.38270661e-01
7.11656034e-01 -7.80334711e-01 -1.84059203e-01 4.17647392e-01
2.77154267e-01 -6.47572458e-01 -1.72400534e+00 4.61685896e-01
8.46652746e-01 -7.46125817e-01 1.26910484e+00 1.12203369e-02
-7.12566376e-02 -4.23846990e-01 -5.50543427e-01 -8.77412558e-01
-5.13087451e-01 -4.67672378e-01 -2.46222332e-01 6.13414288e-01
-1.87206164e-01 -6.43904507e-01 7.81006098e-01 3.64672542e-01
6.68449281e-03 -9.19271171e-01 -1.23757756e+00 -9.03368592e-01
-1.25346899e-01 -2.08442330e-01 7.05542028e-01 1.15457976e+00
6.22700974e-02 -8.80228877e-02 -3.25598747e-01 3.57464254e-01
1.36283684e+00 2.77803957e-01 1.82743056e-03 -1.07177472e+00
-4.23050523e-01 4.48352769e-02 -3.22787970e-01 -1.01888478e+00
-3.08400095e-02 -1.11912823e+00 -3.35056148e-02 -9.71794605e-01
5.07580280e-01 -8.74088109e-01 -5.63328564e-01 -1.36122823e-01
1.75687030e-01 6.44310266e-02 -1.81200773e-01 5.94354093e-01
-1.26541033e-01 2.92903990e-01 1.82764959e+00 -1.75474182e-01
2.59922296e-01 -2.20361233e-01 -3.79027158e-01 3.55784208e-01
4.06413883e-01 -7.04191506e-01 -3.59997183e-01 -2.45583132e-01
-6.24305271e-02 9.96105134e-01 2.96053916e-01 -1.13037992e+00
3.10283095e-01 -2.63223678e-01 4.05711532e-01 -7.04949796e-01
2.49581888e-01 -8.00718367e-01 7.10276604e-01 7.09976077e-01
-3.90252203e-01 -2.40545452e-01 -2.37341240e-01 4.72018123e-01
-1.54878035e-01 -5.49636185e-01 1.05018866e+00 -2.26292029e-01
-2.56257564e-01 5.54387093e-01 -1.56625018e-01 3.05310590e-03
6.47018969e-01 -2.80946642e-01 1.21643074e-01 -1.37433782e-01
-6.20055258e-01 -1.79313555e-01 2.58338511e-01 -1.29728645e-01
7.89765477e-01 -1.50470781e+00 -6.00093901e-01 1.41209826e-01
-1.75940841e-01 -2.72029310e-01 7.26063609e-01 1.56802082e+00
-5.08525372e-01 4.99519467e-01 -1.94396392e-01 -8.07383239e-01
-8.02710891e-01 6.53791249e-01 6.55005097e-01 -1.94713876e-01
-8.40297580e-01 4.92766589e-01 -7.44299740e-02 -3.59425068e-01
1.78145580e-02 -2.68383205e-01 8.80506709e-02 -2.42170155e-01
9.23120379e-01 7.29216576e-01 2.42272213e-01 -7.94552147e-01
-6.89363122e-01 1.04273343e+00 1.07738592e-01 -8.12071562e-02
1.24447215e+00 -2.40669906e-01 -1.25030741e-01 2.91586012e-01
1.54582536e+00 7.37254545e-02 -1.05929196e+00 -5.70203125e-01
-7.83084780e-02 -7.25889921e-01 5.88530421e-01 -5.89611948e-01
-1.37965560e+00 5.39582968e-01 7.69346535e-01 -1.48597524e-01
9.80782270e-01 2.66412627e-02 1.24541187e+00 1.62300080e-01
7.78923273e-01 -9.59071159e-01 -5.73643632e-02 -6.24447055e-02
7.81501889e-01 -7.52123892e-01 2.62413979e-01 -5.67082942e-01
-3.68901253e-01 1.20867825e+00 1.17846191e-01 4.19187509e-02
7.61640191e-01 3.46845061e-01 -2.43433759e-01 -2.54535228e-01
-9.97525007e-02 6.62980020e-01 2.42861286e-01 5.01073837e-01
4.03394639e-01 1.66813090e-01 -4.16214883e-01 3.69681776e-01
-1.28073711e-02 1.80964962e-01 3.40901762e-01 7.98860312e-01
-4.74203318e-01 -7.32199907e-01 -7.57222772e-01 4.86086071e-01
-5.70501328e-01 6.62248507e-02 4.02546138e-01 2.07820535e-01
-2.06441179e-01 4.97343272e-01 -2.43436053e-01 -1.84494644e-01
1.52831554e-01 -3.41126323e-01 1.16630948e+00 -6.10110275e-02
1.44136036e-02 9.69558433e-02 -3.63288343e-01 -8.37835610e-01
-5.82508683e-01 -9.03481483e-01 -1.51273823e+00 -2.16359034e-01
-4.62984473e-01 4.38569546e-01 8.55071485e-01 8.42934906e-01
1.90637127e-01 -6.16501924e-03 1.00892282e+00 -6.24449372e-01
-9.36988294e-01 -6.85940981e-01 -1.10610020e+00 1.79220378e-01
3.96378994e-01 -8.20465624e-01 -1.51801780e-01 -2.05382034e-01] | [13.501082420349121, -2.413736343383789] |
4fe4566a-532f-41f5-a7e0-7d3b723a7010 | deep-learning-of-partial-graph-matching-via | null | null | http://openaccess.thecvf.com//content/CVPR2023/html/Wang_Deep_Learning_of_Partial_Graph_Matching_via_Differentiable_Top-K_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Wang_Deep_Learning_of_Partial_Graph_Matching_via_Differentiable_Top-K_CVPR_2023_paper.pdf | Deep Learning of Partial Graph Matching via Differentiable Top-K | Graph matching (GM) aims at discovering node matching between graphs, by maximizing the node- and edge-wise affinities between the matched elements. As an NP-hard problem, its challenge is further pronounced in the existence of outlier nodes in both graphs which is ubiquitous in practice, especially for vision problems. However, popular affinity-maximization-based paradigms often lack a principled scheme to suppress the false matching and resort to handcrafted thresholding to dismiss the outliers. This limitation is also inherited by the neural GM solvers though they have shown superior performance in the ideal no-outlier setting. In this paper, we propose to formulate the partial GM problem as the top-k selection task with a given/estimated number of inliers k. Specifically, we devise a differentiable top-k module that enables effective gradient descent over the optimal-transport layer, which can be readily plugged into SOTA deep GM pipelines including the quadratic matching network NGMv2 as well as the linear matching network GCAN. Meanwhile, the attention-fused aggregation layers are developed to estimate k to enable automatic outlier-robust matching in the wild. Last but not least, we remake and release a new benchmark called IMC-PT-SparseGM, originating from the IMC-PT stereo-matching dataset. The new benchmark involves more scale-varying graphs and partial matching instances from the real world. Experiments show that our methods outperform other partial matching schemes on popular benchmarks. | ['Junchi Yan', 'Xiaokang Yang', 'Shaofei Jiang', 'Ziao Guo', 'Runzhong Wang'] | 2023-01-01 | null | null | null | cvpr-2023-1 | ['stereo-matching-1', 'graph-matching'] | ['computer-vision', 'graphs'] | [ 6.49842843e-02 1.84024751e-01 1.26649022e-01 -2.51564354e-01
-1.09067190e+00 -1.65233195e-01 3.67442608e-01 1.17770493e-01
-2.23282129e-01 1.99734256e-01 -5.58887757e-02 1.94153428e-01
-2.06466869e-01 -5.46892583e-01 -9.14015055e-01 -7.41691470e-01
-1.76643163e-01 5.65555751e-01 1.84295595e-01 -1.15534872e-01
1.59026280e-01 4.95057046e-01 -1.47932804e+00 -6.85015023e-02
1.23687077e+00 1.20221686e+00 9.09743365e-03 2.33419970e-01
2.23077282e-01 3.66146117e-01 -1.03020355e-01 -7.68086970e-01
7.76045084e-01 -1.48187369e-01 -3.57997566e-01 3.21340799e-01
1.23360062e+00 -3.12766641e-01 -5.30712128e-01 1.51703572e+00
6.13703072e-01 1.76412240e-01 4.73808080e-01 -1.60697901e+00
-4.21665043e-01 3.34165841e-01 -9.63731408e-01 -6.95858374e-02
2.77492493e-01 4.06751752e-01 1.24364543e+00 -1.16954732e+00
5.79777718e-01 1.03877211e+00 1.08791363e+00 7.32029229e-02
-1.21339440e+00 -5.09637654e-01 2.78664827e-01 2.06935048e-01
-1.65862310e+00 -4.01076794e-01 9.26890731e-01 -4.55042064e-01
7.75982141e-01 1.97021425e-01 6.60770416e-01 8.78453314e-01
-3.57543677e-02 7.57312000e-01 5.37856221e-01 3.28363888e-02
5.72242886e-02 -3.58934790e-01 -1.03078753e-01 8.19486618e-01
4.20653880e-01 8.34049806e-02 -5.17894745e-01 -2.87154764e-01
4.13042456e-01 1.20706320e-01 -3.82759869e-01 -6.81354880e-01
-1.17329991e+00 4.96765554e-01 8.85476291e-01 -6.97789043e-02
-5.37527919e-01 2.66541511e-01 3.34166229e-01 3.62196326e-01
5.84866345e-01 3.96710068e-01 -2.68999547e-01 3.42961937e-01
-1.20620668e+00 3.82975549e-01 6.39009893e-01 1.21307075e+00
1.07531953e+00 1.57371126e-02 -1.48165703e-01 7.62365997e-01
3.16886634e-01 3.51109236e-01 3.08467690e-02 -8.44032705e-01
6.95808351e-01 9.34796154e-01 -1.59824640e-01 -1.64274228e+00
-4.06206548e-01 -8.91063929e-01 -1.01612449e+00 -1.05467364e-02
4.47732389e-01 2.85076886e-01 -1.02530706e+00 1.66027617e+00
6.26034558e-01 7.58171380e-01 -4.74437773e-01 1.28380704e+00
8.26245844e-01 2.06936225e-02 -2.48256475e-01 7.06254914e-02
9.91453171e-01 -1.01110625e+00 -3.35351706e-01 -3.07353526e-01
6.70349836e-01 -6.20528281e-01 9.11047816e-01 2.06172347e-01
-9.22056973e-01 -3.75432283e-01 -1.01539338e+00 -2.22348571e-01
-2.39400029e-01 -5.40414788e-02 5.02351403e-01 2.61876792e-01
-8.55836928e-01 8.23845863e-01 -7.27343440e-01 -3.28109264e-01
4.22551960e-01 3.25122416e-01 -6.12574518e-01 -2.58334756e-01
-9.20072138e-01 4.80278134e-01 2.70067066e-01 6.50156319e-01
-5.88981807e-01 -9.45853114e-01 -1.10288095e+00 2.32089050e-02
7.11881518e-01 -9.26856339e-01 5.13197124e-01 -9.92441595e-01
-9.73517239e-01 1.16330755e+00 7.95538127e-02 -3.84849489e-01
9.92388785e-01 -1.89077154e-01 -2.34796554e-01 -7.17383102e-02
3.71821404e-01 6.11625075e-01 1.00450182e+00 -1.06381583e+00
-4.94849950e-01 -6.04946852e-01 -2.55678892e-01 8.12302753e-02
-2.31219098e-01 -4.26029950e-01 -1.09570265e+00 -8.29107821e-01
6.08335733e-01 -9.69316065e-01 -4.50810879e-01 1.90122813e-01
-6.99929774e-01 -5.38987927e-02 4.86909330e-01 -7.39823043e-01
1.08165443e+00 -2.22052360e+00 2.49757320e-01 6.85782313e-01
5.08428395e-01 -5.77390231e-02 -2.89810747e-01 3.67344856e-01
-9.07051042e-02 -3.90008718e-01 -4.00735766e-01 -8.09655309e-01
2.16763794e-01 1.17381632e-01 -3.30956094e-02 9.37707126e-01
2.15488598e-01 9.92833674e-01 -1.04691494e+00 -5.18821895e-01
1.60205245e-01 2.17960835e-01 -6.33355677e-01 7.48630688e-02
-1.90471843e-01 4.00120527e-01 -1.48240075e-01 9.56688046e-01
1.09021270e+00 -5.06636918e-01 -8.05374384e-02 -4.08663094e-01
2.34541953e-01 -1.38171628e-01 -1.50132203e+00 2.09189010e+00
9.37652737e-02 2.00167239e-01 2.29653195e-01 -9.19100106e-01
8.36854815e-01 -1.45227954e-01 7.91801870e-01 -5.91714561e-01
2.10100152e-02 5.27233243e-01 -1.36834338e-01 -2.58811474e-01
5.32739162e-01 4.07090217e-01 4.02964316e-02 -1.39260203e-01
2.41514128e-02 -6.42066821e-02 1.51940495e-01 5.01446307e-01
1.25925994e+00 2.59322077e-02 7.80187249e-02 -3.66922349e-01
5.03009856e-01 1.07861543e-02 9.48181033e-01 7.45447576e-01
-1.90484017e-01 1.08867729e+00 4.62382168e-01 -3.62887651e-01
-7.98305690e-01 -8.30315709e-01 3.80989797e-02 6.56543553e-01
4.83593345e-01 -5.11690378e-01 -5.14392734e-01 -7.12674320e-01
5.05277753e-01 2.14943007e-01 -4.87012476e-01 -1.70619726e-01
-4.56886411e-01 -6.08123779e-01 3.37053120e-01 2.52614498e-01
3.28194439e-01 -6.67458713e-01 1.27945632e-01 1.65358976e-01
-1.38282046e-01 -1.39709699e+00 -9.37604904e-01 -1.07858419e-01
-6.07435346e-01 -1.34191251e+00 -6.17512822e-01 -6.23607695e-01
9.37729716e-01 3.07342499e-01 1.26733005e+00 4.94007885e-01
-3.26248050e-01 2.97516465e-01 -1.72882602e-02 -3.58075388e-02
1.00971512e-01 3.59843671e-01 -7.84714520e-02 3.20347399e-01
2.17858940e-01 -8.04770529e-01 -8.41679990e-01 3.35090697e-01
-6.62705541e-01 7.56556615e-02 5.38118243e-01 8.60461533e-01
8.84178460e-01 -2.84439743e-01 3.16071719e-01 -8.44916642e-01
2.48376921e-01 -5.41094720e-01 -9.31122899e-01 1.74313009e-01
-6.77792430e-01 -2.24699043e-02 5.21974981e-01 -1.66959494e-01
-3.84884208e-01 3.94228935e-01 -9.36781317e-02 -9.30229306e-01
3.42200696e-01 5.23445666e-01 -4.80093390e-01 -4.74890143e-01
4.26246524e-01 3.30118015e-02 4.00246270e-02 -3.88943076e-01
2.45519087e-01 1.20214656e-01 1.01818395e+00 -5.42652369e-01
1.25379002e+00 6.09374464e-01 2.60124564e-01 -5.34770846e-01
-8.95442069e-01 -8.07988644e-01 -4.96799618e-01 -3.86120945e-01
4.59983051e-01 -1.08512640e+00 -6.25486076e-01 6.55894935e-01
-8.11567724e-01 -1.43556148e-01 -1.69661835e-01 2.31570721e-01
-4.45796698e-01 7.75620639e-01 -5.41234553e-01 -4.01679486e-01
-5.12012541e-01 -1.19677591e+00 1.48089182e+00 9.27155390e-02
6.81188405e-02 -7.97477067e-01 -6.31843582e-02 3.48146409e-01
2.84545422e-01 5.62080085e-01 4.57746625e-01 -5.77313185e-01
-9.57750380e-01 -4.80192363e-01 -4.97293979e-01 1.65067554e-01
-4.83501889e-02 2.62702871e-02 -8.20915759e-01 -6.19137228e-01
-4.48301345e-01 -7.18258917e-02 1.12216139e+00 2.43233517e-01
1.04954803e+00 -8.80735442e-02 -3.74162048e-01 1.26326263e+00
1.47422206e+00 -7.16627181e-01 4.87018645e-01 2.30068892e-01
1.12575579e+00 5.12242854e-01 6.35724783e-01 3.05152565e-01
5.34136653e-01 7.84204423e-01 8.58699977e-01 -2.31007218e-01
-1.12732954e-01 -4.03033495e-01 1.61264643e-01 9.50338721e-01
2.38069296e-01 1.45587875e-02 -8.14962566e-01 6.54414117e-01
-2.25483727e+00 -6.26318812e-01 -4.04427737e-01 2.35167289e+00
4.72544223e-01 5.73870055e-02 -8.29595178e-02 -2.76633114e-01
7.97398269e-01 1.70102715e-01 -7.48490989e-01 4.48470533e-01
-3.64471316e-01 -1.91653877e-01 6.57997072e-01 5.19872248e-01
-1.19636536e+00 9.16361988e-01 4.84180689e+00 1.00568712e+00
-9.51507807e-01 -3.25868763e-02 3.93478543e-01 -1.48821190e-01
-3.62029046e-01 2.42566049e-01 -5.80833673e-01 5.85070670e-01
4.21896279e-01 2.37179641e-02 3.63299429e-01 7.87518919e-01
6.17330372e-02 4.05447073e-02 -1.21055627e+00 1.22287428e+00
-2.33959011e-03 -1.15618181e+00 -1.52245790e-01 1.53173760e-01
8.63875806e-01 3.41208071e-01 -1.00411654e-01 2.26987392e-01
1.21128932e-01 -7.56410480e-01 5.56227088e-01 6.28316045e-01
5.89140713e-01 -5.62324762e-01 7.18741894e-01 2.50667989e-01
-1.27477121e+00 2.72957951e-01 -5.62664211e-01 3.37412685e-01
9.11366418e-02 1.05800974e+00 -5.01039922e-01 1.06890857e+00
8.91945720e-01 1.00998068e+00 -7.37015188e-01 1.47057378e+00
-9.66177583e-02 1.70041174e-01 -6.83416545e-01 5.85385978e-01
2.15248585e-01 -6.03840292e-01 7.50648439e-01 9.75487590e-01
4.13195580e-01 -2.72554696e-01 4.12223697e-01 8.36637914e-01
-2.76651561e-01 1.60438702e-01 -4.56779480e-01 3.90406132e-01
4.17201072e-01 1.51349378e+00 -6.87961638e-01 -1.55388892e-01
-4.82320696e-01 1.19696343e+00 6.80870354e-01 3.24885428e-01
-9.02326584e-01 -7.14259371e-02 8.41603518e-01 2.14168921e-01
1.73549026e-01 -6.61350638e-02 -1.72644019e-01 -1.43328083e+00
4.40540731e-01 -1.07526755e+00 5.40402651e-01 -5.11123776e-01
-1.79976785e+00 3.97000104e-01 -3.50199819e-01 -1.26960242e+00
-6.87692314e-02 -3.85990769e-01 -6.16883218e-01 7.59925961e-01
-1.46065629e+00 -1.28303802e+00 -8.51479173e-01 6.38620377e-01
1.08139850e-02 8.61472487e-02 2.41899759e-01 8.68026435e-01
-8.68762672e-01 8.33858252e-01 -1.17097937e-01 1.88948289e-01
1.02877128e+00 -1.31020653e+00 6.38419807e-01 1.22616029e+00
1.21625260e-01 4.95650083e-01 7.30215847e-01 -8.30069959e-01
-1.55810404e+00 -1.32885921e+00 6.70728087e-01 -1.38949007e-01
8.11781406e-01 -4.89288658e-01 -1.15372539e+00 6.71670794e-01
-1.88524604e-01 6.11186802e-01 -9.05235298e-03 -8.49664435e-02
-4.18524325e-01 -2.83706129e-01 -1.01235199e+00 6.45294487e-01
1.62133121e+00 -4.76188600e-01 -6.56061061e-03 5.95934391e-01
6.50323033e-01 -7.81696916e-01 -9.14474249e-01 6.37731373e-01
2.62507647e-01 -1.04326475e+00 1.02615929e+00 -5.21937847e-01
4.47331443e-02 -6.45663440e-01 -9.15432125e-02 -1.13560319e+00
-1.43271357e-01 -8.23446274e-01 -1.42934829e-01 1.29570985e+00
3.30986559e-01 -8.09227884e-01 1.09905183e+00 6.72492445e-01
-4.14086699e-01 -7.04666376e-01 -1.06620574e+00 -8.26459169e-01
-4.05013710e-01 -4.25558805e-01 5.91974556e-01 1.12136042e+00
-5.06942153e-01 -6.60621300e-02 -5.51660180e-01 5.74332237e-01
1.13333714e+00 1.99151263e-01 1.27439332e+00 -1.27380216e+00
-3.38999033e-01 -3.72215390e-01 -7.23609805e-01 -1.07284665e+00
2.67733902e-01 -1.15691471e+00 1.15999967e-01 -1.30043495e+00
1.20710462e-01 -4.25397843e-01 -1.26687035e-01 3.28526527e-01
-5.84364772e-01 2.30913758e-01 6.03569634e-02 3.51269424e-01
-6.47159457e-01 7.86854148e-01 1.03356099e+00 -3.15144360e-01
-2.13294807e-05 -7.23716570e-03 -3.67850423e-01 5.63973308e-01
3.54421020e-01 -5.77653468e-01 -1.22982375e-01 -3.20887268e-01
6.37492061e-01 -2.33054951e-01 5.97471893e-01 -1.00639367e+00
7.20232010e-01 1.57794714e-01 4.20402661e-02 -7.37696111e-01
2.64134198e-01 -1.13060677e+00 4.84240949e-01 1.96745545e-01
6.39030710e-02 1.96000531e-01 -1.57147065e-01 7.96926200e-01
-3.54290992e-01 2.28968281e-02 6.42541707e-01 1.15091652e-01
-8.42951119e-01 8.50163877e-01 4.32540864e-01 2.23059088e-01
8.96179616e-01 -3.94288480e-01 -2.39889279e-01 -3.33949983e-01
-5.56789160e-01 7.84500539e-01 8.12361121e-01 2.71855444e-01
5.73451817e-01 -1.44445825e+00 -9.39664245e-01 2.13374406e-01
5.18001914e-01 4.18895125e-01 3.68651927e-01 1.46539140e+00
-4.37841475e-01 -2.37291649e-01 1.51377007e-01 -8.38921845e-01
-1.06195283e+00 5.15515387e-01 6.10171795e-01 -4.15550351e-01
-8.77310455e-01 7.73861706e-01 2.97142845e-02 -6.30804300e-01
3.58208090e-01 -2.78549522e-01 4.62872654e-01 -4.73025776e-02
8.55367780e-02 3.83945137e-01 4.47901070e-01 -6.53010547e-01
-5.22952974e-01 5.63149214e-01 7.09658265e-02 4.86454487e-01
1.17097688e+00 -9.82049406e-02 -3.40930253e-01 -1.61225796e-02
1.31458771e+00 9.44698155e-02 -1.37298477e+00 -5.44894755e-01
3.36434782e-01 -5.95262945e-01 -1.34653509e-01 -1.94940224e-01
-1.65798604e+00 3.52401584e-01 2.49834433e-01 -7.84308687e-02
1.03620934e+00 -1.05369121e-01 9.31663513e-01 1.83337584e-01
3.40951264e-01 -1.10154426e+00 -1.40433744e-01 2.13991255e-01
9.94851708e-01 -1.49546516e+00 1.59026474e-01 -7.27847993e-01
-2.11313665e-01 8.09159696e-01 8.24281216e-01 -4.36387002e-01
5.85086524e-01 1.72547042e-01 -2.98372395e-02 -5.04814386e-01
-5.17737865e-01 -3.88942599e-01 6.98579013e-01 2.80194938e-01
8.06018058e-03 -1.52709827e-01 -5.96054010e-02 1.88242868e-01
-1.91252694e-01 -3.86627883e-01 4.75629494e-02 4.48143005e-01
-9.77112204e-02 -7.30884969e-01 -2.35495582e-01 6.02696955e-01
-2.03591704e-01 -1.85987487e-01 -4.68498588e-01 8.56896162e-01
7.85604864e-02 6.46933138e-01 4.42318916e-02 -3.19299728e-01
6.12473667e-01 -3.46672535e-01 2.26550534e-01 -3.52225780e-01
-8.10314655e-01 1.09550813e-02 -7.42040724e-02 -1.16614985e+00
-2.52338469e-01 -6.69718623e-01 -1.07305861e+00 -4.13017839e-01
-5.32728553e-01 -9.64514315e-02 2.89812386e-01 6.60356104e-01
6.87063634e-01 3.46201807e-01 3.32855672e-01 -9.45920646e-01
-6.25872552e-01 -7.98125982e-01 -5.46076715e-01 8.82597387e-01
2.62139648e-01 -5.70572019e-01 -6.83781087e-01 -5.23181498e-01] | [7.21090030670166, 6.386640548706055] |
03c31267-d028-4afa-a799-1b262f2171e4 | unitrans-unifying-model-transfer-and-data | 2007.07683 | null | https://arxiv.org/abs/2007.07683v1 | https://arxiv.org/pdf/2007.07683v1.pdf | UniTrans: Unifying Model Transfer and Data Transfer for Cross-Lingual Named Entity Recognition with Unlabeled Data | Prior works in cross-lingual named entity recognition (NER) with no/little labeled data fall into two primary categories: model transfer based and data transfer based methods. In this paper we find that both method types can complement each other, in the sense that, the former can exploit context information via language-independent features but sees no task-specific information in the target language; while the latter generally generates pseudo target-language training data via translation but its exploitation of context information is weakened by inaccurate translations. Moreover, prior works rarely leverage unlabeled data in the target language, which can be effortlessly collected and potentially contains valuable information for improved results. To handle both problems, we propose a novel approach termed UniTrans to Unify both model and data Transfer for cross-lingual NER, and furthermore, to leverage the available information from unlabeled target-language data via enhanced knowledge distillation. We evaluate our proposed UniTrans over 4 target languages on benchmark datasets. Our experimental results show that it substantially outperforms the existing state-of-the-art methods. | ['Jian-Guang Lou', 'Börje F. Karlsson', 'Zijia Lin', 'Qianhui Wu', 'Biqing Huang'] | 2020-07-15 | null | null | null | null | ['cross-lingual-ner'] | ['natural-language-processing'] | [-3.63576366e-03 -3.12041622e-02 -6.38784647e-01 -4.07882631e-01
-1.16649199e+00 -8.74128938e-01 6.80256903e-01 -1.60952449e-01
-6.81206107e-01 1.07792413e+00 3.10031056e-01 -4.14320052e-01
3.58767003e-01 -5.30171156e-01 -6.31732821e-01 -4.36916292e-01
5.45731544e-01 4.58186775e-01 2.76059527e-02 -3.71842146e-01
-1.04539737e-01 9.54008400e-02 -7.35142291e-01 2.02716172e-01
1.23983037e+00 7.15234637e-01 1.62829325e-01 -7.91167393e-02
-6.19878173e-01 7.67558038e-01 -2.33443335e-01 -4.43992764e-01
2.24308491e-01 -4.03337717e-01 -8.53758752e-01 -2.51713187e-01
2.38841977e-02 1.50801763e-01 -1.32961214e-01 1.05415213e+00
5.53849220e-01 3.05192936e-02 6.66541815e-01 -9.33482707e-01
-1.04391754e+00 8.87610435e-01 -4.69759375e-01 -4.19651419e-02
-3.36182676e-03 -1.16731048e-01 7.52064705e-01 -1.19481814e+00
7.19234765e-01 1.14941847e+00 7.48038828e-01 7.95379579e-01
-1.12438190e+00 -8.43702912e-01 2.45604843e-01 -1.82848498e-01
-1.50885689e+00 -7.27842689e-01 8.41072202e-01 -2.47068286e-01
8.13463986e-01 -1.39452860e-01 -2.97893547e-02 1.26420784e+00
-3.10190529e-01 1.04408288e+00 1.35339880e+00 -7.15756774e-01
8.47354680e-02 5.32516897e-01 -1.66147873e-02 6.27331078e-01
1.45053849e-01 2.30151132e-01 -4.47317332e-01 -3.48970369e-02
4.50591534e-01 -2.81510446e-02 -3.75057936e-01 -1.64058968e-01
-1.39595449e+00 6.36924803e-01 2.80890852e-01 6.49398088e-01
-2.01601297e-01 -3.48574549e-01 5.75518191e-01 3.86272401e-01
6.68290317e-01 2.72223920e-01 -1.10106313e+00 -2.18787324e-02
-8.05052161e-01 -3.27094316e-01 9.35818315e-01 1.36683941e+00
1.13738668e+00 1.54274702e-01 -9.64515656e-02 9.21980858e-01
2.84259051e-01 6.59144342e-01 9.62444544e-01 -2.08932430e-01
1.10793388e+00 7.48110414e-01 1.76839396e-01 -3.48039001e-01
5.60610089e-04 -3.41634840e-01 -7.07604408e-01 -5.40942490e-01
1.43579528e-01 -6.06110275e-01 -1.10316479e+00 1.90505445e+00
3.02089930e-01 2.01612115e-01 6.27206564e-01 4.93356496e-01
8.45883250e-01 6.00633681e-01 3.32261354e-01 -2.56399423e-01
1.22634661e+00 -1.27179611e+00 -8.26389968e-01 -4.67886180e-01
1.04614162e+00 -8.45272243e-01 1.10244489e+00 -1.60701483e-01
-5.83938301e-01 -5.08436620e-01 -8.18342149e-01 -1.49704248e-01
-8.16852331e-01 4.24924105e-01 3.98151070e-01 7.80523539e-01
-8.10437620e-01 3.43674690e-01 -7.74746239e-01 -4.46620047e-01
1.50177985e-01 1.34668291e-01 -4.42872435e-01 -4.55579072e-01
-1.43481839e+00 8.52174580e-01 8.24247003e-01 1.09332591e-01
-7.37370193e-01 -7.19723403e-01 -1.14486861e+00 -1.86377838e-01
4.95327532e-01 -4.28711802e-01 1.21005416e+00 -9.79165077e-01
-1.52244222e+00 7.14057446e-01 -2.93702394e-01 -1.01610452e-01
4.41210091e-01 -4.21912581e-01 -6.71891749e-01 -5.16504228e-01
2.66316980e-01 4.82967496e-01 2.62514591e-01 -1.32978439e+00
-7.01179862e-01 -3.17549169e-01 -2.11520061e-01 2.94883251e-01
-7.37532794e-01 2.83986926e-02 -6.86674178e-01 -7.63445079e-01
-2.71769881e-01 -8.13399553e-01 -1.91625178e-01 -5.30661225e-01
-5.19473255e-01 -4.04946148e-01 8.09710264e-01 -6.30476058e-01
1.24203146e+00 -1.92906702e+00 -1.17105998e-01 -1.84296086e-01
-2.43987828e-01 7.19802260e-01 -3.27687651e-01 7.10130215e-01
2.03218341e-01 5.02784610e-01 -2.60063916e-01 -3.81547183e-01
-6.94165155e-02 2.91632682e-01 -3.53502154e-01 9.53328311e-02
2.70950109e-01 1.03355682e+00 -1.12375295e+00 -5.89640915e-01
1.82338301e-02 5.31281173e-01 -1.98331386e-01 3.86124969e-01
-6.59237579e-02 7.08760798e-01 -8.51972640e-01 6.30928695e-01
5.55694163e-01 -1.97925344e-02 3.75189841e-01 -2.32551679e-01
-3.68470997e-01 5.73765635e-01 -9.06579614e-01 1.94295096e+00
-9.43822086e-01 7.34592006e-02 -1.24929570e-01 -8.94048691e-01
1.06054151e+00 7.19824255e-01 1.15650445e-01 -5.15512407e-01
3.22076455e-02 6.08014464e-01 -3.19853038e-01 -2.51025051e-01
3.56969893e-01 -3.47941279e-01 -2.90558130e-01 3.50881904e-01
2.93675780e-01 2.92166084e-01 -3.57090794e-02 -2.09649161e-01
8.14691067e-01 4.63734835e-01 5.56252837e-01 -2.07608834e-01
5.81638813e-01 1.07228838e-01 1.12056315e+00 6.30915284e-01
-2.35676304e-01 1.32116020e-01 -1.53384089e-01 -1.18524872e-01
-6.86556101e-01 -8.81852627e-01 -2.30927393e-02 1.22118485e+00
9.10627842e-02 -2.80468702e-01 -6.59998894e-01 -1.27354097e+00
-3.88211071e-01 8.40865493e-01 -4.96700883e-01 3.37653188e-03
-6.49340272e-01 -5.97169280e-01 8.49748552e-01 6.55789375e-01
7.08036780e-01 -1.12731051e+00 1.38865277e-01 3.41899842e-01
-3.93528402e-01 -1.33738613e+00 -7.65908360e-01 2.36374930e-01
-1.01572430e+00 -7.55641043e-01 -8.30678940e-01 -1.11356986e+00
7.01841295e-01 2.34551653e-01 1.12619150e+00 -3.09472173e-01
1.87504709e-01 2.01469168e-01 -6.52516663e-01 -4.03547317e-01
-4.52038556e-01 5.06374002e-01 1.35463923e-01 1.04363561e-01
7.52128243e-01 -4.72556114e-01 -1.71790808e-01 3.89599949e-01
-7.99998403e-01 -1.52568640e-02 9.23463464e-01 1.05315793e+00
7.43862569e-01 -2.25930586e-01 7.48330235e-01 -1.29242957e+00
3.62388939e-01 -7.21525371e-01 -2.90666640e-01 7.22610474e-01
-6.44719660e-01 4.25873697e-01 9.18447673e-01 -4.98039275e-01
-1.53657722e+00 1.46591604e-01 3.65620181e-02 -4.20569152e-01
-3.17417562e-01 8.20369542e-01 -5.68924665e-01 7.00805783e-02
4.49974775e-01 5.10208726e-01 -6.36828542e-01 -1.08570743e+00
6.54462159e-01 8.98776770e-01 4.43608284e-01 -8.31028819e-01
8.15973222e-01 2.59989470e-01 -5.88889658e-01 -5.94581783e-01
-1.15334165e+00 -6.25554442e-01 -9.15277898e-01 3.37517649e-01
7.79763579e-01 -1.04695189e+00 6.77258223e-02 1.68239385e-01
-1.26418495e+00 -8.81788805e-02 -6.96147829e-02 7.35251665e-01
-1.71754658e-01 2.77632803e-01 -6.53081357e-01 -7.92963207e-01
-4.45284486e-01 -1.00532258e+00 1.08360803e+00 2.74296731e-01
3.12820673e-01 -1.35926378e+00 3.65417600e-01 2.97985107e-01
3.09910029e-01 4.31784838e-02 9.07901525e-01 -1.25947642e+00
-1.85940146e-01 -2.03981429e-01 -1.64603457e-01 4.82323647e-01
5.95519364e-01 -5.61667323e-01 -9.55620050e-01 -2.97864765e-01
-1.46554068e-01 -6.15351379e-01 7.20345497e-01 -2.65563846e-01
5.14738142e-01 -3.32813501e-01 -4.65148985e-01 4.62736964e-01
1.50096118e+00 1.37826756e-01 3.45749766e-01 2.10688889e-01
9.89823520e-01 7.13964105e-01 7.41405547e-01 -1.18537083e-01
6.44004524e-01 6.92957997e-01 -2.17463359e-01 -4.14647222e-01
-1.92018211e-01 -7.10747540e-01 6.53458476e-01 1.57102609e+00
-4.79255207e-02 -2.36078545e-01 -1.03386891e+00 8.67497265e-01
-1.87203693e+00 -5.07237494e-01 2.90080816e-01 2.22141171e+00
1.15879047e+00 -1.87091008e-01 -2.93010831e-01 -5.53241909e-01
1.01599312e+00 -1.19417936e-01 -6.58160388e-01 -7.49835139e-03
-1.16221510e-01 2.21000105e-01 6.58150494e-01 3.71076345e-01
-1.20821762e+00 1.51573122e+00 5.33311605e+00 1.02889311e+00
-1.13083458e+00 4.70740229e-01 2.82565951e-01 4.26857501e-01
-3.47981036e-01 2.30008245e-01 -1.20530772e+00 3.17596793e-01
9.91650164e-01 -3.85083407e-01 2.60382760e-02 9.91653442e-01
1.05984006e-02 5.04746675e-01 -1.18972409e+00 7.82633007e-01
5.90375401e-02 -9.04238999e-01 2.08344400e-01 2.23518908e-02
8.32566023e-01 4.40609396e-01 -1.94890678e-01 8.28647554e-01
6.75610602e-01 -7.54690528e-01 4.35296029e-01 3.40501517e-01
8.93201172e-01 -7.20892549e-01 1.01326990e+00 6.73491418e-01
-1.39277983e+00 3.04597795e-01 -3.03919524e-01 2.73740649e-01
2.82778621e-01 3.99817199e-01 -8.74983907e-01 1.16191876e+00
3.74195427e-01 8.15839410e-01 -4.18529689e-01 7.03640640e-01
-4.47626501e-01 9.60648596e-01 -2.06450790e-01 6.14673309e-02
3.01863670e-01 -1.96236312e-01 3.46438885e-01 1.36727595e+00
4.08583373e-01 -1.61309585e-01 6.56659722e-01 6.58381760e-01
-4.54417855e-01 6.40820205e-01 -7.73907840e-01 -1.60853490e-01
7.29996026e-01 1.07451999e+00 -3.69835019e-01 -3.76171768e-01
-7.62296140e-01 1.07784212e+00 5.91599464e-01 4.32534844e-01
-4.07566100e-01 -5.58686793e-01 3.27144861e-01 -3.01375419e-01
4.32245404e-01 -2.33208120e-01 -2.67282203e-02 -1.68569684e+00
1.94360897e-01 -7.26225793e-01 5.62962890e-01 -3.47353190e-01
-1.65766239e+00 9.04087365e-01 -2.31682673e-01 -1.34296966e+00
-2.92029291e-01 -4.97152954e-01 -2.84763187e-01 1.02054250e+00
-2.12470031e+00 -1.77012765e+00 2.42671043e-01 6.58637941e-01
7.59553313e-01 -2.76437819e-01 1.02204168e+00 5.42507827e-01
-5.73248625e-01 8.21242809e-01 4.15538192e-01 6.28039062e-01
1.06971407e+00 -1.06128442e+00 4.38654244e-01 9.63723183e-01
3.01394016e-01 8.90252113e-01 8.03071484e-02 -8.18772674e-01
-1.45442939e+00 -1.60336447e+00 1.31465900e+00 -6.21884227e-01
6.68031454e-01 -4.77787405e-01 -9.46988881e-01 1.00694072e+00
1.88931867e-01 1.64819553e-01 1.01256061e+00 3.59891593e-01
-6.43123329e-01 6.48918971e-02 -9.83006895e-01 5.85866213e-01
9.88704205e-01 -6.39944077e-01 -9.85246062e-01 1.09958552e-01
9.20490861e-01 -6.69355690e-02 -8.44959021e-01 4.23929363e-01
1.59764007e-01 -3.82316977e-01 5.55899560e-01 -8.69594693e-01
-9.39348340e-02 -4.16206002e-01 -2.96136171e-01 -1.38166201e+00
6.52690455e-02 -5.29830873e-01 2.15010360e-01 1.94160998e+00
7.93019116e-01 -7.66307235e-01 3.83012503e-01 3.99720460e-01
-3.15303117e-01 -4.33428258e-01 -8.89270246e-01 -1.17366731e+00
5.14901578e-01 -3.93433183e-01 7.10541368e-01 1.41572833e+00
-1.00617245e-01 7.37351894e-01 -5.89232564e-01 9.14688632e-02
5.06098270e-01 4.93031181e-02 5.42875707e-01 -1.04667819e+00
-1.44580565e-02 4.32437174e-02 -1.04112409e-01 -1.21976721e+00
5.19906640e-01 -1.28945923e+00 2.40376264e-01 -1.30817795e+00
2.33527049e-01 -8.19934249e-01 -6.62065804e-01 9.82528925e-01
-3.51463109e-01 4.89306748e-02 5.14200665e-02 4.20893937e-01
-6.48609221e-01 8.64890337e-01 9.76060331e-01 5.10863066e-02
-3.98250461e-01 -2.49640793e-01 -7.89326310e-01 7.01498985e-01
5.62861085e-01 -8.59818101e-01 -3.78847510e-01 -7.08434224e-01
-9.89823937e-02 -8.87814984e-02 -1.59333244e-01 -7.06120431e-01
1.12773582e-01 -3.59920919e-01 6.19428419e-02 -2.78364718e-01
-1.13472633e-01 -7.69196987e-01 -2.26428747e-01 1.09777629e-01
-3.32045734e-01 -1.72419474e-01 1.66199848e-01 7.20208943e-01
-4.33576435e-01 -9.19124186e-02 5.56837618e-01 -1.58915803e-01
-8.81694138e-01 3.84310365e-01 -2.44057309e-02 5.42125523e-01
5.86264372e-01 1.65007621e-01 -2.37758696e-01 -7.44405910e-02
-4.08637226e-01 2.58258432e-01 3.16095799e-01 7.94210255e-01
6.89350208e-03 -1.66566253e+00 -8.71808350e-01 -6.33287197e-03
5.35994589e-01 -2.83553839e-01 8.52338821e-02 6.51101828e-01
2.18703076e-01 6.63364232e-01 1.14458039e-01 -3.47396344e-01
-7.73624241e-01 6.08508587e-01 1.34870470e-01 -6.74262226e-01
-2.60347217e-01 4.38549459e-01 5.07541001e-01 -1.18738651e+00
-1.04864441e-01 -7.18258396e-02 -2.57119656e-01 -1.48084402e-01
4.56592590e-01 -2.29288023e-02 1.60761431e-01 -8.04646909e-01
-3.64823073e-01 5.90873003e-01 -3.52553904e-01 -1.47438228e-01
1.21664059e+00 -3.25884104e-01 1.05176046e-01 6.70828342e-01
1.25078547e+00 2.49319404e-01 -8.82198155e-01 -8.80740583e-01
4.48947161e-01 -1.03781655e-01 -9.60902032e-03 -1.07618511e+00
-9.04575229e-01 1.01134753e+00 3.77544522e-01 -1.99749842e-01
1.05259633e+00 2.62115467e-02 1.10374713e+00 6.70355976e-01
5.66830099e-01 -1.22270930e+00 -3.75917852e-01 7.88176060e-01
4.70870405e-01 -1.40410590e+00 -4.46383089e-01 -5.37897348e-01
-8.18967164e-01 8.31697881e-01 6.77190900e-01 3.22101593e-01
6.95812523e-01 1.83939338e-01 5.32824636e-01 1.57224357e-01
-5.84728718e-01 -3.84714097e-01 3.32423329e-01 7.90201426e-01
8.41919661e-01 6.47651181e-02 -3.67713243e-01 1.13122189e+00
6.41982332e-02 1.15744635e-01 2.51452550e-02 1.18657041e+00
-1.97524726e-01 -1.70955741e+00 -1.08737551e-01 3.98938805e-02
-6.08202219e-01 -5.55225909e-01 -4.70536113e-01 9.46862996e-01
1.36185706e-01 9.62444365e-01 -4.57999200e-01 -2.08656847e-01
4.46583837e-01 4.90544438e-01 6.56647012e-02 -8.95494998e-01
-5.99421680e-01 1.43107429e-01 2.47647271e-01 -1.75322875e-01
-5.38879693e-01 -3.83516163e-01 -1.09358239e+00 2.44006842e-01
-5.71274698e-01 5.76336741e-01 6.41076207e-01 1.21474540e+00
6.30292654e-01 1.40829086e-01 6.66133404e-01 -4.38147426e-01
-5.37825763e-01 -1.10172653e+00 -4.42314327e-01 2.47695312e-01
1.42721266e-01 -5.91603100e-01 -2.53390819e-01 3.59795749e-01] | [9.960225105285645, 9.654614448547363] |
acefef7f-f7ef-42c8-a269-8b9102ae2175 | improving-j-divergence-of-brain-connectivity | 2012.11240 | null | https://arxiv.org/abs/2012.11240v2 | https://arxiv.org/pdf/2012.11240v2.pdf | Improving J-divergence of brain connectivity states by graph Laplacian denoising | Functional connectivity (FC) can be represented as a network, and is frequently used to better understand the neural underpinnings of complex tasks such as motor imagery (MI) detection in brain-computer interfaces (BCIs). However, errors in the estimation of connectivity can affect the detection performances. In this work, we address the problem of denoising common connectivity estimates to improve the detectability of different connectivity states. Specifically, we propose a denoising algorithm that acts on the network graph Laplacian, which leverages recent graph signal processing results. Further, we derive a novel formulation of the Jensen divergence for the denoised Laplacian under different states. Numerical simulations on synthetic data show that the denoising method improves the Jensen divergence of connectivity patterns corresponding to different task conditions. Furthermore, we apply the Laplacian denoising technique to brain networks estimated from real EEG data recorded during MI-BCI experiments. Using our novel formulation of the J-divergence, we are able to quantify the distance between the FC networks in the motor imagery and resting states, as well as to understand the contribution of each Laplacian variable to the total J-divergence between two states. Experimental results on real MI-BCI EEG data demonstrate that the Laplacian denoising improves the separation of motor imagery and resting mental states, and shortens the time interval required for connectivity estimation. We conclude that the approach shows promise for the robust detection of connectivity states while being appealing for implementation in real-time BCI applications. | ['Stefania Colonnese', 'Fabrizio De Vico Fallani', 'Danielle S. Bassett', 'Marie-Constance Corsi', 'Gaetano Scarano', 'Tiziana Cattai'] | 2020-12-21 | null | null | null | null | ['connectivity-estimation'] | ['graphs'] | [ 5.37956953e-01 -1.18423887e-01 2.91052818e-01 -1.58758070e-02
-1.28539607e-01 -3.48323882e-01 4.60382581e-01 -9.84452739e-02
-3.57239574e-01 6.00718975e-01 1.30304620e-01 -1.51685297e-01
-7.47051179e-01 -4.92732793e-01 -4.37685609e-01 -7.44677007e-01
-7.53417134e-01 1.12896934e-02 -2.34854385e-01 -9.70569849e-02
1.85960889e-01 5.03834486e-01 -1.13365638e+00 -1.14442175e-02
1.22799528e+00 8.83805811e-01 2.21384108e-01 3.62663865e-01
3.76304507e-01 2.81694770e-01 -6.54081583e-01 -8.23569372e-02
8.26902762e-02 -7.56836116e-01 -6.44952059e-01 -7.77404308e-02
2.56899297e-01 2.98476845e-01 -5.19659340e-01 1.40415430e+00
4.89268929e-01 2.47646868e-01 8.20616543e-01 -1.16884172e+00
3.85992080e-02 4.37195897e-01 -6.45722866e-01 7.74549067e-01
3.16845059e-01 8.77295434e-02 6.05665922e-01 -5.79404593e-01
4.20069695e-01 9.32165980e-01 6.58737481e-01 1.22103661e-01
-1.60822439e+00 -8.51979017e-01 -8.66091475e-02 6.04838550e-01
-1.69624007e+00 -4.06661391e-01 8.09387922e-01 -6.50847852e-01
6.42436862e-01 2.27495566e-01 9.66261744e-01 1.02626574e+00
6.77341878e-01 2.87649542e-01 1.42563713e+00 -1.26781136e-01
1.17408238e-01 -4.67862397e-01 3.35129857e-01 4.65267718e-01
4.12108272e-01 -9.09795538e-02 -5.73913217e-01 -1.17812328e-01
7.57645488e-01 -2.89617538e-01 -9.29438710e-01 -1.38463616e-01
-1.46153975e+00 7.08206296e-01 4.08331037e-01 6.66127741e-01
-7.13234901e-01 8.70067552e-02 3.60963821e-01 4.73722130e-01
6.09302938e-01 3.13014150e-01 3.08610111e-01 -7.55523294e-02
-1.25070632e+00 -2.75046766e-01 7.37858891e-01 4.55703318e-01
5.10964274e-01 -3.16059478e-02 -2.27125540e-01 6.29499793e-01
-3.59302424e-02 3.90215069e-01 4.28635597e-01 -8.31237495e-01
2.58693874e-01 3.23808283e-01 -4.29694593e-01 -1.21835601e+00
-6.44575119e-01 -7.57839322e-01 -1.43408227e+00 -1.34041242e-03
3.18764836e-01 -2.80614972e-01 -4.68165129e-01 1.92364454e+00
-2.08244070e-01 5.49014807e-01 -3.53185028e-01 8.93532455e-01
2.37100527e-01 2.07916856e-01 -1.58896282e-01 -4.94125038e-01
9.79970276e-01 -1.29607648e-01 -9.39984858e-01 -2.45766804e-01
3.36801410e-01 -1.27520248e-01 5.29835761e-01 5.69377184e-01
-1.21174002e+00 -3.79803985e-01 -1.07609081e+00 5.73208272e-01
-8.36407319e-02 -3.46109718e-02 3.95842761e-01 5.39077401e-01
-1.27594805e+00 8.57712030e-01 -1.04922628e+00 -3.80790234e-01
2.78878331e-01 5.59196889e-01 -5.43335497e-01 -6.65187985e-02
-1.32198060e+00 1.08441329e+00 3.36314946e-01 5.33382416e-01
-6.17645204e-01 -4.52028394e-01 -6.25555694e-01 1.27493486e-01
1.16251163e-01 -4.77200717e-01 2.12908596e-01 -1.14164054e+00
-1.16838360e+00 5.73284328e-01 -1.43984720e-01 -4.40864503e-01
4.27520603e-01 2.79245794e-01 -5.45329213e-01 4.52124387e-01
1.11778818e-01 3.29305232e-01 1.00077474e+00 -8.82596314e-01
1.87932774e-01 -5.29929817e-01 -3.08243752e-01 8.10084045e-02
-1.56826213e-01 -3.72013301e-01 -9.25194845e-02 -5.72758853e-01
4.16831493e-01 -8.96764040e-01 3.49421240e-02 -4.25901145e-01
-5.32492757e-01 1.54268831e-01 2.43451223e-01 -8.61855268e-01
1.14151454e+00 -2.31859279e+00 5.99995434e-01 1.00050867e+00
5.66487193e-01 -5.55774644e-02 -3.40384543e-01 2.78996527e-01
-4.64143127e-01 1.82730854e-02 -5.47328770e-01 1.71767503e-01
-3.84485453e-01 1.01058036e-01 2.62438893e-01 1.06679130e+00
1.75530657e-01 8.82107258e-01 -7.89693654e-01 -5.54095916e-02
-1.53332949e-04 7.02076793e-01 -2.98255652e-01 -5.03327139e-02
7.06793249e-01 8.47680748e-01 3.85511070e-02 -2.15687335e-01
6.52873755e-01 3.63399535e-02 4.81639862e-01 -5.80479383e-01
5.89598566e-02 1.27366781e-01 -1.16413987e+00 1.60304308e+00
-2.96653450e-01 9.64587927e-01 4.30867314e-01 -1.70300853e+00
6.12973869e-01 2.63379395e-01 6.92955136e-01 -6.89039648e-01
3.10844779e-01 -5.38182678e-03 8.07473600e-01 -2.70091653e-01
-2.82769531e-01 -2.96067204e-02 2.92198598e-01 4.73948359e-01
2.63410032e-01 -9.06720236e-02 4.76115346e-01 3.73464227e-01
1.35437322e+00 -4.73450661e-01 8.33764225e-02 -9.14034307e-01
6.16734385e-01 -5.15997887e-01 1.24616608e-01 6.08375371e-01
-1.98973373e-01 3.05376112e-01 8.15927744e-01 3.90179962e-01
-1.90983981e-01 -1.24093008e+00 -2.42862493e-01 3.56286168e-01
2.83150464e-01 -1.39248878e-01 -1.04792404e+00 -7.01212287e-02
-1.49169624e-01 6.20104849e-01 -6.36792898e-01 -7.98107803e-01
-4.39642996e-01 -9.28142309e-01 4.30042148e-01 6.21797517e-02
3.85639340e-01 -7.84155548e-01 -4.56806809e-01 2.12887883e-01
-5.71541846e-01 -1.02143955e+00 -4.51049387e-01 3.71447206e-01
-1.06624925e+00 -1.26043105e+00 -6.95962965e-01 -6.41914248e-01
6.76607788e-01 2.98858702e-01 7.98619390e-01 6.96951449e-02
-3.39662760e-01 7.97156215e-01 -2.66473498e-02 2.10783169e-01
-3.10591310e-01 -1.51171237e-01 2.19664454e-01 3.47570539e-01
1.02356248e-01 -9.72808421e-01 -7.97645867e-01 3.11761051e-01
-9.18287933e-01 -6.45977184e-02 6.06013954e-01 7.13175476e-01
3.37408602e-01 4.82275158e-01 7.89785802e-01 -3.75080317e-01
1.33465087e+00 -7.60085523e-01 -1.76127464e-01 1.28721341e-01
-6.26805007e-01 1.48762703e-01 2.35633582e-01 -6.01204157e-01
-7.98075438e-01 -2.77814150e-01 2.69418359e-01 -1.10095114e-01
1.60370637e-02 9.34088051e-01 -8.61592665e-02 -3.74637663e-01
5.19578338e-01 2.11257160e-01 3.04654002e-01 -1.85854733e-01
2.46034145e-01 4.37739313e-01 7.31431246e-01 -2.92488992e-01
3.37680757e-01 4.89343315e-01 1.60160467e-01 -1.25131392e+00
-2.35108405e-01 -4.37643290e-01 -7.99669266e-01 -5.54851413e-01
8.00469697e-01 -7.00235903e-01 -7.48041272e-01 3.39864403e-01
-9.02845621e-01 -3.73658538e-01 -3.68207768e-02 7.31799603e-01
-5.27843475e-01 7.73714781e-01 -5.90086699e-01 -5.94690561e-01
-3.81825060e-01 -1.17798340e+00 4.58406776e-01 -1.87254757e-01
-3.45044464e-01 -1.30403078e+00 4.96942066e-02 -2.27578077e-02
4.49461609e-01 3.80877227e-01 1.11296248e+00 -1.95139065e-01
4.44338471e-03 -1.75854862e-01 -1.55417934e-01 4.90851402e-01
1.40134007e-01 -4.19320017e-01 -6.05198741e-01 -4.58067834e-01
4.30394441e-01 3.38271618e-01 9.10379171e-01 7.83110619e-01
9.31022882e-01 -7.36999363e-02 -3.58801275e-01 3.19449782e-01
1.17107034e+00 -9.67353135e-02 8.73014450e-01 -2.04844236e-01
5.18178821e-01 7.25019991e-01 -6.50559738e-02 1.39193833e-01
-1.28962189e-01 4.21360016e-01 3.34246844e-01 2.37606838e-02
-1.77781016e-01 3.86213243e-01 5.23367107e-01 1.24174523e+00
-3.77748847e-01 -1.96246803e-02 -8.71851683e-01 4.60818499e-01
-1.66435599e+00 -9.30024862e-01 -5.07053077e-01 2.48164701e+00
6.79620028e-01 1.85109720e-01 -8.68585259e-02 3.50056350e-01
9.41612244e-01 -2.08854139e-01 -5.10012627e-01 -5.54257184e-02
-1.46875724e-01 6.38167977e-01 4.07419741e-01 4.71473336e-01
-5.72963178e-01 2.91718900e-01 6.53711033e+00 7.05419123e-01
-1.12517786e+00 4.17925239e-01 3.19930613e-01 -1.57157481e-01
4.26235422e-02 -2.81083941e-01 2.23785266e-01 4.37373430e-01
1.19099951e+00 -3.67287576e-01 9.77310181e-01 -1.20289072e-01
5.81560552e-01 -5.52207828e-01 -1.02525389e+00 1.16774774e+00
9.85242724e-02 -8.15042019e-01 -2.91350156e-01 2.94938385e-01
5.38148284e-01 2.63158441e-01 -1.12908043e-01 -3.05128366e-01
-3.82637769e-01 -1.05000615e+00 5.02918243e-01 9.39018548e-01
1.01262069e+00 -7.12919235e-01 7.57792890e-01 4.25664186e-01
-1.09054160e+00 7.10498095e-02 -1.06836036e-01 -2.00478137e-01
1.13944083e-01 1.00316310e+00 -1.86064303e-01 5.09116650e-01
3.10529172e-01 1.01022601e+00 -5.61280012e-01 1.16289878e+00
-1.70841724e-01 6.13139808e-01 -3.01298141e-01 3.23446989e-01
-1.91816986e-01 -7.97025383e-01 7.12402225e-01 1.11873651e+00
3.77925575e-01 2.85046771e-02 -2.76939452e-01 1.22866821e+00
1.89410508e-01 -2.12713882e-01 -5.62367439e-01 -1.52271539e-01
5.26798144e-02 1.27841496e+00 -9.50077057e-01 -4.95471284e-02
-2.56951749e-01 1.27822578e+00 2.97559798e-01 8.10777903e-01
-6.03267968e-01 -3.92484814e-01 6.22825027e-01 2.04907477e-01
-1.01939283e-01 -5.97628593e-01 -2.53846020e-01 -1.19434547e+00
-3.57170813e-02 -7.00539529e-01 1.12444852e-02 -5.96475065e-01
-1.25007236e+00 5.59404433e-01 3.49050164e-01 -9.28642929e-01
-8.75772983e-02 -5.63912690e-01 -8.98605466e-01 1.06779277e+00
-1.18291986e+00 -2.68442631e-01 -4.69689041e-01 9.33687449e-01
-1.10786647e-01 4.06782359e-01 5.29415965e-01 4.53923583e-01
-7.26496458e-01 2.49362946e-01 2.02373430e-01 -1.21164806e-02
3.96268874e-01 -9.50496256e-01 3.72560509e-02 1.06334352e+00
2.28574455e-01 8.47749829e-01 4.96635169e-01 -7.31830716e-01
-1.25907731e+00 -7.35075593e-01 6.85368404e-02 2.45849133e-01
8.80202115e-01 -3.43257785e-01 -8.73033941e-01 2.37203389e-01
3.93572927e-01 -2.38267720e-01 4.42419082e-01 -2.38834143e-01
1.88247040e-01 1.77407395e-02 -1.04544318e+00 4.59694862e-01
1.27895355e+00 -8.16723704e-01 -4.19933587e-01 4.10914153e-01
-2.43726417e-01 2.56866902e-01 -1.11200404e+00 3.25412959e-01
3.80631328e-01 -9.79016960e-01 8.70720625e-01 -1.97126940e-01
-1.18644036e-01 2.65366770e-02 3.90532762e-01 -1.85193586e+00
-5.74358284e-01 -5.77683687e-01 -1.10151686e-01 7.58204639e-01
1.14550859e-01 -9.39986765e-01 1.74519047e-01 4.47780401e-01
1.28837109e-01 -3.77372473e-01 -1.23073959e+00 -9.06918645e-01
-1.95083488e-02 -5.26013672e-01 -2.32807100e-01 8.02064359e-01
4.88797814e-01 4.02880996e-01 3.87857147e-02 3.09520103e-02
7.45176733e-01 -3.57860923e-01 3.94423530e-02 -1.47079325e+00
-1.15483515e-02 -6.88136041e-01 -7.16063201e-01 -7.56358147e-01
4.34600919e-01 -1.28844154e+00 2.87573896e-02 -1.58425522e+00
2.12266356e-01 -6.68214560e-02 -5.33008337e-01 1.93752974e-01
-9.44355354e-02 3.71728152e-01 -9.28519070e-02 3.44613224e-01
1.27601810e-02 3.72199863e-01 1.00852561e+00 -2.17582196e-01
-1.61527336e-01 -2.17655897e-01 -3.89835060e-01 6.18317544e-01
8.69139850e-01 -4.74008143e-01 -4.90754455e-01 -1.91004798e-02
2.21858785e-01 7.33768418e-02 4.93571281e-01 -1.36469626e+00
2.47244626e-01 4.12953377e-01 2.38872051e-01 -1.01074338e-01
2.63183802e-01 -7.43071318e-01 4.22950387e-01 7.85362005e-01
-1.62814170e-01 -1.11996800e-01 1.25846624e-01 8.00589323e-01
-1.68488622e-01 -1.53847933e-01 7.42087305e-01 9.85417068e-02
-3.13480973e-01 4.31082584e-02 -8.83255422e-01 2.54804730e-01
8.14452410e-01 -2.44330704e-01 2.99332943e-02 -6.94057047e-01
-9.61395264e-01 -6.32433370e-02 -4.39602463e-03 2.38051284e-02
6.72994018e-01 -1.11000979e+00 -6.54366255e-01 4.42354202e-01
-4.00971472e-01 -8.29896212e-01 1.94908947e-01 1.99566698e+00
-2.18366921e-01 1.03160039e-01 -4.44248408e-01 -6.70978725e-01
-9.51897323e-01 2.06164539e-01 5.69841683e-01 -7.91926086e-02
-5.64695895e-01 5.28448164e-01 1.30585060e-01 2.84953773e-01
1.14409238e-01 -4.87985134e-01 -3.60767186e-01 3.51368070e-01
2.82830924e-01 6.93518400e-01 1.77343681e-01 -8.54098439e-01
-5.54011643e-01 4.59414601e-01 4.88380581e-01 -3.37537140e-01
1.11970770e+00 -3.60101521e-01 -5.61827719e-01 4.07388270e-01
1.30824614e+00 -2.22006544e-01 -9.91913915e-01 8.09295475e-02
8.12635049e-02 -6.20742366e-02 4.91010249e-01 -5.85040092e-01
-1.44804621e+00 9.78198469e-01 9.66150284e-01 5.00078738e-01
1.30729055e+00 -2.78117120e-01 4.95397687e-01 2.37490252e-01
4.29991841e-01 -7.86480844e-01 -2.02150270e-01 1.31655872e-01
9.98248398e-01 -7.65718162e-01 5.82929254e-02 -3.58122230e-01
-4.16169733e-01 1.15889180e+00 -5.33151589e-02 -3.28735977e-01
9.41901326e-01 1.56581327e-01 -3.37111980e-01 -6.18555784e-01
-9.20808017e-02 -3.03483456e-01 6.13491476e-01 5.60117960e-01
3.66478533e-01 2.21695527e-01 -6.91136241e-01 3.40266019e-01
6.33034334e-02 -2.94462383e-01 6.45335972e-01 4.83924747e-01
-1.12006225e-01 -6.03156209e-01 -2.54158288e-01 8.56181204e-01
-2.92847216e-01 -3.37739855e-01 -5.28110921e-01 5.30154765e-01
-1.36359751e-01 1.27102995e+00 4.80679870e-02 -3.38925481e-01
2.37293854e-01 1.52679458e-01 7.00156629e-01 -4.14605558e-01
-4.62815285e-01 -2.75555626e-02 -1.17825612e-01 -8.27529192e-01
-7.75780439e-01 -5.72628438e-01 -1.11446488e+00 -1.32772267e-01
-7.21212089e-01 3.27443242e-01 7.28244305e-01 9.34589148e-01
5.36737800e-01 6.57800913e-01 3.42383742e-01 -1.14950728e+00
-9.79777724e-02 -1.14614677e+00 -9.77386236e-01 5.89115024e-01
1.58407271e-01 -9.42543864e-01 -7.03934908e-01 -9.88730937e-02] | [12.51935863494873, 3.4235363006591797] |
08ad7dd6-3f94-4a31-b74d-a2ddfdbedc22 | multiple-confidence-gates-for-joint-training | 2204.00226 | null | https://arxiv.org/abs/2204.00226v1 | https://arxiv.org/pdf/2204.00226v1.pdf | Multiple Confidence Gates For Joint Training Of SE And ASR | Joint training of speech enhancement model (SE) and speech recognition model (ASR) is a common solution for robust ASR in noisy environments. SE focuses on improving the auditory quality of speech, but the enhanced feature distribution is changed, which is uncertain and detrimental to the ASR. To tackle this challenge, an approach with multiple confidence gates for jointly training of SE and ASR is proposed. A speech confidence gates prediction module is designed to replace the former SE module in joint training. The noisy speech is filtered by gates to obtain features that are easier to be fitting by the ASR network. The experimental results show that the proposed method has better performance than the traditional robust speech recognition system on test sets of clean speech, synthesized noisy speech, and real noisy speech. | ['Shilei Zhang', 'Junlan Feng', 'Yingying Gao', 'Weibin Zhu', 'Tianrui Wang'] | 2022-04-01 | null | null | null | null | ['robust-speech-recognition'] | ['speech'] | [ 0.07719503 -0.1609266 0.5118399 -0.37027082 -0.73096496 -0.05172543
0.04116208 -0.24807823 -0.3966903 0.41672313 0.3958122 -0.34207693
0.05006345 -0.47934324 -0.39732856 -0.68303066 0.20990512 -0.3593985
0.18527247 -0.4856685 -0.06374397 0.45238796 -1.6885871 0.37094623
0.8722075 0.9979815 0.7632541 0.8763935 -0.07357769 0.57976985
-1.1076076 0.05843196 0.01449233 -0.2706354 -0.09898696 0.2711313
-0.28363866 -0.18097247 -0.7455589 1.4896231 1.0756506 0.56267107
0.20355462 -0.58307683 -0.520305 0.8723489 -0.09466724 0.4191043
0.25596628 -0.01442383 0.43987635 -1.1380645 -0.16732316 1.5382298
0.40405786 0.6592109 -0.69591594 -0.7645073 0.254746 0.27306104
-1.485225 -1.034211 0.75303316 0.1277866 1.1412345 0.5341865
0.4385537 0.81198305 -0.10029879 0.7979279 1.1091548 -0.46964702
0.18643793 0.3664974 0.3320104 0.15569487 -0.24713519 0.6215426
-0.49321708 0.21577375 0.4284683 -0.35343188 -0.62138534 0.82578605
-0.6760378 0.3428307 0.3273336 0.52403134 -0.33994567 -0.19932172
0.24192983 0.61567056 0.5678289 0.13916557 -0.450532 -0.27527717
-0.87906456 -0.18145128 0.4392746 0.7011681 0.34698856 0.9212107
-0.3366452 1.3314877 0.8180532 0.8017967 0.74463046 -0.28235403
0.5138097 0.1680479 -0.04691314 -0.60133284 -0.06038498 -0.9892157
-0.8948636 0.277436 -0.26105967 -0.30990896 -1.3266501 1.3538939
0.07275457 0.45614588 0.40070745 0.79465526 1.0796012 1.3444378
-0.0877233 -0.52463436 0.9972282 -0.8609702 -1.3622565 -0.45823
0.14275943 -1.2120326 0.74252474 0.5530373 -1.1343048 -1.1984203
-1.2524178 0.35327137 -0.2043576 0.49673188 -0.09434506 1.1202561
-0.9868664 0.42950755 -0.60552645 0.11814204 0.03410986 0.40145707
-0.15422338 -0.10572333 -1.4155309 1.0191064 0.43111327 0.7417329
-0.8682385 -0.2581876 -0.9808156 0.35372412 0.1397341 -0.07408089
1.340501 -0.77552027 -1.9903386 -0.04409677 -0.27114108 -0.22543739
0.20033547 -0.09805007 -1.1585809 -0.22798167 -0.5868275 -0.0993737
1.198458 -1.1304688 -0.6709855 -0.2957515 -0.6144603 0.38409817
-0.17980029 0.487849 -0.12295663 -0.9011981 0.46465272 -0.33384514
-0.2423489 -0.5135567 -0.30255073 -0.11670256 0.9473028 -1.1614261
1.4616369 -2.588881 -0.27277476 0.34787986 -0.41364598 0.9471219
-0.24565151 -0.01031671 -0.27795926 -0.01112377 0.03065713 -0.45892614
-0.05935532 0.14104529 -0.3874039 0.11036242 0.23164791 0.27021345
-0.49031875 -0.18008237 0.39366058 0.6876209 -0.17797586 0.65226406
0.21981989 0.26903966 -0.22044241 0.51689947 0.9333203 0.5082669
-0.26314467 -0.32858312 -0.24516802 0.53828967 -1.7642496 0.88484675
-0.50120956 0.43029058 0.46558878 -0.76426256 1.5962143 0.8208642
-0.30509192 -0.5879996 0.2232717 0.35093033 0.3062311 -0.56582147
0.47547716 -0.20947167 0.42996433 -0.18462543 0.1043838 -0.41808707
-0.25928247 -0.21285526 0.6734156 -0.3027682 0.02880427 0.1244045
0.8072313 -0.70883894 0.8596211 0.60140324 -0.27937615 0.54720694
-0.321413 0.31165814 -0.9031135 -1.0884469 -0.20171635 0.7095231
0.10943339 -0.12484333 -0.5694279 -0.22603914 -0.5049959 0.8470666
-0.08405906 -0.38951162 -0.43408582 -0.5591027 0.5718023 0.5119199
0.6241014 -0.95787776 0.50774103 0.5229804 -0.05149523 -1.0193979
-0.49122047 0.3487855 -0.453915 -0.41331112 -0.66868275 -0.97044885
0.4638195 0.4114528 0.24434039 0.24924041 0.17061834 -0.03816254
-0.6925232 -0.4813978 -1.0347359 -0.48552713 0.3243487 0.29194158
0.03906789 -0.32948717 -0.21819468 0.4485333 -0.71222305 -0.33000085
0.57633346 1.0017948 0.46279985 0.7005115 1.0805806 -0.20107874
0.9077739 -0.11252798 -0.6064899 0.29178032 -0.5140947 -0.41162387
0.6659216 -0.5481562 -1.6341928 -0.19540575 -0.93458605 -0.25623924
-0.2612597 0.62874174 -0.65664065 -0.00913382 0.41417325 0.6377977
-0.03333021 -0.77402174 -0.12594748 1.598364 0.44072777 -0.12244878
0.7792769 -0.35057226 -0.60825896 -1.282723 -0.3552395 -0.6317248
-0.03679873 -0.19493186 0.41833717 -1.0817139 -0.25367594 0.8064811
-1.3048322 -0.04700897 -0.1269252 1.1083409 -0.12398262 0.4415225
-0.71141785 -1.4048041 -0.49326614 -1.419457 0.6368766 0.52800786
0.31495434 -0.50580615 -0.3509 0.33327857 0.6541908 -0.8369698
0.5660642 -0.78387266 -0.2579831 -0.23687938 -0.03437733 1.3176566
0.3963995 0.15148766 -1.4478827 -0.2990751 0.57549214 0.11243345
0.70950145 0.36439702 0.9996067 -0.20311369 -0.07928816 0.35671765
1.0962087 0.92860836 1.1552213 -0.15189628 0.10554395 0.3058671
0.87762994 0.33260417 -0.19605273 0.374213 0.04350983 -0.21973918
-0.40191188 -0.11487001 0.8070822 1.5276636 0.23896974 -0.5016611
-0.36602366 0.47621584 -1.3895905 -0.9984499 0.10313617 2.0959637
0.75274974 0.26141676 -0.4293219 0.7391842 1.1510506 0.09584647
-0.14925961 -0.5365878 -0.30890867 0.49714032 0.10565784 0.759731
-0.7855082 0.74673927 6.535197 1.1778078 -1.3678633 0.09465395
0.56506604 0.22026357 -0.1920113 -0.30626458 -0.83202434 0.46114114
0.8269927 -0.09547246 0.4744737 0.70521146 0.7535062 0.02210346
-0.56692946 1.06642 -0.10747325 -0.7839402 -0.10649168 -0.4584228
0.40831763 -0.01422517 0.23664841 0.50053734 0.03414721 -1.0163622
0.61482 0.47740898 0.60007316 -0.7805686 1.0072275 0.44856903
-1.2009232 -0.22937533 -0.51250196 0.1898947 0.21494327 0.6285296
-0.85820353 0.5353318 0.6628723 0.2661756 -0.3647117 1.4214753
-0.48822144 1.0780878 -0.3518255 -0.0982023 -0.07022084 0.08580504
0.9353689 1.3232249 0.7383857 0.29921508 -0.08016071 0.6612478
0.11527768 0.18018946 -0.25143084 -0.08619753 0.731554 0.9927442
-0.10801112 -0.27967575 -0.1908197 0.6344878 -0.23697299 0.529507
-0.6011241 -0.74479365 0.49498007 -0.16844746 0.41817883 -0.2226085
-0.07461347 -0.8256067 -0.11910437 -1.1628957 -0.06416772 -1.1515336
-1.2410154 1.0432428 -0.6132935 -1.1290312 0.0416724 -0.54631734
-0.80019 1.4185756 -1.716256 -0.8630099 0.03679292 0.57095593
0.85905594 -0.44260564 0.6426225 0.65031815 -0.7596878 0.8145129
0.19846849 -0.13957773 0.49025297 -0.83209604 0.2659312 1.2575387
-0.08588876 0.48145443 0.8447663 -0.7054911 -1.0198921 -1.1885068
0.7920289 0.3931259 0.24032725 -0.12670614 -1.222634 0.30172026
0.06159717 0.02846497 0.5389577 -0.16573824 0.02940663 -0.4673647
-1.2056469 0.4071647 0.56701744 -0.6200564 -0.8614309 -0.06657563
1.0704311 -0.43353406 -0.7887138 0.5706926 0.04886518 -0.54199594
0.68095535 -0.14502415 -0.28179237 -0.8629333 -0.6077629 -1.8479095
-0.15599696 -0.79179174 0.14971335 1.5039117 0.64510053 -0.70167214
0.26025903 0.2114707 -0.7026759 -0.32948244 -0.958931 -0.94960177
-0.5398822 -0.68321806 0.76197267 0.5158061 -0.11990567 0.29364878
-0.29337204 0.80056053 0.27640986 -0.60955435 0.31037253 -0.7007688
-0.45595255 -0.07742915 -0.37728694 -1.2687813 -0.14615016 -0.4662141
0.59738433 -1.5988305 -0.49598545 -0.3410896 -0.4682556 0.20150986
-0.39074332 -0.20449282 -0.02827995 -0.4367005 -0.09355462 1.0472987
1.2970036 -0.3688881 -0.33237886 0.7427462 -0.42257804 0.46427035
0.79463196 -0.40456915 -0.45885995 -0.2609441 -0.35095483 0.39274928
-0.03893266 -1.1943399 0.52984273 0.093833 0.20921719 -0.8368239
0.5269025 -0.89889765 -0.08463574 0.28622234 -0.2133318 -0.37481198
0.3793631 0.36144036 -0.68545574 -0.5191158 1.0372667 0.22989559
-0.57266456 0.09710222 -0.6661678 -0.62302256 0.39733565 -0.39340338
-0.1819115 -0.5492338 -1.0770116 0.01949838 -0.47543237 0.5217932
1.2081376 -1.0832884 -0.896028 0.56324637 -0.1350259 -0.07583712
0.5352607 0.35456163 0.03819761 0.1085512 0.27519682 -0.27863842
-1.47527 0.31920242 0.66906637 0.19401178 -0.11386012 0.95803654
-0.11098529 -0.45127156 0.5053914 -0.33451512 -0.48189607 -0.4128984
0.8195912 0.43936375 0.4466473 -0.70991904 -0.03590393 0.09901308
-0.13110854 -0.20921151 1.3749514 -0.5002954 -0.04591498 0.19686383
0.8520139 0.33204395 -0.8282261 -0.32704696 -0.39653897 -0.42736778
0.5313612 -1.0737189 -0.8903145 0.9730602 1.0998105 0.25962698
1.281546 -0.38901028 0.713253 0.33389065 0.05632155 -1.3734542
0.02711767 0.7131193 1.0925626 -1.0088552 -0.45675004 -0.45238855
-0.38047662 1.1460845 0.86521864 0.17074256 0.928606 0.60224754
0.2201001 0.29960483 -0.56566983 -0.35115764 0.26744655 0.7454935
0.24395333 -0.05641161 -0.09780081 1.0753467 -0.2360787 -0.32360068
0.41068923 0.5360635 -0.94628876 -0.9639787 -0.7515766 0.2545094
-0.56224984 -0.3423328 0.12920722 0.10974761 0.08726915 1.700867
-0.01358691 -0.79502594 0.65077645 -0.05747223 0.23156822 -0.7724772
-0.6002066 0.5204439 0.2967636 0.02625405 -0.01901493 -0.23976408
-1.2866848 0.20247364 -1.1059057 0.41418338 1.1007413 1.0166686
0.05897046 0.9845525 1.2541924 -0.27966097 -0.89787024 -1.3093635
-0.93522495 -0.12780611 0.5161848 -0.2242733 -0.62265116 -0.23619294] | [14.843189239501953, 6.023190975189209] |
0e1d1c2b-107e-4a3c-af30-d0e2c04d660a | simple-deep-random-model-ensemble | 1305.1019 | null | http://arxiv.org/abs/1305.1019v2 | http://arxiv.org/pdf/1305.1019v2.pdf | Simple Deep Random Model Ensemble | Representation learning and unsupervised learning are two central topics of
machine learning and signal processing. Deep learning is one of the most
effective unsupervised representation learning approach. The main contributions
of this paper to the topics are as follows. (i) We propose to view the
representative deep learning approaches as special cases of the knowledge reuse
framework of clustering ensemble. (ii) We propose to view sparse coding when
used as a feature encoder as the consensus function of clustering ensemble, and
view dictionary learning as the training process of the base clusterings of
clustering ensemble. (ii) Based on the above two views, we propose a very
simple deep learning algorithm, named deep random model ensemble (DRME). It is
a stack of random model ensembles. Each random model ensemble is a special
k-means ensemble that discards the expectation-maximization optimization of
each base k-means but only preserves the default initialization method of the
base k-means. (iv) We propose to select the most powerful representation among
the layers by applying DRME to clustering where the single-linkage is used as
the clustering algorithm. Moreover, the DRME based clustering can also detect
the number of the natural clusters accurately. Extensive experimental
comparisons with 5 representation learning methods on 19 benchmark data sets
demonstrate the effectiveness of DRME. | ['Xiao-Lei Zhang', 'Ji Wu'] | 2013-05-05 | null | null | null | null | ['clustering-ensemble'] | ['graphs'] | [-2.14173824e-01 -1.31933033e-01 1.80574507e-02 -3.27183276e-01
-5.07841349e-01 -1.88840598e-01 5.06972909e-01 -2.38455832e-01
-1.87428400e-01 1.30716518e-01 3.98940384e-01 1.67137057e-01
-5.09606957e-01 -8.05319965e-01 -6.21569514e-01 -1.29522681e+00
-1.62400961e-01 6.96232975e-01 -3.03959638e-01 -4.55485433e-02
2.70143926e-01 4.33564901e-01 -1.59936833e+00 5.65672457e-01
8.55143607e-01 9.66325343e-01 2.96875060e-01 3.34479809e-01
-2.55256563e-01 9.24591660e-01 -5.85860074e-01 1.09714180e-01
1.38251513e-01 -6.08200133e-01 -6.19246185e-01 2.62215763e-01
-1.30994916e-01 3.34623083e-03 -5.36073327e-01 1.12982416e+00
8.26948106e-01 3.63348871e-01 9.42584693e-01 -1.21948481e+00
-7.15596735e-01 1.19576371e+00 -6.74901128e-01 3.34774964e-02
-1.43044174e-01 -3.16519529e-01 7.48000324e-01 -1.05649543e+00
3.18289995e-01 1.16578948e+00 7.27511048e-01 3.14150721e-01
-9.94883478e-01 -7.37933040e-01 5.68113662e-02 5.02680182e-01
-1.94626224e+00 -6.83014631e-01 9.42898989e-01 -6.70787811e-01
7.30271518e-01 -3.47210765e-02 3.27058852e-01 9.02321994e-01
-1.72030851e-01 8.83264720e-01 7.37499416e-01 -4.98823762e-01
5.98653913e-01 -6.91222772e-02 5.54334819e-01 6.76546752e-01
1.22354165e-01 1.74766369e-02 -2.04236805e-01 -1.82859197e-01
6.24608994e-01 3.48935127e-01 -2.17627883e-01 -4.68441457e-01
-1.23805106e+00 1.06514871e+00 5.98756015e-01 6.17832661e-01
-5.09285033e-01 3.07126284e-01 3.88812631e-01 2.14450181e-01
5.65384626e-02 2.00053841e-01 -6.12602271e-02 1.94992989e-01
-1.16666424e+00 -2.19292894e-01 7.09210932e-01 8.04631352e-01
8.07998955e-01 3.87251198e-01 -2.77935583e-02 1.06146038e+00
3.88519257e-01 4.03333694e-01 8.56060445e-01 -1.00983930e+00
1.83297601e-03 6.42512202e-01 -1.83925465e-01 -1.47846830e+00
-3.71395290e-01 -7.97246456e-01 -1.66768825e+00 -2.58423626e-01
-1.57752022e-01 -2.89070874e-01 -7.56750941e-01 1.51591647e+00
-8.07481781e-02 7.58230031e-01 2.03652471e-01 5.93406558e-01
1.09718335e+00 8.69035721e-01 -2.99432009e-01 -4.79321450e-01
9.32245672e-01 -7.41346359e-01 -7.56905138e-01 2.11257666e-01
6.82194233e-01 -4.24474508e-01 1.35003924e-01 6.47420228e-01
-7.46215940e-01 -8.91342223e-01 -9.57425714e-01 4.16141301e-01
-4.57345366e-01 5.99336803e-01 7.56597281e-01 4.21766192e-01
-1.34374118e+00 4.62331653e-01 -8.49749506e-01 -3.30444098e-01
4.44453686e-01 6.10195637e-01 -3.28429192e-01 -2.33651884e-02
-1.01184177e+00 6.53018475e-01 7.17176795e-01 3.41898263e-01
-8.77840459e-01 -2.77627617e-01 -7.66007185e-01 2.28052512e-01
5.28816804e-02 -6.14703298e-01 4.67326373e-01 -1.22696412e+00
-1.35942447e+00 7.23759055e-01 -2.67175645e-01 -4.86421049e-01
-2.13600338e-01 -5.98505586e-02 -6.47854447e-01 2.13339224e-01
-2.36491882e-03 4.80636626e-01 8.76679361e-01 -1.66358471e+00
-3.65352184e-01 -2.73300260e-01 -5.90879977e-01 1.03894755e-01
-3.97254437e-01 6.24690158e-03 -4.92228121e-01 -7.51695931e-01
5.30028939e-01 -6.73562229e-01 -5.68991601e-01 -6.71420217e-01
-5.67290068e-01 -3.45482200e-01 9.01712358e-01 -4.56370145e-01
1.65365398e+00 -2.38219285e+00 6.52473330e-01 6.28643692e-01
3.87105495e-01 2.63441801e-01 -3.18629877e-03 3.51028264e-01
-5.22947431e-01 -4.63559739e-02 -4.19835597e-01 -3.16622138e-01
-6.23288639e-02 3.62753630e-01 -4.66231883e-01 6.04942560e-01
-3.19508821e-01 7.82351017e-01 -8.35392714e-01 -4.87071067e-01
4.76028979e-01 5.63817143e-01 -3.46485287e-01 1.98220193e-01
3.54784161e-01 4.72357005e-01 -2.86663085e-01 3.22323501e-01
7.36414909e-01 -3.63476455e-01 3.59335870e-01 -5.96064448e-01
-1.89643223e-02 -6.75963163e-02 -1.73808587e+00 1.91407549e+00
-1.31551296e-01 2.82732099e-01 -2.82504540e-02 -1.64493489e+00
1.09777391e+00 1.60804436e-01 8.35480094e-01 -1.61628768e-01
2.18602642e-01 8.84199701e-03 1.82667568e-01 -3.33481938e-01
8.00262839e-02 1.61044553e-01 -1.51259631e-01 6.84663296e-01
5.34118533e-01 3.65289539e-01 1.07131436e-01 4.77338135e-01
8.71504784e-01 -2.99329162e-01 3.49051148e-01 -3.64405304e-01
6.11751080e-01 -3.97577971e-01 7.47390568e-01 8.32443058e-01
2.27808818e-01 6.38919950e-01 1.60369337e-01 -6.05634868e-01
-7.10486591e-01 -1.09690309e+00 -7.34813586e-02 1.02077830e+00
-4.71850075e-02 -5.42285621e-01 -6.23899341e-01 -5.35597205e-01
-2.40163222e-01 7.45151639e-01 -6.75441921e-01 -4.41363990e-01
-5.92381358e-01 -1.17480087e+00 5.52429795e-01 6.37646616e-01
6.03803873e-01 -1.15975797e+00 -7.94818029e-02 2.65716344e-01
-2.79091030e-01 -5.15070438e-01 -6.43778965e-02 5.54507315e-01
-7.53997087e-01 -9.67378497e-01 -4.97800767e-01 -1.11606681e+00
7.28516698e-01 2.68652648e-01 1.01085448e+00 2.68551350e-01
-3.38891111e-02 4.56475705e-01 -6.54992998e-01 -8.94173831e-02
-2.87732780e-01 -4.87880968e-02 4.32268709e-01 5.09942055e-01
6.71747625e-01 -9.42023456e-01 -5.16341209e-01 1.07189417e-01
-6.31208956e-01 -3.80776346e-01 5.59804738e-01 7.70108759e-01
9.01844919e-01 6.20110512e-01 7.39968359e-01 -8.42161238e-01
4.94718224e-01 -9.15474474e-01 -2.29235932e-01 1.15081914e-01
-2.39386678e-01 1.54603161e-02 7.68546820e-01 -2.97302306e-01
-7.78425574e-01 4.09179866e-01 -3.13242301e-02 -7.87930369e-01
-3.54388893e-01 5.71012735e-01 -3.76318544e-01 2.16091022e-01
5.49722135e-01 7.10313439e-01 -1.34712428e-01 -6.37650669e-01
5.83651364e-01 8.65334094e-01 5.80121696e-01 -6.54897153e-01
6.57889664e-01 6.77956641e-01 -3.32293570e-01 -1.03567660e+00
-7.56420314e-01 -6.46449327e-01 -1.12791097e+00 -1.30697161e-01
8.82240057e-01 -9.57649589e-01 -6.05456412e-01 4.76570040e-01
-1.05907226e+00 -2.08865888e-02 -2.37935886e-01 4.75235164e-01
-5.50952554e-01 6.09091818e-01 -4.11380291e-01 -6.72535896e-01
-2.94030547e-01 -1.10593545e+00 9.05372500e-01 5.00716865e-02
-7.53995329e-02 -1.02032340e+00 1.47816226e-01 -1.50424704e-01
1.20051302e-01 3.35312307e-01 7.70640194e-01 -9.54947293e-01
-1.31167069e-01 -3.99372615e-02 1.51823953e-01 6.37812853e-01
2.30568945e-01 5.28468266e-02 -1.03892136e+00 -4.89528924e-01
7.03088790e-02 -1.23951733e-01 1.35828137e+00 7.18244672e-01
1.77883124e+00 -9.36907604e-02 -6.67838573e-01 1.02872503e+00
1.52812910e+00 4.46092725e-01 8.15685809e-01 -2.72674821e-02
9.81754065e-01 2.52064556e-01 -2.47599632e-02 4.89798814e-01
3.38076830e-01 3.58725458e-01 2.17178106e-01 -3.53810824e-02
5.04702292e-02 -3.73278745e-02 4.87443089e-01 1.89358640e+00
-1.41328305e-01 -5.07710241e-02 -8.56534600e-01 5.82853913e-01
-2.07965088e+00 -1.39500988e+00 -9.43326801e-02 2.09664416e+00
3.32418263e-01 -2.64864981e-01 -5.48897050e-02 4.20149535e-01
9.95686829e-01 1.12103000e-01 -3.86137754e-01 -1.43260032e-01
-2.10063621e-01 3.46917421e-01 9.31447446e-02 1.08947530e-01
-1.38367367e+00 8.86600137e-01 6.29581785e+00 9.31086540e-01
-8.60172510e-01 4.25427444e-02 3.91708404e-01 1.35758430e-01
5.56488559e-02 -3.16469446e-02 -6.82385266e-01 7.68939793e-01
8.42157543e-01 6.59491271e-02 5.96212387e-01 7.26884186e-01
8.74297619e-02 2.48930812e-01 -1.25659752e+00 1.36902750e+00
2.38625780e-01 -1.41190302e+00 3.79567832e-01 -1.35523789e-02
8.50849688e-01 5.03002815e-02 6.07784688e-02 4.92858469e-01
7.89911926e-01 -1.31225646e+00 2.43395284e-01 9.69017386e-01
5.61146915e-01 -1.13052642e+00 8.45318079e-01 5.07657051e-01
-1.35247588e+00 -3.90307277e-01 -7.68486977e-01 1.99346021e-01
-1.30267143e-01 8.71172428e-01 -3.44813257e-01 8.78284156e-01
9.23745215e-01 1.31024134e+00 -4.18039769e-01 9.69252765e-01
-1.06477194e-01 8.43266249e-01 -3.46156985e-01 4.89335507e-01
1.56930193e-01 -4.59382534e-01 3.02397668e-01 1.42933857e+00
3.45491439e-01 2.21930593e-01 3.70402038e-01 6.96518004e-01
-1.49065360e-01 -1.56652927e-01 -7.60066748e-01 1.24675408e-01
9.33808625e-01 1.22717309e+00 -8.32076252e-01 -4.75702912e-01
-8.73975083e-02 8.80445182e-01 4.27040488e-01 4.08952147e-01
-5.46061814e-01 -4.30858284e-01 4.18015361e-01 -3.36890459e-01
7.46131659e-01 -1.12763599e-01 -9.09941792e-02 -1.16933572e+00
-4.59619075e-01 -1.00485933e+00 6.78437948e-01 -5.98683953e-01
-1.57433772e+00 5.60141981e-01 -1.39611036e-01 -1.40191066e+00
-2.05310091e-01 -3.42321843e-01 -8.40830743e-01 4.98332769e-01
-1.13819277e+00 -8.85445297e-01 -1.89233303e-01 9.91739929e-01
3.78580958e-01 -8.00988436e-01 1.12864864e+00 3.03404957e-01
-7.38584936e-01 6.29224718e-01 8.24696243e-01 6.58566415e-01
4.41925049e-01 -1.04062569e+00 -2.48456880e-01 7.52505481e-01
4.07558084e-01 9.92209136e-01 1.65760830e-01 -3.20238292e-01
-1.27289724e+00 -1.43919480e+00 4.89313036e-01 -3.52149606e-01
4.31323349e-01 -2.86491603e-01 -8.93326521e-01 7.59336293e-01
2.82310635e-01 -1.37484834e-01 1.15995502e+00 7.83491433e-02
-1.92863196e-01 -2.94776201e-01 -8.58410478e-01 8.98645073e-02
8.42213452e-01 -3.44561696e-01 -9.38104987e-01 2.45228291e-01
4.97035056e-01 1.07224502e-01 -1.02499592e+00 2.31500238e-01
8.18830729e-02 -1.00786960e+00 1.13390493e+00 -4.03720349e-01
2.95933932e-01 -7.62416780e-01 -4.90532398e-01 -1.62235224e+00
-8.44814539e-01 -2.43850231e-01 -5.19421995e-01 1.32731295e+00
-1.61121383e-01 -3.90253037e-01 3.75604004e-01 -4.11481738e-01
-3.25209826e-01 -6.02833331e-01 -8.17438543e-01 -4.42600459e-01
6.42368346e-02 -3.62132996e-01 6.48888409e-01 1.57596695e+00
-2.37254545e-01 3.33945423e-01 -3.19027901e-01 3.64684492e-01
8.99210274e-01 3.10488522e-01 6.65013909e-01 -1.49054623e+00
-3.51543844e-01 -4.38947141e-01 -4.85408366e-01 -1.05062497e+00
3.90712529e-01 -1.38215768e+00 -1.28890380e-01 -1.68532312e+00
2.57991940e-01 -4.29831982e-01 -8.88599515e-01 3.09926122e-01
2.33200833e-01 -1.78955942e-01 2.86635384e-02 5.25487781e-01
-9.94734704e-01 6.48366749e-01 7.55439281e-01 -2.84891993e-01
-2.32650399e-01 7.40248011e-03 -1.10232770e+00 9.58031178e-01
6.73814476e-01 -5.04574299e-01 -4.82435286e-01 -5.24296761e-01
2.18849387e-02 -2.57201910e-01 1.09493963e-01 -1.17499769e+00
7.35554516e-01 2.37941936e-01 7.31005669e-01 -9.52867448e-01
9.02035460e-02 -9.14030552e-01 1.89321727e-01 3.49663436e-01
-3.06029946e-01 -1.91537112e-01 -2.25703523e-01 7.51370311e-01
-3.92347932e-01 -1.33763820e-01 8.54936779e-01 -4.30606157e-01
-1.00214243e+00 2.94364482e-01 -3.86498839e-01 -1.10063232e-01
9.18905079e-01 -3.21096212e-01 2.06683874e-01 -2.44943917e-01
-1.07672691e+00 3.89369041e-01 7.12720752e-02 2.63609499e-01
8.54644716e-01 -1.55617082e+00 -8.01550269e-01 2.69043148e-01
-1.12489775e-01 2.18048632e-01 2.94391543e-01 6.42422080e-01
-2.52846241e-01 1.71522409e-01 -7.83927143e-02 -7.86334813e-01
-6.23435676e-01 8.58365655e-01 5.59696794e-01 -6.75326362e-02
-6.01527333e-01 9.34894085e-01 2.02789754e-01 -5.92919827e-01
5.02920568e-01 2.11303264e-01 -8.01962018e-01 1.90443844e-01
6.83478892e-01 5.64737618e-01 2.81214137e-02 -9.21473503e-01
-4.60457116e-01 6.11081481e-01 -1.03325821e-01 4.53933030e-01
1.77405596e+00 -1.10655375e-01 -6.96103454e-01 6.06552601e-01
1.28825307e+00 -3.57797056e-01 -6.69500828e-01 -2.78061122e-01
8.54385644e-02 8.78666267e-02 2.18131110e-01 -5.46270370e-01
-1.51092112e+00 9.94316697e-01 8.06491613e-01 -8.45640004e-02
1.33030486e+00 -6.64980412e-02 4.71619308e-01 5.83844662e-01
3.75968486e-01 -1.32068741e+00 -1.05527028e-01 5.67090988e-01
6.73519969e-01 -9.77601230e-01 2.01953426e-01 9.26215574e-03
-5.27696729e-01 1.09566963e+00 5.39215922e-01 -5.80393016e-01
1.22672975e+00 2.06640139e-01 -2.48267978e-01 -4.42677706e-01
-6.04679942e-01 -2.64532775e-01 1.81110457e-01 7.73646355e-01
3.40432584e-01 1.19611852e-01 6.47378489e-02 1.07907271e+00
-1.25398651e-01 -1.78773850e-01 1.54083192e-01 4.98291045e-01
-5.37518680e-01 -8.36326063e-01 -4.32266831e-01 6.93232894e-01
-4.97383438e-02 -1.01050630e-01 -4.07694578e-01 4.93118227e-01
5.72476089e-01 1.09909546e+00 3.04802507e-01 -8.19066167e-01
-6.92908019e-02 7.87073672e-02 1.26518428e-01 -7.01692104e-01
-6.15850329e-01 1.15378149e-01 -5.53259671e-01 -5.24246752e-01
-6.62645221e-01 -3.65837961e-01 -1.36019051e+00 -2.55235314e-01
-2.58394510e-01 4.63911146e-01 2.73944438e-01 9.73775804e-01
5.31082809e-01 5.64161181e-01 1.00687885e+00 -8.28158259e-01
-3.49086881e-01 -9.65652287e-01 -9.07564878e-01 4.61451054e-01
9.01506245e-02 -8.94004285e-01 -4.74842012e-01 3.60174567e-01] | [9.138726234436035, 3.1967079639434814] |
0d90c416-152d-4974-9780-2a5f484544af | fast-searching-for-a-faster-arbitrarily | 2111.02394 | null | https://arxiv.org/abs/2111.02394v2 | https://arxiv.org/pdf/2111.02394v2.pdf | FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation | We propose an accurate and efficient scene text detection framework, termed FAST (i.e., faster arbitrarily-shaped text detector). Different from recent advanced text detectors that used complicated post-processing and hand-crafted network architectures, resulting in low inference speed, FAST has two new designs. (1) We design a minimalist kernel representation (only has 1-channel output) to model text with arbitrary shape, as well as a GPU-parallel post-processing to efficiently assemble text lines with a negligible time overhead. (2) We search the network architecture tailored for text detection, leading to more powerful features than most networks that are searched for image classification. Benefiting from these two designs, FAST achieves an excellent trade-off between accuracy and efficiency on several challenging datasets, including Total Text, CTW1500, ICDAR 2015, and MSRA-TD500. For example, FAST-T yields 81.6% F-measure at 152 FPS on Total-Text, outperforming the previous fastest method by 1.7 points and 70 FPS in terms of accuracy and speed. With TensorRT optimization, the inference speed can be further accelerated to over 600 FPS. Code and models will be released at https://github.com/czczup/FAST. | ['Tong Lu', 'Enze Xie', 'Guo Chen', 'Wenhai Wang', 'Jiahao Wang', 'Ping Luo', 'Zhe Chen'] | 2021-11-03 | null | null | null | null | ['scene-text-detection'] | ['computer-vision'] | [-1.61306083e-01 -4.93712902e-01 2.47276314e-02 -2.68246353e-01
-5.01709819e-01 -4.99646544e-01 4.03462470e-01 -7.04559013e-02
-5.16680658e-01 -2.83676349e-02 -1.15773998e-01 -5.13934791e-01
2.95919508e-01 -7.64899433e-01 -5.54811776e-01 -4.52679962e-01
2.21907392e-01 3.13691050e-01 6.63821995e-01 1.18425995e-01
3.12539250e-01 4.55564708e-01 -1.31811881e+00 2.20516756e-01
8.00146401e-01 1.14095557e+00 4.35569078e-01 1.06213474e+00
-2.64780015e-01 7.25865841e-01 -4.35701907e-01 -6.14824355e-01
3.27168018e-01 1.24033615e-01 -3.96410286e-01 2.19435077e-02
6.58431530e-01 -7.38277733e-01 -7.11099505e-01 8.26257229e-01
6.29729986e-01 -1.44598231e-01 4.83728766e-01 -9.30981934e-01
-3.57866704e-01 6.33327365e-01 -8.91971946e-01 6.04184344e-03
6.66019991e-02 2.49211520e-01 1.14598358e+00 -1.17189705e+00
2.44877353e-01 1.19243622e+00 9.53912497e-01 1.73490360e-01
-8.96724224e-01 -7.26909995e-01 3.88073362e-02 3.98422368e-02
-1.59134269e+00 -4.30080086e-01 2.98787951e-01 -3.83974880e-01
1.14448583e+00 2.13061973e-01 6.33012354e-01 9.66999590e-01
1.64045319e-01 1.17817855e+00 5.28768420e-01 -3.74862522e-01
3.56056318e-02 -3.79532762e-02 1.94547758e-01 1.04812169e+00
4.62850273e-01 -2.50216663e-01 -6.86172068e-01 -6.12243786e-02
9.94813621e-01 -2.10685958e-03 -1.99120462e-01 -8.92317574e-03
-1.36879301e+00 8.71710420e-01 2.31392771e-01 -4.09703180e-02
-4.18429114e-02 3.84880543e-01 8.44900310e-01 -8.96964297e-02
3.91124457e-01 1.61730140e-01 -4.50370789e-01 -2.90804803e-01
-1.02920461e+00 5.17093651e-02 8.45383644e-01 1.14158273e+00
5.40275872e-01 3.78179073e-01 -3.04294378e-01 9.65107262e-01
1.39523581e-01 8.65271270e-01 2.86605358e-01 -4.71703380e-01
7.57104099e-01 5.96779525e-01 -2.67440110e-01 -9.68378425e-01
-5.95973551e-01 -5.30265689e-01 -1.03781486e+00 -7.51387924e-02
5.37094116e-01 -2.12817088e-01 -9.19641793e-01 9.29274440e-01
3.05153638e-01 -8.86777192e-02 -2.80117899e-01 8.62659812e-01
8.24609041e-01 7.19020009e-01 -2.57126659e-01 3.88419926e-01
1.65257752e+00 -1.27875102e+00 -4.81200159e-01 -2.99394101e-01
1.03702164e+00 -1.00003862e+00 1.20351410e+00 7.21520007e-01
-8.66276264e-01 -3.76745969e-01 -1.10357285e+00 -3.53361726e-01
-2.64848173e-01 8.34655762e-01 8.68620336e-01 7.20598042e-01
-1.11969280e+00 3.69423807e-01 -9.38779116e-01 -5.17222643e-01
5.41996598e-01 3.84184897e-01 7.89067522e-02 9.68233719e-02
-5.56988180e-01 2.80234456e-01 4.71930832e-01 1.11518353e-01
-5.20485997e-01 -5.42503417e-01 -6.21354342e-01 1.98828772e-01
7.33511627e-01 -7.16915190e-01 1.32362306e+00 -6.94661558e-01
-1.73429358e+00 6.48521781e-01 -9.60065126e-02 -6.07727587e-01
7.48299956e-01 -4.68810350e-01 -1.75470829e-01 3.82046252e-01
-3.64249311e-02 7.11111426e-01 1.06030512e+00 -4.69152957e-01
-7.58822381e-01 -1.50327727e-01 -3.32973689e-01 1.85030624e-01
-9.06072140e-01 1.39012128e-01 -1.20979142e+00 -7.33263135e-01
4.20774966e-02 -8.48175168e-01 -2.11829823e-02 5.32371938e-01
-7.95911133e-01 -2.07001954e-01 1.05572212e+00 -5.47496855e-01
1.23540020e+00 -2.03006887e+00 -4.28613335e-01 1.89270582e-02
5.69553375e-01 4.71538752e-01 -4.29133512e-02 2.85577148e-01
2.26228297e-01 -6.04092702e-02 1.70246363e-01 -4.40058231e-01
-4.33532707e-02 -1.64983213e-01 -2.90593743e-01 5.47798395e-01
3.43448785e-03 1.06672788e+00 -3.95930797e-01 -6.11272514e-01
5.77136338e-01 4.86341953e-01 -5.07821679e-01 -1.62361354e-01
-1.12659670e-01 -2.87847489e-01 -6.34258628e-01 8.72463644e-01
6.83707654e-01 -5.77267110e-01 -1.13372833e-01 -3.80166441e-01
-2.83355057e-01 1.28163457e-01 -1.21152830e+00 1.54380584e+00
-2.02096477e-01 1.02380788e+00 1.86474264e-01 -7.49351203e-01
8.96649837e-01 -2.17296537e-02 2.55725712e-01 -6.91145420e-01
4.47465003e-01 -2.12759450e-02 -4.53836054e-01 -3.04075956e-01
9.12089467e-01 6.15666986e-01 2.07003057e-01 3.59262884e-01
-2.11435482e-01 2.65575461e-02 2.07027495e-01 3.76311660e-01
1.16035616e+00 -1.01567502e-03 -4.06357720e-02 -2.98152417e-01
2.02198148e-01 9.31254923e-02 2.37876520e-01 9.74960983e-01
-4.65209298e-02 5.76365709e-01 4.68959242e-01 -6.06971085e-01
-1.10367966e+00 -7.94352889e-01 -3.47442001e-01 1.25991964e+00
-7.94899762e-02 -9.30361629e-01 -7.09593892e-01 -3.74257147e-01
3.39956731e-02 3.81248057e-01 -2.46512726e-01 2.93504387e-01
-5.40740609e-01 -8.76227081e-01 9.50500011e-01 7.96952128e-01
8.85375381e-01 -5.86405516e-01 -5.38154542e-01 -8.63910001e-03
1.14648219e-03 -1.53379643e+00 -6.98977172e-01 2.58022159e-01
-9.59273219e-01 -8.89714718e-01 -5.37250817e-01 -6.48864865e-01
4.51521665e-01 6.43306136e-01 8.85792613e-01 1.73100129e-01
-6.71356618e-01 2.33461633e-01 -4.13034230e-01 -3.31030667e-01
-2.60841008e-02 2.25588784e-01 -7.79402629e-02 -1.92658067e-01
1.92413539e-01 2.40016598e-02 -6.43261492e-01 3.41207027e-01
-7.78729320e-01 5.16957223e-01 7.51321137e-01 7.68503308e-01
4.56181258e-01 1.97908998e-01 -3.64489965e-02 -5.75527728e-01
3.77399117e-01 -2.66038533e-02 -1.01543152e+00 1.65932775e-01
-4.93478268e-01 -1.52539521e-01 9.13851142e-01 -4.86143708e-01
-8.69239986e-01 3.18763107e-01 -2.25695893e-01 -5.11525750e-01
6.44019842e-02 1.79830581e-01 3.93082172e-01 -1.67392656e-01
7.53518879e-01 4.48088497e-01 -1.53874949e-01 -3.43866646e-01
1.82338655e-01 7.59289920e-01 4.57911819e-01 -5.99135280e-01
8.28869641e-01 6.66031420e-01 -1.22104056e-01 -1.28261828e+00
-7.35423625e-01 -6.61027074e-01 -5.05367815e-01 -1.96595192e-02
8.37210059e-01 -1.07357883e+00 -9.89861071e-01 9.93770778e-01
-9.90146220e-01 -6.31999314e-01 2.25787401e-01 4.84362334e-01
-2.70103782e-01 6.33787334e-01 -1.05625546e+00 -7.34539270e-01
-9.60812747e-01 -9.70643759e-01 1.43699658e+00 1.14573501e-01
2.01013848e-01 -7.46137798e-01 -3.90282869e-01 1.39309824e-01
4.08510387e-01 -2.84795195e-01 4.07540411e-01 -3.90531480e-01
-6.47913694e-01 -4.20616031e-01 -8.44817638e-01 2.20422968e-01
-2.60277808e-01 4.84147012e-01 -9.33474600e-01 -3.84885162e-01
-3.59842032e-01 -3.35587651e-01 9.28063452e-01 5.57626367e-01
1.48211086e+00 -1.05837114e-01 -4.38097775e-01 1.04610682e+00
1.40584815e+00 -5.88842407e-02 4.72636908e-01 3.50235045e-01
1.03120995e+00 1.42684937e-01 5.15215218e-01 7.02584565e-01
2.98170358e-01 6.46002531e-01 2.57770568e-01 -2.54301906e-01
-6.02474771e-02 -2.21836910e-01 3.46699357e-01 8.87812018e-01
2.27587700e-01 -4.66578841e-01 -1.11028981e+00 1.29206508e-01
-1.91811538e+00 -6.63596869e-01 -6.30119026e-01 2.08632612e+00
3.74829143e-01 3.47064108e-01 2.35512838e-01 -1.02314457e-01
7.39114523e-01 3.27988341e-02 -6.65010035e-01 -1.62286237e-01
-1.81140289e-01 1.02004640e-01 9.20603037e-01 1.30725712e-01
-1.18745220e+00 1.32503879e+00 6.16697407e+00 1.29549384e+00
-1.22763324e+00 -1.04174100e-01 6.06450617e-01 -2.93768227e-01
4.39012021e-01 -1.51195750e-01 -1.27292371e+00 2.90168077e-01
7.53743470e-01 6.91660568e-02 3.25953513e-01 1.13237083e+00
2.10147157e-01 -2.32673943e-01 -8.04720163e-01 1.36943579e+00
3.87803838e-02 -1.53105259e+00 -3.49940471e-02 1.39134731e-02
3.24275315e-01 4.88995612e-01 -1.03732608e-01 3.42783511e-01
2.09327489e-01 -7.46030927e-01 8.77904177e-01 1.15137339e-01
1.08088720e+00 -6.25332475e-01 4.25061435e-01 3.76034111e-01
-1.60123587e+00 1.67064124e-03 -7.06004262e-01 3.32050398e-02
-1.27536207e-01 8.77855718e-01 -9.11243618e-01 3.57335567e-01
1.02667582e+00 8.30489039e-01 -7.07739592e-01 1.01380432e+00
-1.00134112e-01 8.56026769e-01 -8.36730778e-01 -5.30358255e-01
3.14402372e-01 -1.14215523e-01 3.34680051e-01 1.65629029e+00
2.20856234e-01 -3.14425796e-01 3.04539889e-01 6.37162805e-01
-1.73009843e-01 2.33121768e-01 -3.58926445e-01 8.86176061e-03
5.43888330e-01 1.70165300e+00 -1.10365558e+00 -5.61767340e-01
-5.08414090e-01 1.25043237e+00 2.75505126e-01 2.22317010e-01
-1.06120241e+00 -7.42800534e-01 2.80886352e-01 -2.59991474e-02
5.17846048e-01 -4.79855180e-01 -3.88268977e-01 -1.28022683e+00
1.20367818e-01 -5.95608175e-01 2.12774724e-01 -8.40936303e-01
-1.03357255e+00 4.42330748e-01 -3.81633401e-01 -1.03498781e+00
2.52724230e-01 -1.13815498e+00 -6.00733638e-01 6.25248849e-01
-1.29243016e+00 -1.24758852e+00 -4.98117357e-01 6.24238491e-01
8.01385045e-01 1.00069474e-02 4.34627324e-01 2.68363655e-01
-1.09279859e+00 9.12493885e-01 2.61837631e-01 5.47415674e-01
7.15252101e-01 -1.00899494e+00 9.92238402e-01 8.99799466e-01
1.25624120e-01 2.60835052e-01 3.15668762e-01 -6.02367997e-01
-1.89710367e+00 -1.23842800e+00 3.07994872e-01 -3.61152530e-01
9.70729113e-01 -7.62497723e-01 -8.57912600e-01 6.47285104e-01
-2.83364858e-02 -9.10752863e-02 2.23897487e-01 1.21743500e-01
-5.71142972e-01 -7.43266270e-02 -7.31078982e-01 9.04020250e-01
9.94627416e-01 -3.25744838e-01 4.02835235e-02 6.20709717e-01
6.36224866e-01 -7.35141993e-01 -6.18789852e-01 5.68852201e-02
6.28722847e-01 -9.90263522e-01 8.22095871e-01 1.18684515e-01
1.88131601e-01 -1.83518767e-01 2.41909139e-02 -5.47011197e-01
-4.00372684e-01 -6.90334618e-01 -2.60040909e-01 9.03875351e-01
1.90978736e-01 -8.16451371e-01 1.07746744e+00 2.74675712e-02
-2.72412062e-01 -9.72944081e-01 -6.24811828e-01 -7.89833486e-01
-1.11913130e-01 -8.79978836e-01 2.48430207e-01 7.11960435e-01
-3.08594197e-01 4.39236939e-01 -4.06360626e-01 1.51164636e-01
5.72260141e-01 -7.13733062e-02 1.04775751e+00 -1.02572310e+00
-3.74845594e-01 -6.67072296e-01 -2.88582653e-01 -1.66982484e+00
-2.31392175e-01 -7.69723535e-01 -7.14033097e-02 -1.26137424e+00
3.20045590e-01 -4.78150070e-01 3.87321413e-01 7.40669608e-01
-1.46571755e-01 2.50950247e-01 1.10696681e-01 2.82561153e-01
-7.87729204e-01 4.87970352e-01 1.09937918e+00 9.95392054e-02
-1.34257436e-01 -1.54404417e-01 -4.11429197e-01 9.01414871e-01
9.56250012e-01 -2.80929685e-01 -7.35232681e-02 -7.07060635e-01
4.21579778e-01 -1.01910278e-01 4.17761892e-01 -1.10606635e+00
5.41199207e-01 1.77445561e-01 5.81521451e-01 -1.12863743e+00
3.75957400e-01 -4.23591435e-01 -4.40132082e-01 4.91733819e-01
8.31988826e-02 9.52650532e-02 4.15259302e-01 3.82905275e-01
2.63798535e-01 -2.32302472e-01 7.05604613e-01 2.57554799e-01
-6.90292358e-01 4.52709407e-01 -4.86170173e-01 -8.73006135e-02
7.49235332e-01 -5.18216014e-01 -7.39517450e-01 -1.90410644e-01
4.15822379e-02 3.58147144e-01 4.57699478e-01 3.60164613e-01
6.20000958e-01 -8.68874907e-01 -6.79826379e-01 1.34794399e-01
6.83124810e-02 1.47143647e-01 2.65070230e-01 1.01512349e+00
-9.92836535e-01 6.20012820e-01 2.94009477e-01 -1.09012496e+00
-1.18951631e+00 2.67949760e-01 2.62229443e-01 -2.66640574e-01
-1.22180355e+00 7.95757174e-01 2.49452606e-01 -2.93535709e-01
3.73034775e-01 -4.33039665e-01 3.67294997e-01 -2.74068058e-01
6.43358052e-01 4.96754080e-01 2.47327581e-01 -1.24257706e-01
-1.37570009e-01 7.74765134e-01 -3.47054034e-01 2.72682816e-01
9.48927999e-01 1.15157083e-01 6.02042601e-02 1.69687495e-01
9.89758193e-01 5.01930565e-02 -1.31304717e+00 -2.57217526e-01
-2.84402400e-01 -4.23216134e-01 4.16939735e-01 -6.33071423e-01
-1.13198102e+00 7.55352676e-01 5.28438151e-01 1.11877576e-01
1.06511939e+00 -8.62488598e-02 9.09056783e-01 9.42084610e-01
1.84610054e-01 -1.31149018e+00 3.67033690e-01 8.77810538e-01
5.24048388e-01 -1.07820308e+00 2.31942624e-01 -5.81617773e-01
-4.47363496e-01 1.49963284e+00 7.12693930e-01 -8.54729675e-03
2.67230213e-01 7.79236794e-01 -2.01106682e-01 -3.13032925e-01
-7.54096270e-01 9.73210111e-03 1.76045597e-01 1.46271273e-01
3.71559083e-01 5.50446585e-02 2.41554558e-01 1.45946413e-01
-3.01616579e-01 -2.18383446e-01 4.59323645e-01 8.29187274e-01
-6.06194973e-01 -6.00921929e-01 -5.87166488e-01 1.00251496e+00
-4.04446214e-01 -3.89986098e-01 -1.51904255e-01 8.71144772e-01
-4.68705744e-01 7.25906909e-01 2.24873409e-01 -4.64711726e-01
1.73697755e-01 -2.94639945e-01 1.81913376e-01 -4.66475099e-01
-5.19078255e-01 2.92458922e-01 7.92464521e-03 -6.92317963e-01
1.54668570e-01 -3.09616446e-01 -1.26197529e+00 -8.44682038e-01
-9.10538793e-01 -3.90993834e-01 9.17907000e-01 6.63261712e-01
4.51465040e-01 4.65875477e-01 3.86882067e-01 -9.31394935e-01
-4.60491806e-01 -9.00918901e-01 -5.42802691e-01 -3.07483763e-01
-3.24206091e-02 -3.49550962e-01 -3.33957464e-01 -8.59256685e-02] | [12.021595001220703, 2.23667573928833] |
1219f8f3-5010-4088-98cc-2661f8a9f688 | low-resource-named-entity-recognition-with | null | null | https://aclanthology.org/I17-2016 | https://aclanthology.org/I17-2016.pdf | Low-Resource Named Entity Recognition with Cross-lingual, Character-Level Neural Conditional Random Fields | Low-resource named entity recognition is still an open problem in NLP. Most state-of-the-art systems require tens of thousands of annotated sentences in order to obtain high performance. However, for most of the world{'}s languages it is unfeasible to obtain such annotation. In this paper, we present a transfer learning scheme, whereby we train character-level neural CRFs to predict named entities for both high-resource languages and low-resource languages jointly. Learning character representations for multiple related languages allows knowledge transfer from the high-resource languages to the low-resource ones, improving F1 by up to 9.8 points. | ['Kevin Duh', 'Ryan Cotterell'] | 2017-11-01 | low-resource-named-entity-recognition-with-1 | https://aclanthology.org/I17-2016 | https://aclanthology.org/I17-2016.pdf | ijcnlp-2017-11 | ['low-resource-named-entity-recognition'] | ['natural-language-processing'] | [-7.05389306e-02 1.02580048e-01 -2.62538522e-01 -5.71059346e-01
-1.05975413e+00 -7.01937258e-01 1.97308853e-01 3.58174175e-01
-1.05297899e+00 1.37801349e+00 1.02728292e-01 -2.89216995e-01
5.41591108e-01 -8.79429400e-01 -6.22013569e-01 -1.50970742e-01
-8.66386145e-02 8.71831238e-01 4.14398819e-01 1.01736635e-02
9.11719948e-02 7.23822236e-01 -7.98542798e-01 3.06520313e-01
1.22158813e+00 4.15985912e-01 3.70984644e-01 3.92250896e-01
-9.03226018e-01 7.32541978e-01 -7.24646866e-01 -6.31238103e-01
-7.02214614e-02 -2.33874932e-01 -1.18877709e+00 -3.57526690e-01
1.79400042e-01 1.67976499e-01 6.67170435e-02 1.12902677e+00
3.02766174e-01 9.60012749e-02 5.55359602e-01 -6.15618825e-01
-9.19922888e-01 7.66188860e-01 -3.39600742e-01 2.11681560e-01
2.54339933e-01 -2.89490104e-01 1.13233185e+00 -8.74232709e-01
1.02411938e+00 8.97499800e-01 4.24785823e-01 6.11917734e-01
-8.97535801e-01 -7.22956359e-01 9.86005440e-02 -7.86130056e-02
-1.65009284e+00 -4.97296453e-01 3.89930934e-01 -2.79042512e-01
1.49906158e+00 -2.03822434e-01 2.03121409e-01 8.06678474e-01
-1.38060987e-01 6.67790711e-01 1.08502936e+00 -7.26104975e-01
-5.31269377e-03 3.10188204e-01 9.21309888e-02 6.46820068e-01
4.59130198e-01 -4.57410872e-01 -5.48505425e-01 -1.59598187e-01
9.12896633e-01 -4.33874249e-01 7.83857480e-02 3.03288221e-01
-1.23141325e+00 8.06119621e-01 3.44577193e-01 9.12292957e-01
-4.83023614e-01 -1.61759809e-01 3.38473499e-01 6.46176934e-03
6.85995460e-01 7.70255327e-01 -9.03626502e-01 -2.13366270e-01
-7.24541247e-01 -3.28271091e-01 1.10241568e+00 1.41050696e+00
9.64511633e-01 1.27959400e-01 1.20670982e-01 9.23840821e-01
1.91944748e-01 4.26936358e-01 5.20641983e-01 -2.36714229e-01
8.51182818e-01 6.65492713e-01 2.57025301e-01 -4.50734109e-01
-4.19871211e-01 -1.79987282e-01 -8.76515448e-01 -4.53053653e-01
5.55292130e-01 -5.06385863e-01 -9.55669761e-01 1.58878934e+00
1.90259412e-01 -5.75995073e-04 4.66849864e-01 5.50405145e-01
7.42085934e-01 9.54528928e-01 6.92332327e-01 -1.28979996e-01
1.50243747e+00 -9.68788922e-01 -5.67294836e-01 -6.17125869e-01
8.21427643e-01 -8.51379752e-01 8.50002408e-01 -1.94334954e-01
-9.62883055e-01 -4.52970624e-01 -4.84830230e-01 -2.17833385e-01
-6.46691561e-01 4.61150497e-01 7.16164052e-01 6.64088845e-01
-9.20169473e-01 3.83168548e-01 -8.19602847e-01 -4.95356232e-01
2.97880888e-01 3.96791220e-01 -7.93127775e-01 -1.21800490e-01
-1.51476979e+00 1.18223286e+00 9.90139127e-01 6.17001317e-02
-4.65269506e-01 -5.08518994e-01 -8.96645546e-01 1.17634378e-01
1.08239643e-01 -4.04349029e-01 1.03791702e+00 -8.29140425e-01
-1.42540097e+00 1.18132949e+00 -2.63473451e-01 -2.58100629e-01
1.94930881e-01 -4.80366081e-01 -7.56626546e-01 3.49099077e-02
1.38666227e-01 6.20896459e-01 1.20141588e-01 -7.63387322e-01
-7.22025931e-01 -1.32838249e-01 -1.48932725e-01 1.32662535e-01
-4.43142682e-01 7.76319683e-01 -6.04170978e-01 -4.61166084e-01
-3.34549159e-01 -8.56445312e-01 -5.55936754e-01 -3.74619067e-01
-3.64120483e-01 -4.97257560e-01 1.70113191e-01 -9.35176432e-01
9.88960087e-01 -1.86792171e+00 -1.48944810e-01 -2.08589323e-02
-1.63882867e-01 5.19892454e-01 -2.29227319e-01 4.78615701e-01
1.18897021e-01 4.45617408e-01 -2.36979604e-01 -6.00764565e-02
-1.48793533e-01 4.07503784e-01 1.96756478e-02 3.53744149e-01
8.17387164e-01 7.66765296e-01 -9.12568927e-01 -6.82971776e-01
-1.37707800e-01 5.56107879e-01 -2.74402231e-01 5.19664705e-01
-1.50933191e-01 7.81342804e-01 -7.53309071e-01 5.56473911e-01
3.59235466e-01 -3.08001906e-01 5.41315854e-01 2.38887087e-01
-4.03166920e-01 6.63851440e-01 -8.77191961e-01 1.78319550e+00
-7.62067199e-01 3.80867392e-01 -2.73989320e-01 -6.33830130e-01
1.31670320e+00 5.11634409e-01 1.77948520e-01 -4.17845309e-01
-1.47562876e-01 5.91442943e-01 -1.70424297e-01 -3.57755482e-01
8.53205979e-01 -5.34619987e-01 -5.82609594e-01 1.29147112e-01
3.26170027e-01 2.44922116e-01 2.45613560e-01 -1.69970751e-01
9.85149503e-01 2.23629594e-01 8.07125270e-01 -5.06512821e-02
5.34237146e-01 2.14296266e-01 9.06406403e-01 4.32775378e-01
-1.84070438e-01 1.93195418e-01 2.15277821e-01 -4.50643271e-01
-1.14019620e+00 -8.27369750e-01 -1.27656832e-01 1.37833488e+00
-2.53397852e-01 -1.28000692e-01 -6.05648279e-01 -8.02356005e-01
-3.87244433e-01 8.71773422e-01 -2.02628195e-01 3.48838627e-01
-9.11772788e-01 -5.28523028e-01 1.02446032e+00 7.64786243e-01
5.11947632e-01 -1.60161459e+00 -1.77579954e-01 3.70054156e-01
-2.15965450e-01 -1.49852502e+00 -2.02987239e-01 3.58603865e-01
-6.72244310e-01 -4.70112383e-01 -1.04640329e+00 -1.13698542e+00
8.07474494e-01 -4.69017029e-01 1.37474358e+00 -1.53856397e-01
4.11545625e-03 -2.48942271e-01 -5.08536518e-01 -3.11002493e-01
-5.98506689e-01 5.97836971e-01 -6.77650720e-02 -3.09386224e-01
6.31171823e-01 -2.09228680e-01 1.56914994e-01 2.53330525e-02
-7.09964752e-01 -1.03658482e-01 1.00062025e+00 5.03430843e-01
7.70842731e-01 -1.77135199e-01 7.12160289e-01 -1.38001883e+00
1.02564022e-01 -5.25587261e-01 -6.99537575e-01 6.77125156e-01
2.15038960e-03 2.60372132e-01 8.69732380e-01 -3.92168015e-01
-1.45513022e+00 3.92303526e-01 -4.94042873e-01 1.17919520e-01
-7.15086639e-01 5.70753872e-01 -4.55722600e-01 1.05534650e-01
4.55136299e-01 1.44849852e-01 -1.11694705e+00 -7.27263570e-01
5.54069102e-01 8.50987554e-01 3.41161877e-01 -7.03658700e-01
5.69734573e-01 -7.81484991e-02 -5.65675795e-01 -1.17806304e+00
-1.12431979e+00 -6.99100912e-01 -1.19036865e+00 2.40302339e-01
1.28535473e+00 -1.05837560e+00 -2.55250603e-01 2.14489847e-01
-1.67882967e+00 -2.31268793e-01 -2.71807492e-01 6.60063028e-01
-1.85363188e-01 1.61690265e-01 -9.05911088e-01 -6.86975598e-01
-4.00164574e-01 -7.32151031e-01 8.48615289e-01 4.16260332e-01
-1.78010777e-01 -1.07701528e+00 3.85743856e-01 2.43287399e-01
2.49833822e-01 -1.27985761e-01 8.59427333e-01 -1.44361567e+00
-3.89741302e-01 -2.82321721e-01 -3.84788513e-01 1.42969698e-01
5.51018752e-02 -2.97086090e-01 -8.24773252e-01 3.56253423e-02
-4.72258538e-01 -4.81036216e-01 4.70338017e-01 -2.05289096e-01
7.25702167e-01 -2.13455439e-01 -3.42874140e-01 1.45868689e-01
1.42990601e+00 1.13784187e-01 7.71068811e-01 2.71595836e-01
7.15637326e-01 6.07752740e-01 5.52276433e-01 2.49031529e-01
4.49529737e-01 2.95744717e-01 -3.34843814e-01 -1.54144779e-01
6.49161190e-02 -3.84155780e-01 1.97052419e-01 1.34863281e+00
-1.29747167e-01 -4.23222721e-01 -1.49836123e+00 8.15955341e-01
-1.39471078e+00 -6.31280899e-01 -4.78669070e-02 2.06063700e+00
1.29429829e+00 2.92591588e-03 -3.28519315e-01 -6.92327857e-01
1.06509995e+00 -1.90817320e-03 -3.38600010e-01 -3.83519500e-01
-2.23811477e-01 3.62771422e-01 4.91865873e-01 2.02300370e-01
-1.24878597e+00 1.65769053e+00 6.47029495e+00 7.51089990e-01
-9.06513989e-01 1.95497751e-01 5.31642914e-01 5.57602108e-01
-2.20961720e-01 1.01735063e-01 -1.35464692e+00 1.21044591e-01
1.55066693e+00 -2.08123207e-01 -1.37419701e-01 1.07697773e+00
-2.63266206e-01 9.12875980e-02 -7.42673516e-01 5.96896291e-01
1.09642200e-01 -1.05996311e+00 1.45332724e-01 -1.21611170e-01
7.86779821e-01 5.28511465e-01 -6.55234933e-01 5.50516367e-01
7.42602468e-01 -1.05024016e+00 3.38236928e-01 4.75713730e-01
9.92030919e-01 -7.91140974e-01 9.08164203e-01 4.82616007e-01
-1.35824978e+00 5.59100151e-01 -7.50080884e-01 2.33637363e-01
5.31865478e-01 6.89201057e-01 -8.95865440e-01 7.22228944e-01
5.14836848e-01 3.86563480e-01 -3.95113438e-01 1.00428307e+00
-5.80246508e-01 7.32803106e-01 -3.90815318e-01 -3.87800753e-01
2.69223958e-01 -1.84816882e-01 1.08099647e-01 1.85876262e+00
1.94959879e-01 2.13547036e-01 5.98626018e-01 4.51556146e-01
-7.63643861e-01 6.96786761e-01 -5.67653120e-01 -5.74949324e-01
3.49824786e-01 1.26446080e+00 -7.97123492e-01 -6.27506435e-01
-6.21296525e-01 1.26159024e+00 1.11432159e+00 7.29679018e-02
-5.39763033e-01 -6.80834830e-01 3.29556882e-01 -2.84274608e-01
3.72112691e-01 -5.90569913e-01 -5.42647503e-02 -1.45649505e+00
-2.34225780e-01 -5.99026263e-01 4.34262633e-01 -5.21219850e-01
-1.81319714e+00 1.07619298e+00 -3.97457719e-01 -9.47570682e-01
-2.20133096e-01 -6.93387270e-01 -1.25825241e-01 1.19832170e+00
-1.77098215e+00 -1.31335723e+00 2.74618924e-01 3.56775284e-01
3.98806572e-01 -2.19556898e-01 1.42262805e+00 5.54823577e-01
-4.64367151e-01 3.82148772e-01 1.83886170e-01 8.62585664e-01
8.60664308e-01 -1.17169392e+00 6.86416864e-01 8.64365518e-01
5.32394707e-01 7.95360506e-01 2.19068483e-01 -8.57145011e-01
-1.03911078e+00 -1.18486965e+00 1.89027429e+00 -2.76580155e-01
1.00194025e+00 -3.41190666e-01 -1.16680300e+00 8.92381728e-01
2.27570638e-01 3.09386641e-01 1.30523121e+00 4.56365645e-01
-6.43245280e-01 3.90445858e-01 -9.89140809e-01 3.42586040e-01
7.59781361e-01 -8.07786942e-01 -8.02584827e-01 5.01741111e-01
6.84389830e-01 -3.12517494e-01 -1.25736892e+00 2.12528966e-02
2.11996466e-01 -2.80538112e-01 5.22067904e-01 -1.08773434e+00
3.16584885e-01 -2.02356398e-01 -1.82747081e-01 -1.06940174e+00
-2.47423008e-01 -3.59395772e-01 3.18212688e-01 1.68032670e+00
9.26458836e-01 -6.50032759e-01 6.12848103e-01 7.15741694e-01
-1.32349739e-02 -1.85610458e-01 -1.00361502e+00 -8.97096455e-01
4.53897476e-01 -3.83085012e-01 3.54225159e-01 1.37595987e+00
8.92484635e-02 7.64228702e-01 -3.63478482e-01 3.86398852e-01
2.04399079e-01 1.63181037e-01 3.27770561e-01 -1.21432197e+00
-1.73005983e-01 -3.07789315e-02 -3.42022717e-01 -1.00612342e+00
8.04050207e-01 -1.08579290e+00 3.67219806e-01 -1.71171463e+00
2.98698902e-01 -9.22464609e-01 -3.74874443e-01 9.65614736e-01
-2.21599087e-01 2.07240626e-01 1.97927848e-01 1.73554793e-01
-8.07537198e-01 3.30123723e-01 7.56156862e-01 2.14268789e-01
-1.29910782e-01 -2.39275724e-01 -3.60649288e-01 8.06843936e-01
1.02722418e+00 -8.00833344e-01 3.58044565e-01 -8.12729359e-01
2.18059748e-01 1.47348553e-01 -4.29467976e-01 -7.89911509e-01
1.91673785e-01 -3.99622381e-01 6.24241889e-01 -4.73112822e-01
9.35887769e-02 -6.90169513e-01 -1.03588268e-01 1.93869770e-01
-3.53559971e-01 -3.24728601e-02 3.91906917e-01 3.55235159e-01
-3.66285086e-01 -5.99079251e-01 8.08939695e-01 -4.69088107e-01
-9.35201705e-01 3.15678895e-01 -5.18875003e-01 4.75287944e-01
8.83291185e-01 4.13701653e-01 -3.23228598e-01 1.22602850e-01
-5.45643449e-01 -1.34612560e-01 4.20179844e-01 2.99838305e-01
2.57308841e-01 -9.46739912e-01 -8.51790607e-01 -7.36313313e-02
2.53337264e-01 -8.52785483e-02 1.17273889e-01 1.46384165e-01
-7.79497206e-01 6.50521874e-01 -4.37722981e-01 -4.91600931e-02
-9.63823140e-01 4.00049061e-01 -1.27243489e-01 -8.06572437e-01
-2.94909954e-01 8.81807327e-01 -1.08443499e-01 -8.50445867e-01
-2.86179632e-01 1.04547314e-01 -4.57786947e-01 -5.96343614e-02
5.20180821e-01 1.35680428e-03 -5.58622666e-02 -1.01138747e+00
-6.39516175e-01 2.88800657e-01 -2.40293711e-01 -1.41033471e-01
1.40034378e+00 1.21128008e-01 -2.04322532e-01 4.98733640e-01
1.18132532e+00 3.35309982e-01 -6.47802830e-01 -3.30748677e-01
7.30897009e-01 -9.46542528e-03 -3.22629988e-01 -7.50834405e-01
-6.56669855e-01 9.84440506e-01 1.11462682e-01 -2.24628508e-01
8.19158852e-01 2.32910633e-01 8.86531532e-01 8.40438366e-01
7.18883157e-01 -1.00511301e+00 -4.00180489e-01 1.17935956e+00
4.26348418e-01 -1.22406459e+00 -1.06966652e-01 -5.20134091e-01
-8.52386236e-01 1.00679040e+00 5.11072636e-01 -6.14621527e-02
3.99826050e-01 3.12075257e-01 2.01628923e-01 7.69616663e-02
-3.92410427e-01 -3.39106441e-01 3.58151436e-01 5.84786415e-01
1.02674425e+00 1.70301646e-01 -2.22754776e-01 6.40522361e-01
-1.45229669e-02 -1.34076387e-01 3.26199740e-01 1.07682538e+00
-5.70177019e-01 -1.54901469e+00 -1.23149298e-01 2.61233211e-01
-9.47335958e-01 -5.96781433e-01 -3.44044566e-01 7.60184169e-01
3.89789715e-02 8.12606931e-01 1.38546184e-01 2.85037100e-01
2.94640690e-01 4.17897969e-01 1.10662840e-01 -1.18735754e+00
-8.00191462e-01 2.09532548e-02 5.84074676e-01 -1.39558464e-01
-6.25378788e-01 -3.85306269e-01 -1.65914774e+00 -5.23050055e-02
-4.06368941e-01 4.23779219e-01 6.62200570e-01 9.32666421e-01
2.95525491e-01 1.55754343e-01 1.92040011e-01 -4.95596528e-01
-1.15860716e-01 -1.02243090e+00 -7.63009012e-01 1.52398586e-01
-3.22053641e-01 -1.15017682e-01 1.75952092e-01 2.64861166e-01] | [10.051029205322266, 9.803346633911133] |
c7243f62-697a-4f04-94ed-177aa667c1f2 | individual-causal-inference-using-panel-data | 2306.01969 | null | https://arxiv.org/abs/2306.01969v1 | https://arxiv.org/pdf/2306.01969v1.pdf | Individual Causal Inference Using Panel Data With Multiple Outcomes | Policy evaluation in empirical microeconomics has been focusing on estimating the average treatment effect and more recently the heterogeneous treatment effects, often relying on the unconfoundedness assumption. We propose a method based on the interactive fixed effects model to estimate treatment effects at the individual level, which allows both the treatment assignment and the potential outcomes to be correlated with the unobserved individual characteristics. This method is suitable for panel datasets where multiple related outcomes are observed for a large number of individuals over a small number of time periods. Monte Carlo simulations show that our method outperforms related methods. To illustrate our method, we provide an example of estimating the effect of health insurance coverage on individual usage of hospital emergency departments using the Oregon Health Insurance Experiment data. | ['Wei Tian'] | 2023-06-03 | null | null | null | null | ['causal-inference', 'causal-inference'] | ['knowledge-base', 'miscellaneous'] | [-8.42817798e-02 1.45029649e-01 -1.20720279e+00 -2.48634726e-01
-5.20637155e-01 -3.23662490e-01 1.56192258e-01 2.83941060e-01
-4.67883587e-01 1.22747695e+00 6.97148442e-01 -7.30099320e-01
-2.33789518e-01 -8.28629375e-01 -6.86707258e-01 -3.56938154e-01
-3.41919243e-01 4.30331290e-01 -4.81574655e-01 4.59415495e-01
-5.46487458e-02 3.10453251e-02 -9.31643009e-01 -1.09953135e-01
1.14131260e+00 1.56502247e-01 -4.80033308e-02 1.12359293e-01
4.86432791e-01 5.05187452e-01 -2.76317239e-01 -3.94395381e-01
2.42672309e-01 -2.48763666e-01 -3.04999948e-01 2.79525876e-01
-1.58483088e-01 -6.32171929e-01 -1.84634581e-01 7.16987014e-01
5.96094549e-01 1.03187777e-01 1.10239255e+00 -1.28173602e+00
-7.83256829e-01 9.25358057e-01 -5.99613845e-01 -6.14226237e-02
3.72195244e-01 3.24556947e-01 9.03574884e-01 5.27514704e-02
6.03030622e-01 1.67252386e+00 5.55060804e-01 2.72481114e-01
-1.98092759e+00 -5.92815161e-01 1.82114094e-01 -3.31979066e-01
-9.66906846e-01 -2.86985189e-01 2.39381194e-01 -8.88179600e-01
5.05821347e-01 -8.02764595e-02 5.54077983e-01 1.07974017e+00
7.17986941e-01 3.44679445e-01 1.38383436e+00 -4.57105458e-01
2.96656340e-01 1.19253978e-01 4.46379364e-01 2.21829981e-01
8.26482594e-01 8.28888118e-01 3.14655840e-01 -9.12018478e-01
1.14226651e+00 4.24131006e-01 -1.23059258e-01 -1.83979094e-01
-9.84985769e-01 1.11606467e+00 6.20737672e-02 -5.00730649e-02
-1.17628038e+00 2.52239764e-01 2.95549572e-01 2.25610137e-01
4.17491645e-01 1.47493571e-01 -8.24891627e-01 2.80562073e-01
-4.40304905e-01 5.10855794e-01 6.55394852e-01 4.71761256e-01
6.07199907e-01 -2.00453192e-01 -7.57299542e-01 5.72573960e-01
5.66541217e-02 7.89897263e-01 3.00929278e-01 -1.18784142e+00
4.64936137e-01 3.94103199e-01 8.46094549e-01 -4.96641904e-01
-4.36801225e-01 -8.76842253e-03 -7.20757484e-01 2.04781547e-01
6.69361055e-01 -1.15500426e+00 -8.17093313e-01 2.34684730e+00
2.96179771e-01 1.22790113e-01 -8.27055126e-02 3.05658847e-01
8.28459784e-02 2.79783994e-01 7.26193368e-01 -6.70594871e-01
1.44934690e+00 -3.87033910e-01 -9.50436711e-01 2.33094886e-01
9.91566718e-01 -8.76310393e-02 6.69226408e-01 -2.11827472e-01
-1.35235763e+00 -1.70730323e-01 1.39770627e-01 5.35014510e-01
2.78585479e-02 -6.98664561e-02 8.07779312e-01 8.53137910e-01
-7.61804998e-01 5.74017584e-01 -6.56525731e-01 -3.67469311e-01
4.73951429e-01 5.02930820e-01 3.44962068e-02 -1.10909957e-02
-1.12595737e+00 7.11640477e-01 -1.47029579e-01 -6.17603958e-01
-8.56070638e-01 -1.07384706e+00 -6.28935456e-01 7.73820162e-01
4.20942336e-01 -1.44435072e+00 1.21909094e+00 -1.04477382e+00
-1.28341055e+00 2.55942732e-01 -1.60915226e-01 -3.81843746e-01
2.33247414e-01 3.91241193e-01 -2.70531029e-01 -4.11304951e-01
5.31409264e-01 1.14548147e-01 -3.10715716e-02 -1.02102578e+00
-5.31468630e-01 -6.16439581e-01 1.17084607e-02 2.48466551e-01
1.58793211e-01 2.64788181e-01 4.67804909e-01 -7.91544139e-01
-5.95943809e-01 -1.08608127e+00 -7.24712551e-01 -8.54842842e-01
-3.78448546e-01 -1.78939238e-01 -1.63509592e-01 -4.43940371e-01
1.16969693e+00 -2.06920433e+00 -2.04875126e-01 3.87694947e-02
-6.79242313e-02 -5.02540052e-01 -1.10728607e-01 5.71878850e-01
-3.74763757e-01 5.49350977e-01 -1.15213424e-01 -7.64058232e-02
1.78369984e-01 -6.05054945e-02 7.28322938e-02 6.43409252e-01
-3.37925524e-01 8.06510508e-01 -4.61761892e-01 -2.43812487e-01
5.55124208e-02 -1.78931296e-01 -8.46680760e-01 -1.53156370e-01
2.82358617e-01 4.78931487e-01 -6.71398818e-01 2.97401786e-01
6.72707677e-01 -3.10827017e-01 5.91430008e-01 6.65903091e-01
-2.84457475e-01 3.31552833e-01 -9.98543501e-01 8.84157360e-01
-2.51038074e-01 5.15602790e-02 5.08390814e-02 -1.24216378e+00
-3.92193869e-02 9.18187797e-01 7.84049153e-01 -3.53097886e-01
1.95121393e-01 -7.46326819e-02 4.37065959e-01 -6.94789827e-01
-1.88278005e-01 -7.52975166e-01 -3.33297610e-01 5.47215879e-01
-1.76845685e-01 3.91014218e-01 -2.89599113e-02 1.80609807e-01
1.06296885e+00 -3.82720828e-01 7.15864241e-01 -7.87841141e-01
-2.50918984e-01 3.94891351e-02 1.08417380e+00 1.04322743e+00
-1.24042109e-01 -1.06267445e-01 7.71588981e-01 4.25314941e-02
-9.09706414e-01 -9.74785149e-01 -4.75118965e-01 5.79069734e-01
-1.32161677e-01 4.48961645e-01 -6.56919897e-01 -4.37100917e-01
7.29213238e-01 1.01871121e+00 -1.08972335e+00 9.53247547e-02
-9.00366381e-02 -1.28985357e+00 7.03987256e-02 5.75405836e-01
3.15209985e-01 -9.47057009e-01 -3.15385789e-01 2.13851228e-01
-3.13115120e-02 -4.26525414e-01 -4.12291527e-01 -4.08666283e-01
-1.12093008e+00 -8.52655411e-01 -8.35097969e-01 -2.74461001e-01
5.26783466e-01 1.09636597e-01 9.90546763e-01 -8.09365138e-02
2.65824348e-01 4.29073125e-01 1.19622096e-01 -7.66917527e-01
-2.04794526e-01 -2.50497669e-01 2.27267489e-01 -4.51558270e-02
3.91783953e-01 -4.67709124e-01 -7.77307868e-01 -4.85020205e-02
-7.82681704e-01 -2.19759062e-01 3.28260392e-01 9.28263009e-01
-5.08084372e-02 -7.10286573e-02 1.19551945e+00 -1.34058511e+00
5.48552513e-01 -9.84913468e-01 -7.59288013e-01 4.74976748e-01
-7.82509446e-01 -2.87936211e-01 1.34117126e-01 -8.40417743e-01
-1.46747196e+00 -2.21715689e-01 5.61571121e-01 1.07918657e-01
-3.60936493e-01 9.06312525e-01 -3.51610631e-01 5.53395331e-01
4.45262581e-01 -6.32818460e-01 2.64202021e-02 -4.95244652e-01
-8.10110569e-02 5.85485816e-01 -1.18785180e-01 -7.80096889e-01
2.33225212e-01 3.99800658e-01 3.83489430e-02 -1.95510477e-01
-2.62750506e-01 -8.31546634e-02 -6.03689104e-02 1.92174852e-01
9.79709446e-01 -1.40737343e+00 -1.23581898e+00 9.91286933e-02
-6.73431993e-01 -9.34596121e-01 -3.32553536e-01 1.36202776e+00
-5.59898674e-01 -1.04416080e-01 -7.64498472e-01 -1.15140629e+00
4.25442368e-01 -1.17953348e+00 5.67326188e-01 2.43457735e-01
-2.16940537e-01 -1.37498093e+00 3.42449218e-01 7.65311643e-02
9.05789807e-02 4.16834205e-01 1.27808666e+00 -4.06456620e-01
-4.48698223e-01 -3.08012575e-01 8.99651349e-02 -6.07373893e-01
4.33702976e-01 3.61997187e-02 -2.87559688e-01 -3.87973726e-01
-1.32617369e-01 -5.96705265e-03 4.04318154e-01 1.69257283e+00
1.07653868e+00 -5.83542764e-01 -7.74457455e-01 1.17485099e-01
1.57430387e+00 6.42243981e-01 6.55290186e-01 2.71427948e-02
1.54740065e-01 8.81499827e-01 3.38124365e-01 8.11242998e-01
5.90729117e-01 7.75868118e-01 -2.88303588e-02 -4.35883015e-01
7.06109166e-01 -3.49207401e-01 5.30498959e-02 -2.39674211e-01
-1.70381919e-01 -5.82516074e-01 -6.56410754e-01 8.87443781e-01
-1.93712378e+00 -1.36392522e+00 -3.99983168e-01 2.73630357e+00
6.18424118e-01 -2.58402318e-01 7.35852301e-01 -3.87538403e-01
1.01717150e+00 -5.58417737e-01 -3.56530070e-01 -5.41714191e-01
3.09814289e-02 -4.97001037e-02 1.10148621e+00 6.87444687e-01
-5.09937525e-01 3.36385310e-01 8.16995335e+00 3.64285618e-01
-4.20754224e-01 1.90934435e-01 1.25432289e+00 -9.57239512e-03
-4.74192202e-01 4.96722609e-01 -6.43268883e-01 7.05939054e-01
1.16845751e+00 -7.08407283e-01 1.14282174e-02 2.89612174e-01
1.01517332e+00 -5.31888485e-01 -9.91944611e-01 1.54777557e-01
-7.97580063e-01 -9.42069709e-01 -4.19253737e-01 8.37323308e-01
1.40835357e+00 -4.80403692e-01 2.11378887e-01 4.84228581e-01
1.34272122e+00 -9.01414871e-01 2.73355573e-01 4.21591401e-01
8.65665257e-01 -7.28791356e-01 6.20774806e-01 6.05806231e-01
-4.34209287e-01 -6.64292634e-01 -1.04411684e-01 -7.04912841e-01
5.11287451e-01 7.08652437e-01 -4.32997286e-01 2.83093542e-01
2.85833180e-01 1.79159984e-01 2.01950833e-01 1.11070931e+00
9.55752730e-02 9.10563350e-01 4.12763841e-02 6.13445222e-01
1.06232382e-01 -3.82580072e-01 1.60283387e-01 6.68986619e-01
5.41544437e-01 9.68821704e-01 1.83794439e-01 9.09240842e-01
-2.42091134e-01 2.12704927e-01 -8.67070675e-01 -1.61424987e-02
4.12283123e-01 6.14674687e-01 -5.68911970e-01 -8.11933219e-01
-8.11018229e-01 2.26876944e-01 -9.40360576e-02 7.49517620e-01
-7.47306824e-01 2.85922199e-01 6.21572495e-01 3.06173652e-01
-5.76508678e-02 4.35386658e-01 -5.73044896e-01 -1.26690328e+00
-5.85805774e-01 -7.16950476e-01 6.80415571e-01 -4.90780294e-01
-1.21220303e+00 -6.63664401e-01 6.26588285e-01 -6.01803839e-01
-2.39776641e-01 -1.38165906e-01 -6.48203611e-01 1.33491671e+00
-1.08691680e+00 -3.52832049e-01 5.06375670e-01 3.46565813e-01
2.48757377e-01 4.31295753e-01 7.59160638e-01 1.13879912e-01
-1.13644707e+00 3.08158100e-01 5.36517322e-01 -1.52641848e-01
6.27499819e-01 -1.19802558e+00 5.77628287e-03 4.32791084e-01
-7.50590801e-01 7.01034844e-01 6.43807828e-01 -1.23147881e+00
-5.17210305e-01 -7.08816588e-01 1.04480374e+00 -2.05082268e-01
3.74057829e-01 -4.29078266e-02 -6.65885627e-01 1.31633055e+00
3.81705076e-01 -5.08710027e-01 7.20145822e-01 4.84035909e-01
2.09088624e-01 4.87895533e-02 -1.52833438e+00 7.44641244e-01
8.23412001e-01 -1.39865763e-02 -2.34482542e-01 5.13073623e-01
6.36600256e-01 1.59864411e-01 -1.11722398e+00 4.99949217e-01
5.88561833e-01 -7.97413349e-01 7.51802206e-01 -8.83459449e-01
4.50466007e-01 4.74503309e-01 1.41382301e-02 -1.60535383e+00
-9.60923731e-01 -3.26388150e-01 7.24553227e-01 1.22773564e+00
6.52953923e-01 -1.03475702e+00 2.54660130e-01 1.25049233e+00
3.99954230e-01 -2.65525490e-01 -7.05942571e-01 -7.77093112e-01
5.25947690e-01 4.08662409e-01 1.02252412e+00 1.41189241e+00
2.69695610e-01 2.33949795e-01 -4.47703958e-01 2.06422046e-01
6.94817722e-01 2.39137053e-01 5.74313939e-01 -1.45477939e+00
-5.54910004e-01 -2.19792664e-01 2.33562410e-01 -4.18759495e-01
4.73621160e-01 -1.66260689e-01 -3.18226516e-01 -1.30503356e+00
1.12661064e+00 -4.88652736e-01 -2.66062945e-01 2.36526668e-01
-5.24802864e-01 -4.81695354e-01 -8.11152235e-02 -1.44816831e-01
1.94320679e-01 3.31998229e-01 9.35978293e-01 1.29306227e-01
-5.41619360e-01 3.04538518e-01 -7.84080207e-01 5.10356545e-01
7.11479187e-01 -7.73875296e-01 -3.59717816e-01 -1.49760634e-01
-3.93279254e-01 1.33162987e+00 4.00979757e-01 -2.31045976e-01
-3.62767667e-01 -1.22184551e+00 1.70007631e-01 -9.28317010e-02
-1.16050959e-01 -7.96648860e-01 9.07054543e-01 8.50050390e-01
-7.30943322e-01 2.72587061e-01 1.20883808e-01 5.99016607e-01
2.14511409e-01 -3.52514423e-02 6.53775871e-01 -2.70244271e-01
3.65178525e-01 4.33071673e-01 -8.79142523e-01 -1.09527722e-01
1.09117234e+00 1.87544391e-01 -2.81113207e-01 -6.02917194e-01
-8.30844998e-01 2.62931406e-01 8.41491818e-01 -3.41510326e-01
-1.31924257e-01 -1.41271424e+00 -9.59317148e-01 -3.84341806e-01
-1.16701506e-01 -9.77597296e-01 6.49893045e-01 9.22062099e-01
3.15925330e-01 6.44537687e-01 -2.23771423e-01 1.84322804e-01
-1.13086748e+00 1.22565448e+00 3.93115789e-01 -5.30009747e-01
-3.64333779e-01 8.87833983e-02 1.17816103e+00 6.91801980e-02
-3.52990538e-01 -5.85565157e-02 -1.55994028e-01 -1.08257376e-01
3.06973904e-01 6.88301206e-01 -7.54999042e-01 -5.30927598e-01
-1.05944127e-01 1.10138357e-01 4.36371505e-01 -5.39241552e-01
1.25049376e+00 -6.61693156e-01 -1.04616717e-01 6.39121354e-01
8.79049540e-01 -3.44724394e-02 -1.19210815e+00 -5.05172275e-02
-1.54200092e-01 -5.04389524e-01 3.94870900e-02 -7.24853098e-01
-9.45170760e-01 2.87344068e-01 5.72731972e-01 4.06370014e-01
1.09300411e+00 -5.09432316e-01 -1.10076712e-02 -3.63440484e-01
3.33232135e-01 -7.67705917e-01 -4.93156970e-01 -4.30463165e-01
2.72143900e-01 -1.30092502e+00 8.60380828e-02 -4.39484805e-01
-6.13421381e-01 1.18907668e-01 9.61641371e-02 -3.59004021e-01
9.28773940e-01 2.25553606e-02 -3.00501972e-01 1.61711965e-02
-1.19880903e+00 -1.97903335e-01 -6.30294085e-02 4.03022826e-01
5.52230060e-01 8.28764498e-01 -1.25051379e+00 4.71249104e-01
4.08689320e-01 3.41838956e-01 1.00937605e+00 6.23764992e-01
5.99949211e-02 -1.16541338e+00 -7.71595061e-01 9.37970817e-01
-1.22290146e+00 -8.93465281e-02 9.19853617e-03 9.10845816e-01
1.35088399e-01 1.02036929e+00 3.64925295e-01 4.19958442e-01
5.11081934e-01 2.13146076e-01 1.67467251e-01 -7.15948641e-01
-4.76386696e-01 4.80115503e-01 1.87185988e-01 -1.09926932e-01
-8.72715175e-01 -1.15607071e+00 -6.71926975e-01 -7.99875021e-01
-4.49023217e-01 2.08840594e-01 1.15209334e-02 9.19626534e-01
3.19021791e-01 2.87906975e-01 1.07322633e+00 -6.73891783e-01
-6.11820161e-01 -1.00177920e+00 -1.28971922e+00 6.32964253e-01
3.54544401e-01 -9.17925715e-01 -5.06409168e-01 -1.69160947e-01] | [7.894723892211914, 5.172478199005127] |
815344c0-3de5-49fa-bd1c-376b2ccf45bc | 190510752 | 1905.10752 | null | https://arxiv.org/abs/1905.10752v1 | https://arxiv.org/pdf/1905.10752v1.pdf | TIGS: An Inference Algorithm for Text Infilling with Gradient Search | Text infilling is defined as a task for filling in the missing part of a sentence or paragraph, which is suitable for many real-world natural language generation scenarios. However, given a well-trained sequential generative model, generating missing symbols conditioned on the context is challenging for existing greedy approximate inference algorithms. In this paper, we propose an iterative inference algorithm based on gradient search, which is the first inference algorithm that can be broadly applied to any neural sequence generative models for text infilling tasks. We compare the proposed method with strong baselines on three text infilling tasks with various mask ratios and different mask strategies. The results show that our proposed method is effective and efficient for fill-in-the-blank tasks, consistently outperforming all baselines. | ['PengFei Liu', 'Dayiheng Liu', 'Jie Fu', 'Jiancheng Lv'] | 2019-05-26 | tigs-an-inference-algorithm-for-text | https://aclanthology.org/P19-1406 | https://aclanthology.org/P19-1406.pdf | acl-2019-7 | ['text-infilling'] | ['natural-language-processing'] | [ 6.50825739e-01 2.94528957e-02 -1.00564957e-01 -2.65499353e-01
-7.68330753e-01 -3.70628059e-01 8.54932904e-01 -1.21375434e-01
-3.37347895e-01 1.23743498e+00 4.31613833e-01 -6.57833397e-01
4.36068535e-01 -7.92319477e-01 -8.36370111e-01 -4.38514918e-01
6.35839343e-01 6.92314506e-01 3.78269702e-02 -1.12558916e-01
4.95621622e-01 9.50610489e-02 -1.25926113e+00 5.14322221e-01
1.08237541e+00 4.49974269e-01 7.97047615e-01 6.40307367e-01
-7.84314573e-01 7.03838944e-01 -8.65773976e-01 -6.34248853e-01
-1.80721402e-01 -8.48693073e-01 -6.42675698e-01 -9.72608775e-02
5.30188501e-01 -4.33461130e-01 -2.89163828e-01 8.79233301e-01
5.95681667e-01 1.40349612e-01 7.21815109e-01 -7.64398754e-01
-5.84704876e-01 1.17264402e+00 -5.62146544e-01 1.01572238e-01
5.01282156e-01 -3.99811976e-02 9.17357206e-01 -1.29118431e+00
5.11684120e-01 1.44650018e+00 4.65901941e-01 6.78267658e-01
-1.18156493e+00 -5.38605809e-01 3.94875675e-01 -1.19926438e-01
-1.25905848e+00 -5.30133486e-01 4.55753714e-01 -1.17727518e-01
1.21732068e+00 1.81062520e-01 3.91252875e-01 1.15139151e+00
2.97466367e-01 1.03726757e+00 8.65941703e-01 -6.99747562e-01
2.85967112e-01 -1.57250419e-01 -9.31826532e-02 7.31981337e-01
3.77200872e-01 -3.57751220e-01 -7.97668338e-01 -2.17228636e-01
6.06935680e-01 -3.56190391e-02 -6.95819035e-02 3.72321218e-01
-1.23676419e+00 8.49838674e-01 -1.42065689e-01 1.27088860e-01
-3.22879344e-01 4.87397403e-01 3.83216560e-01 -1.71433628e-01
7.57784963e-01 1.44072175e-01 -1.99976444e-01 -2.82355875e-01
-1.63620079e+00 6.14000320e-01 8.86469483e-01 1.19964755e+00
5.40829718e-01 3.36160570e-01 -8.96015882e-01 9.19236124e-01
2.91091114e-01 5.84265232e-01 4.38159347e-01 -5.39023936e-01
8.01366389e-01 2.74207264e-01 2.64577150e-01 -5.09555697e-01
4.55150418e-02 -5.02207041e-01 -1.12010539e+00 -2.07924619e-01
2.27800012e-01 -4.84415114e-01 -1.04994380e+00 1.76480746e+00
1.62995934e-01 3.00134093e-01 -1.98558077e-01 4.15287733e-01
7.92880714e-01 9.15445924e-01 -9.96151567e-02 -5.10271132e-01
1.27243328e+00 -1.14911389e+00 -1.14883196e+00 -7.07558155e-01
3.67993444e-01 -1.00240493e+00 1.20481336e+00 3.31160516e-01
-1.26146078e+00 -5.32268524e-01 -1.05851412e+00 -3.45675737e-01
-1.78705662e-01 4.33463454e-01 6.25177264e-01 5.72518587e-01
-9.67499197e-01 4.62180525e-01 -7.32609093e-01 -6.32395521e-02
2.43638769e-01 -1.69077575e-01 1.07165411e-01 -1.89566106e-01
-1.11761832e+00 8.14370215e-01 7.38674343e-01 4.27908659e-01
-8.38833988e-01 -5.38987100e-01 -8.46739292e-01 1.35282189e-01
3.63852620e-01 -1.09418666e+00 1.52641666e+00 -3.27182531e-01
-1.54509640e+00 5.15372515e-01 -1.08343172e+00 -7.06149638e-01
6.66768909e-01 -5.67411542e-01 -1.15127089e-02 -3.30880702e-01
1.47609338e-01 6.58886135e-01 1.03315330e+00 -1.00724649e+00
-3.76030773e-01 1.07556999e-01 -2.83861220e-01 2.46552750e-01
4.90753539e-02 -1.23504892e-01 -5.65179348e-01 -1.15808976e+00
2.04426590e-02 -6.16643012e-01 -3.59612316e-01 -4.17857975e-01
-7.40145564e-01 -3.51637453e-01 6.74525023e-01 -8.79206598e-01
1.82849789e+00 -1.66643178e+00 7.58663341e-02 -1.88500524e-01
-1.99836984e-01 2.93421000e-01 -8.45326856e-02 8.33068073e-01
2.62712270e-01 2.18396947e-01 -5.56573212e-01 -8.79997134e-01
3.32844317e-01 2.15701222e-01 -8.36183667e-01 -1.61104083e-01
1.44849747e-01 1.13676417e+00 -9.14398670e-01 -6.23735547e-01
1.25452876e-01 2.16634899e-01 -4.53509241e-01 3.35451782e-01
-7.43862510e-01 1.07458517e-01 -1.79668859e-01 3.57039481e-01
6.44606054e-01 -1.28454670e-01 1.57487303e-01 1.93079531e-01
1.09208010e-01 6.20018363e-01 -1.10192025e+00 2.09743309e+00
-6.04149818e-01 6.72411144e-01 -4.54378426e-01 -6.75369143e-01
9.73806381e-01 2.80515045e-01 -3.68674040e-01 -3.19167256e-01
-1.39380634e-01 2.95703083e-01 -3.24356496e-01 -2.01574847e-01
9.66080725e-01 -1.50364101e-01 -6.05724044e-02 7.15796828e-01
-1.60013158e-02 -4.29267377e-01 7.90957570e-01 4.79813010e-01
9.29970801e-01 3.76655787e-01 3.18519592e-01 1.02285817e-02
3.83721113e-01 -3.77080530e-01 2.77461618e-01 1.38845730e+00
4.82543439e-01 7.36048102e-01 4.41640466e-01 -2.31983110e-01
-1.08660412e+00 -1.04326713e+00 8.83332714e-02 8.45040262e-01
-3.08287054e-01 -7.28930652e-01 -9.68982220e-01 -5.63790500e-01
-3.10520113e-01 1.21100760e+00 -3.63357931e-01 -1.26226032e-02
-7.56782115e-01 -8.18449318e-01 5.86015105e-01 5.38190544e-01
6.11058414e-01 -1.39801955e+00 -2.32843295e-01 4.87561464e-01
-5.89104712e-01 -1.16825521e+00 -7.88618028e-01 -1.00144386e-01
-1.02595365e+00 -5.04445612e-01 -8.77347648e-01 -8.24798644e-01
7.83800721e-01 8.93351063e-02 1.38583434e+00 2.83800185e-01
-1.27780795e-01 -9.76767018e-02 -2.72818416e-01 -5.00099063e-01
-6.27312422e-01 3.17601413e-01 -3.45869094e-01 -2.30644748e-01
1.72585338e-01 -3.58024895e-01 -3.51376116e-01 -1.77585557e-01
-1.07208514e+00 6.85304046e-01 6.66431487e-01 1.01871932e+00
6.21618390e-01 -2.43352093e-02 6.21424198e-01 -1.12496102e+00
1.15710068e+00 -3.77753347e-01 -3.50415081e-01 4.43156809e-01
-4.61077601e-01 4.53177243e-01 6.48963094e-01 -2.51447856e-01
-1.32358229e+00 -1.11873001e-01 -4.36524183e-01 9.25560221e-02
-8.44179988e-02 7.75537968e-01 -1.86734602e-01 6.32554889e-01
3.75029355e-01 7.33631432e-01 -4.11203623e-01 -4.99222010e-01
5.40174901e-01 5.37742436e-01 5.90408146e-01 -7.60440290e-01
6.77345395e-01 1.81446671e-01 -1.08746417e-01 -6.97630703e-01
-1.01665974e+00 1.12307342e-02 -4.63778198e-01 1.16054460e-01
6.48244262e-01 -8.05905581e-01 -1.24976419e-01 5.82192361e-01
-1.61565244e+00 -4.38328594e-01 -1.08781077e-01 2.53177762e-01
-4.56228971e-01 6.68228328e-01 -6.68414891e-01 -1.01311874e+00
-8.60685647e-01 -9.90176141e-01 1.18665695e+00 1.56456038e-01
-4.74071205e-01 -9.58996952e-01 1.51505664e-01 1.35880023e-01
4.67643738e-01 -4.81144302e-02 9.17160451e-01 -4.16658610e-01
-5.23145616e-01 -1.31673783e-01 -1.21943124e-01 2.58178204e-01
1.18096925e-01 -1.11223169e-01 -8.13951015e-01 -1.19539887e-01
-1.98727354e-01 -2.40367085e-01 1.21844923e+00 3.88757110e-01
1.30550206e+00 -5.84677935e-01 -3.53251517e-01 3.05277169e-01
1.12882876e+00 8.51291642e-02 8.42225254e-01 -5.45847304e-02
5.18314660e-01 7.24606365e-02 5.20443857e-01 7.04072833e-01
2.15313435e-01 4.24612254e-01 4.40728776e-02 -3.32655795e-02
-2.59128332e-01 -8.13297451e-01 4.02487993e-01 7.67675102e-01
3.35483491e-01 -7.64070094e-01 -5.98413944e-01 5.75904965e-01
-2.05415177e+00 -1.08258092e+00 -5.57041802e-02 2.08901906e+00
1.25569618e+00 4.97559905e-01 -2.11195797e-01 3.12564187e-02
9.16922033e-01 2.96937823e-01 -4.57971960e-01 -6.02932751e-01
-7.35917464e-02 6.81915879e-01 -5.21092489e-02 7.23257124e-01
-7.59005070e-01 1.26909256e+00 7.03263474e+00 1.25578523e+00
-5.58648586e-01 -7.55008236e-02 6.95267737e-01 -1.10423781e-01
-5.74394047e-01 1.70372888e-01 -1.27021646e+00 7.52941430e-01
8.38839412e-01 -1.48283988e-01 4.78324533e-01 5.15879989e-01
5.67783356e-01 -2.07583711e-01 -1.19246435e+00 8.50747645e-01
2.31441021e-01 -1.58333170e+00 5.73359013e-01 -2.61247486e-01
9.77693260e-01 -4.45612043e-01 -1.43854022e-01 3.51639807e-01
4.25939918e-01 -1.21281195e+00 8.23757172e-01 6.06934071e-01
7.76974142e-01 -6.02285862e-01 6.26532555e-01 6.21362865e-01
-1.09284747e+00 4.52994734e-01 -3.33689570e-01 -2.26017222e-01
7.20489502e-01 9.28016484e-01 -1.01982594e+00 3.83193880e-01
1.72721237e-01 4.50111151e-01 -3.25313210e-01 1.01940548e+00
-7.62562811e-01 9.01769161e-01 -2.34264120e-01 -4.37694937e-01
2.06784531e-01 -1.04557179e-01 4.16454703e-01 1.66136408e+00
6.42669797e-01 -2.22737744e-01 4.74188402e-02 1.12094998e+00
-3.15104127e-01 -4.85775545e-02 -5.49828351e-01 -3.06802690e-01
7.18861759e-01 9.88862693e-01 -7.85585880e-01 -6.87812030e-01
-1.06088512e-01 1.18739057e+00 2.62785256e-01 5.04403889e-01
-1.03052294e+00 -6.75982833e-01 7.82872215e-02 -6.73207715e-02
5.26656926e-01 -5.06953299e-01 -4.53752130e-01 -1.17501533e+00
3.41747344e-01 -9.06071305e-01 1.04744181e-01 -1.11009455e+00
-1.04072988e+00 4.88631904e-01 1.92017198e-01 -8.03912699e-01
-6.31472230e-01 -2.54688323e-01 -9.11567509e-01 1.11594808e+00
-1.40339196e+00 -1.08228457e+00 -1.65184915e-01 1.82734385e-01
1.08696926e+00 -4.86215204e-02 6.99204266e-01 -1.71340292e-03
-5.50096393e-01 7.36740053e-01 3.90332825e-02 5.79430796e-02
4.26265627e-01 -1.18269014e+00 1.03718829e+00 1.42863178e+00
4.99158949e-01 9.73296404e-01 9.08718705e-01 -9.36355114e-01
-1.13486826e+00 -1.03210044e+00 1.35188389e+00 -8.14969018e-02
3.87445450e-01 -7.92201102e-01 -8.86080682e-01 7.22175121e-01
7.51030028e-01 -5.93201160e-01 3.73960942e-01 -1.14441521e-01
-8.89014602e-02 2.14297548e-01 -7.02717304e-01 8.65332782e-01
1.12235439e+00 -3.29060793e-01 -7.65295744e-01 5.63804150e-01
9.48807955e-01 -8.43529761e-01 -1.88818544e-01 9.41572711e-02
3.99054617e-01 -5.78477085e-01 7.42286205e-01 -5.09807587e-01
7.17080951e-01 -3.62198830e-01 1.72646698e-02 -1.27233863e+00
-1.28690884e-01 -1.08614028e+00 -5.32929897e-01 1.46039045e+00
6.19218171e-01 -4.72501189e-01 7.00264692e-01 2.33837381e-01
-2.91936517e-01 -6.91225171e-01 -8.06721687e-01 -5.65917671e-01
2.47264653e-02 -5.31873643e-01 7.78346241e-01 3.37793469e-01
-2.45252058e-01 6.82012558e-01 -7.43230641e-01 -2.38695100e-01
5.21946132e-01 1.62205800e-01 7.97497392e-01 -5.64697146e-01
-5.24342358e-01 -4.20073569e-01 3.45606565e-01 -1.65474999e+00
2.71462619e-01 -9.37640846e-01 3.82899225e-01 -2.05573273e+00
3.52825373e-01 -1.67624667e-01 2.82070100e-01 5.14932811e-01
-7.60800719e-01 -1.78824877e-03 1.51261523e-01 -1.57700285e-01
-2.99524546e-01 7.84080029e-01 1.12639141e+00 -2.38345444e-01
3.48736495e-02 6.58069924e-02 -6.89251482e-01 4.94060814e-01
7.70014107e-01 -4.22940999e-01 -5.76932251e-01 -6.56916559e-01
4.75976646e-01 2.71528331e-03 1.26760364e-01 -7.35746562e-01
1.72500238e-01 -2.03962177e-01 3.36225003e-01 -1.18726206e+00
1.35828450e-01 -1.36445642e-01 2.24554036e-02 3.81134689e-01
-4.82663214e-01 2.26174951e-01 3.34642082e-01 4.90784943e-01
-1.14576817e-01 -7.63388693e-01 3.80229890e-01 -3.45538348e-01
-4.12671626e-01 1.61316216e-01 -4.75574106e-01 4.42371607e-01
4.94778842e-01 -8.01449567e-02 -3.12092692e-01 -4.11677957e-01
-4.08187419e-01 4.77115735e-02 1.04247615e-01 3.62435102e-01
8.74673069e-01 -1.26923013e+00 -1.03835237e+00 1.41115949e-01
-3.20791841e-01 2.75330663e-01 7.49795437e-02 3.23365837e-01
-5.43152452e-01 5.44245660e-01 4.46710795e-01 -3.07854265e-01
-1.05091548e+00 3.79036188e-01 5.79384435e-03 -7.26567328e-01
-4.55607980e-01 8.22673321e-01 9.96770784e-02 -1.77961215e-01
2.43136227e-01 -4.83520061e-01 2.13364631e-01 -3.36837560e-01
8.00731897e-01 2.91946262e-01 4.92749922e-02 -5.38896620e-02
-5.63722523e-03 1.20879009e-01 -3.38900924e-01 -3.97720635e-01
8.59377205e-01 -1.72931209e-01 -2.46140480e-01 5.02804935e-01
6.75035357e-01 1.89498533e-03 -8.93542647e-01 -2.47982979e-01
1.07571483e-04 -3.35436463e-01 -1.48083612e-01 -7.72186577e-01
-5.69041014e-01 1.08119929e+00 6.31257286e-03 -2.24934351e-02
8.45739484e-01 -2.55284011e-01 1.05228746e+00 4.51205611e-01
2.22072959e-01 -9.52396274e-01 1.41748056e-01 7.43293881e-01
9.89726841e-01 -9.88765895e-01 1.82217211e-01 -2.08449900e-01
-5.22357941e-01 9.66652393e-01 5.92387676e-01 8.09314027e-02
1.81138217e-01 4.23918873e-01 -4.13180739e-01 1.88008651e-01
-9.94352818e-01 2.95325760e-02 2.16097251e-01 1.84088752e-01
8.29232574e-01 -1.36923194e-01 -7.49639690e-01 4.41897333e-01
-3.78837645e-01 2.16413602e-01 4.69397098e-01 9.57134962e-01
-5.66394985e-01 -1.32641721e+00 -2.17834666e-01 5.50202549e-01
-4.11367118e-01 -7.72390842e-01 -3.53968114e-01 4.24098253e-01
-5.06035499e-02 8.70609283e-01 5.27313491e-03 1.80305094e-01
-1.16124101e-01 4.01316404e-01 6.59181535e-01 -8.35677087e-01
-3.42695683e-01 2.27702767e-01 1.72945827e-01 -3.73897925e-02
-1.62144646e-01 -6.24310732e-01 -1.25236905e+00 -2.52271086e-01
-4.37616408e-01 1.92275420e-01 4.47430849e-01 1.09349263e+00
4.16892618e-01 5.24480999e-01 1.50170356e-01 -6.54480278e-01
-4.87748206e-01 -1.29199421e+00 -2.81128287e-01 2.89060804e-03
1.19267322e-01 -2.99975008e-01 -1.85543209e-01 1.97962001e-01] | [11.969132423400879, 9.072413444519043] |
22d932a6-010d-4cd2-af8b-17a3121df171 | a-cnn-based-patent-image-retrieval-method-for | 2003.08741 | null | https://arxiv.org/abs/2003.08741v3 | https://arxiv.org/pdf/2003.08741v3.pdf | A Convolutional Neural Network-based Patent Image Retrieval Method for Design Ideation | The patent database is often used in searches of inspirational stimuli for innovative design opportunities because of its large size, extensive variety and rich design information in patent documents. However, most patent mining research only focuses on textual information and ignores visual information. Herein, we propose a convolutional neural network (CNN)-based patent image retrieval method. The core of this approach is a novel neural network architecture named Dual-VGG that is aimed to accomplish two tasks: visual material type prediction and international patent classification (IPC) class label prediction. In turn, the trained neural network provides the deep features in the image embedding vectors that can be utilized for patent image retrieval and visual mapping. The accuracy of both training tasks and patent image embedding space are evaluated to show the performance of our model. This approach is also illustrated in a case study of robot arm design retrieval. Compared to traditional keyword-based searching and Google image searching, the proposed method discovers more useful visual information for engineering design. | ['Jianxi Luo', 'Christopher L. Magee', 'Jie Hu', 'Shuo Jiang', 'Guillermo Ruiz Pava'] | 2020-03-10 | null | null | null | null | ['type-prediction'] | ['computer-code'] | [ 5.19266464e-02 -2.70333529e-01 -8.31869125e-01 1.63888961e-01
-3.71810049e-01 -5.97851694e-01 7.06844687e-01 -3.20417173e-02
-5.32989129e-02 4.58625674e-01 -8.55325907e-02 -7.40641654e-01
-1.84532017e-01 -8.81598294e-01 -6.36928976e-01 -4.47496533e-01
4.59775716e-01 -2.37796288e-02 -2.48076811e-01 1.24368526e-01
8.76236200e-01 7.96297669e-01 -1.56918144e+00 2.30795339e-01
5.76554775e-01 1.55795014e+00 6.58072531e-01 -1.65593736e-02
-6.90154850e-01 7.76317477e-01 -8.69187713e-01 -1.65388986e-01
1.66944042e-01 -1.00397021e-02 -6.69365644e-01 -4.06313658e-01
2.77433276e-01 -7.39930198e-02 -6.34334207e-01 1.03299272e+00
5.65174997e-01 -1.13016896e-01 1.03602338e+00 -1.04994845e+00
-1.61232209e+00 1.47309065e-01 -4.48379666e-01 2.69982815e-01
2.94507653e-01 -2.36831740e-01 9.82376635e-01 -1.03480256e+00
7.27345884e-01 1.16970503e+00 2.21735433e-01 9.46499407e-02
-6.61863983e-01 -8.80714357e-01 -7.82486945e-02 4.60722744e-01
-1.30126417e+00 1.83873281e-01 1.36734235e+00 -7.45658457e-01
8.50785315e-01 5.88413067e-02 9.38268423e-01 1.18231678e+00
6.51399434e-01 8.63678455e-01 8.71608675e-01 -4.24332589e-01
5.04359752e-02 4.24583942e-01 8.25935900e-02 6.86418414e-01
2.96451807e-01 3.01430047e-01 -4.33949232e-01 1.98211044e-01
1.07984829e+00 5.62326789e-01 -5.54426350e-02 -3.28918993e-01
-1.13681328e+00 9.73560154e-01 9.65945959e-01 5.56908369e-01
-5.34231842e-01 2.31909588e-01 3.07868332e-01 3.49679083e-01
3.79353791e-01 9.89945471e-01 -2.16834128e-01 2.64088601e-01
-8.76402259e-01 -9.35204735e-04 5.32978594e-01 1.00487566e+00
7.18718648e-01 2.42722407e-01 -4.46291089e-01 9.22272205e-01
4.41918164e-01 5.62619627e-01 8.59674156e-01 -6.17296040e-01
2.16423944e-01 1.04667759e+00 -2.82356411e-01 -1.44484842e+00
-1.66746840e-01 -7.40544379e-01 -3.88743758e-01 2.88546801e-01
-3.68439078e-01 2.37631574e-01 -1.12500989e+00 1.04645157e+00
-3.14641856e-02 -1.40568152e-01 7.74813071e-02 6.17019057e-01
1.43395519e+00 9.24192965e-01 8.71751532e-02 3.40815932e-01
1.41597068e+00 -9.98415887e-01 -6.57604575e-01 1.04731247e-02
3.44207495e-01 -9.09918606e-01 1.11083317e+00 5.58718145e-02
-5.57233453e-01 -9.34310853e-01 -1.28506744e+00 -6.56002993e-03
-1.07469869e+00 7.37795591e-01 7.73845375e-01 5.31211019e-01
-7.00325310e-01 3.84599805e-01 -3.69924214e-03 -1.38131708e-01
9.29573357e-01 3.10270369e-01 -2.03354597e-01 -1.57713234e-01
-1.18788552e+00 9.28805292e-01 4.86041933e-01 -7.19551072e-02
-7.65522122e-01 -7.45223701e-01 -6.71953857e-01 3.80469501e-01
2.47717828e-01 -4.07879800e-01 7.66783059e-01 -7.58528948e-01
-1.45009077e+00 5.46007335e-01 1.66136667e-01 -1.12517223e-01
-1.81613937e-01 1.18802451e-01 -6.82070255e-01 2.82659888e-01
3.17942500e-01 9.23428059e-01 1.08455729e+00 -1.20712173e+00
-5.18737197e-01 -3.39010596e-01 2.06155684e-02 -1.35539398e-02
-7.54088104e-01 -3.50691676e-01 -7.23250568e-01 -9.76012588e-01
-2.05860808e-01 -5.33498168e-01 9.20881480e-02 -1.08587302e-01
-3.28366965e-01 -5.54053128e-01 1.03602588e+00 -4.92304444e-01
1.27981603e+00 -1.99834275e+00 -4.28077765e-03 4.60442394e-01
2.84057468e-01 2.80074567e-01 -3.96047443e-01 2.77253777e-01
-2.10012615e-01 1.93433493e-01 3.27936351e-01 3.86312962e-01
7.77312443e-02 -1.75040558e-01 -4.62851435e-01 8.96941051e-02
2.07530305e-01 1.68359900e+00 -5.08815110e-01 -4.88323390e-01
4.35873270e-01 4.67547953e-01 -9.43986699e-03 -7.34327659e-02
-3.03241760e-02 2.66332775e-01 -9.98432755e-01 1.16658926e+00
4.08403039e-01 -5.62482238e-01 -2.02307165e-01 -4.13054734e-01
-2.38951519e-01 -5.35165630e-02 -5.09048045e-01 1.73234427e+00
-6.06278598e-01 1.03336883e+00 -6.65739417e-01 -1.16930115e+00
1.13111186e+00 1.32901534e-01 5.08501053e-01 -1.45988846e+00
2.93063104e-01 2.41989970e-01 -5.39562404e-01 -6.17398739e-01
4.23322499e-01 3.79822701e-01 5.16244769e-02 2.04813227e-01
-1.26740396e-01 3.58927660e-02 -2.07073689e-01 -9.39270854e-02
7.79330850e-01 1.51865229e-01 -8.55465382e-02 -3.01534627e-02
5.46433508e-01 2.78147310e-01 4.46247682e-02 6.20549679e-01
2.26659119e-01 6.84576184e-02 3.09590399e-01 -8.10032248e-01
-1.04028273e+00 -6.18507087e-01 -2.75326639e-01 6.71041250e-01
5.21822691e-01 -1.16039805e-01 -4.26033437e-01 -6.68366432e-01
2.42099524e-01 3.48778814e-01 -7.36034811e-01 -3.03211629e-01
-2.18967646e-01 1.01577915e-01 2.90874153e-01 4.37404037e-01
4.46422756e-01 -1.64446306e+00 -5.99952459e-01 -8.82049203e-02
4.61043268e-01 -6.04898155e-01 -2.65727758e-01 -6.11560978e-03
-6.10739946e-01 -1.17100155e+00 -1.25392866e+00 -1.43058348e+00
6.00142121e-01 3.63779336e-01 8.31410527e-01 -4.89353538e-02
-6.76442802e-01 5.36358297e-01 -2.76464909e-01 -6.96234703e-01
-5.28349467e-02 3.25569063e-01 -5.18806100e-01 -2.97577769e-01
5.55934548e-01 -2.98073478e-02 -8.53922725e-01 1.02263361e-01
-9.38929915e-01 -3.38709056e-01 1.16209066e+00 9.40603316e-01
6.88151002e-01 1.56793371e-01 8.17808807e-01 -5.45690298e-01
1.15985954e+00 -5.08640707e-01 -8.92513037e-01 5.52263260e-01
-1.08954561e+00 1.23174928e-01 1.80950925e-01 -5.72152078e-01
-7.25026369e-01 -8.06738064e-02 2.81625926e-01 -8.12703311e-01
7.14608142e-03 1.01504576e+00 -3.16128274e-03 -6.07812107e-01
3.51712406e-01 2.43025556e-01 -1.62494089e-02 -4.96454686e-01
4.66681182e-01 9.64206934e-01 1.86888799e-01 -2.61076331e-01
5.32011092e-01 -5.23306355e-02 1.93787068e-01 -8.64633560e-01
-3.88308257e-01 -4.48196113e-01 -3.18701923e-01 -3.56023669e-01
1.01377881e+00 -8.10716212e-01 -1.16234481e+00 -9.45939645e-02
-1.32764697e+00 3.14632237e-01 -1.33173630e-01 5.84639192e-01
-1.63586125e-01 4.10034269e-01 -2.77777553e-01 -6.81353271e-01
-5.66787422e-01 -1.49264693e+00 1.19003010e+00 2.21475914e-01
7.07278550e-02 -7.64834344e-01 -2.03691032e-02 2.61407852e-01
4.28561598e-01 1.08658157e-01 1.38249028e+00 -4.55546618e-01
-7.63586700e-01 -6.24204397e-01 -5.85817814e-01 2.96851069e-01
1.54949322e-01 -4.00383741e-01 -8.86364162e-01 -8.18961114e-02
-2.40759164e-01 -8.70262310e-02 1.06441641e+00 6.03869975e-01
1.51678240e+00 -1.43058941e-01 -6.61439240e-01 4.58693564e-01
1.63911772e+00 8.94660771e-01 7.85996020e-01 7.37234771e-01
1.06861389e+00 6.92509770e-01 5.82326591e-01 7.23733157e-02
-1.20187305e-01 6.74358726e-01 6.36154652e-01 -3.34589213e-01
-7.71365836e-02 -1.74948558e-01 -1.04237504e-01 5.60377955e-01
-1.68128796e-02 -2.01830521e-01 -7.72677481e-01 4.46911871e-01
-1.54129875e+00 -8.00519764e-01 3.32003683e-01 1.85135341e+00
2.70189106e-01 -1.41979307e-01 -2.50124782e-01 -1.62215471e-01
7.23037183e-01 8.17569569e-02 -7.31449187e-01 -4.44242924e-01
-1.40950397e-01 6.49709940e-01 7.71591008e-01 -2.38934472e-01
-1.22314811e+00 1.04889131e+00 6.44716263e+00 1.19022572e+00
-1.22561955e+00 -3.17890644e-02 4.40512806e-01 1.70113459e-01
-5.74086845e-01 -3.51356149e-01 -3.00061136e-01 3.97155792e-01
4.98889059e-01 1.54881209e-01 4.24013257e-01 1.19449413e+00
-4.53599721e-01 4.20831293e-01 -6.83086872e-01 1.53878021e+00
1.14333689e-01 -2.24141479e+00 5.65575600e-01 4.83846605e-01
6.11148059e-01 -2.67491639e-01 6.37109399e-01 3.74985904e-01
-3.11788380e-01 -1.31389332e+00 6.90569401e-01 4.05362427e-01
1.02578926e+00 -9.64401007e-01 4.89390492e-01 -2.54546255e-01
-1.28513122e+00 -6.43529534e-01 -6.15921319e-01 3.94367278e-01
-3.53092730e-01 1.54722720e-01 -5.71962059e-01 7.45709598e-01
7.15865254e-01 1.23713875e+00 -5.53815603e-01 1.02237320e+00
-1.36127934e-01 1.27900571e-01 3.51055205e-01 -3.85090947e-01
3.95654052e-01 -2.20472813e-01 2.82355219e-01 8.99990201e-01
7.30052590e-01 -5.83363473e-01 -5.88795915e-02 1.24441493e+00
-4.28203523e-01 3.98605168e-01 -1.12504363e+00 -8.80158067e-01
5.37502706e-01 1.19880736e+00 -7.65574336e-01 -1.87426075e-01
-6.10228598e-01 8.21325958e-01 -1.05991453e-01 7.10800529e-01
-7.22017527e-01 -8.45090628e-01 2.93306082e-01 -5.18213362e-02
8.33186746e-01 -4.38726493e-05 -2.91803151e-01 -7.45031893e-01
-9.97701436e-02 -5.97497106e-01 -4.27612150e-03 -9.58517492e-01
-1.15852726e+00 6.04277194e-01 -3.11795205e-01 -1.63900697e+00
1.90182552e-01 -1.30121624e+00 -5.89375973e-01 9.33714449e-01
-1.76688361e+00 -1.42951059e+00 -1.79348961e-01 4.74929899e-01
6.16498888e-01 -9.52609360e-01 6.29767001e-01 5.06797194e-01
-5.13449550e-01 5.62452435e-01 2.54037201e-01 3.05871870e-02
4.61472631e-01 -8.92020881e-01 7.47809932e-02 1.07516140e-01
3.47372144e-01 6.85539365e-01 -9.51338261e-02 -7.81743109e-01
-1.74700761e+00 -9.79059935e-01 3.49532604e-01 -3.11584711e-01
6.53289974e-01 1.36565566e-01 -5.06176174e-01 2.07774401e-01
3.48891795e-01 -4.27131504e-01 7.22957015e-01 -2.50588387e-01
-3.92566830e-01 -4.26549874e-02 -7.41668105e-01 4.15967166e-01
4.95728672e-01 -6.42152965e-01 -4.72547352e-01 3.20708096e-01
1.03630698e+00 3.42898406e-02 -1.11992168e+00 2.64457613e-01
1.04849243e+00 -1.06991529e-01 1.49940836e+00 -6.92452192e-01
7.05639660e-01 -2.94003040e-01 -9.62099060e-02 -8.76455486e-01
-4.86797869e-01 1.59441605e-01 1.00060597e-01 7.71526217e-01
6.16723001e-01 -4.91983950e-01 8.14850926e-01 -9.28120762e-02
-2.53490359e-01 -9.66171980e-01 -7.81194866e-01 -7.35145509e-01
-1.75175637e-01 -1.29203066e-01 8.20345283e-01 1.02303112e+00
-3.31308365e-01 6.15351498e-01 -2.83380568e-01 -2.14145377e-01
2.38054648e-01 6.21131837e-01 4.05150056e-01 -1.68082917e+00
8.68776590e-02 -8.48180532e-01 -6.60679758e-01 -8.39096129e-01
3.48607183e-01 -1.14037192e+00 -4.67974275e-01 -1.85878944e+00
6.53832033e-02 -2.83908606e-01 -8.75962913e-01 4.09213990e-01
3.12114537e-01 1.30071551e-01 2.68177944e-03 6.19573116e-01
-2.69148320e-01 5.48937082e-01 1.66389275e+00 -9.34450328e-01
-1.98266700e-01 -2.18956262e-01 -7.98033893e-01 3.87309603e-02
6.07177138e-01 -2.23640263e-01 -5.12489378e-01 -4.24392462e-01
4.74839032e-01 -2.38552213e-01 4.93916571e-01 -7.86538243e-01
1.55386120e-01 7.29204863e-02 9.74994957e-01 -8.65304768e-01
2.03394756e-01 -1.05456233e+00 1.55131280e-01 5.26177585e-01
-4.24307704e-01 3.05688195e-02 3.16324413e-01 8.65767717e-01
-5.41510820e-01 -2.35155851e-01 1.65551472e-02 -1.40288994e-01
-1.11589837e+00 4.51815933e-01 -2.51141459e-01 -7.10231721e-01
8.61348748e-01 -6.45033419e-01 -4.69933063e-01 -5.56912795e-02
-3.18783551e-01 -1.10801637e-01 1.18682548e-01 8.85674655e-01
9.53799009e-01 -1.82790995e+00 -1.08025838e-02 1.84447467e-01
6.66148543e-01 -5.43870032e-01 6.95702881e-02 3.91099483e-01
-6.37117863e-01 9.99307215e-01 -5.04524589e-01 -4.67736334e-01
-9.04785216e-01 1.18110943e+00 -2.16625094e-01 4.46209982e-02
-4.20973659e-01 6.29653752e-01 4.36817139e-01 -9.23699811e-02
4.60718602e-01 -2.32947648e-01 -1.04119086e+00 2.18007311e-01
2.84190655e-01 1.72127977e-01 -1.03781044e-01 -5.56471705e-01
-2.18614668e-01 1.13567984e+00 1.11318529e-01 2.18466535e-01
1.31375813e+00 2.77481705e-01 -2.67457455e-01 1.68071836e-01
1.55995035e+00 -3.98365766e-01 -5.17908573e-01 -8.28961730e-02
4.83814217e-02 -2.59809583e-01 4.90055382e-01 -7.32191503e-01
-1.37791193e+00 1.08873510e+00 1.15674770e+00 1.02491677e-01
9.35071766e-01 1.41242206e-01 7.87857592e-01 8.11734438e-01
1.43191189e-01 -1.29881310e+00 3.97149950e-01 1.60622403e-01
1.10104883e+00 -1.18996024e+00 -5.74796349e-02 -7.55110383e-02
-4.33806807e-01 1.32361186e+00 3.04324657e-01 9.36959758e-02
6.38175368e-01 -4.88889635e-01 2.85412073e-02 -9.23641086e-01
5.23782969e-02 -4.91023622e-03 9.90091383e-01 4.47817892e-01
2.87430555e-01 -7.41163269e-02 -1.73258126e-01 3.51295114e-01
1.38237506e-01 -8.16297010e-02 -2.78138816e-01 1.16151416e+00
-6.27069294e-01 -1.14369786e+00 -2.73027867e-01 6.05853498e-01
-3.49269152e-01 -2.90880471e-01 -6.09668076e-01 7.26128280e-01
5.32818511e-02 7.22263098e-01 -8.57938230e-02 -7.39223719e-01
3.38172287e-01 -1.59705862e-01 4.29182559e-01 -4.45250571e-01
-2.57395744e-01 1.22565962e-01 -4.13209677e-01 -4.28504199e-01
-4.33419198e-01 2.07814798e-01 -8.53367865e-01 1.93125978e-01
-5.16614437e-01 2.16798931e-01 1.37167048e+00 6.80865705e-01
6.42777503e-01 1.05783153e+00 6.53153002e-01 -7.45556831e-01
4.68526743e-02 -7.66271114e-01 -5.22450089e-01 6.97005913e-02
-4.60381918e-02 -1.08764291e+00 7.51482276e-03 -1.72966078e-01] | [10.830292701721191, 0.5859860181808472] |
581b63ac-bb8b-4e7f-827a-6f036669aa04 | hairmapper-removing-hair-from-portraits-using | null | null | http://openaccess.thecvf.com//content/CVPR2022/html/Wu_HairMapper_Removing_Hair_From_Portraits_Using_GANs_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Wu_HairMapper_Removing_Hair_From_Portraits_Using_GANs_CVPR_2022_paper.pdf | HairMapper: Removing Hair From Portraits Using GANs | Removing hair from portrait images is challenging due to the complex occlusions between hair and face, as well as the lack of paired portrait data with/without hair. To this end, we present a dataset and a baseline method for removing hair from portrait images using generative adversarial networks (GANs). Our core idea is to train a fully connected network HairMapper to find the direction of hair removal in the latent space of StyleGAN for the training stage. We develop a new separation boundary and diffuse method to generate paired training data for males, and a novel "female-male-bald" pipeline for paired data of females. Experiments show that our method can naturally deal with portrait images with variations on gender, age, etc. We validate the superior performance of our method by comparing it to state-of-the-art methods through extensive experiments and user studies. We also demonstrate its applications in hair design and 3D face reconstruction. | ['Xiaogang Jin', 'Yong-Liang Yang', 'Yiqian Wu'] | 2022-01-01 | null | null | null | cvpr-2022-1 | ['3d-face-reconstruction', 'face-reconstruction'] | ['computer-vision', 'computer-vision'] | [ 4.65677172e-01 4.94653046e-01 1.17150076e-01 -3.61489922e-01
-5.24925828e-01 -5.89607060e-01 5.58821917e-01 -6.24451518e-01
3.10251623e-01 5.50983310e-01 1.30065903e-01 9.26911905e-02
5.31574845e-01 -8.66601467e-01 -7.36509681e-01 -6.75353408e-01
4.67823058e-01 4.01891768e-01 -4.04879957e-01 -4.31557834e-01
-2.30216250e-01 5.35097420e-01 -1.27769327e+00 1.42761886e-01
6.10938966e-01 8.06004226e-01 -5.39554477e-01 4.03960496e-01
-2.81711705e-02 1.74014181e-01 -7.67111480e-01 -9.12779391e-01
4.38548356e-01 -7.74715066e-01 -4.59808499e-01 6.19781435e-01
9.91912305e-01 -4.57384616e-01 -2.27116466e-01 7.01939464e-01
8.19591761e-01 -6.36673927e-01 7.64865160e-01 -1.44372094e+00
-8.26275229e-01 5.42241752e-01 -1.22133744e+00 -8.50098252e-01
5.31069696e-01 2.29739279e-01 4.27388310e-01 -1.01791346e+00
9.10967231e-01 1.55004382e+00 1.03934300e+00 1.23876607e+00
-1.52333570e+00 -1.19111991e+00 1.17480187e-02 -5.20155370e-01
-1.57526362e+00 -6.67294979e-01 1.07805896e+00 -3.74727964e-01
-1.89062916e-02 4.06457692e-01 1.04836988e+00 1.68140149e+00
1.51172746e-02 9.74498451e-01 1.20051980e+00 -4.76787716e-01
-8.63939077e-02 3.47604156e-02 -7.11238205e-01 8.30601752e-01
-2.40232721e-02 -7.84007683e-02 -7.05467701e-01 -8.45902711e-02
1.11453640e+00 -2.05008939e-01 7.26538301e-02 -5.00505686e-01
-6.28514528e-01 8.27115119e-01 4.02557224e-01 -1.06658317e-01
2.86226999e-02 3.00768375e-01 -2.84700585e-03 -2.73160581e-02
6.77200854e-01 -4.80764732e-02 1.55089051e-01 7.10630178e-01
-1.25018442e+00 5.24551213e-01 8.00883949e-01 1.22516096e+00
6.17555261e-01 2.43306756e-01 -4.06616062e-01 9.91740823e-01
2.69245178e-01 8.41363668e-01 -1.99234962e-01 -6.80132151e-01
1.90039814e-01 3.96049768e-01 -2.33915508e-01 -8.91541898e-01
-1.28513902e-01 -2.02686355e-01 -1.08108974e+00 4.45837915e-01
4.76823628e-01 -2.57515430e-01 -1.34275424e+00 1.87023771e+00
5.89012861e-01 -2.06111327e-01 -4.33497846e-01 7.37498581e-01
1.03462005e+00 9.04665738e-02 -2.87321378e-02 -8.39928910e-03
1.17610228e+00 -7.61663437e-01 -8.30102146e-01 -4.45457280e-01
-6.48793802e-02 -1.03235066e+00 1.17948174e+00 1.90087065e-01
-1.39420164e+00 -4.79874730e-01 -8.64461064e-01 -3.33218873e-01
-1.37039423e-01 1.85545757e-01 6.75753295e-01 8.19123387e-01
-1.00664747e+00 5.68280876e-01 -3.90142828e-01 -5.07692456e-01
9.11965251e-01 5.06732881e-01 -3.51247281e-01 -1.49152294e-01
-8.29520106e-01 5.01588821e-01 -2.41819978e-01 5.33738993e-02
-1.06136203e+00 -8.95178378e-01 -9.07465339e-01 -4.91123796e-01
9.92195755e-02 -1.06397092e+00 1.00365806e+00 -1.19954884e+00
-1.60751176e+00 1.43549931e+00 3.29631194e-02 -9.91018489e-02
8.79536033e-01 -1.99501231e-01 -6.32759631e-02 -7.72788227e-02
7.44647384e-02 1.23758769e+00 1.50894511e+00 -1.50399578e+00
7.91990664e-03 -6.47337198e-01 -4.13509935e-01 -5.19161858e-02
-4.29852724e-01 -2.20211610e-01 -6.83707714e-01 -9.36331809e-01
-1.03901222e-01 -1.20899904e+00 1.65546462e-01 4.58070487e-01
-8.97979677e-01 1.19172364e-01 1.05320728e+00 -9.02334869e-01
9.06526208e-01 -2.02949882e+00 3.28240395e-01 3.12482297e-01
3.64249527e-01 -1.01185583e-01 -1.92328379e-01 4.18486774e-01
-2.49427855e-02 4.02778715e-01 -3.71483028e-01 -9.46705878e-01
-1.10782515e-02 2.20387146e-01 -1.21412188e-01 4.81504649e-01
3.53251487e-01 1.04720116e+00 -4.89562511e-01 -6.63501084e-01
-1.11132190e-01 9.59363818e-01 -4.46664959e-01 2.60962874e-01
-2.20985726e-01 8.81376445e-01 5.67472614e-02 1.16131449e+00
9.99748945e-01 2.37687200e-01 2.68543661e-01 -2.52296418e-01
3.14113289e-01 -3.60973984e-01 -8.74641359e-01 2.04955435e+00
-3.86509389e-01 3.83633375e-01 5.20518959e-01 1.89730123e-01
1.02982533e+00 1.94786310e-01 2.15423375e-01 -5.86727560e-01
3.22837591e-01 4.95904647e-02 -4.34251279e-01 -3.53809595e-01
1.02804922e-01 -4.45473850e-01 -1.01538159e-01 2.97943860e-01
1.41119465e-01 -5.01062989e-01 1.48163810e-01 3.49445604e-02
7.39151478e-01 5.61698735e-01 -2.00354874e-01 -2.66702831e-01
1.48424074e-01 -6.59565747e-01 4.99666005e-01 1.76095873e-01
1.45195022e-01 1.13760567e+00 5.26505649e-01 -5.99505842e-01
-1.34099305e+00 -1.05718529e+00 -7.93715566e-02 1.02487588e+00
-1.21500209e-01 -4.34186518e-01 -1.39097190e+00 -8.41294646e-01
1.50941119e-01 4.12788242e-01 -1.25372684e+00 -1.87760852e-02
-5.07747948e-01 -6.28258705e-01 6.00587010e-01 4.04966801e-01
6.11283123e-01 -1.01409698e+00 -3.37335132e-02 -2.58069217e-01
-3.59965228e-02 -8.22578728e-01 -7.73763418e-01 -3.61634940e-01
-4.05490160e-01 -9.68317747e-01 -1.12579620e+00 -8.04291248e-01
9.66011643e-01 -2.49620408e-01 1.31674612e+00 1.78885296e-01
-6.10088468e-01 1.65209278e-01 5.98642863e-02 -7.42706001e-01
-5.14873266e-01 3.90224040e-01 -1.53004557e-01 3.09639782e-01
3.93157676e-02 -8.47158968e-01 -9.25520360e-01 3.55386257e-01
-9.62299347e-01 4.69173372e-01 5.31261563e-01 6.91285968e-01
4.56449687e-01 -2.93384254e-01 3.35965157e-01 -1.29473245e+00
6.44255817e-01 -1.37873024e-01 -2.38845751e-01 1.35571554e-01
-2.59701699e-01 -2.61861235e-01 3.62978399e-01 -5.79339981e-01
-9.89081502e-01 1.69789702e-01 -3.29512388e-01 -5.11307895e-01
-1.42486036e-01 -3.06816846e-01 -6.06321692e-01 -2.99803585e-01
7.35903680e-01 -1.69217974e-01 3.24865669e-01 -4.32252526e-01
5.03649235e-01 2.80808359e-01 5.95307946e-01 -6.27678752e-01
1.19925380e+00 6.68429375e-01 1.23960860e-01 -6.78380370e-01
-9.25063372e-01 2.81282485e-01 -8.61468315e-01 -2.57036924e-01
8.36441398e-01 -7.33064890e-01 -4.83923674e-01 6.66400790e-01
-9.82312024e-01 -5.27817309e-01 -3.64671916e-01 -4.80913103e-01
-4.79199946e-01 6.25148714e-02 -8.78996670e-01 -7.79353440e-01
-5.92396140e-01 -9.49938178e-01 1.56852186e+00 2.32017398e-01
-4.77300733e-01 -8.74148786e-01 1.28857475e-02 5.22844970e-01
3.30036730e-01 9.39690769e-01 8.40235054e-01 2.03072801e-01
-2.20584303e-01 -1.98398143e-01 -6.39826655e-02 2.19595581e-01
1.34458885e-01 2.05733791e-01 -1.32589459e+00 -5.22218347e-01
-3.38046193e-01 -5.02292097e-01 7.71328628e-01 2.54254818e-01
1.22123504e+00 -4.86970007e-01 -3.91777992e-01 1.06781626e+00
1.19125533e+00 -2.54424810e-01 9.47166622e-01 -2.09818006e-01
1.15745187e+00 9.03427720e-01 1.73581079e-01 3.55781972e-01
1.40172049e-01 4.94206220e-01 4.64694649e-01 -7.68221319e-01
-7.28371739e-01 -7.97867954e-01 1.56037644e-01 4.68581587e-01
-1.26084298e-01 -3.90291177e-02 -5.52052855e-01 3.37024361e-01
-1.34446073e+00 -5.23847699e-01 8.22716951e-02 2.04706240e+00
1.06641662e+00 -1.31874278e-01 5.29538572e-01 2.26322815e-01
7.00797439e-01 -8.00208822e-02 -6.74753666e-01 -2.44352847e-01
-2.07470655e-01 5.62960148e-01 3.59381825e-01 3.30336660e-01
-9.91660893e-01 1.12870872e+00 6.92489910e+00 9.54689622e-01
-1.17295563e+00 -4.70797010e-02 9.57214713e-01 -1.48050919e-01
-5.60940802e-01 -1.51902586e-01 -8.07518423e-01 2.84034491e-01
1.21085413e-01 4.59535748e-01 6.17986262e-01 5.76922715e-01
-3.52237761e-01 3.20010602e-01 -1.12016129e+00 1.16289473e+00
4.82579798e-01 -9.84258711e-01 2.13102117e-01 3.03198934e-01
9.35204983e-01 -7.47795999e-01 4.26661611e-01 -4.54980731e-02
7.09911212e-02 -1.44818306e+00 9.52344000e-01 4.05273438e-01
1.48852491e+00 -8.91873598e-01 1.94518998e-01 -4.20826413e-02
-8.90310705e-01 2.08857402e-01 -6.87600859e-03 2.77886003e-01
-3.25478613e-02 5.51572442e-01 -9.76748407e-01 3.02930236e-01
4.66771185e-01 4.98964071e-01 -8.53789985e-01 4.65031296e-01
-6.57893598e-01 3.42780411e-01 -3.49346876e-01 2.94237703e-01
-3.37517947e-01 -1.90890893e-01 2.63862401e-01 1.03357899e+00
3.33956420e-01 -2.27550074e-01 -3.38592380e-02 1.18191624e+00
-5.29759228e-01 2.01005951e-01 -9.26582754e-01 5.76511137e-02
3.33944112e-01 1.53015196e+00 -7.66457438e-01 3.88952307e-02
6.16668575e-02 1.21572137e+00 1.30320475e-01 3.78339887e-01
-7.00147629e-01 -6.91572502e-02 2.37072662e-01 8.43926609e-01
1.39273122e-01 5.29515557e-02 -5.28990924e-01 -9.08105135e-01
2.77381167e-02 -1.10019755e+00 6.72545880e-02 -8.70579541e-01
-1.49200654e+00 6.64116502e-01 -1.58241361e-01 -7.50840545e-01
-4.56285954e-04 -2.54903227e-01 -7.02289402e-01 8.36725771e-01
-1.03832245e+00 -2.43583751e+00 -6.16136789e-01 4.61663067e-01
4.29249525e-01 -3.57277058e-02 1.10302150e+00 1.88434944e-01
-6.26173079e-01 1.06585062e+00 -5.94429195e-01 1.73619270e-01
9.30335283e-01 -1.11945343e+00 7.23656893e-01 5.11212289e-01
-1.98641419e-02 5.51662624e-01 7.10621655e-01 -8.73338938e-01
-1.53577495e+00 -1.04272628e+00 4.49777663e-01 -5.91482401e-01
4.98699360e-02 -1.08548808e+00 -4.59319055e-01 7.84323931e-01
7.17254102e-01 -2.62950808e-01 9.52424824e-01 2.36973628e-01
-6.51018739e-01 -2.48211578e-01 -1.44451344e+00 8.09432149e-01
1.36878252e+00 -2.72945076e-01 -3.12520638e-02 3.07116568e-01
3.35900694e-01 -5.48604548e-01 -9.37238276e-01 5.36959410e-01
9.58029985e-01 -9.56525147e-01 9.67473388e-01 -1.24080569e-01
6.72702253e-01 -1.35311216e-01 2.80062288e-01 -1.27058709e+00
-1.93330064e-01 -1.10076976e+00 -8.39097332e-03 1.77903247e+00
2.86802351e-01 6.35583373e-03 1.02291000e+00 5.80590487e-01
2.40263671e-01 -7.93089092e-01 -4.12008613e-01 -2.23312452e-01
4.04587358e-01 8.15438628e-02 8.11983049e-01 7.93819904e-01
-7.01610744e-01 6.68842375e-01 -8.95180106e-01 -3.52690727e-01
1.10839009e+00 2.47037977e-01 1.32297790e+00 -1.17390323e+00
-1.92227632e-01 -1.97036654e-01 -8.62081796e-02 -6.63085103e-01
1.77427992e-01 -6.82803750e-01 -1.49316788e-01 -1.25829554e+00
3.03584725e-01 -3.71189415e-01 4.48759824e-01 8.23527098e-01
-6.93487823e-02 1.09408426e+00 2.99081594e-01 -4.13296968e-02
-5.50804026e-02 5.21834970e-01 1.67961657e+00 -3.01285863e-01
-1.30450964e-01 -6.67950278e-03 -9.21297848e-01 7.82334685e-01
6.74475491e-01 -3.67591560e-01 -3.97982031e-01 -5.13492703e-01
3.74809921e-01 -1.77884057e-01 4.21961635e-01 -8.91461194e-01
-1.81578442e-01 1.39705464e-01 1.09003150e+00 -6.48232162e-01
5.79984546e-01 -6.15203857e-01 4.20554578e-01 1.48414016e-01
-2.05958426e-01 -2.26871029e-01 2.96593070e-01 2.44247630e-01
3.22753608e-01 2.34268650e-01 9.65105712e-01 -2.86153913e-01
1.62081942e-01 5.69764853e-01 6.90184534e-02 3.98303894e-03
9.27551866e-01 -1.74355388e-01 8.18032995e-02 -5.66700995e-01
-8.30884457e-01 6.86977357e-02 9.96302545e-01 3.38403761e-01
6.12632692e-01 -1.73347282e+00 -9.16658998e-01 7.62458086e-01
5.08765317e-02 1.21623546e-01 1.13925040e-01 5.69367588e-01
-4.90842253e-01 -4.26128238e-01 -5.16588807e-01 -5.37325680e-01
-1.58445251e+00 5.67699671e-01 1.41629219e-01 3.19900252e-02
-4.81692463e-01 9.25360799e-01 5.23204386e-01 -5.30847073e-01
3.35808694e-01 1.25597283e-01 1.89991131e-01 -3.17016728e-02
4.66995001e-01 4.82345149e-02 -1.76774159e-01 -5.97916842e-01
1.19640594e-02 8.98033321e-01 5.45265600e-02 -2.99567990e-02
1.21725273e+00 -8.37139487e-02 -2.94480443e-01 1.98248133e-01
1.06038892e+00 3.47806811e-01 -1.33894837e+00 2.59674937e-01
-7.96144724e-01 -6.56159580e-01 -4.33999449e-01 -7.22483933e-01
-1.46416962e+00 9.92128491e-01 5.81799209e-01 9.97666717e-02
1.09547710e+00 -5.66214174e-02 1.08079231e+00 -4.06423032e-01
2.89287478e-01 -1.00439835e+00 1.21158212e-01 -5.40916920e-02
1.33054686e+00 -9.71670866e-01 -1.92840770e-02 -8.89266610e-01
-4.22865182e-01 9.65234280e-01 7.07909286e-01 -1.36216804e-01
4.45687294e-01 5.21566927e-01 3.18298280e-01 -2.96799123e-01
-1.85202450e-01 4.92069609e-02 3.79835993e-01 7.73210704e-01
5.49848139e-01 -1.45776197e-01 -7.20225424e-02 3.06432009e-01
-6.20611906e-01 -4.45280597e-02 -7.40990117e-02 8.22104156e-01
3.45902532e-01 -1.57259107e+00 -5.74537456e-01 1.69339120e-01
-4.98676121e-01 1.50261581e-01 -1.09109998e+00 8.94205153e-01
6.33711517e-01 5.67517459e-01 -1.34959430e-01 -4.38455105e-01
2.92647034e-01 1.81350112e-01 1.08186877e+00 -5.05247712e-01
-7.48846829e-01 7.02949464e-01 -4.58615199e-02 -2.62467057e-01
-2.79644161e-01 -3.49660039e-01 -5.49286962e-01 -6.04739428e-01
-7.90287107e-02 -4.10149753e-01 6.69892967e-01 5.13497472e-01
3.39800209e-01 4.39353645e-01 6.18935883e-01 -1.25698817e+00
-4.27097976e-02 -1.10303700e+00 -6.93282306e-01 9.71840560e-01
2.38504745e-02 -5.34069896e-01 -8.64585191e-02 3.69912773e-01] | [12.429119110107422, -0.2864268720149994] |
11df8cf1-a766-4a4a-9e74-ca819438c891 | on-learning-semantic-representations-for | 2007.04101 | null | https://arxiv.org/abs/2007.04101v1 | https://arxiv.org/pdf/2007.04101v1.pdf | On Learning Semantic Representations for Million-Scale Free-Hand Sketches | In this paper, we study learning semantic representations for million-scale free-hand sketches. This is highly challenging due to the domain-unique traits of sketches, e.g., diverse, sparse, abstract, noisy. We propose a dual-branch CNNRNN network architecture to represent sketches, which simultaneously encodes both the static and temporal patterns of sketch strokes. Based on this architecture, we further explore learning the sketch-oriented semantic representations in two challenging yet practical settings, i.e., hashing retrieval and zero-shot recognition on million-scale sketches. Specifically, we use our dual-branch architecture as a universal representation framework to design two sketch-specific deep models: (i) We propose a deep hashing model for sketch retrieval, where a novel hashing loss is specifically designed to accommodate both the abstract and messy traits of sketches. (ii) We propose a deep embedding model for sketch zero-shot recognition, via collecting a large-scale edge-map dataset and proposing to extract a set of semantic vectors from edge-maps as the semantic knowledge for sketch zero-shot domain alignment. Both deep models are evaluated by comprehensive experiments on million-scale sketches and outperform the state-of-the-art competitors. | ['Yi-Zhe Song', 'Tongtong Yuan', 'Timothy M. Hospedales', 'Tao Xiang', 'Peng Xu', 'Liang Wang', 'Yongye Huang'] | 2020-07-07 | null | null | null | null | ['learning-semantic-representations'] | ['methodology'] | [-2.56002605e-01 -5.93981981e-01 -4.28198308e-01 -4.51634496e-01
-7.99069762e-01 -4.78513032e-01 5.51396668e-01 -3.89001906e-01
2.24109907e-02 2.41883233e-01 4.25836027e-01 2.05746740e-01
-5.41233346e-02 -9.69111025e-01 -6.69954062e-01 -3.87300879e-01
2.22598672e-01 5.25236726e-01 -1.56740902e-03 -2.52702773e-01
4.18640882e-01 6.47845507e-01 -1.36275816e+00 1.54277697e-01
2.50322789e-01 1.25188613e+00 1.55175820e-01 5.90973377e-01
-5.70454299e-01 2.92769253e-01 -3.68852764e-01 -5.92645705e-01
3.17077100e-01 -2.04253614e-01 -2.19352156e-01 -3.62032168e-02
9.28434908e-01 -8.70071709e-01 -1.15546155e+00 8.27954650e-01
7.79159188e-01 2.71475539e-02 8.54411960e-01 -1.64441454e+00
-1.27635622e+00 3.14211458e-01 -4.28006351e-01 -2.71580875e-01
3.48336726e-01 4.32236314e-01 1.34220648e+00 -1.26074386e+00
8.10223818e-01 1.47097611e+00 7.11461961e-01 7.54501283e-01
-1.11439753e+00 -1.09135151e+00 1.33199662e-01 -6.34928280e-03
-1.70423830e+00 -4.01654124e-01 1.20492721e+00 -2.41157055e-01
7.11888492e-01 -2.18998298e-01 7.15785801e-01 1.75211704e+00
-2.40768403e-01 1.17856252e+00 3.97512376e-01 1.97762504e-01
2.64028788e-01 -2.95220432e-03 -1.43117979e-01 8.26844215e-01
1.09022014e-01 -5.08493148e-02 -6.70642316e-01 -4.75879461e-01
1.39052415e+00 6.45124376e-01 1.36071876e-01 -5.62889874e-01
-1.14429104e+00 9.33391750e-01 3.86408687e-01 2.39972770e-01
-3.32865477e-01 6.37596011e-01 6.47434831e-01 4.01276857e-01
7.41116032e-02 1.62123978e-01 -1.77372634e-01 -8.27623904e-02
-1.07178140e+00 6.72118783e-01 7.74153292e-01 1.49188983e+00
7.98080266e-01 6.72953725e-02 -5.14334321e-01 1.21689296e+00
2.74141222e-01 7.46615231e-01 3.45905662e-01 -5.94304860e-01
6.25912726e-01 5.72978139e-01 -1.32589325e-01 -1.31573176e+00
2.36639693e-01 7.68357813e-02 -1.17333162e+00 -3.52665454e-01
-2.52323672e-02 3.97608757e-01 -8.77477765e-01 1.87778521e+00
-1.71411172e-01 4.66104656e-01 -1.08552068e-01 1.08510649e+00
1.17221713e+00 7.49456108e-01 1.22306935e-01 5.70976675e-01
1.39634776e+00 -1.02202392e+00 -5.05886614e-01 8.94851238e-02
-3.61395516e-02 -5.40849090e-01 1.45604432e+00 4.86919982e-03
-1.01391602e+00 -7.54623950e-01 -1.19240749e+00 -4.71363902e-01
-6.57761276e-01 3.07297140e-01 5.00654101e-01 2.68568069e-01
-6.71984673e-01 7.09279656e-01 -2.91166872e-01 -5.19146860e-01
7.60342538e-01 4.96267080e-02 -4.00866717e-01 -3.53782117e-01
-1.38346291e+00 2.34556332e-01 6.65960610e-02 -1.33320212e-01
-1.07621467e+00 -8.02652359e-01 -1.01816380e+00 5.36562979e-01
1.66311577e-01 -7.40111351e-01 8.07771444e-01 -3.99512827e-01
-1.38646424e+00 7.97955513e-01 -5.24055995e-02 1.76695243e-01
4.00825411e-01 9.99139249e-02 -3.53381604e-01 2.01853827e-01
1.72300622e-01 8.27336133e-01 1.23106122e+00 -1.20709658e+00
-1.83148637e-01 -3.30493450e-01 -1.22253798e-01 -1.66810766e-01
-7.25550592e-01 -3.14413100e-01 -8.38905990e-01 -1.13884807e+00
-1.21090412e-01 -6.66749835e-01 4.23300751e-02 6.99199080e-01
-3.15818131e-01 -4.53412682e-01 1.01008296e+00 -4.68094468e-01
1.27722061e+00 -2.11940837e+00 2.50833660e-01 2.26573363e-01
2.75536686e-01 1.22007705e-01 -7.36711085e-01 8.11378837e-01
2.18111008e-01 9.39407796e-02 -2.87752926e-01 -5.00829041e-01
7.04281747e-01 3.85127604e-01 -9.83154774e-01 -2.68843770e-02
5.14038444e-01 1.46504855e+00 -9.49231803e-01 -4.88644570e-01
1.47067934e-01 4.16684330e-01 -3.78928632e-01 6.92663193e-01
-1.80006668e-01 -2.76870787e-01 -5.68412125e-01 1.10503495e+00
8.07603598e-01 -3.14092964e-01 2.21115351e-02 -3.92043710e-01
3.59189749e-01 4.89492388e-03 -1.23713481e+00 2.36905456e+00
-4.80489939e-01 4.11774397e-01 -9.93969887e-02 -8.21101248e-01
1.20690334e+00 2.74723023e-01 2.46691868e-01 -8.44155908e-01
-2.45434999e-01 2.61326730e-01 -8.02142441e-01 -1.65124118e-01
5.56383431e-01 -2.81362265e-01 -4.78635639e-01 6.87828302e-01
4.92553532e-01 -2.35604957e-01 -2.12092832e-01 4.94345963e-01
1.08850873e+00 -4.04722355e-02 8.51267204e-02 -9.82165858e-02
3.72712433e-01 -6.82175457e-01 3.48313481e-01 8.18887830e-01
-1.34498447e-01 9.16312695e-01 6.21184826e-01 -8.29249799e-01
-1.43828845e+00 -1.35405004e+00 1.84363768e-01 1.02730346e+00
4.60372448e-01 -6.64859653e-01 -2.32431099e-01 -7.14676559e-01
7.30554521e-01 -8.43923539e-04 -6.35627210e-01 -3.94389570e-01
-5.64055562e-01 7.86650702e-02 7.55190134e-01 8.05993259e-01
3.50117385e-01 -1.27803636e+00 -9.31449905e-02 2.27668986e-01
1.96251243e-01 -1.10447550e+00 -9.97870624e-01 -2.67460972e-01
-5.22075355e-01 -7.81698823e-01 -1.30597889e+00 -8.30142379e-01
3.68991464e-01 5.42461634e-01 1.26530385e+00 3.12152147e-01
-5.94623566e-01 8.05810392e-01 -1.44403994e-01 2.54359525e-02
2.43880883e-01 2.63543576e-01 2.30835918e-02 8.79109502e-02
3.82492453e-01 -9.39262271e-01 -8.96711648e-01 3.60756963e-01
-1.05558324e+00 -2.68959552e-01 8.20501864e-01 1.13393569e+00
4.91689175e-01 -3.93593103e-01 7.07632244e-01 -3.68978083e-01
9.41371024e-01 -6.25724256e-01 -2.05778033e-01 5.40559530e-01
-2.54176557e-01 2.32046902e-01 9.12859499e-01 -7.21167564e-01
-5.43536305e-01 -1.38189048e-01 -8.15711468e-02 -1.29074931e+00
-5.77951558e-02 -1.19383551e-01 -3.96867245e-01 -1.55980036e-01
1.13734454e-01 6.01041615e-01 -1.23567045e-01 -4.96388972e-01
6.25846267e-01 6.54169381e-01 5.79552114e-01 -1.18406975e+00
9.98671055e-01 3.83987546e-01 -2.34464332e-01 -9.07730997e-01
-4.58376348e-01 -5.26930869e-01 -3.22772473e-01 7.45085403e-02
6.11931860e-01 -9.09195423e-01 -7.29972482e-01 4.60921526e-01
-1.35932100e+00 -1.67492494e-01 -2.82005847e-01 -1.52053893e-01
-7.53156483e-01 5.94193518e-01 -7.15076685e-01 -6.62458122e-01
-7.57675707e-01 -1.00650549e+00 1.82279611e+00 1.73494369e-01
-9.59533900e-02 -7.45336115e-01 2.37730816e-01 -1.70828670e-01
4.80542392e-01 -1.20089158e-01 1.03581393e+00 -3.27584296e-01
-8.62148941e-01 -2.17674017e-01 -8.59574437e-01 1.21257439e-01
-2.73346435e-02 -1.24716081e-01 -8.46780419e-01 -3.72586846e-01
-6.12490714e-01 -9.04367089e-01 1.05389524e+00 -1.43562466e-01
1.60608566e+00 -2.85359979e-01 -1.84934646e-01 9.09439325e-01
1.59522200e+00 -2.13839084e-01 7.50999093e-01 -2.66607106e-01
8.51360738e-01 1.67152315e-01 4.47323650e-01 7.88852036e-01
4.36801612e-01 7.78621793e-01 7.70644993e-02 6.98134229e-02
-9.97542217e-02 -1.08926010e+00 -9.36631188e-02 8.68170679e-01
3.75919759e-01 -1.22605890e-01 -5.73302209e-01 7.69066632e-01
-1.99044228e+00 -1.02468622e+00 8.56425226e-01 1.84129488e+00
6.01035476e-01 -5.01415133e-02 2.20696911e-01 -2.95471847e-01
6.68033302e-01 7.49910355e-01 -7.83319592e-01 -1.55947119e-01
-6.36552926e-03 3.70566577e-01 1.66742668e-01 -9.00218636e-02
-1.00197494e+00 1.25709629e+00 5.33447218e+00 1.21730173e+00
-1.08791959e+00 -1.43967211e-01 2.53047854e-01 -9.87704750e-03
-6.96292758e-01 -2.22262725e-01 -7.20449805e-01 9.12538767e-01
3.38950366e-01 -1.45577257e-02 7.06981897e-01 9.65626001e-01
-6.59268558e-01 7.05431104e-01 -1.19434655e+00 1.62181699e+00
2.17709750e-01 -1.63795233e+00 7.67713547e-01 -6.80164322e-02
5.25680602e-01 -4.59722400e-01 7.81637132e-02 7.05850363e-01
1.10560119e-01 -1.03989089e+00 6.26228154e-01 6.12939775e-01
1.45418692e+00 -6.25616312e-01 2.36825630e-01 -5.00913486e-02
-1.84303927e+00 -1.04808718e-01 -7.55183518e-01 2.99401015e-01
3.65224555e-02 5.84138818e-02 -1.08743973e-01 4.24845278e-01
4.77712750e-01 1.15093410e+00 -1.51507229e-01 5.97799599e-01
8.29844624e-02 2.87606400e-02 -1.86816648e-01 -1.42313868e-01
5.44927061e-01 -2.10483506e-01 2.81663001e-01 1.36190784e+00
3.86919200e-01 2.02753395e-01 1.48649797e-01 1.52499616e+00
-5.79491794e-01 -1.24508426e-01 -1.06914210e+00 -6.42383099e-01
1.00589800e+00 1.34427452e+00 -3.62844467e-01 -5.65815210e-01
-3.97520036e-01 1.39040112e+00 6.23280227e-01 4.32128489e-01
-6.49633944e-01 -7.71385610e-01 1.21547031e+00 -1.87383205e-01
6.82590663e-01 -2.17132598e-01 -1.24882281e-01 -1.42964888e+00
2.98870534e-01 -5.92080474e-01 2.53139257e-01 -6.36542559e-01
-2.09706879e+00 2.30612010e-01 -4.45372820e-01 -1.19517255e+00
1.93723723e-01 -6.87196255e-01 -1.10728657e+00 8.80673587e-01
-1.55481589e+00 -1.62116373e+00 -5.64071417e-01 7.21624494e-01
7.15748310e-01 -5.65290749e-01 8.48036647e-01 6.00426018e-01
-3.77787620e-01 1.09664690e+00 -8.17422792e-02 6.16161466e-01
9.01967943e-01 -1.16757643e+00 1.18197572e+00 1.25617012e-01
2.74675846e-01 1.03238833e+00 -1.33678958e-01 -6.24024630e-01
-2.17963815e+00 -1.01785362e+00 5.30687630e-01 -2.90240258e-01
8.04676592e-01 -8.44247997e-01 -9.96900260e-01 3.55765313e-01
-4.10581797e-01 3.32553893e-01 5.52054822e-01 -2.23321584e-03
-1.08230579e+00 -2.66538888e-01 -9.78639543e-01 6.19869173e-01
1.47130919e+00 -1.24439991e+00 -6.16876423e-01 -1.66910246e-03
6.64736569e-01 -5.25536425e-02 -1.05820262e+00 5.37975840e-02
1.34607470e+00 -5.49456358e-01 1.50482559e+00 -9.24255610e-01
7.90118456e-01 5.44526428e-02 -3.92580241e-01 -9.20951486e-01
-4.81332570e-01 -4.53448564e-01 -3.63103062e-01 1.26735985e+00
-1.38214022e-01 -2.35883236e-01 9.97026801e-01 3.78500104e-01
3.02097738e-01 -9.26821232e-01 -7.41052687e-01 -8.96097660e-01
-2.52159480e-02 -1.73643187e-01 1.00389647e+00 1.01014316e+00
-3.18832755e-01 4.08232093e-01 -8.56138945e-01 -2.18150690e-01
8.10529649e-01 6.28829539e-01 1.22098076e+00 -1.26965725e+00
-4.83892225e-02 -7.45579064e-01 -6.44554794e-01 -1.50108266e+00
5.15404224e-01 -7.95522392e-01 -2.40133643e-01 -1.34069836e+00
4.24436837e-01 -6.28283978e-01 -4.94235963e-01 2.96885610e-01
-1.36289284e-01 1.67309120e-01 4.81991172e-01 3.03696245e-01
-8.09190750e-01 1.15784705e+00 1.13904297e+00 -6.25443518e-01
2.67029852e-01 -4.40150380e-01 -5.57272553e-01 9.59140956e-02
2.64815480e-01 -2.87908435e-01 -3.72055590e-01 -4.76635009e-01
1.03548594e-01 2.11579636e-01 6.77094102e-01 -7.11262345e-01
4.50972736e-01 -1.05202399e-01 4.84269559e-01 -7.45846689e-01
5.94460964e-01 -9.61482465e-01 -1.90689549e-01 1.27478167e-01
-4.38095450e-01 -1.32309511e-01 3.23327631e-02 9.81482446e-01
-2.52110124e-01 9.58315283e-02 6.21144116e-01 -3.58111948e-01
-8.65118563e-01 1.03771746e+00 4.06145066e-01 8.29822421e-02
5.87411046e-01 -2.25304887e-01 -1.58416286e-01 -4.40643400e-01
-1.89033061e-01 3.30210745e-01 5.93149066e-01 8.54239345e-01
1.13645077e+00 -2.13160443e+00 -4.83827382e-01 4.63225871e-01
5.98557055e-01 -2.41791889e-01 5.28579772e-01 8.10057446e-02
-2.03826427e-01 4.97222930e-01 -7.12270617e-01 -3.36355388e-01
-7.38258839e-01 7.93893754e-01 -3.78677547e-02 -5.14597748e-04
-8.77490520e-01 8.38017523e-01 3.27204227e-01 -6.89403176e-01
6.21200740e-01 -1.42572492e-01 4.55601215e-01 -3.30819748e-02
6.11619055e-01 2.44651660e-01 -4.58642215e-01 -1.76843062e-01
-2.43508190e-01 9.50471163e-01 -9.05079767e-02 9.77408290e-02
1.22483730e+00 2.94667661e-01 3.01066064e-03 4.46632296e-01
1.69310129e+00 -3.68394822e-01 -1.31505001e+00 -4.86711770e-01
-6.75788447e-02 -7.95887709e-01 -2.70317286e-01 -4.88669544e-01
-1.14613569e+00 1.27030706e+00 3.21897179e-01 -3.12754184e-01
7.24270940e-01 -2.41440292e-02 1.57134819e+00 7.04514444e-01
3.31725717e-01 -1.11416078e+00 5.07151484e-01 3.98439378e-01
1.08614373e+00 -1.36866200e+00 -2.28661180e-01 2.31589079e-01
-5.83359420e-01 1.26483512e+00 6.31834090e-01 -4.59294975e-01
5.72182715e-01 1.76227391e-02 -4.48937118e-01 -4.99307632e-01
-7.79569864e-01 -1.84301753e-02 4.24336016e-01 3.84147078e-01
-3.63820381e-02 1.04093894e-01 1.23152375e-01 9.61969316e-01
3.26587856e-01 1.22566268e-01 -2.39774808e-01 7.68177867e-01
-2.18232259e-01 -1.04861486e+00 4.77443151e-02 4.33062166e-01
3.61261010e-01 -7.89465234e-02 -7.95238554e-01 6.49739146e-01
-1.97302595e-01 2.33773282e-03 5.05523160e-02 -4.84362990e-01
4.92423981e-01 1.29521117e-01 4.42239970e-01 -3.39020193e-01
-2.51362354e-01 -4.26945806e-01 -5.05175531e-01 -6.94196224e-01
2.92549521e-01 -1.60793979e-02 -6.95583940e-01 -4.87515539e-01
2.80818015e-01 -1.20528340e-01 6.24025643e-01 4.85589683e-01
5.50852358e-01 8.73966590e-02 7.72759140e-01 -1.12405217e+00
-9.19360161e-01 -6.44345880e-01 -9.15220141e-01 7.69271672e-01
2.45695353e-01 -9.50816214e-01 -2.57788807e-01 -4.56119835e-01] | [11.670783042907715, 0.5951008796691895] |
39458165-993b-49bc-81bd-c5492cb435b7 | tackling-morpion-solitaire-with-alphazero | 2006.07970 | null | https://arxiv.org/abs/2006.07970v1 | https://arxiv.org/pdf/2006.07970v1.pdf | Tackling Morpion Solitaire with AlphaZero-likeRanked Reward Reinforcement Learning | Morpion Solitaire is a popular single player game, performed with paper and pencil. Due to its large state space (on the order of the game of Go) traditional search algorithms, such as MCTS, have not been able to find good solutions. A later algorithm, Nested Rollout Policy Adaptation, was able to find a new record of 82 steps, albeit with large computational resources. After achieving this record, to the best of our knowledge, there has been no further progress reported, for about a decade. In this paper we take the recent impressive performance of deep self-learning reinforcement learning approaches from AlphaGo/AlphaZero as inspiration to design a searcher for Morpion Solitaire. A challenge of Morpion Solitaire is that the state space is sparse, there are few win/loss signals. Instead, we use an approach known as ranked reward to create a reinforcement learning self-play framework for Morpion Solitaire. This enables us to find medium-quality solutions with reasonable computational effort. Our record is a 67 steps solution, which is very close to the human best (68) without any other adaptation to the problem than using ranked reward. We list many further avenues for potential improvement. | ['Hui Wang', 'Aske Plaat', 'Mike Preuss', 'Michael Emmerich'] | 2020-06-14 | null | null | null | null | ['game-of-go', 'solitaire'] | ['playing-games', 'playing-games'] | [-4.47374791e-01 -1.48733826e-02 -2.08327979e-01 3.84060770e-01
-9.84287024e-01 -7.23428249e-01 3.33985418e-01 -9.74870399e-02
-6.98590398e-01 1.31699932e+00 -9.55207869e-02 -3.36806506e-01
-6.35136485e-01 -7.19977021e-01 -5.29840350e-01 -6.66799724e-01
-2.80351788e-01 6.59025967e-01 5.10631680e-01 -8.51190209e-01
6.09642923e-01 2.35777572e-01 -1.60230649e+00 -3.32748368e-02
6.82365596e-01 7.27811635e-01 3.73614818e-01 8.79627526e-01
2.12238759e-01 1.10181212e+00 -7.40391791e-01 -1.16411008e-01
6.13801360e-01 -6.75978661e-01 -9.06044602e-01 -4.32180315e-01
8.41742456e-02 -3.99540484e-01 -4.83751118e-01 6.39221072e-01
8.88922930e-01 4.67865199e-01 1.82987303e-01 -1.16404569e+00
-1.38354748e-01 6.48065925e-01 -5.45708358e-01 6.32469833e-01
4.20669436e-01 5.38840473e-01 1.31111515e+00 -1.98519975e-01
6.92939281e-01 7.74960995e-01 7.03591168e-01 7.38592923e-01
-9.86522019e-01 -6.97073996e-01 -1.52739763e-01 4.75584149e-01
-1.14400828e+00 -1.22590914e-01 5.58260679e-01 1.69780105e-01
1.35169053e+00 4.01829720e-01 1.21626985e+00 1.08001804e+00
2.59804726e-01 1.19158697e+00 1.41115880e+00 -3.94589335e-01
7.19616890e-01 -2.80951977e-01 -7.31445432e-01 7.63306439e-01
-1.16520740e-01 6.64984465e-01 -8.26685190e-01 -6.98666722e-02
9.01928484e-01 -5.28092027e-01 1.98541954e-01 -4.59154934e-01
-8.31287622e-01 1.03255832e+00 3.90671492e-01 3.68800968e-01
-3.59083235e-01 6.91932678e-01 3.20899308e-01 6.29309833e-01
1.08708672e-01 1.11499071e+00 -5.26978731e-01 -9.85273242e-01
-1.33207536e+00 8.56122732e-01 9.15947735e-01 3.72264236e-01
4.67288673e-01 2.88251430e-01 8.18314105e-02 5.38760781e-01
2.16299836e-02 2.04132423e-01 6.22514129e-01 -1.38678217e+00
2.71380752e-01 2.46553347e-01 2.74397343e-01 -6.78750932e-01
-5.81647992e-01 -6.86863065e-01 -3.24273705e-01 8.86445999e-01
6.05015278e-01 -4.71893251e-01 -3.86910498e-01 1.54314649e+00
1.25891343e-01 1.26215026e-01 -3.67546827e-02 9.59238708e-01
3.96995246e-01 6.41640007e-01 -4.20468092e-01 -1.08917519e-01
7.91257262e-01 -1.01722527e+00 -4.33004290e-01 -4.12795901e-01
4.67171550e-01 -5.01308739e-01 1.01312327e+00 8.09103310e-01
-1.59101152e+00 -2.41645183e-02 -1.25129855e+00 4.43338543e-01
-1.88899726e-01 -3.90192181e-01 9.69976246e-01 7.29118705e-01
-1.22145414e+00 1.08020818e+00 -9.47145104e-01 -2.55217791e-01
3.23087156e-01 5.84776640e-01 4.47001457e-02 3.27900827e-01
-1.31891930e+00 1.24249601e+00 5.58202744e-01 -2.70739943e-01
-1.01974905e+00 -5.11255443e-01 -3.13666523e-01 5.35028055e-02
1.16398156e+00 -4.34177786e-01 1.58265007e+00 -8.44306409e-01
-1.94837737e+00 5.18921912e-01 2.92588413e-01 -7.50169873e-01
7.30661929e-01 2.31039915e-02 7.04825949e-03 -5.39284805e-03
-3.05191725e-02 5.92730761e-01 4.71526265e-01 -8.65509331e-01
-8.98570180e-01 2.42293440e-02 2.43618786e-01 5.33264458e-01
-8.75032023e-02 7.63148740e-02 -2.20135152e-01 -4.18240637e-01
-2.00519919e-01 -9.68299150e-01 -5.37036061e-01 -3.35073888e-01
5.02561741e-02 -4.61648345e-01 1.45990357e-01 -2.74155498e-01
1.49225855e+00 -1.54079771e+00 1.58689708e-01 1.91624716e-01
1.48716584e-01 3.84398162e-01 -2.72054046e-01 9.30626452e-01
1.81869775e-01 4.75113653e-02 6.42529353e-02 1.25799209e-01
1.09291665e-01 3.04623187e-01 -1.81562658e-02 3.15204322e-01
-1.22393891e-01 1.23097265e+00 -1.31601739e+00 -2.98361719e-01
-9.71995518e-02 -3.06475073e-01 -9.10522103e-01 1.10792108e-01
-4.05351400e-01 -5.28074466e-02 -3.66177678e-01 4.84238565e-01
2.13892087e-01 -1.36079922e-01 9.31281745e-02 4.69294816e-01
-5.34742534e-01 3.43764901e-01 -1.33150327e+00 1.81185365e+00
-1.42173231e-01 5.70845008e-01 -7.61420056e-02 -1.07944596e+00
8.50003719e-01 7.68050253e-02 1.00893688e+00 -1.02133274e+00
1.98901802e-01 5.76312125e-01 2.27719396e-01 -4.03998971e-01
6.93453550e-01 -1.60524249e-01 -4.21077833e-02 5.93552709e-01
-2.58822478e-02 -5.52949965e-01 6.86115682e-01 1.65850908e-01
1.64235198e+00 3.86988252e-01 2.97076911e-01 -1.75906420e-01
4.18475317e-03 3.85964155e-01 3.98891181e-01 1.17552745e+00
-3.98505688e-01 4.74566549e-01 6.08241260e-01 -6.16303563e-01
-1.07023895e+00 -8.59803855e-01 4.41273838e-01 1.19940615e+00
-3.01101822e-02 -6.22930825e-01 -7.48318434e-01 -3.71820599e-01
-4.47738431e-02 4.70361352e-01 -5.46598792e-01 -2.38997236e-01
-6.84576809e-01 -6.65754855e-01 6.19514704e-01 3.82584035e-01
5.97297251e-01 -1.68089688e+00 -1.25159311e+00 6.60722136e-01
6.39024377e-02 -5.08425355e-01 -3.08765084e-01 5.81952691e-01
-7.10406005e-01 -1.04812098e+00 -7.98668563e-01 -5.41793108e-01
-2.56432950e-01 -1.17445998e-01 1.16775966e+00 1.01455808e-01
-4.40170527e-01 1.56983256e-01 -3.78701270e-01 -2.31062785e-01
-3.06026667e-01 2.13194475e-01 7.46254027e-02 -7.83318698e-01
1.12474129e-01 -6.37304246e-01 -7.30398059e-01 1.70534894e-01
-5.59126496e-01 -6.59670308e-02 5.94954967e-01 1.03642178e+00
2.26354405e-01 2.42486686e-01 6.40337706e-01 -3.28493685e-01
1.09078431e+00 -4.64508891e-01 -6.38851404e-01 -1.83190361e-01
-8.12387347e-01 2.58692861e-01 6.03078485e-01 -4.73181278e-01
-5.97821951e-01 -8.06858316e-02 -2.81410277e-01 -2.14091048e-01
2.97430873e-01 6.12377524e-01 6.80275559e-01 -1.85989991e-01
1.08508635e+00 3.16263348e-01 1.44037694e-01 -2.61052042e-01
2.33572528e-01 3.08616757e-01 3.48622531e-01 -5.06490409e-01
5.17014503e-01 1.28452837e-01 -9.23760049e-03 -3.59029472e-01
-6.17493510e-01 -3.19653869e-01 5.03929658e-03 -3.27555388e-01
4.74949449e-01 -5.00410497e-01 -1.42591071e+00 4.56361830e-01
-5.60107410e-01 -9.18687880e-01 -6.58604443e-01 2.62710720e-01
-1.25576854e+00 7.12739397e-03 -5.98839104e-01 -9.95900035e-01
-1.36258826e-01 -8.92240524e-01 6.27041817e-01 4.96844530e-01
-2.06945106e-01 -6.85307145e-01 5.53919375e-01 3.56037110e-01
7.63335347e-01 5.22670858e-02 3.56194556e-01 -3.91857356e-01
-6.09083295e-01 -8.56705941e-03 1.72622532e-01 -8.58845096e-03
-7.73115531e-02 -2.12599143e-01 -2.04573229e-01 -5.91383040e-01
-3.12949717e-01 -7.92315841e-01 6.21649265e-01 4.17811245e-01
7.32496619e-01 -2.08986267e-01 1.41131267e-01 3.85383725e-01
1.38308477e+00 5.70993960e-01 7.00270236e-01 1.12776971e+00
-5.06447181e-02 1.26549423e-01 9.24662650e-01 7.70454466e-01
3.25121492e-01 8.51933956e-01 7.95060515e-01 1.23281837e-01
1.74978618e-02 -3.97351444e-01 3.84309947e-01 4.04351950e-01
-3.81269872e-01 -2.46865004e-01 -1.01631784e+00 5.31235218e-01
-2.05138254e+00 -1.29680216e+00 3.59497964e-01 1.91620851e+00
8.89257252e-01 3.73902559e-01 5.93801975e-01 2.12867916e-01
2.46415213e-01 1.20888717e-01 -9.32305455e-01 -6.61766708e-01
-4.11830358e-02 4.62485582e-01 6.96618497e-01 3.97724122e-01
-7.47873664e-01 1.27566338e+00 7.09883547e+00 1.21359003e+00
-1.00043178e+00 -1.25795141e-01 2.95704931e-01 -6.81723535e-01
5.65849207e-02 9.69525650e-02 -5.00855982e-01 5.20766497e-01
8.83740962e-01 -3.70139778e-01 1.24037969e+00 7.18899786e-01
4.09504890e-01 -4.69101578e-01 -5.57691097e-01 1.00067854e+00
-1.01633422e-01 -1.49686706e+00 -6.71538770e-01 8.14700872e-02
9.05473113e-01 1.23892941e-01 1.05446644e-01 6.52547777e-01
9.17264998e-01 -1.13083577e+00 6.68045282e-01 3.30012828e-01
4.27295774e-01 -1.16368377e+00 4.04285997e-01 7.19906092e-01
-8.18386674e-01 -3.41295421e-01 -3.72286290e-01 -4.06069666e-01
-4.50382382e-02 -1.89012244e-01 -7.17917621e-01 3.29211801e-01
8.02281797e-01 3.93833041e-01 -2.39598006e-01 1.48233020e+00
-2.79037394e-02 5.29290974e-01 -4.53013599e-01 -7.57662594e-01
7.88552165e-01 1.51249366e-02 6.21699750e-01 5.55563986e-01
3.66856664e-01 7.04010651e-02 2.55342573e-01 7.25535810e-01
1.71015576e-01 5.55321127e-02 -3.44211310e-01 -1.36271253e-01
3.34997356e-01 8.91870320e-01 -7.07420528e-01 8.69941264e-02
5.84609285e-02 9.27879691e-01 5.19175589e-01 -4.81547341e-02
-8.47885966e-01 -2.72851169e-01 4.08830732e-01 8.02879333e-02
3.69048297e-01 -2.57684141e-01 -1.45654306e-02 -8.66423845e-01
-2.13301256e-01 -1.39090836e+00 3.52719635e-01 -8.01643670e-01
-8.26898515e-01 4.58133489e-01 -2.19108723e-02 -1.15414166e+00
-7.53406584e-01 -4.15834486e-01 -5.50279498e-01 5.69768846e-01
-1.08874619e+00 -6.04860544e-01 2.88883567e-01 5.36167622e-01
7.69856036e-01 -6.47541106e-01 6.55686259e-01 -3.47726829e-02
-1.98133990e-01 5.82982838e-01 5.72459698e-01 -4.76281792e-01
3.55923116e-01 -1.49894321e+00 3.11959743e-01 4.62296247e-01
2.05166072e-01 2.53213733e-01 9.69006777e-01 -5.78689337e-01
-1.61607540e+00 -1.52660549e-01 4.33025569e-01 -1.72304615e-01
1.02834964e+00 -8.09074491e-02 -3.81829917e-01 1.90087631e-01
3.96871924e-01 -3.28910410e-01 2.47169361e-01 1.45944819e-01
2.65288591e-01 -3.01793180e-02 -9.11725104e-01 8.71462703e-01
1.04996717e+00 -9.27515253e-02 -4.27269220e-01 2.60124981e-01
2.41843700e-01 -8.31506312e-01 -4.31711406e-01 3.72003019e-02
5.38751066e-01 -1.32963812e+00 8.98197353e-01 -7.11618900e-01
4.90085393e-01 5.68657406e-02 2.47935191e-01 -1.68642473e+00
-5.22971332e-01 -1.22646999e+00 -2.93236911e-01 6.37655675e-01
2.99091369e-01 -3.63138616e-01 1.46211219e+00 3.48656178e-01
-7.18129128e-02 -1.28704476e+00 -1.17533433e+00 -1.20972717e+00
3.56998950e-01 -2.81318009e-01 3.65560770e-01 5.61446309e-01
2.82346457e-01 6.40130714e-02 -8.23597610e-01 -5.94773889e-01
4.26072717e-01 2.07084626e-01 6.10177100e-01 -8.06283414e-01
-8.10261309e-01 -8.09450507e-01 -1.28593162e-01 -1.04347742e+00
-1.62033543e-01 -7.56741762e-01 1.61189958e-01 -1.65586436e+00
1.31304339e-01 -5.42033195e-01 -3.98359537e-01 6.03839040e-01
7.15245456e-02 2.59562343e-01 5.37630677e-01 9.78737995e-02
-1.00763440e+00 5.43443322e-01 1.52260196e+00 -3.56678180e-02
-3.78511816e-01 5.61866537e-02 -8.52380455e-01 5.63917041e-01
1.14204991e+00 -5.60187280e-01 -3.65862548e-01 -1.22019671e-01
7.30430841e-01 5.15422404e-01 2.74987780e-02 -1.23371601e+00
4.39721406e-01 -5.02563953e-01 -4.22634594e-02 -3.76299471e-01
3.83859962e-01 -4.77740020e-01 3.38236801e-02 8.78508329e-01
-4.13201779e-01 2.70135403e-01 3.12750608e-01 2.95382470e-01
1.61588956e-02 -6.65621281e-01 5.66986442e-01 -5.16516268e-01
-8.18493366e-01 1.20791756e-01 -9.47633624e-01 3.94960344e-01
1.07827616e+00 -2.58128017e-01 -1.31715730e-01 -6.90104306e-01
-7.06230164e-01 3.93619806e-01 2.46447712e-01 3.22617769e-01
4.35534447e-01 -1.20719731e+00 -7.18617141e-01 -2.35119909e-01
-5.06983101e-01 -4.15059119e-01 1.74170554e-01 4.57511961e-01
-6.76641762e-01 3.69759142e-01 -6.29874766e-01 -2.10730112e-04
-1.15853834e+00 4.03362125e-01 5.98004997e-01 -9.26601946e-01
-5.13335109e-01 7.63844430e-01 -6.84786022e-01 -2.55384684e-01
1.73530445e-01 1.99535966e-01 -2.92153209e-02 4.02857736e-02
2.75042474e-01 6.45631850e-01 -6.59604417e-03 8.81889388e-02
-3.42987746e-01 1.83670491e-01 3.32497694e-02 -5.05154848e-01
1.68361187e+00 1.91179529e-01 2.74098665e-01 1.21685192e-01
7.86271214e-01 -6.00433648e-02 -1.57199943e+00 2.36781806e-01
-1.45025095e-02 -5.06083250e-01 1.77303165e-01 -1.11691737e+00
-1.03446639e+00 5.31263888e-01 6.20308816e-01 3.80218744e-01
9.98180568e-01 -2.24866286e-01 8.89177322e-01 7.47153103e-01
7.48946726e-01 -1.51732671e+00 6.38492227e-01 8.85050178e-01
8.86196375e-01 -1.18440807e+00 -8.92751515e-02 5.25326610e-01
-6.79242730e-01 1.02033329e+00 7.95256734e-01 -6.68153763e-01
8.14475119e-02 5.08926928e-01 -2.43661448e-01 -1.92962378e-01
-1.12089527e+00 -4.73804086e-01 -2.43116617e-01 4.74531025e-01
1.09330468e-01 -2.10199460e-01 -5.98557711e-01 3.26077849e-01
-5.97490430e-01 2.21773952e-01 5.50191343e-01 1.09609747e+00
-7.55485237e-01 -1.28136218e+00 -2.03102276e-01 4.73634750e-01
-5.81369340e-01 -1.24680195e-02 -1.85557663e-01 8.23023617e-01
-1.68248221e-01 9.64002967e-01 -1.68101266e-01 -3.77551824e-01
3.16971302e-01 -3.87902379e-01 7.32829690e-01 -2.05228612e-01
-1.02733707e+00 -4.17681783e-02 7.59695023e-02 -9.11561847e-01
-1.41715584e-02 -6.13423645e-01 -1.32538950e+00 -7.51479506e-01
-1.61129028e-01 3.61489832e-01 5.30294120e-01 8.25312853e-01
-4.00500782e-02 5.18962562e-01 5.38212240e-01 -9.90330875e-01
-9.09264207e-01 -4.82104629e-01 -7.07336068e-01 9.26428810e-02
6.77492172e-02 -6.96509898e-01 -1.36368349e-01 -8.56331646e-01] | [3.5836448669433594, 1.5313441753387451] |
dcc2ad54-95fc-4ebd-bc6f-0d5b83fa7290 | signal-is-harder-to-learn-than-bias-debiasing | 2305.19671 | null | https://arxiv.org/abs/2305.19671v1 | https://arxiv.org/pdf/2305.19671v1.pdf | Signal Is Harder To Learn Than Bias: Debiasing with Focal Loss | Spurious correlations are everywhere. While humans often do not perceive them, neural networks are notorious for learning unwanted associations, also known as biases, instead of the underlying decision rule. As a result, practitioners are often unaware of the biased decision-making of their classifiers. Such a biased model based on spurious correlations might not generalize to unobserved data, leading to unintended, adverse consequences. We propose Signal is Harder (SiH), a variational-autoencoder-based method that simultaneously trains a biased and unbiased classifier using a novel, disentangling reweighting scheme inspired by the focal loss. Using the unbiased classifier, SiH matches or improves upon the performance of state-of-the-art debiasing methods. To improve the interpretability of our technique, we propose a perturbation scheme in the latent space for visualizing the bias that helps practitioners become aware of the sources of spurious correlations. | ['Julia E. Vogt', 'Ričards Marcinkevičs', 'Laura Manduchi', 'Moritz Vandenhirtz'] | 2023-05-31 | null | null | null | null | ['domain-generalization'] | ['methodology'] | [ 2.17962459e-01 1.18842065e-01 -1.32761136e-01 -4.28927034e-01
-2.93060064e-01 -4.46964771e-01 7.38814890e-01 -1.13097072e-01
-1.45415947e-01 8.45279157e-01 2.64241397e-01 -1.78480119e-01
-1.28639206e-01 -7.01855302e-01 -6.10775530e-01 -1.15328085e+00
2.23492056e-01 2.91766375e-01 -1.59922317e-01 -1.24929659e-02
4.63101208e-01 1.52066037e-01 -1.60267997e+00 3.54863256e-01
9.30187225e-01 8.14606786e-01 -2.91432261e-01 3.15935582e-01
7.49075115e-02 8.96744370e-01 -7.49936163e-01 -6.70460820e-01
-3.70904244e-02 -5.65254569e-01 -3.20985079e-01 -2.21193761e-01
6.32686198e-01 -2.60287911e-01 -2.73967654e-01 1.63755512e+00
1.86942428e-01 6.84003904e-02 9.43004727e-01 -1.40320218e+00
-1.05076504e+00 8.77268136e-01 -6.38442695e-01 -1.63275022e-02
-1.95905715e-01 2.41674244e-01 1.25108576e+00 -8.19446206e-01
3.25221419e-01 1.42359161e+00 6.45438433e-01 6.72067702e-01
-1.67708063e+00 -1.00728452e+00 2.91994363e-01 2.09713042e-01
-1.22202277e+00 -2.93953270e-01 1.09122312e+00 -6.55872881e-01
3.26610863e-01 3.32205296e-01 4.50770944e-01 1.86380339e+00
2.75051355e-01 7.05983758e-01 1.31919396e+00 -2.15707928e-01
5.47386706e-01 4.15717304e-01 3.41280639e-01 5.19349933e-01
6.96102142e-01 6.52224958e-01 -8.24087381e-01 -4.62575644e-01
5.70333123e-01 2.08641231e-01 -4.37503517e-01 -7.34662712e-01
-1.34588099e+00 9.18206930e-01 5.74569523e-01 1.85772747e-01
-4.16396588e-01 2.68146276e-01 4.86158989e-02 2.44564816e-01
7.47746229e-01 6.70468152e-01 -2.33919710e-01 2.51320805e-02
-1.08698595e+00 4.01190937e-01 5.70302904e-01 4.03151214e-01
6.50538206e-01 6.62798509e-02 -3.96708310e-01 5.35030782e-01
2.19238535e-01 3.75227362e-01 3.71301919e-01 -1.11344635e+00
-6.12585619e-02 4.84936923e-01 3.20762157e-01 -1.25075936e+00
5.16356789e-02 -8.23664844e-01 -1.20164776e+00 6.13609493e-01
5.41996837e-01 4.94228192e-02 -7.61313140e-01 1.99235547e+00
9.63290036e-02 1.47500470e-01 -7.67276734e-02 1.10862422e+00
2.85482109e-01 3.24915320e-01 -5.16678847e-04 -5.82179800e-02
1.07203782e+00 -7.44173467e-01 -1.00455725e+00 -5.89593835e-02
2.38286301e-01 -2.72657782e-01 1.17335904e+00 6.76802516e-01
-7.71743178e-01 -2.37423837e-01 -1.16260993e+00 4.93733175e-02
-2.86428660e-01 -1.81544006e-01 6.69464409e-01 6.84077263e-01
-6.01749837e-01 9.71979201e-01 -8.49554837e-01 9.73561704e-02
6.73073053e-01 -2.84862407e-02 4.60153009e-04 1.11949794e-01
-1.17601895e+00 9.76042449e-01 1.25567108e-01 5.75248413e-02
-1.02841008e+00 -9.99652863e-01 -5.15918493e-01 2.20797271e-01
4.40392911e-01 -7.08972812e-01 1.04528987e+00 -1.26931894e+00
-1.36590016e+00 6.12682462e-01 -3.28031272e-01 -2.91292936e-01
8.38157833e-01 -4.47710842e-01 -2.38421053e-01 -3.20389688e-01
-9.60934833e-02 3.42712373e-01 1.38087797e+00 -1.37602603e+00
-1.22036867e-01 -4.30184841e-01 -1.94155931e-01 -1.72008857e-01
-3.11060578e-01 -5.51938295e-01 4.25380558e-01 -8.73841345e-01
2.83229388e-02 -6.64626360e-01 2.33615711e-02 3.92416328e-01
-6.83440030e-01 -1.94606677e-01 8.70991170e-01 -4.17419612e-01
1.21849978e+00 -2.31493688e+00 2.85781115e-01 2.42365494e-01
6.60773873e-01 -8.57476424e-03 -1.80712231e-02 1.56147361e-01
-2.09202081e-01 2.51071900e-01 -4.41205859e-01 -2.31713995e-01
1.54918447e-01 1.18734434e-01 -5.52484274e-01 5.93450606e-01
3.20966184e-01 5.33939242e-01 -1.14330947e+00 -2.61041895e-02
7.16742454e-03 5.73057950e-01 -6.46914124e-01 1.71957895e-01
-2.33036101e-01 5.14938295e-01 -8.86360258e-02 4.24651921e-01
7.32710302e-01 -3.06909233e-01 6.55915663e-02 -2.79116541e-01
3.84995341e-02 2.07832515e-01 -9.22239721e-01 1.30839598e+00
-2.92628467e-01 8.94918740e-01 -2.04749733e-01 -1.01726270e+00
9.04799700e-01 1.57316089e-01 -2.21331969e-01 -1.79606304e-01
1.63346827e-01 1.83162004e-01 1.14824072e-01 -2.95939356e-01
3.35174412e-01 -3.61671150e-01 3.06699306e-01 5.05028486e-01
-1.45679742e-01 1.50820553e-01 -2.67052323e-01 2.46241435e-01
8.33076119e-01 2.48562768e-02 2.09554970e-01 -4.60861951e-01
6.78719729e-02 -2.50862300e-01 6.28260732e-01 1.04720044e+00
-1.66226506e-01 6.92739069e-01 8.09563577e-01 -4.85160619e-01
-1.11737585e+00 -1.29355085e+00 -5.61512299e-02 7.66620219e-01
3.93023714e-02 -7.31780007e-02 -6.56046808e-01 -8.03663611e-01
2.11315691e-01 1.30752611e+00 -1.03209078e+00 -6.81601524e-01
4.23736535e-02 -7.39361048e-01 3.35014433e-01 3.65182877e-01
3.61277550e-01 -7.49097109e-01 -4.53625858e-01 -4.53297496e-02
-1.17153384e-01 -5.49322844e-01 -3.88667703e-01 1.14249811e-01
-8.47280979e-01 -1.00907362e+00 -7.89715767e-01 -7.40453750e-02
8.39947701e-01 7.54574686e-03 1.23863077e+00 6.82754535e-03
4.54277247e-02 -1.51961520e-01 4.74655479e-02 -3.77298892e-01
-6.48457110e-01 -3.40435892e-01 2.07264468e-01 3.26887876e-01
7.40539968e-01 -8.74162674e-01 -7.33795822e-01 1.62642300e-01
-7.79736340e-01 2.37204172e-02 5.65385818e-01 1.23592889e+00
1.65193304e-01 -1.10175543e-01 3.89554828e-01 -9.69972193e-01
8.56652558e-01 -5.04151285e-01 -5.80781698e-01 1.68071523e-01
-9.14198399e-01 6.38252318e-01 4.55133229e-01 -9.50096607e-01
-1.06727374e+00 -5.22334397e-01 4.76831883e-01 -7.27195144e-01
-1.01304129e-01 2.69042671e-01 -1.23112284e-01 3.12707961e-01
9.25602734e-01 1.25147045e-01 -3.73912677e-02 -4.83522564e-01
5.49236476e-01 6.21552169e-01 4.54509586e-01 -5.48210740e-01
6.81029260e-01 5.95872939e-01 -5.33211678e-02 -5.40061474e-01
-1.13572598e+00 -2.06946060e-02 -3.72529775e-01 -1.04993954e-01
6.55885339e-01 -7.35745966e-01 -7.81408191e-01 4.61595446e-01
-1.27892375e+00 -4.42148298e-02 -3.45015645e-01 3.95857513e-01
-2.60018378e-01 1.47677183e-01 -2.60793358e-01 -1.05037475e+00
-7.46460706e-02 -1.06349897e+00 7.61192918e-01 3.42510752e-02
-7.34332919e-01 -8.54537249e-01 2.02035591e-01 1.62783727e-01
5.57467401e-01 1.95358932e-01 1.24925458e+00 -6.54462993e-01
-4.92776334e-01 -1.68384805e-01 -2.38270774e-01 6.82188213e-01
7.62334540e-02 8.44600499e-02 -1.29854822e+00 -1.93378657e-01
-3.08734868e-02 -1.68711722e-01 1.14097798e+00 3.30939561e-01
1.36048687e+00 -6.42269254e-01 -1.53586119e-01 3.32727492e-01
1.00301623e+00 -6.58046305e-02 4.85916346e-01 5.06448336e-02
6.42163813e-01 7.30703831e-01 4.47637076e-03 3.92996252e-01
-2.84076463e-02 4.01518911e-01 5.27292669e-01 3.28301303e-02
6.43604249e-02 -5.43632805e-01 4.17752892e-01 5.80651045e-01
-1.40847370e-01 -7.57103041e-02 -6.42933309e-01 5.69264650e-01
-1.76648080e+00 -1.08246768e+00 -4.19130027e-02 2.44757104e+00
1.03425360e+00 2.06149504e-01 -2.47077778e-01 1.45852119e-01
8.61455321e-01 3.14273208e-01 -9.48593855e-01 -2.85586298e-01
-1.05200030e-01 3.66213806e-02 2.62864143e-01 6.15625501e-01
-7.93493629e-01 6.24813855e-01 5.93534517e+00 6.96981132e-01
-1.08065236e+00 5.21411374e-02 6.90678239e-01 -2.39628121e-01
-7.11025119e-01 1.27789691e-01 -3.41131181e-01 6.12884045e-01
5.45434296e-01 2.58960407e-02 5.73033333e-01 8.84616911e-01
1.18050940e-01 1.61836036e-02 -1.51070035e+00 9.81543303e-01
-6.54310361e-02 -1.35362339e+00 2.20428586e-01 3.59555408e-02
7.79250383e-01 -3.65263253e-01 5.66130459e-01 1.28114745e-01
6.14165962e-01 -1.08671343e+00 9.38652575e-01 7.44524062e-01
3.25405717e-01 -7.57300615e-01 6.65543616e-01 3.15682769e-01
-1.10948004e-01 1.34894243e-02 -3.48961383e-01 -3.17358434e-01
-3.09092104e-01 1.17451108e+00 -3.08398843e-01 -6.35002032e-02
7.11554110e-01 7.90604055e-01 -4.22349513e-01 6.33823216e-01
-6.18892550e-01 8.40954006e-01 -5.16285524e-02 -2.19442442e-01
3.67533852e-04 -2.41724417e-01 9.10679340e-01 8.49654973e-01
3.22869509e-01 -1.55845255e-01 -4.66454983e-01 1.75222921e+00
-2.01600358e-01 -5.33897460e-01 -7.50839472e-01 -2.73822814e-01
5.28785527e-01 8.74665976e-01 -2.19572753e-01 -3.92155260e-01
-5.12472652e-02 9.80383575e-01 2.91455269e-01 8.02174926e-01
-4.96995747e-01 -2.14490116e-01 9.44396138e-01 -1.28000617e-01
1.66393906e-01 6.96965083e-02 -4.84312713e-01 -1.53440976e+00
1.97356921e-02 -1.08464241e+00 -2.42231134e-02 -7.91228771e-01
-1.77105069e+00 2.98819780e-01 -1.17217645e-01 -1.12980068e+00
5.41081615e-02 -4.50256646e-01 -6.97947919e-01 8.84630382e-01
-1.20996308e+00 -6.80396080e-01 -1.83832616e-01 1.80370763e-01
2.81575561e-01 -7.05941617e-02 7.04300404e-01 -1.87497914e-01
-6.24458969e-01 5.73962212e-01 3.25212926e-01 -7.05679655e-02
8.48128855e-01 -1.28899729e+00 1.00053862e-01 7.31454492e-01
2.99036741e-01 9.55672741e-01 1.25120437e+00 -3.47776085e-01
-7.60690331e-01 -8.76631856e-01 8.07861269e-01 -6.21746719e-01
8.22801113e-01 -5.20035148e-01 -1.20160913e+00 4.40900415e-01
1.42871693e-01 -1.60554230e-01 6.50916874e-01 4.82605636e-01
-9.39758480e-01 -1.50672048e-01 -1.09792137e+00 9.07309353e-01
8.38258624e-01 -6.17826581e-01 -8.42007875e-01 2.91545205e-02
5.72816253e-01 1.84573501e-01 -3.04707468e-01 9.24210548e-02
7.66539514e-01 -1.22445643e+00 8.31268489e-01 -8.38472605e-01
8.29722822e-01 -1.44308463e-01 -2.31374711e-01 -1.68056405e+00
-3.64610016e-01 -4.20978606e-01 -3.61932784e-01 9.70543623e-01
4.82263684e-01 -7.97996879e-01 7.85300732e-01 6.35384679e-01
4.26687211e-01 -5.86440325e-01 -8.36132050e-01 -7.63711393e-01
2.95148581e-01 -3.78982782e-01 7.36650288e-01 1.24033237e+00
-1.42746150e-01 1.94251567e-01 -6.64710522e-01 3.83119404e-01
1.20274174e+00 1.78168222e-01 4.10569638e-01 -1.46392167e+00
-4.74341154e-01 -6.45606995e-01 -2.74472862e-01 -8.96305740e-01
9.81644839e-02 -6.05002582e-01 4.47525363e-03 -6.81023240e-01
3.91979933e-01 -2.78525986e-03 -7.32994974e-01 2.13289380e-01
-3.66777152e-01 5.11564650e-02 3.60308625e-02 2.25589693e-01
-3.06462765e-01 8.53012264e-01 1.12896168e+00 -4.00167197e-01
4.47263680e-02 -7.01467842e-02 -9.49856639e-01 8.85248303e-01
6.71781361e-01 -8.49603593e-01 -2.91869611e-01 -2.37248480e-01
5.81799567e-01 -2.72617251e-01 8.50896418e-01 -6.66550279e-01
1.03034459e-01 -2.48413950e-01 3.37263435e-01 -4.92176682e-01
2.33349219e-01 -6.65304840e-01 6.55449703e-02 4.65038776e-01
-8.55514288e-01 -3.11862260e-01 -1.89380795e-01 7.99511969e-01
-1.91542596e-01 -6.01636842e-02 8.04110944e-01 -1.81972254e-02
-8.03885087e-02 -5.04197627e-02 -3.81527722e-01 -5.95243042e-03
5.47975719e-01 9.04031992e-02 -5.05924761e-01 -6.72719300e-01
-6.89441741e-01 -1.09112589e-03 4.73548591e-01 2.70228773e-01
6.36063039e-01 -1.26573408e+00 -6.77763462e-01 2.33659893e-01
2.00858414e-01 -1.99167296e-01 1.19737521e-01 9.02172446e-01
4.25773300e-03 1.24681070e-01 -7.32351691e-02 -4.84979510e-01
-1.04111540e+00 5.43769240e-01 3.80352944e-01 -8.38912949e-02
-4.10538465e-01 9.16565120e-01 4.90589321e-01 -3.50500971e-01
4.52891409e-01 -3.74242574e-01 1.25468612e-01 1.36710316e-01
5.86785614e-01 4.31798190e-01 -2.20900953e-01 -8.29683542e-02
-1.83053017e-01 2.65810609e-01 -1.74314275e-01 -1.53080136e-01
1.18246388e+00 -6.85653239e-02 -1.30244344e-01 9.07897949e-01
9.54837918e-01 -4.66383845e-02 -1.55900168e+00 -2.76332885e-01
3.82939577e-02 -5.07532239e-01 3.68828028e-01 -1.09386420e+00
-1.01005208e+00 1.38484979e+00 6.86486542e-01 2.16603041e-01
7.86587775e-01 -1.75908267e-01 3.78413916e-01 4.56803769e-01
1.98169485e-01 -7.90816367e-01 1.15662560e-01 1.67523935e-01
8.91513288e-01 -1.38589859e+00 1.13226334e-02 -3.43819819e-02
-6.54964864e-01 9.89389300e-01 3.02215278e-01 -1.77734464e-01
5.92553437e-01 -7.14995945e-03 2.76912421e-01 -2.35704497e-01
-8.84957314e-01 2.88431793e-01 2.41672680e-01 6.27800941e-01
2.64864504e-01 1.08917855e-01 -4.65592556e-03 7.13681042e-01
-1.11247770e-01 -6.59614131e-02 4.45769191e-01 5.51390588e-01
-1.35129914e-01 -8.42521369e-01 -3.83258134e-01 3.11420441e-01
-3.40250582e-01 -2.72267908e-01 -4.94839013e-01 4.72807437e-01
1.07303090e-01 8.20668817e-01 1.19724065e-01 -3.26749682e-01
7.55809024e-02 1.47466198e-01 2.45486677e-01 -2.89943129e-01
-1.98199511e-01 -2.68357962e-01 -1.90250680e-01 -6.87706947e-01
-4.24216181e-01 -5.28292537e-01 -5.23528159e-01 -5.19433439e-01
-4.34918225e-01 1.49576545e-01 4.23415691e-01 8.63228440e-01
3.94202828e-01 4.48183984e-01 5.11796474e-01 -8.04073572e-01
-1.08759105e+00 -1.06237626e+00 -6.25884950e-01 6.55054867e-01
7.58841932e-01 -9.41767156e-01 -9.06976640e-01 -1.18146256e-01] | [8.969643592834473, 4.62490701675415] |
bf3655ef-66f0-421f-92ca-b1499ac84cf7 | adaptive-memory-management-for-video-object | 2204.06626 | null | https://arxiv.org/abs/2204.06626v1 | https://arxiv.org/pdf/2204.06626v1.pdf | Adaptive Memory Management for Video Object Segmentation | Matching-based networks have achieved state-of-the-art performance for video object segmentation (VOS) tasks by storing every-k frames in an external memory bank for future inference. Storing the intermediate frames' predictions provides the network with richer cues for segmenting an object in the current frame. However, the size of the memory bank gradually increases with the length of the video, which slows down inference speed and makes it impractical to handle arbitrary length videos. This paper proposes an adaptive memory bank strategy for matching-based networks for semi-supervised video object segmentation (VOS) that can handle videos of arbitrary length by discarding obsolete features. Features are indexed based on their importance in the segmentation of the objects in previous frames. Based on the index, we discard unimportant features to accommodate new features. We present our experiments on DAVIS 2016, DAVIS 2017, and Youtube-VOS that demonstrate that our method outperforms state-of-the-art that employ first-and-latest strategy with fixed-sized memory banks and achieves comparable performance to the every-k strategy with increasing-sized memory banks. Furthermore, experiments show that our method increases inference speed by up to 80% over the every-k and 35% over first-and-latest strategies. | ['Charalambos Poullis', 'Ali Pourganjalikhan'] | 2022-04-13 | null | null | null | null | ['semi-supervised-video-object-segmentation'] | ['computer-vision'] | [ 8.60992223e-02 -9.85014588e-02 -5.39674520e-01 -3.39528471e-01
-3.69728714e-01 -3.77230853e-01 5.78018166e-02 -7.03617856e-02
-8.43586624e-01 4.74402308e-01 -1.73781604e-01 -3.43424320e-01
1.39972866e-01 -7.68361151e-01 -9.44233954e-01 -4.54471141e-01
-1.44455373e-01 3.52956623e-01 1.08307362e+00 4.21943069e-01
1.25241712e-01 5.53031862e-01 -1.76009071e+00 5.40058911e-01
3.71767372e-01 1.28745520e+00 3.75723660e-01 7.37065196e-01
-3.09010714e-01 6.61246479e-01 -4.75102067e-01 -2.92955667e-01
5.96691370e-01 -1.75006002e-01 -1.02417612e+00 -2.32156441e-02
1.04292536e+00 -7.87167609e-01 -8.88132870e-01 8.62218559e-01
1.60928428e-01 3.33007395e-01 2.36378267e-01 -1.39157879e+00
-2.11167678e-01 4.85679775e-01 -4.03527260e-01 7.08790541e-01
1.06685631e-01 2.11181670e-01 8.18106353e-01 -6.83394670e-01
9.04586852e-01 1.22214043e+00 7.15950847e-01 6.94709003e-01
-8.07215691e-01 -7.91250765e-01 5.66206455e-01 5.26042461e-01
-1.16771662e+00 -5.10963321e-01 3.11763257e-01 -2.83421099e-01
1.20464396e+00 8.26151222e-02 8.75872195e-01 3.90375376e-01
-1.69964194e-01 1.27491224e+00 4.11551237e-01 2.78634252e-03
2.01505110e-01 -1.96611360e-01 3.12475741e-01 9.91858065e-01
1.36215687e-01 -3.14134330e-01 -7.90732563e-01 8.87717307e-02
9.03291166e-01 2.04459459e-01 -7.33263791e-02 1.18969670e-02
-1.18135834e+00 6.73979163e-01 3.04773331e-01 -6.86344877e-02
-3.16769868e-01 4.19351667e-01 5.74679494e-01 2.38244206e-01
3.71265858e-01 -4.03213799e-02 -6.60150111e-01 -2.65185267e-01
-1.56875300e+00 2.00912520e-01 6.77815199e-01 1.29063213e+00
8.03846836e-01 1.87739819e-01 -2.32854098e-01 6.05385005e-01
1.23986810e-01 2.56747216e-01 2.59035707e-01 -1.61024153e+00
5.59161901e-01 4.69356149e-01 -9.01491046e-02 -8.08442235e-01
-1.53224096e-01 1.11773700e-01 -7.11710751e-01 -1.78238943e-01
5.47793984e-01 -1.31964192e-01 -1.22618186e+00 1.54499531e+00
3.95252258e-01 7.03548372e-01 -1.46216363e-01 8.32739174e-01
1.01507640e+00 8.37158203e-01 1.66073740e-01 -3.94245535e-01
1.38139248e+00 -1.30436790e+00 -3.36183578e-01 -1.21691369e-01
3.80004674e-01 -5.17070234e-01 7.27668047e-01 2.08437935e-01
-1.31579041e+00 -8.97808015e-01 -8.55397701e-01 -8.68812352e-02
-1.74678817e-01 -1.04265109e-01 6.51925623e-01 3.52358550e-01
-1.13094246e+00 9.10436928e-01 -9.85700369e-01 -2.05824241e-01
7.70533621e-01 5.45933545e-01 -1.65575027e-01 1.09802723e-01
-9.57695067e-01 2.37896025e-01 6.07730865e-01 -1.14817163e-02
-8.78800988e-01 -9.22431707e-01 -8.16739261e-01 1.11861028e-01
6.87808871e-01 -6.19147241e-01 1.13631320e+00 -9.49331164e-01
-1.17852616e+00 5.38050830e-01 -6.26469553e-01 -8.51434648e-01
4.54797775e-01 -1.12880640e-01 -2.22348556e-01 7.10999608e-01
2.88227219e-02 1.41909325e+00 1.15090001e+00 -6.73190594e-01
-1.02156234e+00 -1.16942190e-01 3.49466920e-01 4.62711491e-02
-4.24517363e-01 -1.45507351e-01 -1.09310007e+00 -5.73718965e-01
2.77512372e-01 -1.03856385e+00 -5.05722165e-02 5.31058192e-01
-1.29491165e-01 -3.66809458e-01 1.14428902e+00 -4.97686356e-01
1.37325287e+00 -2.23222685e+00 -1.81874976e-01 -3.07350047e-02
2.64738500e-01 6.24187052e-01 -1.67280167e-01 -1.70340911e-01
3.41595978e-01 1.42928466e-01 5.87874576e-02 -3.97844106e-01
-3.33442479e-01 4.85418767e-01 -1.11294396e-01 1.95035771e-01
8.83687288e-02 7.76734829e-01 -8.14455211e-01 -1.01127326e+00
1.22109376e-01 3.07869792e-01 -6.88578248e-01 1.19381763e-01
-2.87374675e-01 7.33169690e-02 -2.40438879e-01 6.65582955e-01
4.95661139e-01 -4.03582007e-01 -1.41270766e-02 -5.06430864e-01
-5.22208139e-02 2.60860503e-01 -1.24989879e+00 1.45357966e+00
-4.77810055e-02 8.31460178e-01 -1.31233603e-01 -8.06399107e-01
5.10943651e-01 3.13332051e-01 5.18132031e-01 -4.66275960e-01
-1.52622581e-01 -3.76088843e-02 -2.51928836e-01 -4.32805181e-01
6.22240782e-01 3.92296493e-01 3.31923366e-01 2.76705056e-01
1.17794611e-01 3.97379488e-01 7.82059073e-01 4.08607543e-01
9.59078670e-01 1.24704905e-01 -2.40903080e-01 2.63571925e-02
6.48786187e-01 -1.14522405e-01 9.49506938e-01 8.77245009e-01
-4.54069674e-01 3.75286788e-01 4.18321192e-01 -8.20551813e-01
-1.03647423e+00 -8.65744412e-01 1.34118050e-02 1.28308630e+00
2.59745389e-01 -5.92728376e-01 -9.13452923e-01 -8.18379402e-01
1.98733225e-01 2.64640927e-01 -4.53959286e-01 1.47640824e-01
-8.84740233e-01 -1.85184762e-01 6.81760073e-01 1.01412773e+00
9.65891063e-01 -1.26979268e+00 -9.52100515e-01 1.95041522e-01
-3.23776633e-01 -1.51925170e+00 -7.07049370e-01 -2.52399355e-01
-1.27732718e+00 -1.17410696e+00 -7.90457368e-01 -8.17623615e-01
7.48138547e-01 3.97847593e-01 9.92156863e-01 2.74037898e-01
-2.13862732e-01 2.29159504e-01 -1.37262583e-01 -1.02752233e-02
-1.96265221e-01 1.79563075e-01 -5.66397421e-02 -2.74010181e-01
2.84506381e-01 -2.73680806e-01 -8.57151628e-01 3.32481742e-01
-8.90127301e-01 2.34479040e-01 2.32501641e-01 4.67997640e-01
8.19892824e-01 2.34662946e-02 4.22527999e-01 -8.28078628e-01
-6.72074184e-02 -3.78131330e-01 -5.95561087e-01 1.50039643e-01
-4.43436682e-01 3.99857350e-02 6.47380948e-01 -6.45965457e-01
-7.81739950e-01 6.98152632e-02 -1.71650983e-02 -8.10627759e-01
-1.11281067e-01 8.85676891e-02 3.17419678e-01 8.09149072e-03
-1.60326764e-01 2.18938828e-01 -1.48606881e-01 -4.34292078e-01
1.66209966e-01 5.27205586e-01 7.20285356e-01 -3.18435788e-01
3.33470881e-01 5.60769200e-01 6.06005499e-03 -6.72976077e-01
-8.49820375e-01 -5.74326634e-01 -7.44680345e-01 -3.80481809e-01
9.01167274e-01 -9.82538462e-01 -1.05490935e+00 6.27012074e-01
-1.10092962e+00 -3.95696491e-01 -1.65343300e-01 2.88083583e-01
-4.05613810e-01 5.08041739e-01 -1.09129107e+00 -5.26546180e-01
-4.91916478e-01 -1.25721920e+00 8.11910629e-01 4.71431226e-01
-1.63196236e-01 -6.51426077e-01 -5.43064892e-01 4.91899580e-01
2.32135028e-01 -2.17490569e-01 7.66765714e-01 -4.88906801e-01
-9.92553234e-01 1.60707369e-01 -4.49304670e-01 2.94317007e-01
-2.71929413e-01 1.97119668e-01 -5.97261190e-01 -3.77185941e-01
-4.30396438e-01 3.80791053e-02 1.15033996e+00 3.91562998e-01
1.48072159e+00 -3.55778188e-01 -3.36843491e-01 7.18189538e-01
1.27696228e+00 2.97526807e-01 6.06680989e-01 3.76765430e-01
6.91307247e-01 3.31872851e-02 7.01206326e-01 4.67921346e-01
4.55950946e-01 3.88379723e-01 4.32963878e-01 2.10606337e-01
-2.83758640e-01 -1.26279280e-01 2.44523376e-01 7.31801212e-01
-1.32680580e-01 -2.96295673e-01 -4.99671638e-01 8.29550683e-01
-1.93988252e+00 -1.00874114e+00 8.57285038e-03 2.19246769e+00
8.70804012e-01 4.48925287e-01 3.06284130e-01 -1.82309821e-02
7.95007408e-01 4.45423007e-01 -8.01431954e-01 -4.48945493e-01
1.25706017e-01 5.77245951e-02 6.52205110e-01 2.42067054e-01
-1.10033965e+00 1.18894780e+00 6.41306686e+00 8.57762337e-01
-1.01203883e+00 1.28488317e-01 7.45347619e-01 -4.21027154e-01
1.78957596e-01 -4.82178032e-02 -1.27283382e+00 7.75799453e-01
9.67493534e-01 2.54088342e-01 4.00759488e-01 9.65386689e-01
7.90602639e-02 -4.75628406e-01 -1.21063077e+00 1.04947698e+00
-4.97425273e-02 -1.73176515e+00 3.51412803e-01 -2.05720469e-01
7.87224889e-01 2.54006505e-01 -2.31830239e-01 1.27464473e-01
-2.63345718e-01 -4.11814243e-01 8.57495904e-01 4.96337533e-01
7.31758058e-01 -7.15376019e-01 6.00235999e-01 2.39875644e-01
-1.55074275e+00 -3.93303223e-02 -5.49813688e-01 -9.89688188e-03
3.33111316e-01 3.09684902e-01 -5.06182194e-01 -5.02912290e-02
1.08695579e+00 6.18661940e-01 -5.27012289e-01 9.80399132e-01
2.19884157e-01 8.12978208e-01 -5.84565163e-01 1.30967394e-01
5.49684465e-01 7.56216124e-02 3.63259494e-01 1.43137097e+00
1.63933516e-01 2.69072741e-01 3.86946678e-01 4.85656798e-01
-4.24112946e-01 -2.60764897e-01 -9.10145938e-02 1.17859021e-01
8.21621180e-01 9.98339653e-01 -1.08898306e+00 -1.05891573e+00
-5.24739504e-01 9.24606800e-01 5.18411659e-02 2.51035541e-01
-1.03004372e+00 -3.68359357e-01 6.59755945e-01 2.56265819e-01
9.20235336e-01 -2.60059446e-01 6.57170489e-02 -8.91456425e-01
3.40627730e-02 -5.46191931e-01 5.24423063e-01 -6.17585003e-01
-7.80339241e-01 5.05268633e-01 2.10474357e-01 -9.52092230e-01
-2.07847059e-01 -3.68257672e-01 -3.17354709e-01 2.67057151e-01
-1.51709294e+00 -8.26898515e-01 -3.23731035e-01 7.00168967e-01
1.00186062e+00 -7.01553077e-02 3.50719154e-01 5.24089038e-01
-5.24673820e-01 6.40030503e-01 -3.83257084e-02 5.37149608e-01
5.06683648e-01 -8.72672498e-01 5.79159677e-01 8.08989286e-01
1.63152218e-01 4.14442480e-01 3.12836856e-01 -7.90259182e-01
-1.31871235e+00 -1.09019434e+00 8.06724966e-01 1.18362144e-01
3.98660928e-01 2.68338714e-02 -9.65574324e-01 7.74463594e-01
-8.77882168e-02 5.06786346e-01 4.69450951e-01 -3.36329013e-01
-4.54183728e-01 -2.64306784e-01 -1.18446410e+00 5.66345811e-01
1.33675706e+00 -4.26835507e-01 -3.43965650e-01 1.44980431e-01
8.07510316e-01 -5.09586394e-01 -8.72708082e-01 3.42764795e-01
7.37930298e-01 -8.52919459e-01 9.56287563e-01 -6.34013891e-01
2.01300353e-01 -3.35528344e-01 5.49150072e-03 -3.84254336e-01
-2.98128515e-01 -5.93036950e-01 -7.22136497e-01 1.00179601e+00
1.37490496e-01 -3.71958077e-01 1.21389353e+00 8.36182535e-01
4.76136692e-02 -9.56573129e-01 -9.25743222e-01 -7.19101012e-01
-3.89880866e-01 -5.40938377e-01 5.55406213e-01 4.92584944e-01
-4.53867078e-01 -1.49355486e-01 -2.32838303e-01 9.66129303e-02
5.80192268e-01 2.87287146e-01 6.57207131e-01 -1.03750157e+00
-2.22774133e-01 -2.96914428e-01 -5.94123840e-01 -1.74240756e+00
2.86461234e-01 -5.16246736e-01 -4.00831737e-02 -1.25112689e+00
3.81185651e-01 -4.24981117e-01 -2.82338113e-01 7.30131507e-01
-1.42152473e-01 7.10698605e-01 6.74842179e-01 4.05561239e-01
-1.15164900e+00 -7.08156005e-02 1.13036847e+00 -9.93957818e-02
-2.26160958e-01 -4.15817574e-02 -4.08541324e-04 9.79824662e-01
6.37410879e-01 -5.20717323e-01 -3.44106674e-01 -5.56739032e-01
-1.67764444e-02 1.56948045e-01 4.07644242e-01 -1.11241579e+00
4.71440524e-01 -7.38000348e-02 5.10036469e-01 -1.05935001e+00
4.18563306e-01 -6.92022085e-01 5.47510311e-02 5.16242743e-01
-2.87775189e-01 1.20912462e-01 3.20445180e-01 5.72536767e-01
-8.92826021e-02 -4.61897165e-01 7.48303056e-01 -3.10901999e-01
-1.27630723e+00 6.27433479e-01 -5.45250833e-01 2.76834369e-01
9.15598989e-01 -7.34788537e-01 -4.19564582e-02 -2.21146092e-01
-9.04081225e-01 2.63909101e-01 3.21460187e-01 4.88199115e-01
7.37585366e-01 -1.10040736e+00 -3.49800348e-01 2.66215861e-01
-3.52648228e-01 2.44884744e-01 3.13736022e-01 6.00250363e-01
-6.78680778e-01 3.67717177e-01 -1.42154455e-01 -7.99289703e-01
-1.67181444e+00 4.89713550e-01 3.33700813e-02 9.82837635e-04
-7.05186129e-01 9.17684317e-01 -5.15352897e-02 3.18935901e-01
6.27646446e-01 -5.04515588e-01 3.07254940e-02 6.99437708e-02
5.89755654e-01 7.73341954e-01 -2.20462054e-01 -6.86140537e-01
-3.00079286e-01 5.45655251e-01 -3.10086459e-01 2.30371937e-01
1.18929040e+00 -2.44828194e-01 -6.38693273e-02 3.27285856e-01
1.29291403e+00 -6.00612640e-01 -1.74603593e+00 -3.76264751e-01
-1.83262572e-01 -6.91402376e-01 1.06006920e-01 -2.97395200e-01
-1.59238386e+00 6.32670164e-01 4.15167332e-01 -1.50053039e-01
1.10205936e+00 -1.82735603e-02 1.48431122e+00 5.88003457e-01
5.10181069e-01 -1.25442898e+00 1.28959650e-02 5.90753615e-01
1.03162274e-01 -1.03413308e+00 3.56800072e-02 -4.81782585e-01
-3.35703462e-01 1.29343772e+00 7.15312302e-01 -1.17014229e-01
6.99142456e-01 2.88165301e-01 -3.03434342e-01 1.16539709e-01
-7.51384437e-01 -5.26145883e-02 3.37590516e-01 3.00454080e-01
7.56847337e-02 -2.51994908e-01 -6.84514716e-02 1.78353965e-01
1.42577723e-01 2.35399812e-01 2.56857872e-01 9.38886762e-01
-6.33112967e-01 -7.79492736e-01 -1.74700156e-01 7.02099562e-01
-5.82026005e-01 -1.34619132e-01 1.29449770e-01 7.38193810e-01
3.60499829e-01 6.82238042e-01 6.90425873e-01 -8.64019096e-02
-7.97812566e-02 2.21455470e-01 4.98008251e-01 -2.74449795e-01
-5.23464441e-01 -1.27286557e-02 -1.11347668e-01 -8.52070510e-01
-6.51031017e-01 -6.84377611e-01 -1.63885856e+00 -6.72745228e-01
-3.92435879e-01 -1.31323740e-01 4.16192710e-01 9.34627593e-01
5.70067883e-01 3.41739804e-01 2.32911527e-01 -9.38604116e-01
-1.54932037e-01 -5.29538095e-01 -2.35450760e-01 3.24220121e-01
3.51250112e-01 -4.71503764e-01 -1.09628893e-01 3.86355728e-01] | [9.130205154418945, -0.018118364736437798] |
34df5c10-dd7c-4b21-9125-cd60f7fd1da8 | defensive-ml-defending-architectural-side | 2302.01474 | null | https://arxiv.org/abs/2302.01474v1 | https://arxiv.org/pdf/2302.01474v1.pdf | Defensive ML: Defending Architectural Side-channels with Adversarial Obfuscation | Side-channel attacks that use machine learning (ML) for signal analysis have become prominent threats to computer security, as ML models easily find patterns in signals. To address this problem, this paper explores using Adversarial Machine Learning (AML) methods as a defense at the computer architecture layer to obfuscate side channels. We call this approach Defensive ML, and the generator to obfuscate signals, defender. Defensive ML is a workflow to design, implement, train, and deploy defenders for different environments. First, we design a defender architecture given the physical characteristics and hardware constraints of the side-channel. Next, we use our DefenderGAN structure to train the defender. Finally, we apply defensive ML to thwart two side-channel attacks: one based on memory contention and the other on application power. The former uses a hardware defender with ns-level response time that attains a high level of security with half the performance impact of a traditional scheme; the latter uses a software defender with ms-level response time that provides better security than a traditional scheme with only 70% of its power overhead. | ['Josep Torrellas', 'Nam Sung Kim', 'Bo Li', 'Raghavendra Pradyumna Pothukuchi', 'Hyoungwook Nam'] | 2023-02-03 | null | null | null | null | ['computer-security'] | ['miscellaneous'] | [ 1.78170323e-01 8.64063278e-02 -1.80549785e-01 9.72253531e-02
-6.87604487e-01 -8.74497235e-01 5.23443103e-01 5.73185831e-02
-3.36899728e-01 2.83193171e-01 -3.27474564e-01 -1.21992314e+00
1.88269451e-01 -7.77421951e-01 -5.58036566e-01 -9.04844344e-01
-8.15645754e-01 -1.24667481e-01 3.94529790e-01 -5.00481606e-01
3.33724767e-01 7.66085029e-01 -8.58334541e-01 5.73776424e-01
1.40903845e-01 1.18694723e+00 -8.72571349e-01 1.09131515e+00
2.48555720e-01 7.09702909e-01 -1.33258820e+00 -3.59147489e-02
5.33020318e-01 -3.12184662e-01 -3.92778337e-01 -7.84992516e-01
-1.61928777e-02 -2.16307208e-01 -5.56818664e-01 9.59287524e-01
6.08828902e-01 -7.62620270e-01 1.37628645e-01 -1.74630356e+00
2.25009650e-01 1.21426320e+00 -6.03326023e-01 1.40096635e-01
2.72919595e-01 2.95145273e-01 4.22548622e-01 1.68343827e-01
-7.80670345e-03 1.26716220e+00 4.88583654e-01 8.84960353e-01
-1.40343130e+00 -1.45874715e+00 -8.12403560e-02 1.64370183e-02
-1.48370492e+00 -2.31887639e-01 1.02550304e+00 -2.34922231e-03
6.11238956e-01 6.88828349e-01 1.76954433e-01 1.47862518e+00
7.72077143e-01 4.90448326e-01 1.34893107e+00 -5.49315572e-01
7.75255024e-01 1.75975353e-01 7.77785927e-02 1.86886236e-01
2.75433749e-01 8.92009735e-01 -1.84974402e-01 -8.75872195e-01
1.33858353e-01 -2.19928265e-01 -2.69364148e-01 5.57769137e-03
-6.52599752e-01 9.74394739e-01 2.35637784e-01 2.01671869e-01
2.36601476e-02 5.84821880e-01 7.03235686e-01 8.27266276e-01
-3.13285887e-01 8.27715099e-01 -6.73034191e-01 3.80078852e-02
-8.43919754e-01 2.49138717e-02 1.29668367e+00 3.90846431e-01
2.23665074e-01 6.99400842e-01 1.54887438e-01 -1.97050616e-01
3.18607777e-01 8.11412334e-01 2.31225386e-01 -2.55633593e-01
1.63943917e-01 1.00261472e-01 -3.47243875e-01 -9.18479383e-01
-4.86278027e-01 -4.36992168e-01 -7.46621072e-01 8.17065895e-01
1.44909680e-01 -7.10239172e-01 -7.35729456e-01 1.54393601e+00
2.08041728e-01 5.01485407e-01 2.76797265e-01 4.86857384e-01
-1.22436322e-01 7.21009254e-01 1.36137217e-01 -1.68005392e-01
1.23753643e+00 -3.36200833e-01 -2.89885879e-01 -1.86451867e-01
6.41736805e-01 -5.26909947e-01 4.34599102e-01 7.60715187e-01
-5.28144002e-01 -2.33420163e-01 -1.79286551e+00 1.26728630e+00
-4.43232656e-01 -4.67379987e-01 6.46998525e-01 1.80571210e+00
-7.20712364e-01 3.51119637e-01 -7.29444623e-01 4.27878112e-01
-1.94002111e-02 6.49159908e-01 1.25293702e-01 7.14498401e-01
-1.43921423e+00 5.49737453e-01 3.52639467e-01 -4.36677933e-01
-1.33082688e+00 -6.67300940e-01 -6.37106597e-01 1.21329598e-01
3.43832493e-01 9.10141796e-04 9.77293730e-01 -8.32187772e-01
-1.62784195e+00 3.31246108e-01 8.88333261e-01 -9.78837252e-01
2.44359776e-01 -3.42932940e-02 -1.00478041e+00 1.02270208e-01
-6.55759692e-01 -5.17033264e-02 1.29962063e+00 -1.28593493e+00
-3.23066890e-01 -5.62406629e-02 8.56564790e-02 -7.26641476e-01
-3.76839101e-01 2.25463703e-01 4.38943982e-01 -6.30438149e-01
-1.66113719e-01 -1.18728197e+00 -4.36739862e-01 -5.53768098e-01
-7.67086387e-01 6.88227117e-01 1.30010724e+00 -2.23028868e-01
1.35958695e+00 -2.22838306e+00 -2.20192447e-01 9.68296051e-01
1.71274230e-01 6.86258614e-01 2.76490264e-02 6.32607162e-01
-3.28871667e-01 2.23999500e-01 7.58585781e-02 3.97400379e-01
4.64212745e-02 -8.45982656e-02 -8.93941164e-01 7.07876742e-01
-2.85144478e-01 3.14482450e-01 -3.15439135e-01 1.30098555e-02
-6.45167902e-02 3.20361182e-02 -6.12816989e-01 2.33530134e-01
-2.11592272e-01 3.25862467e-01 -7.14665532e-01 7.36404598e-01
7.76675403e-01 3.24791342e-01 6.67152286e-01 -4.56404090e-02
-1.13957457e-01 3.91547009e-03 -1.34241736e+00 7.04967618e-01
-5.30637085e-01 4.01220292e-01 6.93344712e-01 -9.21380520e-01
1.01802468e+00 3.70197326e-01 2.37503812e-01 -7.06772089e-01
7.35605717e-01 1.99852481e-01 5.10712028e-01 -4.26067673e-02
-2.78371155e-01 1.73569843e-01 -7.46298373e-01 9.56630409e-01
-3.73449445e-01 9.99967381e-02 -6.92154527e-01 2.63101548e-01
1.73042989e+00 -5.87759852e-01 7.51407668e-02 -2.40700498e-01
8.25626254e-01 -3.53463382e-01 3.97096276e-01 1.10542154e+00
-5.92247657e-02 -4.15462136e-01 7.03606188e-01 -5.83270371e-01
-5.32993615e-01 -1.01420808e+00 2.39032626e-01 9.38115001e-01
1.04893707e-01 -5.04515529e-01 -8.93795669e-01 -9.18771327e-01
8.69796574e-02 7.45214760e-01 -3.58146578e-01 -8.72787237e-01
-6.99229658e-01 -6.83963776e-01 1.68557835e+00 2.33982205e-01
4.96065348e-01 -7.37021506e-01 -1.06736851e+00 1.23176984e-01
5.51516891e-01 -9.92483377e-01 -1.33882314e-01 7.61511862e-01
-2.08884567e-01 -9.85636830e-01 2.91611940e-01 -2.55414397e-01
1.90308869e-01 -2.15436891e-02 6.29225552e-01 4.66554046e-01
-5.58339298e-01 1.34804234e-01 -5.21929443e-01 -6.92243218e-01
-1.11009157e+00 -2.04349503e-01 4.83365774e-01 1.55890182e-01
1.73819676e-01 -7.68945217e-01 -1.91253856e-01 3.33395422e-01
-9.86298800e-01 -6.32176518e-01 8.90840054e-01 6.55786872e-01
-5.84444068e-02 5.05618334e-01 3.50737005e-01 -7.49367476e-01
6.00132048e-01 -3.73294324e-01 -9.75773394e-01 9.68284607e-02
-5.93497336e-01 3.46982449e-01 1.07903039e+00 -9.94589031e-01
-4.13419992e-01 7.00593889e-02 -4.88841534e-01 -3.58940363e-01
-6.71892613e-02 -5.42427897e-02 -6.06020749e-01 -8.97404790e-01
1.10214221e+00 2.89922506e-01 -5.37896482e-03 -2.00728312e-01
2.25298360e-01 6.82003736e-01 2.83375174e-01 -7.38399088e-01
1.54124713e+00 4.04435784e-01 3.24771672e-01 -7.55123675e-01
-4.97417569e-01 2.00629711e-01 -7.33411461e-02 6.01964667e-02
3.48083287e-01 -4.43689317e-01 -1.21595800e+00 2.83988774e-01
-7.60388792e-01 -4.53178018e-01 3.18419874e-01 1.68426469e-01
-2.53077209e-01 3.29973042e-01 -6.88763618e-01 -9.62108135e-01
-7.40632176e-01 -1.05297947e+00 5.12392759e-01 4.44951877e-02
-5.25484830e-02 -6.44946277e-01 1.88164577e-01 -1.98497266e-01
7.97370374e-01 7.76106179e-01 1.06938446e+00 -1.22227466e+00
-3.83583605e-01 -6.30262613e-01 2.29234591e-01 1.96576759e-01
-3.03070724e-01 -1.18134141e-01 -1.05267632e+00 -7.17357397e-01
4.01775390e-01 -1.74298078e-01 3.11660528e-01 6.51164800e-02
1.25160253e+00 -7.12769270e-01 -5.14333904e-01 9.13305104e-01
1.34689236e+00 6.81575537e-01 6.93017423e-01 6.29704356e-01
2.85891116e-01 1.99472934e-01 3.09448957e-01 3.98749411e-01
-4.91767049e-01 8.02210867e-01 7.82356918e-01 -1.96030021e-01
5.00741124e-01 -6.26635402e-02 8.35885465e-01 4.45148759e-02
6.04602218e-01 -9.63936523e-02 -9.37588990e-01 -3.58946353e-01
-1.32798588e+00 -1.05549121e+00 1.17249377e-01 2.28615689e+00
6.64796770e-01 1.06122136e+00 1.72567815e-01 6.30420685e-01
2.82736510e-01 2.12759584e-01 -2.16209382e-01 -1.05995309e+00
6.65795654e-02 6.37024999e-01 1.18865728e+00 4.39376354e-01
-1.23836160e+00 8.06121051e-01 6.21129465e+00 1.05203021e+00
-1.63070953e+00 -7.66338184e-02 2.93288231e-01 2.06810281e-01
6.22424930e-02 3.60940158e-01 -7.80649662e-01 5.44473350e-01
1.64699292e+00 -3.82175073e-02 3.13061327e-01 1.12875748e+00
8.05841088e-02 3.00534397e-01 -1.02309549e+00 6.41210794e-01
6.87201926e-03 -1.29202127e+00 -2.11251557e-01 3.06020260e-01
-7.85257742e-02 -3.25426966e-01 9.81611609e-02 4.53387946e-01
4.06590194e-01 -1.08512771e+00 5.24656951e-01 -2.28896767e-01
5.97971737e-01 -1.22385788e+00 8.67513418e-01 5.50908148e-01
-9.67122257e-01 -4.16205704e-01 7.59465620e-02 -1.56790644e-01
-2.73425644e-03 8.67354125e-02 -7.61199176e-01 3.19259584e-01
4.41485703e-01 -6.02229416e-01 -5.99012256e-01 5.50131977e-01
-4.24128652e-01 1.12680995e+00 -3.17566305e-01 -1.73711404e-01
2.06172332e-01 5.31597018e-01 6.70293689e-01 1.37702763e+00
-1.68434963e-01 3.97843979e-02 6.57237649e-01 2.83402950e-01
3.66673201e-01 -4.64469582e-01 -4.97512400e-01 1.89033166e-01
9.53438640e-01 1.27046168e+00 -5.29088616e-01 7.52351135e-02
-6.11111969e-02 6.07971430e-01 -4.40493375e-01 -1.01976134e-02
-1.18748832e+00 -8.81411314e-01 7.77421176e-01 1.39333084e-01
-1.01312332e-01 -2.19812199e-01 -2.07206786e-01 -7.00828016e-01
-5.89170694e-01 -1.57816410e+00 4.58679587e-01 1.81761146e-01
-9.62292314e-01 9.19890821e-01 2.15655714e-02 -1.21962273e+00
-3.02979112e-01 -5.55065751e-01 -9.91553485e-01 5.44235468e-01
-9.80769515e-01 -1.23919594e+00 2.30808869e-01 8.15807462e-01
2.37452492e-01 -5.98444164e-01 9.12764490e-01 4.17499393e-02
-5.33631980e-01 1.14119387e+00 -2.55980194e-01 4.29915071e-01
5.08615077e-01 -9.37770724e-01 6.08414233e-01 9.87157941e-01
1.45954797e-02 6.51586413e-01 1.12831175e+00 -5.10392308e-01
-1.98797584e+00 -7.82283247e-01 -4.88170044e-04 -1.91957086e-01
9.83014405e-01 -7.73275554e-01 -7.09349751e-01 4.25798029e-01
5.68262227e-02 -1.15700357e-01 9.48580801e-01 -1.49953365e-01
-9.13366914e-01 -1.88069373e-01 -1.34908080e+00 7.76907623e-01
1.31356027e-02 -3.64618599e-01 -2.44694933e-01 1.02018584e-02
7.33780921e-01 -2.38921687e-01 -4.04032528e-01 1.19378217e-01
5.98034501e-01 -8.54470015e-01 9.60054755e-01 -5.21916389e-01
-4.96797562e-01 -4.01574075e-01 -2.93849915e-01 -1.08857334e+00
-1.58594951e-01 -1.21749616e+00 -4.04631048e-01 9.24941063e-01
4.94248986e-01 -6.78889930e-01 9.27301109e-01 1.03122436e-01
1.80293158e-01 -4.85888422e-01 -1.02796602e+00 -1.16822803e+00
1.03231616e-01 -5.38356364e-01 5.78647256e-01 8.88246834e-01
2.98770428e-01 2.25390300e-01 -6.88461721e-01 7.79588878e-01
1.00774181e+00 -1.24637084e-02 9.93993461e-01 -7.51004398e-01
-1.00416243e+00 -3.18765819e-01 -5.23031592e-01 -4.77883846e-01
1.68648869e-01 -5.48226893e-01 -1.72318861e-01 -6.16111904e-02
-4.28570360e-01 -4.53531116e-01 -3.36847246e-01 5.24692297e-01
4.21271145e-01 -9.52740461e-02 1.23017639e-01 -1.03424557e-01
-3.99335980e-01 -3.26418877e-02 1.08560897e-01 -3.92195314e-01
-4.07205284e-01 4.18775618e-01 -7.03501463e-01 7.78029323e-01
1.09418178e+00 -8.27480197e-01 -2.50185966e-01 2.80498773e-01
-2.49371126e-01 3.27416271e-01 2.15342775e-01 -1.29554296e+00
2.55495042e-01 -3.47195774e-01 3.99786592e-01 -4.11301814e-02
1.47751942e-01 -1.10422945e+00 4.26319428e-02 1.47264063e+00
-1.96721867e-01 2.02293292e-01 5.54336667e-01 4.96405929e-01
1.60032049e-01 7.85121024e-02 1.14028811e+00 3.46755803e-01
-4.84003186e-01 2.87187868e-03 -8.72900546e-01 -3.79764318e-01
1.37640941e+00 2.56025232e-03 -6.41752303e-01 -4.00561243e-01
-3.92713279e-01 5.27267754e-02 2.58647680e-01 3.72971982e-01
6.57800734e-01 -8.79944086e-01 -5.21738827e-01 5.64895868e-01
-1.01836681e-01 -1.14915383e+00 8.73205140e-02 4.33530599e-01
-5.58588386e-01 1.96338505e-01 -4.43824410e-01 -1.81876540e-01
-1.61656952e+00 1.09704506e+00 5.02455652e-01 -2.96482623e-01
-4.96675521e-01 6.89615607e-01 -1.87684387e-01 1.40657714e-02
4.31194872e-01 6.06957197e-01 -5.89623936e-02 -5.77488840e-01
9.94735956e-01 1.22503407e-01 1.94215158e-03 -1.81800112e-01
-6.89904392e-01 2.21641183e-01 -6.03799522e-02 -1.55531049e-01
7.52522767e-01 5.33668876e-01 3.06091551e-02 -1.94324538e-01
1.05172670e+00 5.82616746e-01 -7.88421035e-01 1.98035717e-01
3.94418567e-01 -3.80452067e-01 1.84122324e-01 -9.09840524e-01
-9.16777611e-01 6.59107804e-01 9.38566208e-01 7.20196366e-01
1.29307199e+00 -4.73858356e-01 7.93766856e-01 3.11453044e-01
7.15872288e-01 -7.91855574e-01 2.34060943e-01 2.84985840e-01
2.77154058e-01 -5.92171192e-01 -8.20452198e-02 -3.68126005e-01
-4.67317730e-01 1.24586523e+00 6.76637352e-01 -4.06061321e-01
9.16779339e-01 1.24481320e+00 9.28478539e-02 -1.05434999e-01
-8.26578140e-01 3.62293750e-01 -1.47534683e-01 8.38455737e-01
-2.13478610e-01 2.39042684e-01 -7.75998831e-02 6.98849618e-01
-5.26064746e-02 -7.98491001e-01 7.33782232e-01 1.25769281e+00
-6.98651016e-01 -1.65160513e+00 -8.29213142e-01 2.74084434e-02
-8.44477415e-01 6.81939796e-02 -7.58485734e-01 6.87343001e-01
-6.12111688e-02 1.29280436e+00 -4.69016135e-01 -1.28011227e+00
2.17424348e-01 -3.18768710e-01 2.65373345e-02 -4.03730124e-01
-8.91767800e-01 1.32339999e-01 1.84056405e-02 -8.39133799e-01
4.99107122e-01 2.03467775e-02 -1.01948261e+00 -5.31102121e-01
-1.75672144e-01 5.73424160e-01 8.52071702e-01 7.31126249e-01
2.83208549e-01 6.29698277e-01 1.33837497e+00 -5.65101087e-01
-1.13694382e+00 -3.92688304e-01 -6.63372099e-01 -2.18580693e-01
3.95959646e-01 -5.40615380e-01 -7.57220566e-01 -6.06441200e-01] | [5.565083026885986, 7.519000053405762] |
46c15249-99c0-4883-b9c1-73cde9b6b61b | task-adaptive-feature-transformation-for-one | 2304.06832 | null | https://arxiv.org/abs/2304.06832v1 | https://arxiv.org/pdf/2304.06832v1.pdf | Task Adaptive Feature Transformation for One-Shot Learning | We introduce a simple non-linear embedding adaptation layer, which is fine-tuned on top of fixed pre-trained features for one-shot tasks, improving significantly transductive entropy-based inference for low-shot regimes. Our norm-induced transformation could be understood as a re-parametrization of the feature space to disentangle the representations of different classes in a task specific manner. It focuses on the relevant feature dimensions while hindering the effects of non-relevant dimensions that may cause overfitting in a one-shot setting. We also provide an interpretation of our proposed feature transformation in the basic case of few-shot inference with K-means clustering. Furthermore, we give an interesting bound-optimization link between K-means and entropy minimization. This emphasizes why our feature transformation is useful in the context of entropy minimization. We report comprehensive experiments, which show consistent improvements over a variety of one-shot benchmarks, outperforming recent state-of-the-art methods. | ['Ismail Ben Ayed', 'Freddy Lecue', 'Imtiaz Masud Ziko'] | 2023-04-13 | null | null | null | null | ['one-shot-learning'] | ['methodology'] | [ 2.13349849e-01 2.26166323e-01 -4.74647999e-01 -4.71567512e-01
-8.61384153e-01 -3.07548344e-01 9.54622388e-01 1.23995520e-01
-5.74686527e-01 6.77812874e-01 6.43218994e-01 2.02208653e-01
-3.79088074e-01 -8.29355121e-01 -5.69495559e-01 -1.05296230e+00
5.45331948e-02 4.54988182e-01 -2.67561413e-02 -3.22612911e-01
1.57293037e-01 2.09562406e-01 -1.60118496e+00 1.32983088e-01
5.42137980e-01 7.44286597e-01 -2.80311882e-01 7.34422982e-01
2.00364605e-01 4.49726641e-01 -2.70633906e-01 -4.98943835e-01
2.16410235e-01 -5.66209197e-01 -7.93708503e-01 1.27081070e-02
2.22237706e-01 -1.01775229e-01 -5.52370369e-01 7.87186742e-01
3.99390936e-01 6.60851598e-01 1.21511388e+00 -1.07314193e+00
-8.82175684e-01 3.47381204e-01 -1.94002286e-01 3.59864414e-01
1.21159010e-01 2.64017969e-01 1.66694403e+00 -9.22312081e-01
5.94308197e-01 1.21044016e+00 5.36103606e-01 6.69912815e-01
-1.51768780e+00 -2.37033889e-01 -9.05399397e-02 3.30998570e-01
-1.19746685e+00 -5.99768162e-01 6.97381914e-01 -5.83651960e-01
1.20600116e+00 3.20426345e-01 5.29566884e-01 1.30544031e+00
2.40950689e-01 6.78781033e-01 8.13154876e-01 -4.68278378e-01
5.31814873e-01 2.45337784e-01 5.09045660e-01 6.21815026e-01
3.05755526e-01 1.90171644e-01 -4.91875619e-01 -2.66488016e-01
2.93527186e-01 2.27538124e-01 -7.66069219e-02 -7.13872969e-01
-9.90841925e-01 1.25771391e+00 3.44091952e-01 4.42668200e-01
-6.95218295e-02 1.47472277e-01 5.30355513e-01 3.36806118e-01
6.55083954e-01 6.95637465e-01 -4.06032652e-01 -3.45541984e-01
-8.75213444e-01 2.07853779e-01 9.75385308e-01 7.00686395e-01
1.08920228e+00 -8.14960748e-02 -7.03999579e-01 8.29863906e-01
7.12179467e-02 1.94916666e-01 2.39909753e-01 -9.76419568e-01
1.59870923e-01 4.67659622e-01 -3.65946032e-02 -3.97517115e-01
-2.92337120e-01 -2.81874686e-01 -7.68875599e-01 -1.51279256e-01
4.17809308e-01 -1.41025215e-01 -7.97256052e-01 1.98526931e+00
1.45608515e-01 2.05699116e-01 9.39922109e-02 7.41682649e-01
5.00532925e-01 6.35452747e-01 -1.79915279e-02 -3.26310933e-01
1.33934188e+00 -1.03329372e+00 -6.68182969e-01 -3.66850317e-01
6.97570622e-01 -2.96405703e-01 1.30735171e+00 -5.69238327e-02
-9.21968520e-01 -1.35726959e-01 -9.98107135e-01 -4.22510147e-01
-7.14445472e-01 -3.58846158e-01 7.96384871e-01 4.24932212e-01
-7.03092754e-01 8.64060760e-01 -7.12523699e-01 -6.52628124e-01
4.51818347e-01 1.09075576e-01 -2.62105733e-01 1.17884889e-01
-1.39662051e+00 1.00457120e+00 3.36345911e-01 -3.16731423e-01
-8.50630462e-01 -8.15209866e-01 -9.52393353e-01 4.18560922e-01
6.06902480e-01 -8.43419433e-01 9.92614210e-01 -2.35724345e-01
-1.61896670e+00 8.52477014e-01 -2.60217637e-01 -3.39619964e-01
2.49783382e-01 -1.78367227e-01 -1.34133860e-01 -9.03360397e-02
-1.96864456e-01 3.28464031e-01 8.95830274e-01 -8.04346979e-01
-8.58671218e-02 -2.13988617e-01 1.26695573e-01 2.60120332e-01
-5.82968116e-01 -1.05739027e-01 -1.91452160e-01 -4.37602192e-01
-4.32954133e-01 -9.24956024e-01 -9.02191252e-02 -2.00480539e-02
-3.25666159e-01 -2.43918285e-01 5.56720257e-01 -8.94818306e-02
1.37247550e+00 -2.02300715e+00 5.33992887e-01 -1.17361248e-01
2.64497012e-01 1.62070036e-01 -1.61127806e-01 7.02035427e-01
5.52229211e-03 1.75704241e-01 -3.73858243e-01 -4.93939996e-01
4.03586626e-01 4.68259186e-01 -2.60542154e-01 4.90046442e-01
4.67003256e-01 1.20205438e+00 -1.06855512e+00 -3.65886062e-01
3.56470764e-01 4.81092751e-01 -7.55606830e-01 1.93943024e-01
-1.46764308e-01 1.02081850e-01 -3.58825773e-01 2.95002162e-01
2.72161186e-01 -3.54776174e-01 -4.31172810e-02 -1.03329599e-01
1.45573661e-01 4.05367285e-01 -6.67365730e-01 1.83559215e+00
-5.60199678e-01 6.65387809e-01 -3.42286348e-01 -1.19782555e+00
4.60811615e-01 1.14855990e-01 4.25147921e-01 -2.37688646e-01
3.03903371e-01 -1.68803230e-01 6.79550916e-02 -5.83367407e-01
5.30954659e-01 -4.49770331e-01 -4.17491615e-01 5.50689936e-01
6.78750694e-01 -1.74885258e-01 1.79878429e-01 2.77193040e-01
1.10564697e+00 -2.01728806e-01 7.56836534e-01 -4.69332725e-01
1.73378259e-01 -6.17733240e-01 4.02911127e-01 9.47834313e-01
-2.15331852e-01 5.38029790e-01 6.15396321e-01 -3.29773933e-01
-1.12017870e+00 -1.14344692e+00 -2.99658328e-01 1.50282764e+00
-8.24488848e-02 -6.73500001e-01 -5.51433384e-01 -7.37285614e-01
7.07086772e-02 1.09207869e+00 -1.17448866e+00 -7.52089024e-01
-8.26432481e-02 -1.16638696e+00 4.08487171e-01 3.11591804e-01
-2.80329641e-02 -8.11726689e-01 -5.65631270e-01 2.26940326e-02
-3.71001363e-02 -8.56822968e-01 -4.61354792e-01 6.89401567e-01
-6.99884057e-01 -8.23036373e-01 -3.79533559e-01 -2.82385677e-01
2.50522882e-01 2.21990887e-02 1.10846364e+00 -2.71921962e-01
-3.32280308e-01 3.98298383e-01 -1.63224563e-01 -2.52681315e-01
-2.94089347e-01 2.95312434e-01 1.29894167e-01 -6.13803882e-03
8.97951186e-01 -7.47400463e-01 -5.11879683e-01 4.48516421e-02
-8.39194059e-01 -3.55358899e-01 3.26760828e-01 1.10640204e+00
2.41578311e-01 -4.49747086e-01 5.09192884e-01 -8.60387385e-01
7.62320876e-01 -7.75435090e-01 2.08506966e-03 5.06554604e-01
-6.37218893e-01 7.28966296e-01 6.32524073e-01 -4.66446638e-01
-8.10443461e-01 -2.94451743e-01 1.53169394e-01 -4.47752386e-01
1.05113886e-01 1.02648795e-01 -5.71417361e-02 4.82155345e-02
8.79247546e-01 2.41703734e-01 -1.35258213e-01 -3.89691442e-01
7.01367199e-01 6.66369140e-01 1.51756749e-01 -6.41363502e-01
7.35641181e-01 3.98111850e-01 8.26795325e-02 -9.10737932e-01
-1.21987951e+00 -5.86970091e-01 -6.96344435e-01 1.18246213e-01
9.02571678e-01 -7.16252685e-01 -6.04381680e-01 8.52321312e-02
-9.98318970e-01 -2.48628110e-01 -8.34762871e-01 4.11982715e-01
-9.05287266e-01 1.02112241e-01 -5.77228427e-01 -8.48816454e-01
-3.42532963e-01 -9.04294908e-01 1.05918026e+00 1.51383221e-01
-3.61814708e-01 -1.35220349e+00 5.92594028e-01 1.18720777e-01
3.96614969e-01 -6.63366839e-02 9.85473454e-01 -1.13729954e+00
-2.34085321e-01 -2.52759725e-01 6.52871327e-03 3.35502237e-01
2.37360056e-02 6.20180257e-02 -1.33182204e+00 -2.77849883e-01
4.69180904e-02 -6.80730641e-01 1.45904255e+00 2.30190471e-01
9.56691027e-01 -3.26596618e-01 -1.36170536e-01 7.51270533e-01
1.34121287e+00 -3.36748928e-01 6.19286120e-01 -3.01535148e-02
6.15033627e-01 3.26482743e-01 2.34969854e-01 5.69046855e-01
3.36649954e-01 7.04547942e-01 1.04732409e-01 1.93442553e-01
1.52887285e-01 -2.44558409e-01 1.66027755e-01 8.82850468e-01
-1.61948845e-01 -1.82202592e-01 -5.47670066e-01 4.92711902e-01
-1.98479664e+00 -1.14481866e+00 5.24733245e-01 2.24286342e+00
9.42178369e-01 1.23827457e-01 1.82150006e-01 -1.98437218e-02
3.59364271e-01 6.45575881e-01 -7.21745312e-01 -7.52185583e-01
9.01172385e-02 3.41818839e-01 3.53797138e-01 7.54133403e-01
-1.07855225e+00 9.19073403e-01 7.20859718e+00 1.08910739e+00
-7.55613387e-01 2.86915123e-01 4.61138785e-01 -5.46662748e-01
-5.32082081e-01 1.28506437e-01 -7.79094458e-01 4.59538639e-01
9.80438530e-01 -5.56949139e-01 6.54593706e-01 8.26045275e-01
-5.26376590e-02 2.14190152e-03 -1.54421687e+00 8.43382537e-01
1.52241841e-01 -1.36823416e+00 -1.53616846e-01 1.52468473e-01
7.13141441e-01 1.31396174e-01 -6.24169931e-02 7.09618688e-01
2.83401757e-01 -9.66401637e-01 2.65354931e-01 7.06964672e-01
6.90188050e-01 -5.68155587e-01 3.82729352e-01 4.17410046e-01
-9.05947685e-01 -7.66871497e-02 -5.66162229e-01 -3.33658010e-01
1.07470565e-02 6.78208709e-01 -5.29324591e-01 3.00107598e-01
2.69752711e-01 7.39806175e-01 -3.59234214e-01 5.80717564e-01
-8.20134804e-02 4.72778678e-01 -2.91706681e-01 -2.84465820e-01
2.02519596e-01 -3.27332675e-01 7.06755877e-01 1.53120077e+00
2.94166785e-02 1.13591561e-02 -1.33610219e-01 1.12987733e+00
-1.99841753e-01 -7.17079565e-02 -8.97936285e-01 -1.70917749e-01
4.06327665e-01 1.28266287e+00 -2.38713667e-01 -3.99021447e-01
-5.17004550e-01 9.14357483e-01 6.94851875e-01 3.57359588e-01
-8.85439277e-01 -6.01549268e-01 9.96791303e-01 -8.20201337e-02
4.22607541e-01 3.40390876e-02 -1.29656345e-01 -1.53976786e+00
-1.65668562e-01 -2.98227072e-01 3.77390891e-01 -2.65806645e-01
-1.52501512e+00 1.41996115e-01 1.88418627e-01 -9.90928173e-01
-5.48334002e-01 -4.28797275e-01 -8.63833189e-01 5.51040828e-01
-1.38117480e+00 -7.40774989e-01 2.13712931e-01 3.19971830e-01
5.02823949e-01 1.07401970e-03 1.19383156e+00 -6.19461723e-02
-8.08897614e-01 8.45361531e-01 5.81405163e-01 -1.99974105e-01
6.43583119e-01 -1.23981380e+00 3.37700903e-01 6.71690464e-01
1.89727202e-01 6.97503567e-01 8.47644746e-01 -1.92661777e-01
-1.34066606e+00 -9.37477112e-01 9.49255168e-01 -5.96336067e-01
8.81576240e-01 -8.57709289e-01 -9.67191279e-01 6.31743312e-01
-4.52069864e-02 2.95939356e-01 1.04269397e+00 4.80368406e-01
-6.28667474e-01 -6.89870119e-03 -1.07364237e+00 5.52503645e-01
1.21528172e+00 -8.76923084e-01 -7.88281739e-01 3.84349704e-01
7.28364885e-01 1.46645397e-01 -9.09723699e-01 8.99299532e-02
7.51314759e-01 -8.78962517e-01 1.13075387e+00 -1.12266743e+00
6.21973932e-01 3.81054044e-01 -2.63634294e-01 -1.52711546e+00
-6.70223951e-01 -6.92298949e-01 -6.57257378e-01 8.50422204e-01
3.63026202e-01 -4.89852518e-01 5.05075157e-01 8.23298633e-01
2.09577248e-01 -1.01166439e+00 -9.64104712e-01 -9.05288517e-01
4.88149017e-01 -2.92992890e-01 2.78150320e-01 8.79696667e-01
5.41002214e-01 7.55138874e-01 -6.85472608e-01 -5.01459360e-01
7.12825239e-01 -1.42098228e-02 4.97447580e-01 -1.26332486e+00
-4.90874887e-01 -3.58509213e-01 -5.16338229e-01 -8.21419358e-01
3.41300160e-01 -1.12960696e+00 1.09523073e-01 -1.07027566e+00
6.78211629e-01 1.19187757e-01 -5.80627739e-01 3.78220767e-01
-3.85563761e-01 3.78241166e-02 2.13155895e-01 5.11168428e-02
-8.74675274e-01 8.81562531e-01 1.07294273e+00 -4.18115631e-02
-1.73277676e-01 -1.90029979e-01 -8.69299889e-01 5.35550892e-01
7.53276289e-01 -6.97652519e-01 -4.62698787e-01 -8.17140266e-02
2.23600581e-01 -3.33584607e-01 2.13781655e-01 -7.42568970e-01
1.18426550e-02 -4.70944554e-01 1.10232279e-01 4.60215434e-02
7.23537683e-01 -4.93074954e-01 -3.12106729e-01 1.56497851e-01
-7.14552283e-01 -7.78027415e-01 -5.15176989e-02 7.57963121e-01
1.73682287e-01 -3.16132247e-01 9.39260304e-01 -1.77658439e-01
-2.86696732e-01 3.34413737e-01 -2.79307276e-01 6.25734746e-01
1.02235210e+00 -1.60062686e-01 -4.45512503e-01 -4.66498792e-01
-6.98533893e-01 1.14756927e-01 5.27178288e-01 2.81194568e-01
3.35895956e-01 -1.44776702e+00 -6.37085021e-01 2.13374466e-01
3.60736609e-01 -4.05417502e-01 1.47527844e-01 8.60544622e-01
2.42244482e-01 4.86998856e-01 -4.46245521e-02 -3.32158625e-01
-7.90973485e-01 7.13089287e-01 1.72783405e-01 -4.94564176e-01
-4.92715240e-01 6.46995306e-01 3.51486742e-01 -3.73576492e-01
1.36416063e-01 -3.05496067e-01 8.18349868e-02 2.05481142e-01
5.74168324e-01 5.41912615e-01 -9.73808616e-02 -3.90113682e-01
-4.19311225e-01 5.15157223e-01 -2.25110501e-01 -1.28646120e-01
1.38361061e+00 -1.22703493e-01 1.35497645e-01 1.03391993e+00
1.76976073e+00 -3.61257523e-01 -1.35910034e+00 -5.01351595e-01
-1.59340531e-01 -2.55943716e-01 7.41645172e-02 -5.26365101e-01
-5.88340044e-01 1.10444999e+00 2.08798543e-01 2.62351066e-01
7.13584423e-01 2.84412742e-01 5.69806397e-01 8.20466697e-01
-1.99903622e-02 -1.29001832e+00 1.27236813e-01 7.66062081e-01
6.34038806e-01 -1.39767921e+00 1.03922300e-01 -1.87466643e-03
-6.41583562e-01 1.09112656e+00 2.06281766e-01 -2.37264201e-01
8.03884983e-01 2.01635718e-01 -4.88182187e-01 -2.92546630e-01
-1.14804518e+00 -3.50347042e-01 4.72788244e-01 4.07144248e-01
3.00149888e-01 1.61161691e-01 -3.64180475e-01 7.18019843e-01
-5.43232895e-02 5.37283868e-02 2.97079623e-01 7.86965191e-01
-6.98415339e-01 -8.37617993e-01 2.15207219e-01 7.76423872e-01
-2.31974721e-01 -2.20242038e-01 -4.48173881e-01 7.19484031e-01
-1.13972753e-01 6.03075087e-01 2.53386527e-01 -5.72698295e-01
2.33856812e-02 6.48173869e-01 5.71872592e-01 -7.62832344e-01
-3.28355551e-01 -2.59334236e-01 3.19932252e-02 -5.15347064e-01
-2.51221150e-01 -5.47874033e-01 -8.01778376e-01 -3.18048477e-01
-4.32013839e-01 1.27432700e-02 3.53747904e-01 1.17371249e+00
4.30657536e-01 3.78332555e-01 6.28557265e-01 -9.62261438e-01
-9.66654241e-01 -1.20834267e+00 -6.35892510e-01 7.04226136e-01
4.07813311e-01 -8.50753307e-01 -8.79149616e-01 -2.37306520e-01] | [9.469466209411621, 3.419149398803711] |
caf53489-76e2-4b81-a9dd-b4cb9783a691 | unsupervised-online-video-object-segmentation | 1810.03783 | null | https://arxiv.org/abs/1810.03783v2 | https://arxiv.org/pdf/1810.03783v2.pdf | Unsupervised Online Video Object Segmentation with Motion Property Understanding | Unsupervised video object segmentation aims to automatically segment moving objects over an unconstrained video without any user annotation. So far, only few unsupervised online methods have been reported in literature and their performance is still far from satisfactory, because the complementary information from future frames cannot be processed under online setting. To solve this challenging problem, in this paper, we propose a novel Unsupervised Online Video Object Segmentation (UOVOS) framework by construing the motion property to mean moving in concurrence with a generic object for segmented regions. By incorporating salient motion detection and object proposal, a pixel-wise fusion strategy is developed to effectively remove detection noise such as dynamic background and stationary objects. Furthermore, by leveraging the obtained segmentation from immediately preceding frames, a forward propagation algorithm is employed to deal with unreliable motion detection and object proposals. Experimental results on several benchmark datasets demonstrate the efficacy of the proposed method. Compared to the state-of-the-art unsupervised online segmentation algorithms, the proposed method achieves an absolute gain of 6.2%. Moreover, our method achieves better performance than the best unsupervised offline algorithm on the DAVIS-2016 benchmark dataset. Our code is available on the project website: https://github.com/visiontao/uovos. | ['Mohan Kankanhalli', 'Yongkang Wong', 'Zhiyong Cheng', 'Tao Zhuo', 'Peng Zhang'] | 2018-10-09 | unsupervised-online-video-object-segmentation-1 | null | null | ieee-transactions-on-image-processing-2019-8 | ['motion-detection', 'unsupervised-video-object-segmentation'] | ['computer-vision', 'computer-vision'] | [ 3.49354893e-01 -1.44870073e-01 -3.06458235e-01 -1.89056858e-01
-6.79790497e-01 -4.48565066e-01 2.82300234e-01 1.01439551e-01
-6.60074890e-01 5.39507568e-01 -2.94295460e-01 1.30677493e-02
1.93752632e-01 -4.39884514e-01 -6.36784852e-01 -8.54289412e-01
4.54072095e-02 8.97807851e-02 8.97407413e-01 3.16411078e-01
2.58943975e-01 1.69621781e-01 -1.42800295e+00 -6.13797791e-02
1.03070724e+00 1.11562419e+00 5.26100576e-01 5.26728153e-01
-2.01013125e-02 8.42717767e-01 -2.64416248e-01 -5.46171069e-02
4.22879755e-01 -3.72073561e-01 -7.53549814e-01 7.84689128e-01
3.52702886e-01 -6.78112507e-01 -3.76973331e-01 1.38555729e+00
2.87067890e-01 4.52396721e-01 3.19044411e-01 -1.00150096e+00
-3.57180595e-01 5.81356585e-01 -8.08917224e-01 6.72000229e-01
1.29611611e-01 3.48894328e-01 8.60264182e-01 -1.07297766e+00
6.93561792e-01 8.34513664e-01 2.24929512e-01 3.67738277e-01
-9.82406914e-01 -3.35677445e-01 6.51738167e-01 4.32958841e-01
-1.54929328e+00 -5.61497152e-01 9.25113559e-01 -4.51154679e-01
5.21989048e-01 1.17478460e-01 5.55321932e-01 6.35052085e-01
-2.33203843e-01 1.35422075e+00 8.07083905e-01 -1.80421561e-01
2.14554593e-01 -2.23510377e-02 1.41342685e-01 6.32336736e-01
2.40854248e-01 -1.43960357e-01 -2.36306667e-01 2.33781084e-01
5.98087668e-01 1.57048270e-01 -4.21962798e-01 -3.78134131e-01
-1.26023829e+00 3.53437394e-01 3.17854613e-01 3.08717757e-01
-5.35217047e-01 1.54743902e-02 3.53686929e-01 -1.41109601e-01
5.29726446e-01 -3.01879942e-01 -3.15889269e-01 -3.14428918e-02
-1.36360919e+00 1.84022856e-03 4.55207407e-01 1.11732352e+00
7.24681497e-01 1.48488879e-01 -3.18432935e-02 6.25583112e-01
3.27319115e-01 3.24524403e-01 4.51854080e-01 -9.96985793e-01
5.21201551e-01 5.58390856e-01 3.35966080e-01 -1.04127586e+00
-2.08791539e-01 -3.40391666e-01 -7.04413295e-01 -1.65022254e-01
5.03828228e-01 -2.05188677e-01 -1.00353348e+00 1.16102386e+00
7.57078886e-01 6.10645175e-01 1.61575042e-02 1.31690276e+00
8.85879755e-01 8.93433452e-01 7.27754831e-02 -6.02512360e-01
1.02271676e+00 -1.22772789e+00 -8.69448066e-01 -3.09928030e-01
3.17363799e-01 -7.50887215e-01 4.19916511e-01 3.89026225e-01
-1.24025095e+00 -6.68329120e-01 -9.12363410e-01 1.26065150e-01
-5.61132692e-02 1.84996128e-01 4.93038803e-01 4.77507085e-01
-8.14785182e-01 2.96786338e-01 -1.14241087e+00 -2.60986209e-01
7.47167706e-01 4.12730366e-01 -9.68540385e-02 -1.91003904e-02
-8.13144505e-01 2.56553024e-01 1.02121103e+00 5.91725290e-01
-9.79012132e-01 -3.23898464e-01 -7.35766411e-01 -2.45998859e-01
1.03702736e+00 -5.13188958e-01 1.29010963e+00 -1.33856297e+00
-1.37478340e+00 4.73415583e-01 -2.98800468e-01 -7.04033971e-01
8.40852082e-01 -3.34709555e-01 -1.49116337e-01 5.82024753e-01
6.90554678e-02 7.43704855e-01 9.58499610e-01 -1.23681355e+00
-1.04646993e+00 -2.09789559e-01 -1.52503308e-02 3.26091558e-01
-3.63833547e-01 1.33680061e-01 -1.06284869e+00 -8.09320271e-01
2.66996235e-01 -8.67252052e-01 -3.87725353e-01 -1.19641535e-01
-4.23813075e-01 -2.19546929e-01 1.12044406e+00 -7.84361959e-01
1.49316061e+00 -2.09784389e+00 1.03617638e-01 8.67166556e-03
1.04870901e-01 5.34142971e-01 2.25380987e-01 -8.44603218e-03
3.46017182e-01 -1.12886094e-01 -6.70101345e-01 -2.03698426e-01
-2.74604231e-01 -6.58533443e-03 4.17426042e-02 6.89694345e-01
1.76146790e-01 7.49539793e-01 -9.52893853e-01 -9.81031835e-01
4.99624133e-01 2.43187532e-01 -4.65543449e-01 6.73737600e-02
-2.52698034e-01 6.77455306e-01 -7.39969909e-01 9.87349927e-01
8.66227269e-01 -2.58853853e-01 1.20936543e-01 3.00410967e-02
-2.32594326e-01 -2.87401170e-01 -1.40888822e+00 1.74207103e+00
1.73864752e-01 5.69401264e-01 1.61012784e-01 -1.23181057e+00
5.74587643e-01 2.94071585e-01 8.33537042e-01 -3.57418954e-01
3.79107267e-01 1.63579255e-01 -3.91339734e-02 -6.85979664e-01
6.37905300e-01 2.53163308e-01 2.74982989e-01 3.88846770e-02
1.92754867e-03 3.20097446e-01 6.45321786e-01 2.35055134e-01
8.27599764e-01 4.38050449e-01 2.10688666e-01 -4.64617945e-02
8.19899797e-01 8.81258324e-02 9.60928738e-01 7.67513692e-01
-7.11345613e-01 7.09434688e-01 -6.82148114e-02 -1.88856050e-01
-7.16965795e-01 -8.54316890e-01 -3.89998108e-02 8.41211081e-01
6.69794202e-01 -1.72384307e-01 -9.91967142e-01 -7.27983952e-01
-3.20608109e-01 3.10733467e-01 -2.09368855e-01 2.89488822e-01
-6.77930117e-01 -7.58047640e-01 1.18519232e-01 6.02608502e-01
7.81961024e-01 -1.01294971e+00 -7.67653704e-01 4.84959006e-01
-5.13015747e-01 -1.62132454e+00 -6.53536022e-01 -3.84038985e-01
-1.08494735e+00 -9.79002535e-01 -9.22202647e-01 -1.03977787e+00
8.37361455e-01 7.18765438e-01 6.55397236e-01 1.89635739e-01
-3.03598523e-01 2.69395620e-01 -5.23999512e-01 -1.26520216e-01
-1.43450931e-01 -6.99529126e-02 -2.38156598e-02 5.42198002e-01
1.95731059e-01 -1.92773893e-01 -1.04412675e+00 3.74370992e-01
-1.09215415e+00 1.69017047e-01 5.62244415e-01 5.49686074e-01
8.23534310e-01 3.20185512e-01 5.22642434e-01 -7.82321155e-01
-2.13173568e-01 -5.16934097e-01 -8.84974778e-01 -9.62497108e-03
-3.69857192e-01 -4.37701344e-01 4.94475335e-01 -3.98019075e-01
-1.30389440e+00 5.61916053e-01 9.51571539e-02 -5.41301548e-01
-2.81679004e-01 3.21057767e-01 -2.39712685e-01 4.55853529e-02
9.10124648e-03 5.13594985e-01 -2.51705319e-01 -4.48934764e-01
3.25695723e-01 6.83306217e-01 7.51615345e-01 -2.11290672e-01
7.47925639e-01 7.98370302e-01 -4.38353091e-01 -9.30900514e-01
-6.63701355e-01 -1.04926515e+00 -7.05641925e-01 -5.38915992e-01
1.09314179e+00 -1.03397036e+00 -4.09821302e-01 6.21481597e-01
-9.65990782e-01 -2.12513685e-01 5.75474985e-02 5.50072432e-01
-5.75107455e-01 8.12313676e-01 -5.82755506e-01 -1.12332451e+00
-3.98444116e-01 -1.17248368e+00 8.27235222e-01 6.07515991e-01
1.53831661e-01 -7.97222912e-01 -5.17654598e-01 7.49135315e-01
-1.70567632e-02 2.47781888e-01 -3.15753520e-02 -5.23414254e-01
-1.09436357e+00 -2.63290167e-01 -2.24465966e-01 4.62475926e-01
1.40912950e-01 1.85074359e-01 -6.48678958e-01 -2.24037722e-01
1.40779484e-02 1.07574694e-01 1.15320802e+00 5.70120990e-01
1.18228865e+00 -2.49551520e-01 -3.73861730e-01 4.03754950e-01
1.44925201e+00 4.62614298e-01 4.26834017e-01 3.13679576e-01
9.46881354e-01 4.93622512e-01 1.17735267e+00 6.55573189e-01
3.40910375e-01 5.86882472e-01 5.02229393e-01 1.08219549e-01
5.67480028e-02 1.03168534e-02 4.43982273e-01 7.77193785e-01
-2.19173208e-01 -2.86083758e-01 -8.10156167e-01 8.87612820e-01
-2.08552170e+00 -8.97263825e-01 -3.47191989e-01 2.09929419e+00
5.49621582e-01 3.85626078e-01 2.69870996e-01 1.10792033e-01
1.05076385e+00 2.30851710e-01 -6.89841211e-01 4.26122725e-01
-1.44901752e-01 -2.34223872e-01 6.49318159e-01 2.97962695e-01
-1.60551155e+00 1.02263641e+00 4.54134750e+00 8.79613519e-01
-7.80446947e-01 3.02057058e-01 6.80716336e-01 -2.66700312e-02
3.04593563e-01 1.66371346e-01 -7.79491663e-01 6.74889028e-01
5.08149147e-01 -9.90490988e-02 1.65483519e-01 6.49304509e-01
7.39055514e-01 -4.44883972e-01 -6.75377369e-01 8.86318266e-01
2.88652107e-02 -1.04615378e+00 -1.50293335e-01 -1.97748914e-01
1.03712499e+00 -8.48105848e-02 2.17213612e-02 -1.74465757e-02
-3.19318414e-01 -3.53074789e-01 1.01575911e+00 5.45984149e-01
1.16754159e-01 -7.59774387e-01 7.46154428e-01 4.73577291e-01
-1.43396354e+00 -1.78629071e-01 -1.58546209e-01 1.35004506e-01
4.03599024e-01 5.49319506e-01 -6.17406011e-01 7.53444374e-01
6.86175525e-01 9.43503857e-01 -4.78358567e-01 1.42850208e+00
-1.06015682e-01 7.97031879e-01 -2.81444728e-01 1.73855603e-01
4.68435317e-01 -3.31111163e-01 9.28004920e-01 1.18495035e+00
1.80893302e-01 2.91693121e-01 5.79121649e-01 5.87707162e-01
-5.96965402e-02 3.79694909e-01 -1.15293331e-01 -1.06300943e-01
3.02696824e-01 1.32425642e+00 -1.33031738e+00 -6.12130225e-01
-5.43056667e-01 1.18638897e+00 -5.94679117e-02 5.42638898e-01
-1.04733360e+00 -9.57170948e-02 1.02370732e-01 6.16582893e-02
7.84936428e-01 -3.67425323e-01 8.78684372e-02 -1.28146935e+00
2.60979682e-01 -5.65259457e-01 4.77587461e-01 -5.65750301e-01
-8.44046950e-01 2.19121218e-01 2.21060645e-02 -1.49919796e+00
9.58942696e-02 -3.14821720e-01 -5.34426570e-01 1.99965641e-01
-1.43836355e+00 -8.18459988e-01 -3.60718936e-01 4.16691601e-01
1.12898064e+00 8.31009150e-02 -3.79401036e-02 5.11490524e-01
-9.29861307e-01 2.53309667e-01 1.74445137e-01 2.57416636e-01
4.74299431e-01 -9.98184144e-01 5.85602000e-02 1.39840317e+00
9.16535109e-02 3.23718667e-01 6.35012150e-01 -7.43138492e-01
-1.38960552e+00 -1.37466991e+00 3.10196579e-01 -1.16981477e-01
6.73360586e-01 -2.44905856e-02 -1.01404572e+00 4.41931367e-01
2.77201444e-01 3.96487713e-01 3.44475240e-01 -7.89019585e-01
3.00570309e-01 -4.86785099e-02 -8.92335832e-01 5.04286647e-01
1.03394353e+00 8.52354690e-02 -3.30259830e-01 3.79066795e-01
7.10473061e-01 -5.46255708e-01 -7.44621694e-01 5.32333791e-01
3.02710980e-01 -8.01710784e-01 9.10767853e-01 -1.51726678e-01
3.12759340e-01 -7.95909524e-01 5.16271703e-02 -6.37259781e-01
4.75144386e-03 -7.70062208e-01 -4.74440724e-01 1.36958766e+00
7.62532428e-02 -4.72934753e-01 8.19993436e-01 4.10505861e-01
-1.52844116e-01 -6.19132578e-01 -7.95134902e-01 -8.18440557e-01
-4.69103575e-01 -4.29723620e-01 1.49602601e-02 7.14451551e-01
-4.28171813e-01 -1.79176912e-01 -3.71858180e-01 4.25756991e-01
9.53909695e-01 3.12363535e-01 6.88816726e-01 -7.28312492e-01
-2.21954405e-01 -4.37022448e-01 -6.88731253e-01 -1.37279665e+00
9.61920917e-02 -8.01297426e-01 3.48432988e-01 -1.57288098e+00
3.94794226e-01 -9.02374536e-02 -5.14632285e-01 1.82105869e-01
-6.01937294e-01 5.11344612e-01 4.26327348e-01 3.51810992e-01
-1.26890635e+00 5.08999407e-01 1.25273228e+00 -2.83972591e-01
-4.12044585e-01 5.62185161e-02 -3.92292649e-01 1.03675616e+00
9.59319651e-01 -4.68039960e-01 -3.10268939e-01 -2.86514580e-01
-5.38331926e-01 3.95446382e-02 4.51127917e-01 -1.05298400e+00
5.24271607e-01 -1.46795988e-01 2.58891255e-01 -9.36808228e-01
5.45963570e-02 -8.15854490e-01 2.78831348e-02 4.28441793e-01
7.45779872e-02 -3.25886220e-01 9.02534947e-02 1.01395905e+00
-3.68728876e-01 -3.14888477e-01 7.17104733e-01 -3.49990800e-02
-1.38684452e+00 5.32430053e-01 -5.22705436e-01 8.61016735e-02
1.32134593e+00 -3.84771526e-01 3.61262932e-02 -9.34347659e-02
-8.03567469e-01 5.50546706e-01 3.65638047e-01 4.38866556e-01
7.77847826e-01 -1.04904354e+00 -6.98201358e-01 -1.90848798e-01
5.49597815e-02 2.62612820e-01 6.16119266e-01 1.26274991e+00
-6.86744988e-01 1.49897635e-01 1.72240689e-01 -8.63584638e-01
-1.31550729e+00 6.88078701e-01 1.41184228e-02 2.07045913e-01
-8.15559983e-01 6.75146580e-01 2.22696453e-01 2.94932097e-01
2.55114794e-01 -2.52621591e-01 -3.15331131e-01 1.16005249e-01
5.76323152e-01 5.31609178e-01 -2.46160641e-01 -9.98445570e-01
-4.13786680e-01 4.22799230e-01 -2.21547842e-01 5.11699989e-02
1.05283725e+00 -5.87166846e-01 3.44622321e-02 4.31884795e-01
1.03731632e+00 -2.94891655e-01 -1.65148079e+00 -3.48967522e-01
6.05490915e-02 -6.69875383e-01 9.37479213e-02 -1.66530535e-01
-1.35388231e+00 4.76693392e-01 6.61179841e-01 1.80630952e-01
1.39136696e+00 -1.48569196e-01 9.84779239e-01 1.55668646e-01
3.43818933e-01 -1.37071288e+00 -1.05347060e-01 1.99370831e-01
3.83602440e-01 -1.59761310e+00 1.14197627e-01 -7.47386932e-01
-6.39837503e-01 9.41888213e-01 6.13036633e-01 -1.18883356e-01
4.76781666e-01 -9.16185826e-02 -6.91201463e-02 2.15160057e-01
-4.23843831e-01 -5.61335683e-01 2.95568943e-01 1.11554682e-01
1.35131061e-01 -7.10823461e-02 -6.47916198e-01 4.44966942e-01
4.73853499e-01 1.54434949e-01 5.91610134e-01 1.13799191e+00
-6.19566441e-01 -6.97841823e-01 -4.33700293e-01 2.89746732e-01
-8.66292894e-01 1.10216036e-01 -8.58232975e-02 6.07375383e-01
1.29090369e-01 1.25634325e+00 -9.30176675e-02 -4.07456942e-02
1.34944860e-02 -1.89977810e-01 3.40279758e-01 -4.48447019e-01
-2.23567307e-01 6.15150869e-01 -1.81038007e-01 -5.78123510e-01
-1.06443965e+00 -8.77282619e-01 -1.62923920e+00 2.01140448e-01
-5.34470916e-01 -3.82294953e-02 4.09844011e-01 8.47113788e-01
6.54237270e-02 5.56252778e-01 6.97519839e-01 -1.16070080e+00
-1.21744797e-01 -6.95172071e-01 -4.12168592e-01 3.58575344e-01
2.44572535e-01 -6.42296314e-01 -2.88449019e-01 6.16633236e-01] | [9.174880981445312, -0.35043877363204956] |
75d97281-9e41-482d-8c78-dd859b57ab47 | a-review-of-co-saliency-detection-technique | 1604.07090 | null | http://arxiv.org/abs/1604.07090v5 | http://arxiv.org/pdf/1604.07090v5.pdf | A Review of Co-saliency Detection Technique: Fundamentals, Applications, and Challenges | Co-saliency detection is a newly emerging and rapidly growing research area
in computer vision community. As a novel branch of visual saliency, co-saliency
detection refers to the discovery of common and salient foregrounds from two or
more relevant images, and can be widely used in many computer vision tasks. The
existing co-saliency detection algorithms mainly consist of three components:
extracting effective features to represent the image regions, exploring the
informative cues or factors to characterize co-saliency, and designing
effective computational frameworks to formulate co-saliency. Although numerous
methods have been developed, the literature is still lacking a deep review and
evaluation of co-saliency detection techniques. In this paper, we aim at
providing a comprehensive review of the fundamentals, challenges, and
applications of co-saliency detection. Specifically, we provide an overview of
some related computer vision works, review the history of co-saliency
detection, summarize and categorize the major algorithms in this research area,
discuss some open issues in this area, present the potential applications of
co-saliency detection, and finally point out some unsolved challenges and
promising future works. We expect this review to be beneficial to both fresh
and senior researchers in this field, and give insights to researchers in other
related areas regarding the utility of co-saliency detection algorithms. | ['Xuelong. Li', 'Huazhu Fu', 'Dingwen Zhang', 'Junwei Han', 'Ali Borji'] | 2016-04-24 | null | null | null | null | ['co-saliency-detection'] | ['computer-vision'] | [ 5.42432606e-01 -2.35910773e-01 -4.57432956e-01 1.70931686e-02
-4.43785429e-01 -1.94205135e-01 3.27901751e-01 1.39158785e-01
2.58513144e-03 4.33340997e-01 5.65664507e-02 9.50678587e-02
-1.04565747e-01 -3.35374832e-01 -4.53407824e-01 -7.68371344e-01
-9.92135480e-02 -2.16732129e-01 1.01705539e+00 -1.32244334e-01
7.76032388e-01 3.02557260e-01 -2.11674547e+00 -6.86345017e-03
1.19064295e+00 7.64380157e-01 7.94264674e-01 5.29073417e-01
3.54901776e-02 6.54299796e-01 -6.46822810e-01 -2.56099198e-02
-9.20015275e-02 -6.98099434e-01 -7.86712289e-01 4.17067885e-01
3.42529953e-01 9.72784236e-02 -1.36146694e-01 1.33929002e+00
3.33565831e-01 2.02322096e-01 5.75045049e-01 -1.57809091e+00
-7.40286648e-01 2.78026193e-01 -9.10824895e-01 1.14938903e+00
2.57860065e-01 -2.79758185e-01 1.02848697e+00 -1.22790623e+00
4.50868428e-01 1.09339988e+00 3.12418014e-01 2.99708456e-01
-5.07172823e-01 -4.26680386e-01 5.08531630e-01 8.73431027e-01
-1.24280703e+00 -1.43346086e-01 1.15860140e+00 -3.31910193e-01
5.53818524e-01 6.38807952e-01 1.02776289e+00 3.97597104e-01
1.31372303e-01 1.53916049e+00 1.01854479e+00 -6.16017640e-01
4.57642302e-02 1.27173644e-02 3.70792001e-01 5.74517608e-01
3.15158427e-01 -2.01712608e-01 -7.31765568e-01 -4.90251482e-02
6.92959666e-01 2.52041757e-01 -1.43872827e-01 -7.13909864e-01
-1.43419492e+00 8.48105133e-01 6.97334349e-01 5.83248973e-01
-3.70043039e-01 -1.33943930e-01 2.07360148e-01 -2.82623291e-01
4.67226386e-01 5.07780313e-01 -8.77732709e-02 2.32256547e-01
-1.11222494e+00 4.16084915e-01 1.39367372e-01 8.56561184e-01
8.06548297e-01 3.00140858e-01 -2.55375415e-01 7.41301119e-01
2.80973703e-01 6.51739061e-01 5.11471570e-01 -7.21830845e-01
-2.53806952e-02 4.95338291e-01 -2.04261634e-02 -1.40597379e+00
-2.82184988e-01 -2.62466908e-01 -3.31087172e-01 1.40255198e-01
-2.05084458e-01 3.17571536e-02 -7.49001801e-01 1.14852369e+00
4.06195521e-01 5.45889258e-01 -2.27095485e-01 1.23003268e+00
1.35036886e+00 4.47493434e-01 1.34066328e-01 -1.75637186e-01
1.50154328e+00 -1.54351091e+00 -8.94433200e-01 -5.27737856e-01
-1.69952229e-01 -1.21433198e+00 7.41566956e-01 -2.46977642e-01
-1.16085923e+00 -5.41632593e-01 -1.19628048e+00 -1.03179418e-01
-4.95076448e-01 2.28532284e-01 8.42714310e-01 3.14261347e-01
-1.05438614e+00 2.24122688e-01 -7.96628892e-01 -6.40834093e-01
5.66344738e-01 1.57473713e-01 3.10000628e-01 -6.99060708e-02
-1.13321960e+00 1.16319513e+00 3.94487977e-01 4.83469367e-02
-9.01636779e-01 -5.67045689e-01 -9.19814050e-01 -2.32194528e-01
6.24617875e-01 -4.80899394e-01 1.17805183e+00 -1.11704075e+00
-8.44559431e-01 9.81990933e-01 -7.35059977e-01 -3.88360053e-01
9.84642282e-02 -3.71311307e-01 -6.00193143e-01 4.14358467e-01
5.48808992e-01 6.81674242e-01 9.46999252e-01 -1.20729625e+00
-1.21493959e+00 -1.47856385e-01 -3.61537486e-02 6.48230672e-01
-1.60376281e-01 7.09074318e-01 -6.41407430e-01 -7.78480709e-01
2.71011651e-01 -6.52101755e-01 -3.09255004e-01 -1.47033215e-01
-3.18571538e-01 -3.32930088e-01 1.41583347e+00 -3.51167798e-01
1.33609569e+00 -2.11923194e+00 -3.36710140e-02 -3.76211435e-01
6.02856576e-01 5.57319403e-01 1.41765401e-01 1.52499467e-01
1.33820083e-02 -8.58575851e-02 -3.50107938e-01 -9.40257451e-04
-4.88896102e-01 -3.17033827e-01 -4.19194072e-01 6.03092194e-01
5.13906181e-01 1.17819512e+00 -1.38573229e+00 -8.33227634e-01
7.80857265e-01 4.26929295e-01 1.89380854e-01 1.01700425e-02
1.43927455e-01 1.45589918e-01 -7.35297382e-01 1.00788713e+00
5.69991529e-01 -3.84362668e-01 -2.93996185e-01 -1.88791640e-02
-3.57087880e-01 1.69221193e-01 -9.46134329e-01 1.16172922e+00
2.54571378e-01 1.14725900e+00 -3.17046903e-02 -1.10631895e+00
9.95693326e-01 3.15119736e-02 6.01327717e-01 -5.17102897e-01
1.71047617e-02 3.74554396e-01 1.45445317e-02 -4.81041998e-01
8.80022347e-01 1.98405951e-01 1.77369297e-01 1.33780181e-01
-1.56709254e-01 -1.72937527e-01 3.61225218e-01 4.56578881e-02
5.21062434e-01 -1.89672738e-01 7.58886158e-01 -4.20111746e-01
7.61205673e-01 3.29800904e-01 6.40065014e-01 6.40605271e-01
-9.75552142e-01 7.89498270e-01 -9.68239456e-02 -4.46549922e-01
-5.91053903e-01 -9.45708930e-01 4.04787138e-02 1.26459289e+00
1.06992233e+00 -3.69746149e-01 -6.95560694e-01 -5.83167076e-01
-1.82228506e-01 1.79729238e-01 -8.34253192e-01 -5.27341589e-02
-6.62139535e-01 -8.64188254e-01 1.48751065e-01 6.00555837e-01
8.13276350e-01 -1.47647417e+00 -1.13852775e+00 -1.92891464e-01
-4.40864354e-01 -9.45924163e-01 -4.17656600e-01 -1.98020682e-01
-1.04609275e+00 -1.28491580e+00 -1.23075235e+00 -1.50542521e+00
6.82751894e-01 1.61315751e+00 1.12141812e+00 5.68005741e-01
-4.26529944e-01 2.85844088e-01 -3.75244051e-01 -8.16236317e-01
3.96839648e-01 -6.14133589e-02 -1.67794496e-01 -1.47687882e-01
7.24380255e-01 -1.00637309e-01 -5.69469512e-01 4.56542224e-01
-7.02213705e-01 1.48838386e-01 5.09116471e-01 5.75803936e-01
8.24219108e-01 -1.24454778e-02 5.79093099e-01 -5.66950977e-01
4.54512417e-01 -5.32191873e-01 -3.23001266e-01 3.31999481e-01
-3.71923327e-01 -4.95521247e-01 -5.42408824e-02 -1.70489386e-01
-8.01273942e-01 -5.50554320e-02 3.02297682e-01 -3.98350626e-01
-3.80111337e-02 2.74492621e-01 -1.54700086e-01 -3.87228251e-01
4.00057375e-01 4.86244231e-01 -3.11883777e-01 -3.24433684e-01
3.75934362e-01 3.41654748e-01 6.41405523e-01 -1.21467002e-01
6.01980507e-01 6.26120150e-01 -1.51810735e-01 -1.12507391e+00
-1.28149819e+00 -1.06954348e+00 -7.09730804e-01 -5.27160645e-01
8.33886862e-01 -6.82822526e-01 -6.95718229e-02 6.50546312e-01
-1.00092483e+00 1.70248240e-01 -2.02842668e-01 3.00065160e-01
-5.19932330e-01 5.40407538e-01 -2.72055209e-01 -6.87433362e-01
-3.56685609e-01 -1.43910122e+00 9.72419381e-01 1.16883874e+00
-3.73929814e-02 -1.19794679e+00 -2.43303150e-01 4.10996199e-01
4.82491940e-01 2.13029355e-01 3.37091506e-01 -4.03693736e-01
-7.45771587e-01 8.99055377e-02 -4.50092465e-01 -1.33990243e-01
3.76117855e-01 1.40552446e-01 -8.97633731e-01 -6.79524764e-02
7.36499717e-03 -3.11122183e-02 9.64304388e-01 8.66048634e-01
1.04373085e+00 1.93882272e-01 -8.08693111e-01 2.45385155e-01
1.12703252e+00 2.57281482e-01 3.37409228e-01 6.61439419e-01
6.32790864e-01 7.08048105e-01 1.28109682e+00 5.89851439e-02
5.04142582e-01 6.87698901e-01 4.85761017e-01 -5.09180844e-01
-5.08183718e-01 -1.44053280e-01 1.37517631e-01 9.70277667e-01
-2.06269100e-01 1.93560928e-01 -8.49378526e-01 9.90670443e-01
-2.01966906e+00 -9.26244617e-01 -4.44135755e-01 1.75045645e+00
5.10915101e-01 -5.07951006e-02 4.02865022e-01 2.39405692e-01
1.46938860e+00 4.61462468e-01 -6.36098564e-01 5.48760183e-02
-5.62835693e-01 -1.35029525e-01 2.40784794e-01 3.38955164e-01
-1.68521821e+00 1.41860926e+00 7.69578838e+00 7.84612536e-01
-1.35879636e+00 1.42315730e-01 4.87235814e-01 2.68936545e-01
-7.98452422e-02 1.72523648e-01 -7.40646958e-01 5.18008947e-01
2.11780488e-01 -5.69971561e-01 -8.03576931e-02 1.22004032e+00
2.87659708e-02 -5.83679974e-01 -5.18546760e-01 1.06915975e+00
6.91359162e-01 -1.50334740e+00 -4.31614965e-02 -3.29073668e-01
1.06905115e+00 1.16170086e-01 3.13345402e-01 -1.10245533e-01
-5.61464354e-02 -9.06849444e-01 8.00357103e-01 2.86482483e-01
3.29996616e-01 -6.08012497e-01 6.82755649e-01 1.67584404e-01
-1.63158560e+00 1.42296612e-01 -5.42958140e-01 -3.43002915e-01
2.38645568e-01 7.30115473e-01 -5.66627920e-01 4.76154029e-01
1.00098503e+00 1.31500614e+00 -9.44616079e-01 1.67812788e+00
-4.31136131e-01 4.43790883e-01 4.29280140e-02 -3.63845021e-01
2.21806660e-01 3.29002514e-02 7.59356260e-01 1.29220808e+00
-9.09673199e-02 -1.31639525e-01 4.13318396e-01 8.40552032e-01
4.46286380e-01 8.82673450e-03 -4.15317833e-01 7.66531229e-02
6.30373836e-01 1.25596130e+00 -1.46812189e+00 -5.57077110e-01
-4.27192837e-01 9.00921166e-01 3.01806512e-03 3.02940965e-01
-8.58494341e-01 -7.67320573e-01 7.02190518e-01 -1.39791653e-01
3.89340937e-01 -2.11266190e-01 -6.01894081e-01 -1.01903796e+00
-4.64880355e-02 -4.87369329e-01 3.78242522e-01 -9.02670264e-01
-8.18869829e-01 4.11697119e-01 1.92981392e-01 -1.36828125e+00
1.57051474e-01 -2.13205874e-01 -8.47453654e-01 7.67912745e-01
-1.96192944e+00 -1.07638061e+00 -4.49366480e-01 3.11145306e-01
1.02042699e+00 -6.43189996e-02 3.14366579e-01 -5.63939624e-02
-5.83370745e-01 1.52083039e-01 -1.26696080e-01 2.21739896e-02
4.93343055e-01 -1.02972889e+00 5.62084019e-01 1.29850888e+00
1.92459732e-01 6.57852411e-01 7.89800525e-01 -7.73327827e-01
-1.08836210e+00 -9.51184809e-01 1.03098190e+00 -3.25100958e-01
5.54239988e-01 2.51964945e-02 -9.68629718e-01 4.31410164e-01
2.74263084e-01 1.94323227e-01 4.82244283e-01 -2.25596696e-01
1.56419739e-01 2.65191197e-01 -9.22870815e-01 7.42977023e-01
8.96056473e-01 -3.09797704e-01 -9.22146440e-01 6.42058030e-02
6.66404188e-01 -4.36350971e-01 -1.69038728e-01 5.03603578e-01
2.70234108e-01 -1.03924036e+00 1.15673494e+00 -1.98966742e-01
3.29775363e-01 -8.17113638e-01 1.62611395e-01 -9.51680064e-01
-6.27764583e-01 -4.06646073e-01 -2.96687990e-01 1.03442454e+00
-1.54484853e-01 -2.33526781e-01 7.14734554e-01 5.98560832e-02
-2.46334746e-01 -1.03110552e+00 -6.11014485e-01 -3.89550179e-01
-3.05366009e-01 -3.33495915e-01 3.32470596e-01 8.05433810e-01
-7.72805139e-03 2.21494794e-01 -2.94928461e-01 6.16601668e-02
6.76241875e-01 5.48443735e-01 4.36426252e-01 -1.12199867e+00
4.51010913e-01 -7.27012098e-01 -5.49665034e-01 -1.27283704e+00
5.00538610e-02 -6.97530925e-01 1.80665448e-01 -1.76873636e+00
5.60175121e-01 -1.81690320e-01 -5.19777775e-01 2.63960958e-01
-8.48885477e-01 6.00109637e-01 5.01300633e-01 3.86240482e-01
-1.18421149e+00 5.05773187e-01 1.56327891e+00 -2.01223537e-01
-7.46145844e-02 7.81892538e-02 -1.10882545e+00 1.05128813e+00
8.34328413e-01 -1.53057426e-01 -3.81751388e-01 -2.28671223e-01
-3.08198482e-01 -5.35655379e-01 3.28830540e-01 -9.29877639e-01
3.00009906e-01 -6.42341971e-01 4.17453676e-01 -1.07965302e+00
8.84208381e-02 -4.38283712e-01 -5.10686100e-01 3.88035744e-01
7.46718273e-02 2.98581839e-01 1.64168805e-01 4.56954598e-01
-5.77982962e-01 -3.42521757e-01 9.29760993e-01 -2.19986781e-01
-1.68004560e+00 -1.68733968e-04 -5.30695915e-01 1.14208892e-01
1.52312362e+00 -4.96632934e-01 -1.76062152e-01 -1.43738955e-01
-3.55456084e-01 4.19927865e-01 6.05005622e-01 9.17027354e-01
1.04712045e+00 -1.42216265e+00 -4.51209486e-01 1.07106231e-01
2.66184807e-01 -9.91322696e-02 3.07867557e-01 8.44486535e-01
-3.68266195e-01 6.05440557e-01 -4.73868519e-01 -8.17933202e-01
-1.61921048e+00 7.72941351e-01 2.61582807e-02 1.92792788e-01
-4.70989197e-01 8.64888370e-01 4.24810559e-01 2.85873115e-01
1.36124194e-01 -8.99309963e-02 -8.30491483e-01 -1.69350699e-01
9.17815924e-01 7.02955782e-01 -1.03040986e-01 -1.18368256e+00
-8.53491664e-01 8.98693562e-01 -2.99921595e-02 2.28959113e-01
9.12081540e-01 -4.57509011e-01 -2.44346350e-01 6.31048799e-01
7.10207105e-01 -1.70667291e-01 -1.19913733e+00 -2.74877787e-01
1.54858127e-01 -6.40929818e-01 -7.05491975e-02 -3.70396823e-01
-1.20633268e+00 9.38422561e-01 4.81425077e-01 1.08650334e-01
1.36977446e+00 4.03261006e-01 6.35887921e-01 -1.36237606e-01
3.89962256e-01 -1.06176019e+00 4.63172168e-01 7.11603820e-01
9.67020392e-01 -1.37981963e+00 3.55832130e-01 -8.11876774e-01
-8.97671223e-01 8.26660872e-01 7.47316599e-01 -4.54384595e-01
9.45535243e-01 -9.13316607e-02 1.80011228e-01 -2.79372752e-01
-1.56986520e-01 -6.64307594e-01 7.36103594e-01 8.93924415e-01
4.70544845e-01 5.45914136e-02 -5.34782469e-01 3.60773563e-01
-1.87688246e-01 -1.76837325e-01 4.42321628e-01 1.16502047e+00
-8.97176027e-01 -7.55486131e-01 -6.78121030e-01 3.71824563e-01
-3.87567252e-01 -1.92146190e-02 -4.95918244e-01 4.51478988e-01
2.36766234e-01 1.19827986e+00 5.40338643e-02 -3.82675767e-01
-4.10190560e-02 -4.35402185e-01 1.72995061e-01 -8.11816275e-01
-2.24092916e-01 7.60813728e-02 -5.64764023e-01 -3.49617392e-01
-1.09736514e+00 -8.17056596e-01 -1.21273267e+00 -8.70511532e-02
-6.28809452e-01 1.65842965e-01 4.69285488e-01 1.01204264e+00
3.49158049e-01 5.93934178e-01 4.13123578e-01 -1.15327203e+00
3.20958525e-01 -7.90964723e-01 -5.65613866e-01 2.25721315e-01
5.00263155e-01 -1.28963363e+00 -1.27900496e-01 2.02957213e-01] | [9.763233184814453, -0.37773048877716064] |
841e5600-dac6-42c6-857d-979f8577aaf1 | probing-for-targeted-syntactic-knowledge | 2210.16228 | null | https://arxiv.org/abs/2210.16228v1 | https://arxiv.org/pdf/2210.16228v1.pdf | Probing for targeted syntactic knowledge through grammatical error detection | Targeted studies testing knowledge of subject-verb agreement (SVA) indicate that pre-trained language models encode syntactic information. We assert that if models robustly encode subject-verb agreement, they should be able to identify when agreement is correct and when it is incorrect. To that end, we propose grammatical error detection as a diagnostic probe to evaluate token-level contextual representations for their knowledge of SVA. We evaluate contextual representations at each layer from five pre-trained English language models: BERT, XLNet, GPT-2, RoBERTa, and ELECTRA. We leverage public annotated training data from both English second language learners and Wikipedia edits, and report results on manually crafted stimuli for subject-verb agreement. We find that masked language models linearly encode information relevant to the detection of SVA errors, while the autoregressive models perform on par with our baseline. However, we also observe a divergence in performance when probes are trained on different training sets, and when they are evaluated on different syntactic constructions, suggesting the information pertaining to SVA error detection is not robustly encoded. | ['Paula Buttery', 'Marek Rei', 'Andrew Caines', 'Christopher Bryant', 'Christopher Davis'] | 2022-10-28 | null | null | null | null | ['grammatical-error-detection'] | ['natural-language-processing'] | [ 1.24219768e-02 2.55253881e-01 1.70174167e-01 -7.97295749e-01
-1.32504916e+00 -6.82847679e-01 4.17670161e-01 8.71495545e-01
-7.90238976e-01 4.62923497e-01 6.46244824e-01 -6.14725828e-01
1.95725530e-01 -7.94026136e-01 -1.01495636e+00 5.46819046e-02
1.58445612e-01 4.07807082e-01 -7.20951706e-02 -3.35218132e-01
3.47370803e-01 -5.14880382e-02 -1.18259358e+00 7.09127247e-01
1.42330694e+00 5.25428295e-01 5.72632514e-02 4.27975208e-01
-9.33974460e-02 1.08083069e+00 -7.47758746e-01 -9.01960731e-01
-3.49997669e-01 -3.91911536e-01 -1.22489583e+00 -6.98377907e-01
8.14720273e-01 -1.06635824e-01 8.68248194e-02 1.04615808e+00
2.82794476e-01 -1.33875147e-01 8.35335314e-01 -6.89775527e-01
-1.17493153e+00 1.07708549e+00 9.82530266e-02 5.37947893e-01
7.28760958e-01 4.19333786e-01 1.52713764e+00 -1.15792441e+00
9.24847245e-01 1.43234026e+00 1.05714440e+00 5.70042551e-01
-1.61727822e+00 -5.41384876e-01 3.08546036e-01 2.62728184e-01
-1.08854437e+00 -6.37424290e-01 3.43966037e-01 -5.88108540e-01
1.55537105e+00 2.13899016e-01 4.32001829e-01 1.46379840e+00
2.81100720e-01 8.15597117e-01 1.62473500e+00 -5.91195524e-01
1.16879731e-01 3.54970805e-03 7.42974699e-01 6.54677093e-01
3.90870035e-01 3.54246229e-01 -7.96790183e-01 -1.19247891e-01
-1.35388654e-02 -7.51109660e-01 -2.73728102e-01 2.78656483e-01
-8.71688783e-01 8.14963579e-01 3.89105201e-01 4.43632483e-01
-2.83886820e-01 9.23859850e-02 5.85504055e-01 8.68684828e-01
3.77647430e-01 9.05447960e-01 -8.27422976e-01 -3.29595119e-01
-6.32825971e-01 4.25311685e-01 7.04190075e-01 7.95987844e-01
5.55116296e-01 9.30243265e-03 -4.49792564e-01 9.99640703e-01
5.20468235e-01 4.14464921e-01 4.99421597e-01 -8.48192513e-01
8.19470465e-01 6.25030398e-01 8.92459527e-02 -7.56426930e-01
-5.66875637e-01 -2.81509340e-01 -1.15264416e-01 -9.92661193e-02
6.01041138e-01 5.63558228e-02 -6.65781081e-01 2.26228952e+00
-8.29665512e-02 -4.59622771e-01 2.20078915e-01 5.35795152e-01
1.00479102e+00 4.14060384e-01 9.43019390e-01 -1.00565925e-01
1.42885470e+00 -3.71649176e-01 -6.48608387e-01 -7.72494853e-01
1.36750042e+00 -7.24579334e-01 1.38841343e+00 7.89544210e-02
-1.41120148e+00 -6.18661761e-01 -8.10747325e-01 -4.11911458e-01
-1.64836839e-01 5.61928516e-03 3.50143254e-01 4.57508773e-01
-1.08526719e+00 4.49664235e-01 -5.19280851e-01 -3.17655772e-01
1.37975141e-01 -1.01896279e-01 -5.61303318e-01 -2.09943533e-01
-1.52840841e+00 1.46056950e+00 3.40268105e-01 1.44186929e-01
-8.56953621e-01 -8.20402265e-01 -1.30224466e+00 1.11589074e-01
-1.00273758e-01 -4.46698159e-01 1.53857148e+00 -9.41716492e-01
-9.86243665e-01 1.58252144e+00 -4.08888936e-01 -3.78065109e-01
1.90270424e-01 -4.18691009e-01 -3.93209875e-01 -3.94394696e-01
3.51369262e-01 3.14386606e-01 3.05309385e-01 -9.95848417e-01
-4.60062712e-01 -5.21866202e-01 -3.19343582e-02 7.75988847e-02
1.76419660e-01 7.05571353e-01 5.51050544e-01 -6.33503854e-01
2.83112526e-01 -6.69632316e-01 1.70512125e-01 -4.36403871e-01
-3.44488233e-01 -9.74714935e-01 -9.33057964e-02 -1.09765315e+00
1.36555231e+00 -2.14827013e+00 2.68220156e-02 2.55409896e-01
-3.28647271e-02 1.15988784e-01 -3.81749928e-01 4.55225378e-01
-3.10488135e-01 7.43686497e-01 -3.22645247e-01 -3.27940375e-01
2.09383711e-01 2.28695795e-01 -3.82132560e-01 3.51202667e-01
5.30983627e-01 9.91482913e-01 -7.76511788e-01 -3.41048360e-01
-3.36384207e-01 -1.50372358e-02 -9.64250088e-01 4.51751769e-01
-3.80364448e-01 2.82034308e-01 5.33481874e-02 5.51168621e-01
4.61030722e-01 7.57719129e-02 4.37809676e-01 1.95712782e-02
-1.67781547e-01 1.38547337e+00 -8.05651248e-01 1.42764568e+00
-7.26519525e-01 5.02551913e-01 1.69855610e-01 -8.54829550e-01
7.54777133e-01 3.44594151e-01 -4.82467681e-01 -1.10347903e+00
-1.64543893e-02 5.81433475e-01 5.48681974e-01 -6.78716063e-01
3.50217879e-01 -2.64838219e-01 -4.51994717e-01 3.68555337e-01
2.68128633e-01 -1.87331855e-01 8.69120359e-02 2.61459440e-01
1.18102300e+00 1.49543742e-02 3.57563943e-01 -5.33633828e-01
2.93463647e-01 -4.48612273e-02 8.00173819e-01 1.11917353e+00
-3.09121013e-01 2.20759273e-01 7.25298882e-01 -3.27828705e-01
-8.80017459e-01 -1.26076829e+00 -5.03472269e-01 1.51329184e+00
-4.13354963e-01 -6.96453154e-01 -5.35680175e-01 -8.66569459e-01
8.87268782e-02 1.41757524e+00 -8.45333993e-01 -3.28171134e-01
-8.88179421e-01 -2.64027655e-01 5.30225694e-01 8.05312335e-01
1.13129213e-01 -1.29275906e+00 -2.31732994e-01 3.13650906e-01
-3.59991074e-01 -8.84130597e-01 -1.64225161e-01 2.81905174e-01
-6.41837716e-01 -1.00847185e+00 2.11722165e-01 -1.02469289e+00
5.13098836e-01 -5.80138862e-01 1.70511031e+00 7.50406981e-01
1.65301427e-01 4.58908349e-01 -3.25277984e-01 -2.70349890e-01
-9.51672971e-01 -7.21273124e-02 -1.20635785e-01 -7.19148517e-01
7.75969625e-01 -3.38495851e-01 -2.22346187e-01 5.95907606e-02
-3.71692061e-01 -3.42510372e-01 3.73341024e-01 1.06868899e+00
1.94733396e-01 -7.66464114e-01 4.81078625e-01 -1.13583314e+00
8.10289025e-01 -5.37493348e-01 -4.44438875e-01 4.28927809e-01
-6.16365194e-01 2.03299984e-01 2.83692807e-01 -1.24242818e-02
-1.18005347e+00 -5.26085496e-01 -6.26627922e-01 3.95055056e-01
-3.28641802e-01 7.72010088e-01 1.52680073e-02 3.19149941e-01
9.96680319e-01 -8.50258470e-02 -2.93482721e-01 -4.14912552e-01
2.56395847e-01 5.67771912e-01 4.31835175e-01 -1.31913817e+00
4.81137782e-01 -3.67698610e-01 -8.21191013e-01 -3.39334786e-01
-1.17964697e+00 -5.11030629e-02 -3.39133650e-01 1.32332131e-01
9.14491534e-01 -1.02331150e+00 -7.45195150e-01 4.47658062e-01
-1.55804360e+00 -5.84073424e-01 -2.15566814e-01 4.27859873e-01
-3.93857867e-01 6.39468506e-02 -1.01331317e+00 -4.35382634e-01
-2.96308011e-01 -1.18616128e+00 7.42158115e-01 -2.62911320e-01
-8.28944266e-01 -9.77933228e-01 1.86913133e-01 3.68247330e-01
5.31069458e-01 -2.74108380e-01 1.61934268e+00 -1.28314209e+00
-1.78962916e-01 4.24469039e-02 -1.45998716e-01 4.45688695e-01
-3.91223490e-01 2.67789625e-02 -8.97128582e-01 -1.68506235e-01
-3.65272444e-03 -8.08724284e-01 9.58802998e-01 1.95879504e-01
9.45128977e-01 -6.11257017e-01 1.75744779e-02 2.51347661e-01
1.10679221e+00 -2.06135526e-01 4.40855950e-01 1.82452187e-01
2.81889141e-01 6.53472781e-01 4.54719454e-01 -1.38294846e-01
7.14484811e-01 5.71555734e-01 -2.43903939e-02 5.33473015e-01
-5.24511412e-02 -5.83394766e-01 6.86093748e-01 9.87702727e-01
2.11465135e-01 -2.00255379e-01 -1.30363119e+00 9.87833679e-01
-1.31081796e+00 -7.71342874e-01 -3.28659743e-01 2.07801414e+00
1.52050197e+00 1.19906701e-01 -4.15304601e-01 -1.39738590e-01
3.91451120e-01 1.38733342e-01 -3.55489179e-02 -1.08575189e+00
-1.07765079e-01 6.08241260e-01 -1.51866183e-01 8.50582957e-01
-8.01496387e-01 1.27031672e+00 6.85630083e+00 2.56210685e-01
-7.56343663e-01 4.65592057e-01 4.98820364e-01 3.33626419e-01
-9.54969287e-01 1.23732284e-01 -7.15041995e-01 4.64041561e-01
1.31452549e+00 9.84771848e-02 1.18394017e-01 6.75409496e-01
-1.47756085e-01 -1.76847264e-01 -1.48947585e+00 2.77590245e-01
-3.22226845e-02 -9.53768671e-01 -1.03505060e-01 -4.36243713e-01
6.53302014e-01 9.56236422e-02 1.07048862e-02 1.03770602e+00
8.04747105e-01 -1.30127633e+00 9.94622827e-01 3.07741821e-01
5.73086083e-01 -3.88565838e-01 8.22142124e-01 2.50759602e-01
-5.36351442e-01 -3.49554755e-02 -3.86829108e-01 -3.77206177e-01
-6.60376102e-02 2.73839802e-01 -7.96016872e-01 -5.22726439e-02
6.66088223e-01 3.86871397e-01 -9.33424234e-01 5.59562564e-01
-1.16249156e+00 1.26317418e+00 -1.33680895e-01 -9.99218673e-02
-8.44579414e-02 2.44934902e-01 3.82991076e-01 1.40457892e+00
3.24498639e-02 1.76666170e-01 -1.74823031e-02 1.08149827e+00
-4.68610466e-01 4.80205536e-01 -4.99330759e-01 -8.70660134e-03
8.73239398e-01 6.16491556e-01 3.65096889e-02 -2.88179010e-01
-7.17294991e-01 6.87441230e-01 1.26350713e+00 2.55621105e-01
-3.49925101e-01 -4.58118133e-02 7.01702237e-01 -5.16106281e-03
5.61274178e-02 -3.38916570e-01 -5.02335370e-01 -1.01133347e+00
9.89822373e-02 -1.04984820e+00 6.01516366e-01 -8.92235339e-01
-1.59647453e+00 1.98451713e-01 -3.34885567e-01 -3.01792502e-01
-3.96249682e-01 -8.72984827e-01 -7.40427256e-01 1.15153241e+00
-1.27756739e+00 -9.44064975e-01 1.04299217e-01 2.26773039e-01
3.67531091e-01 1.34847224e-01 1.08836532e+00 -1.06490813e-01
-4.04810578e-01 9.21105206e-01 -2.91368306e-01 6.78035259e-01
9.69780684e-01 -1.38495517e+00 6.34087861e-01 1.00292039e+00
3.15138072e-01 1.21181977e+00 6.81942344e-01 -9.25969362e-01
-9.75314617e-01 -7.54950345e-01 1.64730608e+00 -9.18701649e-01
6.53656602e-01 -3.28532368e-01 -1.54082453e+00 1.31630266e+00
4.81148243e-01 1.21941723e-01 7.52881944e-01 1.00713706e+00
-9.40056860e-01 2.78008610e-01 -1.06314647e+00 3.30287665e-01
1.34515822e+00 -1.07002282e+00 -1.42495394e+00 2.99093157e-01
8.00873101e-01 -6.05155051e-01 -1.02837145e+00 4.05607611e-01
1.55445904e-01 -9.09195662e-01 5.24673522e-01 -1.36692786e+00
7.12761402e-01 7.59787932e-02 -3.57834101e-01 -1.75464666e+00
-3.85770857e-01 3.62662040e-02 3.83167326e-01 1.29258394e+00
9.58893597e-01 -5.68825066e-01 -6.79495186e-02 9.88447726e-01
-5.21846473e-01 -5.23578942e-01 -1.29380965e+00 -6.22636139e-01
8.00907314e-01 -8.27296019e-01 2.74430603e-01 1.20316422e+00
1.26679629e-01 2.35611200e-01 3.21735173e-01 3.30438644e-01
1.91784292e-01 -1.26217440e-01 2.30019778e-01 -1.18473041e+00
-1.25980273e-01 -2.77264684e-01 -2.61943102e-01 -5.22780597e-01
7.43907154e-01 -1.41281688e+00 3.78785104e-01 -1.49125934e+00
2.66851187e-01 -6.40347838e-01 -1.48447812e-01 7.16589689e-01
-5.77106535e-01 -1.97422206e-01 4.92180660e-02 -1.02304608e-01
-3.41243386e-01 3.70077103e-01 6.06338620e-01 7.07568452e-02
1.55858263e-01 -3.36255431e-01 -8.21411550e-01 8.28134477e-01
5.70315361e-01 -6.53526902e-01 2.51080543e-01 -9.67443824e-01
8.44951093e-01 -1.72997817e-01 5.25504649e-01 -6.85147762e-01
2.79831626e-02 -4.50404026e-02 2.36410275e-01 6.87816292e-02
-2.67762840e-01 -3.35941166e-01 -4.69158977e-01 3.84287566e-01
-9.10561025e-01 4.64038998e-01 3.63105357e-01 1.65061593e-01
-2.32908905e-01 -6.38451755e-01 6.59100056e-01 -2.71368325e-01
-6.68495953e-01 -3.58114630e-01 -5.25675535e-01 9.48696673e-01
3.86232167e-01 3.84180129e-01 -5.80800295e-01 -3.65612134e-02
-9.09056902e-01 1.24714315e-01 3.62022996e-01 4.98746395e-01
5.43924153e-01 -1.11572528e+00 -1.12213743e+00 2.12080866e-01
4.66985524e-01 -2.73942351e-01 2.30273992e-01 7.64896035e-01
-1.96557820e-01 2.81859100e-01 1.91401690e-01 -5.47394931e-01
-1.11760724e+00 3.34325284e-01 4.45001990e-01 -1.98256209e-01
-1.30986169e-01 1.27432835e+00 -1.14338115e-01 -9.20611441e-01
-3.74188870e-02 -7.15624034e-01 1.33994417e-02 6.18091710e-02
3.77777606e-01 -3.12730148e-02 2.67271429e-01 -6.40910089e-01
-4.73845005e-01 3.07872713e-01 -3.68199676e-01 1.05803281e-01
1.12985694e+00 1.36526585e-01 -2.45754689e-01 5.99544525e-01
9.76485729e-01 7.59222925e-01 -5.13569236e-01 -4.88525003e-01
4.05393362e-01 -2.03154519e-01 -1.99261531e-01 -1.33381128e+00
-4.87815380e-01 9.21033680e-01 3.63202959e-01 -1.61923498e-01
3.70674491e-01 3.08989853e-01 2.17812791e-01 3.93280953e-01
3.06640357e-01 -1.23137975e+00 -1.53273627e-01 9.63496268e-01
1.27116370e+00 -1.41099703e+00 -3.51206213e-01 -2.40903333e-01
-6.07951224e-01 7.96398759e-01 1.06466401e+00 -6.30399808e-02
2.70110637e-01 1.68029070e-01 8.49622414e-02 -2.38419488e-01
-9.87439334e-01 -3.47411670e-02 2.13518769e-01 4.73148733e-01
1.12968922e+00 2.12389708e-01 -7.33948708e-01 9.87608254e-01
-5.40460765e-01 -5.63135624e-01 3.57803404e-01 7.79368758e-01
-4.30032909e-01 -8.70826304e-01 -1.63595438e-01 5.19053280e-01
-5.53843796e-01 -7.17591286e-01 -3.99258435e-01 8.48302245e-01
-7.14728162e-02 8.84682477e-01 3.43459159e-01 1.93486195e-02
6.72242045e-01 8.12320590e-01 4.37602699e-01 -9.59494710e-01
-1.08160174e+00 -9.05792356e-01 6.94964468e-01 -7.85122216e-01
-8.65051001e-02 -9.26512301e-01 -1.25093079e+00 -1.37198910e-01
-1.05558045e-01 2.39908621e-01 5.17355025e-01 1.21743214e+00
2.64767259e-01 2.40367338e-01 1.02082580e-01 1.61659643e-01
-8.51214886e-01 -1.32133901e+00 5.61945923e-02 8.54890287e-01
2.74492651e-01 -4.24164325e-01 -3.85289639e-01 -4.73602712e-01] | [10.731304168701172, 9.344642639160156] |
76c54b20-7e18-492e-b557-fa456637c294 | reinforcement-learning-with-tensor-networks | 2209.14089 | null | https://arxiv.org/abs/2209.14089v1 | https://arxiv.org/pdf/2209.14089v1.pdf | Reinforcement Learning with Tensor Networks: Application to Dynamical Large Deviations | We present a framework to integrate tensor network (TN) methods with reinforcement learning (RL) for solving dynamical optimisation tasks. We consider the RL actor-critic method, a model-free approach for solving RL problems, and introduce TNs as the approximators for its policy and value functions. Our "actor-critic with tensor networks" (ACTeN) method is especially well suited to problems with large and factorisable state and action spaces. As an illustration of the applicability of ACTeN we solve the exponentially hard task of sampling rare trajectories in two paradigmatic stochastic models, the East model of glasses and the asymmetric simple exclusion process (ASEP), the latter being particularly challenging to other methods due to the absence of detailed balance. With substantial potential for further integration with the vast array of existing RL methods, the approach introduced here is promising both for applications in physics and to multi-agent RL problems more generally. | ['Juan P. Garrahan', 'Dominic C. Rose', 'Edward Gillman'] | 2022-09-28 | null | null | null | null | ['tensor-networks'] | ['methodology'] | [-1.72539517e-01 9.95624959e-02 -1.19787060e-01 3.67376804e-01
-5.78822911e-01 -3.76573026e-01 8.26536119e-01 -2.61221975e-01
-6.20904148e-01 1.20925355e+00 2.62862388e-02 -2.81954199e-01
-5.87088525e-01 -4.53784436e-01 -3.61389667e-01 -1.34951603e+00
-2.50335068e-01 9.94906843e-01 -1.71398759e-01 -8.09467793e-01
2.41602182e-01 8.29542339e-01 -1.34798336e+00 -2.89730847e-01
8.62469137e-01 8.42375457e-01 -1.71888530e-01 7.46063411e-01
1.40390113e-01 1.05458546e+00 -2.83483714e-01 -1.19296268e-01
2.36878961e-01 -5.50100446e-01 -9.48636413e-01 -1.10742457e-01
-1.65651083e-01 1.38832644e-01 -5.23228168e-01 9.01495993e-01
6.89726174e-01 7.90333509e-01 8.28019440e-01 -1.09820640e+00
-3.21296930e-01 4.11792070e-01 -2.86912501e-01 3.33555460e-01
2.27570906e-01 7.32420564e-01 1.05213296e+00 -3.50520462e-01
5.98016202e-01 1.49611986e+00 5.26658237e-01 8.74263108e-01
-1.50080991e+00 -1.47829875e-01 2.73122221e-01 9.85914096e-02
-8.05864990e-01 -3.48649442e-01 6.45718157e-01 -4.28366840e-01
1.13092983e+00 2.82679409e-01 1.08100331e+00 1.27518463e+00
4.07655627e-01 9.79781508e-01 1.43351650e+00 -3.35441768e-01
7.00406849e-01 -3.21808696e-01 -1.02741703e-01 5.63984811e-01
-6.52204826e-03 6.84211612e-01 -4.23905671e-01 -4.99153137e-01
8.68820906e-01 -3.41322660e-01 1.42357692e-01 -4.32808220e-01
-1.37599170e+00 1.08128428e+00 3.21993798e-01 1.46307349e-01
-5.69860578e-01 6.70977116e-01 5.98608971e-01 4.24125552e-01
7.05958188e-01 7.93900669e-01 -3.96951020e-01 -2.78939813e-01
-6.82560384e-01 1.13783801e+00 9.36825037e-01 4.11317974e-01
3.93680960e-01 7.36141443e-01 -1.28358558e-01 4.78428453e-01
4.57855046e-01 6.70534492e-01 3.23432565e-01 -1.55626917e+00
1.86707735e-01 1.21183179e-01 6.38995469e-01 -3.71831536e-01
-6.15631998e-01 -3.38994235e-01 -9.07655299e-01 6.46415174e-01
8.34324002e-01 -2.90259093e-01 -4.76313740e-01 1.47350931e+00
7.67260253e-01 7.08792284e-02 1.93149358e-01 8.07923019e-01
3.35698277e-01 7.12817371e-01 1.98453516e-02 -5.38968742e-01
1.00270975e+00 -8.35800827e-01 -6.50374532e-01 3.73398699e-02
6.79983914e-01 -3.71068597e-01 8.30138803e-01 6.75805807e-01
-1.46682954e+00 -6.90408051e-02 -6.13786757e-01 2.17036188e-01
-2.64008284e-01 -2.19489515e-01 1.06962228e+00 3.66329044e-01
-1.23812616e+00 1.11879492e+00 -1.17183411e+00 -1.94812819e-01
8.63463134e-02 6.73223972e-01 2.75842369e-01 2.43081167e-01
-1.28395200e+00 1.25595248e+00 1.20021284e-01 4.98096615e-01
-1.16873085e+00 -3.64776433e-01 -5.52823186e-01 -2.90574104e-01
4.45250392e-01 -6.85782671e-01 1.50628066e+00 -8.73225510e-01
-2.29860020e+00 1.81404740e-01 1.07946217e-01 -6.16476178e-01
8.04960012e-01 1.46976784e-01 -4.63953316e-02 1.43608274e-02
-1.49205938e-01 4.47845787e-01 1.01065254e+00 -8.28567564e-01
5.44707850e-02 -2.02010155e-01 1.61103770e-01 3.45476806e-01
3.69968712e-01 -9.23689269e-03 5.07473767e-01 -3.98838401e-01
-2.62596577e-01 -1.34495604e+00 -9.35665309e-01 -2.17383862e-01
-2.62796342e-01 -5.94076276e-01 5.17121494e-01 -2.48490289e-01
7.97773123e-01 -1.49363482e+00 1.01892483e+00 2.64125884e-01
2.87454665e-01 4.04869735e-01 -1.26804724e-01 8.32978725e-01
-2.52888654e-03 -7.62906745e-02 -2.05024511e-01 -1.52165115e-01
1.39774740e-01 4.15200144e-01 -1.47568196e-01 6.80709481e-01
1.92709103e-01 1.09186637e+00 -1.21479309e+00 -2.25800186e-01
2.02966720e-01 3.71321857e-01 -4.35061574e-01 -2.78514680e-02
-9.11257505e-01 1.11026049e+00 -8.61184955e-01 2.37916023e-01
1.93464309e-01 -1.35884076e-01 1.96066901e-01 2.76140332e-01
-3.14093977e-01 1.00501642e-01 -1.22860432e+00 1.37295413e+00
-3.12593073e-01 3.67521077e-01 2.27023125e-01 -1.16732907e+00
6.15131438e-01 4.22294915e-01 8.00278604e-01 -5.66503167e-01
1.58482835e-01 2.90462673e-01 2.92591542e-01 -7.24066496e-01
4.47370708e-01 -4.62496787e-01 -3.98329459e-02 7.05422997e-01
-1.16417408e-02 -5.85299253e-01 4.57913041e-01 2.32345000e-01
1.27605116e+00 5.81493258e-01 1.25559002e-01 -3.91692460e-01
6.24876499e-01 1.26928508e-01 3.20618808e-01 9.11055446e-01
-3.15883934e-01 -1.34903610e-01 8.17594290e-01 -7.13301539e-01
-1.26520240e+00 -6.54497981e-01 1.14201613e-01 7.74646401e-01
-3.56315404e-01 -3.50296348e-01 -5.86011231e-01 -2.81846344e-01
2.16350138e-01 5.01174986e-01 -7.56184280e-01 -4.64680605e-02
-8.64492774e-01 -1.05071521e+00 3.42995107e-01 8.25228728e-03
1.59605145e-01 -1.55382049e+00 -3.32521766e-01 5.78519225e-01
1.16949730e-01 -7.22092152e-01 -8.47150013e-02 1.24760292e-01
-8.52920771e-01 -9.30594921e-01 -8.77008617e-01 -4.52451743e-02
1.16443284e-01 -3.57987225e-01 9.96546566e-01 -1.38397306e-01
-1.85733110e-01 6.54001951e-01 -6.72360659e-02 -1.18918054e-01
-8.58738363e-01 2.01337505e-02 3.37388128e-01 -2.15260834e-02
-2.38885924e-01 -4.88073289e-01 -6.29323423e-01 1.89823627e-01
-8.91364455e-01 -2.53659874e-01 5.33504076e-02 9.16655838e-01
2.58052230e-01 -1.94748610e-01 5.26043355e-01 -5.93667984e-01
1.11517084e+00 -4.89231497e-01 -9.85421777e-01 -1.36822574e-02
-6.31969690e-01 4.80810881e-01 8.45239758e-01 -7.05116570e-01
-9.36055303e-01 -1.06365800e-01 -6.79877177e-02 -2.26624772e-01
1.66305810e-01 4.63797271e-01 4.68969226e-01 -5.93339801e-01
6.55256510e-01 7.42506385e-02 4.77744430e-01 -2.09586039e-01
4.74961221e-01 3.25788409e-02 -1.61955774e-01 -1.02506077e+00
5.60192108e-01 3.96474689e-01 6.59386933e-01 -9.64183331e-01
-7.19647884e-01 -8.13116357e-02 -4.59563881e-01 -5.16528189e-01
8.33441556e-01 -4.04843718e-01 -1.66690528e+00 5.13776183e-01
-1.09203792e+00 -8.61352563e-01 -8.15447330e-01 5.34922063e-01
-1.20879269e+00 2.33720422e-01 -9.12422359e-01 -1.34728098e+00
-1.37728721e-01 -1.34012830e+00 8.46134961e-01 1.16666041e-01
8.22953731e-02 -1.38205385e+00 6.03983641e-01 6.99026659e-02
6.75700963e-01 4.87579435e-01 7.55440712e-01 -3.42513591e-01
-5.73253214e-01 3.14938612e-02 3.39922786e-01 1.74400046e-01
-6.46865904e-01 2.55461007e-01 -6.86975896e-01 -3.83177400e-01
1.11734837e-01 -6.83577836e-01 8.23565364e-01 7.23075569e-01
7.21240163e-01 -3.28633249e-01 -6.62122145e-02 1.36095807e-01
1.25387716e+00 1.48552760e-01 4.35327142e-01 2.27359220e-01
7.41150260e-01 5.79978466e-01 3.71498406e-01 5.41243255e-01
2.77837545e-01 8.31011534e-01 3.52483958e-01 1.39256045e-01
1.68807581e-01 1.28830627e-01 6.37244046e-01 1.10039890e+00
-5.12886703e-01 -1.33538887e-01 -8.84197593e-01 -3.73495445e-02
-2.28689981e+00 -1.20641732e+00 -2.06528783e-01 2.06586099e+00
7.91770458e-01 -9.96728614e-02 6.85972393e-01 -8.82317424e-02
3.20272774e-01 2.83687979e-01 -1.04728818e+00 -7.70434082e-01
-1.62712634e-02 3.72150809e-01 4.52910483e-01 6.27768397e-01
-8.48551631e-01 1.02761722e+00 7.28308296e+00 1.09822702e+00
-8.12562764e-01 1.98704287e-01 5.10418355e-01 -1.91398501e-01
3.88367772e-02 1.03268348e-01 -7.97563195e-01 3.49817067e-01
1.49058819e+00 1.70414463e-01 1.26037371e+00 3.43890727e-01
8.20304930e-01 -3.45669419e-01 -8.47399592e-01 5.79580784e-01
-4.79451984e-01 -1.41470766e+00 -4.02623534e-01 2.94549316e-01
8.12054276e-01 3.62542242e-01 5.95958717e-02 3.87146711e-01
9.88374949e-01 -1.04502809e+00 5.25707483e-01 7.96036243e-01
5.19288659e-01 -6.71732724e-01 3.29131365e-01 3.26262683e-01
-8.76758337e-01 -1.46955848e-01 -3.78486335e-01 -3.38317841e-01
2.15854585e-01 3.96692067e-01 -1.88408598e-01 3.14347893e-01
2.71180779e-01 8.86296690e-01 1.97660476e-02 9.98365998e-01
-1.50898639e-02 6.19049907e-01 -5.56716025e-01 -6.14584088e-01
6.96402788e-01 -8.41609895e-01 9.17599380e-01 6.01099074e-01
-9.99217853e-02 -3.46849188e-02 1.76861122e-01 8.40281546e-01
3.44059199e-01 -8.64019245e-02 -7.63844073e-01 -2.10377499e-01
-1.55572191e-01 1.19056094e+00 -8.89631212e-01 -2.31940240e-01
6.80924505e-02 4.27495748e-01 6.28789961e-01 6.81804657e-01
-6.27613127e-01 2.41197944e-02 5.58128417e-01 -1.99473038e-01
2.57035404e-01 -6.14581227e-01 2.49412000e-01 -1.46597540e+00
-3.84077817e-01 -1.16667545e+00 8.77328590e-02 -5.82817435e-01
-1.45651460e+00 3.97915065e-01 2.10953668e-01 -9.62925732e-01
-6.18062556e-01 -7.08296418e-01 -4.87674981e-01 5.75256705e-01
-1.23809206e+00 -9.66536820e-01 4.79254484e-01 5.85175455e-01
3.12496096e-01 -1.04670599e-01 8.45290661e-01 -1.45668805e-01
-8.08575749e-01 -1.89387783e-01 8.20795476e-01 -4.30423707e-01
-3.04787513e-02 -1.48426580e+00 3.25178981e-01 2.64454782e-01
-2.00696960e-01 2.26883531e-01 1.11207330e+00 -5.79347908e-01
-1.87030661e+00 -6.40499651e-01 1.81229755e-01 -4.39485669e-01
1.18686700e+00 -4.86007065e-01 -5.72679937e-01 4.97252584e-01
1.24360494e-01 1.51742876e-01 4.60366979e-02 3.92182320e-02
1.01116590e-01 4.25025113e-02 -1.10327780e+00 6.42630100e-01
7.69767940e-01 -2.92768747e-01 9.55699105e-03 9.88989413e-01
3.81215900e-01 -4.23836321e-01 -9.78876889e-01 -1.80966765e-01
4.17855352e-01 -8.78288627e-01 1.04621768e+00 -1.00095272e+00
2.09998816e-01 -6.65399656e-02 2.55671352e-01 -1.66652930e+00
-2.06801385e-01 -1.53062928e+00 -4.88248587e-01 7.35145509e-01
2.47555256e-01 -9.88927305e-01 6.65522754e-01 5.46380639e-01
2.51601130e-01 -9.94576931e-01 -1.38375854e+00 -8.52147639e-01
6.19944811e-01 -4.71387982e-01 4.13318992e-01 5.95494270e-01
3.86381545e-03 3.58422160e-01 -6.43139124e-01 -4.46031570e-01
7.57171690e-01 -2.39829466e-01 4.54865456e-01 -1.14128768e+00
-5.63553214e-01 -6.00016415e-01 -2.17351178e-03 -7.48821378e-01
5.02774954e-01 -8.87528360e-01 -2.89100390e-02 -1.15091014e+00
-2.08171561e-01 -6.41481578e-01 -7.12702274e-02 -7.80799165e-02
3.11884254e-01 -4.67941165e-02 2.46912464e-01 3.96775603e-01
-8.93102527e-01 1.06580412e+00 1.72869158e+00 3.22889574e-02
-2.95438111e-01 4.21696931e-01 -2.66602337e-02 4.64929283e-01
7.41360366e-01 -6.70319259e-01 -2.65657455e-01 1.29977688e-01
7.85571277e-01 5.53551257e-01 3.81779641e-01 -7.63894796e-01
1.39535353e-01 -4.72878546e-01 -3.95365851e-03 -1.17999226e-01
5.02135873e-01 -3.43780071e-01 2.12840978e-02 6.57870233e-01
-7.01290488e-01 4.53103960e-01 -6.34925487e-03 8.54360521e-01
3.24501395e-01 -3.38383406e-01 7.86092699e-01 -6.46178961e-01
8.73102620e-02 4.31427628e-01 -9.96601224e-01 4.66955394e-01
7.94340491e-01 2.67654508e-01 -2.32266754e-01 -4.79886532e-01
-1.30677760e+00 2.95628965e-01 1.87487602e-01 -2.17677355e-01
4.10409123e-01 -1.17284441e+00 -5.33011436e-01 -3.61904129e-02
-4.53744262e-01 -1.86787337e-01 2.91018151e-02 1.19701040e+00
-5.51395297e-01 3.44292641e-01 -4.54509258e-02 -2.73546249e-01
-4.70099121e-01 5.84679902e-01 8.14924419e-01 -8.43162239e-01
-5.50294399e-01 4.21481371e-01 -3.59741032e-01 -6.19558811e-01
-5.75772300e-03 -3.53024095e-01 -1.08852431e-01 -6.23751916e-02
3.05029094e-01 8.26448023e-01 -1.62308574e-01 -5.30956149e-01
2.39767227e-02 4.42412257e-01 -1.94654241e-02 -4.87971365e-01
1.62581658e+00 8.86854306e-02 -3.42787266e-01 8.15890074e-01
5.60870767e-01 -4.97834146e-01 -1.49169445e+00 -1.02657080e-01
1.27489686e-01 2.67872900e-01 1.72028244e-01 -6.04100347e-01
-8.53497505e-01 8.36001694e-01 3.13506216e-01 7.18852043e-01
4.18463558e-01 -1.90000281e-01 5.92992663e-01 5.66442966e-01
3.04259479e-01 -1.24510443e+00 2.00427487e-01 8.58730257e-01
8.31992626e-01 -1.06305385e+00 2.65519679e-01 2.51492977e-01
-6.90849483e-01 1.21440923e+00 2.51504958e-01 -6.89440727e-01
6.66103601e-01 3.46586891e-02 -3.67106587e-01 -4.65166420e-01
-1.06310308e+00 -3.82452965e-01 2.39902139e-01 5.31640112e-01
1.41097769e-01 7.92785659e-02 -3.62745255e-01 -4.19363886e-01
1.97569519e-01 -4.07025069e-02 7.10108399e-01 8.50558400e-01
-1.70555994e-01 -1.31441653e+00 -2.55907118e-01 5.28901875e-01
-3.30103576e-01 1.29051783e-04 -7.19146878e-02 9.14464533e-01
-4.04179424e-01 7.36683309e-01 -3.31769615e-01 1.23387456e-01
-9.28311143e-03 6.23188242e-02 7.81548440e-01 -2.88820058e-01
-1.09287000e+00 1.98522419e-01 2.09508881e-01 -6.95771933e-01
-7.24992096e-01 -1.10369265e+00 -9.37460721e-01 -7.01790392e-01
-2.79923290e-01 3.26848000e-01 6.54406548e-01 1.31287014e+00
1.94570929e-01 3.83875817e-01 5.29130161e-01 -1.50273681e+00
-1.07809198e+00 -8.27906251e-01 -7.81595170e-01 -2.09387578e-03
5.18509030e-01 -9.49952006e-01 -5.26333034e-01 -6.92583144e-01] | [4.1019697189331055, 2.165006160736084] |
e3be76c5-d35c-4c18-b4ca-68745428b387 | ssn-sparks-at-semeval-2019-task-9-mining | null | null | https://aclanthology.org/S19-2217 | https://aclanthology.org/S19-2217.pdf | SSN-SPARKS at SemEval-2019 Task 9: Mining Suggestions from Online Reviews using Deep Learning Techniques on Augmented Data | This paper describes the work on mining the suggestions from online reviews and forums. Opinion mining detects whether the comments are positive, negative or neutral, while suggestion mining explores the review content for the possible tips or advice. The system developed by SSN-SPARKS team in SemEval-2019 for task 9 (suggestion mining) uses a rule-based approach for feature selection, SMOTE technique for data augmentation and deep learning technique (Convolutional Neural Network) for classification. We have compared the results with Random Forest classifier (RF) and MultiLayer Perceptron (MLP) model. Results show that the CNN model performs better than other models for both the subtasks. | ['S Milton Rajendram', 'Mirnalinee T T', 'Angel Suseelan', 'Rajalakshmi S'] | 2019-06-01 | null | null | null | semeval-2019-6 | ['suggestion-mining'] | ['natural-language-processing'] | [-1.82976097e-01 4.79299575e-01 -5.47497928e-01 -7.00820565e-01
3.07845503e-01 -2.82228500e-01 9.14814353e-01 7.16280460e-01
-2.83921748e-01 8.95441890e-01 6.36175156e-01 -1.20001781e+00
-6.81183068e-04 -8.68241966e-01 -2.05136333e-02 -2.67599106e-01
-1.70337796e-01 2.04786435e-01 2.53917021e-03 -7.85077333e-01
1.15412712e+00 -6.99609285e-03 -1.74086225e+00 1.03551638e+00
4.86595422e-01 1.12382495e+00 -3.23173791e-01 9.83453393e-01
-6.11325383e-01 1.14597261e+00 -6.09153450e-01 -1.74760491e-01
3.67660761e-01 1.97806090e-01 -1.00057673e+00 -3.32534850e-01
2.17237756e-01 2.57807821e-01 3.37224543e-01 6.92225277e-01
3.98271054e-01 3.90298218e-01 6.82561398e-01 -1.21126664e+00
-6.58088267e-01 9.31163132e-01 -8.35824788e-01 7.59831071e-01
3.25384796e-01 -4.76618558e-01 9.80127156e-01 -1.22052503e+00
4.45688248e-01 1.24876261e+00 6.66594803e-01 2.92755842e-01
-5.57926476e-01 -7.87558317e-01 3.56761783e-01 4.89812106e-01
-4.61471796e-01 -2.75872141e-01 5.98808944e-01 -5.11921644e-01
1.87648761e+00 2.18397066e-01 6.17628515e-01 4.91237789e-01
6.06721818e-01 7.38192320e-01 1.42618167e+00 -7.47738063e-01
3.18425477e-01 8.20963085e-01 1.08277428e+00 3.88294876e-01
4.13465500e-02 1.60057291e-01 -7.34405756e-01 -4.00104940e-01
-2.15942144e-01 1.90512925e-01 4.44360137e-01 6.10146284e-01
-6.85269058e-01 1.12024021e+00 3.10420454e-01 1.74739435e-01
-8.09910655e-01 -4.84892666e-01 8.14330995e-01 1.16315687e+00
9.56552446e-01 5.66037178e-01 -1.18167770e+00 1.36211529e-01
-7.47283995e-01 3.07056010e-01 1.26441753e+00 2.68841654e-01
9.46100533e-01 1.83030233e-01 -1.55494153e-01 8.56607378e-01
5.81992209e-01 -2.72877216e-01 1.00863528e+00 -2.52878368e-01
3.28953832e-01 9.96359825e-01 1.80760384e-01 -1.13932741e+00
-5.79472899e-01 -6.13407969e-01 -6.71136916e-01 5.55243492e-01
-4.00259376e-01 -1.08831453e+00 -1.10875249e+00 5.43075919e-01
2.50509143e-01 -3.48699003e-01 2.23386869e-01 3.25618505e-01
1.47545707e+00 6.14245236e-01 3.05992160e-02 -2.54694015e-01
1.18962777e+00 -1.42300868e+00 -7.41210401e-01 -3.05789918e-01
6.30145967e-01 -1.05332410e+00 4.66866702e-01 9.13922131e-01
-7.49951124e-01 -8.59961748e-01 -1.21273017e+00 3.41567636e-01
-9.74059522e-01 1.12237535e-01 1.06035423e+00 5.11743903e-01
-1.22299170e+00 6.16420925e-01 -1.15845747e-01 -4.42624569e-01
2.84058988e-01 8.92343163e-01 -3.78142476e-01 3.97925109e-01
-1.08915973e+00 9.70027924e-01 1.50717378e-01 4.86325435e-02
-8.72564167e-02 -2.01765046e-01 -4.98984605e-01 -1.19870983e-01
-2.00762272e-01 -2.50696063e-01 1.44466686e+00 -1.54446888e+00
-1.49819112e+00 6.73187375e-01 -2.68782586e-01 -7.48038828e-01
-4.13073879e-03 -3.01537424e-01 -7.58115470e-01 -4.22477484e-01
2.11926941e-02 7.75759220e-01 8.19116116e-01 -6.48930907e-01
-1.18265724e+00 -3.57287109e-01 1.09479494e-01 9.01841521e-02
-2.62522161e-01 5.04219770e-01 6.31669283e-01 -2.62481868e-01
-1.81717753e-01 -6.83286965e-01 -6.26618087e-01 -1.11811352e+00
-7.39647627e-01 -8.66521597e-01 8.91979039e-01 -5.55990577e-01
1.22435367e+00 -1.57519913e+00 -8.77294838e-01 5.11268020e-01
1.60890907e-01 4.17513639e-01 2.35245243e-01 7.33263373e-01
-5.30821383e-01 3.99817199e-01 5.69523096e-01 -1.36040049e-02
-4.05105233e-01 -1.16150618e-01 -1.67120114e-01 -2.78797559e-02
9.20726061e-02 6.20418072e-01 -4.62811559e-01 -2.96293795e-01
4.86870781e-02 1.50165260e-01 -1.23666950e-01 8.01111311e-02
-1.28432274e-01 9.35515575e-03 -3.83687049e-01 5.65161169e-01
5.31708658e-01 -5.94188049e-02 -1.77056566e-01 1.42512545e-01
-5.22641301e-01 8.77142787e-01 -1.07966769e+00 6.61961198e-01
-3.84786993e-01 9.67018843e-01 7.66989635e-03 -1.27844787e+00
1.74819005e+00 4.48701471e-01 -1.30667165e-01 -4.54482079e-01
1.93698391e-01 1.35981873e-01 3.42812002e-01 -7.69261539e-01
4.81466383e-01 -2.58005917e-01 5.97541511e-01 9.65335667e-01
-2.35332951e-01 6.65750146e-01 1.65759549e-01 3.00956756e-01
1.23968661e+00 -3.42050552e-01 5.63497245e-01 -5.34278333e-01
8.26863468e-01 4.03317869e-01 3.49234909e-01 8.66459250e-01
-1.65874615e-01 9.46114585e-02 4.33523118e-01 -1.02520692e+00
-5.42837024e-01 -3.18926483e-01 3.23909014e-01 1.48743463e+00
-6.92832351e-01 -4.92543310e-01 -1.43315330e-01 -1.10830462e+00
8.43369663e-02 6.81513250e-01 -1.09123242e+00 2.50208437e-01
-1.33585647e-01 -5.64027488e-01 -2.39397734e-01 4.28947568e-01
4.48711663e-01 -1.73058498e+00 -2.65526623e-01 2.15865165e-01
5.14150202e-01 -1.91544607e-01 3.02818060e-01 9.74563479e-01
-1.14354992e+00 -1.13821864e+00 7.92083144e-02 -1.00717854e+00
5.86641788e-01 3.61431777e-01 1.26346421e+00 3.83304566e-01
-3.99599858e-02 -8.51852223e-02 -7.35257566e-01 -1.41030693e+00
-1.59946144e-01 2.82416970e-01 -8.22471306e-02 -3.14772278e-01
1.20600712e+00 -7.15135336e-01 -7.79004812e-01 -1.00016773e-01
-1.88010857e-01 -1.16365038e-01 7.19755650e-01 5.92669249e-01
-9.58836153e-02 4.48945276e-02 1.33520377e+00 -1.62658417e+00
1.49359298e+00 -9.91816640e-01 1.43545359e-01 -3.18803579e-01
-1.47343755e+00 -9.74126011e-02 3.69285583e-01 -1.11359119e-01
-9.81629074e-01 -7.52553344e-02 -1.58522666e-01 5.15280247e-01
-4.97249186e-01 9.26517189e-01 8.00756812e-01 2.91437835e-01
1.30865109e+00 -7.50129819e-02 1.39128223e-01 -5.46934545e-01
3.78005989e-02 1.23465848e+00 -1.36434305e-02 1.04910210e-01
3.58076096e-01 3.40052664e-01 -5.33133030e-01 -5.91565728e-01
-9.21571434e-01 -8.89935136e-01 -4.89112586e-01 -2.98846543e-01
3.06358933e-01 -6.29327059e-01 -4.72042173e-01 -1.02583095e-01
-1.09590042e+00 2.63717383e-01 -1.03485174e-01 3.58551443e-01
1.05385035e-01 -2.01770872e-01 -6.10994458e-01 -1.31613266e+00
-1.47857928e+00 -4.27281976e-01 1.11676164e-01 5.77348053e-01
-7.77487457e-01 -9.24662650e-01 3.22066128e-01 4.24356580e-01
8.96147549e-01 -8.86380151e-02 7.88988173e-01 -1.64801431e+00
3.32514912e-01 -7.36810982e-01 -1.06926091e-01 6.86693311e-01
-1.01045482e-01 2.47833818e-01 -1.03261948e+00 3.30514342e-01
-1.36445612e-01 -3.19388866e-01 1.23624861e+00 4.32243943e-01
8.19939077e-01 -7.67137289e-01 -2.22507611e-01 -4.13929939e-01
1.06953824e+00 3.46577555e-01 4.84597325e-01 9.06821966e-01
1.62943274e-01 8.69761467e-01 6.20437086e-01 3.85468334e-01
3.11837822e-01 -2.94043601e-01 3.30305070e-01 3.23210619e-02
3.77207041e-01 1.37283579e-01 5.68739235e-01 7.40238428e-01
4.99906316e-02 -2.27673035e-02 -7.57163048e-01 5.89917719e-01
-2.17408538e+00 -7.69009292e-01 -7.36234963e-01 1.37812686e+00
6.75778747e-01 6.05075836e-01 3.27642590e-01 6.68061852e-01
6.63040280e-01 2.63146162e-02 -2.79581040e-01 -1.81125665e+00
6.97721839e-02 6.00102603e-01 3.09777558e-01 4.67447519e-01
-1.11459577e+00 6.93575621e-01 6.56779957e+00 2.12816060e-01
-1.16610491e+00 2.70437226e-02 1.08922076e+00 6.76766783e-02
-1.94238007e-01 5.11074252e-02 -8.85155141e-01 -2.54602320e-02
1.23573983e+00 -2.49896228e-04 -1.69598639e-01 1.32124531e+00
5.79578221e-01 -4.75423634e-01 -7.45327055e-01 2.81892747e-01
-1.49623737e-01 -1.86830997e+00 -3.07853192e-01 -1.31763205e-01
1.06279826e+00 7.72808313e-01 -2.00190321e-01 5.82963824e-01
7.18114555e-01 -1.04556751e+00 2.03667417e-01 4.49932128e-01
-2.96668440e-01 -9.27363575e-01 1.47731137e+00 4.86340553e-01
-4.05651093e-01 -3.08517128e-01 -4.82264876e-01 -1.26075006e+00
-3.17568123e-01 1.10762012e+00 -1.45350468e+00 2.21415721e-02
1.00492430e+00 1.07778454e+00 -6.09540522e-01 8.93519878e-01
-3.90877396e-01 1.23365080e+00 -5.45263626e-02 -7.46470988e-01
2.66606420e-01 8.53476673e-02 2.13849366e-01 1.65772998e+00
-2.32027858e-01 -2.40725368e-01 1.12061046e-01 8.40243697e-02
1.97358534e-01 6.96130931e-01 -5.37410200e-01 2.98681259e-01
1.24881402e-01 1.79970932e+00 -6.90221727e-01 -4.86495286e-01
-4.40965354e-01 4.66480106e-02 7.49518424e-02 2.62944818e-01
3.89119059e-01 -6.23445749e-01 2.89782137e-01 4.10794497e-01
3.17419887e-01 3.98482174e-01 -9.61428940e-01 -4.41387445e-01
-2.69217938e-01 -9.03244019e-01 4.26332474e-01 -5.74246705e-01
-1.39113438e+00 7.32327223e-01 -4.79643255e-01 -6.53607190e-01
-2.39494234e-01 -6.77076697e-01 -1.85395277e+00 8.87666702e-01
-1.47036719e+00 -7.95693159e-01 -8.15731287e-03 3.81383002e-01
8.23448122e-01 -7.33674347e-01 8.54710281e-01 -5.37809730e-02
-3.62206310e-01 2.61646844e-02 -2.48476744e-01 -1.52504086e-01
6.19171381e-01 -1.63965702e+00 4.76285577e-01 1.76261917e-01
-6.12072609e-02 7.88421333e-01 9.27566886e-01 -6.50823236e-01
-9.69134092e-01 -7.34796643e-01 1.44179857e+00 -1.76985189e-01
6.15990937e-01 1.52745798e-01 -6.10741377e-01 3.78396958e-01
8.96508276e-01 -4.88188565e-01 1.31701684e+00 7.93945074e-01
-2.33522445e-01 -1.81778073e-01 -1.04594684e+00 1.57322317e-01
7.03743380e-03 -5.06960340e-02 -8.37385893e-01 5.75305521e-01
5.02123177e-01 3.90128344e-01 -4.69825864e-01 2.94974893e-01
6.81375682e-01 -1.37143850e+00 2.96144783e-01 -1.13968694e+00
8.97167325e-01 -9.86204073e-02 1.84626088e-01 -1.37090349e+00
-2.01694369e-01 -5.80697536e-01 -3.04288089e-01 1.07600260e+00
1.31289947e+00 -4.78141099e-01 1.18418515e+00 5.11339545e-01
-6.95003942e-02 -1.21920323e+00 -3.26167047e-01 2.25846767e-01
-1.27387047e-01 -6.83324933e-01 1.74828574e-01 8.54810417e-01
6.04988694e-01 1.23085821e+00 -1.84496328e-01 -8.03885877e-01
-3.13121200e-01 -1.90074071e-02 8.73414695e-01 -1.65328479e+00
-3.77576202e-02 -4.86953914e-01 -2.36711591e-01 -5.17125428e-01
-2.68813912e-02 -6.73015535e-01 -1.32143259e-01 -1.92171133e+00
-4.25649174e-02 -1.42766789e-01 -7.90506363e-01 5.74129820e-01
-4.55148630e-02 -2.23974094e-01 -2.66965121e-01 8.10797960e-02
-6.72486484e-01 -1.28856614e-01 8.45586479e-01 -2.02774569e-01
-3.67058605e-01 8.00769389e-01 -1.51446354e+00 8.34105670e-01
1.26974642e+00 -5.13271630e-01 -3.21337849e-01 3.18140119e-01
1.13868999e+00 -5.10716081e-01 -2.92079449e-01 -4.50632453e-01
4.14638758e-01 9.97060910e-02 4.65264320e-01 -1.32828462e+00
-3.17110538e-01 -3.34013194e-01 -7.06321418e-01 4.63238686e-01
-8.82178843e-01 5.64365208e-01 8.73423889e-02 6.76992536e-01
-3.78629506e-01 -4.18114990e-01 2.07863227e-01 -4.78080124e-01
-3.28872353e-01 -1.22331493e-01 -1.07602155e+00 -7.12957621e-01
4.54646409e-01 -3.14314038e-01 -6.36391103e-01 -7.68881261e-01
-7.58973300e-01 4.55681294e-01 -3.59702170e-01 7.09856391e-01
7.76543200e-01 -6.44594371e-01 -9.12731886e-01 2.05063149e-01
6.23030774e-02 -2.26813108e-01 -3.85181636e-01 8.52315247e-01
-3.38920981e-01 4.79999244e-01 -1.45118311e-01 2.43785605e-02
-1.42147827e+00 3.08221310e-01 8.67902562e-02 -4.55416769e-01
-3.41479272e-01 1.19346941e+00 -6.03644788e-01 -7.12480128e-01
4.44633424e-01 -4.18736078e-02 -1.49736142e+00 3.19804966e-01
9.83428717e-01 7.63877809e-01 5.68903804e-01 -4.70607989e-02
-2.88595587e-01 -1.48662597e-01 -7.66224980e-01 -6.97422549e-02
1.78037727e+00 1.24529973e-01 -5.93710899e-01 5.84601164e-01
9.70448911e-01 -1.44817680e-01 -2.36490026e-01 -1.60312667e-01
6.18339479e-01 1.69663548e-01 3.64121020e-01 -1.35298121e+00
-9.61721480e-01 8.33700359e-01 5.22301257e-01 7.84785330e-01
6.45195901e-01 -4.15864170e-01 4.27676678e-01 1.02730417e+00
-4.78158444e-01 -1.69081056e+00 -1.24031417e-01 8.18721771e-01
8.84449482e-01 -1.78523481e+00 3.12660605e-01 3.72568257e-02
-7.74310112e-01 1.58474076e+00 9.22431648e-01 -6.83434963e-01
1.75684357e+00 1.99726939e-01 4.97346342e-01 -7.34676540e-01
-1.74123704e+00 2.76659757e-01 2.85261482e-01 8.61850441e-01
1.16313851e+00 1.60865501e-01 -8.46499324e-01 8.34161222e-01
-4.91340578e-01 1.24481201e-01 6.04213476e-01 1.04211462e+00
-1.14897346e+00 -9.41695392e-01 -1.74460173e-01 1.46852624e+00
-8.91399562e-01 -3.89827132e-01 -1.04677308e+00 2.54438072e-01
3.09423953e-01 1.74322712e+00 -3.64191346e-02 -8.73226821e-01
8.20885450e-02 1.93752795e-01 -6.62825048e-01 -1.13303089e+00
-1.78085256e+00 -1.29396796e-01 8.30549479e-01 -4.93842177e-02
-5.43864727e-01 -5.19700468e-01 -1.07560337e+00 -3.17084342e-01
-7.99784243e-01 8.61208022e-01 1.22993743e+00 1.15908980e+00
4.72815901e-01 3.67441654e-01 8.08731556e-01 -3.88734937e-01
-2.31590971e-01 -1.68670332e+00 -3.41613948e-01 -1.22910820e-01
1.89355999e-01 -2.40877166e-01 -3.68785769e-01 -1.12610400e-01] | [10.98051929473877, 7.283262729644775] |
cc3a3a24-53f6-4ea8-9fc8-874d35df12ae | medtype-improving-medical-entity-linking-with | 2005.00460 | null | https://arxiv.org/abs/2005.00460v4 | https://arxiv.org/pdf/2005.00460v4.pdf | Improving Broad-Coverage Medical Entity Linking with Semantic Type Prediction and Large-Scale Datasets | Medical entity linking is the task of identifying and standardizing medical concepts referred to in an unstructured text. Most of the existing methods adopt a three-step approach of (1) detecting mentions, (2) generating a list of candidate concepts, and finally (3) picking the best concept among them. In this paper, we probe into alleviating the problem of overgeneration of candidate concepts in the candidate generation module, the most under-studied component of medical entity linking. For this, we present MedType, a fully modular system that prunes out irrelevant candidate concepts based on the predicted semantic type of an entity mention. We incorporate MedType into five off-the-shelf toolkits for medical entity linking and demonstrate that it consistently improves entity linking performance across several benchmark datasets. To address the dearth of annotated training data for medical entity linking, we present WikiMed and PubMedDS, two large-scale medical entity linking datasets, and demonstrate that pre-training MedType on these datasets further improves entity linking performance. We make our source code and datasets publicly available for medical entity linking research. | ['Denis Newman-Griffis', 'Ritam Dutt', 'Rishabh Joshi', 'Shikhar Vashishth', 'Carolyn Rose'] | 2020-05-01 | null | null | null | null | ['type-prediction'] | ['computer-code'] | [ 1.59328759e-01 5.18511295e-01 -3.99084747e-01 -2.02704608e-01
-1.10296714e+00 -6.52421057e-01 2.84796715e-01 1.16946018e+00
-6.17029190e-01 9.64144289e-01 3.58148545e-01 -2.50973552e-01
-1.44243419e-01 -7.36250520e-01 -4.66115296e-01 -2.01469079e-01
1.00502610e-01 9.29669321e-01 4.84808296e-01 -8.23401585e-02
2.87126321e-02 4.70141843e-02 -1.18961656e+00 3.40802997e-01
1.33494782e+00 2.93277562e-01 -4.89099771e-02 3.08652312e-01
-2.75107563e-01 3.80615860e-01 -6.19458556e-01 -7.05342174e-01
-3.04833710e-01 -4.77778375e-01 -1.16004860e+00 -5.30679584e-01
2.30483890e-01 2.96519101e-01 7.50922710e-02 1.07790458e+00
6.28442466e-01 -3.40578914e-01 5.75003505e-01 -1.11423397e+00
-6.71373188e-01 9.95749176e-01 -3.38924110e-01 1.15315817e-01
4.71666276e-01 -3.50519419e-01 1.12868071e+00 -7.97265708e-01
1.27282548e+00 8.09073746e-01 9.19096231e-01 8.13747823e-01
-9.79666352e-01 -6.72130644e-01 -2.89447129e-01 -1.93675265e-01
-1.74003029e+00 -3.66394103e-01 9.57405567e-02 -6.12784505e-01
1.03407228e+00 2.96331465e-01 2.90242821e-01 6.88836634e-01
-1.49776652e-01 3.67268622e-01 5.75936198e-01 -3.67029428e-01
1.49363935e-01 4.24999781e-02 4.10118788e-01 8.95564079e-01
8.22804153e-01 -3.56494635e-01 -2.10979387e-01 -8.13981950e-01
5.38066886e-02 -3.36393893e-01 -2.22709045e-01 -2.35841461e-02
-1.40179241e+00 5.85154355e-01 3.57834399e-01 4.01012331e-01
-4.27201569e-01 -2.65829116e-01 5.53761244e-01 -4.79802974e-02
5.40875912e-01 1.00270283e+00 -6.61352038e-01 2.16821849e-01
-9.56323743e-01 1.76444933e-01 8.74266744e-01 1.10140359e+00
5.48003495e-01 -8.72896492e-01 -3.13561946e-01 8.33960950e-01
9.90594178e-02 -6.27174927e-03 6.19888365e-01 -5.73864162e-01
4.63798195e-01 9.55045402e-01 1.30326539e-01 -7.19018221e-01
-7.17078626e-01 -3.46726090e-01 -3.80958527e-01 -4.94365335e-01
2.78018326e-01 -3.13302904e-01 -8.78513515e-01 1.82466209e+00
8.25420439e-01 3.95731002e-01 4.16545838e-01 5.19743979e-01
1.54061222e+00 1.09186217e-01 8.19632888e-01 -7.86354989e-02
1.80998933e+00 -6.91620708e-01 -7.88742065e-01 1.42755613e-01
1.08867729e+00 -9.18995380e-01 4.65664804e-01 -2.68018723e-01
-9.98637855e-01 -1.51204884e-01 -7.78219163e-01 -1.97443768e-01
-8.26633334e-01 7.18688816e-02 5.99658668e-01 3.02145451e-01
-8.90324056e-01 7.18108237e-01 -8.45823348e-01 -6.33811057e-01
4.29522425e-01 3.66752326e-01 -7.00974286e-01 -4.97606210e-02
-1.46279263e+00 1.03750861e+00 7.12824464e-01 -3.57830226e-01
-1.09707221e-01 -1.18958235e+00 -9.19314146e-01 2.62885932e-02
3.92664880e-01 -1.27858603e+00 9.19030845e-01 -4.39342469e-01
-4.73419398e-01 1.20637083e+00 -1.56539112e-01 -3.87628734e-01
1.72590643e-01 -1.54080629e-01 -8.46646488e-01 2.33109519e-01
7.02711999e-01 9.55946982e-01 6.39351457e-03 -9.06720221e-01
-7.82051325e-01 -2.37931430e-01 -1.42881244e-01 1.24577947e-01
-1.99348748e-01 1.33418545e-01 -7.51282156e-01 -6.51759565e-01
7.13320449e-02 -8.09641182e-01 -4.80120778e-01 -2.44137198e-01
-7.67890275e-01 -4.38669592e-01 1.97500393e-01 -7.75482118e-01
1.45234632e+00 -1.85162640e+00 -4.46301252e-02 1.66354075e-01
4.70338434e-01 3.76935869e-01 1.47753125e-02 4.56138194e-01
-3.34252119e-01 4.19690937e-01 -3.03801119e-01 -1.44459397e-01
-2.63705641e-01 1.47607818e-01 3.39061804e-02 9.10251439e-02
3.85440260e-01 1.06233156e+00 -1.34648216e+00 -9.50096488e-01
-3.91819447e-01 5.23356855e-01 -6.43186986e-01 -3.81964222e-02
-2.28998020e-01 3.63960922e-01 -6.52700901e-01 7.98302472e-01
3.88594985e-01 -5.04484594e-01 4.43011701e-01 -4.28107291e-01
1.67452514e-01 7.04544961e-01 -1.02171254e+00 1.76640856e+00
9.76967216e-02 3.64707336e-02 -2.92381108e-01 -5.17585218e-01
6.02683783e-01 6.35556042e-01 9.92401421e-01 -2.32621372e-01
-1.31149977e-01 4.70759928e-01 -1.31436855e-01 -7.18278050e-01
7.76101768e-01 4.41304818e-02 -2.89532274e-01 3.20786834e-01
7.14400783e-02 4.09267753e-01 5.29655159e-01 5.54292202e-01
1.48303664e+00 6.39700098e-03 7.36249030e-01 -2.51163781e-01
5.12523472e-01 4.66496140e-01 7.48711288e-01 4.89425659e-01
2.48073135e-02 4.63381171e-01 3.74058336e-01 -1.16613060e-01
-7.04118729e-01 -9.73881185e-01 -3.27469170e-01 6.90784693e-01
-8.22691433e-03 -7.53403604e-01 -8.02500784e-01 -1.12570310e+00
2.48062000e-01 4.36700165e-01 -6.84374273e-01 -3.36839333e-02
-5.55280387e-01 -1.11135375e+00 9.92877781e-01 3.84639680e-01
-1.90418009e-02 -8.72118354e-01 -4.77762818e-01 3.72096717e-01
-5.71813464e-01 -1.11157596e+00 -6.16629303e-01 6.82560503e-02
-5.88862658e-01 -1.53225338e+00 -5.44497907e-01 -8.01721394e-01
1.16654718e+00 -1.80315763e-01 1.51722825e+00 4.68386769e-01
-4.80458766e-01 2.17134342e-01 -3.85534644e-01 -3.44401211e-01
-5.24574339e-01 4.80696738e-01 -1.71979651e-01 -6.51093900e-01
7.53770411e-01 -8.75252411e-02 -5.71276188e-01 1.86759934e-01
-8.04682314e-01 -4.46021296e-02 4.64385331e-01 7.31204510e-01
8.09970140e-01 -2.11905822e-01 7.69049108e-01 -1.59395933e+00
5.63491285e-01 -8.83670986e-01 -2.34402537e-01 5.60567021e-01
-7.60234952e-01 2.58265942e-01 2.26815701e-01 -1.84242710e-01
-8.26789320e-01 2.64659405e-01 -5.65765560e-01 3.55772525e-01
-1.43378958e-01 8.78106236e-01 3.93807553e-02 2.62756407e-01
7.03701258e-01 -2.84204990e-01 -3.95008266e-01 -6.28913641e-01
5.23327649e-01 4.82834548e-01 7.75617123e-01 -4.86296445e-01
6.64113283e-01 2.25112483e-01 -2.14049853e-02 -1.68201804e-01
-1.02400482e+00 -8.00026000e-01 -7.87685156e-01 4.48882848e-01
1.11564767e+00 -1.13481164e+00 -4.60748106e-01 -2.28160471e-01
-1.00625002e+00 1.00323379e-01 -2.92966962e-01 4.12454695e-01
6.00576364e-02 2.72759438e-01 -5.78296006e-01 -1.52846053e-02
-7.03309596e-01 -8.49195540e-01 1.16169906e+00 2.68912137e-01
-7.39396334e-01 -9.63453829e-01 4.24429029e-01 2.74961025e-01
3.64597291e-02 5.19606590e-01 1.05033195e+00 -1.19302487e+00
-9.60304290e-02 -2.91872472e-01 -1.40113562e-01 -5.01812637e-01
2.65423805e-01 9.82753467e-03 -5.74483693e-01 -9.25025567e-02
-7.93256223e-01 -4.71689040e-03 9.16619897e-01 9.56245661e-02
8.28729987e-01 -2.39102796e-01 -1.12387919e+00 5.68508565e-01
1.33875144e+00 -2.67211981e-02 3.66803288e-01 4.69822913e-01
7.22260535e-01 6.33811176e-01 8.36529791e-01 6.70448989e-02
8.08237910e-01 6.47184789e-01 4.62091044e-02 -4.44491535e-01
-1.20884582e-01 -3.28857690e-01 -3.57368231e-01 7.85941482e-01
1.80479616e-01 -3.41506511e-01 -1.07999027e+00 9.03468192e-01
-1.77410257e+00 -8.00348401e-01 -2.73121655e-01 1.97788346e+00
1.53506768e+00 -3.34664360e-02 1.19094968e-01 -2.99363852e-01
9.04703438e-01 -6.55428767e-01 -3.45403701e-01 9.65351164e-02
-2.81352028e-02 4.89030153e-01 6.24802589e-01 4.90340233e-01
-1.19253528e+00 1.06297588e+00 5.98785639e+00 4.95978594e-01
-6.18136883e-01 3.97903442e-01 3.15223426e-01 1.58972666e-01
-3.10787559e-01 6.53876215e-02 -1.21147168e+00 5.29212534e-01
7.86870480e-01 -3.35225046e-01 -4.21375245e-01 6.53439641e-01
-2.73866743e-01 9.12455656e-03 -1.13954556e+00 3.55676621e-01
-1.14749141e-01 -1.52910888e+00 1.97836254e-02 -1.49923041e-01
6.69927955e-01 1.13969848e-01 -3.47472221e-01 1.64798155e-01
6.77650869e-01 -9.49245989e-01 2.45385244e-01 5.12236118e-01
9.67280865e-01 -5.34284413e-01 9.26635385e-01 -2.47028783e-01
-1.29950142e+00 3.20769936e-01 -2.22952023e-01 7.87309945e-01
2.36494154e-01 8.57887208e-01 -1.23622024e+00 9.37854707e-01
5.89782298e-01 6.88002765e-01 -8.38973522e-01 1.37797117e+00
-3.40377957e-01 4.06454772e-01 -1.88968509e-01 3.85613441e-01
-1.82023868e-02 3.58363211e-01 4.30089712e-01 1.58267927e+00
3.42909187e-01 3.22923332e-01 2.10575432e-01 8.35096955e-01
-3.78342986e-01 3.21460128e-01 -2.46642604e-01 -3.80481295e-02
9.12814498e-01 1.48445761e+00 -9.60700274e-01 -5.20469308e-01
-3.12341720e-01 7.63091683e-01 3.75076920e-01 -1.62899010e-02
-8.46542537e-01 -7.25350022e-01 7.07164049e-01 -8.63477774e-03
9.47818160e-02 2.56665707e-01 -9.43003222e-02 -1.02267170e+00
-1.68458119e-01 -5.65601766e-01 1.10111785e+00 -5.16195416e-01
-1.32038069e+00 6.27273858e-01 -9.48298872e-02 -1.14757359e+00
-1.50560632e-01 -1.34001255e-01 -3.20296824e-01 8.85620654e-01
-1.59245050e+00 -1.11388421e+00 -3.26083481e-01 3.14836174e-01
1.89639479e-02 2.91287482e-01 1.12571335e+00 7.85673976e-01
-5.80606759e-01 8.92907619e-01 -1.59814507e-01 3.80125135e-01
1.28003836e+00 -1.39434826e+00 8.25795293e-01 5.32582760e-01
-1.44079968e-01 1.11291790e+00 6.08117223e-01 -1.19063830e+00
-7.08839118e-01 -1.42207456e+00 1.64218915e+00 -8.22067678e-01
6.36423111e-01 -1.58829197e-01 -1.25593984e+00 6.57238305e-01
-1.93841271e-02 -1.24569759e-01 1.13895166e+00 3.70147042e-02
-3.42578828e-01 4.63017106e-01 -1.35062230e+00 3.66416961e-01
1.16274524e+00 -1.84348509e-01 -8.80731702e-01 4.38214660e-01
7.96771526e-01 -5.30313432e-01 -1.30956781e+00 4.56068128e-01
3.73177826e-01 -2.65375167e-01 1.22532713e+00 -8.16623569e-01
3.52106482e-01 -5.25541186e-01 3.70655388e-01 -1.23985362e+00
-1.66719720e-01 -5.04060745e-01 -6.43196851e-02 1.52527785e+00
1.04342139e+00 -5.78186095e-01 6.16598427e-01 5.35936236e-01
-2.61735559e-01 -6.67257071e-01 -6.39327228e-01 -4.48778123e-01
3.49209495e-02 3.68049145e-01 6.37909830e-01 1.50039673e+00
3.91665697e-01 3.32142711e-01 1.71226248e-01 4.14327383e-01
4.14829284e-01 5.28282076e-02 4.17306930e-01 -1.52413929e+00
-1.51004195e-01 -1.96456775e-01 -1.59245506e-01 -3.64907980e-01
1.43742058e-02 -1.30080891e+00 8.08170214e-02 -1.96216059e+00
4.79067892e-01 -8.66680324e-01 -2.66867757e-01 1.00532091e+00
-9.26706553e-01 3.81995738e-01 -4.11600441e-01 4.26768631e-01
-8.93101156e-01 -8.89990628e-02 6.55351996e-01 2.95019448e-02
-1.91471905e-01 -4.22455668e-01 -1.07224226e+00 6.17635548e-01
5.37910461e-01 -1.00278604e+00 -1.56905070e-01 -4.48991120e-01
3.74632716e-01 -1.47868887e-01 1.03977509e-01 -8.39828491e-01
4.01301056e-01 5.76693006e-02 3.25449526e-01 -7.56971300e-01
-1.42286092e-01 -5.14591694e-01 3.17515522e-01 7.12126851e-01
-4.58912194e-01 2.49214545e-01 3.12103778e-01 3.50840807e-01
-1.75655201e-01 -2.52260596e-01 6.46292150e-01 -2.39051059e-01
-6.86558723e-01 1.63567483e-01 -1.20254137e-01 5.43995738e-01
9.86137986e-01 7.95196220e-02 -8.29050124e-01 4.10427481e-01
-8.73725057e-01 4.82250750e-01 5.28349936e-01 3.99079680e-01
1.45073712e-01 -1.16093111e+00 -8.17065179e-01 -2.81710982e-01
4.59527612e-01 -5.00830933e-02 -6.55270473e-04 8.83743167e-01
-3.52283299e-01 5.31335115e-01 -1.81779452e-02 -2.86887616e-01
-1.58941531e+00 5.99955738e-01 1.47576988e-01 -5.85521996e-01
-6.43981040e-01 1.02727830e+00 -3.44157480e-02 -6.09867394e-01
-5.15863262e-02 6.12754263e-02 -5.81040621e-01 2.20229030e-01
4.75751460e-01 1.34889618e-01 3.29908282e-01 -7.93184519e-01
-8.64790440e-01 4.84762400e-01 -3.17535728e-01 1.53715849e-01
1.33228636e+00 1.05756586e-02 -4.14890319e-01 -1.74180478e-01
9.58943844e-01 2.74222374e-01 -3.12099248e-01 -2.62060493e-01
6.36814773e-01 3.13573107e-02 -4.01624918e-01 -1.07071280e+00
-8.27004075e-01 1.60803393e-01 2.90434122e-01 -7.52095804e-02
1.07673693e+00 3.40949148e-01 8.32824230e-01 1.61790311e-01
5.28865345e-02 -9.92266834e-01 -3.94914895e-01 3.21910560e-01
2.88676023e-01 -1.07871246e+00 2.00159281e-01 -9.31537569e-01
-4.49765146e-01 7.60033250e-01 6.72152817e-01 4.39263523e-01
5.47710359e-01 3.16968650e-01 2.15074271e-02 -7.12755620e-01
-6.63307488e-01 -4.06777322e-01 6.58787072e-01 5.57532310e-01
9.63958919e-01 -4.95212041e-02 -8.55391800e-01 7.23208785e-01
-1.77767292e-01 1.79019272e-02 2.72517622e-01 9.90253985e-01
-2.66034901e-01 -1.38564885e+00 -1.19391643e-01 4.97405529e-01
-9.64222312e-01 -6.02514744e-01 -6.15721583e-01 6.20196581e-01
5.58735549e-01 7.26823926e-01 -6.39620945e-02 -6.54465258e-02
4.79454964e-01 1.42263830e-01 9.88759771e-02 -1.09981287e+00
-9.33891952e-01 -1.66650623e-01 4.76529539e-01 -3.13168854e-01
-5.60594559e-01 -6.18884742e-01 -1.65558600e+00 3.30361426e-02
-5.81825316e-01 5.87388933e-01 3.75878811e-01 9.72195327e-01
8.67217600e-01 7.33880162e-01 -1.26217902e-01 8.38987827e-02
7.10498691e-02 -6.97344422e-01 7.46676177e-02 7.04739213e-01
-1.01116166e-01 -6.58176601e-01 1.69446155e-01 1.70044318e-01] | [8.776663780212402, 8.817458152770996] |
a93c9f69-d10f-4954-b66f-abb7a286d185 | lightweight-wood-panel-defect-detection | 2306.12113 | null | https://arxiv.org/abs/2306.12113v1 | https://arxiv.org/pdf/2306.12113v1.pdf | Lightweight wood panel defect detection method incorporating attention mechanism and feature fusion network | In recent years, deep learning has made significant progress in wood panel defect detection. However, there are still challenges such as low detection , slow detection speed, and difficulties in deploying embedded devices on wood panel surfaces. To overcome these issues, we propose a lightweight wood panel defect detection method called YOLOv5-LW, which incorporates attention mechanisms and a feature fusion network.Firstly, to enhance the detection capability of acceptable defects, we introduce the Multi-scale Bi-directional Feature Pyramid Network (MBiFPN) as a feature fusion network. The MBiFPN reduces feature loss, enriches local and detailed features, and improves the model's detection capability for acceptable defects.Secondly, to achieve a lightweight design, we reconstruct the ShuffleNetv2 network model as the backbone network. This reconstruction reduces the number of parameters and computational requirements while maintaining performance. We also introduce the Stem Block and Spatial Pyramid Pooling Fast (SPPF) models to compensate for any accuracy loss resulting from the lightweight design, ensuring the model's detection capabilities remain intact while being computationally efficient.Thirdly, we enhance the backbone network by incorporating Efficient Channel Attention (ECA), which improves the network's focus on key information relevant to defect detection. By attending to essential features, the model becomes more proficient in accurately identifying and localizing defects.We validate the proposed method using a self-developed wood panel defect dataset.The experimental results demonstrate the effectiveness of the improved YOLOv5-LW method. Compared to the original model, our approach achieves a 92.8\% accuracy rate, reduces the number of parameters by 27.78\%, compresses computational volume by 41.25\%, improves detection inference speed by 10.16\% | ['Yang Chen', 'You Miao', 'Cheng Bao', 'Lai Jiang', 'Fanghua Liu', 'Yongxin Cao'] | 2023-06-21 | null | null | null | null | ['defect-detection'] | ['computer-vision'] | [ 3.31560791e-01 -1.87646315e-01 1.46418795e-01 2.22775787e-01
-3.94029409e-01 -9.56915468e-02 -1.25529200e-01 1.22550964e-01
-1.58138022e-01 7.42681175e-02 5.95372766e-02 3.44899371e-02
-2.33722642e-01 -1.06491661e+00 -4.60550308e-01 -7.28181064e-01
-3.54287704e-03 -4.96799409e-01 7.94678867e-01 -3.82512286e-02
2.67784536e-01 4.29090649e-01 -1.89082885e+00 4.72989321e-01
1.08087564e+00 1.43538892e+00 7.75775850e-01 4.55987364e-01
1.42154962e-01 5.86747408e-01 -5.72036147e-01 -2.33501762e-01
1.53057009e-01 1.91976622e-01 -4.26824987e-01 7.59826303e-02
3.44774544e-01 -7.96186149e-01 -4.21607941e-01 9.41018999e-01
8.54422152e-01 -2.78202146e-01 4.56353068e-01 -8.67675781e-01
-6.96108937e-01 3.02850127e-01 -9.15017366e-01 4.80689034e-02
1.71985388e-01 1.08591437e-01 1.05020249e+00 -9.61264670e-01
3.42605382e-01 1.10398209e+00 9.39609230e-01 2.67732859e-01
-1.06635606e+00 -4.00167495e-01 2.38926560e-01 2.27414027e-01
-1.32648492e+00 8.40956792e-02 9.48090971e-01 -1.84618711e-01
8.56019139e-01 3.06449145e-01 6.73635006e-01 6.00337684e-01
1.69704497e-01 1.07882547e+00 5.26812494e-01 -3.29875737e-01
-2.73034945e-02 -3.06094110e-01 1.67834073e-01 1.21668696e+00
3.02513987e-01 9.24694687e-02 -5.32909691e-01 7.37752318e-02
1.12297177e+00 2.65004396e-01 -6.54160261e-01 -1.57294035e-01
-9.43264306e-01 7.15099633e-01 8.97796631e-01 7.51074776e-02
-3.60618293e-01 7.48272091e-02 4.88143474e-01 1.04586750e-01
2.38458753e-01 1.22609213e-01 -5.56015968e-01 4.71094280e-01
-5.23090184e-01 2.35317238e-02 4.82283235e-01 6.99030519e-01
5.84258080e-01 1.22023150e-01 -2.83244640e-01 1.11970186e+00
3.19861174e-01 4.82453078e-01 4.53408718e-01 -8.53775978e-01
3.71762305e-01 1.01467991e+00 -2.11541787e-01 -1.38500416e+00
-4.02054280e-01 -5.14382660e-01 -1.19652236e+00 2.83460319e-01
-1.51579529e-02 7.83943385e-02 -1.03814423e+00 1.41676641e+00
6.59982190e-02 -2.50231236e-01 -4.13766831e-01 8.10731471e-01
9.03253376e-01 5.97092211e-01 -1.14719346e-02 2.14211211e-01
1.35917914e+00 -9.20845628e-01 -3.08371037e-01 -5.54436259e-02
5.63806772e-01 -7.43906260e-01 1.13991809e+00 4.05275881e-01
-7.58350313e-01 -1.02817452e+00 -1.19981396e+00 -7.08172331e-03
-1.89956859e-01 8.59530032e-01 7.55577028e-01 4.81482446e-01
-9.33112800e-01 4.91437554e-01 -7.48874485e-01 -2.41053373e-01
5.92717469e-01 6.54499352e-01 -3.02840084e-01 -3.98117900e-01
-6.87483132e-01 3.45692307e-01 3.13834220e-01 3.29357028e-01
-7.41106331e-01 -7.45380163e-01 -8.77898097e-01 3.73229057e-01
2.04977214e-01 -7.14103639e-01 7.67386794e-01 -5.13256371e-01
-1.39116871e+00 4.48110044e-01 9.71118547e-03 -1.29579037e-01
2.31115073e-01 -2.86451072e-01 -3.31892282e-01 2.45180190e-01
-1.63365752e-02 8.49207997e-01 7.27732658e-01 -1.08862269e+00
-9.56152499e-01 -4.01843011e-01 1.13957465e-01 7.95801505e-02
-5.49547911e-01 -3.40242088e-01 -4.25647527e-01 -8.20733964e-01
5.08324087e-01 -5.89268208e-01 -6.49538934e-02 4.23806161e-01
-4.41220224e-01 -1.17681839e-01 1.27448916e+00 -8.67784858e-01
1.31304038e+00 -2.48050284e+00 -4.52438146e-02 1.75810188e-01
4.13262665e-01 5.43556035e-01 -3.36426139e-01 4.39316720e-01
1.44647032e-01 2.99304388e-02 -3.61729354e-01 -1.28163889e-01
-2.96043247e-01 1.88605502e-01 -2.01036297e-02 2.85131544e-01
2.77209491e-01 7.63588250e-01 -2.47183636e-01 -2.75097787e-01
5.25579572e-01 5.02009451e-01 -8.31204951e-01 5.25175929e-02
1.99882463e-01 -5.27315512e-02 -5.37483156e-01 1.13347816e+00
8.64041328e-01 -1.04733862e-01 -2.99099833e-01 -6.64534330e-01
-4.85283025e-02 -3.19163263e-01 -1.20480156e+00 1.74093521e+00
-5.26405275e-01 4.21526372e-01 3.65747064e-01 -9.88479555e-01
1.15392315e+00 2.38493588e-02 4.41673309e-01 -6.50569022e-01
5.74685112e-02 7.29542673e-02 -9.77412388e-02 -5.66048741e-01
1.12748466e-01 3.11014295e-01 5.49298786e-02 -1.41397104e-01
1.49397142e-02 1.60595968e-01 -7.66022364e-03 -3.51630244e-03
1.37618780e+00 -5.51312380e-02 -6.40505329e-02 -1.75418586e-01
6.35163248e-01 -4.47540760e-01 7.71401346e-01 5.57557821e-01
-8.76696780e-02 6.54109776e-01 2.08863154e-01 -6.56327009e-01
-5.52694380e-01 -1.18998146e+00 -2.29683965e-02 7.50181735e-01
4.14536923e-01 -3.78415614e-01 -5.95382690e-01 -5.46536088e-01
3.49593997e-01 6.07531518e-02 -3.81972998e-01 -3.74775469e-01
-4.92580771e-01 -9.13603008e-01 3.87302041e-01 8.13386321e-01
1.17145371e+00 -9.00560975e-01 -8.44472468e-01 2.59557962e-01
-8.42447132e-02 -7.74638414e-01 -2.08263874e-01 2.41559803e-01
-9.60680962e-01 -1.11029816e+00 -5.98461330e-01 -1.26504529e+00
8.59666288e-01 5.18832326e-01 6.63141966e-01 4.60902154e-01
-1.01313078e+00 1.14994995e-01 -6.09027863e-01 -1.95587546e-01
2.22374484e-01 -4.80070077e-02 -2.69634038e-01 -1.78457931e-01
9.33665708e-02 -3.42331618e-01 -8.25240195e-01 3.21739107e-01
-9.08197105e-01 -1.10668922e-02 9.74167407e-01 1.26613522e+00
5.16676903e-01 4.22673702e-01 5.83999217e-01 -5.78594863e-01
4.67818767e-01 -4.39489000e-02 -4.25785601e-01 9.05294120e-02
-5.72667122e-01 -2.12588891e-01 6.00458741e-01 -1.70960873e-01
-1.13750958e+00 1.50698632e-01 -3.03842753e-01 -1.14130722e-02
-7.29423761e-02 3.46112132e-01 -3.73750597e-01 -4.04504120e-01
3.64255726e-01 7.17859417e-02 4.36077304e-02 -6.19228601e-01
7.73050636e-02 7.36503482e-01 4.62377012e-01 -2.13359073e-01
6.11860573e-01 3.54332060e-01 -2.20843986e-01 -9.20122743e-01
-3.92929077e-01 -3.50804955e-01 -5.14148712e-01 -1.79267332e-01
5.76481342e-01 -9.81411278e-01 -1.04652452e+00 1.02697325e+00
-1.08286929e+00 2.08945144e-02 -3.12070340e-01 2.59342939e-01
-1.34548709e-01 5.88151693e-01 -9.74706948e-01 -6.56285644e-01
-6.43471062e-01 -9.58590806e-01 1.21701288e+00 2.33174875e-01
1.86846524e-01 -5.55164397e-01 -6.58471584e-01 3.04297537e-01
3.69112372e-01 3.15586239e-01 1.28527594e+00 1.78467140e-01
-5.99890828e-01 -5.38205028e-01 -6.64133370e-01 6.89036489e-01
2.61837274e-01 -5.55213094e-02 -9.70622301e-01 -2.74776489e-01
-1.13874502e-01 -5.30968457e-02 1.35270429e+00 5.93073130e-01
1.51268816e+00 7.17756618e-03 -5.19356012e-01 5.61106920e-01
1.54931498e+00 3.56828541e-01 5.56636810e-01 3.30638210e-03
7.22326756e-01 5.46523988e-01 3.54498863e-01 4.87923265e-01
3.50995123e-01 2.52699792e-01 7.62175143e-01 -6.25129402e-01
-6.03695214e-01 -4.00728405e-01 3.00257266e-01 1.01309812e+00
-5.39154336e-02 -3.18818569e-01 -4.90777016e-01 6.79371774e-01
-1.62133551e+00 -6.41218066e-01 -1.51041180e-01 1.87568235e+00
2.57430434e-01 1.61867261e-01 -2.55892843e-01 5.93801022e-01
7.45919526e-01 8.80451947e-02 -6.24613464e-01 -6.23356402e-02
-2.51206279e-01 1.55668274e-01 5.37084281e-01 9.62743461e-02
-1.15887821e+00 7.26151049e-01 5.68065786e+00 1.00415349e+00
-9.07824159e-01 -6.96909428e-02 5.41805863e-01 1.23034775e-01
-7.31777921e-02 -2.95193791e-01 -6.12403989e-01 4.72370952e-01
6.84978738e-02 4.50821936e-01 2.18397394e-01 8.89910519e-01
-1.20956004e-01 -2.67744631e-01 -7.44019628e-01 1.08420920e+00
1.28633887e-01 -1.23206294e+00 -8.42354260e-03 -6.42168447e-02
3.07706714e-01 -1.20596163e-01 9.64904055e-02 2.02645630e-01
1.47943571e-01 -6.78485811e-01 4.90219474e-01 1.68949693e-01
7.90610194e-01 -8.59197676e-01 1.09423602e+00 1.81291163e-01
-1.59874582e+00 -7.08999455e-01 -7.32345521e-01 -1.25666738e-01
2.21219547e-02 9.66862798e-01 -4.19960320e-01 6.04370654e-01
1.08013320e+00 7.00694561e-01 -5.62987208e-01 1.06474900e+00
-1.56585455e-01 4.97518390e-01 -5.12184441e-01 3.91485393e-02
-3.42421718e-02 4.14754227e-02 2.39350930e-01 8.73682439e-01
4.63083714e-01 -9.96465385e-02 4.05848920e-01 7.12866366e-01
-1.24824211e-01 -1.31016076e-01 -4.68324751e-01 4.44355249e-01
4.37060297e-01 1.25638390e+00 -7.20554650e-01 1.72034666e-01
-4.33187306e-01 1.21992540e+00 1.29026562e-01 1.96056068e-01
-6.13671660e-01 -7.96478450e-01 4.79623765e-01 1.60112530e-01
5.96748114e-01 2.02610306e-02 -2.47161239e-01 -8.14493239e-01
3.25669497e-01 -5.24718881e-01 3.34244221e-01 -6.38466120e-01
-1.19991314e+00 4.52475369e-01 -5.57766795e-01 -1.01729596e+00
5.70953608e-01 -9.19037759e-01 -6.84937954e-01 5.69194973e-01
-1.43611908e+00 -1.57626677e+00 -6.75230622e-01 4.29017693e-01
6.64173782e-01 -7.92321786e-02 6.77628756e-01 5.11265635e-01
-8.48800123e-01 5.27663291e-01 -1.14497863e-01 1.05050474e-01
4.17640209e-01 -9.70229208e-01 2.68610626e-01 9.18633163e-01
-1.67039663e-01 2.14403421e-01 4.37550209e-02 -5.96984744e-01
-1.50114608e+00 -1.15164995e+00 2.63540417e-01 7.44129941e-02
1.32970020e-01 -2.43345156e-01 -7.98937619e-01 1.52219743e-01
-3.76618266e-01 8.31659064e-02 5.15959561e-01 -8.60292241e-02
-2.85118192e-01 -5.55540144e-01 -1.38042641e+00 3.07849348e-01
1.19298649e+00 -5.02366841e-01 -2.07195908e-01 1.65305540e-01
7.91356802e-01 -1.28153637e-01 -8.98421288e-01 7.04486430e-01
6.83138669e-01 -9.10759568e-01 1.12540793e+00 2.00616762e-01
3.57419282e-01 -5.26508331e-01 -3.89163047e-01 -1.05094135e+00
-7.92451262e-01 -1.67545035e-01 -5.39235817e-03 1.37040210e+00
1.30552441e-01 -6.43017113e-01 7.93040931e-01 2.16624260e-01
-5.13143182e-01 -9.99448240e-01 -9.41601396e-01 -4.84951437e-01
-4.48827744e-01 -2.75019675e-01 3.80356431e-01 3.74431074e-01
-2.37739131e-01 -9.54003930e-02 -2.22486049e-01 4.13044959e-01
4.44226682e-01 1.45266265e-01 3.34740072e-01 -1.44914556e+00
-1.15633197e-01 -2.80994534e-01 -6.99057937e-01 -1.24021959e+00
-3.26542228e-01 -7.39350259e-01 3.46267112e-02 -1.61471224e+00
6.93934932e-02 -3.48458856e-01 -4.73393381e-01 5.58221757e-01
-1.89237624e-01 2.86382675e-01 1.14465855e-01 -1.65300723e-02
-1.86756834e-01 6.06917620e-01 1.21551502e+00 -2.06480801e-01
-1.02405146e-01 -2.00809285e-01 -7.19345331e-01 7.70632386e-01
7.14705706e-01 -8.17645565e-02 -2.57330269e-01 -8.01509500e-01
-2.46076435e-01 -1.87970757e-01 6.36643350e-01 -1.33696294e+00
3.95533413e-01 5.99751770e-01 8.52129281e-01 -6.59289718e-01
2.90263444e-01 -8.11916173e-01 -2.32344970e-01 1.06587875e+00
2.25809783e-01 -1.04929067e-01 3.16837519e-01 7.29660094e-01
-3.95196885e-01 -8.22927281e-02 7.80887902e-01 4.77689728e-02
-1.08037174e+00 3.11631292e-01 -1.92418039e-01 -6.42039537e-01
1.09068239e+00 -4.67502236e-01 -3.22542191e-01 1.40859380e-01
-4.35441971e-01 2.04885945e-01 3.27214122e-01 3.45929354e-01
9.87459362e-01 -1.08767509e+00 -4.95166630e-01 8.70366573e-01
6.41414300e-02 1.67130753e-01 7.36382246e-01 5.66683412e-01
-9.51813817e-01 1.75044507e-01 -3.65258247e-01 -5.61731756e-01
-1.18366098e+00 3.65376383e-01 4.28910367e-02 -1.82137519e-01
-7.57617533e-01 1.18309653e+00 3.66702616e-01 -3.25229675e-01
2.55763769e-01 -6.14917517e-01 -2.14169309e-01 9.66865569e-03
3.69112283e-01 8.15846145e-01 1.27566501e-01 -2.06633344e-01
-2.28768960e-01 1.05880976e+00 -1.67494774e-01 6.69703782e-01
1.38183022e+00 -1.53177142e-01 -4.42627296e-02 -2.41694689e-01
9.12521601e-01 -6.73208758e-02 -1.38900125e+00 -6.04098337e-03
-3.42255414e-01 -4.71789628e-01 5.41836321e-01 -9.81606543e-01
-1.53694427e+00 9.69608486e-01 1.16217899e+00 1.34091124e-01
1.57961571e+00 -2.76810914e-01 1.17637682e+00 1.81771472e-01
5.67340970e-01 -1.04026258e+00 1.04761735e-01 1.16416760e-01
1.00008881e+00 -1.16270280e+00 -7.83066675e-02 -9.33719754e-01
-3.94256324e-01 1.08462298e+00 8.82849634e-01 -3.40685904e-01
7.09713936e-01 4.92008924e-01 -1.98123649e-01 -2.50720650e-01
-3.25222164e-01 -1.45074174e-01 1.50864303e-01 7.90430546e-01
9.07038227e-02 1.41056612e-01 -7.58382678e-02 7.67117023e-01
2.18352363e-01 -9.16317776e-02 5.17946556e-02 8.96197677e-01
-5.95414937e-01 -8.77714932e-01 -8.83520171e-02 8.40458035e-01
-3.38835925e-01 -3.78698856e-02 -1.42311320e-01 6.42559528e-01
5.59441090e-01 1.07237864e+00 9.13117174e-03 -8.35053682e-01
5.42502642e-01 -3.76387149e-01 1.67512015e-01 -4.72507507e-01
-4.93248194e-01 9.79836136e-02 -2.08424076e-01 -6.68809295e-01
-7.55268410e-02 -1.59128070e-01 -1.04545832e+00 -2.86828995e-01
-7.84293473e-01 -3.10316741e-01 4.70541686e-01 5.35767019e-01
7.02010810e-01 9.49623823e-01 5.72117269e-01 -6.27732277e-01
-5.14005303e-01 -9.10317898e-01 -8.89139414e-01 1.47199919e-02
1.72770828e-01 -7.34756291e-01 -2.42961556e-01 4.83285896e-02] | [7.509652614593506, 1.6713865995407104] |
e4ef7703-1fad-483b-b146-27e07eff7319 | towards-diverse-relevant-and-coherent-open | 2212.01145 | null | https://arxiv.org/abs/2212.01145v1 | https://arxiv.org/pdf/2212.01145v1.pdf | Towards Diverse, Relevant and Coherent Open-Domain Dialogue Generation via Hybrid Latent Variables | Conditional variational models, using either continuous or discrete latent variables, are powerful for open-domain dialogue response generation. However, previous works show that continuous latent variables tend to reduce the coherence of generated responses. In this paper, we also found that discrete latent variables have difficulty capturing more diverse expressions. To tackle these problems, we combine the merits of both continuous and discrete latent variables and propose a Hybrid Latent Variable (HLV) method. Specifically, HLV constrains the global semantics of responses through discrete latent variables and enriches responses with continuous latent variables. Thus, we diversify the generated responses while maintaining relevance and coherence. In addition, we propose Conditional Hybrid Variational Transformer (CHVT) to construct and to utilize HLV with transformers for dialogue generation. Through fine-grained symbolic-level semantic information and additive Gaussian mixing, we construct the distribution of continuous variables, prompting the generation of diverse expressions. Meanwhile, to maintain the relevance and coherence, the discrete latent variable is optimized by self-separation training. Experimental results on two dialogue generation datasets (DailyDialog and Opensubtitles) show that CHVT is superior to traditional transformer-based variational mechanism w.r.t. diversity, relevance and coherence metrics. Moreover, we also demonstrate the benefit of applying HLV to fine-tuning two pre-trained dialogue models (PLATO and BART-base). | ['Kan Li', 'Yiwei Li', 'Weichao Wang', 'Fei Mi', 'Yitong Li', 'Bin Sun'] | 2022-12-02 | null | null | null | null | ['response-generation'] | ['natural-language-processing'] | [-2.00100064e-01 1.98576465e-01 -3.44052404e-01 -3.77280325e-01
-9.53348637e-01 -7.09576011e-01 9.28014219e-01 -1.99313849e-01
-4.67434488e-02 1.12215555e+00 6.40761256e-01 1.26029521e-01
6.05973490e-02 -8.69669318e-01 -7.61724189e-02 -8.17621768e-01
6.16695344e-01 8.21802795e-01 -3.86508293e-02 -6.38881981e-01
1.69229865e-01 -1.76731452e-01 -1.24206793e+00 3.73644263e-01
1.23225844e+00 5.20370007e-01 2.20139891e-01 4.51050013e-01
-6.78546369e-01 6.10838830e-01 -9.25087631e-01 -4.09359783e-01
-3.37973565e-01 -8.79223645e-01 -8.93817246e-01 -2.59456914e-02
-2.96409279e-01 -2.32645750e-01 -9.67465788e-02 7.72996604e-01
6.94189072e-01 5.84943295e-01 8.09388220e-01 -1.52052081e+00
-9.51946735e-01 9.30906415e-01 -2.55227953e-01 -3.12962770e-01
5.83466172e-01 1.95073903e-01 1.20316398e+00 -6.69502020e-01
6.72696888e-01 1.74402583e+00 4.49995250e-01 1.00405276e+00
-1.58679414e+00 -6.05158269e-01 3.33060056e-01 -1.50506064e-01
-1.09016895e+00 -4.96588379e-01 1.11872840e+00 -5.03511906e-01
7.27703810e-01 3.67951214e-01 2.83522666e-01 1.64024341e+00
-7.01544285e-02 9.75173593e-01 1.08873832e+00 -2.63094813e-01
2.15993702e-01 4.98701841e-01 1.13072149e-01 3.15450013e-01
-6.14830613e-01 -2.29693994e-01 -5.09630084e-01 -6.31529748e-01
6.71154916e-01 -8.23689103e-02 -2.47547686e-01 -4.05879803e-02
-1.10108101e+00 1.30576098e+00 -1.43373892e-01 2.22093195e-01
-4.86958474e-01 -5.15807122e-02 4.84371483e-01 2.35137805e-01
7.48581290e-01 4.79229212e-01 -3.20779622e-01 -4.71245289e-01
-9.51824963e-01 5.64695954e-01 8.78903151e-01 1.10995424e+00
7.98512638e-01 1.91380531e-01 -1.10655856e+00 1.37353206e+00
5.55120289e-01 3.16502988e-01 6.65135920e-01 -1.18601143e+00
3.87126714e-01 5.68374813e-01 2.58458108e-01 -9.25296068e-01
-2.16646343e-02 5.60574941e-02 -8.04771364e-01 -4.55846876e-01
1.93325415e-01 -5.00519633e-01 -6.41003609e-01 2.11839366e+00
3.90920103e-01 -2.04945073e-01 3.45487356e-01 9.02564406e-01
1.07417643e+00 1.00288939e+00 1.43083245e-01 -6.24421477e-01
1.02558613e+00 -1.08400905e+00 -1.28538954e+00 2.91818827e-01
4.32431549e-01 -7.82880127e-01 1.51263404e+00 1.82408601e-01
-1.25009930e+00 -6.45935774e-01 -5.90907633e-01 -1.08691551e-01
2.96654031e-02 -5.37420958e-02 6.28679931e-01 5.13730109e-01
-9.55734849e-01 3.53486359e-01 -5.05476177e-01 5.79139888e-02
-1.85323626e-01 2.07164884e-01 4.22206149e-02 4.27841216e-01
-1.81842768e+00 5.19781113e-01 1.03665050e-02 -1.26571357e-01
-7.83000767e-01 -4.34502572e-01 -8.33062649e-01 2.42665689e-03
1.52658537e-01 -8.19173753e-01 1.36586988e+00 -8.22798550e-01
-2.32856703e+00 4.45172846e-01 -4.67005223e-01 -6.03669360e-02
6.70287490e-01 4.10070494e-02 -4.98265103e-02 4.60396409e-02
2.79672176e-01 6.59021378e-01 8.24822724e-01 -1.28656840e+00
-2.58711606e-01 -1.02443554e-01 2.28235319e-01 2.64466017e-01
-4.37066257e-01 -1.49310887e-01 -5.67434728e-01 -5.84788322e-01
-2.20704172e-02 -8.18862617e-01 -2.15049505e-01 -4.83374745e-01
-3.03159446e-01 -8.64901960e-01 6.61206782e-01 -5.18542230e-01
1.35073471e+00 -1.97288299e+00 6.74441040e-01 -4.03156206e-02
4.36028033e-01 -4.31649759e-02 -1.18179075e-01 5.69503427e-01
4.19201910e-01 2.49387503e-01 -9.12217274e-02 -6.48351252e-01
4.63276178e-01 5.28990388e-01 -4.93133754e-01 1.17881326e-02
2.09882446e-02 1.09373820e+00 -9.41971779e-01 -7.35181570e-01
1.09618291e-01 4.77146417e-01 -9.27110374e-01 4.41226989e-01
-6.51827037e-01 7.06437647e-01 -5.69350123e-01 2.64572412e-01
5.50856471e-01 -1.73860818e-01 4.14492905e-01 1.05166450e-01
1.46571649e-02 3.12881082e-01 -8.94664466e-01 1.68781555e+00
-6.55947745e-01 4.08032149e-01 8.82339776e-02 -7.73625374e-01
1.28991723e+00 6.16583407e-01 4.32573795e-01 -5.93019962e-01
1.43501624e-01 6.67020259e-03 -5.00805140e-01 -5.50321639e-01
9.99952853e-01 -4.43419755e-01 -4.76486266e-01 4.70266253e-01
2.69302547e-01 -2.82851607e-01 8.11867937e-02 5.30628026e-01
3.26703668e-01 1.77677423e-01 -2.41519883e-01 2.55134124e-02
5.76464415e-01 -2.55832851e-01 6.74953043e-01 5.57026803e-01
-2.44965479e-01 4.46806222e-01 1.01200068e+00 3.07825685e-01
-7.63657868e-01 -1.13875306e+00 9.15516019e-02 1.42566133e+00
3.29786092e-01 -5.28810024e-01 -7.29624629e-01 -4.99300629e-01
-1.44834250e-01 1.07557631e+00 -6.07026458e-01 -6.88475221e-02
-3.40564787e-01 -6.31220758e-01 7.28842795e-01 2.79475182e-01
3.44635189e-01 -9.28090036e-01 6.30812049e-02 3.85979235e-01
-1.03632736e+00 -7.90518880e-01 -5.77737391e-01 -1.99882239e-01
-5.34395874e-01 -4.51180339e-01 -9.09065247e-01 -5.66355884e-01
2.92230606e-01 7.68334642e-02 1.00869477e+00 -2.04543024e-01
3.55924308e-01 3.05168390e-01 -6.70777380e-01 8.57427120e-02
-6.82738781e-01 4.64172140e-02 5.41150831e-02 6.74492214e-04
2.22258255e-01 -5.18210053e-01 -2.06092149e-01 2.90594369e-01
-7.22470820e-01 1.34013698e-01 1.43225014e-04 1.31204450e+00
4.46369439e-01 -3.46303850e-01 7.07751274e-01 -9.27717328e-01
1.38230848e+00 -7.12922513e-01 -2.32798934e-01 2.74223626e-01
-6.69675529e-01 2.06119299e-01 6.28118515e-01 -8.17151248e-01
-1.49421561e+00 -6.27522290e-01 -2.39759147e-01 -4.28427845e-01
7.57832602e-02 5.90123475e-01 -1.93953738e-01 6.60708249e-01
4.89390671e-01 3.86903614e-01 9.65289623e-02 -3.53474617e-01
6.91131055e-01 8.14809859e-01 1.42970696e-01 -9.62189376e-01
2.85013288e-01 8.59427080e-02 -5.67129493e-01 -7.32364833e-01
-4.19303209e-01 -2.49411166e-01 -2.84069031e-01 -7.55780786e-02
9.73038793e-01 -7.27813780e-01 -6.31330550e-01 4.16784227e-01
-1.33150482e+00 -5.36538363e-01 -2.57781178e-01 3.51019800e-01
-7.14989305e-01 5.93640268e-01 -8.96723151e-01 -9.94240642e-01
-3.28650475e-01 -1.17969549e+00 1.15041137e+00 2.50404388e-01
-7.23870277e-01 -1.42308390e+00 2.49852374e-01 4.44603145e-01
4.42792028e-01 2.18739003e-01 8.53534341e-01 -5.03592372e-01
-2.65485466e-01 1.39459625e-01 7.06506893e-02 6.80265352e-02
1.10766776e-01 1.88884899e-01 -8.81140590e-01 1.34098297e-02
9.68890116e-02 -6.21108174e-01 7.03501701e-01 4.24062639e-01
6.22864425e-01 -6.09694839e-01 -1.63653970e-01 5.21661758e-01
8.47910881e-01 3.48046646e-02 4.75027710e-01 -8.38902742e-02
5.44700086e-01 9.13085282e-01 7.24693596e-01 8.89629185e-01
7.00011551e-01 9.01944399e-01 -1.65883884e-01 -2.23791283e-02
1.67970791e-01 -5.31266868e-01 5.35925925e-01 1.18355083e+00
7.07065463e-02 -3.18836361e-01 -5.84639907e-01 4.97856051e-01
-2.02521086e+00 -1.12460911e+00 -2.94586778e-01 1.61291134e+00
1.44493580e+00 -2.02245280e-01 1.72535866e-01 -2.44359314e-01
6.98974729e-01 3.72380555e-01 -2.69290239e-01 -6.17712796e-01
-3.20033342e-01 4.15691659e-02 -1.33719519e-01 9.07343447e-01
-5.61389029e-01 1.35203838e+00 5.90382004e+00 1.14528775e+00
-1.10005021e+00 4.97610599e-01 3.95308286e-01 -8.79144743e-02
-1.15961885e+00 5.29930443e-02 -9.05512691e-01 6.60471916e-01
7.65764058e-01 -5.51228583e-01 2.97110558e-01 6.85592949e-01
5.16437590e-01 2.50888675e-01 -7.90976822e-01 7.78268397e-01
-5.76656424e-02 -1.20532537e+00 1.54124767e-01 -1.69846177e-01
9.70263422e-01 -5.48565626e-01 1.44287467e-01 8.36652458e-01
6.99439585e-01 -9.49295223e-01 6.40223444e-01 5.51215768e-01
8.11642647e-01 -6.11969411e-01 6.19874179e-01 5.90994060e-01
-9.88641500e-01 2.19311476e-01 -3.51805806e-01 3.45330909e-02
2.90748626e-01 1.45276397e-01 -7.35992551e-01 5.89464068e-01
1.71490252e-01 3.68182123e-01 -3.54545452e-02 7.63931721e-02
-4.77493554e-01 6.74949586e-01 6.55620173e-02 -3.51872414e-01
4.33530241e-01 -4.33653265e-01 6.36972725e-01 1.16697955e+00
-4.15749922e-02 2.09520057e-01 4.26302254e-01 1.49259508e+00
1.45746723e-01 2.25997657e-01 -1.82118818e-01 -2.49387830e-01
7.25841880e-01 1.13919830e+00 -2.66012520e-01 -4.85300124e-01
2.54955851e-02 1.18915725e+00 1.74962819e-01 7.01939046e-01
-1.14636016e+00 -3.13879758e-01 6.47237360e-01 -3.12667757e-01
-3.23929377e-02 -2.72471279e-01 -3.78646880e-01 -1.35206580e+00
-2.82663435e-01 -8.15073848e-01 1.51122928e-01 -4.94972646e-01
-1.32095349e+00 7.74094462e-01 1.47268504e-01 -8.12941730e-01
-7.89766252e-01 1.38743743e-01 -5.89174986e-01 1.47012842e+00
-1.25893545e+00 -1.27924418e+00 -3.22687566e-01 8.44725490e-01
8.72211576e-01 -1.77790448e-01 9.92487192e-01 6.87590688e-02
-5.00337839e-01 9.23014939e-01 1.91080570e-01 -1.07898191e-01
9.22079265e-01 -1.18486798e+00 1.70909286e-01 2.93380201e-01
-2.46720165e-01 9.53705072e-01 8.42556655e-01 -6.87503994e-01
-1.01458168e+00 -8.02721262e-01 1.08498967e+00 -3.97996455e-01
4.00425375e-01 -4.73863900e-01 -9.59159970e-01 4.17542011e-01
3.20066363e-01 -9.06229675e-01 8.47961903e-01 3.84134889e-01
-8.08115900e-02 4.98153955e-01 -9.97126162e-01 7.45045781e-01
4.96208012e-01 -7.74114966e-01 -7.75386453e-01 2.15449631e-01
1.36341453e+00 -3.28998327e-01 -8.05918217e-01 1.87124342e-01
5.01627266e-01 -8.16526771e-01 7.06696570e-01 -5.55753648e-01
6.54923081e-01 9.06017274e-02 -2.58043170e-01 -1.41555929e+00
-4.01615173e-01 -1.08519781e+00 -1.10735051e-01 1.75097871e+00
2.78139561e-01 -5.76116741e-01 7.48742342e-01 8.42279732e-01
-8.81350487e-02 -7.32042670e-01 -5.91212094e-01 -4.90090877e-01
4.09084678e-01 -2.27352738e-01 5.86860418e-01 1.31410110e+00
2.21354097e-01 6.77356303e-01 -7.02259839e-01 -3.85843873e-01
2.60412693e-01 2.81960517e-01 1.04994500e+00 -8.89154553e-01
-5.43244064e-01 -4.85965431e-01 3.11589599e-01 -1.52866256e+00
5.18616259e-01 -7.59180367e-01 9.88509506e-02 -1.58533406e+00
5.44937477e-02 -6.33876741e-01 3.32569629e-02 2.89630830e-01
-6.07520044e-01 -1.72033101e-01 1.21235609e-01 3.34900022e-01
-4.66681868e-01 1.32237208e+00 1.37090421e+00 -1.22017786e-01
-8.39332581e-01 1.07441563e-03 -7.15647936e-01 3.46983880e-01
8.34320903e-01 -2.66995937e-01 -7.28714168e-01 -2.21790224e-01
-9.52583328e-02 6.01510227e-01 1.10673420e-01 -7.07589015e-02
1.11151405e-01 -6.42585039e-01 -7.22510442e-02 -6.53116882e-01
6.66281223e-01 -1.08629819e-02 -5.81977423e-03 4.97576632e-02
-8.32542121e-01 -3.38415265e-01 -1.44528719e-02 4.75768596e-01
-3.90325665e-01 -3.85653526e-02 6.79970324e-01 -7.92265162e-02
-4.05087978e-01 1.69705346e-01 -6.32079661e-01 4.11181688e-01
7.06441700e-01 -2.64153928e-01 -1.92038640e-02 -7.64816999e-01
-7.38022625e-01 6.34167969e-01 2.51830399e-01 6.29645050e-01
8.14314187e-01 -1.47449756e+00 -8.32281709e-01 1.57705262e-01
-6.98605776e-02 -1.78580165e-01 6.10286355e-01 8.70060742e-01
-1.50163695e-02 3.49280030e-01 -1.88470915e-01 -6.87838316e-01
-1.33478618e+00 3.28801781e-01 3.74339111e-02 -4.24823672e-01
-2.07895517e-01 1.21581268e+00 3.79766375e-01 -8.94294143e-01
1.64095059e-01 4.65607494e-02 -3.64108533e-01 5.32736421e-01
6.89697042e-02 2.85849988e-01 -6.96414530e-01 -5.99539936e-01
8.08849558e-02 9.23170671e-02 5.97882792e-02 -6.82357728e-01
1.01010478e+00 -5.18813372e-01 -2.14921013e-01 8.01804066e-01
1.16565490e+00 3.46150249e-01 -1.09426308e+00 -4.11269963e-01
-2.95847684e-01 -4.52648163e-01 -2.26859778e-01 -4.40466881e-01
-6.21202111e-01 1.09134173e+00 1.16012748e-02 2.66544223e-01
7.19464242e-01 -5.15628569e-02 9.26986992e-01 -6.14299290e-02
1.12562075e-01 -1.29756033e+00 4.40829098e-01 7.75078535e-01
8.81943822e-01 -9.17746961e-01 -4.55611408e-01 -2.74588585e-01
-1.33371162e+00 8.78358066e-01 7.56956935e-01 1.74847484e-01
2.67541230e-01 -8.03974867e-02 2.73600101e-01 -3.73577438e-02
-1.18500924e+00 8.81511644e-02 8.47892985e-02 4.90130812e-01
7.71197736e-01 3.15849274e-01 -6.21029496e-01 1.02784693e+00
-4.14546818e-01 -2.32161477e-01 4.81731206e-01 4.93176013e-01
-2.92519182e-01 -1.51681376e+00 -3.30100894e-01 3.61307189e-02
-2.62175858e-01 -2.18263879e-01 -6.17183030e-01 3.12610239e-01
3.36529352e-02 1.26941931e+00 -9.04554501e-02 -5.61270773e-01
4.45977487e-02 2.52948523e-01 2.63494164e-01 -6.40512049e-01
-6.48355484e-01 5.73809862e-01 2.34628618e-01 -1.84364706e-01
-3.12082797e-01 -4.85216707e-01 -1.30564964e+00 -4.25428748e-01
-6.83875263e-01 7.76659489e-01 2.88692445e-01 8.34974229e-01
2.05050886e-01 7.36432910e-01 9.12440658e-01 -5.27984142e-01
-1.07738900e+00 -1.25032997e+00 -6.05978966e-01 5.52643836e-01
3.41549627e-02 -8.05684745e-01 -4.13181186e-01 -6.53975382e-02] | [12.582223892211914, 8.341238021850586] |
89d7c21f-77dc-4628-9af9-2c37e209aa4c | addressing-leakage-in-self-supervised | 2204.11594 | null | https://arxiv.org/abs/2204.11594v1 | https://arxiv.org/pdf/2204.11594v1.pdf | Addressing Leakage in Self-Supervised Contextualized Code Retrieval | We address contextualized code retrieval, the search for code snippets helpful to fill gaps in a partial input program. Our approach facilitates a large-scale self-supervised contrastive training by splitting source code randomly into contexts and targets. To combat leakage between the two, we suggest a novel approach based on mutual identifier masking, dedentation, and the selection of syntax-aligned targets. Our second contribution is a new dataset for direct evaluation of contextualized code retrieval, based on a dataset of manually aligned subpassages of code clones. Our experiments demonstrate that our approach improves retrieval substantially, and yields new state-of-the-art results for code clone and defect detection. | ['Ulrich Schwanecke', 'Adrian Ulges', 'Viola Campos', 'Johannes Villmow'] | 2022-04-17 | null | https://aclanthology.org/2022.coling-1.84 | https://aclanthology.org/2022.coling-1.84.pdf | coling-2022-10 | ['defect-detection'] | ['computer-vision'] | [ 2.67605484e-01 -2.16746762e-01 -7.77390659e-01 -2.39136189e-01
-1.41408992e+00 -7.66723454e-01 2.45059878e-01 6.88187897e-01
8.93798769e-02 1.44726798e-01 9.42535922e-02 -7.77246416e-01
9.02104154e-02 -4.69734818e-01 -6.19269907e-01 -9.40193087e-02
-2.91716456e-01 -4.56796214e-02 5.44186413e-01 -1.84211478e-01
1.03510153e+00 2.00033821e-02 -1.79594815e+00 8.77161205e-01
1.24666619e+00 3.76802027e-01 4.40996498e-01 7.56378829e-01
-5.15847206e-01 9.94803190e-01 -7.41841137e-01 -3.85474205e-01
7.19447061e-02 -5.22029817e-01 -1.09741056e+00 -3.31905454e-01
6.36488795e-01 2.93316673e-02 4.96592894e-02 1.22861898e+00
2.79995650e-01 -4.26161796e-01 2.39555910e-01 -1.10990846e+00
-7.62361228e-01 9.76749063e-01 -6.62445426e-01 4.44846302e-01
7.45853424e-01 6.94622546e-02 1.34809971e+00 -1.03454125e+00
7.21806347e-01 7.34177887e-01 9.80165899e-01 6.00008965e-01
-1.54436123e+00 -5.75435877e-01 -1.00533031e-01 -1.32506415e-01
-1.46667576e+00 -4.51043904e-01 6.18086874e-01 -7.34712481e-01
1.46112967e+00 3.19178730e-01 8.38147402e-02 7.06239760e-01
1.58555999e-01 8.85481894e-01 6.23398125e-01 -9.44759607e-01
2.39039421e-01 2.86988318e-01 6.51702702e-01 1.00314581e+00
2.38082021e-01 4.27814126e-02 -2.62685418e-01 -1.04363179e+00
8.31604972e-02 -4.12543900e-02 -3.02981049e-01 -7.24263072e-01
-1.05915093e+00 8.36305797e-01 1.23925678e-01 5.24941802e-01
3.93821508e-01 2.23939046e-01 7.09222198e-01 7.46241748e-01
1.32100180e-01 9.41558599e-01 -7.76333332e-01 -3.64391744e-01
-1.15095937e+00 2.91886985e-01 9.37998056e-01 1.44747579e+00
1.27206826e+00 -4.54743147e-01 -1.37712970e-01 7.98144221e-01
2.15721384e-01 4.04536217e-01 7.54096210e-01 -6.91111803e-01
8.45067143e-01 1.11980081e+00 -1.65553808e-01 -7.85016298e-01
4.13516089e-02 -1.35034204e-01 -5.70141487e-02 2.12161496e-01
-2.36728974e-02 3.34979802e-01 -5.06516576e-01 1.47669458e+00
-2.33974710e-01 -3.24685037e-01 -3.53204608e-02 1.87850922e-01
6.80036247e-01 1.89784646e-01 -1.65049881e-01 1.56383827e-01
1.02294338e+00 -1.18556607e+00 -2.19894379e-01 -4.35091615e-01
1.39265072e+00 -9.51133549e-01 1.24678004e+00 4.58989926e-02
-9.01377857e-01 -3.06516171e-01 -1.03069293e+00 -2.09530108e-02
-2.32510060e-01 4.27916855e-01 6.30183935e-01 8.83444488e-01
-1.28032517e+00 5.80562532e-01 -6.67521775e-01 -2.02297539e-01
3.82814229e-01 2.80074060e-01 -3.74267399e-01 1.07913412e-01
-2.67837435e-01 4.86814469e-01 1.76384926e-01 -7.29445636e-01
-7.64317095e-01 -9.01015818e-01 -1.03631794e+00 2.40873635e-01
2.08378762e-01 -8.42900649e-02 1.24541438e+00 -1.05984235e+00
-6.63551867e-01 1.27641714e+00 -2.66335160e-01 -2.62094200e-01
-1.26196489e-01 -1.18875660e-01 -2.91334569e-01 -8.98866132e-02
4.90074873e-01 4.30201173e-01 7.53478110e-01 -1.40680552e+00
-6.30070508e-01 -1.54124931e-01 -2.24122778e-02 -3.62965673e-01
-4.46588457e-01 5.76409161e-01 -4.28292960e-01 -9.31403995e-01
-4.21048738e-02 -8.82281125e-01 -8.88258666e-02 -1.72580779e-01
-3.51474553e-01 3.27151343e-02 7.69439697e-01 -6.47918701e-01
1.88981640e+00 -2.69132304e+00 1.20254710e-01 2.64626354e-01
4.09065664e-01 1.30680665e-01 -6.44613802e-01 5.83573341e-01
-4.63816553e-01 4.45286036e-01 -7.06929803e-01 -3.45790237e-01
-1.62125155e-01 -2.19646275e-01 -6.54258370e-01 3.04666966e-01
2.47218668e-01 9.63399827e-01 -1.00465119e+00 -5.56961715e-01
-4.97117192e-01 -3.43844414e-01 -1.09730566e+00 3.37026209e-01
-3.75946581e-01 -2.66676128e-01 -5.11367142e-01 1.02556777e+00
5.06797373e-01 -3.13883573e-01 -2.67716777e-02 5.77831328e-01
-1.42270669e-01 5.66199541e-01 -8.22249353e-01 2.08061433e+00
-4.75478679e-01 7.20480800e-01 -6.81348145e-02 -6.30647123e-01
1.05160105e+00 -2.34647049e-03 1.94395170e-01 -5.85910201e-01
-4.07754660e-01 5.20789683e-01 -3.18149090e-01 -5.67765057e-01
9.10015643e-01 6.37942970e-01 -5.99582970e-01 9.73125160e-01
-1.07718118e-01 -2.77314186e-01 2.07851827e-01 5.61460257e-01
1.91293669e+00 6.48022303e-03 4.09675926e-01 -5.53829432e-01
6.73645496e-01 3.40232521e-01 2.91762322e-01 1.08664095e+00
-9.40872505e-02 7.17230141e-01 6.84839964e-01 -1.45951003e-01
-9.28165376e-01 -7.79962122e-01 -1.32137060e-01 1.22203279e+00
2.31430262e-01 -1.07417274e+00 -7.33739436e-01 -1.38988900e+00
2.07567990e-01 5.63367605e-01 -6.02683246e-01 -4.06588852e-01
-9.33238447e-01 -3.44725788e-01 8.19941700e-01 6.22642994e-01
7.21048042e-02 -9.13815916e-01 -6.52902484e-01 -8.00277293e-02
-5.37723526e-02 -4.06803787e-01 -1.01819408e+00 5.59511960e-01
-1.02844524e+00 -1.37604117e+00 -2.99255997e-01 -1.36222196e+00
9.49940383e-01 5.33844590e-01 1.61595666e+00 1.02286100e+00
-5.37327886e-01 5.29377818e-01 -6.44200802e-01 3.10357094e-01
-1.20000601e+00 3.57248366e-01 -7.39578962e-01 -9.00727510e-01
5.84968209e-01 -3.92549545e-01 -2.13412374e-01 5.20804644e-01
-8.58224332e-01 -4.32371318e-01 5.03681362e-01 1.22224212e+00
1.11680098e-01 -3.86566222e-01 2.25789890e-01 -1.02320945e+00
6.38308644e-01 -6.30700946e-01 -7.78068900e-01 5.56689501e-01
-7.65393317e-01 4.19894695e-01 3.48264575e-01 -3.16202432e-01
-9.34282601e-01 9.26544890e-02 6.43411502e-02 4.46586832e-02
-2.56738663e-02 5.38803518e-01 1.87311023e-01 -5.22915721e-01
1.27466214e+00 2.48271629e-01 1.19134896e-02 -6.79806113e-01
3.89776528e-01 9.75284755e-01 5.55240154e-01 -9.47146237e-01
7.94510782e-01 -8.24927986e-02 -7.94209540e-01 -1.75907686e-01
1.77491438e-02 -8.53815138e-01 -6.36637151e-01 3.86633337e-01
2.78895319e-01 -9.54892099e-01 4.19512996e-03 2.86617696e-01
-1.22998297e+00 -3.64962280e-01 -3.37581575e-01 -1.32665813e-01
-4.37601924e-01 7.86745071e-01 -6.12964392e-01 -3.41171771e-01
-2.95425683e-01 -1.44824493e+00 1.33979177e+00 -2.83907354e-01
-5.77828169e-01 -5.52099586e-01 5.35681486e-01 6.32245615e-02
4.68608111e-01 -3.31392497e-01 1.55159080e+00 -8.86794150e-01
-1.02586210e+00 -2.78678775e-01 -2.55438864e-01 1.21069968e-01
1.75569654e-01 7.15291277e-02 -7.32210755e-01 -5.27239919e-01
-2.76360899e-01 -5.05979776e-01 8.55687082e-01 -3.81704360e-01
9.78522062e-01 -9.20923874e-02 -8.38000834e-01 4.83194381e-01
1.62171280e+00 2.39894778e-01 6.27256632e-01 4.98414993e-01
4.39919919e-01 1.97930768e-01 5.87206125e-01 6.18133128e-01
1.57009110e-01 5.65757155e-01 3.98011863e-01 2.49516875e-01
-2.27616593e-01 -1.84145540e-01 3.54757369e-01 8.54772866e-01
8.11091304e-01 -1.06939813e-02 -1.16186273e+00 1.06052101e+00
-1.62863290e+00 -6.81466520e-01 -5.03563695e-02 2.29493356e+00
1.16134357e+00 -1.19256392e-01 -1.18067250e-01 4.38507125e-02
8.06376934e-01 -1.45117834e-01 -2.09310338e-01 -2.86716074e-01
9.30703208e-02 3.67449522e-01 4.43053603e-01 3.29311907e-01
-1.14194500e+00 8.89663041e-01 7.65143585e+00 8.86367381e-01
-6.74894571e-01 1.94870442e-01 1.48210034e-01 3.58290553e-01
-7.23028660e-01 4.43289936e-01 -8.22130442e-01 4.83805984e-01
7.08737195e-01 -3.40544254e-01 4.90357220e-01 1.44545579e+00
-7.08961129e-01 -3.48032206e-01 -1.45314491e+00 6.54028893e-01
2.53218740e-01 -1.37679017e+00 -3.41514915e-01 -3.48408669e-01
1.08402061e+00 5.42178117e-02 7.51795061e-03 5.19904435e-01
4.37615722e-01 -4.94013309e-01 7.72730589e-01 -1.28115353e-03
7.71444917e-01 -3.22943360e-01 4.81735557e-01 7.11442903e-02
-1.21911705e+00 -4.49840635e-01 -2.17940673e-01 1.67990923e-01
-6.48851395e-01 4.19142842e-01 -5.20063519e-01 4.62703966e-02
6.11923993e-01 8.78926396e-01 -1.28006971e+00 1.41765773e+00
1.66117754e-02 4.45454478e-01 1.58713654e-01 -1.61023855e-01
-2.33624250e-01 4.53598887e-01 5.24231017e-01 1.63649929e+00
3.46149921e-01 -5.84880054e-01 1.44747049e-01 1.28323281e+00
-1.00401215e-01 1.85538456e-01 -8.50355029e-01 7.58476853e-02
6.96342051e-01 7.62120187e-01 -6.69111252e-01 -3.57083589e-01
-6.71831310e-01 9.00851369e-01 4.82725590e-01 1.59293547e-01
-5.96436799e-01 -9.37841594e-01 6.12991869e-01 -2.22355053e-01
6.87875092e-01 8.27141758e-03 -4.23639804e-01 -1.33089769e+00
4.98485595e-01 -1.34130275e+00 5.50657630e-01 -3.41021359e-01
-9.80737865e-01 4.88316119e-01 2.99755055e-02 -1.33306789e+00
-2.00827211e-01 -2.32007638e-01 -7.45639145e-01 7.53990769e-01
-1.24451017e+00 -5.42166710e-01 -1.57505751e-01 2.32579201e-01
4.55661535e-01 -5.17943382e-01 9.84315634e-01 5.10887325e-01
-1.74832135e-01 1.07182038e+00 2.32520133e-01 7.82943517e-02
8.73369396e-01 -1.44155908e+00 9.76679981e-01 1.05879200e+00
1.46344081e-01 1.31702614e+00 5.06905675e-01 -8.55054438e-01
-1.44906878e+00 -9.71605062e-01 9.67536211e-01 -7.19883859e-01
8.07846069e-01 -4.59446192e-01 -1.19511294e+00 6.48013175e-01
1.15851216e-01 8.61619860e-02 7.30372608e-01 4.27584834e-02
-1.02889705e+00 1.85141638e-01 -1.09203207e+00 3.42413992e-01
1.03573799e+00 -1.00785005e+00 -8.66022289e-01 2.78921425e-01
7.78547943e-01 -3.90512943e-01 -6.06173277e-01 2.77734637e-01
1.54354632e-01 -1.08227479e+00 7.01338470e-01 -3.11753660e-01
7.28782237e-01 -3.51327807e-01 -1.60918504e-01 -1.05437684e+00
-5.61636537e-02 -7.28199661e-01 1.05774231e-01 1.27350390e+00
6.09351039e-01 -3.98545116e-01 8.96684408e-01 5.00758827e-01
-3.01845282e-01 -3.08566004e-01 -6.12022698e-01 -7.88529217e-01
5.95440008e-02 -3.20626229e-01 6.14691854e-01 1.04370046e+00
8.70368481e-01 -2.58971959e-01 1.93694487e-01 -4.92679812e-02
2.57422090e-01 4.84220237e-01 8.17033827e-01 -8.65141392e-01
-6.00702643e-01 -6.85280442e-01 -5.44044673e-01 -1.27514982e+00
4.35326159e-01 -1.22018349e+00 3.92864853e-01 -7.63963461e-01
7.03635573e-01 -7.26030171e-01 -7.80934542e-02 6.68954074e-01
-4.24436301e-01 -2.57735085e-02 -2.36044109e-01 3.88949275e-01
-7.99244761e-01 1.44403860e-01 2.97991157e-01 -5.33226907e-01
-3.50936323e-01 -3.01662851e-02 -7.74017751e-01 1.79283947e-01
5.19504607e-01 -9.35176015e-01 -3.15358698e-01 -4.64926153e-01
3.50639701e-01 1.80635527e-01 2.26957723e-01 -8.98216546e-01
3.41527730e-01 2.97557652e-01 -4.40515190e-01 -6.18232667e-01
-5.31796753e-01 -5.74790776e-01 -1.59345791e-01 5.56219041e-01
-8.09686899e-01 5.04020751e-01 6.10642731e-01 4.53840733e-01
-3.50908875e-01 -9.64539707e-01 4.73635584e-01 -3.14429522e-01
-7.66047657e-01 -2.80932575e-01 -5.11111081e-01 4.99011010e-01
7.69038498e-01 -6.63377419e-02 -7.73320913e-01 3.09702903e-01
-1.88531235e-01 3.36071290e-02 1.04586351e+00 6.47661090e-01
5.94764113e-01 -1.15827882e+00 -3.67214054e-01 5.71054280e-01
8.85039330e-01 -6.25277817e-01 -2.56715447e-01 2.39553154e-01
-5.14322221e-01 3.26143414e-01 3.04153506e-02 -5.84192574e-01
-1.67032790e+00 8.46730828e-01 1.01930395e-01 -2.48126149e-01
-5.18007755e-01 8.88420165e-01 -1.26419947e-01 -6.87253952e-01
3.36695343e-01 -5.23987174e-01 2.05730632e-01 -4.15457785e-01
6.01215541e-01 1.54389933e-01 3.14514935e-01 -2.20291719e-01
-5.06795287e-01 5.32693207e-01 -3.19168687e-01 2.84607679e-01
9.39117312e-01 1.45503446e-01 -7.22884357e-01 2.03859687e-01
1.68429410e+00 5.41422486e-01 -6.54041290e-01 -3.16485286e-01
5.95561266e-01 -9.33121264e-01 -2.07332194e-01 -7.57634282e-01
-7.85121381e-01 3.44527632e-01 5.25366366e-01 1.75206468e-01
9.32570517e-01 1.75892100e-01 4.32631373e-01 8.18544567e-01
6.60322845e-01 -5.68311453e-01 2.83897966e-01 5.61256111e-01
5.56246340e-01 -1.31315351e+00 -5.34538031e-02 -4.99431252e-01
-1.01350173e-02 1.14559245e+00 7.39315867e-01 5.11263460e-02
4.86429155e-01 7.97439814e-01 2.69388594e-02 -2.27493212e-01
-7.57071614e-01 3.41536924e-02 9.16860923e-02 5.51623940e-01
5.81957877e-01 -3.83599758e-01 -2.82750458e-01 3.29081744e-01
2.52817810e-01 -2.62778580e-01 6.08054280e-01 1.80161393e+00
-4.65598941e-01 -1.54061425e+00 -4.32694882e-01 5.48880339e-01
-3.66042435e-01 -6.88472688e-01 -5.24684489e-01 6.51504815e-01
-1.59038424e-01 7.59275913e-01 -1.26044065e-01 -4.78905290e-01
2.56160736e-01 1.59612343e-01 2.90288001e-01 -1.09803009e+00
-1.20816469e+00 -3.99757951e-01 -1.39430970e-01 -6.01460576e-01
-1.17606364e-01 -7.21829355e-01 -1.08700156e+00 1.16516799e-01
-7.97643542e-01 4.06521082e-01 4.81804401e-01 4.84175414e-01
8.46323729e-01 1.91643044e-01 7.39110589e-01 -4.01765853e-01
-6.07460439e-01 -5.49681306e-01 -1.34348884e-01 4.62237507e-01
6.29063547e-01 -2.86687076e-01 -5.36123037e-01 9.12182257e-02] | [7.555962562561035, 8.036498069763184] |
8794e3fb-21c8-4f2f-81c3-5a99c9dc544a | fac-3d-representation-learning-via-foreground | 2303.06388 | null | https://arxiv.org/abs/2303.06388v2 | https://arxiv.org/pdf/2303.06388v2.pdf | FAC: 3D Representation Learning via Foreground Aware Feature Contrast | Contrastive learning has recently demonstrated great potential for unsupervised pre-training in 3D scene understanding tasks. However, most existing work randomly selects point features as anchors while building contrast, leading to a clear bias toward background points that often dominate in 3D scenes. Also, object awareness and foreground-to-background discrimination are neglected, making contrastive learning less effective. To tackle these issues, we propose a general foreground-aware feature contrast (FAC) framework to learn more effective point cloud representations in pre-training. FAC consists of two novel contrast designs to construct more effective and informative contrast pairs. The first is building positive pairs within the same foreground segment where points tend to have the same semantics. The second is that we prevent over-discrimination between 3D segments/objects and encourage foreground-to-background distinctions at the segment level with adaptive feature learning in a Siamese correspondence network, which adaptively learns feature correlations within and across point cloud views effectively. Visualization with point activation maps shows that our contrast pairs capture clear correspondences among foreground regions during pre-training. Quantitative experiments also show that FAC achieves superior knowledge transfer and data efficiency in various downstream 3D semantic segmentation and object detection tasks. | ['Ling Shao', 'Shijian Lu', 'Xiaoqin Zhang', 'Aoran Xiao', 'Kangcheng Liu'] | 2023-03-11 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Liu_FAC_3D_Representation_Learning_via_Foreground_Aware_Feature_Contrast_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Liu_FAC_3D_Representation_Learning_via_Foreground_Aware_Feature_Contrast_CVPR_2023_paper.pdf | cvpr-2023-1 | ['unsupervised-pre-training'] | ['methodology'] | [ 2.65185267e-01 -1.26421541e-01 -1.40893310e-01 -4.92863297e-01
-3.83857131e-01 -4.88613278e-01 6.36242032e-01 2.74747193e-01
-2.65279710e-01 1.69283405e-01 -2.16685385e-01 -6.49948791e-02
7.67317554e-03 -8.71974349e-01 -8.96611035e-01 -7.55466640e-01
4.05154936e-02 4.94256616e-01 8.39412391e-01 -2.18216673e-01
4.34102237e-01 1.06576014e+00 -1.72132993e+00 3.53132844e-01
1.10625422e+00 8.64513338e-01 5.97025275e-01 1.38486788e-01
-7.82517791e-01 4.25339103e-01 -7.35564530e-01 -1.02384901e-02
6.94747329e-01 -1.66801676e-01 -6.87322438e-01 4.29838389e-01
8.50962043e-01 -1.87231719e-01 3.89820598e-02 1.16107225e+00
1.82177156e-01 1.42505363e-01 8.12390387e-01 -1.37593067e+00
-6.09310746e-01 2.21839234e-01 -9.79119718e-01 5.52479088e-01
-7.16191623e-03 3.58636230e-01 9.07151461e-01 -1.06944263e+00
4.82966423e-01 1.31695688e+00 5.26554286e-01 3.10148865e-01
-1.27858710e+00 -6.26901686e-01 6.18795276e-01 1.05133601e-01
-1.19171178e+00 1.47577047e-01 1.34677637e+00 -6.06689215e-01
6.65107906e-01 4.28648233e-01 1.07046533e+00 7.51660526e-01
-1.61983848e-01 1.09132624e+00 1.15320385e+00 -2.42887482e-01
2.42543802e-01 1.25792876e-01 1.49135500e-01 5.51653266e-01
2.98405081e-01 1.10612758e-01 -6.30805016e-01 1.52450010e-01
1.38741553e+00 3.21504295e-01 -1.73076391e-01 -1.02419698e+00
-1.26786947e+00 6.57956302e-01 9.16994035e-01 3.81912172e-01
-1.94829956e-01 -1.37907907e-01 1.19855516e-01 2.06317008e-01
4.36235815e-01 5.65112054e-01 -4.73519117e-01 4.41580206e-01
-8.63904417e-01 1.73118562e-01 2.45037749e-01 1.13573027e+00
1.20089209e+00 4.13369201e-02 -1.58158675e-01 7.75269687e-01
1.68697521e-01 4.58882570e-01 -6.69880072e-04 -1.04189062e+00
3.63812864e-01 1.19251895e+00 -1.96228608e-01 -1.09347975e+00
-2.38594115e-01 -7.13471949e-01 -7.75432706e-01 4.95797664e-01
5.81523240e-01 3.74039620e-01 -1.22596622e+00 1.45708585e+00
6.36802614e-01 4.55543160e-01 -2.36659750e-01 1.12005103e+00
8.41464520e-01 5.40812492e-01 1.22610621e-01 -5.74767636e-03
1.20467579e+00 -8.00624311e-01 -2.78695375e-01 -3.83526087e-01
3.85646909e-01 -5.53140521e-01 1.30374324e+00 -4.43089716e-02
-1.16217005e+00 -8.25733721e-01 -9.42054272e-01 -2.05779523e-01
-4.72443610e-01 -2.48481870e-01 7.68583536e-01 2.86880434e-01
-6.93714917e-01 5.05176485e-01 -7.61038244e-01 -1.05907470e-01
1.03651154e+00 2.08968952e-01 -1.18010104e-01 -2.53245831e-02
-7.79411376e-01 6.37943327e-01 4.01782513e-01 -1.11887284e-01
-7.64186502e-01 -1.21577442e+00 -6.56382978e-01 -2.92377803e-03
3.75343710e-01 -8.00872087e-01 7.45881438e-01 -1.26179993e+00
-1.33494830e+00 1.30803132e+00 8.61798078e-02 -1.84400249e-02
6.49653733e-01 -3.14601153e-01 7.53627568e-02 4.26558733e-01
2.20411599e-01 1.07411540e+00 1.01911759e+00 -1.81218231e+00
-8.23074222e-01 -4.04474080e-01 7.22992644e-02 6.18090630e-01
1.77829918e-02 -3.73688638e-01 -5.59532106e-01 -5.77602863e-01
7.49922156e-01 -7.35787511e-01 -2.01812178e-01 4.34723288e-01
-2.11271882e-01 -3.47531199e-01 1.00435627e+00 1.80293489e-02
5.21138787e-01 -2.09676480e+00 3.14920582e-02 4.13248807e-01
3.37344378e-01 1.50617510e-01 -1.46460176e-01 -1.27278879e-01
-8.83309990e-02 -8.58868659e-02 -3.56282711e-01 1.57192275e-01
-2.22368062e-01 3.82511377e-01 -1.44709721e-01 3.47726762e-01
8.35236490e-01 9.36717689e-01 -1.04685223e+00 -7.40456283e-01
6.42338097e-01 3.86575639e-01 -7.42503881e-01 1.97461918e-01
-3.21303934e-01 8.09743524e-01 -6.39390349e-01 8.28835130e-01
9.88735080e-01 -5.25607288e-01 -2.59691715e-01 -3.71821851e-01
-1.55701563e-01 1.43618479e-01 -1.22479403e+00 1.75408673e+00
-7.35261366e-02 4.74027097e-01 -4.12912294e-02 -1.12816334e+00
9.84354198e-01 -3.57403994e-01 5.95437467e-01 -6.91861987e-01
8.90862122e-02 7.35065863e-02 -3.22057307e-02 -3.42926681e-01
2.39729643e-01 8.75426978e-02 3.78867835e-01 -6.33588210e-02
-2.83321030e-02 -7.78863311e-01 -1.24782681e-01 9.97220054e-02
6.12556577e-01 1.94896981e-01 5.32245031e-03 -5.36568642e-01
4.93191302e-01 1.90939426e-01 6.53397739e-01 7.76097536e-01
-2.74997354e-01 6.70020223e-01 2.91241109e-01 -4.24104065e-01
-7.26902068e-01 -1.58414769e+00 -3.10545206e-01 1.08052242e+00
9.33910787e-01 1.92100722e-02 -4.64784563e-01 -7.64182866e-01
2.68848568e-01 5.93621492e-01 -5.17683148e-01 -1.39260530e-01
-8.02346468e-01 -4.26520765e-01 -1.44941723e-02 5.77966511e-01
7.79895544e-01 -8.79190147e-01 -8.65481794e-01 -9.59977806e-02
-3.14664878e-02 -9.58494186e-01 -3.95028442e-01 5.81589580e-01
-1.25083172e+00 -1.09667087e+00 -7.54125178e-01 -1.03888881e+00
1.07931447e+00 8.41928840e-01 1.35579038e+00 1.31617054e-01
-3.12828422e-01 2.90285975e-01 -2.05462784e-01 -4.32665884e-01
-5.22788130e-02 -1.68186381e-01 -2.74799019e-01 -2.32695758e-01
5.15091419e-01 -5.84705234e-01 -8.40253711e-01 6.52758956e-01
-7.41588056e-01 2.05702007e-01 7.39647269e-01 6.80056632e-01
1.00807130e+00 -3.50610912e-01 8.93862247e-02 -8.61072958e-01
1.63371246e-02 -2.29882017e-01 -5.34673989e-01 1.38301298e-01
-2.37201914e-01 -1.00255981e-01 1.45105302e-01 -6.11133873e-01
-8.52669060e-01 1.21725634e-01 1.37534305e-01 -8.99883986e-01
-4.35779274e-01 -1.41162455e-01 -3.14686418e-01 -1.77080423e-01
6.33602262e-01 2.53239274e-01 -7.25885034e-02 -3.34317625e-01
4.48482990e-01 6.12075217e-02 4.19946492e-01 -7.66000807e-01
9.67366815e-01 6.69820011e-01 3.39475088e-02 -1.06283712e+00
-9.30629015e-01 -6.57863855e-01 -9.95234132e-01 -2.85721123e-01
8.45903337e-01 -1.00343060e+00 -3.22824180e-01 3.53896081e-01
-8.59480858e-01 -4.50910777e-01 -7.36254394e-01 3.61986935e-01
-5.30875266e-01 1.88069955e-01 -3.91412526e-01 -4.67914432e-01
1.76116899e-02 -1.06401479e+00 1.28042603e+00 3.52978885e-01
-9.17007774e-03 -7.35480547e-01 -3.70914102e-01 1.57258004e-01
9.67157334e-02 3.05233568e-01 1.05521572e+00 -4.08661664e-01
-1.03905702e+00 3.55417788e-01 -5.72796583e-01 2.20440045e-01
3.31228107e-01 1.07005529e-01 -8.85493279e-01 -6.93449229e-02
-1.82435550e-02 3.22565138e-02 8.71100783e-01 5.70237994e-01
1.46825552e+00 9.63156596e-02 -5.44878662e-01 7.95579374e-01
1.13255346e+00 6.17571883e-02 3.35948080e-01 1.76751062e-01
9.29257393e-01 7.36205041e-01 6.32039487e-01 -4.09566909e-02
6.47876784e-02 4.10898060e-01 4.43765312e-01 -5.42730510e-01
-2.74472117e-01 -2.34224379e-01 -9.88806784e-02 6.38575733e-01
-1.38144806e-01 1.88339904e-01 -8.60320270e-01 4.25562859e-01
-1.47056675e+00 -7.85897255e-01 -1.80124834e-01 1.93271434e+00
8.34928393e-01 5.08129716e-01 1.41906649e-01 -1.09908264e-02
7.68729687e-01 3.70227918e-02 -7.21430302e-01 3.06623578e-01
-3.95356059e-01 1.33632883e-01 4.37163621e-01 2.03587919e-01
-1.14637589e+00 1.09096742e+00 5.47866964e+00 8.09114873e-01
-1.18943620e+00 -1.10060856e-01 7.43359506e-01 -1.04003273e-01
-4.42820072e-01 -1.50162861e-01 -6.80682540e-01 3.69336784e-01
-1.36565715e-01 2.98459619e-01 -1.40169486e-01 9.60689068e-01
-2.59040948e-02 -9.74938869e-02 -1.25613797e+00 1.33920920e+00
-2.12199256e-01 -1.49105787e+00 4.60339636e-01 -1.09924994e-01
9.18189466e-01 4.25161459e-02 7.62099475e-02 1.03456497e-01
1.50576308e-01 -7.13533700e-01 8.07780325e-01 3.46433759e-01
3.34914565e-01 -7.04135835e-01 2.75227427e-01 2.73067981e-01
-1.22512674e+00 3.28708515e-02 -5.93885481e-01 2.57549226e-01
-1.01674139e-01 8.49297106e-01 -5.98314345e-01 3.48061532e-01
9.85976636e-01 9.06449676e-01 -6.32315755e-01 1.15699708e+00
-1.28061637e-01 2.97595292e-01 -4.36969072e-01 -4.87908721e-02
4.17179644e-01 -4.09300327e-01 7.52763331e-01 1.19030070e+00
2.55806483e-02 1.46527857e-01 4.41179991e-01 1.27372980e+00
1.43561706e-01 5.43437991e-03 -5.08348465e-01 3.42257529e-01
4.91495758e-01 1.07160175e+00 -1.22940743e+00 -3.52083266e-01
-4.27420557e-01 9.18630540e-01 2.63262957e-01 4.09366310e-01
-6.36616230e-01 4.71400134e-02 7.77221978e-01 5.06300688e-01
4.80104715e-01 -4.42605197e-01 -5.99981725e-01 -1.03144526e+00
-1.46869108e-01 -4.98268068e-01 3.64979029e-01 -6.79546595e-01
-1.51555240e+00 2.54904628e-01 3.64239007e-01 -1.47093523e+00
4.42677975e-01 -5.85220873e-01 -5.67936003e-01 6.93305254e-01
-1.77975261e+00 -1.00121117e+00 -5.86958528e-01 7.42989600e-01
7.36203253e-01 1.34665892e-01 1.58345193e-01 1.33662879e-01
-1.72701955e-01 2.22305059e-01 -2.99629748e-01 7.58251771e-02
4.90428269e-01 -1.48864555e+00 2.83035040e-01 6.38143182e-01
2.23028272e-01 7.12260067e-01 3.34596455e-01 -7.05356658e-01
-1.09153056e+00 -1.02297437e+00 1.25546470e-01 -5.40686369e-01
2.20711365e-01 -5.12135088e-01 -1.29581797e+00 2.33326659e-01
-1.08156689e-01 2.20651105e-01 5.05540550e-01 1.48126418e-02
-3.45410168e-01 -1.68604463e-01 -1.05147970e+00 7.62326419e-01
1.46346402e+00 -4.44083452e-01 -7.99182594e-01 2.44049311e-01
7.19176352e-01 -4.51354206e-01 -5.60007751e-01 7.21615493e-01
8.36824402e-02 -9.33205366e-01 1.36105382e+00 -4.24922884e-01
3.17788839e-01 -6.38018906e-01 -3.08281053e-02 -1.20074689e+00
-4.45075005e-01 -2.62736797e-01 -1.14664413e-01 1.08310246e+00
4.32256907e-02 -3.56560320e-01 1.04239154e+00 1.48418486e-01
-3.62334251e-01 -6.31625533e-01 -8.40544105e-01 -7.78682530e-01
1.87622309e-01 -2.40671724e-01 5.83698452e-01 1.21096253e+00
-7.24229872e-01 3.67889732e-01 3.82847190e-01 3.79209250e-01
7.30605960e-01 6.30349696e-01 9.23164189e-01 -1.64335263e+00
4.55518924e-02 -8.53333890e-01 -4.73086298e-01 -1.57202947e+00
-2.08717212e-01 -1.08316267e+00 4.80605327e-02 -1.39017248e+00
7.14886887e-03 -8.76930296e-01 -3.55002761e-01 2.99998313e-01
-4.22801554e-01 1.30209237e-01 1.48097292e-01 4.00648355e-01
-4.49637592e-01 6.00199997e-01 1.94264185e+00 -3.30737352e-01
-5.16985238e-01 4.66803797e-02 -5.22928774e-01 8.50110233e-01
5.98543465e-01 -2.62686670e-01 -5.75210333e-01 -4.47063804e-01
-1.74949929e-01 -3.76769692e-01 6.30350769e-01 -1.03243828e+00
1.26805408e-02 -4.58268374e-01 9.72587228e-01 -1.10088384e+00
3.32285851e-01 -1.02448571e+00 -1.68983370e-01 3.13459963e-01
-1.82113916e-01 -1.74227059e-01 3.16584498e-01 6.53829753e-01
-7.51366988e-02 3.43780182e-02 9.69211400e-01 -4.07538682e-01
-1.00579643e+00 4.04817224e-01 -1.39426226e-02 2.84455061e-01
1.09164715e+00 -7.89839029e-01 -1.10462792e-01 1.39025569e-01
-5.78122616e-01 3.87671918e-01 5.34040093e-01 4.52608228e-01
8.35723996e-01 -1.07075965e+00 -5.26413143e-01 6.17290556e-01
2.35636130e-01 7.89363563e-01 2.74180442e-01 6.83349133e-01
-7.00243592e-01 -8.44603330e-02 -4.51056600e-01 -1.55548441e+00
-1.05396724e+00 5.02308369e-01 4.22314942e-01 2.38971978e-01
-9.28548217e-01 1.09462762e+00 8.86295259e-01 -3.95658404e-01
2.91856140e-01 -6.71965539e-01 -6.32796511e-02 -5.13664596e-02
2.51968622e-01 1.50187254e-01 -9.49619189e-02 -3.81845832e-01
-4.21231091e-01 9.44384634e-01 -7.50278682e-02 3.27492744e-01
1.32465780e+00 -5.85101619e-02 1.25561088e-01 6.70284748e-01
1.10665476e+00 -5.60313798e-02 -1.73929679e+00 -2.57305324e-01
2.99341540e-04 -8.34016502e-01 2.57791821e-02 -5.46348929e-01
-1.19889939e+00 9.55091953e-01 8.42217147e-01 2.38293245e-01
9.80598390e-01 3.14569175e-01 4.06914651e-01 3.25004250e-01
1.56207949e-01 -1.06282866e+00 4.78069037e-01 4.48755085e-01
6.84078693e-01 -1.43313944e+00 1.19274125e-01 -9.42853034e-01
-6.30758107e-01 9.42547441e-01 1.01022482e+00 -5.07761300e-01
6.61065102e-01 4.79963794e-02 1.18701920e-01 -4.47421163e-01
-2.03793198e-01 -5.08623719e-01 5.80243766e-01 8.66339266e-01
-3.56075801e-02 -1.99601859e-01 1.42336190e-01 -3.24592702e-02
-2.11301688e-02 -6.32836759e-01 -6.48840964e-02 1.02834189e+00
-4.93248701e-01 -7.25299180e-01 -2.74362475e-01 4.54657644e-01
1.23394839e-02 7.18044937e-02 -4.71805215e-01 1.12456501e+00
2.95976847e-01 4.62540418e-01 6.61620736e-01 -1.25933597e-02
4.16902006e-01 -1.32031143e-01 7.97716200e-01 -6.39785647e-01
-6.68728352e-01 3.33430737e-01 -5.89965582e-01 -4.92957294e-01
-7.54277229e-01 -5.32972813e-01 -1.51644361e+00 8.63598287e-02
-4.99966085e-01 5.82690425e-02 3.84877294e-01 7.23281920e-01
3.02754194e-01 7.11345732e-01 6.90410197e-01 -1.11612618e+00
-3.70735191e-02 -4.87324715e-01 -4.94424641e-01 6.89653039e-01
2.73660988e-01 -1.12179434e+00 -3.85922521e-01 5.99841587e-02] | [7.957724571228027, -3.2642292976379395] |
6062a337-0d95-4a2d-bc57-b1b6d9fb48f3 | data-augmentation-for-personal-knowledge | 2002.10943 | null | https://arxiv.org/abs/2002.10943v2 | https://arxiv.org/pdf/2002.10943v2.pdf | Data Augmentation for Personal Knowledge Base Population | Cold start knowledge base population (KBP) is the problem of populating a knowledge base from unstructured documents. While artificial neural networks have led to significant improvements in the different tasks that are part of KBP, the overall F1 of the end-to-end system remains quite low. This problem is more acute in personal knowledge bases, which present additional challenges with regard to data protection, fairness and privacy. In this work, we present a system that uses rule based annotators and a graph neural network for missing link prediction, to populate a more complete, fair and diverse knowledge base from the TACRED dataset. | ['Hima Patel', 'Lokesh Nagalapatti', 'MN Thippeswamy', 'Lingraj S Vannur', 'Balaji Ganesan'] | 2020-02-23 | null | null | null | null | ['knowledge-base-population'] | ['natural-language-processing'] | [-2.95525998e-01 7.63107359e-01 -3.96264017e-01 -2.56178826e-01
-5.26798069e-01 -6.71899676e-01 2.30400354e-01 5.35914600e-01
-1.94893003e-01 1.48462749e+00 8.51838812e-02 -6.69096112e-02
-5.28084159e-01 -8.78435671e-01 -4.30038303e-01 -2.34733313e-01
9.20119062e-02 1.21573937e+00 4.63581026e-01 -4.56487924e-01
-1.46436572e-01 3.01486671e-01 -1.03344607e+00 4.15037721e-01
1.12595308e+00 8.32374632e-01 -5.55793703e-01 2.98731834e-01
-2.47776404e-01 1.01296175e+00 -8.40180218e-01 -1.54669976e+00
3.50665808e-01 2.50391275e-01 -1.41985667e+00 -6.67113304e-01
6.12234592e-01 1.13525935e-01 -2.89542854e-01 1.04789674e+00
8.86589170e-01 6.34139925e-02 4.33158040e-01 -1.64929366e+00
-1.07287765e+00 1.16752458e+00 -2.47046381e-01 -9.29145142e-02
5.83650053e-01 -2.60820359e-01 1.11955023e+00 -3.52149963e-01
1.14917839e+00 1.08199883e+00 9.75798130e-01 6.29931271e-01
-1.40742278e+00 -7.07619786e-01 4.24713269e-02 6.24209166e-01
-1.62720180e+00 -4.19243485e-01 3.75216663e-01 -3.02250922e-01
9.03126478e-01 5.36217451e-01 5.14605820e-01 1.17124426e+00
-2.64824003e-01 6.80951834e-01 7.36088336e-01 -5.23222625e-01
2.89438754e-01 5.82454264e-01 7.00543225e-01 4.70600069e-01
9.13105309e-01 -1.58509910e-01 -6.92783654e-01 -7.89258718e-01
-1.17294677e-01 -4.37285662e-01 -3.81620735e-01 -4.85425085e-01
-5.63802540e-01 6.94116116e-01 4.14806843e-01 9.03174467e-03
-3.69727552e-01 -3.36496770e-01 4.03205037e-01 3.50322753e-01
1.92717478e-01 8.21140349e-01 -5.33226252e-01 2.92315960e-01
-8.27091157e-01 5.70530534e-01 1.67519188e+00 1.31068063e+00
5.72368920e-01 -3.73947918e-01 -6.58920109e-01 7.00565100e-01
4.66878004e-02 3.44537944e-01 -2.03668147e-01 -6.73775554e-01
5.47206581e-01 7.43557155e-01 3.03356558e-01 -1.09356487e+00
-4.84338135e-01 -2.16648087e-01 -6.47117734e-01 -5.96926250e-02
5.14673710e-01 -3.75289381e-01 -9.93629694e-01 1.40568960e+00
5.04061878e-01 -1.02874726e-01 3.57179761e-01 5.67148507e-01
1.27434814e+00 3.03361267e-01 4.12739694e-01 -6.53874725e-02
1.39966226e+00 -8.83405566e-01 -1.01989818e+00 -1.06701508e-01
3.64755064e-01 -3.90309483e-01 2.75876760e-01 4.92859572e-01
-1.02107263e+00 -1.26323640e-01 -9.01784003e-01 -2.44211242e-01
-9.53996897e-01 -4.91679430e-01 5.16555786e-01 1.05365980e+00
-1.04093790e+00 3.49667579e-01 -1.41558722e-01 -5.86020589e-01
6.83786094e-01 4.42071468e-01 -6.96172655e-01 -3.56924325e-01
-1.84940088e+00 1.41791546e+00 1.32694936e+00 -1.22847512e-01
-3.99225444e-01 -8.95506144e-01 -5.48885882e-01 8.04407299e-02
9.79622543e-01 -1.10405147e+00 8.12675536e-01 -3.71968567e-01
-1.04508340e+00 4.95630026e-01 3.73532236e-01 -6.78649604e-01
5.68633437e-01 -1.59567833e-01 -7.84637213e-01 -2.20293626e-01
-1.29372150e-01 3.91020060e-01 5.40136933e-01 -1.62928557e+00
-7.66992807e-01 -2.61949748e-01 2.96817962e-02 1.45669833e-01
-2.22981110e-01 7.54427537e-02 -8.10440063e-01 -3.31590384e-01
-6.92799687e-01 -9.50949073e-01 -2.65317410e-01 -4.38854694e-01
-9.99810636e-01 -9.13503692e-02 7.09507942e-01 -1.22032475e+00
1.53220689e+00 -1.47103477e+00 1.42235294e-01 8.79548430e-01
2.57607728e-01 7.70651698e-01 9.77135003e-02 5.06431818e-01
6.90396428e-02 4.00713533e-01 4.32630889e-02 -1.50065050e-02
8.36452767e-02 4.25035685e-01 -2.79026568e-01 -2.35640168e-01
3.66629176e-02 8.93431067e-01 -9.78040993e-01 -6.31669521e-01
-4.63821590e-01 2.65905738e-01 -2.73215026e-01 -5.51516265e-02
-4.89296973e-01 2.37035416e-02 -3.35439473e-01 9.29559231e-01
7.70270228e-01 6.31193146e-02 6.04747236e-01 -1.88540503e-01
5.26848257e-01 -2.20235884e-01 -1.24761701e+00 1.42054236e+00
2.54060417e-01 1.33134201e-01 1.24783535e-02 -5.85627973e-01
9.82447267e-01 3.44988197e-01 1.28904521e-01 -4.32415605e-01
-1.22448906e-01 9.28283259e-02 -1.29377156e-01 -6.91635251e-01
1.01125264e+00 -7.28211179e-02 -9.27408785e-02 1.37530193e-01
2.26907536e-01 6.68091401e-02 6.07908309e-01 5.24012983e-01
1.31059396e+00 -7.52068907e-02 4.09960479e-01 -1.12277120e-02
3.89717758e-01 3.40238780e-01 6.67583287e-01 8.25765848e-01
-2.56699752e-02 3.26788962e-01 6.79234266e-01 -5.23479998e-01
-7.85379052e-01 -9.40896749e-01 -3.92645933e-02 8.65512252e-01
-3.52091715e-02 -6.54283762e-01 -5.50379932e-01 -1.18071592e+00
7.00125515e-01 1.08074594e+00 -4.50558871e-01 -3.14662695e-01
-2.98391551e-01 -8.11556220e-01 1.00965345e+00 3.41550618e-01
1.40386760e-01 -1.09262109e+00 2.38884911e-01 3.36061567e-01
-5.35564303e-01 -1.16535902e+00 -3.97550780e-03 5.10295108e-02
-2.36165032e-01 -1.39594996e+00 -3.48208398e-01 -4.83501613e-01
6.03677452e-01 -4.74207342e-01 1.50690258e+00 1.49988523e-02
-3.59832853e-01 1.45822465e-01 -5.38957119e-01 -1.00231326e+00
-4.81307089e-01 4.34163362e-01 -1.04379609e-01 -2.36657128e-01
6.37418330e-01 -1.17677808e-01 -4.81020845e-02 2.63930172e-01
-9.44995582e-01 -4.41980094e-01 2.23410875e-01 7.95339286e-01
5.57343721e-01 3.91643733e-01 6.82475507e-01 -1.47132146e+00
1.09749496e+00 -5.86026490e-01 -3.91330779e-01 9.37819242e-01
-8.66466939e-01 1.04507923e-01 5.42797387e-01 -4.50367294e-03
-1.23392856e+00 -1.28459362e-02 -8.68770555e-02 -5.10754958e-02
2.20397860e-01 7.43334353e-01 -2.36684814e-01 -1.44098043e-01
1.17603445e+00 -3.27088237e-01 -9.85901505e-02 -4.69914466e-01
7.28844464e-01 5.67214251e-01 6.02723956e-01 -6.97340727e-01
8.56087744e-01 2.17416570e-01 -1.89895034e-01 -4.21972096e-01
-1.05591083e+00 -3.82930994e-01 -6.40342355e-01 8.10419843e-02
5.15059054e-01 -8.15984190e-01 -7.22894788e-01 1.58502579e-01
-9.01670575e-01 -1.57506019e-02 -5.84710717e-01 -1.33041516e-01
-1.77069023e-01 4.18034703e-01 -6.01892956e-02 -7.85179019e-01
-1.01443160e+00 -4.62468565e-01 2.00702846e-01 2.08875045e-01
-4.05086577e-01 -8.46256316e-01 2.85208404e-01 8.56996417e-01
2.94191003e-01 4.79581177e-01 1.06205654e+00 -1.25281465e+00
-3.18495870e-01 -5.63118875e-01 -1.99547604e-01 2.86771208e-01
-8.38095844e-02 6.36665151e-02 -7.90829480e-01 -2.89387763e-01
-9.21864986e-01 -4.96224523e-01 8.09462368e-01 -1.20838895e-01
9.36548650e-01 -5.29882908e-01 -6.88699365e-01 2.71289349e-01
1.45234239e+00 9.08324271e-02 9.42860186e-01 4.37691182e-01
6.31049037e-01 7.31071115e-01 5.83442509e-01 3.28280985e-01
6.63094580e-01 6.73531413e-01 2.78460801e-01 7.28009939e-02
-1.48592725e-01 -1.67523205e-01 -1.53704748e-01 1.21872626e-01
-3.60265762e-01 -7.09989250e-01 -1.40260959e+00 6.56580031e-01
-2.30199504e+00 -1.26304936e+00 -4.16518003e-02 2.05103922e+00
1.13355339e+00 6.90115839e-02 2.65731245e-01 1.07425340e-02
9.14848566e-01 -2.49854103e-01 -4.41543221e-01 -5.50217152e-01
-3.28600109e-01 1.18065521e-01 6.65240943e-01 3.26901823e-01
-9.20244038e-01 9.20156181e-01 7.13226748e+00 8.89516413e-01
-4.34861392e-01 -4.49185371e-02 2.87984639e-01 -1.43749088e-01
-3.46843064e-01 5.96108474e-02 -1.20931840e+00 4.86507028e-01
8.94929111e-01 -4.07730460e-01 4.39916492e-01 8.06772113e-01
-6.01326227e-01 3.10743321e-02 -9.98362064e-01 6.10783458e-01
9.13463086e-02 -1.46603107e+00 2.98390359e-01 -1.74809605e-01
7.73459911e-01 6.61668507e-03 -2.71786839e-01 6.08858824e-01
1.10268176e+00 -9.67057765e-01 5.55811346e-01 9.73003566e-01
6.38676822e-01 -1.05987430e+00 1.09615529e+00 1.59818441e-01
-5.26221216e-01 -2.12153599e-01 -5.34231544e-01 4.58793372e-01
1.69997707e-01 8.25325131e-01 -1.24627495e+00 1.22267568e+00
9.11875725e-01 1.14579894e-01 -8.13873768e-01 1.36382222e+00
-2.86120981e-01 2.13904724e-01 -3.29547644e-01 -4.91423123e-02
-2.75859684e-01 1.39476776e-01 4.24368829e-01 1.24603713e+00
-1.94336497e-03 -8.00116509e-02 1.89752668e-01 5.39208531e-01
-5.65155983e-01 1.49568215e-01 -5.51244974e-01 -1.06277332e-01
7.56293178e-01 1.47416222e+00 -2.40074858e-01 -2.83929914e-01
-1.27268299e-01 5.23684204e-01 8.76370668e-01 4.07641530e-01
-5.66276908e-01 -6.80208504e-01 5.04312694e-01 1.91576518e-02
2.21004650e-01 5.59453666e-01 -1.88933730e-01 -9.89614308e-01
1.15717731e-01 -9.25628066e-01 1.21778750e+00 -6.15013421e-01
-1.75647163e+00 4.76727605e-01 3.29261869e-02 -6.25583172e-01
-1.81494996e-01 -4.01181728e-01 -3.21190089e-01 9.21929955e-01
-1.50444233e+00 -1.39590061e+00 -3.68109763e-01 6.93513989e-01
-5.19693494e-01 -3.48762035e-01 1.14075530e+00 4.19190913e-01
-7.01869547e-01 9.74070489e-01 3.09869260e-01 3.53510231e-01
1.02674139e+00 -1.43914723e+00 2.29412153e-01 6.69331908e-01
-1.03057608e-01 7.68580794e-01 6.68339312e-01 -1.13015282e+00
-1.10956728e+00 -1.33760965e+00 9.30930436e-01 -1.08560181e+00
4.54240173e-01 -2.86260277e-01 -9.76053357e-01 7.17901766e-01
2.23595440e-01 1.98206604e-01 1.12962162e+00 6.35465801e-01
-6.16438210e-01 -2.92731643e-01 -1.66923273e+00 4.06031191e-01
1.00335765e+00 -2.57401645e-01 -6.89660609e-01 2.60886490e-01
5.32863796e-01 -5.19077837e-01 -1.23268390e+00 2.80056149e-01
6.00224018e-01 -5.92265308e-01 1.17002285e+00 -1.08370364e+00
5.72903827e-02 -3.13948870e-01 2.66548932e-01 -1.50779831e+00
-6.24812067e-01 -7.71886528e-01 -7.01250851e-01 1.60676408e+00
9.68906343e-01 -4.78934705e-01 1.02639329e+00 1.41675508e+00
2.18982399e-01 -2.79608935e-01 -8.31356943e-01 -9.95502234e-01
-1.77434251e-01 -9.68514197e-03 9.01345730e-01 1.32525396e+00
4.12112892e-01 5.07615626e-01 -7.61106610e-01 2.07398728e-01
6.08509839e-01 -1.58268094e-01 7.30866671e-01 -1.74883664e+00
7.64411613e-02 -4.24744934e-02 -4.17655110e-01 2.28760704e-01
1.23119161e-01 -1.30442882e+00 -2.62298882e-01 -2.06203055e+00
2.82358140e-01 -7.02925682e-01 -5.56430042e-01 1.05100179e+00
-3.18282694e-01 6.81144446e-02 2.36281708e-01 -7.14875832e-02
-8.85405362e-01 1.03992477e-01 8.87187064e-01 -3.09868574e-01
-3.20709705e-01 -2.52955496e-01 -1.18386257e+00 4.43636805e-01
7.66594291e-01 -6.81905389e-01 -3.71894777e-01 -2.95782924e-01
7.06570685e-01 -5.15664034e-02 2.98786070e-02 -7.81090379e-01
5.95350504e-01 -6.27471954e-02 4.28217143e-01 -6.74526393e-01
1.97659984e-01 -1.03427339e+00 7.56076634e-01 4.31844324e-01
-1.79858997e-01 -3.36853921e-01 3.17775726e-01 7.26894796e-01
-3.52827869e-02 -5.07199049e-01 3.75211418e-01 -8.26253518e-02
-8.11292827e-01 3.56241375e-01 2.38173321e-01 2.55158275e-01
1.24807227e+00 -1.05408557e-01 -9.77746069e-01 -1.10138133e-01
-1.07495248e+00 7.84227252e-01 3.63478243e-01 5.66054583e-01
3.29665154e-01 -1.25949943e+00 -9.33064640e-01 -2.25053117e-01
5.83920419e-01 7.68226460e-02 1.76845118e-01 4.28686708e-01
-5.99355042e-01 2.37531602e-01 -4.97206807e-01 2.85848349e-01
-1.28846860e+00 8.13389957e-01 4.96972390e-02 -5.94379187e-01
-3.65115225e-01 9.32453334e-01 -7.08341360e-01 -5.87791145e-01
5.21526158e-01 4.34163153e-01 -4.94330049e-01 5.41713536e-01
6.06882751e-01 8.04152250e-01 2.69496441e-01 -2.33836800e-01
-4.73204732e-01 -1.20996855e-01 -5.45190334e-01 4.72654253e-01
1.34430933e+00 9.54753011e-02 -3.53624135e-01 -2.71261454e-01
4.18929756e-01 2.21967295e-01 -4.14701670e-01 -4.86328870e-01
4.73959953e-01 -3.42172712e-01 -2.72580445e-01 -1.60837281e+00
-1.10851109e+00 -4.55346284e-03 1.60470545e-01 5.15567064e-01
7.87487328e-01 -9.78390500e-02 6.21966660e-01 9.28024530e-01
4.84536678e-01 -1.54141510e+00 -5.28425813e-01 5.75546563e-01
9.16252792e-01 -1.26243401e+00 5.21456182e-01 -5.51436543e-01
-9.25535679e-01 9.11466837e-01 7.69473135e-01 5.45185506e-01
6.14579678e-01 1.05080962e-01 -8.46642256e-02 -2.57643849e-01
-6.45959079e-01 -2.53662974e-01 3.02917957e-01 1.15880764e+00
-1.44284323e-01 9.85116791e-03 -3.75623792e-01 9.46862578e-01
-2.52569467e-01 2.99680978e-01 3.75457287e-01 9.56829786e-01
-1.19852640e-01 -1.31684041e+00 -4.06845063e-01 7.75420427e-01
-6.62097394e-01 -8.59540105e-02 -9.25275385e-01 6.93165362e-01
3.92442524e-01 1.01805556e+00 -5.10037541e-01 -5.04010916e-01
6.25835598e-01 3.96935493e-01 2.24962592e-01 -5.17979562e-01
-8.13596189e-01 -9.32216465e-01 1.15888453e+00 -2.32347980e-01
-1.72288388e-01 -2.25675538e-01 -9.82455671e-01 -6.49360240e-01
-5.53101361e-01 4.28897262e-01 3.30779195e-01 4.72430646e-01
5.08423328e-01 2.47251555e-01 -1.09467812e-01 1.92144625e-02
-4.16048348e-01 -5.95894873e-01 -7.24288702e-01 6.25229001e-01
-1.87949613e-01 -4.40117210e-01 3.58867019e-01 -4.18304019e-02] | [9.296930313110352, 8.433512687683105] |
0d5393f2-be37-4f91-976a-be7e4d65341a | leveraging-unlabeled-whole-slide-images-for | 1807.11677 | null | http://arxiv.org/abs/1807.11677v1 | http://arxiv.org/pdf/1807.11677v1.pdf | Leveraging Unlabeled Whole-Slide-Images for Mitosis Detection | Mitosis count is an important biomarker for prognosis of various cancers. At
present, pathologists typically perform manual counting on a few selected
regions of interest in breast whole-slide-images (WSIs) of patient biopsies.
This task is very time-consuming, tedious and subjective. Automated mitosis
detection methods have made great advances in recent years. However, these
methods require exhaustive labeling of a large number of selected regions of
interest. This task is very expensive because expert pathologists are needed
for reliable and accurate annotations. In this paper, we present a
semi-supervised mitosis detection method which is designed to leverage a large
number of unlabeled breast cancer WSIs. As a result, our method capitalizes on
the growing number of digitized histology images, without relying on exhaustive
annotations, subsequently improving mitosis detection. Our method first learns
a mitosis detector from labeled data, uses this detector to mine additional
mitosis samples from unlabeled WSIs, and then trains the final model using this
larger and diverse set of mitosis samples. The use of unlabeled data improves
F1-score by $\sim$5\% compared to our best performing fully-supervised model on
the TUPAC validation set. Our submission (single model) to TUPAC challenge
ranks highly on the leaderboard with an F1-score of 0.64. | ['Janne Heikkilä', 'Simon Graham', 'Saad Ullah Akram', 'Talha Qaiser', 'Nasir Rajpoot', 'Juho Kannala'] | 2018-07-31 | null | null | null | null | ['mitosis-detection'] | ['medical'] | [ 3.99986386e-01 3.17402780e-01 -4.87287790e-01 -1.68405920e-01
-1.39202225e+00 -6.08893514e-01 1.79261625e-01 7.57744730e-01
-7.32413888e-01 9.44737136e-01 -9.27694663e-02 -2.82584518e-01
2.81800985e-01 -6.59655213e-01 -4.19229984e-01 -1.20477998e+00
2.71072745e-01 7.98750937e-01 5.48229456e-01 3.15817982e-01
2.57084593e-02 4.26277161e-01 -9.86433864e-01 1.50054038e-01
6.35982990e-01 7.18507767e-01 2.81942964e-01 9.22623217e-01
4.49068211e-02 5.50506592e-01 -2.26248398e-01 -2.35230878e-01
-1.24912031e-01 -5.31368256e-01 -6.84529245e-01 2.30260834e-01
2.11802378e-01 -9.95977595e-02 -2.44922061e-02 1.02902508e+00
4.43453521e-01 -4.72460330e-01 8.07378054e-01 -8.24704945e-01
3.02848607e-01 5.44899225e-01 -8.19533765e-01 3.09300035e-01
-2.52645016e-01 2.32321247e-01 9.49619651e-01 -6.45527601e-01
8.69453013e-01 2.28519216e-01 7.31838107e-01 5.72345614e-01
-1.32248580e+00 -6.61026537e-01 -4.71379071e-01 -1.47339389e-01
-1.50264955e+00 -4.62584525e-01 3.92226934e-01 -3.78255218e-01
6.02175534e-01 2.15113178e-01 7.32687593e-01 5.64882874e-01
1.74661502e-01 8.51072252e-01 1.06608737e+00 -4.84004200e-01
4.31205750e-01 4.50175136e-01 2.34762743e-01 8.04850042e-01
4.72054601e-01 -2.68386990e-01 -2.57530093e-01 5.20699732e-02
5.37319541e-01 2.19003141e-01 -3.01886678e-01 -2.68996358e-01
-1.24667168e+00 6.28003955e-01 2.61362702e-01 4.91203755e-01
9.45130177e-03 -1.09877124e-01 4.63496923e-01 3.02602425e-02
4.79286730e-01 2.49659121e-01 -2.08369210e-01 1.05251923e-01
-1.33611870e+00 -2.15741768e-01 6.81985676e-01 3.96454155e-01
7.33781338e-01 -7.15101600e-01 -5.17967641e-02 6.24601901e-01
2.17175737e-01 5.13599217e-01 5.51008940e-01 -4.20804352e-01
-1.38347596e-01 1.02765381e+00 -2.98591107e-02 -5.55386901e-01
-7.07132638e-01 -6.84115410e-01 -1.17470145e+00 1.30599868e-02
1.03487718e+00 1.61042631e-01 -1.19584191e+00 1.30911875e+00
4.94165272e-01 -8.59163627e-02 -1.53595939e-01 8.36077511e-01
5.03088653e-01 1.99674606e-01 2.22850412e-01 -3.80437016e-01
1.38422418e+00 -7.78922796e-01 -4.87774223e-01 -1.95595682e-01
1.09799969e+00 -6.76132202e-01 9.11319256e-01 3.50702375e-01
-8.82430494e-01 -9.71532892e-03 -1.09717309e+00 1.14617757e-02
-2.94577599e-01 6.59335315e-01 5.75369596e-01 6.07869804e-01
-1.11773956e+00 3.38786125e-01 -1.19561577e+00 -6.48573816e-01
8.62512946e-01 7.64227986e-01 -6.07459843e-01 -2.10556865e-01
-4.01169538e-01 5.67945600e-01 3.19019556e-01 -9.02549922e-02
-7.26486623e-01 -6.66091621e-01 -6.94030046e-01 -1.23000666e-01
4.96435583e-01 -3.10121238e-01 1.29913342e+00 -5.39715052e-01
-1.09222019e+00 1.42976761e+00 -3.89254600e-01 -4.09357607e-01
5.44209659e-01 5.35059214e-01 5.14205508e-02 4.58215922e-01
9.14935693e-02 7.85030723e-01 4.27364558e-01 -9.83067989e-01
-9.99646783e-01 -4.35983509e-01 -5.17571568e-01 -2.21531451e-01
-3.47714901e-01 -4.93326277e-01 -6.77827656e-01 -3.42796534e-01
1.54785737e-01 -1.09184301e+00 -4.25963610e-01 2.81556696e-01
-3.69003177e-01 -1.62552267e-01 5.46577156e-01 -5.97040296e-01
1.06560743e+00 -2.24149942e+00 -1.19899236e-01 3.61095756e-01
4.94604379e-01 3.29349369e-01 2.30833873e-01 -1.56804591e-01
2.25138366e-01 2.53571898e-01 -2.20479190e-01 -5.06526113e-01
-2.33893767e-01 -3.23618241e-02 4.30656880e-01 7.61971891e-01
2.16545150e-01 8.65533054e-01 -1.06416726e+00 -1.03574622e+00
4.47081663e-02 1.64176807e-01 -4.03143495e-01 8.83427355e-03
-1.80267364e-01 4.20887053e-01 -7.51885548e-02 9.84146059e-01
3.65997761e-01 -8.53384316e-01 4.80543882e-01 -1.61713019e-01
2.39256680e-01 -8.87851641e-02 -9.22841311e-01 1.45788574e+00
1.89522412e-02 5.44868410e-01 1.63121503e-02 -8.76786649e-01
6.10808015e-01 3.18424255e-01 5.28762102e-01 -2.57990658e-01
3.38057935e-01 6.14447713e-01 3.80677208e-02 -3.27447891e-01
-1.57900285e-02 -3.76637042e-01 -1.25119492e-01 5.01674891e-01
4.63976488e-02 -2.73642123e-01 6.69570148e-01 2.83049256e-01
1.55897212e+00 -3.61681789e-01 7.37818062e-01 -2.59665400e-01
6.10887051e-01 3.33253741e-01 9.01432633e-01 2.57307380e-01
-4.95005101e-01 6.01387978e-01 6.88786745e-01 -3.56015921e-01
-1.09656072e+00 -7.75684595e-01 -1.22605078e-01 5.34025133e-01
-1.39820695e-01 -2.00595662e-01 -5.74261963e-01 -1.09827328e+00
-2.75388937e-02 1.04498215e-01 -9.62877989e-01 9.17833745e-02
-3.90312344e-01 -1.03181481e+00 5.07898927e-01 6.09290063e-01
3.22728485e-01 -7.63279319e-01 -5.37085235e-01 1.58690616e-01
-1.79315016e-01 -9.42599297e-01 -4.60560024e-01 5.01116574e-01
-8.08625340e-01 -1.36657000e+00 -8.91749263e-01 -1.01612878e+00
1.29357016e+00 2.75137097e-01 9.41010892e-01 3.19382697e-01
-8.37984741e-01 -1.47664055e-01 -2.48490244e-01 -5.27420342e-01
-4.72164035e-01 3.08099717e-01 -3.92074615e-01 -3.03105954e-02
6.56842828e-01 -1.51827499e-01 -7.13712692e-01 3.43843520e-01
-8.25848818e-01 1.59559160e-01 9.88409936e-01 1.12209332e+00
1.15346587e+00 1.03608951e-01 5.51501513e-01 -1.27105892e+00
-2.95226187e-01 -3.50743175e-01 -6.46788478e-01 1.52772039e-01
-2.74339408e-01 -2.08313391e-01 5.47389328e-01 -2.65853435e-01
-7.46040463e-01 3.59237492e-01 9.40419957e-02 -1.29954472e-01
-4.03427295e-02 5.97276807e-01 1.01715706e-01 -1.14973553e-01
6.59924448e-01 6.10625185e-02 3.00837845e-01 -3.81171182e-02
-3.30582410e-01 6.57244861e-01 6.16383374e-01 2.45313808e-01
5.51890910e-01 8.21563542e-01 2.85351962e-01 -7.12522745e-01
-9.34126914e-01 -1.02258348e+00 -5.37753940e-01 -1.74539834e-01
5.69904506e-01 -7.41364777e-01 -5.45936525e-01 5.61001480e-01
-5.04717767e-01 -7.26528943e-01 -2.99174309e-01 6.17276192e-01
-2.37756208e-01 2.20393538e-01 -9.21725214e-01 -5.33816397e-01
-4.13013428e-01 -9.33508217e-01 1.26729572e+00 4.78766441e-01
-4.76333678e-01 -8.97379339e-01 2.67518252e-01 5.84456623e-01
2.53440529e-01 3.65557581e-01 8.36038828e-01 -6.78110898e-01
-5.51039100e-01 -7.13087738e-01 -2.08565578e-01 -1.33512869e-01
3.08651298e-01 2.25386649e-01 -9.47107017e-01 -3.74375671e-01
-4.07988131e-01 -4.77638215e-01 1.14712727e+00 4.57991511e-01
1.03344190e+00 1.76127031e-01 -1.05742741e+00 4.51592743e-01
1.52768230e+00 6.80988742e-05 2.92604476e-01 1.22441016e-01
1.52809575e-01 5.85660517e-01 6.92116022e-01 1.52150705e-01
2.18353868e-01 1.19327288e-02 1.80270225e-01 -4.99771297e-01
-2.23788738e-01 -2.77611986e-02 -2.58982837e-01 4.57698911e-01
1.59198314e-01 -3.89802665e-01 -1.25900376e+00 9.57180500e-01
-1.58480740e+00 -5.36279380e-01 6.00837655e-02 2.10493517e+00
1.25021422e+00 3.52432281e-01 1.14553273e-01 6.12462699e-01
6.99921906e-01 -4.19188499e-01 -6.67935669e-01 3.38588387e-01
5.45446984e-02 3.88456464e-01 5.71755886e-01 2.77313113e-01
-1.16621399e+00 7.01627791e-01 5.75762701e+00 8.57224703e-01
-1.22747946e+00 1.26281884e-02 1.22493911e+00 -3.58315438e-01
1.97563902e-01 -2.09574550e-01 -1.01036906e+00 4.10537094e-01
8.91734660e-01 -1.00918755e-01 -1.30688608e-01 5.25105178e-01
1.97612107e-01 -8.03724945e-01 -1.16690755e+00 9.62561607e-01
2.22618058e-02 -1.50774837e+00 -4.47927326e-01 4.03348267e-01
7.46088088e-01 -8.02768096e-02 -2.67848969e-01 1.09451011e-01
3.68837088e-01 -1.01972246e+00 -1.02910504e-01 4.10520375e-01
1.18716002e+00 -6.72457039e-01 1.40620005e+00 4.90136206e-01
-9.12707448e-01 1.03929274e-01 -1.94549441e-01 4.66638118e-01
-2.05186531e-01 1.03332186e+00 -1.39837551e+00 -1.19172065e-02
3.20966333e-01 4.64778602e-01 -7.62280226e-01 1.37012720e+00
1.55560831e-02 7.97554195e-01 -5.24403393e-01 -2.66744584e-01
-5.08802347e-02 4.06542420e-01 -7.43439347e-02 1.32630479e+00
3.05395365e-01 1.27188057e-01 2.20127568e-01 2.73855001e-01
-2.24165842e-01 1.38666689e-01 -1.16795629e-01 -3.39353114e-01
3.75736535e-01 1.86317313e+00 -1.59552550e+00 -3.53317589e-01
-1.68703735e-01 4.83239383e-01 3.60765010e-01 -1.68935731e-01
-5.36666095e-01 -2.46807426e-01 -1.84037477e-01 3.95272255e-01
6.09420873e-02 2.50657082e-01 -3.18852246e-01 -7.88422763e-01
-3.39048713e-01 -6.59042716e-01 5.37334502e-01 -8.17799121e-02
-1.12083137e+00 2.21328899e-01 -5.08382499e-01 -1.21470845e+00
-1.58463959e-02 -3.98689300e-01 -5.76461673e-01 4.46708977e-01
-1.63743961e+00 -1.11694849e+00 -5.90816081e-01 1.32720947e-01
3.87438923e-01 1.87813997e-01 8.86458158e-01 2.93932080e-01
-6.88538492e-01 7.58161485e-01 -9.05484930e-02 2.79927552e-01
9.86016452e-01 -1.52854705e+00 -2.07385242e-01 4.55792964e-01
-3.13055634e-01 3.03064436e-01 4.46148127e-01 -4.77387518e-01
-1.14527273e+00 -1.11803973e+00 1.08866656e+00 -1.56190559e-01
4.93948698e-01 -2.01577440e-01 -7.19309092e-01 4.33054656e-01
-1.70184791e-01 3.51769030e-01 1.32916081e+00 -2.40557492e-01
1.17486641e-01 -1.53004989e-01 -1.26044762e+00 4.74057913e-01
5.25281906e-01 -2.10972190e-01 9.27317049e-03 4.81686413e-01
5.25719523e-02 -4.02742565e-01 -8.73492658e-01 4.31090623e-01
4.30451900e-01 -6.81299984e-01 3.96060884e-01 -4.03665826e-02
2.81418085e-01 -4.84839231e-01 2.61173755e-01 -9.54554915e-01
-2.06161574e-01 -2.79908508e-01 2.45394334e-01 9.73845363e-01
6.58932269e-01 -4.03279305e-01 1.53109872e+00 3.82965177e-01
4.50200215e-02 -1.00096285e+00 -7.60914326e-01 -2.76414752e-01
-7.97304213e-02 1.80602998e-01 -4.40886198e-03 6.32687867e-01
4.84252989e-01 3.08960348e-01 2.85885334e-01 -6.70928285e-02
7.65332043e-01 -7.77344033e-02 8.74109149e-01 -1.26165605e+00
-2.60624319e-01 -4.24989402e-01 -6.61830008e-01 -3.67469251e-01
-2.28554502e-01 -9.64419305e-01 1.94968805e-01 -1.35433328e+00
7.43231237e-01 -4.22940344e-01 -2.62110263e-01 7.47339845e-01
-4.57211763e-01 9.42543149e-01 -2.98491031e-01 3.84914696e-01
-9.09145832e-01 -1.65269643e-01 1.23442602e+00 -3.57060611e-01
-5.34537062e-02 -3.00814901e-02 -6.78749084e-01 8.31596792e-01
9.37918663e-01 -4.86834675e-01 -2.08920583e-01 3.32192600e-01
1.76761553e-01 6.30527362e-02 2.71763861e-01 -1.16003430e+00
5.03351331e-01 -6.73527047e-02 7.98937917e-01 -8.89342308e-01
2.21930668e-01 -4.45806295e-01 -3.20107955e-03 6.63736522e-01
-1.22859187e-01 -6.73472226e-01 1.56370848e-01 6.19720817e-01
-1.61043435e-01 -1.86597273e-01 1.00871837e+00 -3.13788414e-01
-2.37755194e-01 2.45298147e-01 -4.23845798e-01 -1.63373083e-01
1.31891191e+00 -4.23077315e-01 -2.94148147e-01 1.43158346e-01
-8.34476948e-01 3.78083199e-01 8.33155394e-01 -4.56334263e-01
1.68815568e-01 -8.06564510e-01 -7.06493974e-01 4.48715352e-02
2.77163714e-01 4.34302211e-01 2.50783831e-01 1.22154832e+00
-6.51107132e-01 4.41829532e-01 7.32124075e-02 -9.12102818e-01
-1.56318736e+00 3.35212886e-01 1.27647117e-01 -1.03373241e+00
-4.12093639e-01 1.12424302e+00 1.91072356e-02 -1.00650467e-01
2.34870166e-01 -3.86752307e-01 -1.93531632e-01 1.23423904e-01
6.01276577e-01 2.20684275e-01 2.85942703e-01 -1.89720839e-01
-3.05362433e-01 3.27846348e-01 -6.05645120e-01 4.73243855e-02
1.18245673e+00 1.73187315e-01 -1.04738787e-01 4.23055381e-01
1.11469781e+00 6.06007921e-03 -1.23694682e+00 -3.30617726e-01
2.19619378e-01 -2.42720202e-01 1.33904234e-01 -8.44065666e-01
-1.07258630e+00 6.03559971e-01 4.84779865e-01 -5.04723378e-02
1.15804124e+00 8.21449533e-02 6.59308136e-01 2.92866558e-01
3.60313147e-01 -1.07575154e+00 1.40556857e-01 6.06525457e-03
1.99283212e-01 -1.49822831e+00 7.94741213e-02 -6.28815472e-01
-2.76866853e-01 9.91430342e-01 3.82704228e-01 2.85950210e-02
5.27747214e-01 7.04888940e-01 2.63524234e-01 -2.46821761e-01
-9.52823997e-01 -1.54099450e-01 -1.91530034e-01 3.62718791e-01
5.54802239e-01 1.13190033e-01 -4.64749426e-01 5.87261796e-01
-2.59916615e-02 3.59713584e-01 4.00060087e-01 1.12924600e+00
-6.89133227e-01 -1.10428751e+00 -2.98087865e-01 8.71233761e-01
-6.91206694e-01 2.00113967e-01 -4.81512904e-01 9.11494434e-01
2.74387319e-02 7.07713366e-01 2.12410465e-02 3.64158154e-02
-2.09253892e-01 -8.01929161e-02 3.79045218e-01 -7.29423583e-01
-4.33395386e-01 3.49076241e-01 -3.68923359e-02 -1.50768384e-01
-4.59309161e-01 -7.92207658e-01 -1.43601310e+00 -1.59000739e-01
-5.32682359e-01 1.07476324e-01 5.18098176e-01 7.81653345e-01
-6.99682981e-02 5.18988311e-01 4.92372125e-01 -5.91929495e-01
-2.33558744e-01 -1.01586223e+00 -8.22266400e-01 1.45554185e-01
3.71532679e-01 -4.32720065e-01 -5.00374615e-01 5.51777899e-01] | [14.962019920349121, -3.0587403774261475] |
e4f2321c-8ba6-475d-87a0-35a282db2f4b | joint-motion-correction-and-super-resolution | 2107.03887 | null | https://arxiv.org/abs/2107.03887v1 | https://arxiv.org/pdf/2107.03887v1.pdf | Joint Motion Correction and Super Resolution for Cardiac Segmentation via Latent Optimisation | In cardiac magnetic resonance (CMR) imaging, a 3D high-resolution segmentation of the heart is essential for detailed description of its anatomical structures. However, due to the limit of acquisition duration and respiratory/cardiac motion, stacks of multi-slice 2D images are acquired in clinical routine. The segmentation of these images provides a low-resolution representation of cardiac anatomy, which may contain artefacts caused by motion. Here we propose a novel latent optimisation framework that jointly performs motion correction and super resolution for cardiac image segmentations. Given a low-resolution segmentation as input, the framework accounts for inter-slice motion in cardiac MR imaging and super-resolves the input into a high-resolution segmentation consistent with input. A multi-view loss is incorporated to leverage information from both short-axis view and long-axis view of cardiac imaging. To solve the inverse problem, iterative optimisation is performed in a latent space, which ensures the anatomical plausibility. This alleviates the need of paired low-resolution and high-resolution images for supervised learning. Experiments on two cardiac MR datasets show that the proposed framework achieves high performance, comparable to state-of-the-art super-resolution approaches and with better cross-domain generalisability and anatomical plausibility. | ['Wenjia Bai', 'Daniel Rueckert', 'Yike Guo', 'Stuart Cook', "Declan O'Regan", 'Chen Chen', 'Nicolo Savioli', 'Chen Qin', 'Shuo Wang'] | 2021-07-08 | null | null | null | null | ['cardiac-segmentation'] | ['medical'] | [ 4.91308093e-01 1.21373266e-01 -6.09529279e-02 -3.79507571e-01
-1.16871941e+00 -5.00120401e-01 2.30167195e-01 -1.47541732e-01
-3.09199363e-01 6.19831681e-01 2.99038887e-01 6.69968277e-02
-4.13410813e-01 -4.57298011e-01 -2.13027969e-01 -8.68916273e-01
-1.63035721e-01 5.89243829e-01 4.50011402e-01 3.89567494e-01
-9.24054310e-02 3.79845828e-01 -6.64239764e-01 2.97838837e-01
7.90673137e-01 5.19864857e-01 7.89992034e-01 6.62577689e-01
1.67965800e-01 6.38282299e-01 -1.92524423e-03 7.53961131e-02
1.95466027e-01 -5.94254553e-01 -1.25181556e+00 3.57901841e-01
2.24013001e-01 -4.62852627e-01 -2.31423415e-03 1.03901410e+00
6.22107387e-01 -1.58424079e-02 4.06407386e-01 -3.38959545e-01
-2.60422349e-01 4.71241385e-01 -8.20162416e-01 8.85854185e-01
1.85549092e-02 -4.38311324e-03 6.54465437e-01 -7.98729181e-01
8.59446168e-01 6.89531803e-01 6.31875932e-01 4.14529145e-01
-1.63233006e+00 -2.65621156e-01 -1.81477144e-01 -2.73602664e-01
-1.22670090e+00 -1.34079948e-01 8.55602741e-01 -6.19884551e-01
5.03398836e-01 1.51574656e-01 5.12233317e-01 6.39225543e-01
5.12323081e-01 2.65447885e-01 1.28402030e+00 -1.26360700e-01
-1.35027871e-01 -1.99091136e-01 -1.75751567e-01 4.64131117e-01
8.22555795e-02 8.04230496e-02 -3.13742518e-01 -2.85170794e-01
1.47751737e+00 -6.39268896e-03 -5.94804049e-01 -6.91616297e-01
-1.71623516e+00 5.52885056e-01 5.20370901e-01 5.49302578e-01
-7.93908238e-01 -2.81794280e-01 3.35955173e-01 -4.61253792e-01
4.58954036e-01 3.86431575e-01 -1.11739717e-01 1.61852524e-01
-1.31702459e+00 -1.75043687e-01 4.56482768e-02 4.78800982e-01
2.56020039e-01 6.54575154e-02 -3.63890901e-02 8.59609365e-01
3.20699096e-01 6.09048121e-02 6.57761931e-01 -1.36271584e+00
2.39049971e-01 1.67534903e-01 -1.32118195e-01 -6.67771518e-01
-5.73480129e-01 -6.88862920e-01 -1.35157001e+00 -1.53333684e-02
3.30412269e-01 2.89229691e-01 -8.71429801e-01 1.41018665e+00
7.10959375e-01 4.76984024e-01 -2.03563765e-01 1.58784997e+00
6.12258017e-01 2.30670780e-01 1.18480846e-01 -8.33243072e-01
1.43182659e+00 -7.60830879e-01 -8.58163357e-01 -1.77370265e-01
3.20529461e-01 -7.14196265e-01 7.10950255e-01 4.44658808e-02
-1.44225633e+00 -4.41082478e-01 -1.00883412e+00 1.74765944e-01
6.21137023e-01 -2.20153287e-01 2.61710078e-01 4.51452494e-01
-7.32849479e-01 8.05886090e-01 -1.34836853e+00 2.99774289e-01
5.45839489e-01 1.83854505e-01 -5.06084323e-01 -2.14417487e-01
-1.09643996e+00 9.43277478e-01 4.63046968e-01 2.03424558e-01
-8.31972003e-01 -1.23338306e+00 -7.70370185e-01 -5.21664143e-01
4.02431548e-01 -8.67982328e-01 8.56806874e-01 -4.47618663e-01
-1.24271345e+00 1.06161821e+00 -5.81534244e-02 -2.50782162e-01
7.33002424e-01 1.33184269e-02 -2.92237431e-01 8.61729085e-01
4.60627526e-01 2.71686435e-01 9.79902029e-01 -1.17401922e+00
-1.13683179e-01 -5.65318763e-01 -3.56906742e-01 3.07082117e-01
5.97865880e-01 1.34021774e-01 -3.72198164e-01 -8.48318577e-01
9.26535487e-01 -9.34135795e-01 -6.77862883e-01 -2.49035746e-01
-1.42553344e-01 8.26063931e-01 3.17926466e-01 -1.13648927e+00
1.12802529e+00 -1.84045601e+00 3.08529437e-01 1.64734915e-01
6.75387204e-01 4.54149768e-03 3.33998322e-01 -6.47086442e-01
-4.23802584e-01 5.43288924e-02 -8.02303612e-01 -1.32648185e-01
-8.22164774e-01 2.57032067e-01 1.29067168e-01 8.54940236e-01
8.35567489e-02 9.79572535e-01 -1.01605856e+00 -9.42798495e-01
4.03060496e-01 8.55702162e-01 -4.50268626e-01 3.28301311e-01
5.46878636e-01 1.69604957e+00 -6.05350733e-01 2.65604168e-01
7.07091331e-01 -5.88505626e-01 4.94571388e-01 -4.96185362e-01
-9.05606374e-02 3.59364003e-01 -1.09578073e+00 2.26168036e+00
-3.58353615e-01 -2.23053530e-01 4.69809592e-01 -8.91910136e-01
4.50956792e-01 7.51150548e-01 1.10035849e+00 -7.62752175e-01
-1.35173768e-01 1.91403449e-01 -7.36931488e-02 -3.91461849e-01
-4.66721877e-02 -7.30538189e-01 2.61275381e-01 6.00121260e-01
-6.54042214e-02 -2.81011403e-01 -1.85591668e-01 1.15158252e-01
7.55353451e-01 4.25359935e-01 3.13596129e-01 -3.68091702e-01
7.32111633e-01 -1.49159402e-01 8.86886120e-01 5.41786015e-01
-2.84882456e-01 1.20849800e+00 3.59024554e-02 -4.92188781e-01
-1.37735629e+00 -1.24958813e+00 -6.42670453e-01 3.45877230e-01
1.47246169e-02 -1.58608302e-01 -8.63096416e-01 -3.79502505e-01
-5.36365032e-01 1.87675312e-01 -4.36958641e-01 1.47807240e-01
-1.16097391e+00 -1.30606008e+00 3.02029073e-01 4.55763102e-01
3.28411758e-01 -6.69181049e-01 -1.23121500e+00 5.54811954e-01
-8.82213950e-01 -1.55051184e+00 -4.66291457e-01 4.70724069e-02
-1.61314034e+00 -9.74186599e-01 -9.76599693e-01 -3.00137043e-01
6.84669554e-01 1.52575895e-01 1.30840063e+00 -2.36746185e-02
-7.03785539e-01 1.28132313e-01 -1.16254762e-02 5.02034485e-01
-5.61610639e-01 -2.79124945e-01 -7.84049332e-02 -1.88653879e-02
-5.29184759e-01 -7.85867870e-01 -8.04587543e-01 2.70251304e-01
-8.67735744e-01 4.10785139e-01 6.10128701e-01 1.03936565e+00
1.25858021e+00 9.43784192e-02 2.10868523e-01 -1.06291437e+00
-5.96028827e-02 -4.08987820e-01 -4.74896252e-01 8.96208584e-02
-5.73956847e-01 1.61108956e-01 2.18043074e-01 -2.11559072e-01
-1.35021269e+00 1.19116075e-01 5.53675480e-02 -5.46238959e-01
-2.00995028e-01 3.77960742e-01 1.06770635e-01 -4.64205369e-02
4.95255560e-01 3.02123696e-01 1.27062142e-01 -5.15205443e-01
1.88647270e-01 -1.61737278e-02 9.15127218e-01 -5.36374032e-01
6.00732386e-01 7.81616688e-01 4.02280778e-01 -6.08267903e-01
-9.87891436e-01 -6.16648674e-01 -1.58101654e+00 -1.12288043e-01
1.31424201e+00 -8.01734269e-01 -1.18332438e-01 7.59066567e-02
-8.60420883e-01 -7.51187503e-02 -2.62859881e-01 9.17876482e-01
-6.98849618e-01 7.67207623e-01 -8.83764684e-01 -4.47032809e-01
-4.27894920e-01 -1.54296505e+00 9.80414033e-01 -1.46439761e-01
-1.93807140e-01 -1.13766885e+00 8.25616047e-02 7.11620510e-01
3.38675320e-01 6.59904301e-01 9.27189052e-01 -5.52164949e-02
-8.65677416e-01 2.89337516e-01 -1.16524756e-01 3.47919345e-01
2.31421337e-01 -6.80563450e-01 -7.78652251e-01 -2.18475580e-01
5.86504281e-01 1.05945840e-01 4.78001714e-01 1.10882592e+00
9.01813984e-01 7.02410564e-02 -4.73595709e-02 1.00790930e+00
1.35014582e+00 -2.04584241e-01 3.59334171e-01 1.81325853e-01
9.65538502e-01 5.50514281e-01 6.79046869e-01 3.40316176e-01
3.63540828e-01 7.93151140e-01 2.81512737e-01 -4.03605014e-01
-2.19036698e-01 1.46934316e-01 -3.36589992e-01 9.56528842e-01
-4.46487993e-01 8.80614281e-01 -1.16893554e+00 6.20403886e-01
-1.55867791e+00 -8.45715582e-01 -4.35277194e-01 2.36937547e+00
1.07749009e+00 -5.31669706e-02 1.09066978e-01 4.73225825e-02
8.19870710e-01 1.99458852e-01 -5.70641756e-01 3.02222371e-01
7.11461483e-03 1.58107951e-01 5.25947869e-01 8.10057402e-01
-1.14567435e+00 3.26791376e-01 6.45870209e+00 3.18417370e-01
-1.08538198e+00 7.35672057e-01 6.94897115e-01 -9.53086466e-02
-1.35072261e-01 -8.78242999e-02 -3.54016751e-01 2.27301061e-01
8.35890710e-01 -1.24677690e-02 4.08507258e-01 2.41489425e-01
3.62611204e-01 -2.03601032e-01 -8.63892436e-01 9.67202783e-01
-1.39832735e-01 -1.31808436e+00 -3.17474425e-01 9.56218317e-03
6.75377369e-01 1.49172083e-01 -1.39723551e-02 -4.27004933e-01
-2.68845469e-01 -1.14189923e+00 4.36536580e-01 5.16948879e-01
1.09143114e+00 -6.95994437e-01 5.66939831e-01 4.72862095e-01
-1.19520330e+00 2.73352742e-01 -1.48241282e-01 4.66923624e-01
6.24417186e-01 8.17217588e-01 -6.76762581e-01 9.22434688e-01
7.80527711e-01 6.11386120e-01 -2.95362175e-01 6.46734893e-01
-1.78862438e-01 3.60565335e-01 -5.12741506e-02 1.40140307e+00
7.63102323e-02 -4.95313317e-01 7.88274467e-01 9.81849551e-01
-4.91024554e-03 6.02779567e-01 3.28470767e-01 1.20336378e+00
4.28397238e-01 -8.97229165e-02 -2.11435348e-01 4.62189317e-01
1.18201576e-01 1.53366423e+00 -9.52975571e-01 -4.06843096e-01
-3.57607037e-01 9.08089280e-01 3.54111567e-02 3.22295129e-01
-7.97151387e-01 6.70627594e-01 4.50413078e-02 6.10381007e-01
5.15041575e-02 -2.01278880e-01 -5.17052054e-01 -1.26259482e+00
1.10656179e-01 -6.74641550e-01 5.87032259e-01 -5.59133232e-01
-1.01694477e+00 7.15960741e-01 2.31477618e-01 -1.06748807e+00
-4.35274005e-01 1.38639435e-01 -2.00302958e-01 1.41460085e+00
-1.69468343e+00 -1.01117873e+00 -1.34494573e-01 4.53254431e-01
3.30897242e-01 3.82833689e-01 8.18646550e-01 3.80741894e-01
-1.43777966e-01 -7.75567861e-03 -2.48554543e-01 9.15603489e-02
6.68617547e-01 -1.31070662e+00 9.29019526e-02 9.33177710e-01
-2.39354596e-01 6.49089456e-01 3.86694551e-01 -8.44645202e-01
-8.90845954e-01 -1.05226970e+00 5.59102535e-01 -5.15008271e-01
3.00557941e-01 1.92276537e-01 -1.28856754e+00 5.94630778e-01
-2.95307755e-01 7.83463538e-01 7.69962847e-01 -5.28907657e-01
5.49649335e-02 3.76039803e-01 -1.41855407e+00 7.28021935e-02
7.53166676e-01 -6.85563266e-01 -7.86410153e-01 9.28213913e-03
4.25750792e-01 -9.66370106e-01 -1.58281112e+00 6.43026531e-01
4.58871275e-01 -9.74360704e-01 1.53402948e+00 -2.80553132e-01
4.81334269e-01 -4.24840122e-01 2.43795261e-01 -9.25282538e-01
-6.02234662e-01 -5.83296239e-01 -3.34264338e-02 6.52050555e-01
-6.78593619e-03 -2.67846107e-01 6.21577680e-01 7.56603181e-01
-1.08634539e-01 -6.04373276e-01 -1.12821269e+00 -3.43156189e-01
1.43023685e-01 -3.24823380e-01 -2.56175883e-02 1.33543289e+00
-3.97195935e-01 2.39292428e-01 -2.40680948e-01 6.22254789e-01
1.49264252e+00 3.26742113e-01 -1.06561840e-01 -1.17744088e+00
-4.09876227e-01 -7.87490606e-02 -7.06015527e-02 -7.45260000e-01
-2.79941354e-02 -1.10450244e+00 -1.70338422e-01 -1.43312931e+00
5.55170000e-01 -4.46409941e-01 -5.07172406e-01 -3.19629237e-02
-3.92329782e-01 5.50598741e-01 -6.46574795e-02 6.26287937e-01
-3.40828300e-01 9.94422734e-02 1.87718046e+00 3.42686027e-01
-1.60143629e-01 -1.56197771e-01 -2.38204047e-01 9.45008099e-01
4.50363725e-01 -7.07479775e-01 -3.31064165e-01 -3.37808251e-01
-2.12338522e-01 8.78302932e-01 6.94607556e-01 -6.73632503e-01
6.50629774e-02 2.10916065e-02 6.09825492e-01 -6.19571388e-01
2.74172366e-01 -8.10231626e-01 3.84818852e-01 4.04977888e-01
-3.83173019e-01 -7.86523372e-02 -8.85327980e-02 3.23381603e-01
-2.12018654e-01 -1.41994953e-01 1.41895092e+00 -7.01584280e-01
-7.20150769e-02 6.09316409e-01 -1.20945431e-01 4.20888692e-01
6.26417041e-01 -3.29421014e-01 4.38232958e-01 1.72781646e-01
-1.38516760e+00 -4.14259508e-02 4.02668029e-01 1.00024164e-01
7.57140279e-01 -1.21306205e+00 -9.10240471e-01 1.01114497e-01
-3.50427032e-01 5.94900429e-01 8.14316094e-01 1.65920591e+00
-4.39880669e-01 4.33218360e-01 -3.36622477e-01 -1.16668510e+00
-1.09966195e+00 4.07364547e-01 8.33335459e-01 -7.72424519e-01
-1.30627394e+00 5.39237678e-01 6.48633480e-01 -3.53385627e-01
-3.85970533e-01 -2.09811598e-01 -2.03114718e-01 -2.53573418e-01
6.13308370e-01 1.96245447e-01 6.79845139e-02 -1.25994837e+00
-5.95976472e-01 9.40171421e-01 4.37491313e-02 -2.21199289e-01
1.37321150e+00 -7.26627886e-01 -1.49877027e-01 4.74573076e-01
9.34923828e-01 -2.58102864e-01 -1.57930005e+00 -5.08533299e-01
4.06180620e-02 -5.45244455e-01 6.47236228e-01 -6.62772059e-01
-1.14839792e+00 1.07738972e+00 7.04027951e-01 -4.07496423e-01
1.10426235e+00 -7.06502125e-02 7.09091067e-01 -6.09978080e-01
3.73524308e-01 -7.30962753e-01 -1.04985945e-01 5.80054261e-02
7.40505636e-01 -1.21141136e+00 3.31210434e-01 -6.93765938e-01
-9.50633943e-01 9.07961071e-01 1.82121173e-01 8.88427496e-02
4.31698203e-01 4.36993182e-01 9.74824354e-02 -2.54265010e-01
-3.92524153e-01 1.80082679e-01 4.76342261e-01 6.13963485e-01
6.25237167e-01 -3.47261056e-02 -6.20125309e-02 4.25373703e-01
1.82823107e-01 -5.44190779e-02 4.48264956e-01 7.64488697e-01
-1.13917172e-01 -8.80019784e-01 -5.98634958e-01 2.75758188e-02
-1.11660576e+00 -2.28701502e-01 5.11307538e-01 2.94293404e-01
8.15856736e-03 3.78848851e-01 -1.09567404e-01 5.51397085e-01
7.24299252e-02 9.69703346e-02 8.10501695e-01 -7.65737951e-01
-4.97852921e-01 7.12226033e-01 -2.55363196e-01 -7.84886897e-01
-6.69433773e-01 -7.33622491e-01 -1.70353770e+00 3.41360390e-01
-1.21305339e-01 3.35040577e-02 5.71619630e-01 9.15076077e-01
1.25448763e-01 8.93393219e-01 5.75785995e-01 -8.93547773e-01
-3.45451415e-01 -6.17819965e-01 -6.92283750e-01 5.13346672e-01
5.37585676e-01 -5.62110245e-01 -1.03686377e-01 4.25645977e-01] | [13.883010864257812, -2.4060428142547607] |
57569fde-517c-4fc1-b546-9de526aeb333 | a-tutorial-introduction-to-reinforcement | 2304.00803 | null | https://arxiv.org/abs/2304.00803v1 | https://arxiv.org/pdf/2304.00803v1.pdf | A Tutorial Introduction to Reinforcement Learning | In this paper, we present a brief survey of Reinforcement Learning (RL), with particular emphasis on Stochastic Approximation (SA) as a unifying theme. The scope of the paper includes Markov Reward Processes, Markov Decision Processes, Stochastic Approximation algorithms, and widely used algorithms such as Temporal Difference Learning and $Q$-learning. | ['Mathukumalli Vidyasagar'] | 2023-04-03 | null | null | null | null | ['q-learning'] | ['methodology'] | [-2.30553553e-01 1.73140168e-01 -4.58635509e-01 -1.09013841e-01
-6.96303070e-01 -4.09392715e-01 3.32536846e-01 -1.18399680e-01
-7.89468586e-01 1.46249485e+00 -1.54157847e-01 -4.83656943e-01
-3.52134496e-01 -5.21421790e-01 -3.94110709e-01 -7.77131498e-01
-6.18267536e-01 4.76073384e-01 1.22548975e-01 -3.33697200e-01
5.48074305e-01 5.90385079e-01 -1.35682428e+00 -1.95246264e-01
8.91770184e-01 9.96295154e-01 1.50598168e-01 1.01557517e+00
-5.20302832e-01 1.28391767e+00 -9.77546930e-01 -4.14688796e-01
2.30654776e-01 -7.46859968e-01 -9.54913735e-01 -2.27321908e-01
-5.72077215e-01 -5.49262464e-01 -7.18061805e-01 8.99358809e-01
5.69911122e-01 7.23999381e-01 9.57894027e-01 -1.41179168e+00
-2.71924585e-01 8.31117511e-01 -3.62376004e-01 5.58392048e-01
6.28905475e-01 7.15792179e-01 7.29418218e-01 -5.04864216e-01
2.41404191e-01 1.48505783e+00 5.33794522e-01 9.58032727e-01
-1.20467865e+00 -3.12950581e-01 5.52295923e-01 3.20097297e-01
-9.43306506e-01 3.88842672e-02 6.19249046e-01 -3.38930748e-02
1.26848817e+00 -1.81925938e-01 1.05138850e+00 9.25223827e-01
7.65837729e-01 1.50466359e+00 1.63691556e+00 -5.77028453e-01
8.79957378e-01 -2.97692895e-01 -1.69333935e-01 5.73008418e-01
-1.73471034e-01 8.87242138e-01 -3.78176421e-01 -4.32874262e-01
1.04034877e+00 -2.18127798e-02 3.37796271e-01 -4.05820340e-01
-7.21253574e-01 1.07736611e+00 -1.22998826e-01 -9.96690430e-03
-7.72856116e-01 7.70072162e-01 3.60222012e-01 6.20450675e-01
-2.37728050e-03 2.64333427e-01 -3.41846615e-01 -7.69498825e-01
-7.63987124e-01 9.17069554e-01 9.09616351e-01 7.65997708e-01
5.26674807e-01 9.71855164e-01 -5.78391373e-01 5.10477185e-01
1.47194445e-01 1.03206968e+00 6.56036854e-01 -1.52234316e+00
9.23301056e-02 -4.99207497e-01 6.64933145e-01 -5.91515154e-02
-2.72586942e-01 1.46636799e-01 -4.87217069e-01 5.84124506e-01
6.94602206e-02 -7.28294373e-01 -6.21705770e-01 1.56011891e+00
-5.76051883e-02 9.60075706e-02 4.52010155e-01 3.65997583e-01
2.44904175e-01 7.88074017e-01 2.34068826e-01 -1.00822365e+00
4.12364781e-01 -9.17008340e-01 -1.25801599e+00 2.31164277e-01
-6.43232092e-03 -5.38951159e-01 7.65705645e-01 5.65450907e-01
-1.61736047e+00 -5.41579723e-01 -6.76889181e-01 8.09196353e-01
-1.10811964e-02 -7.65582025e-01 9.10109699e-01 6.49943054e-01
-1.32157266e+00 1.21326768e+00 -9.64495242e-01 -1.54240042e-01
1.24858409e-01 4.44762558e-01 7.62225628e-01 -6.10214695e-02
-1.40475512e+00 1.09439766e+00 2.71419704e-01 -2.88040191e-01
-1.44493961e+00 -1.24333099e-01 -6.34397149e-01 -4.55329210e-01
4.07720834e-01 -3.06671292e-01 2.27484536e+00 -7.85389185e-01
-2.61847496e+00 2.32825115e-01 -5.86460531e-02 -9.14985418e-01
6.63775086e-01 -1.73368379e-01 -5.61090946e-01 3.76732707e-01
-4.42965776e-01 5.00303745e-01 1.03864622e+00 -9.77334440e-01
-1.03853118e+00 6.19240664e-02 -5.42959012e-02 4.08676982e-01
3.96894127e-01 2.99835175e-01 5.32561660e-01 -5.64562380e-01
-4.70128238e-01 -7.41261065e-01 -9.33216989e-01 -4.85412270e-01
2.76309580e-01 -8.60547721e-01 4.32674646e-01 -3.03241521e-01
1.49454701e+00 -1.68460560e+00 9.29210633e-02 1.19496174e-01
-2.19025269e-01 1.78880438e-01 -1.24000676e-01 9.91918445e-01
8.58090073e-02 -2.75051981e-01 -3.25558722e-01 9.97389853e-03
2.28110313e-01 7.06526756e-01 -6.68077290e-01 2.16613203e-01
2.09766179e-02 1.10482633e+00 -1.35646963e+00 -4.68133211e-01
2.14817896e-01 -2.60634571e-01 -1.42635316e-01 5.30343235e-01
-5.55476367e-01 1.70005217e-01 -4.89856273e-01 7.35217929e-01
1.55388489e-01 3.54646474e-01 1.54935554e-01 8.73059094e-01
-4.33232546e-01 1.23861201e-01 -1.34843671e+00 1.34431529e+00
-3.69722128e-01 2.32724085e-01 -2.39173695e-01 -1.07862067e+00
9.34544146e-01 6.33337736e-01 8.73244226e-01 -6.31056428e-01
-8.22999887e-03 3.85335565e-01 -1.34461924e-01 -3.56733084e-01
5.88564396e-01 -5.93204141e-01 -1.24742366e-01 9.60780859e-01
2.23988563e-01 -6.29767478e-01 3.14330369e-01 -8.25192779e-02
1.02427936e+00 4.02728349e-01 6.11929655e-01 -2.22394288e-01
3.00877392e-01 2.45356597e-02 4.10510957e-01 1.21690476e+00
-9.32191014e-01 -3.17845672e-01 4.85549986e-01 -6.53240561e-01
-6.14369452e-01 -1.48402989e+00 4.20284748e-01 1.07928979e+00
1.58968382e-02 1.37496488e-02 -4.68775570e-01 -8.59047830e-01
2.34677836e-01 9.85015094e-01 -5.82286239e-01 -2.16125458e-01
-6.28728449e-01 -6.60280168e-01 2.97999769e-01 6.76505268e-01
3.84891331e-01 -1.80808306e+00 -7.77243555e-01 7.24197388e-01
-1.64486244e-02 -2.61829644e-01 -1.15762845e-01 5.22638500e-01
-1.35875106e+00 -5.79423904e-01 -9.83645737e-01 -4.89225149e-01
3.79171520e-02 8.06063935e-02 1.26162851e+00 -5.88897288e-01
3.53224277e-02 1.00920010e+00 -2.33090997e-01 -6.05900347e-01
-3.88220310e-01 -5.26821613e-01 4.91142154e-01 -4.44097489e-01
2.52923965e-01 -3.94214958e-01 -6.68604970e-01 4.39327471e-02
-6.83307171e-01 -9.22138691e-01 3.87520909e-01 1.03724992e+00
8.42839181e-01 2.00356543e-01 4.81955975e-01 -4.39468026e-01
1.28627110e+00 -1.67545989e-01 -6.72328413e-01 1.04402721e-01
-9.72697675e-01 2.90761977e-01 6.93656623e-01 -6.54923618e-01
-1.29718935e+00 -3.62981528e-01 -2.10990846e-01 -3.39228094e-01
6.16929121e-02 1.80287138e-01 4.97104883e-01 2.07622070e-02
7.01043427e-01 7.22650409e-01 2.51672477e-01 -1.48357302e-01
4.94900018e-01 4.59002942e-01 -4.38785776e-02 -8.11936677e-01
3.37018341e-01 3.41017038e-01 3.65840383e-02 -4.34116155e-01
-5.07335007e-01 -6.91897348e-02 -1.84357509e-01 -4.72159445e-01
4.44711506e-01 -3.74589562e-01 -1.10702443e+00 3.96282464e-01
-7.29016125e-01 -1.03560972e+00 -1.18867147e+00 6.22442842e-01
-1.64273345e+00 1.23431034e-01 -1.06604218e+00 -1.61731136e+00
-2.77586758e-01 -1.05005753e+00 2.89532602e-01 5.62947750e-01
2.10707206e-02 -1.11256313e+00 6.92847669e-01 -4.53546405e-01
2.25393876e-01 4.94644493e-02 6.66613281e-01 -3.57692897e-01
-2.23191172e-01 6.68363795e-02 5.30756414e-01 4.74786311e-01
-1.71203867e-01 1.99221447e-01 -4.87493634e-01 -4.63516474e-01
-2.03815103e-02 -7.57478952e-01 5.51114559e-01 8.89823198e-01
9.92439508e-01 -1.32281303e-01 -5.89301288e-02 1.13922894e-01
1.56895804e+00 9.97813523e-01 3.61906677e-01 2.27074370e-01
-6.46775067e-02 3.99163753e-01 1.04597485e+00 1.00641680e+00
1.51036426e-01 -1.38737159e-02 3.74682099e-01 4.10705835e-01
4.50572908e-01 -3.20566297e-01 6.93952203e-01 7.07791865e-01
-4.26694542e-01 -4.23754789e-02 -7.28615046e-01 1.63859427e-01
-1.99437177e+00 -1.57481194e+00 4.88961041e-01 2.06745124e+00
8.27008545e-01 1.54397890e-01 8.17818820e-01 2.47765675e-01
6.58587515e-01 2.01389045e-01 -9.16926146e-01 -1.23680174e+00
-1.15523459e-02 7.11266220e-01 5.37019432e-01 6.92708135e-01
-7.18059897e-01 1.15472567e+00 9.40705872e+00 1.11017203e+00
-6.42934859e-01 -1.67012498e-01 5.51906645e-01 -1.17071040e-01
-1.89730674e-01 -1.77234709e-01 -7.69647956e-01 3.63088340e-01
1.38055170e+00 -3.12932521e-01 8.85889828e-01 1.03040111e+00
5.19533634e-01 -3.76187265e-01 -7.60048270e-01 9.88903940e-01
-4.06923354e-01 -1.02817261e+00 -3.03456723e-03 -1.74099788e-01
9.80695844e-01 -6.56427592e-02 3.57320338e-01 1.17758894e+00
1.36241484e+00 -9.82143879e-01 6.02287054e-01 5.96840262e-01
6.00301921e-01 -1.47979581e+00 5.55497944e-01 2.02093720e-01
-9.31528926e-01 -5.38381040e-01 -4.61495787e-01 -4.47751254e-01
2.73887753e-01 2.47595422e-02 -2.21937507e-01 3.57209593e-01
7.04613388e-01 4.79343235e-01 2.84661919e-01 8.72538507e-01
-4.55555528e-01 6.05046749e-01 -1.49783874e-02 -7.86887825e-01
8.91671479e-01 -3.92426729e-01 3.47224891e-01 1.08475578e+00
8.49937722e-02 3.96722794e-01 5.75518310e-01 3.68080318e-01
5.57070017e-01 -4.87691797e-02 -5.20398080e-01 -1.29505023e-01
2.35661849e-01 5.62586069e-01 -5.43678820e-01 -5.01297116e-01
-3.13110530e-01 1.03835320e+00 2.42002353e-01 5.74733853e-01
-8.23166370e-01 -5.96043050e-01 8.21536303e-01 -4.85682547e-01
3.99576694e-01 -2.65640408e-01 1.64536119e-01 -6.74823225e-01
-6.40613794e-01 -8.43076468e-01 5.26269913e-01 -3.60143036e-01
-1.19236660e+00 4.77732599e-01 1.34201050e-01 -1.20010459e+00
-1.07725477e+00 -5.06987631e-01 -4.20923471e-01 8.68694067e-01
-1.63228202e+00 -1.91792157e-02 5.84848940e-01 7.87678123e-01
8.00153852e-01 -2.76222467e-01 7.35009730e-01 -5.38392544e-01
-3.72891724e-01 4.06450003e-01 8.18604946e-01 -3.80110443e-01
2.51478195e-01 -1.72861910e+00 3.20043445e-01 3.28636795e-01
-1.06307715e-01 1.09372683e-01 9.69300210e-01 -3.59222472e-01
-1.44744241e+00 -6.68211639e-01 2.71324784e-01 -3.77836078e-02
7.97522366e-01 2.85652608e-01 -4.59083527e-01 3.84549022e-01
3.14204514e-01 -2.03657284e-01 6.04584396e-01 -3.15113127e-01
3.38969111e-01 -4.02420834e-02 -1.31606317e+00 8.66345286e-01
9.43318546e-01 -1.79469079e-01 -5.29053926e-01 2.30305418e-01
6.94706738e-01 -3.37458909e-01 -7.62797058e-01 2.00504467e-01
4.83857393e-01 -9.31455553e-01 8.43818009e-01 -1.04536164e+00
1.11552112e-01 2.15687484e-01 -5.51698022e-02 -1.72436929e+00
-5.17855406e-01 -1.46544158e+00 -8.78117442e-01 4.48378623e-01
7.26110339e-02 -5.93974292e-01 8.29989076e-01 3.37291211e-01
1.70715023e-02 -1.10826004e+00 -1.05259418e+00 -1.42756784e+00
4.57484692e-01 -3.83257061e-01 4.01221395e-01 2.37525582e-01
4.82340306e-01 -1.89984202e-01 -2.87879646e-01 -5.33521950e-01
7.87783921e-01 3.33302110e-01 1.44501805e-01 -6.21586323e-01
-6.38903439e-01 -8.03957880e-01 2.16386184e-01 -1.39554989e+00
2.60266930e-01 -2.78986901e-01 5.43277442e-01 -1.54233658e+00
-2.64508069e-01 -1.88704029e-01 -8.66464555e-01 -1.97614700e-01
-1.51858628e-01 -3.80798906e-01 1.40268207e-01 2.25289389e-02
-1.01330614e+00 1.10869503e+00 1.34997201e+00 7.11942986e-02
-5.65362573e-01 8.24347913e-01 -1.34946838e-01 6.57441616e-01
1.23805153e+00 -4.87578839e-01 -5.35607576e-01 4.06950712e-02
-1.86172456e-01 8.73277187e-01 -2.46760398e-01 -7.06231415e-01
-1.32430121e-01 -8.76256287e-01 1.98494181e-01 -6.79870784e-01
2.79512793e-01 -3.51532161e-01 -3.95284712e-01 1.28766608e+00
-5.35262048e-01 7.37989902e-01 5.07635362e-02 9.27677810e-01
-2.04056650e-01 -5.96048415e-01 9.02777612e-01 -6.41190469e-01
-8.04977655e-01 6.15529537e-01 -1.35394979e+00 4.25794095e-01
1.10577953e+00 -1.84339345e-01 4.14359033e-01 -6.42711282e-01
-9.35479879e-01 4.18033838e-01 -1.34142280e-01 -1.09324835e-01
9.99766946e-01 -1.19890881e+00 -4.06351179e-01 -4.71362174e-01
-3.30757648e-01 -6.44771993e-01 1.65772274e-01 3.86098713e-01
-2.14667603e-01 5.70310831e-01 -3.59496355e-01 -1.36737287e-01
-6.39126003e-01 8.70036066e-01 4.65914100e-01 -7.50463128e-01
-2.79880762e-01 8.04784298e-01 -5.86919248e-01 -1.39559731e-01
6.30449057e-01 -2.24611148e-01 -2.09329173e-01 -2.53096640e-01
6.57871604e-01 9.28858221e-01 -6.68618619e-01 1.83672965e-01
6.92001134e-02 8.32207873e-02 1.20426781e-01 -1.01196897e+00
9.73281980e-01 -2.30650648e-01 1.45828038e-01 1.11432672e+00
4.53052342e-01 -8.78436804e-01 -1.88675416e+00 -4.87802386e-01
1.92512929e-01 -1.38009906e-01 -2.65056014e-01 -6.70151293e-01
-5.00095844e-01 8.89140606e-01 7.90211558e-01 2.02093884e-01
8.72000337e-01 -2.78650284e-01 6.05873346e-01 7.24707305e-01
8.70584130e-01 -1.89172852e+00 5.48227191e-01 8.56046557e-01
6.55583084e-01 -9.03490365e-01 -1.44812480e-01 4.20337051e-01
-9.94532883e-01 1.08886814e+00 5.22049963e-01 -6.21271431e-01
8.86744261e-01 1.14277408e-01 -1.49866655e-01 3.45794439e-01
-1.03261590e+00 -5.73134542e-01 -4.99715865e-01 1.19208086e+00
1.09775521e-01 2.37223402e-01 -3.44636291e-01 2.24678814e-01
1.47094026e-01 1.66761011e-01 5.86385190e-01 1.49128699e+00
-7.80391812e-01 -1.11214828e+00 -3.80186856e-01 2.92638272e-01
-6.31086051e-01 -1.09067090e-01 1.41788512e-01 5.49389243e-01
-4.69879657e-01 1.04531276e+00 1.11439392e-01 -2.29229450e-01
3.51714581e-01 1.61305070e-01 1.09046483e+00 -3.67136717e-01
-6.01547718e-01 -1.65661454e-01 -2.54119486e-01 -7.34713376e-01
-3.94795239e-01 -1.06867874e+00 -1.31620347e+00 -5.15773714e-01
2.41705254e-01 5.46374619e-01 4.50378954e-01 7.98728287e-01
-2.29911998e-01 4.15854335e-01 1.03906345e+00 -6.51549160e-01
-1.50101054e+00 -7.95875549e-01 -1.17368853e+00 -2.94139236e-01
1.53950274e-01 -6.63273394e-01 -2.50260383e-01 -2.51780391e-01] | [4.173821926116943, 2.3896148204803467] |
0714be03-3aff-45fe-a7dc-31bff484d1a7 | understanding-and-exploring-the-whole-set-of | 2303.16047 | null | https://arxiv.org/abs/2303.16047v1 | https://arxiv.org/pdf/2303.16047v1.pdf | Understanding and Exploring the Whole Set of Good Sparse Generalized Additive Models | In real applications, interaction between machine learning model and domain experts is critical; however, the classical machine learning paradigm that usually produces only a single model does not facilitate such interaction. Approximating and exploring the Rashomon set, i.e., the set of all near-optimal models, addresses this practical challenge by providing the user with a searchable space containing a diverse set of models from which domain experts can choose. We present a technique to efficiently and accurately approximate the Rashomon set of sparse, generalized additive models (GAMs). We present algorithms to approximate the Rashomon set of GAMs with ellipsoids for fixed support sets and use these ellipsoids to approximate Rashomon sets for many different support sets. The approximated Rashomon set serves as a cornerstone to solve practical challenges such as (1) studying the variable importance for the model class; (2) finding models under user-specified constraints (monotonicity, direct editing); (3) investigating sudden changes in the shape functions. Experiments demonstrate the fidelity of the approximated Rashomon set and its effectiveness in solving practical challenges. | ['Cynthia Rudin', 'Margo Seltzer', 'Chudi Zhong', 'Zhi Chen'] | 2023-03-28 | null | null | null | null | ['additive-models'] | ['methodology'] | [ 2.54648566e-01 1.03843279e-01 -1.92697048e-01 -2.20855370e-01
-9.96046901e-01 -7.44995415e-01 2.08772078e-01 -7.44484887e-02
2.62206830e-02 7.87886500e-01 -3.52355987e-01 -2.98259079e-01
-6.09403968e-01 -4.60352927e-01 -8.04707050e-01 -5.40984213e-01
-1.69128075e-01 1.16262126e+00 -7.71628916e-02 -1.78759377e-02
5.75460196e-01 4.86389995e-01 -1.64479113e+00 2.18750447e-01
1.50766397e+00 1.00568604e+00 3.56155783e-01 5.37037730e-01
-1.06232390e-01 1.46901339e-01 -5.13094544e-01 -1.86982378e-01
5.47861874e-01 -3.32556456e-01 -5.57328522e-01 2.84530520e-01
3.02735150e-01 -2.28940845e-01 3.97975206e-01 1.17980635e+00
1.23918697e-01 4.01097298e-01 1.06190169e+00 -1.89763200e+00
-6.31214142e-01 4.50154126e-01 -8.38210821e-01 -1.60390601e-01
3.91716063e-01 -5.04005812e-02 7.22726941e-01 -1.11466289e+00
3.68568063e-01 1.37222636e+00 9.39232528e-01 2.56593019e-01
-1.49470198e+00 -5.44086933e-01 3.04508656e-01 1.15915567e-01
-1.88350534e+00 -3.34487349e-01 6.24522746e-01 -5.85281014e-01
6.72173500e-01 6.46420896e-01 4.28194851e-01 5.47315478e-01
-1.52897179e-01 5.76694548e-01 9.01471138e-01 -4.83206302e-01
6.98435247e-01 5.82881808e-01 6.89625219e-02 5.23996890e-01
2.97979027e-01 -1.88787028e-01 -3.77772748e-01 -1.01842248e+00
8.51844251e-01 -7.60420710e-02 -4.37814087e-01 -6.65561020e-01
-7.82264352e-01 8.90438318e-01 -6.18868619e-02 -2.42669150e-01
-5.08819580e-01 -9.35308188e-02 -9.91794467e-02 3.31405282e-01
3.54871303e-01 8.67012322e-01 -5.91294169e-01 -9.68124494e-02
-1.04281902e+00 4.85136479e-01 1.11143255e+00 1.19519639e+00
5.19179344e-01 4.81545441e-02 3.31221640e-01 1.03847122e+00
3.39133412e-01 4.09328073e-01 -1.00675225e-01 -1.09669435e+00
3.39610070e-01 6.11381352e-01 7.87416458e-01 -1.11574244e+00
-2.36214027e-01 -1.36719346e-01 -7.90857732e-01 2.23829925e-01
7.15897083e-01 -1.46070719e-01 -6.76404297e-01 1.75954139e+00
7.70266891e-01 2.31034458e-01 -2.36678585e-01 8.35954249e-01
2.00710520e-01 8.19235384e-01 -1.02081522e-01 -3.81038219e-01
8.70887101e-01 -6.49530768e-01 -2.30772987e-01 -3.42337340e-01
5.35631001e-01 -6.68664217e-01 1.12168872e+00 7.02101827e-01
-1.29699123e+00 -2.37006634e-01 -9.66039598e-01 1.63148329e-01
-9.60500166e-02 -2.98820883e-02 4.72094327e-01 4.77947921e-01
-1.07991707e+00 7.01033354e-01 -6.47375047e-01 1.05689920e-01
2.55396456e-01 5.08266866e-01 -1.89124614e-01 -3.83686423e-01
-1.03783739e+00 1.13333023e+00 2.09042922e-01 5.11330739e-02
-4.43802536e-01 -1.19433093e+00 -9.17397678e-01 1.72341183e-01
6.80369735e-01 -7.13220477e-01 1.15463042e+00 -1.18395829e+00
-1.01109302e+00 4.77008134e-01 -3.88234526e-01 -2.89975524e-01
6.91482127e-01 8.31513181e-02 -1.18697621e-01 -1.73827499e-01
-1.55323192e-01 3.13768297e-01 1.05569363e+00 -1.44528794e+00
-4.20675248e-01 -2.11570799e-01 -1.30494773e-01 3.30558002e-01
-1.25789300e-01 5.97922541e-02 -2.34958410e-01 -5.50316393e-01
1.95349529e-01 -9.41225886e-01 -5.73950291e-01 4.30592090e-01
-3.54668409e-01 -2.15297639e-01 5.49112797e-01 -7.84439027e-01
1.61816263e+00 -1.93565834e+00 3.50161314e-01 8.18275273e-01
1.51335359e-01 1.74630046e-01 -1.42017275e-01 4.50827807e-01
-1.67282835e-01 1.90754682e-01 -6.61797345e-01 -3.45364958e-01
2.14994445e-01 1.24636382e-01 -3.29720199e-01 3.53018016e-01
1.49148285e-01 2.94332504e-01 -8.54932547e-01 -3.83251995e-01
5.82237840e-02 3.12875062e-01 -7.98976123e-01 1.16171084e-01
-2.73212761e-01 -2.28575319e-02 -4.30252671e-01 7.07243085e-01
1.07342267e+00 -4.19174969e-01 2.88554281e-01 1.02301784e-01
2.74294347e-01 -1.90530896e-01 -2.02861762e+00 9.12640631e-01
-5.69803357e-01 1.67416528e-01 4.78155136e-01 -1.00461507e+00
1.07163358e+00 -9.71974358e-02 4.13751811e-01 1.18412068e-02
-2.49234363e-01 3.18249851e-01 -2.24659815e-01 -3.01015079e-02
2.83388525e-01 -6.94813728e-02 -6.02757372e-02 7.53407657e-01
-3.69647026e-01 -2.93360114e-01 1.74981520e-01 2.47443337e-02
7.91796446e-01 -1.03771277e-01 7.58263230e-01 -6.07865334e-01
3.72902378e-02 1.76533818e-01 5.10732293e-01 8.38415861e-01
2.14248583e-01 7.60520220e-01 5.49598157e-01 -5.18974483e-01
-1.03331780e+00 -1.00637877e+00 -2.16754377e-01 1.05300975e+00
-1.64531544e-02 -1.49929047e-01 -7.87716269e-01 -3.45240057e-01
4.09351856e-01 1.09530866e+00 -5.43808401e-01 -2.01400146e-01
-3.49511594e-01 -5.93152761e-01 -9.68538150e-02 5.24937510e-01
-9.05649960e-02 -5.53231955e-01 -5.46420038e-01 1.42174914e-01
-7.95373470e-02 -5.83131671e-01 -8.82365584e-01 1.86881110e-01
-7.98395455e-01 -1.19411802e+00 -7.10057020e-01 -6.35409117e-01
1.05134833e+00 2.55430639e-01 1.40909374e+00 6.11393228e-02
-7.52773210e-02 2.96906322e-01 1.14626043e-01 -5.20220041e-01
-5.79496443e-01 -4.42764133e-01 1.96333095e-01 -1.04569659e-01
1.83869317e-01 -6.12809956e-01 -1.77151784e-01 5.95338464e-01
-7.48000026e-01 2.51289904e-01 2.35001922e-01 7.97834992e-01
8.74117613e-01 2.21484244e-01 8.10051739e-01 -1.01531589e+00
8.47995579e-01 -8.70149076e-01 -9.64508593e-01 6.11015797e-01
-6.21549904e-01 3.96327637e-02 8.81967366e-01 -9.42616224e-01
-7.65861034e-01 1.79006770e-01 5.28011799e-01 -6.08830690e-01
5.43285348e-02 7.05623150e-01 -2.65094757e-01 -9.23119411e-02
9.38968062e-01 4.51270901e-02 1.22926705e-01 -4.07626510e-01
3.35444063e-01 7.31781542e-01 3.97825748e-01 -7.90156901e-01
6.11070335e-01 4.05497141e-02 2.09557623e-01 -7.18017340e-01
-5.40633202e-01 -4.25624609e-01 -5.49895883e-01 8.56197849e-02
-1.52393701e-02 -6.13439560e-01 -4.21932489e-01 2.06187014e-02
-1.07677126e+00 -3.55903208e-01 -2.63764679e-01 5.83485998e-02
-7.33051240e-01 3.85466963e-01 1.19205944e-01 -1.13189220e+00
-2.86060452e-01 -9.71804142e-01 9.98358309e-01 -3.93613353e-02
-7.15315163e-01 -1.09067595e+00 -1.27632126e-01 1.12099715e-01
3.40987504e-01 5.26526928e-01 1.20359099e+00 -7.48529375e-01
-3.43503654e-01 -2.07088858e-01 1.93342529e-02 1.31700397e-01
1.27729878e-01 9.84162316e-02 -4.81880635e-01 -2.90780008e-01
-8.62762183e-02 -7.16971159e-02 2.72247136e-01 8.87761950e-01
1.35562289e+00 -6.23113692e-01 -4.69055384e-01 5.41934073e-01
1.14652073e+00 5.10711432e-01 3.86546195e-01 -2.60984134e-02
2.53439248e-01 5.91743946e-01 7.63465762e-01 8.53383958e-01
2.39943087e-01 6.64351523e-01 2.26689354e-01 -1.00202253e-02
5.80731273e-01 -8.49784166e-02 -7.45130563e-03 2.70646751e-01
1.82697237e-01 -1.77186891e-01 -1.16704679e+00 5.49367428e-01
-2.09686732e+00 -7.19647825e-01 1.78925470e-01 2.61229587e+00
9.89647865e-01 -1.30716607e-01 3.06306750e-01 -5.82923628e-02
8.81191611e-01 -4.16394234e-01 -9.23370600e-01 -6.69777811e-01
3.82767051e-01 -1.19366214e-01 1.46191880e-01 5.08070588e-01
-1.03562343e+00 1.86540991e-01 7.00184345e+00 9.48110521e-01
-6.48615897e-01 -3.82570654e-01 7.37164795e-01 -1.90612197e-01
-4.83584553e-01 -5.73136620e-02 -6.31856263e-01 6.76827908e-01
7.97555983e-01 -7.79620469e-01 9.46516216e-01 1.30014467e+00
2.50215143e-01 -8.39080438e-02 -1.54617727e+00 1.06627250e+00
3.45731452e-02 -1.22314227e+00 -1.24790706e-02 -5.85955344e-02
1.02291226e+00 -5.63861310e-01 2.67849654e-01 2.11127684e-01
4.65515107e-01 -1.37939203e+00 4.99964118e-01 7.13690162e-01
9.98094618e-01 -1.04025054e+00 3.20936590e-01 6.77723110e-01
-1.16827285e+00 -1.27764657e-01 -5.21407068e-01 1.38793617e-01
5.02279364e-02 4.61521000e-01 -9.44155395e-01 2.17781812e-01
4.30301279e-01 2.72913814e-01 -7.92086869e-02 1.26825988e+00
5.72407961e-01 3.14987838e-01 -9.73270655e-01 1.39978845e-02
8.92617702e-02 -3.83109450e-01 5.65340400e-01 1.04884887e+00
6.32252455e-01 2.63977051e-01 3.09502721e-01 1.31612444e+00
1.76567569e-01 -5.62009076e-03 -5.71315765e-01 -9.40900221e-02
9.77415383e-01 7.34821141e-01 -5.64919293e-01 -1.30297512e-01
-1.72817647e-01 5.61234534e-01 1.71699211e-01 5.45538425e-01
-8.17085445e-01 -3.58126640e-01 8.91960680e-01 3.38562220e-01
1.32672727e-01 -6.62576184e-02 -5.79613864e-01 -7.47794092e-01
-5.73342033e-02 -1.40135229e+00 6.54796124e-01 -8.88526261e-01
-1.32603383e+00 3.76911312e-01 5.60143411e-01 -1.31231666e+00
-5.85935354e-01 -3.16136181e-01 -5.88118970e-01 1.26971126e+00
-7.38400400e-01 -7.63103664e-01 -2.60054231e-01 4.69020486e-01
5.63560545e-01 -4.07739319e-02 7.97784865e-01 -1.24594942e-01
-1.91841051e-01 5.86267531e-01 4.90211248e-01 -8.75692546e-01
4.40544784e-01 -1.33431208e+00 3.40322316e-01 5.90475261e-01
-4.21028910e-03 9.75082338e-01 1.02103555e+00 -5.87723315e-01
-1.35324955e+00 -1.05652785e+00 6.06929004e-01 -5.23680925e-01
5.19838154e-01 -1.22497439e-01 -1.20557904e+00 6.60085201e-01
-4.94589657e-01 -2.10491791e-01 6.80308223e-01 7.82937258e-02
-1.34058133e-01 2.23605603e-01 -1.59121525e+00 7.86930561e-01
8.83949757e-01 -1.04964949e-01 -3.07183087e-01 4.31608200e-01
4.99144107e-01 -6.51499569e-01 -7.38942027e-01 1.67725980e-01
4.48519289e-01 -5.12191117e-01 1.14425969e+00 -9.69410121e-01
3.20015609e-01 -3.11915576e-01 -1.67337805e-01 -1.60138297e+00
-2.48781115e-01 -9.09446061e-01 -5.68607271e-01 8.27691495e-01
4.44251835e-01 -6.06357634e-01 6.69426262e-01 1.40433502e+00
5.47730960e-02 -1.28745735e+00 -9.31879878e-01 -7.06507921e-01
-6.70907870e-02 -3.89124721e-01 8.25575054e-01 9.95850086e-01
-1.18275456e-01 -3.02164018e-01 -5.12961149e-01 2.82943755e-01
6.75844252e-01 4.34492975e-01 7.48378336e-01 -1.43291843e+00
-5.63467383e-01 -3.03368896e-01 -1.42259272e-02 -9.70869482e-01
-1.76501602e-01 -6.56754613e-01 -9.70900208e-02 -1.47918308e+00
2.07158983e-01 -9.27136362e-01 -4.82757874e-02 3.67681772e-01
-1.39676183e-01 -2.47199520e-01 1.20667659e-01 1.95293929e-02
-5.53834796e-01 2.32592478e-01 9.06231999e-01 1.21665128e-01
-4.92098421e-01 3.63861710e-01 -1.24674344e+00 1.09410822e+00
5.52216768e-01 -5.42095602e-01 -7.82349825e-01 -2.19787702e-01
2.98268229e-01 1.76136106e-01 2.79109329e-01 -4.26052362e-01
2.63771594e-01 -8.05101693e-01 3.62563074e-01 -5.18554628e-01
3.59344125e-01 -1.01163578e+00 7.67741323e-01 1.22284725e-01
-3.35233063e-01 1.17845125e-01 5.08585274e-01 5.98733425e-01
9.54844952e-02 -5.96421957e-01 8.21068883e-01 2.33807769e-02
-5.23885131e-01 1.95211083e-01 -2.06175089e-01 2.47622579e-01
1.12421644e+00 -4.30551499e-01 -5.21891098e-03 -6.84884548e-01
-7.72608638e-01 5.36372006e-01 5.43711901e-01 4.66481745e-02
6.72719598e-01 -1.24477518e+00 -6.32061660e-01 3.03255409e-01
-1.91768587e-01 3.04031998e-01 2.36740142e-01 5.92858016e-01
-3.49413544e-01 -6.52848335e-04 7.36357346e-02 -4.76261497e-01
-1.45141900e+00 5.69452345e-01 4.10636574e-01 -3.04479718e-01
-1.82656258e-01 9.17630732e-01 1.82464153e-01 -3.86134446e-01
3.08426112e-01 -3.55246961e-01 1.66377991e-01 -1.04156218e-01
7.23979652e-01 7.65020490e-01 -1.69070587e-01 -1.76449329e-01
-3.40587288e-01 3.04721057e-01 1.81416851e-02 2.43904114e-01
1.56029749e+00 1.03891492e-02 -2.16552857e-02 2.13072404e-01
8.96212935e-01 -3.40084344e-01 -1.45009458e+00 -1.56452268e-01
-3.69693488e-02 -7.38943875e-01 -9.40161422e-02 -7.88726509e-01
-5.70004344e-01 4.64763254e-01 2.12433890e-01 1.99836537e-01
1.17564690e+00 -1.03375040e-01 3.60269070e-01 5.27815938e-01
5.40797889e-01 -1.23417521e+00 -1.89924166e-01 2.09863991e-01
1.43997359e+00 -1.05534446e+00 1.43556729e-01 -4.47753072e-01
-9.26609695e-01 1.02048039e+00 5.73161125e-01 -6.36460483e-02
7.84215450e-01 5.23650646e-01 -3.73033196e-01 1.73229024e-01
-9.19845283e-01 4.24610853e-01 6.16465032e-01 9.39905286e-01
-1.54571801e-01 1.54094607e-01 2.42029186e-02 1.02269566e+00
-1.79673791e-01 2.96106897e-02 3.64784509e-01 7.99593031e-01
-5.70861161e-01 -6.85624003e-01 -8.53366077e-01 8.84158731e-01
2.86605582e-02 -9.00860503e-02 -3.19050580e-01 7.57299840e-01
-7.04784840e-02 9.20390606e-01 -9.43793878e-02 -3.27590667e-02
2.59065270e-01 2.35864278e-02 3.51775259e-01 -5.85515857e-01
-8.44735727e-02 -7.74653479e-02 8.21631998e-02 -3.74655217e-01
3.56980473e-01 -6.65992975e-01 -9.50997531e-01 -4.80678946e-01
-3.20356131e-01 2.80186832e-01 3.85935575e-01 7.77624190e-01
4.80015904e-01 -2.82571688e-02 6.21418059e-01 -8.97414923e-01
-1.18253696e+00 -6.40042305e-01 -7.27326989e-01 2.38160014e-01
1.56095535e-01 -9.02976274e-01 -5.44499636e-01 9.46080405e-03] | [7.202791213989258, 4.304013252258301] |
21254564-80d8-44ae-8bad-21d60e2cb12b | effective-pseudo-labeling-based-on-heatmap | 2303.05269 | null | https://arxiv.org/abs/2303.05269v1 | https://arxiv.org/pdf/2303.05269v1.pdf | Effective Pseudo-Labeling based on Heatmap for Unsupervised Domain Adaptation in Cell Detection | Cell detection is an important task in biomedical research. Recently, deep learning methods have made it possible to improve the performance of cell detection. However, a detection network trained with training data under a specific condition (source domain) may not work well on data under other conditions (target domains), which is called the domain shift problem. In particular, cells are cultured under different conditions depending on the purpose of the research. Characteristics, e.g., the shapes and density of the cells, change depending on the conditions, and such changes may cause domain shift problems. Here, we propose an unsupervised domain adaptation method for cell detection using a pseudo-cell-position heatmap, where the cell centroid is at the peak of a Gaussian distribution in the map and selective pseudo-labeling. In the prediction result for the target domain, even if the peak location is correct, the signal distribution around the peak often has a non-Gaussian shape. The pseudo-cell-position heatmap is thus re-generated using the peak positions in the predicted heatmap to have a clear Gaussian shape. Our method selects confident pseudo-cell-position heatmaps based on uncertainty and curriculum learning. We conducted numerous experiments showing that, compared with the existing methods, our method improved detection performance under different conditions. | ['Ryoma Bise', 'Kazuhide Watanabe', 'Kazuya Nishimura', 'Hyeonwoo Cho'] | 2023-03-09 | null | null | null | null | ['cell-detection'] | ['computer-vision'] | [ 2.44815245e-01 -2.33729854e-01 6.62976876e-02 -8.36234465e-02
-5.37267327e-01 -3.64140868e-01 3.48205268e-01 5.70667744e-01
-2.61030197e-01 9.55794871e-01 -2.51115531e-01 1.20214775e-01
2.34504163e-01 -8.58754337e-01 -8.83587301e-01 -1.34016573e+00
2.76615441e-01 7.86238492e-01 5.73134959e-01 1.20813772e-01
4.87725616e-01 5.62389135e-01 -1.06783390e+00 3.66605729e-01
9.53889608e-01 6.98014498e-01 3.90469581e-01 5.58516324e-01
-2.69696563e-01 -8.00615996e-02 -8.97403777e-01 3.70862663e-01
5.18351374e-03 -3.90295714e-01 -2.93502152e-01 -1.65177494e-01
-6.85908943e-02 2.96256915e-02 -3.10775135e-02 1.38002694e+00
3.29023689e-01 -1.35669619e-01 1.24140632e+00 -9.58061457e-01
-2.73295373e-01 2.01904550e-01 -6.47720635e-01 -8.07431042e-02
4.19068746e-02 -4.77339998e-02 3.33105415e-01 -6.88278317e-01
8.98586333e-01 8.27893794e-01 6.22568965e-01 5.06437063e-01
-1.53890848e+00 -8.70817661e-01 -7.63959885e-02 -6.53839260e-02
-1.52169251e+00 1.86052807e-02 6.93331420e-01 -7.78093100e-01
2.30565637e-01 -2.81846104e-03 4.97816980e-01 9.88588572e-01
6.76367879e-01 5.57906151e-01 1.15112519e+00 -3.64302516e-01
5.41130126e-01 4.76301789e-01 -3.40701431e-01 3.52574736e-01
2.33046860e-01 1.57334446e-03 7.24326447e-02 -7.23036975e-02
8.93022597e-01 1.38778016e-01 -3.90427172e-01 -6.10433102e-01
-1.37327206e+00 6.86674118e-01 3.47501308e-01 7.58208930e-01
-1.88421775e-02 -3.50569338e-01 3.24346781e-01 -2.32276395e-01
2.56191790e-01 7.80853510e-01 -5.01797974e-01 2.98260152e-01
-9.77190316e-01 3.34460288e-01 5.59484780e-01 6.89214051e-01
7.90784419e-01 -2.63593286e-01 -4.57502663e-01 7.00380623e-01
-1.48499915e-02 3.69054437e-01 7.58386314e-01 -5.57507515e-01
-1.71359852e-01 7.86048710e-01 3.08957070e-01 -1.21126139e+00
-7.20457017e-01 -6.33746386e-01 -1.06148899e+00 2.37412855e-01
1.22656953e+00 -4.17841375e-01 -1.06333828e+00 1.64705098e+00
4.10905868e-01 3.96103263e-01 1.16072565e-01 9.38791156e-01
6.28217041e-01 9.33822989e-01 4.98179048e-02 -3.24294150e-01
1.06402659e+00 -3.94870996e-01 -5.91524363e-01 -6.84833303e-02
7.87808716e-01 -5.29497325e-01 6.95655823e-01 3.44299704e-01
-3.25547487e-01 -4.71463442e-01 -1.15043378e+00 6.04990959e-01
-5.11937201e-01 1.18625440e-01 9.24455598e-02 2.09739715e-01
-5.46730042e-01 5.83498120e-01 -5.66201627e-01 -5.82332671e-01
4.47537273e-01 9.17953998e-02 -2.90160775e-01 9.85426307e-02
-1.17616987e+00 6.24066710e-01 5.78124821e-01 -3.63395125e-01
-5.89120924e-01 -8.44416320e-01 -4.98132765e-01 1.02186799e-01
1.12317599e-01 -2.66596556e-01 7.81119704e-01 -7.33290374e-01
-1.31380010e+00 9.09258962e-01 -6.68792427e-02 -3.90804350e-01
5.05609393e-01 3.20358336e-01 -4.19118106e-01 -4.68281992e-02
1.93515092e-01 5.62415063e-01 5.93209386e-01 -1.52957165e+00
-9.64905083e-01 -2.00176179e-01 -5.23922265e-01 -5.66169620e-02
-3.42756435e-02 -2.99578965e-01 -3.30909163e-01 -5.47071934e-01
2.15639204e-01 -7.79235542e-01 -1.69187754e-01 -1.31114304e-01
-4.76222545e-01 -6.44674227e-02 8.69576693e-01 -3.33027393e-01
9.81466711e-01 -2.43150020e+00 -2.18643874e-01 4.38682288e-01
1.26817897e-01 1.37039632e-01 3.47784102e-01 1.39743254e-01
2.12096442e-02 3.09417211e-02 -3.06292981e-01 2.47260630e-01
-4.08876777e-01 -1.75623819e-01 -5.24048023e-02 5.08060694e-01
5.25066406e-02 2.99269110e-01 -1.10436058e+00 -7.40409911e-01
-2.35916506e-02 3.06616426e-01 -2.02246919e-01 1.05462275e-01
-2.95657158e-01 8.26108217e-01 -2.99906403e-01 4.87316817e-01
7.59895802e-01 -2.68720090e-01 1.45132735e-01 -3.11013401e-01
-2.22760975e-01 -3.56763393e-01 -1.03671944e+00 1.11552119e+00
-1.11252919e-03 8.15070331e-01 -1.44675881e-01 -1.22519720e+00
1.39768529e+00 1.96483389e-01 7.63695538e-01 -3.87466669e-01
8.64762366e-02 4.06716436e-01 1.95013568e-01 -1.99345514e-01
1.47311822e-01 -3.81853610e-01 -1.04274484e-03 6.52685016e-02
-2.63361156e-01 -1.23680808e-01 2.34757796e-01 -7.86653683e-02
7.83789694e-01 -1.31075114e-01 2.65817612e-01 -3.63911569e-01
4.89576995e-01 1.38085172e-01 9.48335707e-01 6.44034505e-01
-4.29877549e-01 9.35116231e-01 7.40160227e-01 -3.22876751e-01
-1.01152635e+00 -9.44642544e-01 -6.31787598e-01 7.29865372e-01
4.39190239e-01 1.87231854e-01 -8.36065829e-01 -5.65769911e-01
1.79024711e-01 5.80937147e-01 -6.57733798e-01 -2.40888625e-01
-4.59606886e-01 -9.59442616e-01 2.27699876e-01 3.08054805e-01
3.92377496e-01 -9.01939571e-01 -4.58027601e-01 4.12447006e-01
-3.07646058e-02 -8.77591431e-01 -4.47016776e-01 4.56383258e-01
-6.62275732e-01 -1.01237297e+00 -1.07875431e+00 -1.17892623e+00
9.72728252e-01 -1.16245165e-01 5.35581410e-01 -8.00748691e-02
-2.19933391e-01 -3.32877845e-01 -2.42490202e-01 -6.26788974e-01
-5.90008497e-01 -2.33107191e-02 -7.06767067e-02 3.72306295e-02
4.19781268e-01 -7.58967772e-02 -6.34508550e-01 6.25336647e-01
-6.87277317e-01 -4.68928069e-02 4.30504769e-01 1.14369631e+00
1.03171074e+00 6.33296609e-01 8.20896626e-01 -1.11422086e+00
3.84885520e-01 -3.54290038e-01 -6.56064093e-01 2.01020911e-01
-5.23338079e-01 1.03564881e-01 6.83063626e-01 -6.67302787e-01
-9.17156398e-01 3.45612109e-01 2.79859930e-01 -2.93148071e-01
-5.18928587e-01 3.81677836e-01 -1.74278677e-01 8.18640441e-02
9.59462881e-01 4.42802995e-01 2.90479869e-01 -3.12445238e-02
-4.45610255e-01 6.52194321e-01 5.61986804e-01 -2.94973254e-01
5.77255785e-01 4.71459359e-01 2.03552514e-01 -6.53063476e-01
-4.32801187e-01 -5.12377739e-01 -6.01277411e-01 -2.77188122e-01
7.69558430e-01 -5.84581435e-01 -2.54623860e-01 7.39495158e-01
-1.00720298e+00 -4.07662541e-01 4.19220515e-02 4.95644242e-01
-2.95281529e-01 2.66500652e-01 -4.63092834e-01 -6.73159719e-01
-1.45091265e-01 -1.15071917e+00 8.43781352e-01 6.32117808e-01
-2.69390494e-01 -1.10157692e+00 1.42231449e-01 -3.41325134e-01
2.20968798e-01 4.96576726e-01 1.25062525e+00 -1.04507053e+00
-2.49049887e-01 -4.99616355e-01 -1.55419484e-01 -1.67033300e-01
2.88450927e-01 3.49691749e-01 -9.61187899e-01 -2.20559373e-01
-1.83362156e-01 2.95650572e-01 7.76036680e-01 9.68845069e-01
1.26384223e+00 8.18102285e-02 -1.24767661e+00 4.71822649e-01
1.22761941e+00 7.85648823e-01 6.94068253e-01 4.11127329e-01
2.53067374e-01 5.63682616e-01 7.48612642e-01 2.92148650e-01
-2.51975179e-01 7.35103488e-01 2.11979672e-01 -2.79111892e-01
1.79895550e-01 -9.12594274e-02 -5.65498322e-02 4.33717482e-02
3.88245910e-01 -3.45017500e-02 -9.93876696e-01 5.58926523e-01
-1.77366590e+00 -7.52164125e-01 -6.71973452e-02 2.49281931e+00
1.07523894e+00 3.81031543e-01 1.90069258e-01 -2.11004633e-02
1.19272649e+00 -4.58330244e-01 -7.63822019e-01 -1.66488007e-01
-1.46905631e-01 -2.68691361e-01 5.31995058e-01 3.70643586e-01
-1.18341672e+00 6.51621819e-01 5.91619015e+00 9.14474189e-01
-1.46091855e+00 -4.96346742e-01 9.54638839e-01 3.75199974e-01
2.10615546e-02 -2.57265359e-01 -9.54412639e-01 7.03802288e-01
2.95191407e-01 -2.77702302e-01 -6.45341426e-02 7.54484713e-01
2.65615731e-02 -3.60941261e-01 -1.19683206e+00 9.49987471e-01
-1.67902514e-01 -1.29310536e+00 -3.12410921e-01 1.91162210e-02
7.64944196e-01 -5.34360886e-01 -2.90407557e-02 4.31956410e-01
-1.90118663e-02 -9.43096519e-01 3.45984131e-01 5.71485937e-01
8.20121765e-01 -6.68067932e-01 1.02288783e+00 5.44291556e-01
-8.26021194e-01 2.09994614e-01 -4.79520649e-01 3.66025895e-01
-1.78386942e-01 1.06685495e+00 -1.65061080e+00 1.06237881e-01
5.91620505e-01 4.14751977e-01 -2.84523934e-01 1.48831093e+00
9.12750587e-02 4.57403868e-01 -3.00818294e-01 -3.90009522e-01
-3.30341578e-01 -1.78564399e-01 6.69419348e-01 1.37817478e+00
7.07787931e-01 -2.75935352e-01 8.08563977e-02 1.12955451e+00
1.93997234e-01 1.87752396e-01 -5.63885033e-01 8.29379261e-02
6.92339003e-01 1.18946791e+00 -1.14268625e+00 -4.16730672e-01
-1.39787998e-02 4.72154677e-01 -1.39650866e-01 4.37113225e-01
-6.76442921e-01 -6.40179276e-01 3.42314780e-01 3.66044909e-01
1.20440833e-01 2.83514559e-01 -7.56314278e-01 -6.03115559e-01
-5.15422106e-01 -3.84149373e-01 3.65436465e-01 -5.79149604e-01
-1.27877486e+00 1.92527846e-01 -1.22825112e-02 -1.38337100e+00
-8.66630822e-02 -7.90368378e-01 -5.86155832e-01 8.29182029e-01
-1.01539254e+00 -4.12843674e-01 -2.45781794e-01 1.51606932e-01
3.30626100e-01 -1.82700932e-01 8.83738041e-01 5.16813733e-02
-4.75407302e-01 5.40670514e-01 6.34208381e-01 2.51048446e-01
1.08795011e+00 -1.34950614e+00 -2.43409544e-01 4.82961774e-01
-2.37196550e-01 3.61468792e-01 1.02657557e+00 -8.76960695e-01
-5.82100987e-01 -1.09104133e+00 6.01870358e-01 -8.94883415e-04
3.26633811e-01 -3.57733637e-01 -1.24771130e+00 1.48074672e-01
-2.33031780e-01 -1.00058913e-01 6.11585617e-01 -4.45538834e-02
1.14259511e-01 -2.77002126e-01 -1.41666651e+00 7.08379805e-01
2.17471793e-01 -2.12105457e-02 -1.65842831e-01 4.56739455e-01
2.31184840e-01 -6.88228548e-01 -6.24927759e-01 5.82126558e-01
6.60551786e-01 -6.73095345e-01 6.48123443e-01 -2.53416687e-01
1.93374425e-01 -6.50322795e-01 2.25574970e-01 -1.68306375e+00
-6.86383903e-01 -4.01159562e-02 1.91063210e-01 9.76180613e-01
5.95269501e-01 -6.85412288e-01 9.40409303e-01 4.36561227e-01
-4.11388129e-02 -8.09436202e-01 -8.13238502e-01 -6.16427124e-01
2.45738536e-01 2.45981619e-01 5.39028287e-01 8.21266234e-01
4.26206440e-01 3.11106965e-02 1.51842952e-01 2.58881688e-01
3.51022154e-01 3.39724869e-01 7.37545609e-01 -1.49430346e+00
-8.78988206e-02 -6.03059888e-01 -3.63972604e-01 -9.00177836e-01
-1.15728572e-01 -7.19821751e-01 6.02776825e-01 -1.34409165e+00
2.27285266e-01 -5.45283258e-01 -4.47420746e-01 2.14395151e-01
-1.69703349e-01 -7.98385143e-02 -2.03044727e-01 2.14534864e-01
-2.31066734e-01 2.39718631e-02 1.43342805e+00 -3.92945439e-01
-5.93597412e-01 1.35157436e-01 -4.13168371e-01 5.55911720e-01
9.78771687e-01 -4.90226716e-01 3.12896222e-02 2.48228744e-01
-5.97352274e-02 -3.90188545e-02 5.62352687e-02 -1.46116209e+00
3.71731669e-01 -2.86797255e-01 8.44963789e-01 -6.47363603e-01
-5.95660396e-02 -7.82454193e-01 2.70594358e-01 7.47080803e-01
-2.12073639e-01 -6.85942769e-01 2.10273415e-01 5.90436041e-01
-7.53777847e-02 -4.02427822e-01 1.16036201e+00 -1.37845445e-02
-4.74532247e-01 -2.48201117e-02 -6.65610313e-01 -1.03987239e-01
1.29095531e+00 -6.65897012e-01 -3.90873075e-01 -2.48441130e-01
-6.60413027e-01 2.33008996e-01 6.61462724e-01 -5.66121787e-02
5.11831284e-01 -1.29054737e+00 -6.27371907e-01 3.81065786e-01
3.25910300e-01 4.51971978e-01 1.83785021e-01 8.66434574e-01
-4.49648231e-01 3.77513885e-01 -1.60137445e-01 -9.81162488e-01
-9.49712574e-01 4.87984508e-01 6.10942125e-01 -1.71571225e-01
-3.84970337e-01 7.57004917e-01 8.23346436e-01 -2.19004184e-01
5.64852729e-02 -5.08425891e-01 -5.29893935e-01 1.19050480e-02
4.73124325e-01 6.92972466e-02 -8.03355277e-02 -3.68166327e-01
-2.66167611e-01 5.77298045e-01 -2.41740018e-01 3.16727936e-01
9.29248214e-01 2.63861626e-01 2.08117321e-01 6.22862101e-01
8.90631855e-01 2.97109168e-02 -1.51087749e+00 -5.01576327e-02
-1.49315715e-01 -3.47118258e-01 -1.70322046e-01 -8.57576132e-01
-1.00998998e+00 5.66383302e-01 8.20674956e-01 2.48420537e-01
9.05839860e-01 -7.65226185e-02 4.73873168e-01 1.80683099e-02
2.84438074e-01 -1.39629495e+00 4.50508259e-02 5.54735601e-01
6.63542211e-01 -1.06782603e+00 -1.35287061e-01 -4.08722728e-01
-6.26230240e-01 1.30167842e+00 8.04507911e-01 -7.59947523e-02
7.38422871e-01 3.47079277e-01 1.52050853e-01 5.55193834e-02
-2.41131932e-01 2.78971285e-01 -7.32409209e-02 1.01215518e+00
2.59522021e-01 1.93042278e-01 -2.07031429e-01 5.94952822e-01
1.56185791e-01 7.06092492e-02 3.32954079e-01 5.35372138e-01
-7.21865714e-01 -6.77149594e-01 -7.02006459e-01 5.39234400e-01
-3.49899769e-01 3.77047449e-01 -2.13277012e-01 7.23666251e-01
2.82707095e-01 5.59149265e-01 3.63727957e-01 -4.52870846e-01
4.80576232e-02 1.56944498e-01 1.74299300e-01 -6.69592977e-01
1.94217071e-01 3.99325103e-01 -4.49478894e-01 1.16687983e-01
3.09303235e-02 -7.04957247e-01 -2.06639242e+00 1.50068656e-01
-4.75190341e-01 1.22857369e-01 3.02712619e-01 9.06867802e-01
1.76136419e-01 3.91330808e-01 5.31221628e-01 -5.83319485e-01
-4.01284285e-02 -1.04483223e+00 -9.26871896e-01 5.76532066e-01
1.90987363e-01 -9.11161304e-01 -4.72422570e-01 2.40176052e-01] | [14.773587226867676, -3.204890012741089] |
6c38255c-be82-481a-914f-f26fe51ab48b | caesynth-real-time-timbre-interpolation-and | 2111.05174 | null | https://arxiv.org/abs/2111.05174v1 | https://arxiv.org/pdf/2111.05174v1.pdf | CAESynth: Real-Time Timbre Interpolation and Pitch Control with Conditional Autoencoders | In this paper, we present a novel audio synthesizer, CAESynth, based on a conditional autoencoder. CAESynth synthesizes timbre in real-time by interpolating the reference sounds in their shared latent feature space, while controlling a pitch independently. We show that training a conditional autoencoder based on accuracy in timbre classification together with adversarial regularization of pitch content allows timbre distribution in latent space to be more effective and stable for timbre interpolation and pitch conditioning. The proposed method is applicable not only to creation of musical cues but also to exploration of audio affordance in mixed reality based on novel timbre mixtures with environmental sounds. We demonstrate by experiments that CAESynth achieves smooth and high-fidelity audio synthesis in real-time through timbre interpolation and independent yet accurate pitch control for musical cues as well as for audio affordance with environmental sound. A Python implementation along with some generated samples are shared online. | ['Sukhan Lee', 'Aaron Valero Puche'] | 2021-11-09 | caesynth-real-time-timbre-interpolation-and-1 | https://ieeexplore.ieee.org/document/9596414/keywords#keywords | https://arxiv.org/pdf/2111.05174.pdf | ieee-mlsp-2021-9 | ['pitch-control', 'timbre-interpolation'] | ['audio', 'audio'] | [ 1.41697735e-01 -5.00127859e-02 3.33240479e-01 -7.32848570e-02
-1.01935542e+00 -7.98759162e-01 4.35793430e-01 -2.44588912e-01
-9.91590396e-02 7.76166916e-01 5.19909799e-01 2.97573805e-01
-2.10366547e-01 -7.73656249e-01 -9.38192427e-01 -7.77471423e-01
-1.69008389e-01 1.64799765e-01 -2.50447392e-01 -2.99183987e-02
-2.55379260e-01 4.19546962e-01 -2.05178881e+00 2.80373514e-01
6.42203212e-01 8.52025390e-01 1.35491818e-01 1.26803231e+00
5.06483555e-01 6.22705042e-01 -8.70001793e-01 -1.79551035e-01
3.79496992e-01 -4.28512484e-01 -2.48875707e-01 -3.13765198e-01
4.01546955e-01 -3.23478729e-01 -4.21741195e-02 8.81381452e-01
8.45971704e-01 6.28731191e-01 9.00730550e-01 -9.96006191e-01
-5.23241878e-01 9.09532905e-01 3.69335264e-01 -7.18856812e-01
7.48042285e-01 2.35732436e-01 1.14318848e+00 -7.91194737e-01
5.13081789e-01 1.13022041e+00 1.18496966e+00 5.99451184e-01
-1.82451236e+00 -1.00866783e+00 -5.58950782e-01 -1.62759185e-01
-1.55096471e+00 -4.77725536e-01 1.01534927e+00 -5.46271384e-01
6.90597951e-01 6.72802567e-01 8.16059768e-01 1.32748926e+00
4.88395877e-02 6.53315723e-01 9.95793700e-01 -7.81632781e-01
3.71054709e-01 2.89099570e-02 -7.71931887e-01 2.17170030e-01
-7.52467513e-01 7.83316672e-01 -7.72864580e-01 -4.48385686e-01
1.08441722e+00 -5.14507890e-01 -3.35205972e-01 -1.16134219e-01
-1.44334972e+00 6.50993705e-01 3.80842805e-01 2.55712897e-01
-5.50903797e-01 4.73267674e-01 2.12170005e-01 3.52380306e-01
1.56236008e-01 9.93997931e-01 -3.81303698e-01 -4.00238752e-01
-1.17639744e+00 7.32780874e-01 7.89142787e-01 6.97788775e-01
4.39130932e-01 7.40082979e-01 -6.69842586e-02 8.45781028e-01
1.35113418e-01 6.43556535e-01 7.22715974e-01 -1.39045763e+00
-6.64661154e-02 -4.74076152e-01 2.89898276e-01 -7.77705610e-01
-2.04417735e-01 -3.05441856e-01 -6.75487161e-01 7.11929142e-01
3.76468629e-01 -3.70214254e-01 -6.45395994e-01 2.11271524e+00
1.67347088e-01 4.87342685e-01 1.12964973e-01 1.01127803e+00
4.62662518e-01 8.16222012e-01 5.23671601e-03 -1.54264957e-01
1.08820772e+00 -5.37766457e-01 -1.01740789e+00 4.00812745e-01
-1.52653724e-01 -1.06897283e+00 1.37682521e+00 8.68576348e-01
-1.32473803e+00 -1.24306202e+00 -1.21502495e+00 -9.43175703e-02
-9.50088575e-02 5.22658005e-02 5.77115417e-01 8.39764893e-01
-7.39295900e-01 1.27236676e+00 -8.17790449e-01 3.31304640e-01
-2.62234449e-01 1.99736908e-01 -3.15452069e-01 8.64576340e-01
-1.50027227e+00 3.62078339e-01 3.79243940e-01 -1.96613193e-01
-1.09091198e+00 -1.07776439e+00 -8.23589623e-01 1.36069387e-01
-2.38844544e-01 -8.14779043e-01 1.50855017e+00 -1.15457726e+00
-2.37003517e+00 1.60323709e-01 5.15140116e-01 -6.50482178e-01
4.95892346e-01 -5.85427880e-01 -7.13389039e-01 3.31467353e-02
-5.21379113e-02 7.82988608e-01 1.44734371e+00 -1.16995001e+00
-2.45288610e-01 2.93206394e-01 -3.44080627e-01 1.40529498e-01
-1.61118105e-01 -2.91689217e-01 3.31678629e-01 -1.14840245e+00
8.55571628e-02 -7.88285971e-01 -6.21762127e-02 -9.70667005e-02
-4.22729105e-01 1.60689726e-01 3.60998303e-01 -6.81583762e-01
1.08103418e+00 -2.24785972e+00 3.70246530e-01 4.76200171e-02
-2.01502949e-01 -2.09063798e-01 1.82553425e-01 4.08222318e-01
-4.30090666e-01 -1.32531509e-01 -1.91920236e-01 -4.57105249e-01
4.99524802e-01 -1.66886494e-01 -9.40212548e-01 2.35876739e-01
1.55922532e-01 4.94095683e-01 -9.49494183e-01 -8.80862623e-02
4.00413990e-01 8.56735766e-01 -1.14261496e+00 5.89691579e-01
-4.99124855e-01 9.16388392e-01 1.64174631e-01 3.64862829e-01
4.26835746e-01 7.10094988e-01 -2.04561472e-01 -3.61550659e-01
-3.63696218e-01 3.43651533e-01 -1.43182409e+00 1.99476159e+00
-8.31267178e-01 6.25127912e-01 1.83968440e-01 -1.98826686e-01
1.22237957e+00 8.03141057e-01 2.83007622e-01 3.53780668e-03
-1.56077687e-02 3.15938652e-01 -2.10848033e-01 -3.09898853e-01
6.45031869e-01 -9.47731555e-01 -2.27726683e-01 3.98692757e-01
5.10088325e-01 -8.53081286e-01 -5.53385556e-01 -4.57643062e-01
5.42110622e-01 5.90894401e-01 4.21959944e-02 -1.04171716e-01
1.78097010e-01 -7.40101039e-01 3.10705692e-01 6.22088075e-01
9.03134793e-02 9.34018612e-01 -1.84236020e-01 -7.83361867e-02
-1.37954164e+00 -1.61496615e+00 -2.55638510e-01 1.05139911e+00
-5.12183845e-01 -4.03538585e-01 -7.37736344e-01 2.77091384e-01
8.33032131e-02 1.09382117e+00 -5.30305386e-01 -2.58185625e-01
-5.33254504e-01 1.45344406e-01 9.38880026e-01 4.30264443e-01
9.22421515e-02 -1.45686972e+00 -5.53944528e-01 6.01846695e-01
-1.17891409e-01 -6.25021935e-01 -4.68690008e-01 4.38328624e-01
-6.32686079e-01 -3.00796300e-01 -8.05771351e-01 -5.35180986e-01
-2.44279742e-01 -7.68226147e-01 8.99819136e-01 -8.10873568e-01
-4.45512384e-01 3.80737662e-01 -1.70227930e-01 -6.24061823e-01
-7.06396222e-01 -1.11889511e-01 7.36629426e-01 2.60298222e-01
-4.44235146e-01 -1.35439730e+00 -5.04460275e-01 7.66241997e-02
-9.14789796e-01 1.53561071e-01 -1.29951015e-01 9.12696421e-01
7.11290836e-01 1.11172266e-01 7.83141613e-01 -3.83995831e-01
7.94393599e-01 -2.00738966e-01 -3.81278604e-01 -4.72969055e-01
-1.15708619e-01 1.38572529e-02 7.94164062e-01 -8.87767017e-01
-1.12420702e+00 2.09441051e-01 -7.14942992e-01 -7.82382071e-01
-4.24778104e-01 2.10041612e-01 -1.92753077e-01 4.37489092e-01
1.30199921e+00 -2.55447719e-02 -1.82483599e-01 -6.54838502e-01
8.25723767e-01 7.28405893e-01 1.06827283e+00 -8.70828986e-01
7.41608799e-01 2.28306547e-01 -1.36787534e-01 -9.15851474e-01
-4.23819214e-01 2.49307025e-02 -5.97898841e-01 -2.19426975e-01
8.81712317e-01 -9.76009965e-01 -9.42083478e-01 2.76581317e-01
-8.03110480e-01 -5.47034025e-01 -1.03425860e+00 9.39900756e-01
-1.51703012e+00 -7.86589980e-02 -6.56596720e-01 -1.12870944e+00
-2.61715621e-01 -9.47255850e-01 9.45584834e-01 -5.12611978e-02
-7.18388259e-01 -7.61194468e-01 4.54046100e-01 -2.34440431e-01
2.71652341e-01 6.79394424e-01 6.03173971e-01 9.35314037e-03
-4.61750060e-01 -2.88230628e-01 7.63245106e-01 5.65170348e-01
1.58137798e-01 1.48938015e-01 -1.65404820e+00 -1.94445074e-01
1.33576602e-01 -6.16077542e-01 5.15620291e-01 5.29186070e-01
1.01207066e+00 -5.00725865e-01 3.48678678e-01 8.55203390e-01
1.09975791e+00 7.04818070e-02 6.70761287e-01 -1.08490720e-01
3.88075083e-01 5.27512133e-01 4.49282229e-01 7.69427001e-01
-4.47528005e-01 8.01261663e-01 2.45467111e-01 1.73837990e-01
-2.83338726e-01 -7.39398837e-01 4.54363763e-01 1.12072623e+00
-1.35597453e-01 2.13432431e-01 -3.54965240e-01 4.18519914e-01
-1.31499231e+00 -1.15831852e+00 3.86508673e-01 2.25474048e+00
1.34349000e+00 -5.56134582e-02 2.42187232e-01 7.71499634e-01
5.85213602e-01 -2.15623781e-01 -3.41751486e-01 -6.85129583e-01
-1.17919929e-01 8.71198952e-01 -1.02695003e-01 8.70591342e-01
-1.14760566e+00 7.66294956e-01 6.51190376e+00 8.96434605e-01
-1.18398416e+00 1.86332017e-02 -1.85154397e-02 -1.43979549e-01
-4.60475147e-01 -2.14651793e-01 -2.25120619e-01 1.58669293e-01
1.11660850e+00 9.90919862e-03 9.60457325e-01 8.19724739e-01
-5.10513373e-02 4.02880490e-01 -1.16000259e+00 9.65285182e-01
-3.26883286e-01 -1.30276573e+00 -1.17061414e-01 -6.14200473e-01
6.77524447e-01 -5.22133172e-01 7.41674542e-01 2.96595842e-01
1.31066829e-01 -1.20762491e+00 1.39815176e+00 9.00496781e-01
1.41701245e+00 -1.03086102e+00 5.18013500e-02 3.92723411e-01
-1.07747602e+00 -2.47858316e-02 -8.51545036e-02 -3.08566391e-01
9.62141976e-02 1.41407862e-01 -1.06410670e+00 2.63786554e-01
5.27917206e-01 3.52762818e-01 1.87860712e-01 9.72406387e-01
-4.24109161e-01 8.35272431e-01 -3.34522158e-01 1.62333354e-01
-1.61285013e-01 1.64557755e-01 7.92119741e-01 1.20606446e+00
5.80354273e-01 -3.03299678e-03 2.69803088e-02 1.24512327e+00
1.05570018e-01 1.81669176e-01 -3.16019475e-01 -1.34535283e-01
6.46795690e-01 9.36544478e-01 -4.62902784e-02 2.90062744e-03
3.49698752e-01 9.02387142e-01 -1.50637642e-01 5.04520535e-01
-7.60175228e-01 -7.99760222e-01 7.60308683e-01 4.47882600e-02
1.61074281e-01 -3.07119757e-01 -1.70779482e-01 -9.02011275e-01
-4.32597101e-01 -7.90818572e-01 -3.65931056e-02 -1.33517504e+00
-1.23825729e+00 7.39473701e-01 -1.32608548e-01 -1.43491852e+00
-8.60994577e-01 -2.48092979e-01 -3.52187216e-01 1.19065881e+00
-8.13566804e-01 -1.09678948e+00 1.24596067e-01 6.56673133e-01
3.26931894e-01 -3.07098478e-01 1.64597428e+00 1.87485889e-01
3.42894912e-01 6.57301724e-01 1.54555097e-01 -1.34225667e-01
9.49100852e-01 -1.49699140e+00 3.85921568e-01 1.32415354e-01
5.62381446e-01 6.66603386e-01 1.37095356e+00 -3.59397113e-01
-8.57693970e-01 -9.01660860e-01 4.92240578e-01 -3.13007802e-01
7.17507720e-01 -6.16442680e-01 -8.22141171e-01 5.49542010e-01
2.79939681e-01 -2.24396020e-01 9.48040903e-01 6.22598901e-02
-4.71234441e-01 -2.46381965e-02 -1.19780278e+00 7.42738664e-01
4.92155910e-01 -1.16516066e+00 -7.91051924e-01 1.43767163e-01
1.09704208e+00 -6.20008945e-01 -1.09759057e+00 2.94134587e-01
9.02407944e-01 -9.88369882e-01 1.31243682e+00 -3.94190699e-01
4.24940974e-01 -4.15095836e-01 -3.38729590e-01 -1.46493375e+00
-3.27579647e-01 -1.36877573e+00 -7.86467791e-02 1.10251009e+00
1.09878108e-01 -1.71615005e-01 3.98974925e-01 2.55461838e-02
-2.96018809e-01 -9.13483202e-02 -1.05331564e+00 -7.16731191e-01
3.86302024e-01 -1.02198875e+00 8.29293668e-01 8.28823388e-01
6.45387918e-02 9.33919102e-02 -7.14183211e-01 3.61183077e-01
5.49888134e-01 1.11563124e-01 8.71214449e-01 -1.08997607e+00
-1.02215075e+00 -1.96895674e-01 -2.82533467e-01 -6.81286752e-01
2.16216996e-01 -8.03863049e-01 3.61901700e-01 -7.10250854e-01
-6.06978178e-01 -5.22800446e-01 -3.27940851e-01 1.70077682e-01
2.43227586e-01 5.53546846e-01 3.93356323e-01 5.67691848e-02
1.47351190e-01 8.50962400e-01 1.24020195e+00 2.56053470e-02
-5.52492380e-01 2.33768001e-01 -7.46658891e-02 9.31603551e-01
5.42465150e-01 -3.62118363e-01 -2.62698025e-01 8.99173170e-02
3.20800871e-01 6.45943999e-01 5.33424914e-01 -1.63914251e+00
2.92749312e-02 1.37291640e-01 5.99692643e-01 -3.66290838e-01
9.51682866e-01 -7.10077167e-01 7.11873531e-01 1.78117827e-01
-7.90630996e-01 -5.34610331e-01 4.93090183e-01 4.14179116e-01
-2.32190549e-01 -1.58890337e-01 7.21602082e-01 8.22337419e-02
1.82094518e-02 -5.32371514e-02 -4.33450788e-01 -2.96219587e-01
3.70248169e-01 -4.03425433e-02 4.36786532e-01 -8.09325695e-01
-1.68491626e+00 -8.07877898e-01 1.09601229e-01 1.92491874e-01
5.16831160e-01 -1.87751126e+00 -7.04669774e-01 7.38486767e-01
-7.74587020e-02 -3.59948397e-01 4.68463928e-01 7.66092166e-02
-5.41730046e-01 1.70361653e-01 -4.26720619e-01 -4.78560716e-01
-9.14547861e-01 6.94042146e-01 4.85580772e-01 2.95409024e-01
-6.37289107e-01 1.11224616e+00 1.93679079e-01 -6.05743468e-01
4.06588346e-01 -5.02668262e-01 4.70562987e-02 -3.38811316e-02
4.90688443e-01 7.11481348e-02 -2.51541793e-01 -4.77497429e-01
1.58245385e-01 4.24683005e-01 8.97316515e-01 -8.73789072e-01
1.12198305e+00 4.98128794e-02 2.45343819e-01 1.11853659e+00
1.04070497e+00 7.68232167e-01 -1.53321183e+00 1.72224626e-01
-3.54273438e-01 -4.81988043e-01 6.70225620e-02 -9.47318494e-01
-5.23113549e-01 1.03929770e+00 6.40222073e-01 2.65415877e-01
1.18905294e+00 -4.19413686e-01 6.56859457e-01 8.14147070e-02
4.06084388e-01 -9.56784725e-01 3.33583474e-01 3.03501517e-01
1.37985969e+00 -3.07848573e-01 -3.56577575e-01 3.41652818e-02
-5.84252834e-01 1.24222755e+00 9.79648903e-03 -5.24251997e-01
9.28483963e-01 5.01112878e-01 2.26055428e-01 5.58757484e-01
-5.48303723e-01 1.73507303e-01 4.18128490e-01 7.93521762e-01
5.15300453e-01 4.09761459e-01 3.77166718e-01 1.04740775e+00
-1.09517586e+00 -2.04024985e-01 2.95066953e-01 2.99737573e-01
-2.00832844e-01 -1.09424233e+00 -8.41751993e-01 -3.00786078e-01
-4.92420912e-01 -1.63500130e-01 1.53213218e-01 4.41000521e-01
5.02581418e-01 6.31147742e-01 2.52813041e-01 -6.30988538e-01
3.97456437e-01 4.53435481e-01 7.54063427e-01 -3.54647636e-01
-8.39057148e-01 7.05630183e-01 2.08774023e-02 -3.28702897e-01
-8.55855867e-02 -7.58730769e-01 -1.23387694e+00 2.68407119e-03
-3.66254896e-01 2.62009501e-01 7.14463890e-01 3.10997874e-01
2.21940875e-03 9.36546564e-01 7.13474154e-01 -1.28176570e+00
-7.78652787e-01 -1.10629094e+00 -1.07545924e+00 3.61739188e-01
7.16589749e-01 -4.98015404e-01 -5.38579762e-01 4.49762404e-01] | [15.671387672424316, 5.867585182189941] |
f8593254-864e-4912-bfb1-98dd7523435b | spatial-semantic-embedding-network-fast-3d | 2007.03169 | null | https://arxiv.org/abs/2007.03169v1 | https://arxiv.org/pdf/2007.03169v1.pdf | Spatial Semantic Embedding Network: Fast 3D Instance Segmentation with Deep Metric Learning | We propose spatial semantic embedding network (SSEN), a simple, yet efficient algorithm for 3D instance segmentation using deep metric learning. The raw 3D reconstruction of an indoor environment suffers from occlusions, noise, and is produced without any meaningful distinction between individual entities. For high-level intelligent tasks from a large scale scene, 3D instance segmentation recognizes individual instances of objects. We approach the instance segmentation by simply learning the correct embedding space that maps individual instances of objects into distinct clusters that reflect both spatial and semantic information. Unlike previous approaches that require complex pre-processing or post-processing, our implementation is compact and fast with competitive performance, maintaining scalability on large scenes with high resolution voxels. We demonstrate the state-of-the-art performance of our algorithm in the ScanNet 3D instance segmentation benchmark on AP score. | ['Young Min Kim', 'Sang Kyun Cha', 'Junha Chun', 'Dongsu Zhang'] | 2020-07-07 | null | null | null | null | ['3d-instance-segmentation-1'] | ['computer-vision'] | [ 1.22910626e-01 2.92603731e-01 1.56845540e-01 -6.59494102e-01
-7.99924374e-01 -5.17479599e-01 5.11379719e-01 4.04044747e-01
-4.75317001e-01 8.08755085e-02 8.63096341e-02 -1.55936986e-01
-4.22119409e-01 -1.08124101e+00 -7.64659345e-01 -3.53949308e-01
-3.39125097e-01 9.01426494e-01 5.53893149e-01 2.65389621e-01
8.00865889e-02 7.92073846e-01 -1.56841826e+00 1.12112693e-01
7.29307055e-01 1.26551247e+00 2.33999327e-01 7.50001013e-01
-5.87374747e-01 3.87944728e-01 -5.71809292e-01 2.73433533e-02
5.78328609e-01 -4.33591418e-02 -1.09116912e+00 2.07055911e-01
7.61546910e-01 -2.29553491e-01 -4.83182371e-01 1.08216572e+00
5.73705971e-01 2.10155353e-01 8.32335532e-01 -1.06066251e+00
-5.11375546e-01 2.09004760e-01 -4.69568342e-01 3.64753790e-02
3.28662574e-01 -6.70893341e-02 1.06360471e+00 -1.16872847e+00
7.58087993e-01 1.22135139e+00 7.76327252e-01 2.47437954e-01
-1.37597370e+00 -1.81827307e-01 1.47861525e-01 1.61456183e-01
-1.50277293e+00 -1.92912843e-03 7.76071429e-01 -5.68611026e-01
1.12813103e+00 3.05062503e-01 8.88324559e-01 7.53294706e-01
-9.02014673e-02 9.10952866e-01 9.19567108e-01 5.88575527e-02
6.31014407e-01 -1.01704665e-01 3.67051065e-01 6.35582745e-01
1.92420512e-01 -1.58164710e-01 -2.43474603e-01 8.18615854e-02
1.04888380e+00 8.93675461e-02 1.25454932e-01 -9.11976933e-01
-1.52123737e+00 6.86911285e-01 9.47261930e-01 9.92054269e-02
-4.24088329e-01 4.23058271e-01 1.28745154e-01 1.05157547e-01
5.95849812e-01 5.20409465e-01 -5.85699558e-01 2.88955774e-02
-1.15698516e+00 4.05813932e-01 7.40462422e-01 1.16830039e+00
8.11011553e-01 -3.27186078e-01 -2.97348443e-02 7.29081571e-01
1.77028880e-01 3.27048719e-01 6.51360154e-02 -1.27889884e+00
2.91329443e-01 9.07904029e-01 1.19155366e-02 -9.90537524e-01
-6.38786554e-01 -5.00157714e-01 -6.43378556e-01 4.89909589e-01
3.33935112e-01 4.42486852e-01 -1.19907475e+00 1.24678767e+00
7.98619151e-01 5.58036149e-01 -1.14740305e-01 1.10152054e+00
1.21332586e+00 5.16792297e-01 -6.82304129e-02 7.13562906e-01
1.27548313e+00 -9.14898992e-01 -1.61735803e-01 -3.24676514e-01
4.34863329e-01 -4.02466357e-01 1.01842213e+00 2.93311477e-02
-9.75681126e-01 -5.56854784e-01 -1.25333393e+00 -6.15771949e-01
-6.62300289e-01 -3.30338538e-01 8.72577369e-01 4.65308607e-01
-9.62354004e-01 6.34651065e-01 -1.04901326e+00 -3.52619618e-01
9.07609403e-01 4.20766473e-01 -5.88547528e-01 -2.08494276e-01
-6.02818727e-01 5.98032475e-01 3.27871740e-01 -3.28890562e-01
-8.11585307e-01 -1.10817647e+00 -1.16354585e+00 -1.86642498e-01
3.36398929e-01 -8.51313293e-01 9.68385398e-01 -2.96519488e-01
-1.05618870e+00 1.26946688e+00 -7.92812631e-02 -3.75896513e-01
5.75953960e-01 -3.32941979e-01 -1.32116396e-03 2.32184976e-01
3.47894937e-01 1.02244544e+00 5.83460748e-01 -1.45957160e+00
-6.31659925e-01 -6.85457528e-01 2.56822079e-01 2.88790435e-01
2.24354446e-01 -4.03947115e-01 -5.60677350e-01 -2.67420888e-01
1.08570421e+00 -6.14413381e-01 -7.52729714e-01 4.03635830e-01
-5.64914227e-01 -3.02141249e-01 8.72140348e-01 -4.44963396e-01
4.75263536e-01 -2.17460489e+00 1.35181457e-01 3.31594378e-01
6.52253568e-01 -4.24134493e-01 -9.02659222e-02 -1.30912140e-01
1.05422668e-01 1.83046266e-01 -6.57902062e-01 -3.52613419e-01
5.40257394e-01 4.10155028e-01 1.27513409e-02 6.47869945e-01
2.20209390e-01 1.06621909e+00 -1.14758933e+00 -5.72770536e-01
6.95761263e-01 5.82635045e-01 -6.20994270e-01 1.04739234e-01
-2.44151518e-01 3.75483423e-01 -5.43400049e-01 7.45483577e-01
7.27737129e-01 -4.03988391e-01 -3.46487403e-01 -3.15488487e-01
1.41987756e-01 5.35646915e-01 -1.60895193e+00 2.37528348e+00
-3.63714457e-01 5.82027912e-01 -3.97127792e-02 -1.04579377e+00
7.09210157e-01 -1.47452965e-01 9.62594867e-01 -7.83347547e-01
-4.98099551e-02 2.27722302e-01 -5.11662066e-01 -3.44479978e-01
5.70298076e-01 3.33005100e-01 -3.13738018e-01 5.28421104e-01
9.06109363e-02 -6.99142814e-01 -6.92533553e-02 1.16705216e-01
1.43325400e+00 1.19782880e-01 4.58296426e-02 -3.07899982e-01
5.08773401e-02 2.45212674e-01 3.11314166e-01 8.15748334e-01
-1.44045174e-01 9.20615077e-01 2.07977369e-01 -6.97939754e-01
-1.02063978e+00 -1.72685838e+00 -4.41823721e-01 7.11632729e-01
7.67993927e-01 -2.65805990e-01 -7.49940872e-01 -8.06037366e-01
2.57121623e-01 4.26989496e-01 -4.67310816e-01 3.21164459e-01
-5.44299364e-01 -6.48315132e-01 2.83436090e-01 6.67614102e-01
6.92511261e-01 -7.92537332e-01 -8.62929940e-01 3.69779527e-01
8.73915032e-02 -1.42067444e+00 -1.96587384e-01 3.94691110e-01
-1.02211607e+00 -1.24090409e+00 -4.27536190e-01 -1.01275957e+00
8.63359809e-01 2.62466699e-01 1.49369812e+00 -6.80590123e-02
-6.40854716e-01 6.20706141e-01 -7.50335231e-02 -2.48521984e-01
1.37149781e-01 1.45323098e-01 -1.28606647e-01 -3.61427873e-01
4.53872681e-01 -7.81651258e-01 -8.97788465e-01 2.95179039e-01
-6.54184341e-01 5.00842519e-02 4.75823492e-01 3.98791671e-01
1.40158653e+00 2.41217434e-01 -1.66426152e-01 -7.54190326e-01
7.40870908e-02 -2.98589855e-01 -5.70719063e-01 -1.00751288e-01
-1.66373804e-01 2.62555759e-02 7.74615034e-02 -1.16611766e-02
-5.10297775e-01 1.92443937e-01 -2.37797707e-01 -2.70050675e-01
-6.40040636e-01 6.58399388e-02 -2.06797853e-01 -5.66306338e-02
5.59039235e-01 1.06143795e-01 -4.83550787e-01 -6.23518288e-01
7.42695689e-01 3.74108464e-01 7.10584641e-01 -7.11818457e-01
1.05327153e+00 7.81671643e-01 1.03795037e-01 -9.94098127e-01
-8.67944419e-01 -5.75614870e-01 -1.28468132e+00 7.35685462e-03
1.31169510e+00 -8.50546539e-01 -2.87702709e-01 1.77003771e-01
-1.19095361e+00 -5.65058470e-01 -7.44711339e-01 5.08861303e-01
-7.33951449e-01 -7.50758946e-02 -4.73413646e-01 -3.00326079e-01
-6.77924789e-03 -1.02909255e+00 1.52231765e+00 -6.93321377e-02
-1.71670139e-01 -9.00938332e-01 -2.94849053e-02 3.68026674e-01
1.00227274e-01 7.51138389e-01 9.08500135e-01 -4.15812224e-01
-1.22332907e+00 -1.60987511e-01 -4.49914247e-01 9.65505093e-02
-9.47289467e-02 -2.95535684e-01 -9.94511008e-01 4.44721170e-02
-5.38894795e-02 -8.46325606e-03 6.90625906e-01 5.52126825e-01
1.60594141e+00 -2.02846974e-02 -4.67011303e-01 1.02052617e+00
1.57529163e+00 -1.95644513e-01 5.04223704e-01 3.61653268e-01
1.07659471e+00 6.30624771e-01 5.32256484e-01 2.06766039e-01
6.31324232e-01 3.59336615e-01 6.96746051e-01 -3.91287118e-01
-3.35582197e-01 -2.04580963e-01 -2.37699598e-01 5.54378033e-01
1.78767294e-01 -2.88353078e-02 -1.01935792e+00 7.80114472e-01
-1.70386744e+00 -6.63356662e-01 -3.13566089e-01 1.95544875e+00
3.57588738e-01 3.27025980e-01 -1.34357005e-01 3.16509008e-01
3.24718118e-01 4.56548333e-01 -7.13827729e-01 -6.11151978e-02
8.43027979e-03 6.63456619e-01 8.93677294e-01 4.81098771e-01
-1.34481800e+00 9.16891217e-01 6.53283405e+00 4.75967050e-01
-7.27194190e-01 1.20883547e-01 6.30585015e-01 -3.00667077e-01
-3.65671664e-01 -3.43167216e-01 -3.59766722e-01 1.62949011e-01
5.16402364e-01 1.30472019e-01 2.48654634e-01 1.08154166e+00
-4.44355123e-02 -2.41282955e-02 -1.44651794e+00 1.22190380e+00
-2.30381846e-01 -1.46168220e+00 -2.10353419e-01 1.31356329e-01
7.38873184e-01 5.01347959e-01 2.00998951e-02 4.30954471e-02
4.31731313e-01 -1.19148397e+00 9.39629614e-01 2.29980290e-01
5.63483715e-01 -5.91707349e-01 3.78354788e-01 1.89343542e-01
-1.36219621e+00 1.67326629e-01 -5.31198978e-01 1.41870469e-01
1.20530546e-01 8.86482775e-01 -6.04084492e-01 2.89384633e-01
1.01417637e+00 8.19454908e-01 -5.36162317e-01 1.12863648e+00
-2.70398766e-01 2.29339644e-01 -6.15771234e-01 1.63498998e-01
4.77412552e-01 -2.78426766e-01 6.36071384e-01 1.02149367e+00
2.79694498e-01 1.74895212e-01 3.74628425e-01 1.14156997e+00
-1.73881084e-01 -1.00879557e-01 -6.82729602e-01 2.47101530e-01
4.33400065e-01 1.08816588e+00 -1.19010222e+00 -4.10020739e-01
-2.76674360e-01 1.25253439e+00 2.70964473e-01 3.26061934e-01
-9.10856366e-01 -3.98312479e-01 1.19653344e+00 2.43857518e-01
4.83517200e-01 -7.99690545e-01 -9.72110271e-01 -7.86809742e-01
2.27548346e-01 -1.56499326e-01 2.33446509e-01 -5.74554384e-01
-1.19552124e+00 3.63758594e-01 -4.23177704e-02 -1.09853590e+00
3.19993496e-01 -8.13391447e-01 -3.55688363e-01 4.14864868e-01
-1.48772848e+00 -9.94339705e-01 -5.91343045e-01 5.77221513e-01
6.75265610e-01 3.69332612e-01 7.29637027e-01 3.43057960e-01
-1.24533758e-01 1.91448167e-01 5.12078255e-02 3.30562353e-01
4.79392223e-02 -1.86127925e+00 7.32688665e-01 6.26517594e-01
5.88992357e-01 2.31407672e-01 3.95574480e-01 -3.52388680e-01
-1.28097510e+00 -1.32155299e+00 8.19702327e-01 -9.02693808e-01
3.39592636e-01 -8.00140202e-01 -7.55311430e-01 6.06511533e-01
-2.94504791e-01 4.08564717e-01 6.28709078e-01 1.14805415e-01
-4.59876299e-01 -1.12149073e-02 -1.45367861e+00 5.12758791e-01
1.77693176e+00 -6.49659574e-01 -8.26677024e-01 5.99307716e-01
1.10234380e+00 -8.87691975e-01 -1.24040115e+00 3.35366398e-01
3.09707373e-01 -1.04388678e+00 1.66033494e+00 -3.81981522e-01
3.80063206e-01 -5.06570220e-01 -5.97519994e-01 -1.03334486e+00
-4.06694144e-01 -4.96404618e-02 -6.13595769e-02 5.82620144e-01
2.83636898e-01 -3.51267904e-01 1.12078619e+00 8.09578776e-01
-4.91323709e-01 -7.57135510e-01 -1.16986358e+00 -8.15784276e-01
-9.21432599e-02 -1.11864865e+00 1.12218499e+00 9.77733970e-01
-6.12067103e-01 -2.68543679e-02 4.15217042e-01 6.72309339e-01
1.11089683e+00 4.90900218e-01 8.98793459e-01 -1.38506281e+00
4.48482558e-02 -5.09478271e-01 -1.00583756e+00 -1.37436712e+00
7.75675401e-02 -1.18634009e+00 2.10031793e-02 -2.15791392e+00
-2.24692419e-01 -7.85133541e-01 -4.74162877e-01 2.78703630e-01
2.35716596e-01 5.26794314e-01 -2.22487330e-01 -5.80186732e-02
-9.78748500e-01 4.69462574e-01 1.20158672e+00 -6.07289195e-01
-3.62996757e-01 -4.01379079e-01 -3.72726530e-01 7.35154986e-01
4.83294934e-01 -5.84044218e-01 -4.01942194e-01 -9.27238286e-01
-1.09834909e-01 -4.03789103e-01 6.83438420e-01 -1.26933420e+00
1.23951390e-01 -4.45667580e-02 6.70638323e-01 -1.07363987e+00
5.84051192e-01 -1.12794638e+00 1.68241441e-01 2.40908220e-01
-1.73169181e-01 -2.23862246e-01 3.43913399e-02 5.05440593e-01
5.42381546e-03 1.08587012e-01 6.89548135e-01 -4.93968248e-01
-1.14860821e+00 8.13509285e-01 5.76296933e-02 1.74269468e-01
1.13131905e+00 -6.88989997e-01 6.29320815e-02 1.61931634e-01
-9.09089327e-01 2.92528778e-01 6.67718410e-01 4.47174340e-01
9.17342961e-01 -1.47410893e+00 -4.97701913e-01 2.68871903e-01
1.23681828e-01 8.38531673e-01 1.44314513e-01 4.73078549e-01
-8.99056554e-01 1.19095460e-01 2.81906389e-02 -1.15782070e+00
-9.08977807e-01 1.90268114e-01 4.98808086e-01 1.19447842e-01
-1.14724517e+00 1.06903899e+00 3.38969916e-01 -8.98514509e-01
2.71949023e-01 -7.27663457e-01 1.95732385e-01 -1.36581793e-01
2.63876200e-01 5.45629799e-01 5.50796576e-02 -6.41886830e-01
-5.26199877e-01 8.57374549e-01 4.16728437e-01 -1.13178886e-01
1.65764689e+00 7.54256845e-02 -1.55127451e-01 4.64354366e-01
1.54136074e+00 -3.96959633e-01 -1.28073442e+00 -2.26461530e-01
-6.51808176e-03 -7.64894605e-01 3.44810039e-01 -5.08906364e-01
-1.04699218e+00 8.04677486e-01 7.63259530e-01 1.44047499e-01
6.49344027e-01 4.93881911e-01 1.14478469e+00 4.56835657e-01
7.21160054e-01 -1.12645698e+00 1.41975105e-01 5.52888095e-01
6.73435390e-01 -1.34072328e+00 2.17119940e-02 -4.36689019e-01
-2.99301110e-02 8.37770998e-01 4.92451906e-01 -3.54288638e-01
1.02029324e+00 7.83620700e-02 4.26188186e-02 -7.70324171e-01
-6.59264275e-04 -2.69871622e-01 4.69180554e-01 8.54725778e-01
-8.37204531e-02 3.23180825e-01 2.68098742e-01 3.39544088e-01
-5.57304859e-01 -6.14347994e-01 4.14974988e-02 9.19837952e-01
-5.64702272e-01 -7.56242990e-01 -2.43724048e-01 6.19239271e-01
2.78182719e-02 1.15668364e-01 -2.72043645e-01 8.48541915e-01
2.65420407e-01 5.81980228e-01 4.93288934e-01 -3.94189566e-01
6.17198825e-01 -5.77197634e-02 5.88420749e-01 -6.81422114e-01
-1.03640623e-01 -2.67681301e-01 -3.63833427e-01 -1.19199777e+00
-4.26478237e-01 -6.35906696e-01 -1.73354971e+00 -1.14084631e-01
2.18912676e-01 -2.80944556e-01 1.01594579e+00 6.96764112e-01
4.80353296e-01 5.11439085e-01 6.01101398e-01 -1.21105099e+00
9.29544270e-02 -3.75076592e-01 -6.45472407e-01 5.80742598e-01
2.36492842e-01 -7.30298281e-01 -2.85233378e-01 -2.07303047e-01] | [8.09653377532959, -3.196101427078247] |
e072f61d-9c5d-48ea-9bba-7bf00c828289 | lexical-event-ordering-with-an-edge-factored | null | null | https://aclanthology.info/papers/N15-1122/n15-1122 | https://www.aclweb.org/anthology/N15-1122 | Lexical Event Ordering with an Edge-Factored Model | null | ['Omri Abend', 'Shay B. Cohen', 'Mark Steedman'] | 2015-05-01 | null | null | null | hlt-2015-5 | ['temporal-information-extraction'] | ['natural-language-processing'] | [-2.44508207e-01 3.89024585e-01 -2.65282035e-01 -2.15905145e-01
-8.60921741e-02 -7.76765764e-01 4.48510379e-01 -7.23253429e-01
-5.48377395e-01 1.31954515e+00 3.66348401e-02 -9.49533224e-01
-2.40340635e-01 -1.05564880e+00 -8.44053447e-01 -8.75781775e-01
-7.42435038e-01 6.86515033e-01 1.44298598e-01 -6.52004302e-01
8.47113907e-01 6.29996777e-01 -1.62287033e+00 5.77558100e-01
6.81926727e-01 5.49597681e-01 5.66466339e-02 1.04480565e+00
6.13576770e-02 1.59847176e+00 -6.05450153e-01 -4.56729174e-01
1.43710867e-01 -1.88022718e-01 -5.35770595e-01 -3.36825520e-01
7.91536197e-02 -5.88986814e-01 -3.10633808e-01 6.07356608e-01
1.14270830e+00 -5.53557873e-02 1.07025254e+00 -1.51067197e+00
-7.97060013e-01 5.44234335e-01 2.30399013e-01 1.20632313e-01
6.10446513e-01 -1.79472014e-01 3.51212233e-01 -1.36621380e+00
6.32680655e-01 8.64870071e-01 9.47046995e-01 5.99273622e-01
-1.20647264e+00 -5.60552299e-01 -7.17137158e-01 -2.57835120e-01
-1.59567785e+00 -6.30042374e-01 6.23617955e-02 -3.34735900e-01
1.69142139e+00 6.72587395e-01 1.35647905e+00 1.47341585e+00
1.36649036e+00 4.72517729e-01 1.13904178e+00 -3.16340059e-01
3.65351558e-01 6.43754780e-01 -2.29148138e-02 5.51889002e-01
1.19686353e+00 6.76312149e-01 -5.60917914e-01 -7.92779744e-01
1.11365557e+00 -3.12949389e-01 2.60771513e-01 -7.06989706e-01
-1.06566191e+00 7.88466871e-01 -1.33808568e-01 2.20882341e-01
-3.46203089e-01 -9.92989019e-02 2.24548295e-01 4.12961632e-01
-2.94082850e-01 4.77862865e-01 -8.89795244e-01 -1.87686309e-01
-1.01472771e+00 1.91863477e-01 1.30615628e+00 1.45811689e+00
4.72722203e-03 3.70100170e-01 1.66493103e-01 2.89523482e-01
5.29429734e-01 1.05927479e+00 4.36951250e-01 -1.33365822e+00
-9.55119133e-02 4.20878716e-02 6.12505674e-01 -1.13473701e+00
-7.12930143e-01 -2.41300568e-01 -9.54724610e-01 5.18771529e-01
-1.59270123e-01 -4.59180593e-01 -7.59686172e-01 4.79171664e-01
-1.81984559e-01 -3.19489211e-01 5.45619786e-01 2.69257158e-01
7.28577793e-01 1.45816103e-01 9.25893057e-03 -4.73371416e-01
8.38296890e-01 -1.26899445e+00 -1.11426139e+00 2.33304992e-01
6.82289064e-01 -1.11704338e+00 4.06794518e-01 4.29345578e-01
-1.87200487e+00 -9.69799384e-02 -1.08317137e+00 1.19022481e-01
-7.97166348e-01 -2.57194549e-01 6.66624963e-01 1.49922979e+00
-1.63972652e+00 6.47176981e-01 -6.05021298e-01 5.40335439e-02
-4.73218322e-01 9.07614470e-01 -3.47142309e-01 2.90655226e-01
-1.28751957e+00 9.52449858e-01 -1.04219764e-01 3.03568155e-01
-3.30324143e-01 -1.84501112e-01 -1.03385210e+00 -6.24127567e-01
-2.59417385e-01 -1.07312346e+00 1.36302292e+00 1.09815992e-01
-1.30379498e+00 1.00962746e+00 -4.14315462e-01 -1.14109404e-01
6.61359608e-01 -1.52096570e-01 -8.18174660e-01 1.58321753e-01
2.87593510e-02 3.47163439e-01 5.30214727e-01 -9.69232559e-01
-6.31655395e-01 -1.00153096e-01 -2.54188716e-01 2.48816490e-01
1.43904146e-03 2.05205917e-01 6.38460100e-01 -9.53604504e-02
1.91062525e-01 -8.61961663e-01 -3.21419090e-01 -7.61404037e-01
-1.77353188e-01 -3.84736449e-01 5.59762537e-01 -2.92163283e-01
1.33848476e+00 -1.84365213e+00 -4.54683900e-01 2.94456817e-03
2.53527164e-01 3.89016122e-02 2.33101144e-01 1.23831403e+00
-2.43849114e-01 1.06667292e+00 3.98064762e-01 9.52541176e-03
3.13382268e-01 5.07273376e-01 2.86674555e-02 2.51206756e-01
9.43330750e-02 1.15280068e+00 -1.09214199e+00 -5.20940363e-01
5.57267785e-01 4.60968643e-01 -3.84588838e-01 6.24593735e-01
9.30012882e-01 2.99172718e-02 -8.94021094e-02 1.33241999e+00
1.05377018e+00 -2.90812284e-01 -2.46628653e-02 3.44544262e-01
-6.44541264e-01 6.00222833e-02 -7.41804183e-01 9.49429095e-01
1.31349964e-02 1.99757457e-01 1.74998537e-01 -6.80728197e-01
3.94654661e-01 1.10545933e+00 4.45659161e-01 -9.24297392e-01
-1.69569388e-01 9.00310338e-01 1.03920102e-01 -9.62983370e-01
6.40755475e-01 -1.72503099e-01 7.69825354e-02 1.40852645e-01
-3.61913115e-01 -3.84120375e-01 5.52458428e-02 4.19574350e-01
1.00224769e+00 -1.33707494e-01 5.48685491e-01 -8.72328460e-01
6.88946426e-01 -2.01589778e-01 3.00597936e-01 1.23978651e+00
-2.90862411e-01 9.21478033e-01 3.19022417e-01 -5.37574828e-01
-5.72780192e-01 -1.20098352e+00 -4.80556756e-01 9.86326516e-01
-1.58388391e-01 -3.87671381e-01 -7.10155964e-01 -2.63155133e-01
-1.85071185e-01 4.93483543e-01 -5.93591571e-01 5.23159266e-01
-5.14705896e-01 -6.85349941e-01 7.65277922e-01 3.17972392e-01
6.49624616e-02 -1.49228525e+00 -5.70813596e-01 3.34754556e-01
-1.07132711e-01 -5.39811492e-01 -1.57590687e-01 6.12205923e-01
-1.19366693e+00 -1.01526415e+00 -8.75530541e-01 -1.03603637e+00
7.18889177e-01 2.44572356e-01 1.18463767e+00 2.84042627e-01
-2.87709564e-01 6.49904788e-01 -1.31030381e-02 -2.67939389e-01
-3.16364132e-02 -5.78750819e-02 5.01899779e-01 -1.07851994e+00
6.46393359e-01 -3.92214775e-01 -1.00622165e+00 5.35368681e-01
-3.85821253e-01 -2.04611585e-01 7.85597622e-01 1.04281425e+00
1.42742008e-01 -1.33182362e-01 2.41971418e-01 -1.16735983e+00
1.11498392e+00 -4.85333532e-01 1.30341025e-02 2.62960047e-01
-8.53362501e-01 -4.53584343e-01 2.27881029e-01 -1.33037493e-02
-6.33760452e-01 -2.51423597e-01 -4.14545417e-01 3.22794706e-01
-4.06866640e-01 3.34393084e-02 2.24873364e-01 -5.26000679e-01
5.21005154e-01 1.50258854e-01 -4.07556817e-02 1.21735156e-01
6.30975887e-02 3.75316888e-01 3.67786407e-01 -7.01224983e-01
5.19809723e-01 1.93325520e-01 -3.00995037e-02 -1.28791058e+00
2.66786069e-01 -1.64763972e-01 -6.60750508e-01 -2.69126564e-01
7.75121748e-01 -9.50198710e-01 -6.35705650e-01 5.21708906e-01
-9.54387486e-01 -1.36340424e-01 -5.25335252e-01 6.30965471e-01
-7.02474535e-01 2.94757709e-02 -4.17852998e-01 -1.17976546e+00
-4.78847146e-01 -1.38869667e+00 9.59160328e-01 3.85633826e-01
-2.05811903e-01 -1.19790471e+00 3.38975668e-01 8.99848714e-02
4.29175526e-01 7.39965662e-02 5.03531873e-01 -2.80004859e-01
-2.98304528e-01 -2.85470933e-01 1.59573574e-02 -1.65876746e-01
-3.09365243e-02 4.80770379e-01 -5.48391879e-01 -4.37962860e-01
2.44738594e-01 -4.86500829e-01 -7.51581416e-02 7.11018145e-01
4.09077644e-01 -6.02920949e-01 -7.12544680e-01 5.56552589e-01
1.27709007e+00 4.72497940e-01 5.88340998e-01 6.77552521e-01
9.28869769e-02 5.17871320e-01 4.48484391e-01 2.08113909e-01
1.07506551e-01 1.08483166e-01 1.14217520e-01 -2.17137560e-01
9.45900455e-02 -1.75644696e-01 4.44304377e-01 1.09980536e+00
-8.23837519e-01 -2.15223029e-01 -4.88283783e-01 6.68056428e-01
-1.69156313e+00 -1.31570101e+00 -6.70152605e-01 9.76623476e-01
3.78074080e-01 3.38884830e-01 1.58041969e-01 1.21232189e-01
5.47049284e-01 -4.76693302e-01 -1.27305150e-01 -9.16352808e-01
-3.82233024e-01 4.21136379e-01 7.14838982e-01 6.51479006e-01
-6.10602558e-01 5.08361042e-01 1.25505590e+01 5.49717486e-01
-5.10860123e-02 1.27314463e-01 3.14815581e-01 1.99914098e-01
-2.24020526e-01 1.62632689e-01 -1.04165065e+00 9.26906094e-02
1.64754760e+00 -4.36735392e-01 3.93212438e-01 6.44444764e-01
3.94769877e-01 -1.96429595e-01 -9.78620768e-01 5.21999180e-01
1.12709433e-01 -1.50397491e+00 -3.76861334e-01 9.64888573e-01
5.80093563e-01 -1.81895152e-01 6.81941509e-01 4.33209062e-01
7.39425182e-01 -1.14519978e+00 7.88427055e-01 3.18626672e-01
8.87266219e-01 -8.18826377e-01 1.07315242e+00 2.76328743e-01
-1.04355454e+00 1.15487054e-02 -5.90443075e-01 -9.22581375e-01
1.62244976e-01 2.02188566e-01 -5.61583102e-01 4.06869203e-01
6.19966745e-01 2.15250537e-01 -8.53338003e-01 1.45581341e+00
-4.29231405e-01 4.22687940e-02 -3.00976187e-01 -4.39922899e-01
4.83290881e-01 5.28384298e-02 3.26606274e-01 1.02295005e+00
2.89785862e-01 5.45581102e-01 -2.62208551e-01 2.92286038e-01
6.63503349e-01 5.84383309e-03 -1.38496268e+00 -8.68683681e-02
3.54620427e-01 9.69217539e-01 -3.99494410e-01 -3.56748998e-02
-6.11456513e-01 6.62115753e-01 -1.66942239e-01 5.75146377e-01
-2.92761147e-01 -8.89406264e-01 4.33969021e-01 -2.30117049e-02
-2.88469017e-01 -1.35338321e-01 -3.23372036e-01 -1.02459288e+00
-6.06416404e-01 -3.90122980e-01 7.32179061e-02 -8.63867998e-01
-1.79313409e+00 6.85844719e-01 -3.50327820e-01 -1.03178740e+00
-9.29034173e-01 -9.72984850e-01 -2.90242791e-01 6.67755604e-01
-6.59234166e-01 -1.30476952e+00 1.77773401e-01 4.65253919e-01
1.82401046e-01 -5.88519633e-01 1.09945750e+00 -1.42670095e-01
-3.18711132e-01 1.08882773e+00 3.51253241e-01 -3.05974871e-01
8.32965672e-01 -1.16271269e+00 4.23785180e-01 1.09189451e-02
-6.59259140e-01 1.28460062e+00 6.81860268e-01 -6.81785107e-01
-1.16639841e+00 -1.52893141e-01 1.38110602e+00 -9.15031374e-01
5.27874053e-01 -1.02810413e-01 1.46300867e-01 6.57190859e-01
9.07351732e-01 -5.09461224e-01 1.08935249e+00 -1.13616258e-01
4.93123859e-01 8.23448122e-01 -1.38805819e+00 5.99272728e-01
1.21222687e+00 -4.81854081e-01 -8.27020168e-01 3.28937650e-01
9.94227231e-01 -5.60786903e-01 -1.34066856e+00 6.99696779e-01
1.14132881e+00 -1.04883564e+00 1.54237401e+00 -1.16912270e+00
4.08816934e-01 1.70770600e-01 -2.18455836e-01 -6.52935803e-01
-5.24122715e-01 -8.07741940e-01 -3.97186369e-01 3.40213925e-01
8.64144087e-01 -1.16800272e+00 5.53630650e-01 1.39166510e+00
-3.58625412e-01 -4.51715857e-01 -8.63314152e-01 -1.06149745e+00
-4.66870656e-03 1.81816339e-01 3.85523647e-01 8.18638563e-01
7.09820271e-01 2.02331603e-01 -2.05201268e-01 -4.37674336e-02
6.59316301e-01 -4.87512946e-01 5.82342803e-01 -8.74011815e-01
2.49149442e-01 -3.85976404e-01 -2.94489831e-01 -2.90531307e-01
-3.26795012e-01 -4.87107635e-01 -8.59293997e-01 -1.53084815e+00
8.49664882e-02 1.97367743e-01 -2.33256146e-01 3.44385266e-01
6.77823246e-01 2.45789096e-01 -2.27382466e-01 1.17054820e-01
-1.86927155e-01 -9.06788632e-02 9.18564737e-01 1.10557920e-03
-2.17724726e-01 3.92412305e-01 -1.01716614e+00 9.60534811e-01
7.85404518e-02 -6.09580100e-01 -5.51487744e-01 1.64939865e-01
5.71885824e-01 3.57775450e-01 4.49405432e-01 -9.70509052e-01
9.87445056e-01 -4.32846308e-01 6.12257063e-01 -1.43385470e+00
-2.65485227e-01 -9.00021493e-01 3.61826897e-01 6.29400074e-01
4.08042461e-01 3.53796333e-01 2.26811334e-01 -9.13986787e-02
-2.01378226e-01 -6.44658387e-01 4.43001747e-01 -7.45609999e-01
-4.55086857e-01 -3.90635371e-01 -1.02313221e+00 -9.67486948e-03
7.88939476e-01 -4.61360693e-01 -6.47767961e-01 -1.95199717e-02
-1.44608676e+00 7.30553791e-02 8.33678424e-01 1.34732248e-02
6.98022604e-01 -1.44444084e+00 -1.11748077e-01 6.18802965e-01
-5.42432606e-01 -4.98194307e-01 2.20674574e-01 1.06156552e+00
-1.31528914e+00 1.26915324e+00 -5.21896303e-01 -2.05424115e-01
-8.96585643e-01 7.19095051e-01 3.15837234e-01 1.75098896e-01
-3.93049836e-01 7.00151443e-01 1.12239346e-02 -6.58594191e-01
1.24011397e-01 2.97559589e-01 -4.87763733e-01 -2.52128989e-01
9.22277749e-01 1.17918742e+00 1.32526740e-01 -3.91715407e-01
-5.58191121e-01 4.39446121e-01 2.08604142e-01 -6.12734079e-01
1.38662541e+00 -2.20989168e-01 -7.44191706e-01 5.00318348e-01
7.50003338e-01 -7.33495504e-02 -3.28966863e-02 1.27011871e+00
4.80256975e-03 -5.47239244e-01 -3.11336160e-01 -7.82959402e-01
-3.42039734e-01 5.93214929e-01 6.17733955e-01 4.59080935e-01
9.69997942e-01 -6.47576571e-01 7.70057857e-01 7.64313757e-01
7.83133745e-01 -1.22997761e+00 -6.36951804e-01 3.69483382e-01
9.39836800e-01 -9.30032432e-01 1.30400375e-01 -3.14221263e-01
-5.55534840e-01 1.08689797e+00 5.38173854e-01 -1.62142932e-01
1.17091513e+00 6.96656525e-01 -1.75611794e-01 -4.53250527e-01
-1.04549897e+00 9.83817875e-02 2.10727617e-01 1.32608259e+00
8.26061428e-01 1.41275764e-01 -1.40426433e+00 8.46834600e-01
-6.34404242e-01 4.76996809e-01 9.47268784e-01 1.55018985e+00
-5.12773991e-01 -1.41692734e+00 -4.41162825e-01 2.31142953e-01
-2.82854587e-01 -3.49930972e-01 -8.62966061e-01 1.09929097e+00
1.97083559e-02 1.37330377e+00 -4.48758155e-01 -4.63731140e-01
8.27405751e-01 3.19600284e-01 4.41484600e-01 -1.25142381e-01
-8.94688129e-01 6.48750722e-01 3.84649158e-01 -1.05672014e+00
-7.94483483e-01 -1.27049911e+00 -1.08188879e+00 -1.16740942e+00
-3.07217389e-01 4.65777248e-01 2.31674016e-01 3.19533348e-01
-2.09303293e-02 1.56313553e-03 6.67447627e-01 -1.00111675e+00
-1.97397545e-01 -6.07395887e-01 -1.20775104e+00 -5.56946039e-01
3.94425273e-01 -5.85869551e-01 -1.15223050e+00 -6.24734610e-02] | [-1.5392122268676758, 15.86922550201416] |
5948485c-e55f-48be-af22-e6cfb26f54ad | out-of-domain-human-mesh-reconstruction-via | 2111.04017 | null | https://arxiv.org/abs/2111.04017v1 | https://arxiv.org/pdf/2111.04017v1.pdf | Out-of-Domain Human Mesh Reconstruction via Dynamic Bilevel Online Adaptation | We consider a new problem of adapting a human mesh reconstruction model to out-of-domain streaming videos, where performance of existing SMPL-based models are significantly affected by the distribution shift represented by different camera parameters, bone lengths, backgrounds, and occlusions. We tackle this problem through online adaptation, gradually correcting the model bias during testing. There are two main challenges: First, the lack of 3D annotations increases the training difficulty and results in 3D ambiguities. Second, non-stationary data distribution makes it difficult to strike a balance between fitting regular frames and hard samples with severe occlusions or dramatic changes. To this end, we propose the Dynamic Bilevel Online Adaptation algorithm (DynaBOA). It first introduces the temporal constraints to compensate for the unavailable 3D annotations, and leverages a bilevel optimization procedure to address the conflicts between multi-objectives. DynaBOA provides additional 3D guidance by co-training with similar source examples retrieved efficiently despite the distribution shift. Furthermore, it can adaptively adjust the number of optimization steps on individual frames to fully fit hard samples and avoid overfitting regular frames. DynaBOA achieves state-of-the-art results on three out-of-domain human mesh reconstruction benchmarks. | ['Xiaokang Yang', 'Bingbing Ni', 'Yunbo Wang', 'Michelle Z. He', 'Jingwei Xu', 'Shanyan Guan'] | 2021-11-07 | null | null | null | null | ['3d-absolute-human-pose-estimation'] | ['computer-vision'] | [-1.50645303e-03 -5.72807752e-02 -2.63867438e-01 -2.18180895e-01
-9.62915003e-01 -3.27670366e-01 8.53615254e-02 -1.15605652e-01
-2.96003670e-01 4.88300949e-01 2.68519074e-01 1.90119058e-01
8.55538994e-02 -4.88712043e-01 -1.03488064e+00 -4.97111738e-01
3.61859798e-02 7.07526386e-01 7.51085758e-01 -2.06153587e-01
-1.80322886e-01 4.32877868e-01 -1.50579476e+00 2.93892503e-01
7.88994014e-01 9.24916863e-01 1.89382747e-01 6.41149998e-01
-1.42920524e-01 3.83324116e-01 -5.07726312e-01 -5.10441899e-01
5.30180395e-01 -2.26320714e-01 -5.39965451e-01 4.12943631e-01
8.05167437e-01 -5.32013416e-01 -4.08843756e-01 9.08406615e-01
8.18052232e-01 2.04954952e-01 4.12518054e-01 -1.05935669e+00
-3.01954418e-01 2.49809071e-01 -8.86474371e-01 2.49608576e-01
3.54524732e-01 4.55892175e-01 5.90754032e-01 -1.11218750e+00
9.58445728e-01 1.44457781e+00 9.96405780e-01 6.62876725e-01
-1.49564314e+00 -5.46235144e-01 4.79658902e-01 1.82770967e-01
-1.32268667e+00 -5.75089395e-01 8.30296755e-01 -6.08526289e-01
6.22111976e-01 1.06185682e-01 1.05205667e+00 1.29707015e+00
1.94858629e-02 7.32798398e-01 7.05758929e-01 -1.33899063e-01
2.13855028e-01 -2.78477550e-01 -3.97737116e-01 7.95737267e-01
-4.81247194e-02 7.55511373e-02 -7.79509008e-01 -1.43744215e-01
1.06113887e+00 -1.93721876e-01 -3.61721784e-01 -8.43692124e-01
-1.23767269e+00 4.98490065e-01 1.22291304e-01 -1.87844157e-01
-2.81004667e-01 9.77499411e-02 5.26714087e-01 1.70955211e-01
5.81541300e-01 2.46182531e-01 -7.86016107e-01 -1.51528880e-01
-9.85131741e-01 3.28169286e-01 5.36821485e-01 8.71812165e-01
7.04904497e-01 1.07712112e-01 -2.61816919e-01 1.15547013e+00
2.69614965e-01 5.12631893e-01 3.23454350e-01 -1.23454583e+00
5.02833307e-01 2.88564295e-01 -1.66860558e-02 -1.09696364e+00
-2.75345981e-01 -4.47024971e-01 -5.35393655e-01 2.50275671e-01
8.20796549e-01 6.89719617e-02 -9.53912318e-01 1.86869574e+00
9.56166983e-01 5.21727264e-01 -4.49511528e-01 1.06852341e+00
7.92017043e-01 3.99374545e-01 -1.18750103e-01 -2.03741372e-01
1.07453346e+00 -1.34559405e+00 -6.29215419e-01 -3.79846960e-01
1.25438705e-01 -8.42447400e-01 1.38065076e+00 2.68281311e-01
-1.41906786e+00 -5.77142358e-01 -7.94201612e-01 -8.44149664e-02
2.76953310e-01 -2.09558517e-01 1.83513209e-01 1.44690678e-01
-6.86541021e-01 5.99070072e-01 -1.11936235e+00 -1.31789491e-01
4.24662352e-01 2.75078654e-01 -3.70868474e-01 -2.71730423e-01
-7.95925140e-01 5.00585794e-01 9.33899358e-03 1.33543953e-01
-8.62075746e-01 -1.08250117e+00 -9.97698009e-01 -2.60548919e-01
7.44498014e-01 -9.40766752e-01 1.22785103e+00 -1.08215594e+00
-1.76792753e+00 9.87561822e-01 -1.84477255e-01 -5.64866811e-02
1.05237710e+00 -2.61590123e-01 -1.57657951e-01 1.52682409e-01
1.43897310e-01 7.02719390e-01 1.17572463e+00 -1.32230067e+00
-4.65289861e-01 -2.87767887e-01 -4.08127420e-02 2.30217025e-01
-1.12113141e-01 -2.90145159e-01 -1.15442383e+00 -1.14247668e+00
2.05138534e-01 -9.78424132e-01 -2.46051475e-01 5.28432131e-01
-1.47472933e-01 2.13579670e-01 6.31152391e-01 -6.27509415e-01
1.21214020e+00 -2.14238739e+00 5.13243437e-01 1.73638344e-01
1.06817484e-01 7.45116547e-03 -2.72734582e-01 -2.71288455e-01
1.15397342e-01 -1.75540388e-01 -2.13044465e-01 -7.34979212e-01
-2.03585163e-01 5.88491738e-01 -9.87587348e-02 4.25208658e-01
1.89005837e-01 8.17853332e-01 -1.05692959e+00 -7.40548372e-01
1.54511943e-01 5.79848230e-01 -1.07935381e+00 2.21913621e-01
-4.64227825e-01 8.93195570e-01 -1.71540752e-01 9.19826627e-01
6.94093645e-01 -5.01104474e-01 1.04644738e-01 -5.47350645e-01
2.07320943e-01 -2.27360800e-02 -1.53922343e+00 2.05230379e+00
-2.55257159e-01 3.59864563e-01 2.63958097e-01 -8.60599518e-01
5.14079630e-01 6.91987798e-02 8.57106745e-01 -6.56208217e-01
3.64527926e-02 3.47733706e-01 -2.68179655e-01 -8.19850564e-01
8.86074826e-02 -1.18197501e-02 4.00528014e-01 1.03574015e-01
1.33055151e-01 -4.29230891e-02 8.41737464e-02 -2.14060441e-01
9.66626048e-01 4.94102716e-01 5.55536039e-02 -1.42576247e-01
2.61753201e-01 -1.35956675e-01 1.08186698e+00 5.18049538e-01
-2.45989665e-01 1.08350050e+00 3.13763857e-01 -7.17820227e-01
-1.10870433e+00 -1.09908795e+00 -5.03668338e-02 1.03150380e+00
3.17957312e-01 -4.23575222e-01 -6.45916164e-01 -7.03333080e-01
1.84364781e-01 -3.70854442e-03 -7.03873694e-01 -1.33683264e-01
-9.83212471e-01 -6.39204562e-01 2.46019349e-01 4.91338581e-01
1.43433332e-01 -8.17970335e-01 -7.02625573e-01 3.88903707e-01
-5.11729777e-01 -1.38539660e+00 -9.32940602e-01 -9.37052518e-02
-9.97774661e-01 -1.11358953e+00 -9.06219363e-01 -4.99248028e-01
7.54417241e-01 1.55590609e-01 1.51310980e+00 2.93193281e-01
-4.64748412e-01 4.90477353e-01 -2.90757209e-01 -1.21190019e-01
-4.56695080e-01 1.73253268e-02 -2.47579329e-02 6.41416609e-02
-2.48808041e-01 -6.95150793e-01 -8.68160725e-01 7.15946198e-01
-7.81891823e-01 8.48672092e-02 1.18359044e-01 1.02200842e+00
1.21514964e+00 -2.15681210e-01 2.98220128e-01 -6.74457133e-01
-4.42286395e-02 -4.29974049e-01 -5.08221984e-01 2.11659685e-01
-3.79088134e-01 3.47642824e-02 4.49499190e-01 -1.00709844e+00
-7.97850430e-01 1.63034543e-01 -2.15377256e-01 -1.00050950e+00
2.27920339e-01 1.65207058e-01 -1.78796336e-01 -1.00364774e-01
6.07800543e-01 -1.23295121e-01 1.54240459e-01 -5.03916323e-01
6.01223521e-02 -1.48310382e-02 7.60877550e-01 -8.78629267e-01
6.86695635e-01 6.74123228e-01 -8.27495083e-02 -5.04813433e-01
-9.76529002e-01 -2.97275335e-01 -5.15388966e-01 -5.22682011e-01
7.71103561e-01 -9.63433802e-01 -3.23188454e-01 5.91806293e-01
-8.83040786e-01 -8.34843636e-01 -6.31444871e-01 2.96639532e-01
-6.41629755e-01 5.80354750e-01 -5.91261804e-01 -4.25702363e-01
-1.45621374e-01 -1.43811595e+00 1.21397793e+00 2.86297617e-03
-2.08124667e-01 -9.23106611e-01 1.10419013e-01 4.46073174e-01
1.34114474e-01 4.24824774e-01 7.04525411e-01 -5.09965457e-02
-4.96302962e-01 1.14340514e-01 1.69388548e-01 2.66238391e-01
3.66036780e-02 4.56463061e-02 -7.52682388e-01 -5.20159841e-01
-1.94351971e-01 -3.62692386e-01 5.68510413e-01 5.52430689e-01
1.32101536e+00 -2.05255359e-01 -1.40794039e-01 1.14124203e+00
9.62503612e-01 -2.67038167e-01 4.80152845e-01 5.12774706e-01
7.64820039e-01 4.35084701e-01 7.70149112e-01 6.21568203e-01
5.35884559e-01 1.11798704e+00 5.15028954e-01 -9.58809629e-02
-5.59371054e-01 -4.63321805e-01 2.28181064e-01 8.28909516e-01
4.57759090e-02 -7.92611390e-02 -8.51596951e-01 5.55386841e-01
-2.06083465e+00 -7.10822701e-01 1.55304939e-01 2.30607891e+00
1.08356929e+00 1.92961663e-01 3.42023075e-01 -1.69211969e-01
5.32671213e-01 -9.33761126e-04 -9.06426132e-01 3.33817989e-01
-2.11101830e-01 1.37906015e-01 4.06351388e-01 6.33539140e-01
-9.47301149e-01 7.95516074e-01 6.13745308e+00 9.15558040e-01
-1.15342522e+00 2.58253187e-01 6.97175980e-01 -5.03115594e-01
-3.07029039e-01 -7.67828822e-02 -6.29857779e-01 7.22047746e-01
2.72357583e-01 2.72563517e-01 6.55731261e-01 6.33937836e-01
2.46905312e-01 2.53029703e-03 -1.17569590e+00 1.24734676e+00
1.18818991e-01 -1.35809660e+00 -2.91487970e-03 -1.40238255e-01
8.01314771e-01 4.00226526e-02 -1.36943504e-01 1.22505873e-01
6.84856325e-02 -7.20691621e-01 1.13792670e+00 5.41208982e-01
7.64617503e-01 -4.37134892e-01 3.74342203e-01 3.00948054e-01
-1.21986771e+00 -5.84847927e-02 -2.24064812e-01 2.90907025e-01
5.32440305e-01 6.77251339e-01 -4.28038657e-01 2.85129458e-01
1.03034210e+00 6.69602692e-01 -4.55766320e-01 1.20016253e+00
4.04839404e-02 3.41464370e-01 -6.07838094e-01 6.84025645e-01
-3.03977013e-01 -1.13878381e-02 8.62902403e-01 1.06850600e+00
3.36147577e-01 -3.22443955e-02 4.97691780e-01 5.47780097e-01
-2.24407054e-02 6.37872294e-02 -4.42591421e-02 6.02385283e-01
5.70194721e-01 8.16882312e-01 -6.99842989e-01 -2.51244277e-01
-3.44765157e-01 9.36912596e-01 4.11325902e-01 4.15653944e-01
-9.65850949e-01 4.71583426e-01 6.47925019e-01 6.24940515e-01
3.87793362e-01 -1.79285213e-01 -1.85139477e-01 -1.41443980e+00
2.94357181e-01 -1.22954845e+00 5.73844433e-01 -5.71751356e-01
-1.39877963e+00 3.34558845e-01 9.90720540e-02 -1.32365942e+00
-6.08714446e-02 -2.99364984e-01 -2.18200073e-01 3.44929129e-01
-1.52807808e+00 -1.03392649e+00 -5.43083131e-01 6.20860934e-01
1.04582608e+00 -7.55071780e-03 4.20428425e-01 8.34375858e-01
-6.65375352e-01 8.41445029e-01 -2.34828413e-01 -1.55283734e-01
1.08407021e+00 -9.87855494e-01 3.29128057e-01 6.04272366e-01
-6.99647814e-02 1.71977669e-01 7.35566497e-01 -6.97016835e-01
-1.31730926e+00 -9.17457342e-01 3.34666908e-01 -3.85692000e-01
4.21766579e-01 -3.76289964e-01 -1.13599443e+00 5.56728959e-01
-4.57171142e-01 7.30885327e-01 4.28381085e-01 -1.89611822e-01
-4.59028929e-01 -1.27266318e-01 -1.09916747e+00 6.51173770e-01
1.52007198e+00 -1.69408277e-01 -1.07851557e-01 3.43987823e-01
7.69264936e-01 -1.18055356e+00 -1.01288867e+00 7.28822708e-01
7.39604652e-01 -8.69147122e-01 1.39103234e+00 -6.06934845e-01
5.85714579e-01 -3.71394366e-01 -2.89197445e-01 -1.10154212e+00
-2.22457990e-01 -8.18178773e-01 -6.57958269e-01 9.28386629e-01
1.63378298e-01 -2.44040772e-01 8.32808197e-01 6.67989910e-01
-1.21976726e-01 -1.10707474e+00 -1.10764349e+00 -7.25564003e-01
-9.45855454e-02 -5.08989334e-01 5.52092075e-01 8.81830335e-01
-3.96563560e-01 6.11332282e-02 -6.10461116e-01 1.15039445e-01
7.46918023e-01 -1.74903810e-01 1.22597909e+00 -1.02639663e+00
-7.27490783e-01 -2.72365391e-01 -2.51583397e-01 -1.44657779e+00
6.79626092e-02 -5.46478271e-01 5.27456217e-02 -1.05124402e+00
1.04155175e-01 -6.58958733e-01 3.57400551e-02 4.14698839e-01
-3.82335156e-01 2.91631013e-01 3.07801515e-01 3.70498985e-01
-6.26580894e-01 6.90162778e-01 1.58143067e+00 -8.81952494e-02
-3.17963988e-01 1.37774146e-03 -2.70970851e-01 9.28969681e-01
3.11160684e-01 -5.49405754e-01 -4.39422429e-01 -8.91140580e-01
3.70112479e-01 2.05464125e-01 6.32547915e-01 -9.05478597e-01
1.04454681e-01 -2.77317166e-01 3.34714025e-01 -5.29807568e-01
3.57543200e-01 -9.60599303e-01 4.26711053e-01 2.48596504e-01
-6.78281188e-02 2.92442977e-01 2.65448272e-01 7.63772786e-01
-2.08834130e-02 -6.77353740e-02 1.06248164e+00 -1.93771943e-01
-5.52365899e-01 6.01889551e-01 5.25941066e-02 5.84876001e-01
7.85054743e-01 -4.69501734e-01 2.12211519e-01 -2.18882114e-01
-1.02418911e+00 4.81950760e-01 8.66289020e-01 5.95650256e-01
5.82921803e-01 -1.35268295e+00 -6.27719104e-01 3.11724693e-01
1.10984617e-03 6.57317400e-01 5.79179406e-01 1.02097309e+00
-5.16290605e-01 -5.93391001e-01 1.87188461e-02 -9.91421640e-01
-1.20215285e+00 4.06994194e-01 5.85697651e-01 -3.10519367e-01
-1.07288766e+00 8.84762228e-01 2.18697637e-01 -5.24311006e-01
5.59941173e-01 -2.38366202e-01 2.38646358e-01 -5.13163246e-02
3.01505804e-01 5.57763934e-01 8.46519321e-02 -5.57130277e-01
-3.06832612e-01 9.72139299e-01 9.49779153e-02 1.44694913e-02
1.36698210e+00 -2.59802401e-01 2.38153547e-01 4.10281658e-01
1.08928883e+00 2.28762597e-01 -1.88837409e+00 -4.08902019e-01
-4.84525204e-01 -9.18942153e-01 -2.23584011e-01 -6.43041790e-01
-1.44675553e+00 5.35329401e-01 6.15751803e-01 -3.57392639e-01
9.17213321e-01 -8.89706332e-03 1.22365320e+00 1.90057810e-02
2.32516378e-01 -1.33759844e+00 5.21003187e-01 4.31777179e-01
9.05536592e-01 -1.31569934e+00 1.36903048e-01 -5.16575277e-01
-4.45165247e-01 1.10219538e+00 7.67908037e-01 2.41285879e-02
6.90233231e-01 4.46794748e-01 9.45009291e-02 -1.87316030e-01
-5.91243744e-01 1.91104457e-01 3.19374233e-01 5.55913568e-01
1.95487998e-02 -3.24476838e-01 5.46520874e-02 3.47909898e-01
-8.63142833e-02 8.02522674e-02 1.48916200e-01 7.91498005e-01
-7.25264624e-02 -9.12978351e-01 -5.22897363e-01 1.64212361e-01
-4.53347087e-01 3.00307661e-01 1.53507128e-01 7.38135755e-01
2.57552177e-01 4.87145990e-01 -2.32121740e-02 -1.68232217e-01
7.63457537e-01 -3.12915713e-01 5.87349772e-01 -3.84290248e-01
-2.37522244e-01 4.49309856e-01 -1.79843619e-01 -7.68479407e-01
-4.74103689e-01 -7.64029145e-01 -1.14602613e+00 -1.76685631e-01
-2.95234412e-01 -3.11976671e-01 2.06552982e-01 8.28183651e-01
6.00913405e-01 7.50609696e-01 4.31563556e-01 -1.23399997e+00
-6.62234604e-01 -4.37940925e-01 -1.39806410e-02 7.32908666e-01
5.09516656e-01 -1.09065020e+00 -2.65746832e-01 2.81546354e-01] | [7.219239234924316, -0.7991213202476501] |
196b60a6-a699-4432-bd7d-55e7a55b6b13 | high-resolution-image-synthesis-and-semantic | 1711.11585 | null | http://arxiv.org/abs/1711.11585v2 | http://arxiv.org/pdf/1711.11585v2.pdf | High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs | We present a new method for synthesizing high-resolution photo-realistic
images from semantic label maps using conditional generative adversarial
networks (conditional GANs). Conditional GANs have enabled a variety of
applications, but the results are often limited to low-resolution and still far
from realistic. In this work, we generate 2048x1024 visually appealing results
with a novel adversarial loss, as well as new multi-scale generator and
discriminator architectures. Furthermore, we extend our framework to
interactive visual manipulation with two additional features. First, we
incorporate object instance segmentation information, which enables object
manipulations such as removing/adding objects and changing the object category.
Second, we propose a method to generate diverse results given the same input,
allowing users to edit the object appearance interactively. Human opinion
studies demonstrate that our method significantly outperforms existing methods,
advancing both the quality and the resolution of deep image synthesis and
editing. | ['Jan Kautz', 'Jun-Yan Zhu', 'Ming-Yu Liu', 'Ting-Chun Wang', 'Andrew Tao', 'Bryan Catanzaro'] | 2017-11-30 | high-resolution-image-synthesis-and-semantic-1 | http://openaccess.thecvf.com/content_cvpr_2018/html/Wang_High-Resolution_Image_Synthesis_CVPR_2018_paper.html | http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_High-Resolution_Image_Synthesis_CVPR_2018_paper.pdf | cvpr-2018-6 | ['sketch-to-image-translation', 'fundus-to-angiography-generation'] | ['computer-vision', 'computer-vision'] | [ 6.67866647e-01 2.59984732e-01 3.60912859e-01 -2.23946944e-01
-8.39041770e-01 -8.41364264e-01 6.52867496e-01 -5.28358817e-01
-9.46739838e-02 1.02767169e+00 2.52135042e-02 2.42240489e-01
3.49377334e-01 -1.01291120e+00 -1.01323843e+00 -6.67591810e-01
4.74664479e-01 2.80644625e-01 3.59602809e-01 -1.76358119e-01
-6.24230951e-02 6.16716802e-01 -1.46525562e+00 3.22939217e-01
1.05419278e+00 8.72966349e-01 2.50143498e-01 6.79931164e-01
5.46105579e-02 7.11363077e-01 -9.69323039e-01 -6.22155249e-01
4.05978709e-01 -6.51800752e-01 -5.13615668e-01 2.08460063e-01
6.57432079e-01 -5.85061133e-01 -1.30922854e-01 1.00649548e+00
5.25392473e-01 3.30508128e-02 6.80265665e-01 -1.32445526e+00
-9.77362931e-01 2.84474850e-01 -6.20389581e-01 -3.61142546e-01
4.88871455e-01 5.76376975e-01 6.32103860e-01 -4.96868521e-01
7.98152566e-01 1.54643583e+00 4.30539906e-01 9.74501133e-01
-1.44610512e+00 -9.10696089e-01 8.57706070e-02 -1.71901569e-01
-1.23212886e+00 -3.16933304e-01 8.88499379e-01 -4.32933629e-01
2.57069588e-01 4.60972995e-01 5.99446774e-01 1.43549132e+00
1.39120847e-01 6.27537668e-01 1.42499840e+00 -2.39398673e-01
2.25293264e-01 1.50696427e-01 -8.45028341e-01 7.06491232e-01
7.70503730e-02 1.68548658e-01 -1.42924249e-01 3.99825089e-02
1.39964974e+00 -2.63568163e-02 -3.24483961e-01 -2.42195547e-01
-1.18981397e+00 8.17026138e-01 5.44444561e-01 -5.06046638e-02
-3.01744491e-01 5.75477779e-01 -1.12952227e-02 -8.08231309e-02
4.10253018e-01 6.14020169e-01 8.34807232e-02 2.59965241e-01
-7.77297139e-01 2.71588236e-01 3.71680379e-01 1.04552388e+00
6.78643107e-01 4.18741465e-01 -6.31301880e-01 7.81036556e-01
5.65878712e-02 8.04138601e-01 1.79441825e-01 -1.35334170e+00
1.08217478e-01 2.81713307e-01 2.54625022e-01 -8.66878152e-01
1.04687236e-01 -1.26867294e-01 -9.21033800e-01 8.67035091e-01
2.46054977e-01 -2.69888103e-01 -1.41637623e+00 1.79412580e+00
3.19424599e-01 4.87493910e-02 -6.74138516e-02 9.08934295e-01
9.99844909e-01 6.00765646e-01 2.32143834e-01 8.99791270e-02
1.14412320e+00 -1.06459045e+00 -8.70680332e-01 -2.56432086e-01
-2.49370247e-01 -7.45541036e-01 1.34712386e+00 3.18800509e-01
-1.41110861e+00 -6.33384287e-01 -9.54544425e-01 -1.42491013e-01
-2.22776502e-01 1.77691858e-02 7.03442276e-01 6.34381652e-01
-1.02962172e+00 3.36783528e-01 -6.50484860e-01 1.88724641e-02
6.80460334e-01 5.22217043e-02 -1.98088273e-01 -1.68674126e-01
-1.04122114e+00 6.36658370e-01 2.95514792e-01 -2.56891578e-01
-1.15570033e+00 -7.58713841e-01 -9.12329733e-01 5.30879498e-02
2.57677943e-01 -9.41060424e-01 9.76364315e-01 -1.20352423e+00
-1.91253293e+00 7.45084167e-01 1.36962205e-01 -2.43131265e-01
9.37940359e-01 3.42423804e-02 -2.33566985e-01 2.47721210e-01
8.47002044e-02 1.17716169e+00 1.09797132e+00 -1.81196594e+00
-3.59328151e-01 -3.88772003e-02 2.94020206e-01 2.31983706e-01
-3.38988937e-02 -1.51005372e-01 -5.45203269e-01 -1.13567543e+00
-2.79486030e-01 -8.85143638e-01 -3.44399631e-01 4.40590471e-01
-5.95257223e-01 4.06253189e-01 1.01919198e+00 -7.56605089e-01
5.66946208e-01 -2.34763551e+00 1.25653461e-01 -3.57707590e-02
3.07148039e-01 1.79117486e-01 -2.75787562e-01 2.01864596e-02
2.37456486e-01 4.46369618e-01 -5.14541328e-01 -4.55861121e-01
-1.74441021e-02 1.79702342e-01 -4.30274338e-01 1.02958903e-02
3.53770703e-01 1.40096319e+00 -8.21337819e-01 -5.17671525e-01
4.78041679e-01 8.11926663e-01 -5.29758751e-01 3.22750270e-01
-4.77859199e-01 8.61968338e-01 -2.17703819e-01 6.35598004e-01
8.48116279e-01 -2.43390173e-01 -8.07528570e-02 -2.24053130e-01
3.71298254e-01 -2.49979913e-01 -1.07186878e+00 1.56157219e+00
-5.60310364e-01 5.79753458e-01 1.15744807e-01 -3.32430542e-01
8.32685053e-01 8.94130319e-02 1.87057540e-01 -6.90099895e-01
-1.05249677e-02 -1.34730726e-01 -2.74602592e-01 -1.38898745e-01
4.48398829e-01 -2.79248148e-01 -8.58636573e-02 4.34711248e-01
-1.76037878e-01 -8.74004185e-01 3.53385732e-02 8.71359706e-02
7.74403274e-01 1.75924510e-01 1.24013901e-01 1.97822899e-01
1.61334157e-01 -2.64954329e-01 4.64411408e-01 6.65440381e-01
1.38984919e-01 1.23813748e+00 4.03488487e-01 -9.66400579e-02
-1.18292284e+00 -1.55232239e+00 -1.23000173e-02 6.44978702e-01
3.73589694e-01 1.36870608e-01 -9.05301392e-01 -7.27998734e-01
-3.56139615e-02 7.67271280e-01 -7.01327920e-01 -8.14953148e-02
-5.28472900e-01 -4.51102704e-01 5.67108631e-01 6.73452318e-01
7.15281308e-01 -1.36328721e+00 -5.11363626e-01 -8.08906276e-03
-1.65885404e-01 -1.32693136e+00 -6.09857202e-01 -3.54370505e-01
-6.62741899e-01 -7.46441483e-01 -9.43485260e-01 -7.08651006e-01
9.41514134e-01 -1.09662183e-01 1.14187086e+00 -1.55799448e-01
-5.69821417e-01 2.36089393e-01 -2.26250529e-01 -4.52466935e-01
-7.13879228e-01 -2.66980290e-01 -2.26441965e-01 -5.50334230e-02
-5.66581130e-01 -6.25186145e-01 -7.89864600e-01 3.95381153e-01
-1.23019898e+00 4.52953011e-01 7.53517628e-01 6.53884411e-01
7.95355916e-01 3.99758108e-02 6.69996738e-01 -1.07657588e+00
5.95309556e-01 5.65157235e-02 -5.88891447e-01 1.74332082e-01
-3.10423762e-01 -2.10206956e-02 8.25170338e-01 -6.64826393e-01
-1.32627094e+00 8.33822712e-02 -1.66521385e-01 -6.09451115e-01
-2.67886668e-01 -2.57049352e-01 -3.74491572e-01 -2.72402674e-01
6.83383465e-01 2.32100159e-01 2.42978800e-02 -1.46475166e-01
7.94055104e-01 2.18354970e-01 7.60174632e-01 -5.22482753e-01
1.05535150e+00 7.58998215e-01 -2.04838485e-01 -4.26884830e-01
-4.48830277e-01 3.38094562e-01 -3.71797234e-01 -1.35026008e-01
1.15705585e+00 -9.04714823e-01 -4.95978802e-01 6.01269186e-01
-1.01354635e+00 -7.22682059e-01 -6.53635561e-01 6.28008321e-02
-7.46004164e-01 7.49910474e-02 -6.67048752e-01 -4.59123611e-01
-3.79217327e-01 -1.27425587e+00 1.32350385e+00 3.22587401e-01
2.69584674e-02 -7.91507900e-01 -3.60178709e-01 4.25715774e-01
5.90607882e-01 9.86279309e-01 7.09534526e-01 1.40688300e-01
-9.32487309e-01 1.42539311e-02 -4.01030988e-01 4.83491898e-01
2.67846882e-01 2.08285321e-02 -8.86030257e-01 -3.64540100e-01
-3.75554055e-01 -4.44945633e-01 6.20096743e-01 3.79504442e-01
1.54842710e+00 -3.85650069e-01 -3.32087517e-01 8.29303801e-01
1.29255259e+00 2.88243294e-01 1.02215457e+00 -8.36257823e-03
9.69583929e-01 3.11577618e-01 4.67147857e-01 1.85837582e-01
1.04109943e-02 7.47318327e-01 4.86101985e-01 -5.73547006e-01
-6.52545750e-01 -3.91692668e-01 2.04257131e-01 2.07011595e-01
-1.25052333e-01 -3.78472209e-01 -3.70536059e-01 2.86016375e-01
-1.40830529e+00 -8.74543667e-01 2.74580508e-01 1.89481604e+00
9.70200598e-01 8.98793861e-02 -3.10494024e-02 -1.63699567e-01
7.48468578e-01 1.62955046e-01 -8.59214604e-01 -2.30716169e-01
-1.45810410e-01 5.61975241e-01 4.90398437e-01 4.42503065e-01
-7.87894964e-01 1.15448153e+00 6.68308544e+00 8.85039449e-01
-1.29152632e+00 1.53948471e-01 7.63354421e-01 -1.69410825e-01
-6.61175668e-01 -4.13454890e-01 -4.65638787e-01 5.28794408e-01
2.22834006e-01 -1.85722895e-02 4.23767835e-01 7.14678824e-01
2.24880222e-02 4.33628261e-02 -8.16675842e-01 9.68387246e-01
2.96943367e-01 -1.44311786e+00 4.88296181e-01 2.80134548e-02
1.20471704e+00 -5.35950482e-01 4.55715925e-01 1.12418632e-03
6.41962767e-01 -1.30173445e+00 8.38796377e-01 5.83288550e-01
1.52146983e+00 -9.05328333e-01 3.15426528e-01 -1.25763584e-02
-9.54222322e-01 1.49715677e-01 1.58516821e-02 4.89644676e-01
4.88271534e-01 4.03559059e-01 -6.15510941e-01 3.60579699e-01
7.23772407e-01 3.26417714e-01 -6.65751517e-01 6.85939431e-01
-6.64664745e-01 2.25842685e-01 -1.28149733e-01 2.24990308e-01
-8.07117000e-02 -1.12750575e-01 3.83820355e-01 1.00198054e+00
2.18637556e-01 1.64721102e-01 -2.48433091e-02 1.45586002e+00
-2.92672038e-01 -1.96280733e-01 -5.65405667e-01 5.32885268e-02
4.49809074e-01 1.26338053e+00 -7.60600686e-01 -2.86261320e-01
-3.33163403e-02 1.51570356e+00 6.09237105e-02 5.27175844e-01
-1.14153457e+00 -5.30139744e-01 6.31803632e-01 2.67203718e-01
3.47092986e-01 -1.37995288e-01 -4.80814219e-01 -1.01082528e+00
7.00863376e-02 -1.02201855e+00 -1.12006083e-01 -1.11684489e+00
-1.19838405e+00 6.30191743e-01 -1.38148561e-01 -1.07140911e+00
-2.57695109e-01 -3.76247287e-01 -5.32127500e-01 8.20196390e-01
-1.35004914e+00 -1.46138084e+00 -6.80661023e-01 5.09651363e-01
5.65876067e-01 5.90194948e-02 7.41854548e-01 2.55387187e-01
-1.21717297e-01 8.72657597e-01 -5.30435368e-02 9.58529264e-02
8.20555806e-01 -1.24881077e+00 6.38919652e-01 7.99007595e-01
2.87836846e-02 3.55094492e-01 5.62556565e-01 -5.53220749e-01
-9.24728572e-01 -1.37336648e+00 -1.00806609e-01 -5.06204247e-01
2.42435466e-02 -6.36352718e-01 -7.29396164e-01 7.03703582e-01
2.98858732e-01 8.41299519e-02 2.20769599e-01 -5.58617473e-01
-1.45549789e-01 -1.70155428e-02 -1.68861556e+00 9.36083198e-01
1.09173870e+00 -4.29512441e-01 -8.00837055e-02 2.24080652e-01
9.62179244e-01 -6.97427988e-01 -7.40396142e-01 6.39058769e-01
5.33426881e-01 -9.94304180e-01 1.26431334e+00 -8.08974579e-02
6.91795766e-01 -4.82675850e-01 6.77169189e-02 -1.45380104e+00
-2.50569165e-01 -5.94443202e-01 2.12484710e-02 1.19463122e+00
2.56646574e-01 -6.45907938e-01 7.31823385e-01 5.20576715e-01
2.63003521e-02 -6.15927577e-01 -5.21140397e-01 -6.55013263e-01
2.03192420e-02 -1.39700640e-02 7.54497170e-01 8.78048360e-01
-7.54011929e-01 5.08351065e-02 -5.86173058e-01 1.52316481e-01
7.54963696e-01 1.66916847e-01 9.35533941e-01 -7.98485398e-01
-4.32695061e-01 -4.41365629e-01 -3.96958351e-01 -8.89559269e-01
1.07085310e-01 -5.17181873e-01 1.33996591e-01 -1.72598362e+00
1.40456736e-01 -5.04088879e-01 1.50737956e-01 4.28045452e-01
-2.77482957e-01 9.52529430e-01 3.03711385e-01 4.70687710e-02
-2.80187517e-01 5.37532151e-01 1.93953872e+00 -2.27359831e-01
-2.90008392e-02 -1.68274045e-01 -6.89710736e-01 6.40396535e-01
6.94134474e-01 -2.02988878e-01 -6.24956369e-01 -4.85342056e-01
-1.39626145e-01 -8.12361091e-02 6.09417200e-01 -1.08969915e+00
-2.04862595e-01 -2.58261502e-01 8.51444721e-01 -1.89041242e-01
5.67604184e-01 -6.50299728e-01 4.66964602e-01 3.41096580e-01
-3.48592430e-01 -1.82806849e-01 3.51823479e-01 7.24002898e-01
-9.24928933e-02 1.48089185e-01 1.29722333e+00 -2.54302531e-01
-5.83229125e-01 3.71181577e-01 -4.24665362e-02 1.88576281e-01
1.34874809e+00 -1.38900399e-01 -3.39832604e-01 -5.87262273e-01
-7.69337714e-01 -5.54560162e-02 9.46872294e-01 6.18955672e-01
7.54070699e-01 -1.47880459e+00 -7.50388741e-01 3.24306518e-01
4.89023328e-03 3.19885910e-01 3.35786462e-01 2.21664056e-01
-7.42932379e-01 -8.83246288e-02 -5.45983374e-01 -5.45180500e-01
-1.12401354e+00 5.51778853e-01 2.46707454e-01 -4.38641980e-02
-7.66908467e-01 7.03953862e-01 6.74567282e-01 -4.18736011e-01
1.79738179e-02 -1.36858374e-01 5.18201431e-03 -2.90158153e-01
4.02207494e-01 6.39903247e-02 -2.59923220e-01 -3.22318017e-01
3.77504863e-02 5.57774603e-01 -5.76010980e-02 -2.58103043e-01
1.04440999e+00 -5.92009090e-02 9.82285365e-02 1.64985374e-01
8.26554179e-01 2.12092251e-01 -1.76256394e+00 1.25434864e-02
-7.90248454e-01 -8.03214431e-01 -2.87124097e-01 -1.12112939e+00
-1.39305592e+00 6.78924620e-01 6.42595589e-01 4.01349971e-03
1.16370082e+00 9.54553112e-02 9.21628177e-01 -2.42863938e-01
4.28317279e-01 -8.80697072e-01 4.63846236e-01 2.11601555e-02
1.10103941e+00 -1.19465125e+00 -6.46984950e-02 -6.83247447e-01
-8.58967662e-01 6.12887502e-01 8.72717977e-01 -2.78629899e-01
7.04844818e-02 5.24945319e-01 2.80981779e-01 4.72782291e-02
-3.46878588e-01 -1.32880986e-01 3.17413211e-01 8.89711380e-01
1.80288747e-01 1.56406596e-01 -1.52016580e-01 1.63138330e-01
-2.95404017e-01 -1.90709114e-01 5.71649969e-01 6.36995971e-01
-5.38817421e-02 -9.88646448e-01 -3.36211056e-01 3.20148975e-01
-5.21393239e-01 -1.05114855e-01 -3.19482833e-01 7.48770654e-01
2.15878740e-01 6.52317345e-01 2.99380757e-02 -2.10415959e-01
2.76282221e-01 -3.03597569e-01 6.76713943e-01 -6.47963226e-01
-3.04005593e-01 -4.47161235e-02 -1.96792394e-01 -5.55283427e-01
-2.50330150e-01 -3.87954146e-01 -1.09520769e+00 -1.52101800e-01
-1.28882110e-01 -2.32223287e-01 7.21452355e-01 5.85434198e-01
4.74123865e-01 9.47800279e-01 5.52737713e-01 -1.06611705e+00
-1.86334357e-01 -8.31005275e-01 -4.87311542e-01 6.37324512e-01
2.32184619e-01 -6.24313235e-01 -3.89573514e-01 4.25383955e-01] | [11.678472518920898, -0.5233043432235718] |
b34c71a4-32e0-4f74-90e7-f2b75ba872b4 | humor-as-circuits-in-semantic-networks | null | null | https://aclanthology.org/P12-2030 | https://aclanthology.org/P12-2030.pdf | Humor as Circuits in Semantic Networks | null | ['Igor Labutov', 'Hod Lipson'] | 2012-07-01 | null | null | null | acl-2012-7 | ['humor-detection'] | ['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] | [-7.202491283416748, 3.781613349914551] |
d51357fc-df87-432a-9f79-5262ee2b4bea | fault-localization-for-framework-conversions | 2306.06157 | null | https://arxiv.org/abs/2306.06157v1 | https://arxiv.org/pdf/2306.06157v1.pdf | Fault Localization for Framework Conversions of Image Recognition Models | When deploying Deep Neural Networks (DNNs), developers often convert models from one deep learning framework to another (e.g., TensorFlow to PyTorch). However, this process is error-prone and can impact target model accuracy. To identify the extent of such impact, we perform and briefly present a differential analysis against three DNNs used for image recognition (MobileNetV2, ResNet101, and InceptionV3), converted across four well-known deep learning frameworks (PyTorch, Keras, TensorFlow (TF), and TFLite), which revealed numerous model crashes and output label discrepancies of up to 100%. To mitigate such errors, we present a novel approach towards fault localization and repair of buggy deep learning framework conversions, focusing on pre-trained image recognition models. Our technique consists of four primary stages of analysis: 1) conversion tools, 2) model parameters, 3) model hyperparameters, and 4) graph representation. In addition, we propose a number of strategies towards fault repair of the faults detected. We implement our technique on top of Apache TVM deep learning compiler, and we test it by conducting a preliminary fault localization analysis for the conversion of InceptionV3, from TF to TFLite. Our approach detected that the tf2onnx tool used in the conversion process introduced precision errors to model weights for convolutional layers in particular, which negatively affected the model accuracy. We then repaired the target model by replacing the affected weights with those from source model. | ['Ajitha Rajan', 'José Cano', 'Perry Gibson', 'Nikolaos Louloudakis'] | 2023-06-10 | null | null | null | null | ['fault-localization'] | ['computer-code'] | [-2.07329229e-01 3.01968679e-03 4.86673489e-02 -9.92928371e-02
-3.43029261e-01 -7.68729329e-01 3.28643858e-01 -2.33951002e-01
-8.05103555e-02 2.14851916e-01 -3.97448450e-01 -1.18483388e+00
9.14564580e-02 -7.69906878e-01 -1.33956456e+00 -6.44921139e-02
4.51026373e-02 -2.11368278e-02 4.88975823e-01 3.54230171e-03
3.48445296e-01 3.82391632e-01 -1.45847011e+00 4.84471887e-01
6.70129061e-01 8.11286807e-01 -1.84656546e-01 1.03109503e+00
-1.63509667e-01 1.48562014e+00 -1.11380506e+00 -8.97638381e-01
1.28287375e-01 -3.67686823e-02 -1.05323744e+00 -2.98734039e-01
6.42326117e-01 -6.57097220e-01 -1.35085523e-01 1.21778595e+00
2.72022963e-01 -6.26481831e-01 2.09891662e-01 -1.79380345e+00
-3.13646972e-01 9.86411631e-01 -1.95012346e-01 1.98840484e-01
-1.95856079e-01 4.71668601e-01 5.46843767e-01 -6.93405926e-01
5.15971005e-01 1.17106020e+00 1.16244912e+00 5.00346899e-01
-1.03050947e+00 -7.16174781e-01 -4.03944135e-01 2.73503680e-02
-1.18253493e+00 -5.81745327e-01 3.61590356e-01 -8.22507441e-01
1.74025512e+00 9.98247564e-02 4.58412886e-01 1.49465811e+00
6.16775393e-01 7.27365315e-02 7.33440280e-01 -5.88899076e-01
4.53433424e-01 6.90035000e-02 5.41633248e-01 1.02955055e+00
6.88446343e-01 -9.58939716e-02 -2.43489921e-01 -5.09532928e-01
5.88500321e-01 -2.35317737e-01 -5.33435494e-02 2.46301085e-01
-6.00916088e-01 3.76542658e-01 1.45634055e-01 5.64311028e-01
-2.33646512e-01 9.37344909e-01 7.62547374e-01 5.63907623e-01
2.45310694e-01 3.88494939e-01 -7.94521213e-01 -5.42577922e-01
-7.40432084e-01 -1.37453124e-01 1.03345764e+00 1.12458050e+00
1.05825746e+00 3.40131402e-01 1.28725633e-01 5.63215315e-01
4.74791020e-01 -6.69862702e-02 6.21840477e-01 -7.24838555e-01
6.34789824e-01 9.87129748e-01 -4.63927507e-01 -7.80835211e-01
-2.49741286e-01 -3.33905101e-01 -3.11848462e-01 3.33338410e-01
1.66435480e-01 -4.45160687e-01 -9.56967890e-01 1.25845003e+00
-8.84266198e-03 2.73526698e-01 1.19098663e-01 4.21817183e-01
8.43176007e-01 2.26012886e-01 -4.87130582e-02 6.36023283e-01
1.10062206e+00 -1.07868195e+00 -2.04450428e-01 -3.43606830e-01
1.44431615e+00 -7.43701160e-01 1.14325345e+00 3.34173471e-01
-9.71348763e-01 -4.34224755e-01 -1.34848857e+00 8.72368887e-02
-2.91259706e-01 3.63147140e-01 5.52753747e-01 9.93584514e-01
-1.69497275e+00 9.20138538e-01 -1.14722288e+00 -3.40860993e-01
2.94268012e-01 3.31349701e-01 -4.33300883e-01 -1.22576123e-02
-7.73674071e-01 7.49341369e-01 3.60778391e-01 -1.11793339e-01
-1.33983755e+00 -8.72079253e-01 -6.40241563e-01 2.65798837e-01
1.52193988e-02 -6.22526109e-01 1.46447909e+00 -1.21116149e+00
-1.21022677e+00 6.27776325e-01 4.48227912e-01 -4.47483808e-01
3.92690897e-01 -1.41270250e-01 -3.98352951e-01 -1.80003375e-01
-1.51737511e-01 2.23107398e-01 7.51434445e-01 -9.59738195e-01
-3.70406002e-01 -1.30917523e-02 3.76245201e-01 -6.48489714e-01
-5.69136500e-01 3.70798498e-01 -3.44018549e-01 -2.14607850e-01
-4.69227344e-01 -8.46930921e-01 2.12652221e-01 -4.13383573e-01
-6.29762590e-01 8.77060220e-02 5.68450391e-01 -8.18800926e-01
1.20027959e+00 -2.28925037e+00 -3.05076450e-01 2.63774544e-01
5.00938714e-01 4.31686014e-01 -3.59756619e-01 3.61745566e-01
-4.33283716e-01 7.66272783e-01 3.74670103e-02 -3.00676733e-01
-9.44477469e-02 1.00995205e-01 -2.11834952e-01 2.45850429e-01
2.05899209e-01 6.21190250e-01 -6.29754543e-01 -1.21986814e-01
-1.38419881e-01 4.39221084e-01 -7.11411715e-01 2.83208787e-01
-2.40328163e-01 -5.05239308e-01 -6.18455559e-02 8.35473299e-01
6.73000097e-01 -8.00981149e-02 1.56085804e-01 -2.35288605e-01
-1.52589932e-01 6.98171794e-01 -7.36828387e-01 1.16180599e+00
-6.23351574e-01 8.07007909e-01 -1.34966895e-01 -6.17298126e-01
9.10863042e-01 3.20432127e-01 -5.84256910e-02 -5.04519165e-01
1.00196861e-01 4.02023077e-01 2.01950639e-01 -5.75622141e-01
5.42058170e-01 5.69042265e-01 1.53932542e-01 3.89980882e-01
5.58965087e-01 2.54562825e-01 1.34758487e-01 1.31724998e-01
2.02647567e+00 -2.99155731e-02 -1.95158184e-01 -2.39059746e-01
6.70589730e-02 1.75028801e-01 3.64317983e-01 7.66963065e-01
1.58761609e-02 5.19694209e-01 1.43248808e+00 -5.05845666e-01
-1.15932024e+00 -6.20678246e-01 2.97847420e-01 8.03707361e-01
-7.22430408e-01 -8.61699402e-01 -1.52226794e+00 -1.01249421e+00
-2.86714643e-01 9.65275288e-01 -5.99836469e-01 -7.92770684e-01
-2.87118554e-01 -6.10800982e-01 1.55663943e+00 6.33637190e-01
5.90995610e-01 -8.63015950e-01 -7.16283202e-01 1.28600046e-01
4.16838527e-01 -8.41311216e-01 -2.41400793e-01 4.25881952e-01
-9.20363903e-01 -1.48399699e+00 -9.82243270e-02 -6.06450737e-01
6.70243382e-01 -2.84079835e-02 1.47500241e+00 9.72088873e-01
-2.62550622e-01 3.77437860e-01 -4.43395913e-01 -4.47413884e-02
-1.06994998e+00 1.99795380e-01 -2.02457383e-01 -3.96183878e-01
2.70127207e-01 -4.80234951e-01 -6.90052286e-02 2.69188702e-01
-1.14591324e+00 -1.41036864e-02 4.41181242e-01 6.83379054e-01
-1.96724892e-01 1.48006260e-01 7.89553896e-02 -1.14442861e+00
7.55491138e-01 -6.20724380e-01 -1.14378786e+00 3.52660239e-01
-8.32982183e-01 -1.04369149e-01 6.27954721e-01 -3.36127758e-01
-7.10333645e-01 -2.20019132e-01 -5.60460448e-01 -7.44561672e-01
-4.20598201e-02 7.92549074e-01 -1.64144158e-01 -4.28775072e-01
1.27542460e+00 -2.19096914e-01 4.12229151e-02 -4.23477173e-01
-2.24242121e-01 6.70272052e-01 3.04154366e-01 -5.32211781e-01
3.86827976e-01 -3.44331801e-01 -4.24928993e-01 -5.19542396e-01
1.95549980e-01 2.65109062e-01 -1.34449139e-01 -2.43314952e-01
4.65192080e-01 -8.17533076e-01 -5.91465533e-01 1.02634680e+00
-1.64319706e+00 -9.34812129e-01 1.78823248e-01 9.16102454e-02
1.29540145e-01 2.02250201e-02 -9.55927193e-01 -3.38550389e-01
-4.00458217e-01 -1.67275202e+00 6.93509161e-01 -1.61289647e-02
-1.00619823e-01 -1.00966167e+00 1.05372824e-01 1.40326172e-01
8.54325771e-01 1.11378087e-02 1.34819686e+00 -7.92237580e-01
-4.16624218e-01 -9.36217234e-02 -2.73900360e-01 9.23172951e-01
-2.28816912e-01 9.39805925e-01 -1.05970967e+00 -2.22603157e-01
9.97343007e-03 -1.49059311e-01 4.00415659e-01 -9.36873034e-02
1.03713644e+00 -4.39939648e-01 -7.04963207e-02 7.20171392e-01
1.59443057e+00 1.64015949e-01 9.93226290e-01 7.14819252e-01
1.02950394e+00 1.26034021e-01 -1.91467553e-01 1.27686590e-01
3.40601265e-01 3.47155511e-01 9.00352895e-01 5.34840599e-02
-2.14069933e-01 2.07358524e-02 1.03775418e+00 6.34454429e-01
6.68937340e-02 -1.58313587e-01 -1.49505079e+00 3.70449781e-01
-1.42058933e+00 -2.30151042e-01 -5.12996495e-01 2.03757501e+00
6.81008279e-01 4.60542530e-01 -2.21755370e-01 1.55398861e-01
5.24683654e-01 -4.80959594e-01 -3.59479666e-01 -7.21685052e-01
4.05105501e-01 2.23451018e-01 6.34192050e-01 1.59581289e-01
-5.08167744e-01 1.00767851e+00 5.88008308e+00 4.17171150e-01
-1.52007818e+00 3.29990715e-01 4.96929377e-01 2.61307687e-01
-3.08487535e-01 4.00960654e-01 -7.36337662e-01 5.19299209e-01
1.57638705e+00 2.25688577e-01 5.32014728e-01 1.38103747e+00
-3.35028350e-01 -6.43537566e-02 -1.37235785e+00 6.24270976e-01
-1.86772019e-01 -1.61624765e+00 -1.07781187e-01 3.87744941e-02
3.99902523e-01 4.04441297e-01 -3.91634077e-01 5.98172545e-01
4.89037097e-01 -8.83200109e-01 1.07969403e+00 2.87322342e-01
8.50570858e-01 -8.08950603e-01 1.02138042e+00 4.38003205e-02
-5.79440773e-01 7.26530328e-02 -4.57255602e-01 -1.04742534e-01
-6.44844055e-01 8.35274279e-01 -1.22990811e+00 2.20794648e-01
1.09056342e+00 3.74652416e-01 -1.18693960e+00 9.19863105e-01
-1.92867026e-01 1.06298018e+00 5.15316464e-02 2.02632084e-01
-9.33502614e-02 3.06784540e-01 5.59364371e-02 1.16634095e+00
4.11661416e-01 -6.85079694e-01 -4.18592036e-01 1.10149002e+00
-4.56372738e-01 -4.05908614e-01 -5.14418781e-01 -3.19896728e-01
6.08169734e-01 1.36876023e+00 -8.63590300e-01 -4.29715991e-01
-5.70741296e-01 7.12343454e-01 4.94366288e-01 2.26984650e-01
-1.05793428e+00 -3.83660108e-01 8.01089466e-01 1.67454854e-01
2.28213556e-02 7.35386387e-02 -2.28659958e-01 -1.08585739e+00
2.58025527e-01 -1.29931056e+00 -9.24194884e-03 -8.01770806e-01
-8.16785991e-01 9.60014403e-01 -1.15199924e-01 -6.55866265e-01
-1.16492689e-01 -6.56051397e-01 -9.72726405e-01 7.93203294e-01
-9.97680187e-01 -7.89380610e-01 -3.19444835e-01 3.04872513e-01
-3.65010984e-02 -3.50326329e-01 7.04027772e-01 7.66081810e-01
-1.01696277e+00 9.41911221e-01 -2.31969863e-01 2.84073323e-01
4.72423285e-01 -1.12900233e+00 1.30857062e+00 1.20931697e+00
-3.92597049e-01 9.24752474e-01 3.46116513e-01 -8.29937220e-01
-1.62378240e+00 -1.43864512e+00 7.35702574e-01 -2.05445141e-01
9.41311955e-01 -6.12099111e-01 -1.07933986e+00 1.05221581e+00
2.43625998e-01 5.40943407e-02 3.27318370e-01 -1.75821126e-01
-5.77387929e-01 -5.00491001e-02 -1.14165235e+00 4.81600285e-01
7.00833619e-01 -6.43656373e-01 -2.68857311e-02 3.87043059e-01
9.05653238e-01 -6.82841539e-01 -9.44832742e-01 -9.85643361e-03
3.29767346e-01 -1.38859582e+00 4.76803869e-01 -4.15988326e-01
8.04780960e-01 -1.84620723e-01 -1.87297851e-01 -1.34678936e+00
-6.83772117e-02 -4.78431404e-01 -1.07622281e-01 1.66888583e+00
6.20453298e-01 -9.68303502e-01 5.13609588e-01 9.28839922e-01
-7.59952545e-01 -6.02715790e-01 -5.77814639e-01 -7.03661025e-01
5.94033301e-02 -7.40249217e-01 8.45768631e-01 9.04698074e-01
-3.53092611e-01 1.57127991e-01 7.84281492e-02 1.92330092e-01
7.59970322e-02 -8.65327656e-01 8.80115390e-01 -9.14079607e-01
-5.89531064e-01 -4.60930586e-01 -5.64829648e-01 -1.99834734e-01
1.81835458e-01 -8.55241895e-01 2.35480554e-02 -1.17631900e+00
8.82160366e-02 -4.97577250e-01 1.52032360e-01 1.21633410e+00
1.77632883e-01 2.53701825e-02 4.93841469e-02 2.12656647e-01
-1.33563459e-01 -3.20816487e-01 2.57475168e-01 -3.27761620e-02
1.57530725e-01 -2.97959507e-01 -6.15156829e-01 7.19069421e-01
6.81201696e-01 -8.27508032e-01 -1.56541824e-01 -1.01384950e+00
6.78686619e-01 -1.17095791e-01 7.69524038e-01 -1.51599038e+00
2.27872670e-01 4.68572229e-01 1.07629299e-01 -2.92741451e-02
-4.62131947e-01 -7.68851042e-01 6.77670896e-01 6.31748199e-01
-8.05638283e-02 6.59149647e-01 5.25188029e-01 -2.58969188e-01
-7.20808655e-02 -9.12048459e-01 4.88646001e-01 -1.84528474e-02
-6.99860275e-01 -8.28391463e-02 -6.54782355e-01 -1.79262131e-01
7.05026448e-01 8.93717781e-02 -1.04761875e+00 3.84857595e-01
-1.85649335e-01 -2.57429570e-01 9.94573712e-01 4.46171790e-01
4.82900769e-01 -9.00567532e-01 -2.63856441e-01 4.30340409e-01
8.70700404e-02 -1.30882189e-01 8.07550177e-02 7.48741448e-01
-1.24809134e+00 1.88407525e-02 -2.83220232e-01 -5.89919269e-01
-1.08155060e+00 3.70603561e-01 8.09897840e-01 -1.07607201e-01
-4.08046216e-01 7.71600723e-01 -1.69673815e-01 -6.29526138e-01
2.02199951e-01 -8.91202807e-01 2.75435925e-01 -3.87953758e-01
3.74608099e-01 5.51544011e-01 1.09460330e+00 -2.55412310e-01
-5.84823608e-01 5.42150922e-02 -1.44657671e-01 1.92428485e-01
1.27890098e+00 6.09099329e-01 -6.69852614e-01 2.00813860e-01
1.43229330e+00 -3.46409380e-01 -1.04621708e+00 4.23307627e-01
1.99416608e-01 4.27630953e-02 2.44599611e-01 -8.82242978e-01
-1.52079320e+00 8.52949500e-01 5.68287432e-01 3.58967006e-01
9.45146680e-01 -2.85684824e-01 5.14512956e-01 3.66856039e-01
4.30503786e-01 -6.73999727e-01 -8.48552659e-02 7.20011413e-01
5.53593993e-01 -6.65178061e-01 -4.22267556e-01 -1.87338307e-01
-1.90940648e-02 1.48681653e+00 9.99292672e-01 -7.02939183e-02
4.05460447e-01 7.99337029e-01 5.32436483e-02 -3.71445566e-01
-1.09759390e+00 6.01517081e-01 -2.52891451e-01 4.62788612e-01
6.42975330e-01 -1.55586749e-01 2.36158088e-01 7.67290950e-01
-1.57948375e-01 3.33736002e-01 1.21083260e+00 1.18460548e+00
2.61044372e-02 -1.01939631e+00 -5.05313933e-01 6.65572822e-01
-6.16780460e-01 -4.86656249e-01 -5.88148534e-01 8.79014909e-01
1.37714237e-01 7.86495864e-01 3.97980884e-02 -1.04533732e+00
5.14435172e-01 1.32555053e-01 3.28947604e-01 -5.56922793e-01
-1.33718252e+00 -3.21496129e-01 3.94358039e-01 -7.62900472e-01
3.75922203e-01 -3.22524607e-01 -1.06886339e+00 -6.24538660e-01
-4.22488213e-01 -1.23487920e-01 1.15567279e+00 8.88825238e-01
8.02037716e-01 1.07574451e+00 2.36039132e-01 -7.25476623e-01
-5.00228584e-01 -9.65361774e-01 -2.51530230e-01 -1.64474249e-02
1.99303091e-01 -4.98298734e-01 -5.99658132e-01 1.57009289e-01] | [7.521404266357422, 7.701530933380127] |
a3634ba7-fccf-490f-ab2b-03218b601ecc | hierarchical-temporal-transformer-for-3d-hand | 2209.09484 | null | https://arxiv.org/abs/2209.09484v4 | https://arxiv.org/pdf/2209.09484v4.pdf | Hierarchical Temporal Transformer for 3D Hand Pose Estimation and Action Recognition from Egocentric RGB Videos | Understanding dynamic hand motions and actions from egocentric RGB videos is a fundamental yet challenging task due to self-occlusion and ambiguity. To address occlusion and ambiguity, we develop a transformer-based framework to exploit temporal information for robust estimation. Noticing the different temporal granularity of and the semantic correlation between hand pose estimation and action recognition, we build a network hierarchy with two cascaded transformer encoders, where the first one exploits the short-term temporal cue for hand pose estimation, and the latter aggregates per-frame pose and object information over a longer time span to recognize the action. Our approach achieves competitive results on two first-person hand action benchmarks, namely FPHA and H2O. Extensive ablation studies verify our design choices. | ['Wenping Wang', 'Taku Komura', 'Jia Pan', 'Lei Yang', 'Hao Pan', 'Yilin Wen'] | 2022-09-20 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Wen_Hierarchical_Temporal_Transformer_for_3D_Hand_Pose_Estimation_and_Action_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Wen_Hierarchical_Temporal_Transformer_for_3D_Hand_Pose_Estimation_and_Action_CVPR_2023_paper.pdf | cvpr-2023-1 | ['3d-hand-pose-estimation', '3d-hand-pose-estimation'] | ['computer-vision', 'graphs'] | [ 4.44411263e-02 -1.86368570e-01 -4.42748755e-01 -2.65895456e-01
-6.80637658e-01 -4.95391548e-01 2.98696250e-01 -7.55319118e-01
-3.43408495e-01 6.51775777e-01 7.54773974e-01 2.42205665e-01
3.17237228e-02 -2.67246038e-01 -6.53822899e-01 -7.48377085e-01
-1.89639013e-02 3.56485158e-01 4.94084984e-01 9.27339122e-02
-5.48251681e-02 5.01878619e-01 -1.38685763e+00 5.23361206e-01
2.30787441e-01 1.13691866e+00 1.52558878e-01 8.81858349e-01
2.42176861e-01 1.19915104e+00 -2.49095038e-01 -3.12107682e-01
3.59542727e-01 -2.34784856e-01 -1.08700204e+00 4.49157804e-01
7.75640488e-01 -1.10470808e+00 -1.17842591e+00 6.59861386e-01
8.92329097e-01 2.70821124e-01 4.19645458e-01 -1.25967860e+00
-3.07243884e-01 4.07518774e-01 -5.18478692e-01 3.38130832e-01
6.42033041e-01 7.55988240e-01 1.12514031e+00 -7.70903826e-01
8.97851944e-01 1.54608691e+00 5.77583551e-01 5.49416602e-01
-9.66129303e-01 -5.18455863e-01 6.78367436e-01 5.15885353e-01
-1.36740005e+00 -5.92342079e-01 6.59500957e-01 -5.75012028e-01
1.15463006e+00 -1.07067883e-01 9.19964850e-01 1.87778354e+00
8.98072775e-03 1.53459466e+00 8.40717137e-01 -3.05737890e-02
-1.82482805e-02 -6.07896507e-01 -1.23853814e-02 6.67246342e-01
-1.22666404e-01 1.00776307e-01 -8.84460866e-01 2.56996453e-01
1.17451465e+00 2.32570305e-01 -4.06931341e-01 -6.08895004e-01
-1.33993196e+00 4.04512703e-01 4.85184640e-01 7.65312612e-02
-6.05601490e-01 6.50609016e-01 3.88924778e-01 4.62002642e-02
1.39774561e-01 -1.97019381e-03 -6.86778605e-01 -5.22348225e-01
-6.44174993e-01 3.42660367e-01 4.73911613e-01 1.12102461e+00
2.23085880e-01 -2.09031850e-02 -7.51764715e-01 3.35735708e-01
1.79007649e-01 4.44602400e-01 1.72812715e-01 -1.10189092e+00
8.67678702e-01 2.59649485e-01 3.28999281e-01 -6.93186224e-01
-4.78594989e-01 -3.26429278e-01 -2.33518720e-01 -1.01319492e-01
9.89396930e-01 7.18864650e-02 -1.13619316e+00 1.90589046e+00
3.55336279e-01 -2.09907405e-02 -3.42960656e-01 1.20082247e+00
5.04844427e-01 3.40367071e-02 4.62324291e-01 7.14318380e-02
1.39280176e+00 -1.15774703e+00 -8.41531157e-01 -3.30283999e-01
3.55136484e-01 -4.14836377e-01 1.03433430e+00 1.74908876e-01
-9.07031298e-01 -4.57225770e-01 -6.26256108e-01 -4.63365614e-01
-4.14498597e-02 3.10998261e-01 8.93302679e-01 3.06302756e-01
-6.81080818e-01 6.55932486e-01 -1.29327476e+00 -4.53693926e-01
7.21768320e-01 2.83887863e-01 -4.24845994e-01 -3.09650656e-02
-9.83260512e-01 5.75371325e-01 3.03285152e-01 2.33127981e-01
-9.34972882e-01 -4.42716807e-01 -8.44529986e-01 4.50697914e-02
5.93931854e-01 -7.53170729e-01 1.25879264e+00 -6.66836977e-01
-1.55498600e+00 6.18405640e-01 -3.75902593e-01 -2.52942294e-01
1.09211338e+00 -5.65900147e-01 -3.50473337e-02 4.02635783e-01
1.59731463e-01 7.66021430e-01 1.01315224e+00 -7.18442619e-01
-6.76139176e-01 -7.38496661e-01 4.22550768e-01 3.83909225e-01
-3.10342073e-01 -2.54387379e-01 -7.88487732e-01 -9.29202557e-01
3.35173875e-01 -1.09147477e+00 4.04179059e-02 2.68860221e-01
-4.16120738e-01 -4.57648307e-01 8.05827379e-01 -9.81460869e-01
9.53333199e-01 -1.87627888e+00 6.20522141e-01 -4.77392506e-03
1.45997077e-01 -1.34012043e-01 -2.61323512e-01 6.18151575e-02
-6.52744398e-02 -4.29061055e-01 1.84787959e-01 -3.74319643e-01
6.50148988e-02 2.00472549e-01 -4.39499199e-01 4.32290018e-01
1.87319323e-01 1.33919048e+00 -1.02588260e+00 -5.52643776e-01
4.17442381e-01 7.76564956e-01 -7.34988272e-01 2.35885933e-01
-3.48538727e-01 7.84668505e-01 -6.59576774e-01 1.04164302e+00
1.52562588e-01 -3.96478772e-01 2.09711000e-01 -7.83555746e-01
1.99540481e-01 4.17186052e-01 -1.27825332e+00 2.23503065e+00
-3.41874093e-01 5.25519669e-01 -2.94625521e-01 -3.68321270e-01
2.43784301e-02 5.97598553e-01 8.79859984e-01 -5.84899008e-01
3.34592611e-01 3.24530015e-03 -1.75526708e-01 -7.13049352e-01
2.02350304e-01 3.34302604e-01 3.39938402e-01 4.47095484e-01
2.47517183e-01 3.47956568e-01 1.16424784e-01 -2.58926507e-02
1.21036041e+00 8.69252443e-01 1.81064501e-01 1.52344540e-01
2.64342815e-01 -3.65846336e-01 5.62507272e-01 6.35068297e-01
-7.98560917e-01 7.44340062e-01 5.23862362e-01 -7.97484636e-01
-9.12037551e-01 -1.14131546e+00 2.07353562e-01 1.21205115e+00
9.12821013e-03 -2.83682317e-01 -5.22688746e-01 -1.07773495e+00
1.53982848e-01 2.30601072e-01 -8.46767902e-01 2.70661209e-02
-8.79672408e-01 -2.39946797e-01 3.22600007e-01 1.36153340e+00
7.06101000e-01 -1.16877508e+00 -7.17508733e-01 1.79138586e-01
-6.45448685e-01 -1.53658319e+00 -9.95676458e-01 -9.31978002e-02
-5.95179975e-01 -1.28400588e+00 -9.45806623e-01 -4.01429832e-01
2.40859360e-01 2.86097169e-01 9.75609183e-01 -3.48399967e-01
-4.67761099e-01 7.43391871e-01 -3.84090126e-01 6.62917197e-02
3.43908519e-01 2.77863175e-01 2.49951422e-01 8.45046341e-02
4.08325106e-01 -7.74385333e-01 -1.04957128e+00 2.87478089e-01
-2.80747086e-01 -2.52631038e-01 6.57772362e-01 6.79091513e-01
3.65845948e-01 -4.94578570e-01 1.11475654e-01 -1.72713950e-01
-2.26155035e-02 2.02302426e-01 -2.20142335e-01 3.81963193e-01
-5.36064357e-02 1.96851864e-01 -4.62045595e-02 -5.68481565e-01
-1.10060191e+00 7.31029093e-01 -1.23813681e-01 -7.80259073e-01
7.61321858e-02 -2.32248217e-01 -5.82339108e-01 4.12698872e-02
4.10936803e-01 1.74227342e-01 -2.90515870e-01 -5.21133244e-01
4.67061669e-01 4.13631141e-01 7.74680853e-01 -6.08659029e-01
5.14637709e-01 8.54202211e-01 -1.99798226e-01 -4.40489948e-01
-1.19709015e+00 -4.07497853e-01 -1.12140906e+00 -3.78954411e-01
1.34919059e+00 -1.10967994e+00 -1.16926897e+00 6.27552450e-01
-1.52046525e+00 -4.19833809e-01 -2.16967031e-01 6.79341316e-01
-9.43797231e-01 2.98483700e-01 -7.32954800e-01 -7.15520978e-01
-2.07252339e-01 -1.32274628e+00 1.61275327e+00 -1.39472321e-01
-3.66845131e-01 -5.43751597e-01 -1.27200842e-01 4.01630610e-01
-7.36264735e-02 2.39980504e-01 3.57211351e-01 -4.03953977e-02
-9.47592378e-01 -2.16500893e-01 -3.98327768e-01 7.97997564e-02
3.27329785e-01 -5.02111018e-01 -1.15242612e+00 -4.47601020e-01
-3.18397790e-01 -5.59240401e-01 9.64466929e-01 5.22428274e-01
1.36166549e+00 -1.80257350e-01 -2.81440347e-01 7.98325062e-01
7.90498912e-01 -2.18687542e-02 7.73101985e-01 2.17498377e-01
1.23127520e+00 5.36050439e-01 4.71811771e-01 5.09781599e-01
5.03078461e-01 1.04557490e+00 2.66215175e-01 3.12587678e-01
-4.87474203e-01 -4.11451578e-01 3.50509942e-01 2.66182035e-01
-5.77495575e-01 -3.63512784e-02 -6.72612309e-01 3.47380906e-01
-1.97422659e+00 -1.17597497e+00 4.31300402e-01 1.94110548e+00
7.07005322e-01 6.23293184e-02 4.77197409e-01 5.15209958e-02
3.55091363e-01 4.91251588e-01 -9.32884812e-01 4.86222446e-01
-1.21139504e-01 5.83351366e-02 5.00485480e-01 5.11608124e-01
-1.45733774e+00 1.29203749e+00 6.48694515e+00 4.95555967e-01
-9.38310266e-01 1.10766441e-01 1.20738961e-01 -5.55247664e-01
2.71329075e-01 -2.26086229e-01 -6.77797377e-01 2.45113760e-01
9.97989327e-02 4.26755369e-01 4.38701928e-01 8.11958432e-01
1.16722979e-01 1.61199927e-01 -1.49771869e+00 1.13113916e+00
-1.89544130e-02 -8.85713518e-01 1.33236066e-01 -6.36965334e-02
4.17494833e-01 -5.52497730e-02 -1.43877305e-02 1.60639003e-01
2.96825916e-01 -8.68618548e-01 8.84911776e-01 4.98352408e-01
7.16802061e-01 -4.48009044e-01 3.26480120e-01 -9.52429324e-02
-1.77014577e+00 -3.47126901e-01 2.20603585e-01 -7.18754902e-02
2.60179311e-01 -1.19923569e-01 -3.13235879e-01 1.90507799e-01
8.99619162e-01 1.16919219e+00 -5.46790063e-01 5.56428671e-01
-5.51488459e-01 1.01588428e-01 -2.14934841e-01 4.31071639e-01
1.33784190e-01 3.11628640e-01 6.81703627e-01 1.00032866e+00
-8.28786790e-02 3.86316121e-01 2.96894014e-01 6.43018305e-01
2.19287321e-01 -4.48886782e-01 -2.52085418e-01 -7.57970586e-02
2.85137981e-01 8.29706073e-01 -5.69197416e-01 -3.07530880e-01
-3.46277773e-01 1.53776705e+00 4.13819253e-01 5.90642273e-01
-8.69743526e-01 -4.73174714e-02 1.02059555e+00 1.40884072e-02
6.46998465e-01 -5.39250910e-01 -1.59325704e-01 -1.53487730e+00
5.61101198e-01 -9.32839513e-01 5.91674626e-01 -7.15929389e-01
-1.05008328e+00 3.96334499e-01 -1.15575649e-01 -1.34392416e+00
-5.04315853e-01 -8.69914055e-01 -1.09674715e-01 6.00811541e-01
-1.14655018e+00 -1.48624313e+00 -6.59540474e-01 8.12974274e-01
8.55792940e-01 1.43687741e-03 5.96571863e-01 3.11465114e-01
-6.21547997e-01 7.58841872e-01 -5.97546637e-01 4.77925003e-01
7.26635993e-01 -1.20954609e+00 4.77073669e-01 7.34328389e-01
1.62801340e-01 4.73132223e-01 5.45950830e-01 -6.15040362e-01
-1.51930726e+00 -8.93002272e-01 7.03734815e-01 -1.23221338e+00
5.17835736e-01 -4.44825828e-01 -5.13374984e-01 1.07859361e+00
-2.60011077e-01 2.33025849e-01 1.43551573e-01 8.21552649e-02
-8.89538825e-01 1.67613029e-02 -7.73814857e-01 7.29828298e-01
1.83621049e+00 -9.45581138e-01 -5.44535875e-01 5.30931234e-01
6.82546735e-01 -6.38356149e-01 -7.46458292e-01 3.52728873e-01
1.31785870e+00 -8.77911806e-01 1.29109788e+00 -1.04799438e+00
5.42491972e-01 -1.00352265e-01 -4.14159507e-01 -7.35413730e-01
-3.49395692e-01 -5.04730284e-01 -5.46703756e-01 8.39512527e-01
-2.77608663e-01 -2.39841491e-01 1.09304392e+00 8.05808663e-01
4.01529223e-01 -5.63892484e-01 -1.01422977e+00 -1.04737842e+00
-1.99925601e-01 -4.47774291e-01 4.39882696e-01 5.67921937e-01
-7.01511428e-02 1.37527719e-01 -6.95878625e-01 8.43451098e-02
7.79893398e-01 9.54272076e-02 8.76822233e-01 -8.62436175e-01
-4.40423459e-01 -4.38451648e-01 -6.26075387e-01 -1.58927679e+00
3.25506806e-01 -3.91383737e-01 1.33776307e-01 -1.18515050e+00
2.47429416e-01 1.01045415e-01 -3.19155842e-01 6.33072853e-01
-2.60478228e-01 3.23613435e-01 3.14636528e-01 4.54639971e-01
-5.92818916e-01 6.93323672e-01 1.39938891e+00 -1.86857536e-01
-3.25146019e-01 2.53636413e-03 -8.26743767e-02 8.17318022e-01
2.58738965e-01 -2.62599438e-01 -4.33201283e-01 -7.17041731e-01
-2.41154835e-01 2.30503321e-01 8.84357989e-01 -9.90943313e-01
1.30758822e-01 -2.49125376e-01 5.94421387e-01 -7.59113431e-01
5.49139023e-01 -7.59836495e-01 -2.98258603e-01 5.07447302e-01
-4.96081591e-01 2.78658536e-03 -8.37745070e-02 7.96142876e-01
7.36806691e-02 6.17142558e-01 6.07104301e-01 -1.32166743e-01
-1.05783057e+00 8.69360447e-01 1.31762192e-01 6.47906661e-02
8.31815243e-01 -3.02049845e-01 -5.26094884e-02 -5.80905020e-01
-1.01051116e+00 2.41807610e-01 2.79418647e-01 8.68196368e-01
4.05579656e-01 -1.56461489e+00 -4.34836268e-01 -1.14089465e-02
2.56733835e-01 -1.01926051e-01 4.32092518e-01 1.12971926e+00
-1.11422025e-01 6.22879207e-01 -4.44418162e-01 -7.47110784e-01
-1.16319239e+00 5.16798913e-01 5.38886130e-01 -1.59107193e-01
-1.07986152e+00 1.07616210e+00 4.84040022e-01 7.30910003e-02
8.44769418e-01 -2.47142509e-01 6.11141697e-02 4.72015003e-03
6.59270227e-01 4.31074023e-01 -2.32250601e-01 -7.58883297e-01
-6.11846328e-01 7.26454854e-01 -3.89556289e-02 -1.92720830e-01
1.05819905e+00 -2.01873124e-01 2.39174202e-01 3.24510306e-01
1.25262702e+00 -4.58224535e-01 -2.06732702e+00 -4.56922472e-01
-5.37860952e-02 -8.59476745e-01 -1.07427910e-01 -7.25137532e-01
-1.26029623e+00 8.10391665e-01 6.75956905e-01 -6.52794540e-01
8.74108851e-01 1.39677197e-01 9.53994036e-01 6.82267666e-01
4.82350677e-01 -1.30014980e+00 6.00927830e-01 5.31734407e-01
1.17544019e+00 -1.42710662e+00 3.60449739e-02 -5.71572006e-01
-6.23248756e-01 9.81015384e-01 9.24850285e-01 7.22221956e-02
3.65995288e-01 8.25686753e-02 -3.00388597e-02 -1.37477264e-01
-4.95399415e-01 -5.65608144e-01 4.34813350e-01 6.27877772e-01
2.84869075e-01 -8.30350444e-02 2.24435665e-02 4.31058228e-01
-6.69474825e-02 3.04623812e-01 -1.69626877e-01 1.03494787e+00
3.27025056e-02 -8.91944706e-01 -8.14352855e-02 1.90457001e-01
-1.59659430e-01 2.78564841e-01 -4.82510328e-01 8.70940447e-01
4.97513264e-02 5.46669781e-01 1.09217823e-01 -5.48467278e-01
5.59746027e-01 2.77608871e-01 1.06871879e+00 -2.49989316e-01
-1.89950109e-01 5.85548952e-02 -5.45673706e-02 -1.32014000e+00
-6.31620109e-01 -9.14100409e-01 -9.63482201e-01 -8.03640708e-02
6.34026900e-02 -5.76640666e-01 1.07537299e-01 1.27171755e+00
3.03404063e-01 6.80509984e-01 2.80711800e-01 -1.28907633e+00
-8.38776350e-01 -8.98658931e-01 -5.30893743e-01 5.86276948e-01
5.03808975e-01 -1.23562956e+00 -1.83038414e-01 9.91296172e-02] | [7.720499515533447, 0.11806896328926086] |
7bb2941b-14f9-4ac6-978c-330298435456 | progressive-modality-reinforcement-for-human | null | null | http://openaccess.thecvf.com//content/CVPR2021/html/Lv_Progressive_Modality_Reinforcement_for_Human_Multimodal_Emotion_Recognition_From_Unaligned_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Lv_Progressive_Modality_Reinforcement_for_Human_Multimodal_Emotion_Recognition_From_Unaligned_CVPR_2021_paper.pdf | Progressive Modality Reinforcement for Human Multimodal Emotion Recognition From Unaligned Multimodal Sequences | Human multimodal emotion recognition involves time-series data of different modalities, such as natural language, visual motions, and acoustic behaviors. Due to the variable sampling rates for sequences from different modalities, the collected multimodal streams are usually unaligned. The asynchrony across modalities increases the difficulty on conducting efficient multimodal fusion. Hence, this work mainly focuses on multimodal fusion from unaligned multimodal sequences. To this end, we propose the Progressive Modality Reinforcement (PMR) approach based on the recent advances of crossmodal transformer. Our approach introduces a message hub to exchange information with each modality. The message hub sends common messages to each modality and reinforces their features via crossmodal attention. In turn, it also collects the reinforced features from each modality and uses them to generate a reinforced common message. By repeating the cycle process, the common message and the modalities' features can progressively complement each other. Finally, the reinforced features are used to make predictions for human emotion. Comprehensive experiments on different human multimodal emotion recognition benchmarks clearly demonstrate the superiority of our approach. | ['Guosheng Lin', 'Lixin Duan', 'Yanyong Huang', 'Xiang Chen', 'Fengmao Lv'] | 2021-06-19 | null | null | null | cvpr-2021-1 | ['multimodal-emotion-recognition', 'multimodal-emotion-recognition'] | ['computer-vision', 'speech'] | [ 3.16813201e-01 -4.05968696e-01 -4.42459919e-02 -2.82427400e-01
-9.04996157e-01 -4.84318376e-01 8.09136569e-01 1.88413888e-01
-5.13122618e-01 7.55471945e-01 5.32391965e-01 3.33617538e-01
1.09659694e-01 -3.43803138e-01 -5.02540946e-01 -1.04527080e+00
2.63154805e-01 -9.11705643e-02 -1.36197940e-01 -2.86757350e-01
1.58031568e-01 1.31683543e-01 -1.64103734e+00 4.75179434e-01
7.15510070e-01 1.10549545e+00 2.48627171e-01 7.96815395e-01
-4.70741123e-01 1.03730464e+00 -3.00355792e-01 -5.09512842e-01
-2.72898316e-01 -6.26346111e-01 -5.81011832e-01 2.10514635e-01
-2.81953216e-01 -3.37731749e-01 -3.82574201e-01 8.06298494e-01
5.88968098e-01 4.43562120e-01 3.95680815e-01 -1.75818336e+00
3.46740335e-02 8.54991436e-01 -7.66837955e-01 -2.22209543e-02
8.44325900e-01 1.49335772e-01 8.72903466e-01 -8.32869291e-01
5.05946100e-01 1.30884373e+00 3.65265906e-01 6.35023296e-01
-9.94415998e-01 -6.38331056e-01 4.48365778e-01 2.39093691e-01
-8.78644288e-01 -5.12921989e-01 1.18985486e+00 -1.44412115e-01
5.45780897e-01 3.29365045e-01 6.94904029e-01 1.25117719e+00
-2.35134125e-01 1.16352689e+00 9.28432465e-01 -2.30580091e-01
3.85682583e-02 -4.62547280e-02 -6.42346591e-02 4.27711308e-01
-7.44953275e-01 -1.28738824e-02 -1.07673788e+00 -2.63536930e-01
3.12756151e-01 4.58609909e-01 -2.43575335e-01 6.05903715e-02
-1.74836290e+00 5.31233370e-01 2.74117380e-01 4.75871176e-01
-7.28830576e-01 1.02644488e-01 6.75978005e-01 4.04280275e-01
6.05081059e-02 -6.08108230e-02 -2.09412709e-01 -5.48306644e-01
-4.74635392e-01 -1.49715930e-01 5.27813435e-01 5.55835187e-01
9.19600070e-01 -1.23378739e-01 -1.68740936e-02 1.00428462e+00
5.70418537e-01 7.57261992e-01 5.65804660e-01 -9.90029871e-01
6.20501041e-01 6.39452577e-01 4.29769121e-02 -1.25637531e+00
-3.86161745e-01 3.55940014e-01 -1.14567447e+00 -2.16665208e-01
1.60143450e-01 -5.47415495e-01 -5.97387373e-01 2.00681925e+00
4.88642871e-01 5.87306440e-01 4.23183978e-01 1.03512883e+00
9.77884173e-01 9.80073929e-01 5.02626061e-01 -4.42197382e-01
1.17987263e+00 -8.34550202e-01 -7.92768419e-01 2.33533144e-01
4.43863273e-01 -9.08718705e-01 5.71541369e-01 2.56568104e-01
-1.07262087e+00 -5.75564682e-01 -6.90111637e-01 3.27193141e-01
-1.17628634e-01 -1.99518949e-01 3.75589192e-01 8.50414112e-02
-7.71461666e-01 1.61366329e-01 -7.74885356e-01 -1.75883710e-01
1.09955585e-02 3.92104298e-01 -5.99504292e-01 -6.98396787e-02
-1.21118367e+00 4.42093819e-01 4.36912924e-01 2.82712579e-01
-9.29777622e-01 -4.26025450e-01 -1.00521374e+00 -1.53646335e-01
3.14280927e-01 -5.31046033e-01 1.38336897e+00 -1.20546591e+00
-1.70896423e+00 2.29348287e-01 -5.00766277e-01 1.08547024e-02
2.42345020e-01 -2.92419605e-02 -6.27449811e-01 3.73850107e-01
-2.38200352e-01 1.01181149e+00 1.12481153e+00 -1.52881920e+00
-1.19412363e+00 -1.71232522e-01 -8.20292309e-02 5.62488556e-01
-4.49546635e-01 3.94768976e-02 -4.40631777e-01 -4.36447203e-01
1.08780034e-01 -7.57355273e-01 -1.28547102e-01 -6.40742779e-01
-3.54464948e-01 -2.41472960e-01 9.09968615e-01 -3.47083211e-01
1.11157596e+00 -2.48311353e+00 7.20239162e-01 3.13636005e-01
4.47498024e-01 -2.86040992e-01 -5.82423866e-01 7.03534007e-01
-4.40093502e-02 -1.63597688e-01 -1.61304250e-01 -6.82293594e-01
1.18468724e-01 2.83630997e-01 -4.21786159e-01 1.46639258e-01
2.88090318e-01 7.92384803e-01 -9.64653790e-01 -7.34300673e-01
4.61239249e-01 5.73625207e-01 -1.56767920e-01 4.69039083e-01
8.33609607e-03 8.52457285e-01 -4.83605921e-01 6.66507125e-01
6.39392495e-01 -1.19991213e-01 1.35981306e-01 -4.78146017e-01
-5.16116992e-02 -2.46748328e-01 -1.13269830e+00 1.90408051e+00
-2.88356334e-01 3.12814415e-01 2.90919635e-02 -1.05670226e+00
8.35890353e-01 6.96021438e-01 7.90564597e-01 -7.01484859e-01
3.82492691e-01 1.33950874e-01 -2.35486805e-01 -7.08733976e-01
5.64909160e-01 -2.24225327e-01 -3.34796876e-01 6.66548312e-01
2.79773146e-01 1.51760891e-01 2.15130806e-01 4.95300621e-01
7.92514920e-01 -1.02028757e-01 -1.60005111e-02 6.07044041e-01
8.87379646e-01 -2.80880511e-01 6.19585991e-01 5.20226777e-01
-3.32873404e-01 3.19659114e-01 4.32561636e-01 -1.06156379e-01
-8.98126900e-01 -1.07207847e+00 4.66580033e-01 1.48487294e+00
4.57433164e-01 -3.71065855e-01 -3.29558223e-01 -7.28359342e-01
-1.34338155e-01 6.84503168e-02 -7.06067383e-01 -2.74513781e-01
-3.16675574e-01 -5.03150344e-01 5.47300100e-01 5.32903254e-01
4.54413772e-01 -1.33628881e+00 -4.86425370e-01 3.15539867e-01
-8.06936324e-01 -1.00930548e+00 -4.67797786e-01 -8.39320794e-02
-5.36445081e-01 -7.38236547e-01 -9.25220907e-01 -6.92555130e-01
4.42401648e-01 3.68875682e-01 6.69037640e-01 -3.60171795e-02
1.90138891e-01 9.79059219e-01 -6.66291893e-01 -8.18578526e-02
-4.87540722e-01 -1.72515899e-01 1.99282095e-02 7.41168618e-01
1.12776071e-01 -4.87351418e-01 -4.71822262e-01 2.88839906e-01
-1.15895569e+00 1.67738184e-01 4.25299436e-01 1.08892584e+00
4.09684479e-01 8.18039104e-03 8.16787779e-01 -1.99725837e-01
7.16864944e-01 -7.54816651e-01 -1.31441206e-01 3.47647458e-01
1.60504609e-01 -7.80977681e-03 3.93916965e-01 -9.06354845e-01
-1.53371656e+00 1.96283892e-01 4.62667942e-02 -4.43990231e-01
-3.60926658e-01 8.42563391e-01 -1.95985824e-01 2.09623888e-01
-6.19017780e-02 4.05903727e-01 7.75994956e-02 -1.10484436e-01
5.39968669e-01 6.41611934e-01 7.23542809e-01 -6.08402014e-01
5.04521906e-01 4.43400919e-01 -3.25874716e-01 -6.87215269e-01
-1.61106274e-01 -3.38365912e-01 -2.05769509e-01 -8.06757867e-01
8.96580756e-01 -9.31184113e-01 -1.38364506e+00 8.74639034e-01
-1.18245733e+00 2.20511500e-02 -8.95545445e-03 7.75043905e-01
-4.30929244e-01 5.58883548e-01 -8.80476654e-01 -1.06744015e+00
-4.13763411e-02 -1.25692666e+00 1.26428568e+00 7.93054521e-01
-2.55858302e-01 -9.31206822e-01 1.99814171e-01 3.27424556e-01
2.61083782e-01 6.03564493e-02 5.50800979e-01 -4.70445305e-01
-3.01859170e-01 -5.36664762e-02 -8.24379548e-02 1.03369445e-01
2.05524012e-01 2.37927362e-01 -9.35092092e-01 2.50067506e-02
-2.36237004e-01 -7.42114782e-01 7.65049398e-01 1.25029311e-01
6.88618183e-01 1.81888655e-01 -1.91859230e-01 -1.85297616e-02
9.17068541e-01 3.97289991e-01 4.13636446e-01 3.03526521e-02
6.10126495e-01 7.29990840e-01 7.32442021e-01 8.79647434e-01
6.36968970e-01 1.34354353e-01 4.44706738e-01 -1.98047832e-01
4.64386582e-01 -2.25872785e-01 6.52952254e-01 1.33805168e+00
-4.27330844e-02 -3.16744030e-01 -6.78796232e-01 3.43019217e-01
-2.18349981e+00 -1.18699074e+00 1.83185235e-01 1.90608323e+00
7.27317214e-01 -3.68950665e-01 2.27083787e-01 2.45250866e-01
7.48552918e-01 2.19028935e-01 -4.63661849e-01 -1.30720228e-01
-5.17769277e-01 -4.64336991e-01 -3.42543036e-01 3.06718677e-01
-1.05005407e+00 6.51670635e-01 5.32116985e+00 7.46614933e-01
-1.16219807e+00 -1.14321105e-01 4.51454341e-01 -4.79416065e-02
-4.95074034e-01 -1.19235300e-01 -3.48968148e-01 5.59030473e-01
6.83809638e-01 -3.04991901e-01 6.85582221e-01 1.80913657e-01
5.05196601e-02 -3.85390848e-01 -9.26895082e-01 1.37411690e+00
1.15251623e-01 -1.05249584e+00 3.42776328e-02 -2.49141306e-01
5.83914876e-01 -5.51662818e-02 9.77963507e-02 3.01835090e-01
2.78530002e-01 -6.34413481e-01 5.30203700e-01 1.04966855e+00
2.89971739e-01 -1.02925992e+00 8.17312598e-01 3.41439635e-01
-1.42797935e+00 -1.23819731e-01 1.35417059e-01 2.91520953e-01
6.47389233e-01 2.28639096e-01 -3.10889125e-01 9.41518545e-01
5.57026029e-01 9.84947920e-01 -3.16795349e-01 7.11563706e-01
1.99305788e-01 3.80463511e-01 -4.31897908e-01 -2.60142088e-02
1.30903870e-01 -1.10532276e-01 5.24014711e-01 1.13357198e+00
3.99389029e-01 1.79358780e-01 3.59003693e-01 1.86656848e-01
-2.13072494e-01 2.18393281e-01 -4.96277988e-01 -3.90985042e-01
5.39051652e-01 1.48378241e+00 -3.76819819e-01 -6.85550451e-01
-6.36215806e-01 9.92717564e-01 1.71987280e-01 5.55084646e-01
-1.00113606e+00 -1.70907453e-01 4.03673083e-01 -1.01873589e+00
4.29819971e-02 2.97567323e-02 1.05803777e-02 -1.43571162e+00
-8.15828964e-02 -1.11882842e+00 6.37685835e-01 -8.79423499e-01
-1.69116735e+00 6.67801023e-01 -1.65849328e-01 -1.43557024e+00
-3.02041113e-01 1.40074792e-03 -5.40299237e-01 6.87110484e-01
-1.23079526e+00 -1.00358403e+00 -4.90152895e-01 1.18615651e+00
3.69837195e-01 -2.02435508e-01 7.47894526e-01 4.23668385e-01
-5.49652815e-01 4.67545718e-01 -2.33754203e-01 -1.16467355e-02
8.38116229e-01 -8.65764201e-01 -5.13050258e-01 4.66366053e-01
-1.77606702e-01 2.93815672e-01 4.37903166e-01 -6.23605072e-01
-1.71370900e+00 -5.88772833e-01 5.96357882e-01 5.58258779e-02
9.06441748e-01 1.36843473e-01 -9.93204951e-01 2.94881135e-01
7.27211177e-01 -3.00548434e-01 1.02519798e+00 -7.85327628e-02
-3.44748348e-01 -2.25710467e-01 -8.14738274e-01 7.53700137e-01
5.38771212e-01 -6.14368498e-01 -5.58729172e-01 -2.60240793e-01
7.59992480e-01 -1.48543701e-01 -1.10219479e+00 5.81678987e-01
6.55756891e-01 -7.29569435e-01 6.41568840e-01 -6.89807236e-01
6.99004471e-01 -4.13638055e-01 -5.05371034e-01 -1.35793853e+00
1.29165545e-01 -5.86592674e-01 -1.61712971e-02 1.65745759e+00
3.56233776e-01 -7.07219422e-01 3.30330819e-01 7.38988280e-01
1.76484272e-01 -1.67605460e-01 -9.44684267e-01 -2.31026024e-01
-4.98091549e-01 -6.13203108e-01 7.63154685e-01 1.13628244e+00
6.19466662e-01 3.77791077e-01 -9.00564551e-01 3.66202518e-02
5.27391374e-01 3.08636487e-01 8.11213315e-01 -8.63974094e-01
-1.91080973e-01 -4.82961506e-01 -2.90790826e-01 -9.55074191e-01
1.96232155e-01 -6.29488885e-01 2.21128359e-01 -9.36841190e-01
5.11414587e-01 -1.37213752e-01 -6.59296691e-01 5.29901743e-01
-3.89585644e-01 1.40167430e-01 6.34230793e-01 4.03068885e-02
-1.07571852e+00 1.05270767e+00 1.24672818e+00 -2.32539713e-01
-3.58912170e-01 -4.74313557e-01 -4.47016925e-01 7.20879972e-01
8.19970369e-01 -2.57146835e-01 -2.93102652e-01 -3.10573936e-01
7.18520284e-02 6.33021593e-01 3.90078336e-01 -7.60992765e-01
4.58545297e-01 -2.56600142e-01 3.91497970e-01 -8.05023909e-01
6.13594115e-01 -1.06060255e+00 1.38910338e-01 2.22861081e-01
-5.47129869e-01 1.70694247e-01 2.69653618e-01 7.67727196e-01
-6.91619992e-01 2.66593009e-01 3.68916273e-01 2.99654663e-01
-7.80092955e-01 2.37763077e-01 -6.20233059e-01 -2.22025380e-01
1.03232908e+00 1.03101641e-01 -3.69301856e-01 -7.36707211e-01
-1.07828391e+00 6.74368799e-01 9.30887982e-02 6.31332397e-01
8.81725967e-01 -1.76267374e+00 -6.56594455e-01 6.31045029e-02
1.55906841e-01 -4.41959620e-01 9.12627518e-01 1.29178774e+00
3.96263897e-01 -1.12904869e-01 -2.97670573e-01 -8.59011352e-01
-1.53255582e+00 5.04327476e-01 1.12275757e-01 -6.52593896e-02
-7.85922259e-02 3.86481762e-01 1.46794379e-01 -3.52603167e-01
2.54767537e-01 1.70292363e-01 -3.60533655e-01 4.91666585e-01
6.87648118e-01 3.28917056e-01 -4.50070530e-01 -8.75656843e-01
-2.86013931e-01 4.26418543e-01 -2.53423564e-02 -6.91793978e-01
1.21055913e+00 -5.85205197e-01 -2.73371547e-01 9.00950909e-01
1.27941573e+00 -9.02305637e-03 -1.14847231e+00 -4.99455661e-01
-3.08290988e-01 -3.09668899e-01 -4.14530307e-01 -6.21733904e-01
-1.12444878e+00 8.13216507e-01 4.10844505e-01 2.29459330e-01
1.70322347e+00 7.43269473e-02 8.81088912e-01 3.45632315e-01
1.43788859e-01 -1.10513031e+00 3.16017985e-01 4.07896936e-01
7.62533069e-01 -1.42104542e+00 -7.31961906e-01 -9.15870667e-02
-1.22325039e+00 9.90405083e-01 4.99805003e-01 3.42587531e-01
5.18336177e-01 2.70210594e-01 2.34769866e-01 1.46312937e-01
-1.04512811e+00 -2.58571506e-01 1.36694089e-01 4.18520480e-01
1.49355933e-01 5.34023866e-02 -1.49937660e-01 7.52139568e-01
3.58677506e-01 1.06467120e-01 1.82266340e-01 1.02479959e+00
-2.84385085e-01 -1.20728302e+00 -6.71263039e-01 1.72310904e-01
-4.06763554e-02 1.05187565e-01 -3.20005000e-01 3.35854709e-01
4.97791059e-02 1.28615475e+00 -1.39101416e-01 -8.89372349e-01
1.86715707e-01 2.08294228e-01 2.29170397e-01 2.51392722e-01
-5.82888007e-01 5.23946345e-01 -1.05294585e-01 -5.06360233e-01
-9.65023279e-01 -9.17772233e-01 -1.43423915e+00 -4.15301830e-01
-6.27814308e-02 2.83554286e-01 5.16807199e-01 9.76459026e-01
3.77982885e-01 5.18274486e-01 1.02661598e+00 -1.09809482e+00
-1.00989088e-01 -8.14612806e-01 -2.88021922e-01 8.26883852e-01
4.35867250e-01 -5.81186414e-01 -3.44000578e-01 3.06718081e-01] | [13.208966255187988, 5.067779541015625] |
82690220-e400-43b9-87ad-320ac9f8a6ff | temporal-film-capturing-long-range-sequence-1 | null | null | http://papers.nips.cc/paper/9217-temporal-film-capturing-long-range-sequence-dependencies-with-feature-wise-modulations | http://papers.nips.cc/paper/9217-temporal-film-capturing-long-range-sequence-dependencies-with-feature-wise-modulations.pdf | Temporal FiLM: Capturing Long-Range Sequence Dependencies with Feature-Wise Modulations. | Learning representations that accurately capture long-range dependencies in sequential inputs --- including text, audio, and genomic data --- is a key problem in deep learning. Feed-forward convolutional models capture only feature interactions within finite receptive fields while recurrent architectures can be slow and difficult to train due to vanishing gradients. Here, we propose Temporal Feature-Wise Linear Modulation (TFiLM) --- a novel architectural component inspired by adaptive batch normalization and its extensions --- that uses a recurrent neural network to alter the activations of a convolutional model. This approach expands the receptive field of convolutional sequence models with minimal computational overhead. Empirically, we find that TFiLM significantly improves the learning speed and accuracy of feed-forward neural networks on a range of generative and discriminative learning tasks, including text classification and audio super-resolution. | ['Zayd Enam', 'Sawyer Birnbaum', 'Volodymyr Kuleshov', 'Pang Wei W. Koh', 'Stefano Ermon'] | 2019-12-01 | null | null | null | neurips-2019-12 | ['audio-super-resolution', 'audio-super-resolution'] | ['audio', 'music'] | [ 6.83068514e-01 -2.20945925e-01 -1.65183023e-01 -6.71860814e-01
-6.71905220e-01 -4.43163306e-01 6.21937215e-01 -2.49797016e-01
-4.14265424e-01 5.72974324e-01 6.13027334e-01 -1.63214266e-01
6.56493679e-02 -5.68294644e-01 -7.90116251e-01 -6.67841554e-01
-2.38129377e-01 -6.05248846e-02 1.42953932e-01 -2.93842018e-01
1.19998097e-01 3.75508755e-01 -1.43386745e+00 8.96298051e-01
1.39681771e-01 8.71974766e-01 1.66475743e-01 1.14174068e+00
-1.72726229e-01 1.16367924e+00 -4.69091266e-01 8.26325268e-02
-1.18588589e-01 -7.21784830e-01 -7.99605310e-01 -4.27190036e-01
3.67103219e-01 -2.96643525e-01 -7.16917694e-01 5.13672411e-01
7.55329132e-01 3.61517638e-01 6.31483555e-01 -6.89141393e-01
-9.64724123e-01 8.25886786e-01 -2.55866855e-01 5.86725831e-01
4.33352450e-03 -2.71929372e-02 1.14320195e+00 -1.02893972e+00
5.41857421e-01 1.19215143e+00 1.07919323e+00 9.06761050e-01
-1.49503732e+00 -7.89809883e-01 2.80333877e-01 -1.28254637e-01
-1.21091855e+00 -8.00184608e-01 5.04172862e-01 -3.05682570e-01
1.66113925e+00 3.32326740e-01 3.93166512e-01 1.40088642e+00
5.68034530e-01 5.34916222e-01 4.49091107e-01 -3.08840543e-01
-2.79351347e-03 -3.92492056e-01 2.56223287e-02 4.49957877e-01
-7.39833638e-02 5.66750485e-03 -9.99780715e-01 -1.83691293e-01
1.13118160e+00 1.74756676e-01 -9.84837636e-02 3.65267277e-01
-9.13354099e-01 7.54012942e-01 4.30318683e-01 7.14948356e-01
-2.94782549e-01 7.39691496e-01 4.19968098e-01 7.58542180e-01
5.87387919e-01 5.53186536e-01 -5.34729898e-01 -2.57839411e-01
-1.08476734e+00 -3.47205065e-02 5.99337101e-01 6.91086173e-01
5.36892295e-01 3.94516826e-01 -3.62820387e-01 8.42092395e-01
-7.05168322e-02 1.95607573e-01 8.32790732e-01 -5.11557460e-01
2.36941263e-01 2.46639475e-01 -2.74025679e-01 -6.58993423e-01
-7.27955520e-01 -6.93695128e-01 -1.07165873e+00 -1.95082888e-01
1.96581110e-02 -2.83030719e-01 -1.04707956e+00 1.81278288e+00
-3.60558555e-02 2.92353958e-01 -1.20596262e-02 6.68874323e-01
1.05267477e+00 7.86035419e-01 1.46956056e-01 -2.81758606e-01
9.37779605e-01 -6.32864475e-01 -7.39985824e-01 -2.67703950e-01
7.09496140e-01 -6.61321282e-01 9.36872363e-01 5.01251221e-02
-1.09850931e+00 -6.48462951e-01 -8.65710795e-01 -3.20534527e-01
-3.26852649e-01 9.11328644e-02 6.38878524e-01 3.37478131e-01
-1.23279595e+00 7.15092361e-01 -7.94637620e-01 -1.66163057e-01
4.50900495e-01 7.47928262e-01 -3.40017796e-01 1.27930537e-01
-1.21510577e+00 4.61584479e-01 1.58617094e-01 -1.35519430e-01
-7.94172168e-01 -8.88708353e-01 -5.17555475e-01 4.16236281e-01
-2.47680113e-01 -6.44465804e-01 1.29543626e+00 -1.17630780e+00
-1.81481802e+00 4.41485554e-01 -3.85886699e-01 -5.83610833e-01
-6.82101250e-02 -2.53276646e-01 -3.41156006e-01 -1.58029959e-01
-3.53822917e-01 7.07027137e-01 1.10460067e+00 -4.95958656e-01
-4.54849988e-01 -1.19238861e-01 -2.31883839e-01 1.33641630e-01
-7.25374758e-01 3.83351855e-02 -1.14380091e-01 -1.01665783e+00
4.77752015e-02 -9.65906203e-01 -3.91266555e-01 -3.15440863e-01
-1.04127293e-02 -1.50288418e-01 1.07070076e+00 -4.37449753e-01
1.22622061e+00 -2.49397993e+00 1.79167107e-01 -2.58565677e-04
1.06391847e-01 2.38784283e-01 -4.79294181e-01 3.83947849e-01
-3.09799373e-01 1.09152393e-02 1.33757368e-01 -3.27911913e-01
-3.52257311e-01 3.13894719e-01 -6.38897181e-01 2.51203358e-01
6.20987773e-01 1.32722080e+00 -5.44042230e-01 1.52385281e-02
-5.38527891e-02 9.18367505e-01 -6.52415276e-01 7.52943456e-02
-5.10733068e-01 4.53856319e-01 -2.13581339e-01 2.07816422e-01
2.94948459e-01 -5.16004324e-01 1.25661701e-01 1.06488004e-01
-7.17066601e-02 7.59027779e-01 -7.07478940e-01 1.99183190e+00
-6.15271330e-01 1.09528518e+00 -3.06571454e-01 -9.96958137e-01
1.01508021e+00 4.17889655e-01 3.65936637e-01 -7.04173267e-01
6.87860921e-02 1.83306262e-02 -6.42599314e-02 -2.60114253e-01
4.55828905e-01 -7.43338987e-02 -1.78070217e-01 4.97151911e-01
6.78229868e-01 3.81531000e-01 -1.43018305e-01 -9.40806344e-02
1.39617872e+00 5.40251471e-02 3.68820410e-03 5.20852022e-02
1.65984109e-01 -4.22928602e-01 4.69815999e-01 9.22943115e-01
4.20534134e-01 7.43223846e-01 3.76150638e-01 -5.51790178e-01
-1.02579856e+00 -6.73572183e-01 -7.50589222e-02 2.10168934e+00
-5.76166987e-01 -5.40743053e-01 -5.16507149e-01 -3.04759413e-01
-1.62427157e-01 4.00149316e-01 -9.26865280e-01 -4.20141399e-01
-7.21266806e-01 -8.57341945e-01 9.06629562e-01 8.87572587e-01
1.86438337e-01 -9.93053138e-01 -5.06201804e-01 5.10497928e-01
1.02539677e-02 -1.04285300e+00 -6.35718703e-01 8.09021831e-01
-1.00443578e+00 -4.98052955e-01 -7.15076983e-01 -9.35875535e-01
5.16021788e-01 2.89439820e-02 9.31340277e-01 -3.86277258e-01
-5.75149596e-01 4.24865223e-02 -3.73809189e-01 -1.59309864e-01
-4.15447950e-01 4.92355347e-01 -1.94138214e-01 1.34515300e-01
3.14811766e-01 -6.68900728e-01 -3.97885859e-01 3.84614952e-02
-9.01669800e-01 -4.48905528e-02 8.79218400e-01 1.09471345e+00
5.77009261e-01 -2.24181309e-01 6.63079977e-01 -9.94221568e-01
6.37109637e-01 -3.10742319e-01 -2.64739662e-01 1.10828720e-01
-1.89464465e-01 4.12754118e-01 7.26905227e-01 -8.48166466e-01
-9.65950608e-01 1.12267293e-01 -2.70425081e-01 -5.13804018e-01
-6.94286972e-02 5.70547879e-01 2.59411573e-01 -1.25100598e-01
9.03191864e-01 4.55163747e-01 -1.97647050e-01 -5.13865650e-01
3.62486452e-01 6.83876216e-01 5.81570745e-01 -1.12107240e-01
2.75901556e-01 3.77960652e-01 1.90536156e-01 -9.68251288e-01
-7.60502636e-01 -4.19528842e-01 -5.80363572e-01 9.61165205e-02
6.58401489e-01 -1.07653904e+00 -4.72560227e-01 3.83613646e-01
-1.22612262e+00 -4.42492992e-01 -4.82806832e-01 5.63867807e-01
-3.89767259e-01 -3.57808828e-01 -1.05963922e+00 -5.93446732e-01
-5.84452212e-01 -6.85405135e-01 9.66099083e-01 2.95119166e-01
-4.46467429e-01 -8.73027384e-01 2.36466214e-01 -3.66972744e-01
9.65848446e-01 -5.43703772e-02 9.13744509e-01 -8.40864420e-01
-2.51875401e-01 -9.46616754e-02 1.10372826e-01 1.49775386e-01
1.71064302e-01 -3.37722749e-02 -1.40440190e+00 -3.90369177e-01
-3.02901179e-01 -2.88986713e-01 1.33587861e+00 6.42552972e-01
1.22892225e+00 -4.50825185e-01 -3.89078975e-01 8.47913206e-01
9.38953698e-01 5.61141185e-02 6.18838370e-01 -1.95694700e-01
5.26759923e-01 2.35348135e-01 -2.26825997e-02 6.58280015e-01
-2.77070105e-01 4.61854666e-01 1.34319335e-01 -9.14363712e-02
-4.13500041e-01 -3.85722220e-01 4.43142325e-01 9.93065774e-01
-7.70297498e-02 -1.23241335e-01 -4.29813504e-01 3.47494215e-01
-1.86329007e+00 -1.07385683e+00 3.03071797e-01 1.90788627e+00
1.05937171e+00 1.02519326e-01 4.89422232e-02 7.49953389e-02
5.10890722e-01 1.73141554e-01 -6.32999837e-01 -4.83839482e-01
-1.57743156e-01 7.83679068e-01 3.26445341e-01 2.83017904e-01
-1.11123037e+00 8.87492001e-01 7.16127157e+00 7.41102457e-01
-1.75553441e+00 2.43509352e-01 6.04726672e-01 -5.52335620e-01
-3.45405042e-01 -4.37462300e-01 -1.05357742e+00 1.58287391e-01
1.62221909e+00 2.29532227e-01 4.04177636e-01 7.14711666e-01
2.81444769e-02 4.51692492e-01 -1.27000487e+00 9.56609190e-01
1.97349768e-03 -1.92505980e+00 1.79110229e-01 -1.34969160e-01
8.20988834e-01 2.75767624e-01 6.15229249e-01 3.22826266e-01
2.75522441e-01 -1.49079168e+00 4.18712735e-01 6.46412790e-01
1.31516874e+00 -7.97972143e-01 4.34861064e-01 1.28209531e-01
-1.18892908e+00 -3.69572669e-01 -4.83338922e-01 -7.98959509e-02
-3.74250203e-01 4.70017403e-01 -8.69520009e-01 -4.95896041e-02
7.80361354e-01 7.90523589e-01 -2.94502944e-01 5.67122638e-01
9.17691216e-02 8.07936251e-01 -1.53759748e-01 -2.95906752e-01
2.39137486e-01 5.27567387e-01 2.10110098e-01 1.91984427e+00
2.33379245e-01 5.31275384e-02 -3.06998938e-01 7.18528390e-01
-4.30349559e-01 -1.52163431e-01 -6.01083994e-01 -4.47032183e-01
4.44830328e-01 1.03123677e+00 -4.57883418e-01 4.39032279e-02
-4.19297218e-01 1.00400627e+00 3.72895420e-01 5.44085681e-01
-5.13052404e-01 -6.76331460e-01 8.46764088e-01 1.24929689e-01
6.56250298e-01 -3.18464428e-01 -2.58828044e-01 -8.46651912e-01
-3.71126622e-01 -6.72871232e-01 3.14904958e-01 -4.75419104e-01
-1.01364994e+00 6.47992015e-01 -4.01096642e-01 -8.51635396e-01
-7.48677790e-01 -4.84587610e-01 -3.68553817e-01 8.20865095e-01
-1.11745667e+00 -1.26321065e+00 -2.71897577e-02 6.87332511e-01
7.88197756e-01 -2.73639470e-01 1.15481079e+00 1.94567740e-01
-2.35946134e-01 9.15441275e-01 4.35840458e-01 1.98867619e-01
4.41826284e-01 -7.56151021e-01 8.52128148e-01 5.78413188e-01
5.43255508e-01 6.97732568e-01 4.43613052e-01 -3.41205359e-01
-1.34519827e+00 -1.47730827e+00 1.08674681e+00 -7.51397386e-02
3.85962576e-01 -6.70275450e-01 -1.12490165e+00 8.88459921e-01
4.48679253e-02 3.77267629e-01 1.01374519e+00 3.33524883e-01
-6.81505024e-01 -1.15366757e-01 -7.65764534e-01 4.06373531e-01
1.16177583e+00 -1.08003438e+00 -1.87929988e-01 9.38796178e-02
8.58597159e-01 -4.18883264e-01 -7.23003209e-01 2.73192912e-01
8.43030214e-01 -5.07473171e-01 8.36819410e-01 -9.39857185e-01
1.75319403e-01 1.47102445e-01 -1.26236379e-01 -1.22098589e+00
-8.73507082e-01 -1.05060387e+00 -1.97521284e-01 9.44849610e-01
4.82612014e-01 -6.11139476e-01 6.51148379e-01 1.36116426e-02
-1.67174995e-01 -6.92223072e-01 -1.07195461e+00 -4.43824440e-01
1.22944631e-01 -2.98037171e-01 4.55225587e-01 9.14299130e-01
5.71352579e-02 5.17466128e-01 -4.42465454e-01 1.41568752e-02
-7.97725171e-02 -5.97929582e-02 3.09179395e-01 -1.13233590e+00
-5.75333536e-01 -4.28876311e-01 -4.34632272e-01 -1.24804437e+00
1.39703676e-01 -8.56657445e-01 -4.55666892e-02 -1.08152962e+00
1.34926155e-01 -2.91688681e-01 -5.89842677e-01 7.50858128e-01
1.28413200e-01 4.16858882e-01 -3.95569727e-02 2.11157933e-01
-5.01762927e-01 4.14838403e-01 8.88770998e-01 -7.84368888e-02
-2.62450129e-01 2.20573712e-02 -5.67663968e-01 3.76759320e-01
7.58608401e-01 -5.67052484e-01 -3.36398244e-01 -5.65751493e-01
4.15656984e-01 8.24509468e-03 2.38370702e-01 -9.60703254e-01
2.07083836e-01 1.24099255e-02 8.86332512e-01 -3.34979177e-01
4.54621047e-01 -4.19416696e-01 1.62904978e-01 3.94272864e-01
-1.18194175e+00 -4.97675352e-02 3.71354848e-01 5.58187366e-01
-1.13127291e-01 -3.53725068e-02 6.46688700e-01 -1.00110836e-01
-5.76697230e-01 2.77858287e-01 -6.96280599e-01 -6.30666241e-02
4.24707383e-01 1.83422253e-01 -3.15452754e-01 -4.97384638e-01
-7.74152398e-01 -4.07836288e-01 -1.25158414e-01 5.90602100e-01
7.28103518e-01 -1.30891347e+00 -8.00057590e-01 7.15339899e-01
-2.23009467e-01 -2.18142748e-01 7.78851435e-02 4.97705251e-01
-3.47120725e-02 7.76512802e-01 -1.36330605e-01 -6.32590771e-01
-1.35321343e+00 8.73687267e-02 6.11723483e-01 -1.93026885e-01
-4.94842708e-01 1.44658387e+00 3.27740699e-01 -2.15908393e-01
2.63325661e-01 -5.24533629e-01 -1.24970764e-01 -1.80792317e-01
7.24740326e-01 -4.12544422e-02 3.05687487e-01 -4.43294823e-01
-3.42221439e-01 3.41119707e-01 -3.77863377e-01 3.15276831e-02
1.38025129e+00 1.57715440e-01 1.73380256e-01 7.84962535e-01
1.35242283e+00 -4.57390249e-01 -1.34333026e+00 -5.04081786e-01
-1.59317628e-01 -1.23004980e-01 2.86653608e-01 -5.96569479e-01
-7.72926748e-01 1.14875782e+00 6.90378785e-01 9.31224227e-02
1.07055712e+00 5.76690175e-02 6.40815973e-01 6.02735281e-01
-8.16770941e-02 -1.07068717e+00 4.87095535e-01 9.69280303e-01
8.20101261e-01 -6.22420013e-01 -3.44417602e-01 1.29450724e-01
-2.39941284e-01 1.37408936e+00 4.11249846e-01 -1.16791740e-01
7.42905796e-01 7.92844653e-01 -2.61819661e-01 2.15174049e-01
-1.46839547e+00 -1.36788920e-01 3.59283119e-01 5.07632554e-01
1.04816127e+00 -2.02261329e-01 1.96830109e-01 6.07188761e-01
-2.08106786e-02 3.45806688e-01 1.34341285e-01 1.06361973e+00
-4.93024915e-01 -8.96315277e-01 -3.73797072e-03 6.65037155e-01
-6.50897205e-01 -3.28632265e-01 -3.12091112e-01 1.58642635e-01
1.83048137e-02 8.08467448e-01 5.79271674e-01 -7.61868477e-01
-6.18361235e-02 2.28099003e-01 5.63319921e-01 -6.39259458e-01
-8.87550414e-01 3.69809479e-01 -1.07951716e-01 -5.27511060e-01
-2.90229738e-01 -4.77355033e-01 -1.43375993e+00 -2.58883983e-01
-3.31996351e-01 -7.25148395e-02 5.88595390e-01 8.72715950e-01
6.42161369e-01 9.42232490e-01 6.64519846e-01 -9.25279260e-01
-7.36512125e-01 -1.22305393e+00 -4.68871891e-01 7.14193210e-02
7.40402341e-01 -2.01134980e-01 -1.03902563e-01 2.76591003e-01] | [10.9157133102417, 6.539568901062012] |
ca0a8f41-ec6b-49ca-9360-d53beb6d11ef | segmentation-of-structural-parts-of-rosebush | 2012.11489 | null | https://arxiv.org/abs/2012.11489v2 | https://arxiv.org/pdf/2012.11489v2.pdf | Segmentation of structural parts of rosebush plants with 3D point-based deep learning methods | Segmentation of structural parts of 3D models of plants is an important step for plant phenotyping, especially for monitoring architectural and morphological traits. Current state-of-the art approaches rely on hand-crafted 3D local features for modeling geometric variations in plant structures. While recent advancements in deep learning on point clouds have the potential of extracting relevant local and global characteristics, the scarcity of labeled 3D plant data impedes the exploration of this potential. We adapted six recent point-based deep learning architectures (PointNet, PointNet++, DGCNN, PointCNN, ShellNet, RIConv) for segmentation of structural parts of rosebush models. We generated 3D synthetic rosebush models to provide adequate amount of labeled data for modification and pre-training of these architectures. To evaluate their performance on real rosebush plants, we used the ROSE-X data set of fully annotated point cloud models. We provided experiments with and without the incorporation of synthetic data to demonstrate the potential of point-based deep learning techniques even with limited labeled data of real plants. The experimental results show that PointNet++ produces the highest segmentation accuracy among the six point-based deep learning methods. The advantage of PointNet++ is that it provides a flexibility in the scales of the hierarchical organization of the point cloud data. Pre-training with synthetic 3D models boosted the performance of all architectures, except for PointNet. | ['David Rousseau', 'Gilles Galopin', 'Helin Dutagaci', 'Kaya Turgut'] | 2020-12-21 | null | null | null | null | ['plant-phenotyping'] | ['computer-vision'] | [-1.91043064e-01 1.05252109e-01 1.00768536e-01 -3.11271995e-01
-3.17277312e-01 -9.00201857e-01 3.02794874e-01 4.87295806e-01
1.98633879e-01 1.04740106e-01 -8.78719151e-01 -6.51550412e-01
-2.65935779e-01 -1.13942814e+00 -6.62609279e-01 -2.87829340e-01
-4.73677695e-01 1.01343131e+00 5.33801854e-01 -5.52065730e-01
-1.41799664e-02 1.49325848e+00 -1.23750579e+00 1.79953039e-01
6.42600179e-01 9.82135832e-01 6.64198160e-01 5.42196333e-01
-5.62682807e-01 -2.44350716e-01 -6.21334136e-01 5.55704534e-02
5.44313610e-01 7.94132873e-02 -7.38086462e-01 5.24545968e-01
3.50518763e-01 -2.93454081e-01 2.27852076e-01 6.58199131e-01
3.66218448e-01 -2.99316406e-01 4.45746839e-01 -1.40951622e+00
-6.68180883e-01 4.60086137e-01 -7.80652940e-01 -2.48268753e-01
-2.41488576e-01 3.28176320e-01 9.77972567e-01 -7.95382321e-01
5.91113031e-01 1.18115699e+00 1.18699229e+00 1.42026514e-01
-1.53120613e+00 -1.50581419e-01 6.09509982e-02 -3.98653634e-02
-1.36034858e+00 2.83293426e-01 8.56124878e-01 -5.39499998e-01
8.84584844e-01 2.15315804e-01 9.78026688e-01 4.80799019e-01
-1.66759994e-02 6.70749843e-01 7.32801139e-01 -3.39358121e-01
3.76722246e-01 -3.88032049e-01 3.68801989e-02 7.57337570e-01
-3.48866358e-02 1.39959916e-01 2.65894234e-01 -1.41920611e-01
1.39935887e+00 -6.10516816e-02 -4.79794145e-02 -9.49351668e-01
-8.79242361e-01 7.79576659e-01 1.07607651e+00 4.61597383e-01
-4.90485609e-01 9.79851857e-02 3.68394494e-01 2.84578055e-02
4.80410218e-01 6.10335350e-01 -1.18913627e+00 3.11684698e-01
-1.07006180e+00 4.86500978e-01 4.74046588e-01 1.16278005e+00
8.08127820e-01 1.70245126e-01 3.87944654e-02 7.97732294e-01
3.36306989e-01 3.18288982e-01 -1.57529823e-02 -1.24892890e+00
-1.94939710e-02 1.19298863e+00 7.54865110e-02 -9.20509994e-01
-6.47867262e-01 -4.20662224e-01 -6.70019150e-01 7.22305357e-01
5.12470007e-01 1.46009102e-01 -1.33563852e+00 1.20857346e+00
2.79048324e-01 -9.87007916e-02 -4.92681742e-01 6.25669003e-01
9.26360428e-01 6.05662107e-01 8.93108323e-02 5.54233134e-01
1.09337604e+00 -3.55906337e-01 2.52016764e-02 1.99651420e-02
8.97826314e-01 -6.18886173e-01 1.15100765e+00 1.18529685e-01
-1.18110967e+00 -7.58947313e-01 -9.30440843e-01 1.20054908e-01
-5.19176245e-01 3.71029645e-01 8.69414628e-01 5.45415044e-01
-1.16283870e+00 1.04575670e+00 -1.06002092e+00 -4.59123939e-01
1.04413223e+00 6.47238672e-01 -4.58894610e-01 1.24751680e-01
-4.60750520e-01 7.90957630e-01 6.63419962e-01 2.61747599e-01
-1.08369553e+00 -1.06827569e+00 -5.96894860e-01 3.92282605e-01
2.56279320e-01 -3.96002263e-01 1.34691000e+00 -5.19690394e-01
-1.56194162e+00 1.23479974e+00 4.77787435e-01 -3.51254910e-01
2.35878721e-01 2.22568139e-01 1.88860029e-01 7.04569370e-02
-1.30362073e-02 1.21932280e+00 4.79194760e-01 -1.57140279e+00
-5.09451389e-01 -7.26309180e-01 3.15321349e-02 -2.48558253e-01
1.32773563e-01 -3.91484350e-01 -1.60815358e-01 -1.26778558e-01
6.80951536e-01 -1.16715837e+00 -4.21832919e-01 7.20601261e-01
-3.91034544e-01 -2.25389656e-02 1.42418563e+00 -5.50868392e-01
-3.19252461e-02 -2.00398755e+00 -3.64219211e-02 1.38432264e-01
1.87472612e-01 8.02348852e-01 -5.43870151e-01 4.32909042e-01
-4.04488385e-01 6.02915287e-01 -4.42914307e-01 -9.14310887e-02
-8.90191644e-02 4.78546023e-01 8.95782262e-02 3.22871476e-01
5.90061367e-01 1.16291475e+00 -6.15650415e-01 -3.31512511e-01
6.56042159e-01 5.02504647e-01 -3.88664901e-01 1.18575394e-02
-7.26361632e-01 4.06445086e-01 -3.90492439e-01 9.03867960e-01
1.06452048e+00 -1.59896255e-01 -2.37806350e-01 -5.44801466e-02
-2.80005723e-01 -1.67282388e-01 -7.74052560e-01 1.70502520e+00
-4.26121205e-01 4.90929484e-01 4.30557013e-01 -9.97212768e-01
1.35284233e+00 3.20452332e-01 8.85855079e-01 -5.44833392e-02
1.11741468e-01 1.39536783e-01 2.16599077e-01 1.70703992e-01
-2.78030839e-02 1.43884808e-01 3.77267152e-01 -2.17143804e-01
4.41837192e-01 -1.16317439e+00 1.07110664e-01 -6.12193346e-02
9.60979700e-01 4.36389804e-01 -5.86723946e-02 -3.78450871e-01
3.93751740e-01 5.42613804e-01 3.73619765e-01 3.14456165e-01
-2.55726874e-01 7.97047555e-01 7.21477330e-01 -5.73565364e-01
-1.47496140e+00 -9.63368237e-01 -3.99267077e-01 7.42737114e-01
-1.62643537e-01 -4.02276009e-01 -5.79927444e-01 -6.54908836e-01
2.79383093e-01 7.76199162e-01 -5.23266137e-01 9.82823372e-02
-3.23252261e-01 -7.78615355e-01 5.31711638e-01 6.37789488e-01
6.53427303e-01 -1.16396296e+00 -8.17269444e-01 2.02550575e-01
4.62443620e-01 -1.13473022e+00 5.40421665e-01 5.12978852e-01
-1.28866196e+00 -9.77435172e-01 -5.93079329e-01 -7.48261571e-01
4.37595636e-01 3.44293475e-01 1.37092888e+00 2.00203329e-01
-4.38791156e-01 1.13265403e-01 -4.56611216e-01 -8.33389282e-01
-5.16668558e-01 5.40691733e-01 -7.36277640e-01 -8.33184242e-01
1.95047095e-01 -8.57353806e-01 -2.99947262e-01 3.16361874e-01
-7.18145370e-01 -1.07909322e-01 4.50141579e-01 7.46317625e-01
8.81438375e-01 1.51518986e-01 2.20290512e-01 -8.24863672e-01
2.73752660e-01 -1.11229531e-01 -1.03214455e+00 -1.10248633e-01
-1.63191184e-01 -4.74209189e-01 5.52723587e-01 -1.83214620e-02
-5.31597733e-01 3.31397027e-01 -4.81719345e-01 -4.16877687e-01
-7.92536318e-01 6.19458854e-01 -4.25364286e-01 -4.52585459e-01
8.47536147e-01 -3.29346061e-01 1.89849123e-01 -5.35550773e-01
5.80656469e-01 2.36546561e-01 3.59855324e-01 -5.47219157e-01
1.03018630e+00 5.51952720e-01 5.30632257e-01 -1.22223234e+00
-4.87354130e-01 -2.71179229e-01 -1.43304861e+00 1.39943184e-02
7.56290078e-01 -5.45152009e-01 -7.59363294e-01 5.30729353e-01
-1.24571002e+00 -7.36771762e-01 -8.17301810e-01 -1.40293390e-01
-7.14168906e-01 1.37651727e-01 -5.78184903e-01 -2.88163900e-01
-1.82185695e-01 -1.18532598e+00 1.43864024e+00 1.17621720e-01
-2.08634585e-02 -1.08211911e+00 -1.77521855e-01 -1.05312467e-01
1.85995325e-01 7.23272860e-01 1.34656072e+00 -5.92919350e-01
-6.71809554e-01 -6.35754704e-01 -4.77033734e-01 2.92964250e-01
4.28512432e-02 7.56347358e-01 -9.16717291e-01 -4.24008891e-02
-2.31181890e-01 -2.23011091e-01 3.60981733e-01 6.53513253e-01
1.48614752e+00 4.68938231e-01 -3.80787820e-01 7.54865646e-01
1.47664428e+00 1.38726354e-01 4.76410955e-01 2.78959364e-01
9.32378471e-01 5.16633153e-01 3.63954663e-01 2.80908197e-01
-1.24085568e-01 5.34067154e-01 1.20012796e+00 -5.65386653e-01
7.32685328e-02 -6.94493875e-02 -5.15805781e-01 1.69734821e-01
-2.23594636e-01 -1.08666636e-01 -1.39786184e+00 5.75142026e-01
-1.54274499e+00 -7.67427981e-01 -5.39688051e-01 1.81645250e+00
2.93323934e-01 2.70132452e-01 -3.49498577e-02 5.30389786e-01
4.08451021e-01 -6.10680506e-02 -8.15896392e-01 -3.65019619e-01
-2.06475392e-01 5.31471908e-01 6.63648844e-01 1.70467511e-01
-1.12422872e+00 1.29686403e+00 6.25045443e+00 5.63338637e-01
-1.13599825e+00 -1.03383787e-01 5.37837446e-01 5.01208127e-01
6.58244118e-02 1.91538349e-01 -6.64520681e-01 -1.86924607e-01
7.21080005e-01 4.33989465e-01 2.04275046e-02 1.17330503e+00
3.21170479e-01 2.06080496e-01 -9.30961370e-01 5.30176163e-01
-6.08956039e-01 -1.50107074e+00 -6.46106154e-02 2.75790334e-01
6.05994284e-01 4.21637714e-01 -1.85998231e-01 2.38869727e-01
6.36922836e-01 -1.00318265e+00 4.58431512e-01 -1.44898370e-01
6.34007990e-01 -6.62403226e-01 6.22834802e-01 5.65362513e-01
-1.05646563e+00 -1.69253722e-02 -7.90172875e-01 2.53247857e-01
-6.09518401e-02 6.08395338e-01 -1.37117147e+00 6.27951086e-01
7.08231270e-01 7.29300141e-01 -1.04520071e+00 1.06407857e+00
-1.32557943e-01 7.01387286e-01 -7.46453285e-01 3.46680284e-01
6.43416643e-01 -2.86911041e-01 4.33014452e-01 8.11246157e-01
4.18168694e-01 -2.61243016e-01 3.72240543e-01 1.38768673e+00
1.39347941e-01 -1.25246242e-01 -6.45478666e-01 -1.20455131e-01
4.32694107e-01 1.58909011e+00 -1.29969299e+00 2.35608250e-01
-1.12561524e-01 6.45970643e-01 2.66200036e-01 -9.89265367e-02
-4.54513669e-01 -1.51217893e-01 3.48279297e-01 4.54748064e-01
6.53673351e-01 -8.25015247e-01 -6.21377945e-01 -3.14835995e-01
-5.78353226e-01 -6.32323980e-01 -1.09446287e-01 -1.27119470e+00
-1.22701812e+00 3.57446462e-01 1.34767428e-01 -7.54541159e-01
5.19537777e-02 -1.11434460e+00 -8.22755158e-01 1.17241323e+00
-1.02769351e+00 -1.83241379e+00 -7.74479866e-01 2.10675195e-01
6.04196310e-01 -2.49130458e-01 1.22718084e+00 -2.37866864e-01
-3.33850294e-01 -1.57320812e-01 1.26481846e-01 2.46181153e-02
-6.51404960e-03 -1.57846928e+00 1.02206981e+00 4.37617272e-01
1.78437561e-01 6.48896471e-02 3.99267375e-01 -5.84522724e-01
-1.05088460e+00 -1.26327407e+00 2.20798224e-01 -5.84672034e-01
3.84835035e-01 -3.09563160e-01 -1.06466675e+00 8.10153067e-01
-1.65256828e-01 2.26763889e-01 4.87720609e-01 7.67486244e-02
1.62901938e-01 7.40933716e-02 -1.39649272e+00 6.06484354e-01
9.63366389e-01 -2.14318916e-01 -6.56331107e-02 5.85949540e-01
7.85562098e-01 -4.33632791e-01 -1.18721640e+00 7.29153872e-01
1.38528943e-01 -9.58219945e-01 1.20019329e+00 -5.39233983e-01
3.90633464e-01 -3.24844688e-01 -2.39690199e-01 -1.74713111e+00
-8.04888546e-01 -3.42712134e-01 1.84933305e-01 1.24866295e+00
2.98777461e-01 -1.18470162e-01 1.32376754e+00 2.27222964e-01
-5.43453097e-01 -6.66864514e-01 -5.82380414e-01 -6.71690226e-01
5.86922944e-01 -3.50899071e-01 1.09826469e+00 8.32617402e-01
-8.22103500e-01 1.23965301e-01 6.89018726e-01 4.30614710e-01
3.22016299e-01 1.35307133e-01 9.16755140e-01 -1.85485697e+00
1.16894998e-01 -5.43760002e-01 -8.47256839e-01 -7.44910955e-01
3.58168930e-01 -1.05524313e+00 -2.92121232e-01 -1.75994122e+00
-6.05532587e-01 -8.78637671e-01 4.26120281e-01 6.41587853e-01
1.87575012e-01 -4.13850546e-02 3.01456511e-01 1.85690716e-01
4.32258666e-01 4.78087842e-01 1.51159120e+00 -1.54890120e-01
-4.28525418e-01 5.12251377e-01 -2.11977720e-01 9.82414782e-01
1.14875841e+00 -1.84530094e-01 -5.51684022e-01 -6.97752357e-01
-1.96789905e-01 -1.81847811e-01 7.73530483e-01 -1.19339561e+00
-3.85624558e-01 -6.74625188e-02 7.84760594e-01 -1.20002592e+00
5.77482343e-01 -1.15620697e+00 -5.92073798e-02 4.21742857e-01
-1.02609590e-01 1.38167664e-01 6.16472840e-01 5.45901805e-02
2.20107511e-01 -5.83395004e-01 9.47615504e-01 -5.90875268e-01
-6.96465552e-01 6.87618315e-01 -1.71607286e-01 -4.29977566e-01
1.17774737e+00 -4.67417508e-01 -2.00871378e-02 5.60918190e-02
-8.32345963e-01 9.43975449e-02 6.41363859e-01 2.45624274e-01
4.53834206e-01 -9.68827546e-01 -6.33149981e-01 3.68299812e-01
2.41217986e-02 7.50898838e-01 -1.41366154e-01 1.42232120e-01
-1.47800648e+00 4.70940471e-01 -8.20392787e-01 -1.36629951e+00
-8.75219107e-01 6.09494567e-01 6.22240245e-01 3.77983134e-03
-5.58366656e-01 7.92813003e-01 -9.21623707e-02 -1.12860107e+00
-5.13901114e-02 -8.51055801e-01 -1.10716231e-01 -2.63933331e-01
-3.75409484e-01 1.96940690e-01 5.31422257e-01 -5.45233667e-01
-1.61834165e-01 3.85445803e-01 2.44070128e-01 6.29466623e-02
1.64686084e+00 4.62080240e-01 -6.87648496e-03 5.40247858e-01
6.88774109e-01 -6.84648514e-01 -1.34184277e+00 2.13887155e-01
3.24779719e-01 -5.34499943e-01 3.94640058e-01 -7.29027689e-01
-1.18714333e+00 1.17144251e+00 8.47512543e-01 6.40801430e-01
8.46168220e-01 -1.00019217e-01 4.73509580e-01 3.00516099e-01
3.66424114e-01 -5.28589129e-01 -3.28823119e-01 5.29711246e-01
9.78518367e-01 -1.19667518e+00 -1.27086818e-01 -8.09296787e-01
-4.14348915e-02 1.10568261e+00 7.98724532e-01 -4.20211911e-01
9.58025992e-01 3.01294684e-01 5.32150082e-02 -5.18653512e-01
-4.86352891e-01 -3.12732339e-01 -9.36101004e-02 1.35313380e+00
5.69686174e-01 9.68824774e-02 3.19817424e-01 8.13659355e-02
-3.40428025e-01 -4.87029143e-02 2.97119766e-01 1.10670185e+00
-3.50652725e-01 -1.20444858e+00 -4.95515376e-01 3.45424861e-01
-4.14915085e-02 2.63875335e-01 -7.55037069e-01 1.28162789e+00
3.34525019e-01 3.04692924e-01 1.57052264e-01 1.16483709e-02
7.46979475e-01 -4.54901271e-02 5.98799407e-01 -9.85723019e-01
-7.27820635e-01 -8.94335285e-02 -3.68404686e-01 -2.57427663e-01
1.08639106e-01 -6.14862382e-01 -1.12549293e+00 -4.08445269e-01
-4.81455654e-01 -2.60124862e-01 1.15061355e+00 6.54777408e-01
4.22655344e-01 7.94956267e-01 3.33415419e-01 -1.50086224e+00
-4.94774908e-01 -8.72564018e-01 -7.82431841e-01 2.47282892e-01
-6.11643121e-02 -5.05888820e-01 2.19732046e-01 9.63635296e-02] | [8.949522972106934, -1.8209409713745117] |
ff76657d-0bb6-4c73-992a-8fb3d4c5d26d | dynamic-inference-with-neural-interpreters | 2110.06399 | null | https://arxiv.org/abs/2110.06399v1 | https://arxiv.org/pdf/2110.06399v1.pdf | Dynamic Inference with Neural Interpreters | Modern neural network architectures can leverage large amounts of data to generalize well within the training distribution. However, they are less capable of systematic generalization to data drawn from unseen but related distributions, a feat that is hypothesized to require compositional reasoning and reuse of knowledge. In this work, we present Neural Interpreters, an architecture that factorizes inference in a self-attention network as a system of modules, which we call \emph{functions}. Inputs to the model are routed through a sequence of functions in a way that is end-to-end learned. The proposed architecture can flexibly compose computation along width and depth, and lends itself well to capacity extension after training. To demonstrate the versatility of Neural Interpreters, we evaluate it in two distinct settings: image classification and visual abstract reasoning on Raven Progressive Matrices. In the former, we show that Neural Interpreters perform on par with the vision transformer using fewer parameters, while being transferrable to a new task in a sample efficient manner. In the latter, we find that Neural Interpreters are competitive with respect to the state-of-the-art in terms of systematic generalization | ['Bernhard Schölkopf', 'Francesco Locatello', 'Yoshua Bengio', 'Peter Gehler', 'Shruti Joshi', 'Muhammad Waleed Gondal', 'Nasim Rahaman'] | 2021-10-12 | null | http://proceedings.neurips.cc/paper/2021/hash/5b4e9aa703d0bfa11041debaa2d1b633-Abstract.html | http://proceedings.neurips.cc/paper/2021/file/5b4e9aa703d0bfa11041debaa2d1b633-Paper.pdf | neurips-2021-12 | ['systematic-generalization'] | ['reasoning'] | [ 4.36715394e-01 1.94154605e-01 1.29462391e-01 -5.19348323e-01
-5.74581046e-03 -8.19382846e-01 7.44029939e-01 2.49984637e-02
-4.60723251e-01 3.73068392e-01 7.32175633e-02 -6.27462089e-01
-2.60880321e-01 -1.17917943e+00 -1.25197959e+00 -5.57716548e-01
1.08671822e-01 7.48542309e-01 2.58679569e-01 -5.35977662e-01
1.58317402e-01 6.32054269e-01 -1.73850513e+00 5.54218054e-01
6.98362052e-01 1.03656888e+00 2.90005088e-01 8.00937355e-01
-5.51222675e-02 1.21533847e+00 -4.23838884e-01 -7.37356663e-01
1.85620815e-01 -5.19047141e-01 -1.05247617e+00 -7.91993514e-02
8.09773862e-01 -4.10090566e-01 -4.54500824e-01 8.36776674e-01
1.05326839e-01 2.29393840e-01 8.85553956e-01 -1.02831066e+00
-1.51287138e+00 8.36617112e-01 -6.79616034e-02 3.90592366e-01
1.85013518e-01 4.04721826e-01 1.30341554e+00 -1.00120485e+00
4.54372883e-01 1.29917407e+00 7.04336345e-01 7.66479969e-01
-1.45217872e+00 -1.91454455e-01 1.62267238e-01 1.64930642e-01
-1.06340945e+00 -4.69182104e-01 4.11143452e-01 -3.93624693e-01
1.26483107e+00 2.46671498e-01 6.25715017e-01 1.02878106e+00
-1.25671625e-01 8.95914257e-01 9.16801333e-01 -3.43858331e-01
5.14128566e-01 -9.61610698e-04 2.13935703e-01 1.02357507e+00
1.95925817e-01 -1.97386020e-03 -6.62512541e-01 2.54454643e-01
7.81877756e-01 1.12178311e-01 -4.60301220e-01 -3.73646885e-01
-1.14632881e+00 7.28361964e-01 9.55713212e-01 1.12638153e-01
-2.08127245e-01 3.53320777e-01 3.75119507e-01 5.29808640e-01
7.78005123e-02 5.33952296e-01 -4.55173641e-01 1.36348680e-01
-8.26419711e-01 1.95597336e-01 9.72507179e-01 9.52758014e-01
7.31070280e-01 2.02914044e-01 -1.99816953e-02 6.34977281e-01
1.43292129e-01 4.58388627e-01 8.94039750e-01 -8.91873360e-01
3.65966052e-01 7.11826861e-01 -4.54521209e-01 -5.81882179e-01
-3.61924976e-01 -7.00761795e-01 -9.14370894e-01 1.94238067e-01
4.24517065e-01 2.28984177e-01 -9.59724188e-01 1.94856119e+00
-1.40650868e-01 -1.74592823e-01 3.33359420e-01 7.57706940e-01
7.21951067e-01 6.43010437e-01 -1.04769178e-01 1.98558465e-01
1.22136343e+00 -1.17650867e+00 -1.16202295e-01 -5.45821071e-01
2.11228177e-01 -8.49019960e-02 1.57214725e+00 4.80038106e-01
-1.59623432e+00 -8.58487248e-01 -1.30338764e+00 -4.01680231e-01
-6.24736428e-01 -1.89985365e-01 7.60990560e-01 7.13187695e-01
-1.58842266e+00 7.85144508e-01 -6.99464738e-01 -4.92588907e-01
9.17551696e-01 3.60988766e-01 -2.41452992e-01 -3.10996361e-03
-8.38119984e-01 1.05439425e+00 6.10554814e-01 9.00951698e-02
-1.01351821e+00 -7.50552595e-01 -8.82219553e-01 4.89619702e-01
1.52268589e-01 -1.30222332e+00 1.40900528e+00 -1.19203222e+00
-1.58737540e+00 8.42825830e-01 3.08731180e-02 -6.96591079e-01
2.93206602e-01 -9.74313989e-02 -2.12033279e-02 1.00160934e-01
-3.22728425e-01 7.83152819e-01 1.10233557e+00 -1.17135429e+00
-6.30022824e-01 -5.65474510e-01 5.41050494e-01 1.70837361e-02
-5.17763317e-01 -5.23916006e-01 -2.12552711e-01 -5.43431699e-01
2.58705858e-02 -6.67297900e-01 5.74285500e-02 1.95125610e-01
-1.63298637e-01 -2.07918257e-01 3.63459051e-01 -2.35133603e-01
9.52262759e-01 -1.97943664e+00 6.93830132e-01 9.15970206e-02
3.12339425e-01 1.38346344e-01 -8.90046880e-02 3.12066078e-01
-3.40657383e-02 -2.43450459e-02 -4.63672698e-01 -2.87517756e-01
4.49177086e-01 6.48715019e-01 -6.94000721e-01 3.22785884e-01
3.25899392e-01 1.23157811e+00 -8.32255602e-01 8.10227264e-03
-1.25639409e-01 6.81395903e-02 -8.65264356e-01 2.21001789e-01
-6.66661739e-01 3.25959809e-02 1.27965257e-01 4.16730881e-01
4.30932671e-01 -5.79019725e-01 1.61657646e-01 -1.44965902e-01
2.58834541e-01 1.89192951e-01 -6.24755263e-01 1.96498418e+00
-7.00974107e-01 6.40650749e-01 -2.41121754e-01 -1.31832647e+00
6.71022236e-01 -1.93915159e-01 -5.01813948e-01 -6.17637396e-01
5.73137514e-02 2.31793121e-01 2.80397862e-01 -4.36235458e-01
5.45228660e-01 -3.19930047e-01 7.66166076e-02 6.32713735e-01
5.72731137e-01 -1.21054634e-01 3.21395814e-01 1.06090225e-01
1.09226704e+00 2.30494291e-01 2.94550955e-01 -3.78982425e-01
6.96036041e-01 -1.48175806e-01 -9.63814110e-02 1.06826687e+00
2.56592005e-01 2.63953894e-01 3.51479590e-01 -7.10483015e-01
-1.08613312e+00 -1.74809790e+00 8.52920562e-02 1.68820870e+00
-1.39041662e-01 -2.56834447e-01 -8.19762588e-01 -5.06551325e-01
-4.43219356e-02 8.34928334e-01 -7.45826542e-01 -4.06211942e-01
-6.07705772e-01 -5.66979945e-01 7.57017493e-01 9.29254770e-01
6.52897120e-01 -1.22271621e+00 -1.05568051e+00 4.09276085e-03
3.73013228e-01 -9.85524714e-01 -9.82047394e-02 4.07398015e-01
-1.04867101e+00 -9.48032916e-01 -6.97019339e-01 -9.77521956e-01
8.31689954e-01 -5.26492633e-02 1.45136857e+00 1.05279423e-01
-2.11872950e-01 5.85252702e-01 -6.92435578e-02 -2.46599004e-01
-6.52092397e-01 4.71333057e-01 -7.62291327e-02 -4.59004752e-02
1.66218504e-01 -1.03433132e+00 -5.94130218e-01 1.68330118e-01
-1.32102370e+00 2.27700531e-01 7.35768199e-01 1.02357471e+00
2.03043222e-01 -4.25689891e-02 5.22795856e-01 -9.65787828e-01
7.02982187e-01 -4.51199353e-01 -4.63424385e-01 2.30978206e-01
-5.11835158e-01 5.49959779e-01 1.22955930e+00 -4.09212500e-01
-1.00399518e+00 -1.27998486e-01 -2.41459608e-02 -1.41435519e-01
-1.31120801e-01 4.98893499e-01 2.56960485e-02 -6.59029856e-02
1.12081242e+00 7.12589264e-01 8.80756006e-02 -3.80602479e-01
8.90259326e-01 3.36083531e-01 9.51135814e-01 -8.72109711e-01
7.74689376e-01 5.59219658e-01 1.67209387e-01 -6.77965999e-01
-9.06407714e-01 1.97022438e-01 -5.62194586e-01 1.86055586e-01
7.17094243e-01 -6.51791871e-01 -8.84348571e-01 3.58392388e-01
-1.07921135e+00 -7.38628626e-01 -6.45199358e-01 -5.74856065e-02
-8.79213512e-01 3.65390368e-02 -8.35493624e-01 -3.98012340e-01
-2.95116305e-01 -1.06202185e+00 7.01151967e-01 2.82390356e-01
-1.61051348e-01 -1.13363111e+00 -2.38833785e-01 1.38931662e-01
8.46044183e-01 -1.92526922e-01 1.51301730e+00 -7.07699597e-01
-7.84969926e-01 1.18506253e-01 -3.35393816e-01 7.07832038e-01
-2.32697085e-01 -3.94344449e-01 -1.16258883e+00 -1.88143909e-01
9.35794115e-02 -6.68477178e-01 1.25838625e+00 -1.20820060e-01
1.58559668e+00 -5.07881284e-01 8.09707493e-02 9.45589960e-01
1.37765992e+00 -1.16286702e-01 7.27897525e-01 3.04959595e-01
5.15896738e-01 4.08643126e-01 -4.24391329e-01 1.50742173e-01
5.90775013e-01 2.63792545e-01 5.71926057e-01 2.09965751e-01
-2.47742295e-01 -2.71115303e-01 3.92894119e-01 7.16194630e-01
-3.00991327e-01 -1.45437166e-01 -8.36026669e-01 5.01763463e-01
-1.79735494e+00 -1.05720842e+00 5.06735086e-01 1.97754300e+00
9.91984546e-01 2.95400590e-01 7.84560945e-03 2.06845000e-01
2.98542738e-01 8.48916322e-02 -6.34514868e-01 -7.34259605e-01
-1.77809849e-01 7.53681362e-01 2.47563973e-01 3.87311280e-01
-5.83523750e-01 8.82620573e-01 6.54312038e+00 6.42822862e-01
-9.72177148e-01 4.41548824e-02 5.29962182e-01 8.98799822e-02
-4.32993978e-01 -1.81042776e-01 -4.51806515e-01 8.40234309e-02
1.09806657e+00 3.69443558e-02 1.04824960e+00 7.88483858e-01
-6.74703062e-01 1.61454052e-01 -1.83256805e+00 9.00683701e-01
4.38610613e-01 -1.67591012e+00 5.19141912e-01 -3.06249917e-01
4.96064425e-01 1.51458144e-01 5.01839399e-01 6.34043574e-01
2.79777676e-01 -1.38597035e+00 1.02180684e+00 4.86441195e-01
5.84888756e-01 -4.30183500e-01 3.04256976e-01 4.49421167e-01
-7.18401730e-01 -3.91975254e-01 -5.15551567e-01 -3.85473102e-01
-2.29519978e-01 -9.35693160e-02 -8.00794005e-01 1.41946703e-01
5.84697783e-01 4.68541473e-01 -8.19780409e-01 7.33868659e-01
-4.88270938e-01 3.40691715e-01 -9.60060433e-02 -1.59649506e-01
2.65758306e-01 1.01443894e-01 9.40122530e-02 1.22040915e+00
3.12734187e-01 -1.98207036e-01 -2.49623448e-01 1.31388772e+00
-5.02678514e-01 -3.03169727e-01 -6.14570081e-01 -3.56151275e-02
1.31708205e-01 9.47636604e-01 -5.27276456e-01 -5.01248121e-01
-4.42006975e-01 1.31154513e+00 7.64293611e-01 5.18862367e-01
-7.50046253e-01 -2.76164830e-01 2.81381339e-01 4.64611389e-02
7.62722075e-01 -1.35604292e-01 -3.61506999e-01 -1.15632057e+00
7.77287781e-02 -8.87700319e-01 3.07560027e-01 -1.01113498e+00
-1.53776693e+00 7.31975019e-01 -9.19095203e-02 -6.99789822e-01
-3.02697331e-01 -1.38149905e+00 -3.11203450e-01 7.39598393e-01
-1.35084212e+00 -1.29003143e+00 -2.78276384e-01 7.94601262e-01
5.49148560e-01 -4.98559594e-01 1.13543916e+00 -2.37313174e-02
-1.56800658e-01 7.73711681e-01 -1.42941952e-01 9.09612626e-02
1.50366068e-01 -1.48482263e+00 6.90348864e-01 7.01032758e-01
4.85054821e-01 1.00784111e+00 5.42613566e-01 8.07145759e-02
-1.68862438e+00 -1.01961720e+00 4.72066522e-01 -5.03908098e-01
8.52537632e-01 -4.67374265e-01 -9.84150231e-01 1.03325498e+00
3.23394924e-01 2.27147102e-01 6.85568094e-01 2.06180096e-01
-1.01293933e+00 -2.66773999e-01 -1.01192391e+00 8.48637998e-01
1.38892186e+00 -7.52063394e-01 -1.12571681e+00 1.54634774e-01
7.46861339e-01 -4.89020079e-01 -6.45498157e-01 -5.65976165e-02
6.23989701e-01 -1.14624274e+00 1.09841347e+00 -9.82209384e-01
8.65326762e-01 -2.87450790e-01 -5.23475528e-01 -1.41214943e+00
-4.28429276e-01 -5.29880404e-01 -4.41100448e-01 7.30320632e-01
6.19912744e-01 -7.21799672e-01 6.66575968e-01 3.01802516e-01
-3.12957108e-01 -8.38256121e-01 -7.77395368e-01 -7.81150758e-01
2.99836487e-01 -5.99760890e-01 7.87003160e-01 5.14079392e-01
2.69315094e-02 8.67060959e-01 1.32183850e-01 7.98618793e-02
3.93251508e-01 2.63057053e-01 5.94084084e-01 -1.09021437e+00
-8.86123240e-01 -9.00690198e-01 -5.44129968e-01 -1.41232622e+00
3.32800716e-01 -1.34974205e+00 -1.41366601e-01 -1.42831147e+00
2.53970236e-01 -1.09748669e-01 -1.98922247e-01 5.09385586e-01
1.24491036e-01 2.45426133e-01 3.03137064e-01 2.14656457e-01
-6.22835219e-01 4.76543725e-01 1.17244279e+00 -3.39838356e-01
6.77988380e-02 -4.50603105e-02 -1.05637276e+00 8.78979743e-01
7.11788058e-01 4.45442162e-02 -7.28220820e-01 -9.34839010e-01
6.08798206e-01 -2.99348772e-01 7.05532372e-01 -1.18492711e+00
4.64813024e-01 1.25347629e-01 6.19232476e-01 -2.66439945e-01
3.14801216e-01 -7.83307910e-01 -2.98989922e-01 3.36681902e-01
-5.90851784e-01 3.99290383e-01 3.94751281e-01 5.27053356e-01
8.97260979e-02 -3.82573992e-01 5.76184213e-01 -3.49229544e-01
-8.29548001e-01 1.54610381e-01 -2.59974778e-01 2.55812317e-01
4.94462699e-01 -2.53591865e-01 -7.36551166e-01 -1.94389313e-01
-7.42626786e-01 -2.05508903e-01 5.34166336e-01 2.27658316e-01
7.47885585e-01 -1.11968732e+00 -4.43225443e-01 3.65681767e-01
2.06152678e-01 2.08630830e-01 1.68072388e-01 5.99038959e-01
-8.08719933e-01 2.97725201e-01 -4.19586837e-01 -6.36197269e-01
-4.85873610e-01 8.54271233e-01 4.91257429e-01 3.28823961e-02
-5.48066497e-01 1.13465953e+00 4.02251095e-01 -5.07513821e-01
2.30911091e-01 -8.23346794e-01 2.35093802e-01 -2.88699239e-01
7.47154415e-01 9.92776901e-02 3.27545069e-02 -2.19233572e-01
-9.62369069e-02 2.97294199e-01 -4.62979302e-02 -4.42819707e-02
1.38887048e+00 1.24566972e-01 -4.27802473e-01 5.03245711e-01
1.00294042e+00 -3.29654455e-01 -1.26449072e+00 -4.33804959e-01
-5.05199470e-03 -1.30857572e-01 -2.63361096e-01 -1.00230551e+00
-1.03326309e+00 1.02569437e+00 1.49922282e-01 3.58128995e-01
1.28359509e+00 1.96350694e-01 5.63951373e-01 8.87881696e-01
1.44058615e-01 -7.51012325e-01 3.02057892e-01 7.41842806e-01
9.99908984e-01 -9.33371365e-01 -2.67237276e-01 4.67909873e-02
-4.08543408e-01 1.42551589e+00 5.26585877e-01 -3.48566473e-01
3.59658718e-01 1.32656097e-01 -4.51583654e-01 -1.52429476e-01
-9.35567856e-01 -1.33936509e-01 3.11697364e-01 7.59268522e-01
2.83473462e-01 -1.05442792e-01 4.31868404e-01 7.22058058e-01
-5.68774521e-01 1.19863473e-01 4.44523484e-01 7.85713673e-01
-4.61892724e-01 -6.26200438e-01 -7.59119838e-02 3.52769703e-01
-1.10730305e-01 -4.09788638e-01 -4.53872941e-02 7.81900346e-01
2.00307742e-01 3.79875958e-01 2.02202782e-01 -1.86226159e-01
3.36478472e-01 2.11346343e-01 1.02976239e+00 -6.72220051e-01
-6.56352580e-01 -7.28934884e-01 -4.56707217e-02 -4.69478339e-01
-2.17893496e-01 -2.11088538e-01 -1.20422268e+00 -3.33165646e-01
3.35805684e-01 -3.40830743e-01 4.24555868e-01 1.04992664e+00
2.01799169e-01 8.24322820e-01 5.71382679e-02 -7.60938287e-01
-1.09814763e+00 -7.22581327e-01 -3.41041654e-01 4.90180612e-01
5.07016778e-01 -2.95719057e-01 -2.54608095e-01 2.93564230e-01] | [9.545156478881836, 7.121941566467285] |
72d82175-004a-43c6-abc5-9a11e74bf9e3 | a-high-speed-real-time-vision-system-for | 1812.04115 | null | http://arxiv.org/abs/1812.04115v1 | http://arxiv.org/pdf/1812.04115v1.pdf | A High-Speed, Real-Time Vision System for Texture Tracking and Thread Counting | In garment manufacturing, an automatic sewing machine is desirable to reduce
cost. To accomplish this, a high speed vision system is required to track
fabric motions and recognize repetitive weave patterns with high accuracy, from
a micro perspective near a sewing zone. In this paper, we present an innovative
framework for real-time texture tracking and weave pattern recognition. Our
framework includes a module for motion estimation using blob detection and
feature matching. It also includes a module for lattice detection to facilitate
the weave pattern recognition. Our lattice detection algorithm utilizes blob
detection and template matching to assess pair-wise similarity in blobs'
appearance. In addition, it extracts information of dominant orientations to
obtain a global constraint in the topology. By incorporating both constraints
in the appearance similarity and the global topology, the algorithm determines
a lattice that characterizes the topological structure of the repetitive weave
pattern, thus allowing for thread counting. In our experiments, the proposed
thread-based texture tracking system is capable of tracking denim fabric with
high accuracy (e.g., 0.03 degree rotation and 0.02 weave-thread' translation
errors) and high speed (3 frames per second), demonstrating its high potential
for automatic real-time textile manufacturing. | [] | 2018-12-10 | null | null | null | null | ['template-matching'] | ['computer-vision'] | [ 4.21137452e-01 -6.26214504e-01 -5.12025207e-02 1.78248987e-01
1.62740201e-01 -6.42524660e-01 -1.11837806e-02 2.96857119e-01
-7.47477859e-02 2.17753902e-01 -5.75922489e-01 -5.79004884e-02
-4.09284413e-01 -9.47614133e-01 -5.19572139e-01 -6.81443274e-01
7.77425840e-02 4.54896599e-01 6.14934325e-01 -1.79365784e-01
5.81348300e-01 8.69275987e-01 -1.45686066e+00 1.75245836e-01
5.30714750e-01 1.12611055e+00 6.09718025e-01 7.45340109e-01
9.19573307e-02 2.18981355e-01 -2.75342375e-01 -9.36309993e-02
-2.18132883e-03 -3.08587700e-01 -2.91180313e-01 7.10388958e-01
4.08903867e-01 -2.75158584e-01 2.31771857e-01 9.65828001e-01
1.71442211e-01 -2.49529734e-01 5.73622763e-01 -6.51835918e-01
-1.08366862e-01 -8.76147151e-02 -7.80705333e-01 -1.15273878e-01
5.82969129e-01 -6.15149289e-02 6.17240131e-01 -7.89929688e-01
1.23926616e+00 7.00283229e-01 5.96449792e-01 1.57782659e-01
-1.40080154e+00 -3.60054374e-01 -3.91329825e-01 1.01174958e-01
-1.35769725e+00 -1.35037526e-01 1.25126839e+00 -6.56743884e-01
5.15318155e-01 7.18908370e-01 1.10099220e+00 6.14470899e-01
8.67867351e-01 3.05963576e-01 1.23170185e+00 -8.27360570e-01
2.79990703e-01 -9.77637470e-02 1.83286384e-01 1.05496633e+00
3.05297673e-01 -6.24480993e-02 -3.42512935e-01 7.16808438e-02
1.72120810e+00 -5.21329381e-02 -1.42112002e-01 -7.67591655e-01
-1.42495096e+00 3.15561503e-01 1.63515806e-01 4.68138248e-01
-3.55283260e-01 3.32391821e-02 2.44944260e-01 1.37600988e-01
2.71374911e-01 3.31768900e-01 -1.12679787e-01 -1.95445880e-01
-6.22701705e-01 1.23446472e-01 8.91328096e-01 9.33548331e-01
5.63608766e-01 -1.49682358e-01 1.28192604e-01 7.28156626e-01
1.54403195e-01 6.38763666e-01 2.43789726e-03 -7.25934684e-01
1.56692386e-01 7.66467631e-01 2.76579648e-01 -1.62563789e+00
-5.42398930e-01 -1.24924429e-01 -8.99296403e-01 5.54592311e-01
5.75254619e-01 2.79247224e-01 -6.87481701e-01 8.62214446e-01
5.53431273e-01 -3.37363929e-01 -4.67855036e-01 1.06601357e+00
2.16843918e-01 3.43591005e-01 -5.31181872e-01 -4.05663937e-01
1.75942314e+00 -4.55094248e-01 -8.75639498e-01 4.34997439e-01
3.47627074e-01 -1.58065212e+00 1.00117350e+00 6.62215114e-01
-8.70758116e-01 -5.81120729e-01 -1.07445300e+00 3.98577631e-01
7.35392421e-02 7.38306284e-01 3.51772785e-01 4.68608111e-01
-4.14391965e-01 6.16702557e-01 -9.55634594e-01 -5.57010472e-01
-2.01482967e-01 3.44779789e-01 -2.73580492e-01 1.39962494e-01
-2.67245740e-01 6.72732651e-01 1.21759728e-01 4.31871176e-01
7.50700608e-02 -2.76866734e-01 -5.89860559e-01 -1.78428873e-01
3.64156663e-01 -4.89967048e-01 8.08626950e-01 -6.24041021e-01
-1.81738865e+00 8.05852175e-01 -1.41288117e-01 2.11658418e-01
7.30724514e-01 -1.77784115e-01 -3.58562261e-01 1.72471195e-01
-4.70098332e-02 -2.52576023e-01 9.45046961e-01 -1.15765834e+00
-3.90346378e-01 -4.78043586e-01 -3.80667716e-01 -1.26054570e-01
7.40724523e-03 -9.23076179e-03 -3.30022335e-01 -9.11828518e-01
8.63109648e-01 -7.90802121e-01 4.05340642e-02 3.88622612e-01
-3.96136969e-01 1.08579900e-02 1.11579180e+00 -7.03100801e-01
1.15761232e+00 -2.13808441e+00 1.15863815e-01 7.00340331e-01
1.49655819e-01 -1.25708073e-01 2.70259678e-01 3.85289580e-01
5.04846513e-01 -6.14721060e-01 -2.77963642e-04 2.26055384e-01
-2.91175932e-01 -5.59294857e-02 1.92356318e-01 6.72001600e-01
-5.48899807e-02 2.57089734e-01 -7.42966056e-01 -3.90440822e-01
4.38022405e-01 2.96838075e-01 -3.05700004e-01 -4.59754318e-02
-5.44207590e-03 2.57439137e-01 -5.16453862e-01 1.04610503e+00
7.86921144e-01 -1.23685673e-01 4.52932060e-01 -7.70165145e-01
-6.18154287e-01 -4.65254247e-01 -1.59536254e+00 1.32794917e+00
-3.85542840e-01 6.52388453e-01 4.31184173e-01 -3.38228375e-01
1.57264256e+00 1.69362634e-01 6.16176486e-01 -6.27906144e-01
2.48448461e-01 5.01598716e-01 -1.34489551e-01 -5.39883435e-01
5.77678323e-01 1.69588864e-01 1.81706786e-01 3.86045486e-01
-4.02012140e-01 -9.49410275e-02 1.24237679e-01 -6.11720324e-01
7.28851318e-01 3.11405897e-01 1.96624517e-01 -5.90180397e-01
5.91029346e-01 1.94080874e-01 4.96793360e-01 2.26999387e-01
2.04030082e-01 3.26867551e-01 1.39237180e-01 -1.05017579e+00
-1.41540766e+00 -1.08709955e+00 -1.78622678e-01 4.10754055e-01
7.23612428e-01 -3.44111621e-01 -7.01197863e-01 -3.96679975e-02
2.05419555e-01 -8.70652646e-02 -3.19597483e-01 1.93746760e-01
-9.24245775e-01 -4.57611769e-01 -2.85425663e-01 3.30318362e-01
4.90443856e-01 -9.20313716e-01 -1.23101854e+00 5.57969213e-01
9.96180996e-02 -1.01050258e+00 -4.70128179e-01 -2.26082519e-01
-9.76080418e-01 -1.24376345e+00 -6.35032117e-01 -9.74517703e-01
9.29092884e-01 1.82149261e-01 7.93747783e-01 -4.17805975e-03
-9.50066328e-01 2.92489022e-01 -5.04512966e-01 1.66710243e-02
-3.83838266e-01 -2.09627837e-01 -5.03614359e-02 1.97941229e-01
-4.03659120e-02 -3.65942806e-01 -8.03889871e-01 9.07015860e-01
-3.98060590e-01 3.28623831e-01 5.26660681e-01 6.08817995e-01
8.32535028e-01 8.89032520e-03 -1.54236481e-01 -3.44060153e-01
5.35802007e-01 2.80025721e-01 -9.69945788e-01 3.22088987e-01
-3.05177003e-01 -3.35054755e-01 3.19016635e-01 -6.35685146e-01
-7.45998800e-01 9.25525054e-02 1.23473145e-01 -2.33547479e-01
-1.41122401e-01 1.88078955e-01 2.77631164e-01 -4.00605381e-01
3.20225209e-01 4.00764763e-01 2.93423086e-01 -6.22714698e-01
-1.25892237e-01 5.21703482e-01 3.48461717e-01 -6.14131331e-01
5.69153547e-01 5.10135949e-01 2.21564233e-01 -1.24838269e+00
-9.75132063e-02 -4.07175124e-01 -9.46675062e-01 -9.15839136e-01
9.48651373e-01 -3.12289745e-01 -1.28014600e+00 7.22066939e-01
-1.35052872e+00 -7.77249187e-02 3.80069539e-02 7.11319089e-01
-7.64198840e-01 5.13922095e-01 -9.20774102e-01 -8.03387165e-01
-4.52518314e-01 -1.03013253e+00 9.02374983e-01 3.31453700e-03
-5.21467030e-01 -7.14724720e-01 -8.05040523e-02 4.24867533e-02
2.34134465e-01 7.22193599e-01 7.84249842e-01 5.48408806e-01
-8.62567306e-01 -2.93579400e-01 -1.82780847e-01 -1.11650705e-01
5.37690818e-01 3.73315394e-01 -2.10151151e-01 -3.28346193e-01
-6.86503109e-03 2.95986384e-01 2.40534157e-01 4.61258382e-01
7.26155698e-01 -6.85627237e-02 -3.12225997e-01 4.52159971e-01
1.59410298e+00 3.95511538e-01 3.97509336e-01 6.41560137e-01
6.95787370e-01 5.73744118e-01 1.17081022e+00 5.99697173e-01
-2.15571851e-01 1.20224953e+00 2.69538313e-01 -2.37999573e-01
-1.23928390e-01 2.21524715e-01 1.62564397e-01 1.05103362e+00
-8.22066665e-01 3.23972166e-01 -7.39461660e-01 1.36714637e-01
-1.69907928e+00 -7.04876482e-01 -5.49346089e-01 2.24857378e+00
4.83768344e-01 1.70660004e-01 2.16256648e-01 3.24305713e-01
9.70335066e-01 -2.61024714e-01 -2.20889136e-01 -7.42777348e-01
2.01022834e-01 5.81656508e-02 5.12778580e-01 4.61104244e-01
-8.57515037e-01 8.55039895e-01 5.71913004e+00 6.29569352e-01
-1.35064816e+00 -3.43152404e-01 4.14588079e-02 4.59134132e-01
1.76499933e-02 -4.18340415e-01 -6.42634273e-01 4.85865355e-01
-9.97766107e-02 3.10505986e-01 1.93539158e-01 7.23268330e-01
2.75640011e-01 -3.11629087e-01 -6.51733220e-01 1.08611333e+00
-3.79123166e-02 -1.40399241e+00 -8.95069987e-02 2.18898371e-01
3.56253624e-01 -6.98387086e-01 -2.83141285e-01 -7.80377328e-01
-2.90666133e-01 -3.23901176e-01 9.72426355e-01 6.84036613e-01
7.80035079e-01 -7.61307001e-01 5.56100845e-01 6.05350100e-02
-1.60343826e+00 2.07292348e-01 -3.29731703e-01 -1.15156278e-01
1.98593378e-01 8.36637318e-01 -7.88777053e-01 6.09394014e-01
4.41774547e-01 4.96119618e-01 -6.67597428e-02 1.13921773e+00
1.45771578e-01 3.02136112e-02 -4.41852272e-01 -3.69234681e-01
-4.36816476e-02 -7.60443330e-01 8.25624049e-01 1.15374923e+00
7.08727896e-01 -1.10542804e-01 2.59615451e-01 8.60976577e-01
5.94989896e-01 2.69219667e-01 -2.31139198e-01 3.13027710e-01
4.44618464e-01 1.18695974e+00 -1.49298728e+00 1.09368879e-02
6.43199012e-02 1.18875802e+00 -1.07766435e-01 -5.52831264e-03
-5.94353795e-01 -5.49922705e-01 4.21623766e-01 4.28464949e-01
4.34798121e-01 -8.46213639e-01 -5.12194932e-01 -8.56979430e-01
2.75924206e-01 -4.09954458e-01 -1.40477628e-01 -6.67318165e-01
-9.13818419e-01 4.30444568e-01 -5.00215352e-01 -1.39958668e+00
2.02663600e-01 -9.25746560e-01 -3.88190061e-01 6.41931295e-01
-8.56599867e-01 -1.37628365e+00 -4.89567459e-01 6.81083977e-01
6.95058167e-01 1.31737307e-01 8.81806374e-01 1.14019819e-01
-4.82870698e-01 1.89048707e-01 1.76393121e-01 -3.18823941e-02
5.01507521e-01 -9.80160415e-01 8.30458254e-02 5.08638203e-01
-5.46142720e-02 5.10730386e-01 8.25688422e-01 -9.09444273e-01
-1.87764895e+00 -6.14618242e-01 6.31298363e-01 -1.13480978e-01
5.50586641e-01 -4.78949100e-01 -5.94106436e-01 2.55024940e-01
-3.36677253e-01 -2.98618972e-01 3.82278204e-01 -3.21575161e-03
5.17750829e-02 -3.61256778e-01 -1.06698477e+00 4.42914099e-01
8.51717949e-01 -1.40392855e-01 -2.12673157e-01 1.18304655e-01
-1.45558655e-01 -6.85859919e-01 -1.32210326e+00 5.27563572e-01
1.35536766e+00 -8.04961801e-01 9.00372982e-01 2.60778129e-01
2.31492832e-01 -5.50194204e-01 1.86292425e-01 -6.54255748e-01
-7.06276238e-01 -6.92238867e-01 7.03062117e-02 9.53349054e-01
-8.39088857e-02 -5.08723199e-01 9.05857623e-01 -1.93812642e-02
7.56651536e-02 -6.71885312e-01 -8.92292261e-01 -8.78044486e-01
-9.00594532e-01 1.25629470e-01 2.64885217e-01 9.07294035e-01
-8.11360031e-02 -1.46096691e-01 -2.79243946e-01 2.77549505e-01
9.15801764e-01 7.75018275e-01 6.35001481e-01 -1.32594943e+00
-1.06460720e-01 -3.23083103e-01 -8.02366614e-01 -8.44748735e-01
-6.34246230e-01 -3.28621984e-01 -1.30070612e-01 -1.16416705e+00
-5.54950163e-03 -6.42352462e-01 -3.87049280e-02 6.58514723e-02
4.26356256e-01 4.31517392e-01 1.91576764e-01 3.31974417e-01
-9.57525447e-02 1.02902604e-02 1.60096920e+00 1.80852979e-01
-3.57434243e-01 2.25158073e-02 3.15650582e-01 7.06831336e-01
8.20271552e-01 5.17414184e-03 1.41062394e-01 -4.18852597e-01
6.82648420e-02 -3.40397633e-03 5.20552099e-01 -1.11556768e+00
2.23690867e-01 -1.69559106e-01 5.33605397e-01 -6.70362771e-01
2.39780307e-01 -1.07933056e+00 4.29544300e-01 9.79169905e-01
1.97841316e-01 2.38807783e-01 6.54608905e-02 3.56565416e-01
6.50758296e-02 -5.65741807e-02 7.16037333e-01 -5.26566617e-02
-7.63650835e-01 -1.02934740e-01 -5.18873155e-01 -1.01627505e+00
1.29441726e+00 -9.04506087e-01 -5.78186959e-02 3.72366011e-01
-1.03909373e+00 -3.29263508e-01 9.19460654e-01 1.65287092e-01
9.68930602e-01 -1.34888875e+00 -3.50406796e-01 6.84990585e-01
2.07912438e-02 -3.09058696e-01 1.89876795e-01 8.33366215e-01
-1.26881146e+00 -4.75097112e-02 -7.38958061e-01 -1.08349371e+00
-1.79618430e+00 1.32760435e-01 -4.61036013e-03 -1.48596913e-02
-8.07241380e-01 3.16377461e-01 -4.30453032e-01 -1.27166938e-02
-1.99242860e-01 -6.97165966e-01 -3.58636715e-02 -1.69498473e-01
3.29720646e-01 7.03682363e-01 7.03120902e-02 -4.26974744e-01
-2.50849456e-01 1.51860785e+00 2.20094383e-01 2.41533875e-01
1.23155713e+00 -2.06618503e-01 -4.58838850e-01 4.07324910e-01
7.43341804e-01 3.95549744e-01 -1.04181695e+00 2.63541371e-01
7.89705813e-02 -7.94354856e-01 -1.86230645e-01 -5.69548666e-01
-7.19920158e-01 5.11146545e-01 7.21388042e-01 4.77800101e-01
1.02746034e+00 -1.60350472e-01 7.61366963e-01 5.26235141e-02
8.42888355e-01 -1.13404477e+00 -8.56173262e-02 1.97591797e-01
1.00426960e+00 -6.48798704e-01 7.36640170e-02 -1.11912906e+00
-4.01382670e-02 1.86025321e+00 3.00455809e-01 -4.07947451e-01
4.47165281e-01 4.61006463e-01 -1.53229674e-02 -4.59165961e-01
-4.91494909e-02 1.40936762e-01 3.74572068e-01 3.94014925e-01
2.78385580e-01 2.76475638e-01 -6.88620090e-01 -1.32611394e-01
-1.37039617e-01 1.44344121e-01 3.29423606e-01 1.23412180e+00
-7.82146871e-01 -1.10331535e+00 -6.90676868e-01 1.31597534e-01
-1.27718270e-01 6.42520308e-01 -4.84429374e-02 9.33689356e-01
7.81098381e-02 5.48078358e-01 3.78457516e-01 -5.15719712e-01
6.53735936e-01 -4.65424508e-01 1.02351093e+00 -2.08157495e-01
-2.45788425e-01 6.29930317e-01 1.72402948e-01 -7.40323365e-01
-3.79958391e-01 -5.51538587e-01 -7.80397117e-01 -3.90518725e-01
-3.62377137e-01 -8.17946494e-02 9.72268283e-01 5.27909160e-01
1.89510375e-01 3.52868646e-01 6.72845900e-01 -6.91241384e-01
-9.89960954e-02 -5.03130376e-01 -1.09018922e+00 4.76535112e-01
2.17923452e-03 -8.07444692e-01 1.61218449e-01 2.73540676e-01] | [8.455044746398926, -2.3744373321533203] |
e7e11b4a-81a5-4714-973b-3c75d566f74d | towards-sequence-utility-maximization-under | 2212.10452 | null | https://arxiv.org/abs/2212.10452v1 | https://arxiv.org/pdf/2212.10452v1.pdf | Towards Sequence Utility Maximization under Utility Occupancy Measure | The discovery of utility-driven patterns is a useful and difficult research topic. It can extract significant and interesting information from specific and varied databases, increasing the value of the services provided. In practice, the measure of utility is often used to demonstrate the importance, profit, or risk of an object or a pattern. In the database, although utility is a flexible criterion for each pattern, it is a more absolute criterion due to the neglect of utility sharing. This leads to the derived patterns only exploring partial and local knowledge from a database. Utility occupancy is a recently proposed model that considers the problem of mining with high utility but low occupancy. However, existing studies are concentrated on itemsets that do not reveal the temporal relationship of object occurrences. Therefore, this paper towards sequence utility maximization. We first define utility occupancy on sequence data and raise the problem of High Utility-Occupancy Sequential Pattern Mining (HUOSPM). Three dimensions, including frequency, utility, and occupancy, are comprehensively evaluated in HUOSPM. An algorithm called Sequence Utility Maximization with Utility occupancy measure (SUMU) is proposed. Furthermore, two data structures for storing related information about a pattern, Utility-Occupancy-List-Chain (UOL-Chain) and Utility-Occupancy-Table (UO-Table) with six associated upper bounds, are designed to improve efficiency. Empirical experiments are carried out to evaluate the novel algorithm's efficiency and effectiveness. The influence of different upper bounds and pruning strategies is analyzed and discussed. The comprehensive results suggest that the work of our algorithm is intelligent and effective. | ['Philip S. Yu', 'Wensheng Gan', 'Gengsen Huang'] | 2022-12-20 | null | null | null | null | ['sequential-pattern-mining'] | ['natural-language-processing'] | [ 3.04163188e-01 -2.27757528e-01 -7.94957101e-01 -3.02413553e-01
-1.43562809e-01 -1.66486934e-01 -4.62580994e-02 1.87702939e-01
-1.94968611e-01 1.08549666e+00 6.76668808e-02 -3.26548159e-01
-7.15348959e-01 -1.09740150e+00 -1.50616944e-01 -6.67559206e-01
-4.62794989e-01 3.94728631e-01 3.62738967e-01 1.80157591e-02
4.72984880e-01 4.69631225e-01 -2.02090716e+00 3.14450502e-01
6.59113765e-01 1.26208103e+00 5.33598363e-01 4.68771271e-02
-3.80531192e-01 7.30167985e-01 -5.74618578e-01 -1.29736245e-01
2.16729879e-01 -5.71970701e-01 -9.18016791e-01 2.08462238e-01
-1.06342733e+00 -2.56245553e-01 3.03609371e-01 8.06885540e-01
1.08076625e-01 1.53835282e-01 5.37802696e-01 -1.55681908e+00
-1.38412774e-01 7.42439270e-01 -7.05886126e-01 4.27041680e-01
5.16922534e-01 -2.76490301e-01 1.09370458e+00 -6.73330724e-01
6.02701068e-01 8.39369535e-01 1.77181214e-01 -6.54918477e-02
-8.69040847e-01 -4.94607300e-01 1.14055380e-01 7.20371187e-01
-1.61637664e+00 -1.99706424e-02 5.56488514e-01 -1.81705847e-01
1.22717369e+00 7.18491435e-01 6.75942779e-01 3.91296566e-01
2.08806723e-01 8.46892715e-01 8.76029432e-01 -5.94176829e-01
4.12474036e-01 2.92021602e-01 3.68753731e-01 2.65041769e-01
7.32182860e-01 3.79348397e-02 -7.16518939e-01 -4.22582269e-01
3.87851387e-01 2.33498052e-01 -2.36597821e-01 -3.55611891e-01
-7.06047714e-01 6.08096242e-01 -3.58839214e-01 4.03317004e-01
-5.76097846e-01 -5.85248291e-01 2.71512091e-01 4.90482926e-01
-8.89607370e-02 1.37828842e-01 -4.09378767e-01 -4.97736841e-01
-6.93306148e-01 3.23869914e-01 1.03760076e+00 1.36683345e+00
8.43827128e-01 -3.02634272e-03 -1.80699989e-01 6.02657735e-01
8.83053094e-02 6.84562547e-04 5.82278490e-01 -4.62444872e-01
9.84635763e-03 1.21316600e+00 9.50767249e-02 -8.11075985e-01
-3.21863264e-01 -1.39080599e-01 -6.94412112e-01 -1.70380175e-01
-3.12248841e-02 1.56727452e-02 -4.10607219e-01 1.49845803e+00
3.73665571e-01 -3.13033044e-01 -8.70254636e-02 6.83441043e-01
2.86512136e-01 8.43346775e-01 -1.19475618e-01 -1.17230380e+00
1.27182496e+00 -3.24035645e-01 -7.80692756e-01 4.04397368e-01
3.16492349e-01 -6.39112294e-01 8.29402983e-01 5.94384074e-01
-1.01623940e+00 -1.98802650e-01 -1.07892406e+00 5.20399332e-01
-3.99311334e-01 -5.38597167e-01 6.15385234e-01 7.69324780e-01
-5.08572996e-01 5.18702447e-01 -4.73660380e-01 -2.76354283e-01
1.18308030e-01 3.06478888e-01 5.19297272e-02 -6.64364398e-02
-1.17190862e+00 7.47672737e-01 7.15840042e-01 -5.01657069e-01
-3.87900412e-01 -4.94496077e-01 -4.87858891e-01 3.60983461e-01
7.22680390e-01 -3.20751131e-01 9.60492432e-01 -5.05571008e-01
-8.18424463e-01 4.13550198e-01 -3.40626448e-01 -5.19763887e-01
2.12344259e-01 3.01331520e-01 -6.47964656e-01 -1.37784228e-01
1.33437112e-01 -2.52242714e-01 2.92311609e-01 -9.68861580e-01
-1.26025879e+00 -3.34723055e-01 -2.36460954e-01 1.06657401e-01
-4.36070919e-01 1.40554518e-01 -3.55538636e-01 -5.74956894e-01
3.85809280e-02 -3.44608605e-01 -3.47246975e-01 -7.02795506e-01
-1.10099733e-01 -3.55736852e-01 7.34504998e-01 -3.68070751e-01
2.28094029e+00 -1.87347496e+00 -2.37146765e-01 8.87641907e-01
-3.05816770e-01 -1.72161520e-01 3.31587613e-01 7.40063846e-01
1.81354582e-01 9.97316167e-02 -4.80771780e-01 5.49930453e-01
-9.68616754e-02 6.91852391e-01 -4.92898189e-02 5.21199852e-02
-3.61412913e-02 4.82273161e-01 -6.04959905e-01 -6.33973837e-01
1.81107130e-02 -2.08801955e-01 -5.22319317e-01 2.74817526e-01
-9.89056453e-02 -3.30486685e-01 -4.57204461e-01 1.22220445e+00
5.86011112e-01 -1.72864527e-01 7.43812025e-01 9.25807431e-02
-5.17674565e-01 -7.48223439e-02 -1.57552695e+00 8.99948061e-01
-1.16527192e-01 -2.25422129e-01 -2.24726766e-01 -1.14574754e+00
1.19736540e+00 2.52221256e-01 1.04556417e+00 -8.13960552e-01
6.24773130e-02 2.57155508e-01 1.72704518e-01 -5.74659944e-01
6.85986161e-01 3.45873157e-03 -1.07456870e-01 7.29062855e-01
-3.15630883e-01 7.21337616e-01 7.60339558e-01 -1.13764711e-01
1.30414414e+00 -1.92028299e-01 1.17144251e+00 -6.04377747e-01
5.80270350e-01 1.81681260e-01 9.12074208e-01 4.41431582e-01
-1.02915494e-02 1.27994284e-01 6.53472960e-01 -3.93083990e-01
-9.91462052e-01 -9.01021719e-01 -2.67155349e-01 8.84786546e-01
4.70350772e-01 -4.50951725e-01 -2.41295904e-01 -2.58413166e-01
1.91863522e-01 6.96944177e-01 -2.53289610e-01 5.69355302e-02
-3.10024977e-01 -1.14330459e+00 -2.82669347e-02 3.71360451e-01
2.26682633e-01 -1.26006591e+00 -1.13614035e+00 5.97310603e-01
-2.69090801e-01 -5.26226878e-01 -2.69528687e-01 3.75908107e-01
-7.15598345e-01 -1.26069462e+00 -1.70964062e-01 -5.44849336e-01
4.24487889e-01 3.53568733e-01 1.12919736e+00 4.60048728e-02
-4.42042410e-01 9.63432044e-02 -9.05378580e-01 -6.83544576e-01
-5.47532327e-02 6.72769248e-02 1.32198811e-01 1.40438274e-01
8.93426001e-01 -9.28275645e-01 -3.25075120e-01 6.69915020e-01
-9.60346937e-01 -3.50940913e-01 9.02818561e-01 7.91414917e-01
8.77587616e-01 6.34979963e-01 9.69842017e-01 -7.84428656e-01
9.74222839e-01 -8.22924435e-01 -4.56577271e-01 4.38708603e-01
-1.34112811e+00 1.76428080e-01 4.24024701e-01 -2.19349653e-01
-1.06313527e+00 -9.01341662e-02 1.06178708e-01 -2.12260336e-01
1.15167089e-01 7.84443319e-01 -6.14199698e-01 5.03811717e-01
2.55558074e-01 6.41197085e-01 6.18963204e-02 -3.87778223e-01
-1.54367298e-01 6.58222854e-01 2.67890543e-01 -6.93482101e-01
3.28110427e-01 1.72580674e-01 1.33992404e-01 -9.24970627e-01
-1.39663398e-01 -9.45205331e-01 -2.16231570e-01 -1.87868610e-01
1.71380699e-01 -1.42626449e-01 -1.03220403e+00 -1.63182944e-01
-7.10105002e-01 3.58487427e-01 -3.23028862e-01 1.03747398e-01
-5.46658397e-01 4.55354929e-01 -3.62964123e-02 -1.37024987e+00
-4.70643640e-01 -8.46566141e-01 1.35540411e-01 3.17609496e-02
-5.95706522e-01 -2.58343637e-01 -2.07895070e-01 -8.43521506e-02
7.30335116e-02 3.40274692e-01 1.10635519e+00 -7.32583404e-01
-8.55416417e-01 -1.30126074e-01 -8.36732984e-02 -2.89066564e-02
2.86850601e-01 -2.78981417e-01 -1.71575353e-01 6.13039881e-02
3.81594263e-02 2.31002811e-02 2.81741560e-01 3.15435559e-01
1.31830013e+00 -7.62378037e-01 -4.61024165e-01 2.51904935e-01
1.53295612e+00 9.83773053e-01 7.50230312e-01 6.22303963e-01
-1.75994903e-01 8.29246223e-01 1.40689564e+00 1.29305077e+00
8.92738253e-02 5.65709114e-01 4.11394089e-01 5.03903747e-01
6.25315785e-01 -8.99767429e-02 1.16643526e-01 5.92420280e-01
-2.90856183e-01 -3.69188875e-01 -4.73003566e-01 8.32518756e-01
-2.08059263e+00 -1.29181647e+00 -4.52558622e-02 2.31852341e+00
8.04883778e-01 3.86750907e-01 5.58814466e-01 8.66402745e-01
6.15912855e-01 -3.22236598e-01 -5.34570515e-01 -8.49636495e-01
-1.01182915e-01 8.95933807e-02 4.66490239e-01 1.89078897e-01
-4.97988641e-01 1.32141143e-01 6.17452765e+00 1.10076535e+00
-4.98130530e-01 -2.59741902e-01 5.35183966e-01 -2.23866343e-01
-5.57194769e-01 -7.44644031e-02 -9.37276602e-01 7.02402174e-01
6.70607865e-01 -8.18000734e-01 2.31747195e-01 1.20240164e+00
2.33811200e-01 -4.92642850e-01 -9.92978036e-01 9.02365863e-01
-3.39221865e-01 -1.17650640e+00 2.20596611e-01 4.69463736e-01
4.48054522e-01 -6.51518226e-01 -2.26155862e-01 -8.71217623e-02
1.25607237e-01 -6.90566123e-01 4.79198933e-01 4.72467721e-01
6.63223386e-01 -1.46537018e+00 7.78003514e-01 6.32456064e-01
-1.44953239e+00 -2.99957722e-01 -4.17844981e-01 -4.03193474e-01
3.89032215e-01 7.60153651e-01 -9.41534698e-01 9.68721986e-01
9.59468365e-01 2.53956199e-01 1.36595666e-01 7.80312657e-01
3.65061134e-01 3.94513935e-01 -4.13367271e-01 -5.62699974e-01
1.06919743e-02 -4.23517376e-01 3.98776442e-01 1.10639954e+00
5.64274907e-01 3.79765719e-01 2.38531396e-01 6.16559148e-01
3.58291566e-01 4.96227354e-01 -6.39320672e-01 1.87955096e-01
8.76842558e-01 9.33100402e-01 -5.71199536e-01 -1.51673183e-01
-5.07535100e-01 2.76661754e-01 -9.59082693e-02 -1.07274860e-01
-5.25505543e-01 -4.19506848e-01 7.28964627e-01 5.16315758e-01
3.25390846e-01 9.58689228e-02 -7.01613843e-01 -4.93723273e-01
3.69296551e-01 -7.52514958e-01 9.22984660e-01 -5.30992672e-02
-1.02138901e+00 4.00284797e-01 4.27049190e-01 -1.43497717e+00
-4.95011777e-01 -3.07668626e-01 -3.79107207e-01 8.42703104e-01
-1.14566720e+00 -4.82167214e-01 8.19821358e-02 5.40205479e-01
5.31954408e-01 -4.57807690e-01 6.30366504e-01 2.66869932e-01
-4.80424821e-01 7.22659528e-01 1.34498119e-01 -5.06460130e-01
-5.15549183e-02 -8.42414975e-01 -3.50435823e-01 9.62803125e-01
-2.42361739e-01 7.32344270e-01 7.03372538e-01 -7.36202180e-01
-1.38984239e+00 -7.78079391e-01 1.08168542e+00 -3.70804332e-02
4.36763227e-01 6.67623878e-02 -9.36108232e-01 2.10956872e-01
-6.40343353e-02 -6.02609575e-01 1.01162338e+00 8.78562927e-02
8.16523954e-02 -2.92981446e-01 -1.46015060e+00 5.59753001e-01
1.32313442e+00 2.45545596e-01 -5.35422683e-01 -3.71652514e-01
5.63743532e-01 3.56952429e-01 -1.01073074e+00 6.39188766e-01
9.28785741e-01 -1.30955660e+00 7.80647516e-01 -3.22583348e-01
2.70009011e-01 -3.14147115e-01 -4.21272099e-01 -5.82747221e-01
-5.36921263e-01 -4.86330271e-01 -5.56796312e-01 1.19774032e+00
4.90751863e-01 -4.56744671e-01 9.40528333e-01 5.45838475e-01
5.28849661e-02 -1.25452304e+00 -9.37329471e-01 -1.11931789e+00
-7.80025244e-01 -4.32418883e-01 1.06520855e+00 8.13005447e-01
6.93372726e-01 2.23859131e-01 -5.57113349e-01 -1.87704101e-01
6.86075032e-01 5.64685225e-01 5.88412166e-01 -1.16373241e+00
-2.86112428e-01 -5.57068467e-01 -2.78406769e-01 -6.53490007e-01
-5.18406749e-01 -6.31747186e-01 -3.58993739e-01 -1.27729416e+00
3.54611725e-01 -3.28473151e-01 -5.74478030e-01 4.03380752e-01
8.06459188e-02 -3.39462489e-01 -1.53070003e-01 4.18792963e-01
-4.18149441e-01 2.14898929e-01 9.01657760e-01 1.52242512e-01
-4.90658969e-01 2.95520127e-01 -7.45331407e-01 3.23492616e-01
8.11247766e-01 -2.87774712e-01 -7.98992574e-01 4.70061630e-01
1.03902318e-01 2.33651251e-01 -4.65964466e-01 -5.73544800e-01
1.67227089e-01 -7.38509417e-01 1.57386541e-01 -1.14919329e+00
-3.26565541e-02 -1.13701904e+00 6.55559123e-01 7.19144523e-01
5.34904711e-02 2.32793421e-01 -1.94990471e-01 5.27359664e-01
-4.09037799e-01 -5.36939442e-01 5.41718304e-01 -2.65497208e-01
-1.23424113e+00 3.57948720e-01 -4.40057129e-01 -4.19311732e-01
1.32872105e+00 -7.86441982e-01 3.17592293e-01 -1.51792839e-01
-4.83410269e-01 4.34236854e-01 2.70811230e-01 2.71632463e-01
8.09621155e-01 -1.38766146e+00 -4.53684390e-01 2.67323285e-01
4.34655666e-01 -1.29969567e-01 2.02024549e-01 4.92212504e-01
-2.31557459e-01 5.68212032e-01 -4.15766686e-01 -3.61493945e-01
-1.41869462e+00 9.88231063e-01 -4.19046938e-01 -7.00547576e-01
-2.93438286e-01 2.75908560e-01 -2.78198086e-02 1.39237329e-01
2.10894585e-01 -8.15807283e-02 -3.94509554e-01 2.57723600e-01
7.14626312e-01 8.88648033e-01 1.40586309e-02 -1.40770271e-01
-4.58156139e-01 5.88624142e-02 -6.15962334e-02 1.76910944e-02
1.18534017e+00 -2.70527244e-01 -3.92728388e-01 2.70339280e-01
6.78336799e-01 -3.17175686e-01 -5.43621719e-01 -2.86256284e-01
7.56931245e-01 -5.43500185e-01 -6.29493237e-01 -5.61293066e-01
-5.12575686e-01 2.86909807e-02 2.18560696e-01 5.68347335e-01
1.57509577e+00 -1.96963087e-01 6.75929606e-01 2.08322406e-01
8.03402662e-01 -1.24239314e+00 -1.66293725e-01 2.82172948e-01
7.28439629e-01 -7.96894491e-01 6.17432855e-02 -4.79474157e-01
-7.19340801e-01 7.58008420e-01 7.46236384e-01 4.19493228e-01
7.92643189e-01 6.28879786e-01 -5.34244776e-01 1.08317278e-01
-9.26559806e-01 -4.85398263e-01 -1.22171663e-01 6.60389423e-01
3.09419096e-01 1.88715041e-01 -1.19123876e+00 1.16758394e+00
-2.59643346e-01 2.29126409e-01 4.49139059e-01 1.44620299e+00
-8.62940788e-01 -1.45711267e+00 -5.21297753e-01 7.87037849e-01
-5.14611900e-01 1.59498259e-01 -2.15930969e-01 6.85118556e-01
2.71243334e-01 9.71998632e-01 -1.21899359e-02 -6.01133049e-01
5.72166324e-01 1.81082308e-01 1.42764375e-01 -2.82977402e-01
-8.72461870e-02 1.35006711e-01 2.10026860e-01 -4.22091156e-01
-3.63927364e-01 -7.47234523e-01 -1.25416565e+00 -5.33580780e-01
-1.49677500e-01 6.15909755e-01 2.44521022e-01 7.02523828e-01
2.86187202e-01 4.39616084e-01 8.74499917e-01 1.64206058e-01
-4.98715073e-01 -7.48401582e-01 -1.08860826e+00 2.30597243e-01
-2.79240936e-01 -7.71362066e-01 1.21005543e-01 -9.08725932e-02] | [8.297300338745117, 6.281327247619629] |
9d37eee8-ca44-42ae-9f5f-e98cc7522bc7 | ufact-unfaithful-alien-corpora-training-for | null | null | https://aclanthology.org/2022.findings-acl.223 | https://aclanthology.org/2022.findings-acl.223.pdf | uFACT: Unfaithful Alien-Corpora Training for Semantically Consistent Data-to-Text Generation | We propose uFACT (Un-Faithful Alien Corpora Training), a training corpus construction method for data-to-text (d2t) generation models. We show that d2t models trained on uFACT datasets generate utterances which represent the semantic content of the data sources more accurately compared to models trained on the target corpus alone. Our approach is to augment the training set of a given target corpus with alien corpora which have different semantic representations. We show that while it is important to have faithful data from the target corpus, the faithfulness of additional corpora only plays a minor role. Consequently, uFACT datasets can be constructed with large quantities of unfaithful data. We show how uFACT can be leveraged to obtain state-of-the-art results on the WebNLG benchmark using METEOR as our performance metric. Furthermore, we investigate the sensitivity of the generation faithfulness to the training corpus structure using the PARENT metric, and provide a baseline for this metric on the WebNLG (Gardent et al., 2017) benchmark to facilitate comparisons with future work. | ['Bill Byrne', 'Alexandru Coca', 'Tisha Anders'] | null | null | null | null | findings-acl-2022-5 | ['data-to-text-generation'] | ['natural-language-processing'] | [ 2.38720715e-01 8.90414059e-01 -5.90811968e-02 -2.43964821e-01
-1.14846396e+00 -8.73814046e-01 1.23757482e+00 1.65985718e-01
-1.62470639e-01 8.98240030e-01 9.23352420e-01 -6.48792908e-02
3.09202313e-01 -9.46031094e-01 -8.95707190e-01 -2.33968765e-01
3.16620767e-01 8.90669823e-01 5.02012037e-02 -6.10731304e-01
-1.57585725e-01 -1.00549191e-01 -1.37053692e+00 5.28242290e-01
1.02968562e+00 7.58973837e-01 1.33715823e-01 5.51941097e-01
-1.78942084e-01 8.52767229e-01 -1.01172984e+00 -7.48083055e-01
3.24850678e-01 -6.01613343e-01 -1.21389794e+00 5.04224002e-02
4.13227230e-01 -1.21590771e-01 -1.89295605e-01 6.90669119e-01
5.29446363e-01 1.73973843e-01 8.95168245e-01 -1.20627558e+00
-7.47061908e-01 1.39656365e+00 1.08534493e-01 -6.93115517e-02
3.41536015e-01 -5.47140986e-02 1.21198177e+00 -8.86881411e-01
1.17005301e+00 1.41290128e+00 3.22435200e-01 8.69044602e-01
-1.35864675e+00 -5.11029303e-01 -7.16368705e-02 -2.40050897e-01
-1.00387931e+00 -8.82167697e-01 5.27354300e-01 -1.87428087e-01
1.10954213e+00 2.12782294e-01 3.71563494e-01 1.52316403e+00
-3.24118763e-01 9.44359958e-01 6.88573241e-01 -6.44611537e-01
1.34635806e-01 1.72031105e-01 -3.94024700e-02 4.51779664e-01
1.95673525e-01 1.17148168e-01 -5.98651290e-01 -1.36489347e-02
3.39425266e-01 -7.01114714e-01 -2.84327537e-01 -1.39855117e-01
-1.18327427e+00 1.12345529e+00 4.93132174e-01 4.62494135e-01
-1.34474710e-01 2.72091478e-01 5.65568089e-01 3.91741127e-01
6.60363376e-01 9.09832060e-01 -3.99549246e-01 -4.14437294e-01
-7.93425143e-01 6.24460459e-01 9.10824537e-01 1.58771586e+00
5.03653586e-01 1.89127997e-01 -3.23010474e-01 8.39088917e-01
1.68681536e-02 4.37584132e-01 7.84908056e-01 -1.05859482e+00
7.96267450e-01 3.07368010e-01 1.15865923e-01 -4.51518714e-01
-7.48402029e-02 -3.29481721e-01 -4.65820521e-01 -3.67508858e-01
5.72729945e-01 -3.15390199e-01 -7.01696038e-01 2.10308552e+00
1.86701491e-01 -2.97947645e-01 8.17823708e-01 4.78735536e-01
9.35707331e-01 7.59739578e-01 -6.67801574e-02 -1.32777737e-02
1.01332808e+00 -1.15018213e+00 -6.44006848e-01 -3.18234324e-01
1.10210717e+00 -7.83844769e-01 1.39406943e+00 1.56325568e-02
-1.15712786e+00 -4.56550032e-01 -9.14992154e-01 -3.52756321e-01
-4.92856622e-01 2.19751950e-02 4.24604446e-01 4.83045399e-01
-8.92651439e-01 7.30022609e-01 -6.84444904e-01 -5.99463284e-01
3.38383198e-01 -3.95560771e-01 -2.94284880e-01 -2.48657569e-01
-1.41381657e+00 9.60765004e-01 8.49372566e-01 -5.55259824e-01
-1.03458130e+00 -8.62091482e-01 -9.73781109e-01 -1.53211430e-01
3.90751779e-01 -5.90216935e-01 1.64755440e+00 -7.79496253e-01
-1.11794782e+00 6.70742691e-01 -5.33031709e-02 -7.27769375e-01
5.72477400e-01 -1.37022659e-01 -2.69175500e-01 1.38637200e-01
4.27131802e-01 9.77108240e-01 4.62674767e-01 -1.36839581e+00
-4.92923379e-01 6.30641580e-02 2.02827260e-01 1.64967969e-01
-3.29801023e-01 -1.47854388e-01 -2.21529037e-01 -9.40246046e-01
-3.26290429e-01 -8.38814855e-01 1.25851676e-01 -4.37673360e-01
-7.10524321e-01 -4.09322917e-01 4.97254580e-01 -5.97550690e-01
1.15013480e+00 -1.99389875e+00 1.93621218e-01 -1.42938122e-01
-9.55300871e-03 -6.17595986e-02 -5.35822570e-01 7.09858477e-01
1.81367084e-01 5.85614085e-01 -2.90003926e-01 -5.91235876e-01
4.15067494e-01 4.00826901e-01 -4.55657363e-01 -2.48947471e-01
3.13520342e-01 9.85947430e-01 -1.09763122e+00 -5.20202518e-01
-1.32796511e-01 1.77834451e-01 -7.17141747e-01 3.47699463e-01
-5.76789141e-01 1.77442148e-01 -3.08219820e-01 3.74197841e-01
1.73885122e-01 -1.16378576e-01 7.50183165e-02 -2.12651104e-01
1.77014500e-01 8.86158824e-01 -7.02318132e-01 1.89409614e+00
-8.11730862e-01 4.74589288e-01 -4.44130540e-01 -4.64858711e-01
8.39457512e-01 5.54611564e-01 6.00285754e-02 -6.48837984e-01
1.08812876e-01 4.60529178e-01 -7.08024055e-02 -1.32010102e-01
9.06282723e-01 -4.62371349e-01 -4.81309295e-01 8.43227684e-01
4.60261136e-01 -6.66477323e-01 7.85914660e-01 5.94990015e-01
1.10648787e+00 1.41919896e-01 1.80379733e-01 -2.65023023e-01
-1.26098290e-01 2.19732746e-01 2.88489699e-01 6.80657208e-01
1.77779853e-01 7.65601277e-01 4.80668902e-01 -1.16323428e-02
-1.31682992e+00 -1.00984144e+00 -1.63799822e-01 9.23045993e-01
-2.45709628e-01 -7.59663463e-01 -1.06722856e+00 -1.11922491e+00
-8.43072087e-02 1.57820559e+00 -7.93819785e-01 -3.07851702e-01
-4.20979828e-01 -5.35824060e-01 1.00905645e+00 5.87881505e-01
3.45989376e-01 -1.16056287e+00 -4.07518893e-01 3.86347502e-01
-7.31434107e-01 -1.24880528e+00 -5.19839764e-01 1.24270596e-01
-5.43511450e-01 -9.18951392e-01 -4.23702419e-01 -5.37765563e-01
4.73658562e-01 5.71572781e-02 1.56647027e+00 -4.81133386e-02
2.68514872e-01 1.62622243e-01 -8.93620074e-01 -5.35809875e-01
-1.30871475e+00 4.17368054e-01 -2.13767380e-01 -5.30281305e-01
9.26095247e-02 -4.64962065e-01 3.83278877e-02 1.07711263e-01
-1.22565627e+00 4.58737671e-01 1.03656046e-01 7.60230184e-01
3.62650037e-01 5.53753600e-03 8.30655873e-01 -1.22559965e+00
7.62893140e-01 -6.59564793e-01 -2.24273667e-01 2.18191117e-01
-4.12822187e-01 4.00638968e-01 8.55916619e-01 -3.16339403e-01
-1.18372273e+00 -4.89592612e-01 -2.02690914e-01 -3.48388970e-01
9.89860669e-02 6.17391348e-01 -3.86972666e-01 5.93765736e-01
1.00065625e+00 7.24629313e-02 -2.01025635e-01 -6.11082792e-01
8.97521257e-01 6.92391038e-01 6.12674832e-01 -9.88946676e-01
7.82840788e-01 9.25705060e-02 -4.64258105e-01 -4.66142476e-01
-1.24752903e+00 2.35661473e-02 -5.71610510e-01 8.84817317e-02
5.55471420e-01 -8.72578621e-01 3.04754853e-01 2.31309459e-02
-1.31078935e+00 -5.73978961e-01 -7.94992685e-01 5.54624610e-02
-6.65028930e-01 -4.89961077e-03 -6.25279903e-01 -5.12053192e-01
-5.32318175e-01 -8.81566584e-01 1.24944508e+00 -2.98961818e-01
-5.86122453e-01 -1.24663150e+00 2.09648013e-01 4.53595161e-01
2.87416011e-01 5.48897564e-01 1.14166462e+00 -1.09561074e+00
-2.55594075e-01 -5.59823476e-02 -4.65208218e-02 3.28285962e-01
3.43467534e-01 -8.61260518e-02 -1.06127942e+00 -2.36614928e-01
-6.71932066e-04 -8.67951453e-01 7.90047348e-01 -2.19686851e-01
7.40039408e-01 -8.76871526e-01 -8.45197439e-02 3.42735350e-01
1.25526369e+00 -5.41383922e-02 6.03510380e-01 2.57330507e-01
7.25349545e-01 8.43374193e-01 6.88136220e-01 3.22923213e-01
5.98885238e-01 7.10588992e-01 1.21619202e-01 7.12309498e-03
-5.77133179e-01 -8.86011302e-01 4.18326676e-01 1.17499816e+00
3.15457761e-01 -7.01318920e-01 -7.91173160e-01 8.11712325e-01
-1.70792246e+00 -1.00808012e+00 5.20564280e-02 2.12689066e+00
1.38414860e+00 2.38026470e-01 4.35595997e-02 1.36822343e-01
3.91986936e-01 9.08322856e-02 -2.16814578e-01 -2.87437350e-01
-4.25378680e-01 2.86415845e-01 1.04128167e-01 5.90380073e-01
-8.08723509e-01 1.22988343e+00 6.29673290e+00 1.07149804e+00
-7.92428434e-01 2.80151010e-01 5.46676099e-01 -2.25718543e-01
-8.90290141e-01 1.13442592e-01 -7.47064948e-01 6.07234478e-01
1.23361528e+00 -6.43480539e-01 4.06798750e-01 7.03812540e-01
2.01348439e-02 1.97855264e-01 -1.44569087e+00 3.98173541e-01
1.56388134e-01 -1.38541853e+00 3.79049510e-01 7.53341466e-02
8.63744378e-01 1.02653645e-01 -3.24251413e-01 6.32925034e-01
7.11180747e-01 -8.79940271e-01 1.27213669e+00 1.09190337e-01
8.12774837e-01 -7.45515406e-01 5.54143429e-01 3.96737158e-01
-9.11113322e-01 2.43373409e-01 -4.54964131e-01 1.89098895e-01
4.18993235e-01 6.24474704e-01 -1.15730560e+00 7.73107886e-01
2.81542301e-01 6.17617786e-01 -6.70789003e-01 3.43433350e-01
-3.50883573e-01 6.68717206e-01 -4.04561043e-01 -9.99772456e-03
2.84219980e-01 1.96059287e-01 5.37645757e-01 1.12907696e+00
4.83053565e-01 -1.55950010e-01 9.93370637e-02 1.21504104e+00
-6.25871062e-01 1.42458245e-01 -9.36570108e-01 -6.13157272e-01
7.38063872e-01 1.07806468e+00 -3.46689463e-01 -6.35786772e-01
-2.30644673e-01 8.16424727e-01 6.21149123e-01 2.43648842e-01
-5.55177093e-01 -3.45842004e-01 4.19869334e-01 1.34703055e-01
1.46274313e-01 -2.39589941e-02 -1.52162090e-01 -1.29385340e+00
-3.94228250e-02 -1.22048795e+00 4.00733978e-01 -1.00816667e+00
-1.41512895e+00 8.63416016e-01 1.68639228e-01 -9.91880119e-01
-8.82978499e-01 -1.92707792e-01 -4.08115625e-01 8.26604247e-01
-1.37548375e+00 -1.37835276e+00 -1.12519398e-01 3.13417494e-01
7.73432672e-01 -1.85761392e-01 1.02539754e+00 -8.87137130e-02
-4.07979012e-01 6.45952165e-01 9.68662947e-02 1.82815552e-01
5.81584692e-01 -1.47765231e+00 9.02464092e-01 1.07959354e+00
4.27973688e-01 5.41970551e-01 8.18693280e-01 -8.05727720e-01
-1.03604662e+00 -1.47667849e+00 1.23931825e+00 -6.75300002e-01
7.60247529e-01 -6.30535603e-01 -6.71042860e-01 8.66815507e-01
4.13272709e-01 -2.88313538e-01 6.02322280e-01 1.55285031e-01
-5.76677203e-01 3.35457474e-01 -1.06717050e+00 7.01816797e-01
1.10014141e+00 -4.70969498e-01 -7.45157838e-01 4.57452893e-01
1.36001766e+00 -4.39841151e-01 -1.00931430e+00 2.00970545e-01
1.16996594e-01 -6.81423962e-01 4.82015818e-01 -6.37852728e-01
8.49312484e-01 -1.57072973e-02 -5.21960616e-01 -1.80721521e+00
8.11664909e-02 -6.71352625e-01 -8.94752815e-02 1.72594023e+00
8.08828294e-01 -3.35363418e-01 4.64799494e-01 6.09614193e-01
-4.90860701e-01 -3.88582557e-01 -9.08071876e-01 -1.02216613e+00
6.39660716e-01 -5.20686507e-01 1.05966938e+00 9.10022557e-01
3.49412113e-01 8.51991415e-01 -1.18045807e-01 -4.77653682e-01
4.39426392e-01 8.75457749e-02 9.21261728e-01 -9.70559955e-01
-2.95231253e-01 -1.93837270e-01 1.93359002e-01 -7.48031318e-01
3.33281338e-01 -1.36144722e+00 1.22153968e-01 -1.53538465e+00
1.12126641e-01 -7.17314720e-01 2.28073925e-01 7.25353241e-01
-2.54279613e-01 1.34328067e-01 4.38889444e-01 2.60836035e-01
-3.18720281e-01 8.92265081e-01 1.28705025e+00 -3.88834104e-02
1.40593867e-04 -3.84269476e-01 -1.07627285e+00 2.44795918e-01
8.01258981e-01 -4.99558955e-01 -8.10647428e-01 -7.82286346e-01
-2.67834626e-02 -7.61479661e-02 1.33264944e-01 -9.73502338e-01
-3.22881699e-01 -5.83918057e-02 3.46026160e-02 -9.42767709e-02
3.13136995e-01 -3.24536413e-01 8.40130076e-02 8.83649960e-02
-7.37719119e-01 -3.21029872e-02 3.17851335e-01 1.67197824e-01
-2.28285983e-01 -5.16366184e-01 6.56401157e-01 -1.97500959e-01
-2.09774062e-01 8.58411491e-02 -1.28335118e-01 8.10913444e-01
4.27861273e-01 9.56651047e-02 -7.43502259e-01 -5.27722418e-01
-2.99916208e-01 4.41691019e-02 7.39786983e-01 5.78731358e-01
2.33813629e-01 -1.70579255e+00 -1.01398182e+00 -2.77985409e-02
5.51383913e-01 1.16645560e-01 -2.11536989e-01 1.75682142e-01
-2.18442783e-01 4.67350334e-01 4.52761026e-03 -8.81121904e-02
-7.31779754e-01 4.58967954e-01 2.74322748e-01 -4.26822364e-01
-4.64956492e-01 6.21552348e-01 2.26557940e-01 -8.03849638e-01
-9.99342650e-02 -2.89269775e-01 7.00124428e-02 4.79880907e-02
4.04436082e-01 2.02837691e-01 1.98821768e-01 -6.97652817e-01
1.49996281e-01 -2.96495199e-01 -1.50529236e-01 -6.56927764e-01
1.23804009e+00 9.13683232e-03 3.21738541e-01 3.65270048e-01
1.19929624e+00 1.27621606e-01 -1.03512597e+00 -2.98093885e-01
-8.20727572e-02 -2.97059000e-01 -1.00164577e-01 -1.07239068e+00
-8.37858617e-01 6.96601570e-01 -1.43874049e-01 4.53934133e-01
7.92505026e-01 1.77132323e-01 9.71061587e-01 3.45904410e-01
4.70627993e-01 -1.14016557e+00 2.34253734e-01 7.07370996e-01
1.34159970e+00 -9.80120838e-01 -4.34130907e-01 -4.28725988e-01
-8.59058022e-01 8.14213395e-01 5.63777089e-01 1.80619597e-01
2.49084443e-01 3.75951469e-01 9.21727866e-02 -6.58250377e-02
-1.15366697e+00 -3.43775868e-01 7.56749213e-02 8.80893648e-01
6.60293758e-01 7.18033314e-02 -1.12221129e-01 7.60327756e-01
-1.03142953e+00 -2.24316224e-01 7.20981121e-01 8.40315342e-01
-3.02839309e-01 -1.38396096e+00 9.77988727e-03 4.31611896e-01
-2.15417370e-01 -4.57821786e-01 -7.28287935e-01 1.01474988e+00
-6.81854188e-02 1.19886363e+00 4.04986404e-02 -2.64036864e-01
3.39794576e-01 2.47158691e-01 4.94025767e-01 -9.40929234e-01
-5.12468874e-01 8.12641084e-02 8.84213388e-01 -3.32200259e-01
-2.48517752e-01 -6.42931104e-01 -1.35532260e+00 -3.94076943e-01
-2.07152992e-01 5.01908600e-01 4.95835394e-01 1.04959428e+00
3.27124417e-01 4.23789144e-01 4.35227722e-01 -6.08793378e-01
-4.48608547e-01 -1.46299171e+00 -4.26977038e-01 7.31453300e-01
-1.06435403e-01 -5.30180752e-01 -3.17111671e-01 2.03504682e-01] | [11.444321632385254, 8.87647819519043] |
2d5c8d2e-38c0-475d-b49c-ecb4cdd59fc1 | nature-natural-auxiliary-text-utterances-for | 2111.05196 | null | https://arxiv.org/abs/2111.05196v2 | https://arxiv.org/pdf/2111.05196v2.pdf | NATURE: Natural Auxiliary Text Utterances for Realistic Spoken Language Evaluation | Slot-filling and intent detection are the backbone of conversational agents such as voice assistants, and are active areas of research. Even though state-of-the-art techniques on publicly available benchmarks show impressive performance, their ability to generalize to realistic scenarios is yet to be demonstrated. In this work, we present NATURE, a set of simple spoken-language oriented transformations, applied to the evaluation set of datasets, to introduce human spoken language variations while preserving the semantics of an utterance. We apply NATURE to common slot-filling and intent detection benchmarks and demonstrate that simple perturbations from the standard evaluation set by NATURE can deteriorate model performance significantly. Through our experiments we demonstrate that when NATURE operators are applied to evaluation set of popular benchmarks the model accuracy can drop by up to 40%. | ['Mehdi Rezagholizadeh', 'Philippe Langlais', 'Abbas Ghaddar', 'Ahmad Rashid', 'David Alfonso-Hermelo'] | 2021-11-09 | null | null | null | null | ['slot-filling'] | ['natural-language-processing'] | [ 4.51487303e-01 4.05901432e-01 -1.41265839e-01 -6.96662366e-01
-8.75933051e-01 -7.23835766e-01 1.02065957e+00 -3.74151045e-03
-4.77003485e-01 7.85764754e-01 6.50802553e-01 -3.88038337e-01
2.40348160e-01 -3.05207819e-01 -2.22927913e-01 -2.60436505e-01
-3.21061075e-01 9.00705695e-01 4.51718479e-01 -6.74293339e-01
1.18479006e-01 1.38441026e-01 -1.60275769e+00 5.19830763e-01
5.17736912e-01 5.34421325e-01 -3.63907546e-01 8.65322709e-01
-3.20005864e-01 9.32502627e-01 -9.46414173e-01 -5.10031283e-02
1.50699615e-01 -3.93648148e-01 -1.12306809e+00 3.66222978e-01
2.78024048e-01 -3.70828629e-01 -2.34042227e-01 6.04933321e-01
6.21056259e-01 5.66485941e-01 1.43570900e-01 -1.57277453e+00
-9.59025770e-02 8.49980235e-01 1.12061515e-01 1.26760885e-01
9.30474520e-01 2.34670192e-01 1.14062476e+00 -7.60001898e-01
7.92382658e-01 1.53307080e+00 5.71407855e-01 7.76991904e-01
-1.41376626e+00 -1.40379637e-01 1.61070824e-01 -1.23594075e-01
-1.07060742e+00 -8.80847156e-01 5.05485296e-01 -4.43370566e-02
1.60562944e+00 7.98980892e-01 1.94684297e-01 1.13829386e+00
-3.17555219e-01 9.93864238e-01 1.12336004e+00 -3.58636320e-01
3.78020912e-01 2.47019634e-01 4.87266004e-01 6.09921336e-01
-4.80796099e-01 2.23489925e-02 -7.70215273e-01 -4.95144635e-01
2.05407720e-02 -5.41621387e-01 -2.26209939e-01 -2.90767610e-01
-1.19035995e+00 8.43492746e-01 -6.74527958e-02 7.28798434e-02
-1.59709007e-01 -1.14330508e-01 7.28187919e-01 4.86106634e-01
4.30825055e-01 5.04071414e-01 -5.11774838e-01 -8.89022171e-01
-6.82015598e-01 6.72906399e-01 1.65672958e+00 8.97001028e-01
5.55295706e-01 -2.09726393e-01 -5.71075737e-01 1.07852197e+00
5.08702127e-03 1.20535761e-01 6.09061956e-01 -1.20874250e+00
3.02624851e-01 6.61328614e-01 1.51855230e-01 -5.33490539e-01
-6.76276743e-01 -2.98193574e-01 -3.09165359e-01 -1.67889103e-01
3.79987776e-01 -2.72764474e-01 -8.52379501e-01 1.73288608e+00
3.49459499e-01 2.58203316e-02 2.94079393e-01 6.80697381e-01
9.00694430e-01 5.94816685e-01 7.85773247e-02 -2.70087183e-01
1.36589205e+00 -1.35437822e+00 -7.63616025e-01 -7.60083914e-01
8.39456618e-01 -8.20419133e-01 1.49976683e+00 1.94692582e-01
-1.08861649e+00 -2.87233070e-02 -7.78628528e-01 -1.63384303e-01
-2.40063280e-01 -4.85778123e-01 8.58311832e-01 7.23306715e-01
-1.08454847e+00 1.97222814e-01 -7.45522559e-01 -8.31006885e-01
-1.91056371e-01 2.83806175e-01 -3.82949054e-01 6.24246337e-02
-1.23426890e+00 9.83876050e-01 1.22969963e-01 -2.15222299e-01
-8.33249986e-01 -6.20860279e-01 -9.30867016e-01 -6.64034560e-02
7.70718932e-01 -5.07693470e-01 2.25447226e+00 -4.93888140e-01
-1.79277670e+00 9.18992937e-01 -4.55524266e-01 -7.22743988e-01
5.88558733e-01 -2.78111696e-01 -3.33979726e-01 -1.87670439e-01
1.18381903e-01 7.23682940e-01 4.02151823e-01 -1.07029045e+00
-8.12843740e-01 6.34688064e-02 3.83979261e-01 2.89090514e-01
-1.81269929e-01 3.63959789e-01 -4.53627288e-01 -4.26642448e-01
5.04081184e-03 -1.16995430e+00 -2.21456155e-01 -2.56601185e-01
-5.32355249e-01 -2.94317365e-01 9.13121998e-01 -5.16360998e-01
1.19133961e+00 -2.17589808e+00 8.67633894e-03 -2.88571455e-02
-6.86041266e-02 1.02860928e-01 -4.35572058e-01 8.16778004e-01
1.44701898e-01 1.21492386e-01 -4.19552445e-01 -8.12905014e-01
4.73472267e-01 6.37486756e-01 -4.72415209e-01 2.05225214e-01
4.54098471e-02 7.60805964e-01 -7.91236162e-01 -1.47478223e-01
2.70681888e-01 6.72687516e-02 -9.59494233e-01 5.12236953e-01
-6.46625340e-01 8.12007114e-02 -1.19327389e-01 3.80304426e-01
1.30039334e-01 -7.03743622e-02 2.60282248e-01 2.41659418e-01
5.12391236e-03 1.18095863e+00 -1.05173314e+00 1.79953396e+00
-5.11720598e-01 6.02032900e-01 2.64622241e-01 -7.38772810e-01
6.68594539e-01 3.69965583e-01 8.20522159e-02 -6.78057790e-01
-1.09776795e-01 4.36871834e-02 2.13803217e-01 -3.01864058e-01
7.16188848e-01 -1.61420807e-01 -3.78312528e-01 8.04796576e-01
-1.43709943e-01 -3.02316546e-01 4.17118788e-01 1.84439063e-01
1.35842037e+00 -5.07474601e-01 3.55929136e-01 -3.33489031e-02
4.65964049e-01 2.44820267e-01 3.26018691e-01 1.02608013e+00
-3.77997369e-01 5.14432490e-01 6.46583855e-01 -5.96124351e-01
-8.92092288e-01 -8.53472650e-01 -6.41998947e-02 1.84327674e+00
-2.16722861e-01 -5.94587207e-01 -8.62348437e-01 -5.96329033e-01
-5.17045613e-04 1.20123935e+00 -4.26184684e-01 -7.26538226e-02
-4.98232365e-01 -7.16110468e-01 8.98386002e-01 2.25989029e-01
6.23212397e-01 -1.28898680e+00 -5.92995346e-01 4.12949085e-01
-1.67391509e-01 -1.57554018e+00 -4.10218924e-01 2.17693105e-01
-4.23364729e-01 -7.99703240e-01 -3.54819030e-01 -6.06300056e-01
6.80332035e-02 1.60703883e-01 1.36496425e+00 1.00749560e-01
-2.13172942e-01 5.43427408e-01 -3.76655608e-01 -2.45624006e-01
-9.23435271e-01 4.43072468e-01 3.70194614e-02 -2.29891658e-01
4.03165609e-01 -5.98737061e-01 -1.78019121e-01 4.77858275e-01
-8.50241721e-01 7.94502646e-02 9.20785517e-02 8.51341665e-01
-2.03359112e-01 -3.21917623e-01 3.79764944e-01 -1.08391166e+00
1.15646124e+00 -1.85364783e-01 -1.76097900e-01 -4.21619462e-03
-4.20764983e-01 1.69154435e-01 1.85424879e-01 -2.49709889e-01
-9.22107100e-01 -1.50224045e-01 -2.97592312e-01 1.50893494e-01
-4.69514310e-01 4.41988081e-01 6.01380281e-02 1.39857754e-01
6.55950129e-01 3.57577741e-01 1.77425623e-01 -4.84306484e-01
5.49043834e-01 8.63138556e-01 6.43413007e-01 -5.91383696e-01
3.74488384e-01 3.34063798e-01 -5.22648394e-01 -1.11154854e+00
-7.05273628e-01 -6.96637034e-01 -5.80741502e-02 2.65398324e-01
4.82903361e-01 -7.97509432e-01 -7.39026308e-01 2.64544666e-01
-1.17024994e+00 -7.96767175e-01 -3.11977446e-01 1.17762022e-01
-6.97539032e-01 4.35923338e-01 -6.58904552e-01 -9.69082952e-01
-4.58731532e-01 -1.21236706e+00 1.24059415e+00 2.77781729e-02
-8.91666412e-01 -8.64039123e-01 2.04023227e-01 5.66354573e-01
5.70020318e-01 -3.93981338e-01 7.42616951e-01 -1.41016400e+00
-1.90956160e-01 -8.81620198e-02 2.04985961e-01 2.40551993e-01
2.14309573e-01 -9.71890241e-02 -1.20933247e+00 -2.61592120e-01
-2.14064330e-01 -4.43546653e-01 6.61571622e-01 -2.16799989e-01
5.93636036e-01 -6.10081971e-01 -2.48307243e-01 1.49003848e-01
4.41978693e-01 2.66918331e-01 5.96245229e-01 5.14367640e-01
6.50602058e-02 6.14139020e-01 5.23875475e-01 6.23977005e-01
5.34564614e-01 1.02504945e+00 1.70759216e-01 6.21031374e-02
4.02668826e-02 -1.60386607e-01 4.43853825e-01 5.84007680e-01
4.13840175e-01 -4.19687986e-01 -1.04895294e+00 6.30087912e-01
-1.87341928e+00 -9.12737489e-01 3.44529748e-01 1.86750245e+00
9.12241578e-01 4.48369712e-01 4.98953640e-01 1.83792844e-01
4.06439334e-01 4.30907905e-01 -2.90306300e-01 -7.43190050e-01
1.97778363e-02 1.05188444e-01 2.07461342e-02 9.04258788e-01
-1.12687528e+00 1.28960943e+00 7.23725080e+00 3.81653219e-01
-1.03766787e+00 5.49880974e-02 3.67846340e-01 -1.40277147e-01
-3.86970460e-01 -1.97284799e-02 -6.82648003e-01 9.24351886e-02
9.31012750e-01 -3.28391731e-01 6.82269454e-01 1.01652515e+00
2.74360150e-01 -1.92948841e-02 -1.34247220e+00 8.35082829e-01
1.72032356e-01 -1.07154930e+00 -1.62750348e-01 -1.66885078e-01
4.73277867e-01 1.76035672e-01 -1.30129948e-01 7.61233985e-01
5.69821179e-01 -1.16150331e+00 4.55111355e-01 9.58285481e-02
2.30080158e-01 -4.68153387e-01 6.59302950e-01 5.22518575e-01
-6.39721394e-01 2.14435291e-02 -7.74514750e-02 -4.53802615e-01
4.83638197e-01 2.30914220e-01 -1.49390244e+00 -6.54582530e-02
3.76252353e-01 1.77830413e-01 -3.82301480e-01 8.03407192e-01
-4.29857010e-03 7.74111688e-01 -6.89808607e-01 -2.96444207e-01
4.99363065e-01 2.08879754e-01 8.27513874e-01 1.40777743e+00
-9.43605080e-02 1.08939357e-01 4.20518994e-01 6.92192197e-01
-6.34211749e-02 -3.71495523e-02 -5.45213401e-01 -2.64614016e-01
5.59495687e-01 1.03409648e+00 -3.63976777e-01 -3.96198779e-01
-4.51456845e-01 1.05320942e+00 2.20625758e-01 3.32690150e-01
-5.83552957e-01 -1.50811136e-01 1.34229505e+00 -5.08650653e-02
7.05600753e-02 -2.65375912e-01 -5.69539331e-02 -9.38846588e-01
9.64774787e-02 -1.53271508e+00 3.07228178e-01 -4.61381018e-01
-9.99061704e-01 6.23079836e-01 4.11737226e-02 -5.40420830e-01
-8.90895009e-01 -3.36771250e-01 -7.54987597e-01 6.28529549e-01
-1.17580736e+00 -7.03783631e-01 -3.76650512e-01 2.93340683e-01
1.01990306e+00 -3.29567969e-01 1.29743040e+00 1.13964520e-01
-5.34710169e-01 7.24252105e-01 -2.62773752e-01 1.11293718e-01
5.86440265e-01 -1.15030861e+00 1.17035592e+00 4.90992457e-01
3.77226412e-01 7.39317894e-01 1.38904536e+00 -2.79790163e-01
-1.33709168e+00 -7.52958179e-01 1.07470047e+00 -4.42722529e-01
7.42794335e-01 -5.40624738e-01 -1.02340293e+00 9.72558856e-01
4.13750827e-01 -1.77416489e-01 5.75539172e-01 4.11328167e-01
-4.13259417e-01 2.20714107e-01 -9.63885486e-01 8.88311863e-01
1.19769442e+00 -8.21345448e-01 -1.01095831e+00 4.97751266e-01
9.83681619e-01 -7.46984601e-01 -5.46644688e-01 4.21806425e-01
5.87219656e-01 -1.19437778e+00 9.36227679e-01 -9.93324041e-01
4.62441184e-02 -4.42722999e-02 -3.23340118e-01 -1.40196657e+00
2.11325601e-01 -1.09968650e+00 4.59204316e-02 1.29047918e+00
5.96296728e-01 -7.45782316e-01 8.70703638e-01 9.03159976e-01
-2.44100779e-01 -3.53217483e-01 -1.08695030e+00 -6.07264221e-01
-1.83817610e-01 -6.05893314e-01 5.88625014e-01 6.40146971e-01
4.76867825e-01 6.88687325e-01 -3.12929928e-01 3.59733589e-02
3.17713737e-01 -5.46858422e-02 1.26969147e+00 -8.75774205e-01
-3.67711008e-01 -4.88435864e-01 -3.15256000e-01 -1.23994625e+00
3.53827059e-01 -5.24112582e-01 2.67996430e-01 -1.46425891e+00
-9.54802483e-02 -3.25834870e-01 1.91022947e-01 7.16384232e-01
-7.73352431e-03 1.05746835e-01 3.60249043e-01 -5.66338822e-02
-7.94777751e-01 6.68466628e-01 6.94062233e-01 -2.25297451e-01
-4.10738647e-01 1.21862087e-02 -5.19881725e-01 6.63990080e-01
9.00753260e-01 -2.21240833e-01 -5.43716013e-01 -1.60299703e-01
-1.86910536e-02 6.73304871e-02 3.27232659e-01 -8.64246905e-01
2.65659630e-01 -2.04561710e-01 -6.02118790e-01 -2.45732158e-01
6.17015958e-01 -3.86294186e-01 -2.33405381e-01 2.93149769e-01
-6.84963346e-01 -1.63809992e-02 5.96666276e-01 1.40919834e-01
-3.82825792e-01 1.39107462e-02 3.69848609e-01 -2.68289242e-02
-9.48814273e-01 -3.16328555e-01 -6.33978546e-01 3.89678657e-01
7.69026279e-01 2.27562189e-01 -4.53600883e-01 -9.23301935e-01
-6.92756653e-01 2.65341967e-01 3.28147888e-01 7.41420984e-01
1.93062752e-01 -9.14954126e-01 -6.94945097e-01 2.55771220e-01
3.06884915e-01 -3.64098221e-01 -7.06600770e-02 5.93311191e-01
-4.47246909e-01 7.03901947e-01 1.10771656e-01 -6.48061633e-01
-1.56411195e+00 2.75311023e-01 5.23881376e-01 -1.34549201e-01
-5.43784440e-01 7.54470885e-01 5.02815880e-02 -1.03035522e+00
7.55748928e-01 -6.83198273e-01 1.78859919e-01 -1.47975311e-01
6.21289492e-01 2.28246987e-01 3.41748923e-01 -3.93437535e-01
-6.84671283e-01 -1.69963568e-01 -1.89463794e-01 -5.38117528e-01
1.10337913e+00 -2.08439380e-01 1.83492154e-02 5.49248636e-01
9.60383058e-01 -7.01375678e-02 -7.05751479e-01 -3.47571433e-01
3.00862849e-01 -1.61543190e-01 -3.40666503e-01 -8.72119308e-01
-3.42753649e-01 6.58431113e-01 4.67154324e-01 5.78783572e-01
7.71474421e-01 4.49860618e-02 8.24673891e-01 9.48385239e-01
4.25822467e-01 -9.05781567e-01 -7.08474591e-02 9.80754077e-01
1.07594371e+00 -1.28856635e+00 -4.02177423e-01 -5.10661542e-01
-7.80694485e-01 6.80144608e-01 5.02302647e-01 2.30487674e-01
3.13242286e-01 5.97361743e-01 4.63763088e-01 -4.31804061e-02
-1.24277532e+00 -2.50673383e-01 -9.48647633e-02 4.51095790e-01
5.99969387e-01 1.30342513e-01 -2.43012249e-01 3.64830792e-01
-5.57190120e-01 -6.34908751e-02 5.31046689e-01 1.04587936e+00
-5.40027559e-01 -1.26517904e+00 -2.76476830e-01 2.12928697e-01
-2.10925549e-01 -2.33808875e-01 -6.12964094e-01 8.87717068e-01
-5.07560849e-01 1.10180461e+00 2.62845695e-01 -2.70027488e-01
6.69189930e-01 6.13506436e-01 2.66906098e-02 -9.16135371e-01
-7.14271665e-01 -2.35228211e-01 1.01609206e+00 -7.61047721e-01
-2.91984707e-01 -5.74211061e-01 -1.42992032e+00 -4.21896487e-01
1.79274008e-01 3.37231815e-01 4.05090660e-01 1.11252832e+00
3.55693072e-01 3.39221507e-01 2.60831207e-01 -6.11920416e-01
-9.06795800e-01 -1.12164223e+00 -4.30481166e-01 5.85730016e-01
4.05199349e-01 -6.07495904e-01 -6.36964262e-01 -1.24764457e-01] | [12.747001647949219, 7.844950199127197] |
3fcdd040-5448-4e68-bdf7-e402f7ffc187 | aspect-extraction-and-sentiment | 1712.03430 | null | http://arxiv.org/abs/1712.03430v1 | http://arxiv.org/pdf/1712.03430v1.pdf | Aspect Extraction and Sentiment Classification of Mobile Apps using App-Store Reviews | Understanding of customer sentiment can be useful for product development. On
top of that if the priorities for the development order can be known, then
development procedure become simpler. This work has tried to address this issue
in the mobile app domain. Along with aspect and opinion extraction this work
has also categorized the extracted aspects ac-cording to their importance. This
can help developers to focus their time and energy at the right place. | ['Sharmistha Dey'] | 2017-12-09 | null | null | null | null | ['aspect-extraction'] | ['natural-language-processing'] | [-4.12481986e-02 4.67158228e-01 -6.19938314e-01 -5.06568253e-01
2.84828972e-02 -4.21378016e-01 2.55948067e-01 6.17982686e-01
-1.14302531e-01 5.54239690e-01 2.43512496e-01 -3.78929853e-01
-6.85025454e-02 -9.50692654e-01 -3.31490003e-02 -1.85927778e-01
4.24637914e-01 2.22168580e-01 8.64956230e-02 -5.02701819e-01
7.91074872e-01 1.91232339e-01 -1.64895380e+00 5.14784694e-01
4.53872323e-01 5.95265567e-01 3.27488482e-01 4.72637862e-01
-5.52000940e-01 8.43959868e-01 -7.16982603e-01 -7.01101065e-01
-7.23906010e-02 -4.43314672e-01 -6.98407590e-01 1.48310632e-01
-2.45114490e-01 1.28135681e-02 8.40882778e-01 9.43532169e-01
1.84036344e-01 -3.43306720e-01 6.14015102e-01 -1.01103973e+00
-5.19434325e-02 5.32665610e-01 -3.90147924e-01 1.99332833e-02
6.34689033e-01 -6.01158977e-01 1.08889389e+00 -8.44924748e-01
5.86500347e-01 7.24803925e-01 3.39581132e-01 1.04268298e-01
-6.31327927e-01 -5.11039555e-01 2.44504124e-01 2.79259682e-01
-9.41272318e-01 -1.09491371e-01 8.59490991e-01 -6.07703149e-01
1.16355288e+00 1.74165323e-01 9.51965153e-01 5.02573013e-01
2.80985475e-01 6.50788009e-01 1.01878631e+00 -8.77292871e-01
3.63415003e-01 7.74213016e-01 5.33820152e-01 6.48317635e-02
6.79984570e-01 -7.33628690e-01 -3.34145099e-01 1.25345618e-01
2.42770582e-01 -7.09534138e-02 -9.04421732e-02 6.73582330e-02
-4.83805031e-01 8.03458333e-01 -3.62793356e-01 8.87928128e-01
-7.01352775e-01 -2.44686782e-01 3.99607450e-01 2.42801189e-01
5.12427807e-01 8.07568789e-01 -8.16360831e-01 -7.86052227e-01
-8.83042991e-01 8.85131489e-03 1.25942349e+00 7.95971394e-01
6.54278219e-01 -1.61948726e-01 4.47682917e-01 5.52295744e-01
5.62413394e-01 1.79405466e-01 1.55586004e-01 -7.27595150e-01
2.56747097e-01 1.20064270e+00 1.03599280e-01 -9.56605732e-01
-4.93754707e-02 -5.18748760e-01 -1.00301497e-01 2.51041323e-01
-4.23570536e-03 -2.50451654e-01 -6.43595815e-01 8.42586756e-01
1.50418878e-01 -4.52282548e-01 2.49979138e-01 4.17923838e-01
7.91342676e-01 5.22624195e-01 -6.55997694e-02 -2.56939977e-01
1.76408803e+00 -5.58031976e-01 -1.13024998e+00 -1.84112161e-01
4.59424108e-01 -1.27187967e+00 6.20523453e-01 8.92836452e-01
-6.53547764e-01 -2.64823675e-01 -1.13642287e+00 3.41558039e-01
-6.12070143e-01 2.18615860e-01 6.87905610e-01 9.98745024e-01
-7.55050838e-01 4.35978174e-01 -6.92772388e-01 -5.48378646e-01
2.50443190e-01 4.11182046e-01 4.37928550e-02 2.23345995e-01
-1.01578796e+00 1.02730596e+00 2.79483438e-01 8.05226807e-03
4.90519032e-02 -5.15012205e-01 -5.23257673e-01 -1.76392972e-01
3.05382162e-01 -4.08608198e-01 1.30361760e+00 -1.23596466e+00
-1.50500286e+00 3.99108380e-01 -1.02867439e-01 -4.85634878e-02
-6.92910329e-02 -4.29236919e-01 -6.21721089e-01 3.06411292e-02
2.48183802e-01 1.05801582e-01 6.67615116e-01 -9.91587400e-01
-1.08401644e+00 -3.93224567e-01 4.44034815e-01 8.25866014e-02
-6.21980369e-01 4.76457119e-01 -4.67666894e-01 -4.28975284e-01
-1.58945203e-01 -6.77860141e-01 -2.07411930e-01 -6.42851591e-01
-8.38296562e-02 -1.97120368e-01 8.28393281e-01 -8.36535275e-01
1.54446352e+00 -1.72679710e+00 1.27023589e-02 3.83533239e-01
5.32014184e-02 3.07967186e-01 3.70240539e-01 7.82420456e-01
-1.50034383e-01 3.73614043e-01 4.55102801e-01 2.96427637e-01
-1.05386466e-01 9.10848826e-02 1.79397892e-02 -1.93084627e-01
2.83258855e-01 3.91566783e-01 -6.00897193e-01 -4.61887300e-01
-5.12068458e-02 7.32036889e-01 -4.21236455e-01 -4.99742851e-02
-9.08982307e-02 -2.21975334e-02 -8.53782475e-01 6.22756481e-01
3.52668941e-01 -3.34011391e-02 4.18694586e-01 -1.97165519e-01
-3.63206118e-01 5.42860746e-01 -1.02957714e+00 8.19526553e-01
-9.04574037e-01 6.86556220e-01 -1.68907747e-01 -8.87984872e-01
9.82413888e-01 4.79435205e-01 6.13400280e-01 -6.75475359e-01
3.49959970e-01 2.78503269e-01 2.08553940e-01 -5.83940387e-01
6.23936653e-01 -4.96053696e-02 1.40826613e-01 3.18224490e-01
-3.95538509e-01 -1.83727324e-01 4.83273268e-01 7.63962716e-02
4.89678949e-01 4.37135905e-01 7.54457474e-01 -1.59234986e-01
6.34193301e-01 1.18037328e-01 3.31368804e-01 -2.92852640e-01
2.45586753e-01 3.04664105e-01 9.68528211e-01 -2.60119528e-01
-7.68717527e-01 -6.07080348e-02 4.71121669e-02 7.12693274e-01
-4.55498546e-01 -1.05222201e+00 -7.77713835e-01 -6.83578372e-01
-5.78646243e-01 7.37442017e-01 -5.66411376e-01 3.46115798e-01
-2.88324177e-01 -3.20150942e-01 -2.69730061e-01 3.73986334e-01
2.10156739e-01 -8.48140955e-01 -1.13554621e+00 3.92408341e-01
1.08695298e-01 -8.45653296e-01 1.48928776e-01 3.47970039e-01
-8.00114810e-01 -1.19543588e+00 -6.55826509e-01 -6.20283604e-01
7.32621014e-01 1.28199682e-01 1.18106937e+00 8.70757699e-02
9.55476090e-02 4.67735052e-01 -1.04472220e+00 -1.16400111e+00
-3.32515717e-01 3.99128914e-01 -5.10607421e-01 -3.62365693e-01
7.71010876e-01 -5.66260397e-01 -4.50985104e-01 3.46627802e-01
-6.79045320e-01 -6.70043603e-02 4.82968420e-01 1.65758297e-01
4.38405752e-01 4.99881208e-01 5.79200923e-01 -1.29904950e+00
1.01814759e+00 -4.56469804e-01 -3.73676240e-01 1.63710460e-01
-1.04371107e+00 -9.85933542e-02 3.40033621e-01 4.19790782e-02
-9.29037631e-01 -1.78220600e-01 -3.48286837e-01 4.62047249e-01
-2.29695931e-01 8.03574145e-01 -3.09395909e-01 1.09927982e-01
2.18875617e-01 -3.34269226e-01 -9.52537879e-02 -2.89473265e-01
-3.95689867e-02 8.42335343e-01 -4.58134204e-01 -2.49989212e-01
2.48231173e-01 1.05228782e-01 -4.37534004e-02 -9.39396739e-01
-4.53305244e-01 -6.44385457e-01 -5.73864281e-01 -3.44175190e-01
7.95864761e-01 -4.71926570e-01 -3.93304884e-01 1.26936749e-01
-1.10624266e+00 2.54998237e-01 -2.59467274e-01 4.91629124e-01
-1.32269841e-02 3.28645051e-01 3.74943390e-02 -1.10630274e+00
-3.43502373e-01 -1.04264224e+00 4.00565773e-01 4.35307622e-01
-9.53687608e-01 -9.96219933e-01 -7.81693216e-03 3.38416129e-01
5.79626977e-01 1.51751256e-02 8.08604181e-01 -6.51015282e-01
-1.79457411e-01 -4.94329393e-01 8.05537701e-02 5.30851722e-01
5.31142890e-01 4.34415489e-01 -6.72415853e-01 2.26031050e-01
3.10669005e-01 4.11569208e-01 2.23840505e-01 3.37719560e-01
3.22414726e-01 -2.29892150e-01 -3.54641467e-01 -1.34754762e-01
1.39363217e+00 6.23464286e-01 7.32388496e-01 8.69771063e-01
4.17661190e-01 1.10656559e+00 1.30873215e+00 4.25388098e-01
3.88192028e-01 7.85856724e-01 3.85216296e-01 4.07171585e-02
1.23863757e-01 2.36871883e-01 4.19057786e-01 1.10586488e+00
-4.95905042e-01 -9.32038277e-02 -9.12445724e-01 6.36729181e-01
-1.69419408e+00 -6.80732608e-01 -5.27926564e-01 1.64790213e+00
4.37519521e-01 4.67044413e-01 2.38402337e-01 5.60120881e-01
3.10960233e-01 -1.14566647e-01 8.23976099e-02 -1.02587330e+00
4.29411352e-01 4.20454055e-01 4.50747907e-01 1.60937592e-01
-6.13215446e-01 4.28337812e-01 5.94691324e+00 5.31386614e-01
-1.28259087e+00 1.26514271e-01 6.10501528e-01 1.32171080e-01
-6.18530393e-01 3.57903063e-01 -1.07917249e+00 4.22125876e-01
8.65863204e-01 -2.21729070e-01 -1.41685568e-02 1.12033212e+00
3.12439263e-01 -5.53921402e-01 -8.62939477e-01 6.62267804e-01
9.80218127e-02 -1.02138042e+00 -3.29472214e-01 2.40252972e-01
4.94545698e-01 -4.43283617e-01 -1.52414069e-01 2.45262571e-02
-2.36123025e-01 -8.25893104e-01 5.01559615e-01 3.86186212e-01
1.15885518e-01 -1.20180786e+00 1.41804123e+00 2.30079353e-01
-1.13327074e+00 1.87679380e-01 -6.19902089e-02 -5.65505981e-01
2.92429119e-01 6.74031198e-01 -1.28791606e+00 7.49518335e-01
7.33610272e-01 5.21219671e-01 -3.97763014e-01 7.95132458e-01
-5.25431693e-01 5.67127824e-01 -2.02197582e-01 -5.81779599e-01
-1.05712600e-01 -3.98356855e-01 1.48357421e-01 1.17178643e+00
5.86861610e-01 1.26960659e-02 -3.18390340e-01 3.51237833e-01
5.67010045e-01 7.06929386e-01 -5.02214074e-01 -3.82931083e-01
1.30353093e-01 1.40487278e+00 -1.30440140e+00 1.68772079e-02
-6.73437655e-01 7.14734495e-01 -3.90184730e-01 5.68935722e-02
-3.75444889e-01 -6.29567206e-01 5.67314982e-01 6.24500692e-01
7.39406705e-01 -1.10658631e-01 -3.48365515e-01 -6.38781190e-01
7.24231526e-02 -9.59103942e-01 4.20971550e-02 -4.31105584e-01
-6.35063291e-01 7.81531096e-01 -3.60504128e-02 -1.17736530e+00
-3.95287722e-01 -7.72065282e-01 -7.34171271e-01 7.82385349e-01
-1.52190173e+00 -1.20863903e+00 2.84172148e-01 -8.44971910e-02
5.32271266e-01 -1.95329964e-01 8.08443844e-01 3.65388721e-01
-3.04198712e-01 1.65961891e-01 -5.24530053e-01 -2.69437551e-01
4.16562706e-01 -1.05493510e+00 1.35944054e-01 7.77511299e-01
9.59786251e-02 1.00083768e+00 9.47665572e-01 -7.98960626e-01
-1.08367431e+00 -4.39739466e-01 1.48255754e+00 -5.11786163e-01
6.02762222e-01 -7.53125548e-02 -5.68276584e-01 1.41933784e-01
7.24457979e-01 -8.01825643e-01 1.10188818e+00 4.27381635e-01
1.43777832e-01 -2.12239861e-01 -9.01072681e-01 3.35880131e-01
2.01327473e-01 -3.56806040e-01 -5.20134926e-01 -4.31654323e-03
4.01301742e-01 -1.16669415e-02 -1.03839076e+00 5.35577983e-02
5.91531217e-01 -1.18523383e+00 4.42821085e-01 -3.76102954e-01
5.84656477e-01 -2.30801299e-01 -1.24608688e-02 -1.04115379e+00
1.76968262e-01 -3.60770553e-01 2.43665069e-01 1.67895901e+00
9.54205334e-01 -4.97326940e-01 9.29197252e-01 8.19122195e-01
1.22117661e-01 -9.38917637e-01 -2.96485901e-01 -2.99727529e-01
-4.38468307e-01 -8.72947037e-01 7.11602449e-01 6.69040740e-01
2.32527256e-01 4.62011397e-01 -1.18735440e-01 -2.18573332e-01
-5.43397889e-02 1.45928964e-01 5.96370101e-01 -1.36977625e+00
-3.60849619e-01 -3.79836321e-01 -5.55294096e-01 -3.80833715e-01
-2.76316285e-01 -2.53331631e-01 -3.81055325e-01 -1.88429356e+00
1.13068543e-01 -6.53866529e-02 -2.39245132e-01 4.20536757e-01
3.96124795e-02 1.70315191e-01 3.90637554e-02 -1.01052731e-01
-4.10208583e-01 -3.53556991e-01 9.79795218e-01 2.38153800e-01
-2.99471408e-01 6.20245814e-01 -1.37114644e+00 8.79268408e-01
1.14873195e+00 -5.49449503e-01 -5.83011806e-01 1.48158237e-01
1.31381321e+00 -4.25013006e-01 -2.68070906e-01 -6.49213612e-01
1.46502018e-01 -1.68584689e-01 -2.01248601e-01 -7.07468927e-01
-3.24798152e-02 -1.43435299e+00 3.49784493e-01 3.71604919e-01
1.38216496e-01 2.96625674e-01 6.49888515e-02 -2.27468298e-03
-4.40836519e-01 -9.31339145e-01 1.40047729e-01 -6.14679232e-02
-8.47390771e-01 -4.75193530e-01 -6.54702783e-01 -3.97776604e-01
1.18085706e+00 -7.79390335e-01 1.37935907e-01 -6.93075776e-01
-4.43276882e-01 1.87779479e-02 5.82359672e-01 4.37524825e-01
3.90919864e-01 -8.31808269e-01 -3.03891271e-01 -4.25810106e-02
-5.82916953e-04 -4.26388919e-01 -7.05755427e-02 8.38594615e-01
-6.36752188e-01 5.15124619e-01 -3.80505562e-01 -1.35623470e-01
-1.66337836e+00 3.15539688e-01 -1.43205166e-01 -3.85817230e-01
4.99408729e-02 5.88756740e-01 -4.42706615e-01 1.75676391e-01
-8.16889033e-02 -4.25257027e-01 -1.14594436e+00 7.86107481e-01
4.96588498e-01 5.01455784e-01 3.98241639e-01 -6.52513862e-01
-4.87530023e-01 9.43192005e-01 -1.54418185e-01 -2.03791782e-01
1.71606839e+00 -7.45645687e-02 -4.13971394e-01 4.21141058e-01
8.43627632e-01 8.18916023e-01 -5.29227495e-01 6.28210545e-01
4.47161138e-01 -4.61071908e-01 2.00527817e-01 -7.67258763e-01
-1.17811644e+00 6.94845498e-01 5.40139675e-01 7.06057370e-01
1.28940988e+00 -1.33988550e-02 3.65386307e-01 -2.43272074e-02
3.50800961e-01 -1.42791915e+00 -2.06088886e-01 4.11593795e-01
5.32199740e-01 -1.06319869e+00 4.38006133e-01 -8.65249455e-01
-8.39897037e-01 1.54317129e+00 4.00091827e-01 2.50545651e-01
9.81246054e-01 6.07579410e-01 1.43158987e-01 -5.49972713e-01
-4.45975274e-01 -1.97716698e-01 3.42723757e-01 8.72094274e-01
1.20826709e+00 -7.92065337e-02 -1.01318085e+00 1.06257606e+00
-6.49976032e-03 3.18890244e-01 5.10574043e-01 1.06558502e+00
-4.84249175e-01 -1.97148883e+00 -1.68440044e-01 4.82987970e-01
-1.12653768e+00 -1.22473411e-01 -7.32865274e-01 8.42390537e-01
6.06899917e-01 1.21101630e+00 -3.57932806e-01 -3.74592453e-01
6.29157007e-01 -2.06242934e-01 1.93669423e-01 -9.65142846e-01
-7.69312441e-01 2.12766126e-01 7.81794846e-01 -2.73739576e-01
-8.10310721e-01 -6.81136608e-01 -1.10639763e+00 8.97936672e-02
-5.35228312e-01 6.40046954e-01 1.14650559e+00 8.79176438e-01
3.02833408e-01 7.20708787e-01 3.32108319e-01 -3.06837231e-01
2.54049957e-01 -7.46182203e-01 -3.31930757e-01 -1.34892002e-01
-2.95641541e-01 -3.55602503e-01 2.03364938e-02 1.35054708e-01] | [11.107365608215332, 6.789151191711426] |
2f286656-6f71-4a0e-baf6-287b31703f31 | sygns-a-systematic-generalization-testbed | 2106.01077 | null | https://arxiv.org/abs/2106.01077v1 | https://arxiv.org/pdf/2106.01077v1.pdf | SyGNS: A Systematic Generalization Testbed Based on Natural Language Semantics | Recently, deep neural networks (DNNs) have achieved great success in semantically challenging NLP tasks, yet it remains unclear whether DNN models can capture compositional meanings, those aspects of meaning that have been long studied in formal semantics. To investigate this issue, we propose a Systematic Generalization testbed based on Natural language Semantics (SyGNS), whose challenge is to map natural language sentences to multiple forms of scoped meaning representations, designed to account for various semantic phenomena. Using SyGNS, we test whether neural networks can systematically parse sentences involving novel combinations of logical expressions such as quantifiers and negation. Experiments show that Transformer and GRU models can generalize to unseen combinations of quantifiers, negations, and modifiers that are similar to given training instances in form, but not to the others. We also find that the generalization performance to unseen combinations is better when the form of meaning representations is simpler. The data and code for SyGNS are publicly available at https://github.com/verypluming/SyGNS. | ['Kentaro Inui', 'Koji Mineshima', 'Hitomi Yanaka'] | 2021-06-02 | null | https://aclanthology.org/2021.findings-acl.10 | https://aclanthology.org/2021.findings-acl.10.pdf | findings-acl-2021-8 | ['systematic-generalization'] | ['reasoning'] | [ 3.80419075e-01 1.95644364e-01 -1.35214701e-01 -7.58124053e-01
-2.81075388e-01 -9.21287060e-01 6.20797276e-01 2.21192330e-01
-3.07865232e-01 8.09632182e-01 4.48533088e-01 -5.21340609e-01
-1.00656226e-01 -1.21511149e+00 -6.61884069e-01 -2.82510430e-01
1.65363908e-01 5.01490891e-01 1.49554238e-01 -7.76433766e-01
-5.85781001e-02 3.22356403e-01 -1.70747030e+00 7.32820332e-01
5.30979216e-01 9.13773060e-01 2.64321923e-01 2.53013581e-01
-5.26510954e-01 7.95542419e-01 -6.30384147e-01 -4.24896032e-01
8.60931948e-02 -2.64097571e-01 -1.20044994e+00 -5.96327662e-01
6.34825051e-01 -5.22987023e-02 -1.05352804e-01 1.31607890e+00
3.93828042e-02 1.69003636e-01 5.27542353e-01 -1.24434781e+00
-1.22614026e+00 1.00143087e+00 1.74258247e-01 3.89014870e-01
5.90655088e-01 1.63458258e-01 1.68119347e+00 -6.94963098e-01
6.06443882e-01 1.97992611e+00 5.72874546e-01 1.02357030e+00
-1.30487430e+00 -5.70181310e-01 1.32637084e-01 2.38390401e-01
-1.13819265e+00 -2.64152199e-01 5.14835715e-01 1.28103673e-01
1.37244725e+00 2.21066996e-01 4.16245192e-01 1.35610187e+00
1.88974991e-01 7.88389981e-01 9.88307238e-01 -4.82557565e-01
2.15206221e-01 -2.55512893e-01 5.10151744e-01 4.59659666e-01
5.41764438e-01 -8.39812309e-02 -2.31556118e-01 -1.01426557e-01
3.97379875e-01 -3.49258035e-02 -2.67583817e-01 4.30148607e-03
-1.17930901e+00 1.08965158e+00 6.50736988e-01 6.62816644e-01
-1.79019839e-01 5.30727446e-01 3.40659380e-01 5.59065104e-01
1.65311500e-01 8.89169514e-01 -1.02364635e+00 7.69823045e-02
-3.24445516e-01 7.17820406e-01 8.31737161e-01 9.69339907e-01
8.46908510e-01 -9.58310589e-02 -3.92924342e-03 7.29134977e-01
9.01390463e-02 6.11400962e-01 8.13246787e-01 -9.34548020e-01
4.20862675e-01 8.79658282e-01 -1.57667086e-01 -7.55910695e-01
-6.75507724e-01 -2.75640488e-02 -5.74486852e-01 -4.13307816e-01
4.06153440e-01 -1.89600959e-01 -7.01733530e-01 2.37040901e+00
5.67727443e-03 -1.09149419e-01 5.34526944e-01 8.53736639e-01
1.28161645e+00 5.91442645e-01 4.85534072e-01 2.76544064e-01
1.70302916e+00 -3.48875314e-01 -4.45124596e-01 -7.03733802e-01
1.08993065e+00 -1.64852515e-01 1.77632821e+00 -1.48993447e-01
-9.90481615e-01 -3.34572941e-01 -9.34706628e-01 -5.32572508e-01
-9.02777374e-01 -5.23377478e-01 9.36366439e-01 3.92223895e-01
-1.24215782e+00 6.63510799e-01 -3.92663419e-01 -4.94332790e-01
4.57461417e-01 2.81618327e-01 -2.74105579e-01 -2.07220346e-01
-1.97861266e+00 9.82105076e-01 9.44956243e-01 -1.87019005e-01
-6.50974929e-01 -7.09548235e-01 -1.40765297e+00 4.80770797e-01
4.72023606e-01 -1.03520322e+00 1.50810575e+00 -1.30219245e+00
-8.23794007e-01 1.07257652e+00 -4.12518680e-01 -7.35141337e-01
-3.44183058e-01 -1.11920536e-02 -5.66257477e-01 -8.05111136e-03
3.46970469e-01 9.71472859e-01 3.60149145e-01 -8.84047985e-01
-4.08303410e-01 -4.57671255e-01 7.67293513e-01 1.37663156e-01
-2.54011393e-01 1.56225979e-01 5.72185159e-01 -6.05256319e-01
1.24460354e-01 -6.03387117e-01 6.94065839e-02 -1.64186090e-01
-3.78120303e-01 -6.75479949e-01 5.45824468e-01 -7.23485723e-02
8.32143784e-01 -2.04059315e+00 -1.30424388e-02 -1.60499960e-01
1.56450480e-01 1.08767539e-01 -4.12020624e-01 3.28844398e-01
-4.14191604e-01 5.64068079e-01 -4.66761410e-01 8.95798877e-02
5.17214656e-01 8.00753415e-01 -7.57606268e-01 -1.83760539e-01
5.57309508e-01 1.41512644e+00 -1.08252561e+00 -7.46086463e-02
-1.94437042e-01 1.18603250e-02 -5.00936329e-01 -1.54630274e-01
-9.11710799e-01 -6.20155558e-02 -3.50585341e-01 5.68918824e-01
4.53228503e-01 -4.43061084e-01 3.59152079e-01 5.49844950e-02
5.38542330e-01 1.02712905e+00 -8.83020997e-01 1.47575986e+00
-5.02702951e-01 5.66980243e-01 -1.98124975e-01 -1.09418547e+00
7.97765255e-01 2.08819240e-01 -3.87645990e-01 -6.92840874e-01
1.91427186e-01 2.33761892e-01 2.43511125e-01 -3.77779782e-01
5.64120471e-01 -1.05790806e+00 -5.00682056e-01 4.58067119e-01
2.38499448e-01 -3.86972994e-01 4.66656566e-01 2.66118199e-01
1.16368735e+00 -9.00203288e-02 5.98547578e-01 -5.62659442e-01
5.00158191e-01 1.43080279e-01 6.75612032e-01 8.72900665e-01
-6.35357946e-02 2.60778636e-01 7.70058692e-01 -6.12435043e-01
-6.71385586e-01 -1.41591513e+00 -1.70444757e-01 1.26335597e+00
2.81421155e-01 -5.82326114e-01 -4.81556207e-01 -6.66694105e-01
3.91712151e-02 1.29639030e+00 -4.15848851e-01 -4.08319235e-01
-5.14320195e-01 -5.28884768e-01 8.23536277e-01 8.19814324e-01
4.76887912e-01 -1.69317567e+00 -5.01245201e-01 1.07682660e-01
-2.90561855e-01 -1.19061375e+00 5.52928299e-02 3.02368045e-01
-9.07882333e-01 -1.03079987e+00 -5.66155799e-02 -9.25931454e-01
3.71916592e-01 -1.41987532e-01 1.75023818e+00 3.20673823e-01
5.26167732e-03 2.96045095e-01 -4.05569822e-01 -6.13827944e-01
-6.65793538e-01 6.53651729e-02 6.57874271e-02 -5.33195734e-01
9.90204096e-01 -7.13796198e-01 -1.21466957e-01 -5.91887347e-02
-1.44228065e+00 -1.88700363e-01 1.51233584e-01 7.37691879e-01
4.00514305e-01 1.55579090e-01 7.68473864e-01 -1.06118715e+00
9.77400661e-01 -5.88382542e-01 -4.26562876e-01 2.49158919e-01
-2.09687591e-01 4.49988872e-01 1.03684914e+00 -2.07385644e-01
-1.10477328e+00 -5.09064317e-01 -2.48276472e-01 3.76615748e-02
-6.12855434e-01 7.36491382e-01 -4.15470272e-01 5.32975137e-01
8.08407366e-01 2.14105800e-01 -2.89048105e-01 -3.05434883e-01
5.76738954e-01 1.87482402e-01 3.90978307e-01 -1.18168473e+00
5.29100120e-01 4.16207105e-01 1.12680383e-01 -6.47982717e-01
-1.41152632e+00 -7.97453448e-02 -3.97716224e-01 6.18882835e-01
8.51873219e-01 -6.79256141e-01 -4.90696400e-01 1.50176048e-01
-1.38494289e+00 -3.28628898e-01 -5.58754385e-01 4.07094099e-02
-5.59010863e-01 2.99854010e-01 -6.73859477e-01 -3.18403631e-01
-3.17343950e-01 -8.69129956e-01 1.06283545e+00 1.48369238e-01
-6.42400861e-01 -1.28343117e+00 -4.02193725e-01 -2.91919727e-02
4.00188416e-01 5.89360520e-02 1.71132517e+00 -1.05459428e+00
-3.28835130e-01 1.23592943e-01 -3.37322026e-01 5.20139277e-01
2.22477317e-01 -4.87233639e-01 -8.44963908e-01 2.62590367e-02
2.04872310e-01 -6.19675279e-01 9.20786858e-01 2.54488081e-01
1.19208193e+00 -5.56675732e-01 -7.22232535e-02 3.95628273e-01
1.37487125e+00 -5.18487096e-02 6.29165709e-01 3.34141850e-01
3.72196496e-01 5.55974603e-01 3.30137163e-01 -2.62075160e-02
6.00402951e-01 1.32895663e-01 4.78165835e-01 2.74623275e-01
-1.37270540e-01 -2.12463662e-01 3.29503745e-01 2.01941893e-01
3.56324375e-01 -3.74642730e-01 -1.09800303e+00 7.40929186e-01
-1.49083304e+00 -1.04217446e+00 8.80411714e-02 1.69065464e+00
1.16149700e+00 2.68282473e-01 -2.56020218e-01 -1.76423881e-02
6.95578873e-01 3.13961804e-01 -5.15311122e-01 -7.60380745e-01
-6.09063148e-01 7.19132245e-01 5.88778453e-03 4.79319930e-01
-9.39810812e-01 1.37596631e+00 5.88669395e+00 6.26661777e-01
-9.20417368e-01 -7.64282793e-03 3.84456187e-01 7.16128349e-02
-8.42212081e-01 2.99908459e-01 -8.19121957e-01 9.19403583e-02
1.00078380e+00 -1.56760037e-01 4.66747671e-01 6.69525027e-01
-1.06715105e-01 2.05749035e-01 -1.54685140e+00 6.13051593e-01
-6.27328977e-02 -1.41001058e+00 6.95677280e-01 -4.78156805e-01
4.01476264e-01 1.85325339e-01 -1.89613655e-01 5.26653528e-01
4.97178108e-01 -1.23021019e+00 6.89455509e-01 2.29786366e-01
5.62630832e-01 -5.27449548e-01 7.74227977e-01 2.02869579e-01
-1.03757024e+00 -2.45968133e-01 -6.64275527e-01 -5.70686340e-01
6.33857399e-02 3.24707091e-01 -6.96838379e-01 2.72992373e-01
5.88771284e-01 8.11170757e-01 -6.48941398e-01 2.80654877e-01
-9.58166480e-01 5.85682809e-01 -4.08897787e-01 -3.12922239e-01
3.63994569e-01 3.06268036e-01 6.08840704e-01 1.14075136e+00
1.54807016e-01 8.38632807e-02 -1.00843318e-01 1.62293839e+00
-3.20081711e-01 -1.30121067e-01 -7.43288100e-01 -2.75799811e-01
5.70875704e-01 8.92281234e-01 -5.03482461e-01 -6.14809275e-01
-2.79124677e-01 5.70785880e-01 4.15526062e-01 5.08389592e-01
-6.61610544e-01 -3.63622338e-01 9.06249106e-01 1.31429642e-01
9.34307724e-02 -2.89421696e-02 -2.47270197e-01 -1.47354293e+00
2.96471685e-01 -7.31875122e-01 7.74564385e-01 -1.14817011e+00
-1.62407255e+00 4.09368545e-01 3.79647672e-01 -5.40636480e-01
-3.27132195e-01 -1.13358390e+00 -7.71481276e-01 8.40128362e-01
-1.55229998e+00 -1.06343985e+00 1.19527131e-01 5.97111762e-01
3.94454658e-01 1.12812147e-02 1.10911345e+00 -2.54590392e-01
-7.09257647e-02 4.19714332e-01 -5.06758332e-01 2.46266603e-01
2.16312289e-01 -1.30645990e+00 6.63846016e-01 7.14682639e-01
2.35315219e-01 1.18258727e+00 7.26907074e-01 -4.09327507e-01
-1.05826247e+00 -1.11661196e+00 1.45500541e+00 -5.10407925e-01
8.64879668e-01 -3.94642562e-01 -1.08495080e+00 1.11881030e+00
1.51912168e-01 1.06888294e-01 5.80350935e-01 4.16152060e-01
-8.70057166e-01 1.45691469e-01 -1.28862131e+00 8.22884858e-01
1.38166702e+00 -6.72743917e-01 -1.60100579e+00 3.13675702e-01
1.24919331e+00 -1.48640558e-01 -5.17116666e-01 4.87818390e-01
2.08095983e-01 -7.47805715e-01 8.96568716e-01 -1.22844982e+00
7.57699788e-01 -2.47635454e-01 -7.13021338e-01 -1.32650435e+00
-1.30035847e-01 -9.31529552e-02 8.66397619e-02 9.21898127e-01
6.09858751e-01 -9.66594458e-01 4.41538692e-01 7.25081921e-01
-1.96327239e-01 -5.87122023e-01 -1.01430500e+00 -9.79040027e-01
5.67256033e-01 -7.01018453e-01 1.14430785e+00 9.92119908e-01
1.32103458e-01 9.43102479e-01 5.03276885e-01 -9.25333798e-03
2.52645224e-01 4.28551257e-01 1.60801798e-01 -1.47300172e+00
-2.53453046e-01 -5.42710960e-01 -6.43199742e-01 -9.47172105e-01
9.31405008e-01 -1.52055168e+00 -3.13046575e-01 -1.73643589e+00
1.17833316e-01 -1.11064091e-01 -1.91845417e-01 1.12028646e+00
-9.88704041e-02 -1.51222516e-02 2.66512930e-01 -2.08144218e-01
-5.44400811e-01 5.99030554e-01 1.23300087e+00 -3.76158714e-01
1.55389339e-01 -4.42727327e-01 -1.21414030e+00 9.76788819e-01
1.09149718e+00 -4.56475407e-01 -5.58128297e-01 -7.09567249e-01
6.95600569e-01 -3.22635353e-01 8.09272587e-01 -6.63419425e-01
-2.86824822e-01 -3.67202818e-01 9.78142023e-02 -2.48920590e-01
2.56194353e-01 -6.72730088e-01 -3.53698939e-01 2.89498925e-01
-6.45567834e-01 3.03669363e-01 5.50361335e-01 2.11567894e-01
-2.85556376e-01 -4.77066398e-01 5.83747387e-01 -5.29740393e-01
-1.02996981e+00 -8.42219517e-02 -3.11343253e-01 6.62987769e-01
4.28890079e-01 -7.95635507e-02 -6.33206725e-01 -3.41494352e-01
-8.23002100e-01 4.91284020e-02 3.63218367e-01 7.14830816e-01
7.29068756e-01 -1.31146884e+00 -5.52926898e-01 5.82615882e-02
3.24152261e-01 2.21982092e-01 -1.25309136e-02 2.65764683e-01
-3.15130353e-01 4.60573167e-01 2.70899478e-02 -2.19731554e-01
-8.39204311e-01 6.36246383e-01 5.41412771e-01 -1.46391727e-02
-5.29798508e-01 9.37993824e-01 7.16022551e-01 -9.10212874e-01
-3.17721993e-01 -1.14745843e+00 -3.52119431e-02 -3.04199249e-01
5.83073616e-01 -2.11878315e-01 -3.10202092e-02 -6.44616365e-01
-5.10739684e-01 1.76661909e-01 -2.47218944e-02 3.38039666e-01
1.20933104e+00 -1.32542392e-02 -5.81431210e-01 4.61452961e-01
1.20467365e+00 -4.28315014e-01 -4.62261736e-01 -3.65835845e-01
4.09128875e-01 1.37869138e-02 -3.09532166e-01 -8.74084771e-01
-8.04537296e-01 7.34547317e-01 1.94501746e-02 2.98826426e-01
1.06063187e+00 5.44562280e-01 9.92396057e-01 8.88787985e-01
3.13466638e-01 -6.90645576e-01 -2.09140345e-01 1.20724833e+00
1.03233790e+00 -9.48695540e-01 -5.20239532e-01 -2.60582358e-01
-2.74875611e-01 1.13409078e+00 6.12500429e-01 -2.04159185e-01
4.18880850e-01 7.63999075e-02 -8.52827057e-02 -5.41392207e-01
-9.62043703e-01 -2.76424497e-01 -3.78534794e-02 5.69204986e-01
5.51720738e-01 3.57760668e-01 -2.08460823e-01 9.34241593e-01
-7.31597006e-01 -1.65924996e-01 5.63975155e-01 8.12671840e-01
-4.32271272e-01 -1.07635415e+00 -2.58316845e-01 5.24591565e-01
-4.77317631e-01 -8.32382977e-01 -6.31173849e-01 7.53737032e-01
1.16327286e-01 8.39452982e-01 5.79873770e-02 2.51811687e-02
2.88890064e-01 4.90808308e-01 6.10797763e-01 -1.14385402e+00
-4.49272901e-01 -7.31116295e-01 3.94492239e-01 -5.53415537e-01
-4.46260661e-01 -4.25251871e-01 -1.91988230e+00 -2.45544776e-01
8.17748830e-02 4.52173762e-02 1.22340955e-01 1.23997951e+00
3.15711856e-01 3.19974363e-01 -1.33960530e-01 -3.11604261e-01
-7.96342492e-01 -8.00196886e-01 -6.03466809e-01 9.20861125e-01
3.05890560e-01 -6.28509760e-01 -5.00319898e-01 -1.62018046e-01] | [10.124689102172852, 8.457382202148438] |
1b199aea-de33-4e9f-b783-cf5e9be6a229 | fast-and-accurate-head-pose-estimation-via | null | null | http://openaccess.thecvf.com/content_iccv_2015/html/Lee_Fast_and_Accurate_ICCV_2015_paper.html | http://openaccess.thecvf.com/content_iccv_2015/papers/Lee_Fast_and_Accurate_ICCV_2015_paper.pdf | Fast and Accurate Head Pose Estimation via Random Projection Forests | In this paper, we consider the problem of estimating the gaze direction of a person from a low-resolution image. Under this condition, reliably extracting facial features is very difficult. We propose a novel head pose estimation algorithm based on compressive sensing. Head image patches are mapped to a large feature space using the proposed extensive, yet efficient filter bank. The filter bank is designed to generate sparse responses of color and gradient information, which can be compressed using random projection, and classified by a random forest. Extensive experiments on challenging datasets show that the proposed algorithm performs favorably against the state-of-the-art methods on head pose estimation in low-resolution images degraded by noise, occlusion, and blurring. | ['Ming-Hsuan Yang', 'Songhwai Oh', 'Donghoon Lee'] | 2015-12-01 | null | null | null | iccv-2015-12 | ['head-pose-estimation'] | ['computer-vision'] | [ 5.22491038e-01 -2.02264741e-01 1.15430750e-01 -5.32239437e-01
-6.00161791e-01 6.03536032e-02 2.18072191e-01 -9.03375030e-01
-3.72333109e-01 9.06371415e-01 6.60984516e-01 3.90886873e-01
7.08692595e-02 -2.03131959e-01 -6.48754239e-01 -9.98371422e-01
1.00802287e-01 -1.40692070e-01 -2.79560030e-01 2.41315842e-01
3.76287162e-01 4.72205728e-01 -2.25027227e+00 6.53209835e-02
7.55398273e-01 1.16258705e+00 3.12119812e-01 1.60420731e-01
5.47856629e-01 5.54153025e-01 -2.89623439e-01 -3.18691097e-02
2.81346768e-01 -3.86878669e-01 -2.84255147e-01 4.36507583e-01
9.96782780e-01 -5.73297679e-01 -5.05759358e-01 1.27549183e+00
9.21990216e-01 3.44036669e-02 5.30700147e-01 -7.64488637e-01
-4.72398341e-01 4.13628183e-02 -8.71541917e-01 1.58126742e-01
1.09810460e+00 -1.90064400e-01 3.16847801e-01 -1.44606817e+00
6.90390229e-01 1.46991658e+00 3.82429570e-01 6.45102620e-01
-9.10852015e-01 -9.52997327e-01 -1.21126145e-01 7.59645998e-01
-1.87685621e+00 -1.14421129e+00 8.07955086e-01 -1.34843424e-01
3.35372984e-01 3.67838770e-01 6.71400130e-01 8.93685758e-01
1.26547366e-01 6.77354634e-01 1.55029678e+00 -5.60638666e-01
2.14101195e-01 -1.30198881e-01 -1.28203869e-01 8.31986785e-01
5.42195976e-01 1.02818705e-01 -1.16377747e+00 -3.11603814e-01
4.59849119e-01 -1.32825570e-02 -1.07373095e+00 -2.03591824e-01
-7.28696764e-01 5.49665272e-01 2.47146845e-01 9.56707075e-02
-6.83494925e-01 -3.80899400e-01 -5.22429228e-01 -1.82943553e-01
4.91768003e-01 -1.65714696e-01 2.56017238e-01 3.11763287e-01
-1.15189195e+00 2.20574498e-01 7.25643754e-01 8.91022801e-01
5.23267627e-01 -2.07526032e-02 -1.28611311e-01 8.74175072e-01
6.91913962e-01 1.01059055e+00 6.11794233e-01 -8.36282015e-01
1.32709265e-01 -1.34683400e-01 1.12796701e-01 -9.74745393e-01
-4.81759727e-01 -3.67747188e-01 -8.54051292e-01 4.23923396e-02
2.27758542e-01 -3.91487718e-01 -7.01657236e-01 1.63038468e+00
6.69669807e-01 3.29835445e-01 -2.35748187e-01 1.38419616e+00
7.88436592e-01 3.45327348e-01 -4.19346273e-01 -8.65861058e-01
1.45665705e+00 -4.76802886e-01 -1.17244327e+00 -5.83596110e-01
-4.30294812e-01 -8.49422097e-01 2.79717416e-01 6.52691483e-01
-1.22048664e+00 -4.79326665e-01 -8.90752852e-01 1.41214030e-02
3.49598318e-01 3.66464764e-01 4.28916126e-01 8.38680863e-01
-1.12777209e+00 3.81777138e-02 -4.30020720e-01 -4.01825011e-01
4.34778959e-01 3.70070755e-01 -5.66066027e-01 -7.11274743e-01
-8.72702718e-01 8.18580866e-01 -3.08979392e-01 5.11599541e-01
-3.77172559e-01 -4.41734381e-02 -1.10076416e+00 -1.72677428e-01
-1.71659086e-02 -7.27249920e-01 8.87912273e-01 -5.16120136e-01
-1.63185072e+00 6.88552082e-01 -9.68693078e-01 -1.73028275e-01
1.61838576e-01 -4.09054220e-01 -5.70607483e-01 5.16673088e-01
7.92443827e-02 3.84061307e-01 1.82134068e+00 -1.00199997e+00
-3.82422209e-01 -1.16974413e+00 -7.94963360e-01 5.47948182e-01
-3.06970775e-01 2.51186997e-01 -4.25706327e-01 -3.16128910e-01
6.85431659e-01 -9.85706627e-01 1.98429719e-01 -9.14307088e-02
-2.83758849e-01 9.61259529e-02 8.13293219e-01 -9.39830959e-01
1.21882689e+00 -2.20266891e+00 7.69801587e-02 1.80631071e-01
1.48122787e-01 -6.94894418e-02 1.31617323e-01 -2.90942013e-01
-1.38562635e-01 -6.90398097e-01 3.19965146e-02 -3.61311018e-01
-2.97548980e-01 -2.65852541e-01 -1.01876497e-01 1.37657034e+00
-3.08913708e-01 5.03983259e-01 -4.53598559e-01 -6.47628903e-01
3.47737269e-03 8.61319423e-01 -5.13427496e-01 3.42479259e-01
5.85951924e-01 6.67151570e-01 -3.36559623e-01 8.84876251e-01
1.17893386e+00 -1.77211121e-01 4.78965342e-02 -3.72083604e-01
-5.77400848e-02 -4.96935770e-02 -1.45774901e+00 1.58269882e+00
5.11656143e-02 5.66559494e-01 4.22812343e-01 -6.09865189e-01
8.11286092e-01 2.49228254e-01 1.81323662e-01 -6.13478839e-01
2.84875900e-01 5.78840077e-02 -3.39254767e-01 -5.94383359e-01
4.52521414e-01 -1.59595296e-01 2.58413136e-01 4.64833260e-01
-9.42019373e-02 1.54083058e-01 -2.16665551e-01 -2.06015930e-01
7.66475976e-01 -5.85526340e-02 3.78168195e-01 2.94503872e-03
7.85927474e-01 -7.99959183e-01 5.49718320e-01 4.00953561e-01
-2.43353888e-01 7.87563324e-01 -2.82236934e-01 -1.66167378e-01
-4.89134073e-01 -8.56972754e-01 -5.83599806e-01 9.87121165e-01
1.39762133e-01 -1.03718363e-01 -1.08198094e+00 -1.50918752e-01
-4.85894233e-02 3.21422905e-01 -5.31535327e-01 5.77773191e-02
-5.88470340e-01 -7.71193147e-01 -6.07649563e-03 7.67748505e-02
6.95382237e-01 -7.80667186e-01 -6.56681240e-01 -1.24572203e-01
-4.73682344e-01 -1.28300965e+00 -7.84657001e-01 -6.41987681e-01
-5.72979927e-01 -8.89201462e-01 -1.16839898e+00 -1.03403533e+00
1.07232153e+00 7.31122255e-01 6.81366503e-01 -3.09617013e-01
-4.33574468e-01 4.05927330e-01 -6.59192130e-02 -2.84388632e-01
5.24426520e-01 -4.82610255e-01 5.12873530e-01 7.83403516e-01
6.89974606e-01 -3.45355093e-01 -9.93500292e-01 3.36048812e-01
-3.64880532e-01 -1.73774958e-01 6.92806065e-01 8.47459793e-01
5.59777081e-01 -4.71313223e-02 1.76771343e-01 -4.90826458e-01
5.97110569e-01 -2.59856135e-01 -5.22107184e-01 -1.06461592e-01
-4.64645863e-01 -6.11165874e-02 1.76066950e-01 -4.68334794e-01
-1.42105770e+00 4.20526087e-01 1.74602374e-01 -2.92973250e-01
-3.00268948e-01 1.34687066e-01 -2.83377707e-01 -4.62482840e-01
6.28213465e-01 5.35357833e-01 3.45084548e-01 -5.32954097e-01
1.89726576e-01 1.12521791e+00 7.13936090e-01 6.52829856e-02
7.13649094e-01 7.52527773e-01 1.62073046e-01 -1.29691172e+00
-6.90264344e-01 -5.53342283e-01 -5.58606744e-01 -2.23117396e-01
6.49415910e-01 -1.29036880e+00 -6.64502323e-01 7.15550601e-01
-7.79944003e-01 4.48778510e-01 2.49712989e-01 8.75120699e-01
-3.62555593e-01 4.72836554e-01 -3.62331331e-01 -9.91330743e-01
-5.27437747e-01 -8.74302387e-01 1.11494482e+00 5.56430876e-01
2.74676178e-02 -1.86328411e-01 -1.78975016e-01 5.52277565e-01
4.83251423e-01 -1.62773743e-01 1.32631034e-01 2.17357382e-01
-5.83723664e-01 -2.86619097e-01 -1.33157065e-02 3.14256921e-02
6.56014532e-02 -6.44379318e-01 -1.29609501e+00 -6.49547935e-01
5.62797964e-01 -2.22119659e-01 5.69486082e-01 8.88870239e-01
1.15732086e+00 -4.62428898e-01 -3.93154442e-01 9.52710986e-01
1.15449035e+00 -2.00011075e-01 6.99963689e-01 -1.07549436e-01
5.90004504e-01 6.73348188e-01 6.71283662e-01 7.92971790e-01
5.29330373e-01 6.29287183e-01 1.13864802e-01 1.79807514e-01
-1.50604755e-01 -1.00325823e-01 3.38139921e-01 5.52446663e-01
-1.00302458e-01 1.78342476e-01 -3.32441628e-01 2.96862423e-01
-1.45177388e+00 -1.12253237e+00 2.18678668e-01 2.21044803e+00
6.57353282e-01 -5.07464230e-01 5.73327765e-02 2.20024347e-01
9.46636558e-01 1.70424536e-01 -5.10004044e-01 2.95109749e-01
-2.23726198e-01 3.33710939e-01 4.06700283e-01 3.70855033e-01
-8.41111720e-01 6.59314752e-01 6.88309574e+00 3.54513407e-01
-1.25411880e+00 1.12414978e-01 3.06794018e-01 -3.60082328e-01
7.16280518e-03 -5.01544058e-01 -1.16802835e+00 5.66408753e-01
6.63798332e-01 -8.61851349e-02 6.28642142e-01 6.00004196e-01
1.39599502e-01 -4.51249897e-01 -7.48357356e-01 1.77980971e+00
9.77203608e-01 -6.96051359e-01 -4.01999027e-01 2.16013834e-01
5.87584615e-01 -1.75746575e-01 4.64372218e-01 -3.08995426e-01
-3.22928041e-01 -1.20738971e+00 5.49753368e-01 7.45777667e-01
1.10052288e+00 -5.16032875e-01 4.20295715e-01 3.26132506e-01
-9.88675594e-01 -3.43520463e-01 -4.85286564e-01 -7.57924691e-02
1.05973236e-01 5.63110471e-01 -5.57579041e-01 -3.78292538e-02
8.70217860e-01 6.11948490e-01 -4.71736044e-01 1.16266251e+00
-3.55493248e-01 4.21091914e-01 -3.94397140e-01 4.61734012e-02
-5.62909186e-01 -1.26376495e-01 5.41342199e-01 7.72932768e-01
5.25091469e-01 7.25880027e-01 -2.35234544e-01 5.16111910e-01
8.04223958e-03 1.43211871e-01 -7.48182297e-01 6.06426597e-01
4.26183701e-01 1.21931481e+00 -6.21537007e-02 -4.94505800e-02
-5.20372629e-01 9.64248419e-01 4.44679447e-02 5.74972808e-01
-4.52965856e-01 -2.81831443e-01 4.93939131e-01 3.91496062e-01
6.17078662e-01 4.81674150e-02 -5.10544367e-02 -1.36265135e+00
3.33257496e-01 -9.01899338e-01 1.15841076e-01 -9.77946401e-01
-1.03025520e+00 6.40636981e-01 -1.69796199e-01 -1.38923931e+00
-5.29564142e-01 -2.91729927e-01 -1.15382425e-01 1.12671781e+00
-1.74072421e+00 -9.25955951e-01 -6.59559846e-01 1.30306411e+00
2.15295359e-01 -2.76043326e-01 7.45446622e-01 1.36668384e-01
-4.20072854e-01 7.74567723e-01 -2.44514202e-03 -2.29713231e-01
9.35865998e-01 -7.09321976e-01 -2.57204354e-01 9.00001764e-01
-2.51864523e-01 6.41404390e-01 8.42672944e-01 -5.16808391e-01
-1.86345613e+00 -7.73181081e-01 9.03846204e-01 6.04198538e-02
-9.62786376e-02 -3.62443388e-01 -7.40829110e-01 4.01737124e-01
8.71338621e-02 5.00722349e-01 8.76328468e-01 -1.21961929e-01
-1.49130791e-01 -2.19894469e-01 -1.54265654e+00 8.38446766e-02
1.11137295e+00 -5.14743924e-01 -7.52771199e-01 2.92900950e-01
-1.94005206e-01 -5.30881107e-01 -6.00784063e-01 2.51368970e-01
9.94665027e-01 -9.48034823e-01 9.18336034e-01 7.46287778e-02
-1.39633805e-01 -2.40911961e-01 -2.04287738e-01 -1.18104076e+00
-6.06408715e-01 -9.04067755e-01 -3.12304199e-01 6.06480777e-01
-1.83872998e-01 -7.48824060e-01 8.26023698e-01 5.47736108e-01
4.28963423e-01 -2.70859778e-01 -9.54612434e-01 -2.11352333e-01
-5.77701926e-01 8.31229314e-02 6.01454258e-01 4.50191230e-01
1.43409997e-01 4.84020025e-01 -6.95039749e-01 4.28342581e-01
1.40120530e+00 4.13318157e-01 6.66133940e-01 -1.34963989e+00
-4.35990840e-02 2.34992117e-01 -5.70457458e-01 -1.29232860e+00
4.59472775e-01 -1.96345940e-01 9.65527147e-02 -1.12172437e+00
3.48192185e-01 1.61230683e-01 -5.13248928e-02 -2.53429152e-02
-2.18117684e-01 5.55351794e-01 -7.21265599e-02 3.89440805e-01
-4.85062182e-01 5.52090526e-01 1.27761066e+00 5.52287586e-02
3.49991620e-02 2.32124165e-01 -8.97025406e-01 8.83608401e-01
3.28792423e-01 -1.95518598e-01 -1.44369125e-01 -2.95756012e-01
-1.82061210e-01 5.14491975e-01 1.40361711e-01 -1.11922228e+00
6.12292349e-01 4.53592539e-02 9.87486720e-01 -7.93968499e-01
7.71531105e-01 -6.75555468e-01 1.80012248e-02 2.07796574e-01
2.21004877e-02 -1.76627383e-01 -2.45040283e-01 5.95697343e-01
-2.75209010e-01 1.42979383e-01 1.04187119e+00 2.00196020e-02
-5.77699482e-01 4.36038345e-01 -1.21459708e-01 -2.09259599e-01
7.69982338e-01 -3.16923440e-01 -1.05652139e-01 -7.73332000e-01
-6.68871284e-01 -2.82752067e-01 4.31257695e-01 1.68292508e-01
1.14639866e+00 -1.35849226e+00 -9.26396310e-01 1.08198452e+00
-1.52151629e-01 -2.93037325e-01 3.95866483e-01 8.45460176e-01
-8.10956284e-02 5.08117020e-01 -4.53685105e-01 -7.10902989e-01
-1.64011157e+00 3.59917879e-01 2.08914950e-02 4.93321210e-01
-5.06802917e-01 9.36128020e-01 1.48705661e-01 1.86908364e-01
2.49092847e-01 2.24796548e-01 -7.09159970e-01 8.20571482e-02
1.23407996e+00 3.79540443e-01 7.09233433e-02 -1.37046981e+00
-5.43930411e-01 1.07400882e+00 9.87516716e-02 -2.31671557e-01
1.38792992e+00 -6.16766214e-01 -2.75801659e-01 -1.08303465e-01
1.24017954e+00 3.69259834e-01 -1.06391418e+00 -5.87528288e-01
-6.17969453e-01 -1.03089213e+00 2.96939850e-01 -3.01811635e-01
-1.16896582e+00 6.04245901e-01 1.13187253e+00 -3.11330378e-01
1.59838593e+00 -1.15510970e-01 5.96883297e-01 1.33626431e-01
7.21999884e-01 -8.70567203e-01 -2.17192873e-01 1.48087293e-01
9.25963700e-01 -1.30625355e+00 4.07136530e-01 -4.49591935e-01
-3.39527041e-01 8.96328568e-01 3.09874028e-01 -4.98725055e-03
8.92440319e-01 1.72918737e-01 -5.87570556e-02 -4.98081967e-02
-4.97267723e-01 -3.07365596e-01 5.49891055e-01 6.65976524e-01
1.94328263e-01 6.37223497e-02 -2.89864272e-01 2.56090015e-01
-5.41818917e-01 3.12351435e-01 4.50952679e-01 8.46642852e-01
-7.24680960e-01 -5.28404057e-01 -1.16377044e+00 4.01468456e-01
-6.91342235e-01 -4.37544249e-02 5.87545969e-02 1.25661641e-01
-1.07548974e-01 1.19052052e+00 -1.45991100e-02 -1.94509983e-01
9.17929411e-02 -5.53861149e-02 8.54073584e-01 -4.09361333e-01
3.10834795e-01 5.78505814e-01 -2.33151123e-01 -7.97240853e-01
-5.77252984e-01 -9.46399748e-01 -5.98694146e-01 -9.59993675e-02
-3.91499102e-01 1.44610405e-02 7.48756170e-01 6.85283244e-01
3.93457383e-01 -1.55643195e-01 8.98223162e-01 -1.24581361e+00
-6.37441576e-01 -1.22798753e+00 -1.08555794e+00 4.30309087e-01
6.81628704e-01 -8.43141079e-01 -6.28569603e-01 1.26031507e-02] | [13.16814136505127, 0.2411925047636032] |
d38d9bd0-4dd5-4494-b8c9-f6585a70c0cf | fostering-generalization-in-single-view-3d | 2104.00476 | null | https://arxiv.org/abs/2104.00476v1 | https://arxiv.org/pdf/2104.00476v1.pdf | Fostering Generalization in Single-view 3D Reconstruction by Learning a Hierarchy of Local and Global Shape Priors | Single-view 3D object reconstruction has seen much progress, yet methods still struggle generalizing to novel shapes unseen during training. Common approaches predominantly rely on learned global shape priors and, hence, disregard detailed local observations. In this work, we address this issue by learning a hierarchy of priors at different levels of locality from ground truth input depth maps. We argue that exploiting local priors allows our method to efficiently use input observations, thus improving generalization in visible areas of novel shapes. At the same time, the combination of local and global priors enables meaningful hallucination of unobserved parts resulting in consistent 3D shapes. We show that the hierarchical approach generalizes much better than the global approach. It generalizes not only between different instances of a class but also across classes and to unseen arrangements of objects. | ['Thomas Brox', 'Volker Fischer', 'Maxim Tatarchenko', 'Jan Bechtold'] | 2021-04-01 | null | http://openaccess.thecvf.com//content/CVPR2021/html/Bechtold_Fostering_Generalization_in_Single-View_3D_Reconstruction_by_Learning_a_Hierarchy_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Bechtold_Fostering_Generalization_in_Single-View_3D_Reconstruction_by_Learning_a_Hierarchy_CVPR_2021_paper.pdf | cvpr-2021-1 | ['single-view-3d-reconstruction', '3d-object-reconstruction'] | ['computer-vision', 'computer-vision'] | [ 2.80199805e-03 2.03178525e-01 1.05235100e-01 -3.48888874e-01
-5.44266045e-01 -8.96346509e-01 6.89641118e-01 1.91588506e-01
3.15403976e-02 4.96702403e-01 3.63122016e-01 2.06693694e-01
-1.48811817e-01 -9.49491799e-01 -7.14335382e-01 -7.09081829e-01
2.50310361e-01 8.52667332e-01 7.26662695e-01 8.86896253e-02
1.85790315e-01 1.16725826e+00 -1.61793387e+00 4.45463002e-01
1.91156924e-01 3.96436632e-01 4.76114333e-01 4.83474255e-01
-3.47338505e-02 3.49688411e-01 -3.31140220e-01 -2.62255341e-01
4.45282459e-01 -7.84666389e-02 -6.07814729e-01 6.50486469e-01
6.44057691e-01 -3.91275227e-01 -3.00839782e-01 9.49642599e-01
2.95603395e-01 2.83559382e-01 7.47391939e-01 -8.25275183e-01
-7.48087287e-01 -1.22012809e-01 -4.79810447e-01 1.27339745e-02
4.74495322e-01 2.10571572e-01 9.21380997e-01 -1.16706443e+00
9.29408669e-01 1.36005199e+00 6.31598175e-01 5.87738931e-01
-1.58576512e+00 -1.47846013e-01 5.76399267e-01 -1.87834144e-01
-1.33621085e+00 -4.21536237e-01 9.26826835e-01 -5.98837972e-01
1.02325511e+00 2.58418340e-02 6.59248412e-01 1.14757252e+00
2.08559081e-01 7.81871080e-01 1.18383312e+00 -1.75134182e-01
3.56531680e-01 1.55327410e-01 -2.04567149e-01 6.50527298e-01
4.69323814e-01 2.14269280e-01 -4.81937885e-01 -1.70404851e-01
1.22412312e+00 3.50550056e-01 -3.16836566e-01 -9.91879284e-01
-1.07752025e+00 5.91285229e-01 4.27721649e-01 3.10648084e-01
-4.49297547e-01 -6.74412251e-02 -1.73235983e-01 -9.70284194e-02
5.63920438e-01 3.50950778e-01 -6.85676575e-01 3.50736320e-01
-6.94719970e-01 2.75821298e-01 6.11789644e-01 1.11462080e+00
1.06708670e+00 -6.87873289e-02 1.80083752e-01 6.81963027e-01
2.76428074e-01 4.71090049e-01 -4.44853678e-02 -1.08441746e+00
7.79976994e-02 5.66398144e-01 1.56765744e-01 -9.41752136e-01
-4.06642973e-01 -5.98774672e-01 -5.88458121e-01 6.94311678e-01
4.01927769e-01 2.51340300e-01 -1.33317888e+00 1.82215321e+00
3.42551529e-01 -2.59804893e-02 -1.36217728e-01 6.72980011e-01
6.34558856e-01 3.40892076e-01 -1.07547112e-01 4.41416651e-02
1.01521754e+00 -6.36503994e-01 -2.83906102e-01 -4.35158342e-01
-4.35160547e-02 -6.40307128e-01 9.14442062e-01 4.45513785e-01
-1.35120189e+00 -5.74477196e-01 -8.24296951e-01 -3.33933055e-01
-4.41577017e-01 -3.38875890e-01 4.94337469e-01 5.09681165e-01
-1.17410028e+00 5.41379094e-01 -9.76469934e-01 -6.16817534e-01
6.02091312e-01 2.85095036e-01 -6.34139180e-01 -4.40077424e-01
-2.21584126e-01 8.71497273e-01 3.75499964e-01 -2.94747889e-01
-1.18584561e+00 -6.78152978e-01 -8.45839977e-01 2.18162034e-02
3.71156424e-01 -1.14967418e+00 1.12105823e+00 -6.00961924e-01
-1.35202098e+00 1.06302106e+00 -4.17575598e-01 -4.11523059e-02
3.35062176e-01 -7.35266879e-02 1.33592486e-01 1.61769822e-01
-2.53588296e-02 8.61296475e-01 8.80420625e-01 -1.96398091e+00
-3.27998668e-01 -7.93157399e-01 2.32399032e-01 2.52962559e-01
1.44354790e-01 -5.20313323e-01 -3.58459353e-01 -4.59793717e-01
7.60447025e-01 -6.21412694e-01 -2.98595190e-01 2.98081577e-01
-1.26306802e-01 -9.76831559e-03 8.47060800e-01 -2.96540022e-01
3.64061862e-01 -2.10581183e+00 4.03454632e-01 2.06655920e-01
3.12421083e-01 -2.62089431e-01 -5.94764873e-02 4.68178988e-01
1.66418537e-01 -1.29629612e-01 -3.49182516e-01 -6.27289951e-01
-7.30402470e-02 7.70178735e-01 -4.55539614e-01 4.15197909e-01
1.02232099e-01 9.71338511e-01 -1.04155838e+00 -1.61396816e-01
3.92667055e-01 6.02129459e-01 -7.75207400e-01 1.21687688e-01
-3.73309374e-01 8.06045175e-01 -3.90563935e-01 6.47837698e-01
9.27322209e-01 -3.38960230e-01 3.05302143e-02 -1.26496572e-02
2.66393181e-02 1.26869410e-01 -1.20971954e+00 1.98651505e+00
-4.21705842e-01 2.49869093e-01 1.03403419e-01 -8.24966252e-01
9.26187754e-01 2.79902130e-01 3.22653323e-01 -2.77200431e-01
-1.30791962e-01 1.03301756e-01 -3.96605462e-01 -4.21000123e-01
1.26496583e-01 -7.59854734e-01 2.47354031e-01 2.08530560e-01
4.43407029e-01 -5.61107993e-01 -4.76060003e-01 3.17819193e-02
9.84203756e-01 4.69739646e-01 5.56849539e-01 -1.81763813e-01
8.55873749e-02 -1.15838863e-01 5.03224313e-01 9.72528160e-01
3.96499783e-02 1.08674419e+00 8.94639269e-03 -6.42401814e-01
-1.06928051e+00 -1.89706230e+00 -2.71581471e-01 7.61152923e-01
3.83107245e-01 -1.50496736e-01 -7.40357339e-02 -7.63523042e-01
1.72797233e-01 7.40849972e-01 -7.28252769e-01 9.79405120e-02
-3.79856795e-01 -3.98724884e-01 -1.05075128e-02 7.09481716e-01
2.45405972e-01 -1.05213082e+00 -8.23415399e-01 1.73389018e-01
2.00656801e-01 -1.30844557e+00 -9.11200568e-02 2.53945589e-01
-1.39578450e+00 -6.96348250e-01 -7.57181406e-01 -5.89623749e-01
9.35720205e-01 6.46086395e-01 1.28376830e+00 -7.96291977e-02
-2.28826165e-01 9.90554631e-01 -1.85118496e-01 -2.86811054e-01
-1.66794077e-01 -2.16537192e-01 -7.04324171e-02 2.41629276e-02
1.95731327e-01 -1.04274523e+00 -2.98575163e-01 2.08029777e-01
-8.10217321e-01 -7.80770881e-03 6.49294913e-01 7.72461355e-01
7.31256723e-01 -1.66175142e-01 2.62748271e-01 -9.16997254e-01
1.07713500e-02 -4.29640770e-01 -5.29644012e-01 2.19841093e-01
-1.07534103e-01 9.98110324e-02 4.80156422e-01 -2.56135792e-01
-1.45165336e+00 2.06464842e-01 -7.61743039e-02 -5.93573391e-01
-8.44888926e-01 1.20747723e-01 -4.19920355e-01 -2.44439654e-02
7.39416063e-01 2.05642283e-01 -1.90561056e-01 -7.90081501e-01
4.67854172e-01 -3.58341299e-02 3.94132107e-01 -8.15319777e-01
8.01253974e-01 1.16626763e+00 1.25763357e-01 -1.07271302e+00
-9.16588843e-01 -4.71123308e-01 -1.13239658e+00 -8.00864100e-02
9.17850256e-01 -8.45370650e-01 -2.49860927e-01 1.13198996e-01
-1.16592181e+00 -2.20411465e-01 -7.68129230e-01 3.71029079e-01
-8.24338377e-01 3.84684205e-01 -2.80632257e-01 -7.54950941e-01
5.30620992e-01 -9.55517232e-01 1.41977561e+00 3.24202329e-02
-6.58450425e-02 -1.28029585e+00 6.20885529e-02 1.04831457e-01
1.26971915e-01 3.30366164e-01 8.81767154e-01 -3.41494799e-01
-1.11293840e+00 -1.41520992e-01 -2.19653547e-01 1.04353644e-01
2.62311459e-01 -3.30484122e-01 -1.33955288e+00 -4.08564806e-01
2.93645024e-01 -4.07496020e-02 9.97580469e-01 4.01441395e-01
1.05794168e+00 -8.44287127e-02 -4.83968854e-01 5.83177447e-01
1.52737880e+00 -8.84258933e-03 5.56216776e-01 -7.42135420e-02
5.10801911e-01 6.63589418e-01 -1.88160576e-02 3.30778390e-01
2.69060165e-01 5.87465286e-01 6.47792935e-01 4.05206606e-02
-2.60171413e-01 -5.08530974e-01 7.38933906e-02 4.26655322e-01
-2.86390871e-01 -2.41474763e-01 -8.48625541e-01 8.20542812e-01
-1.61076641e+00 -9.02189851e-01 1.82198301e-01 2.31712055e+00
4.00461376e-01 1.92959204e-01 -1.81929171e-01 -2.98075527e-01
3.74966711e-01 -1.78045519e-02 -5.59584558e-01 -1.05177149e-01
-2.17514887e-01 3.31085831e-01 2.11952284e-01 8.01769197e-01
-6.91928685e-01 8.62346113e-01 6.81314373e+00 4.80505049e-01
-6.06970429e-01 2.26291046e-01 2.12202996e-01 1.45005897e-01
-7.58386672e-01 3.61682713e-01 -7.35225379e-01 -2.99724609e-01
1.09078608e-01 2.48157978e-01 3.01604837e-01 6.06349170e-01
-2.33403236e-01 -1.45055577e-01 -1.29385912e+00 7.85619557e-01
2.25688562e-01 -1.16331995e+00 4.07712907e-01 2.44419307e-01
1.01224649e+00 5.49011566e-02 2.85215173e-02 -3.07896305e-02
4.27741259e-01 -9.84134316e-01 6.15929663e-01 7.16093063e-01
2.36612558e-01 -4.17469919e-01 4.26396489e-01 8.19771469e-01
-1.00904977e+00 2.06194166e-02 -6.49475634e-01 -2.13117287e-01
1.78506032e-01 5.77006757e-01 -7.45067239e-01 6.21786118e-01
6.14399672e-01 7.82182157e-01 -4.75664854e-01 1.17265141e+00
-4.24691200e-01 3.20579088e-03 -5.49192667e-01 4.23760027e-01
1.05577186e-01 -6.91756234e-02 8.17682803e-01 9.11479831e-01
3.83480430e-01 3.83807659e-01 3.38160485e-01 1.09754944e+00
1.44064024e-01 -1.74419537e-01 -1.12853086e+00 4.48185295e-01
1.86182842e-01 9.52448547e-01 -9.04746890e-01 -2.60913849e-01
-5.53954244e-01 9.74735796e-01 5.69611132e-01 6.76893651e-01
-2.36576557e-01 3.10525745e-01 4.85378772e-01 3.15246642e-01
5.71861029e-01 -7.35219300e-01 -5.60890734e-01 -1.37905622e+00
-1.04372144e-01 -2.52130538e-01 3.59046280e-01 -1.02025175e+00
-1.66146016e+00 3.02150100e-01 4.38357234e-01 -1.18426704e+00
1.95959017e-01 -8.00521314e-01 -4.09155011e-01 6.54618442e-01
-1.46496904e+00 -1.28386569e+00 -3.91778111e-01 5.84195554e-01
6.89147174e-01 1.45831451e-01 9.04973745e-01 -2.82217294e-01
2.55925864e-01 4.61678691e-02 -1.69859976e-01 -3.19116354e-01
5.15829325e-01 -1.29367518e+00 1.76348880e-01 6.70469880e-01
6.43488944e-01 7.80729294e-01 6.92114353e-01 -6.60785913e-01
-1.03017759e+00 -7.50352383e-01 6.50000215e-01 -9.26878273e-01
2.70471752e-01 -4.11069602e-01 -1.05251598e+00 9.72380280e-01
2.71374553e-01 2.00246572e-01 4.70824540e-01 2.54559100e-01
-6.40696406e-01 1.93779796e-01 -1.26289368e+00 5.53053856e-01
1.21561134e+00 -6.76068425e-01 -8.44402671e-01 2.43566677e-01
3.18533659e-01 -4.75499332e-01 -7.41899729e-01 4.08611238e-01
5.10575593e-01 -1.54895401e+00 1.33108151e+00 -6.17503703e-01
7.23347440e-02 -3.82954150e-01 -6.30991042e-01 -1.16557002e+00
-4.17132705e-01 -1.46102697e-01 -1.38982683e-01 7.42775619e-01
1.96135506e-01 -6.54303551e-01 7.46135533e-01 3.43666226e-01
-3.32791120e-01 -5.93845308e-01 -8.70492518e-01 -9.74265575e-01
2.69074380e-01 -4.38016713e-01 4.67406392e-01 8.49784672e-01
-2.95901686e-01 2.85365880e-01 -6.78331852e-02 5.97775757e-01
9.37107146e-01 4.55230623e-01 8.31046879e-01 -1.51143265e+00
-6.08173430e-01 -3.31802100e-01 -6.76302612e-01 -1.51502371e+00
5.83228953e-02 -9.55992818e-01 1.21958368e-01 -1.71169531e+00
4.19887811e-01 -1.86065987e-01 8.12338144e-02 4.60136414e-01
-1.20652898e-03 4.45993304e-01 1.71928063e-01 1.67610824e-01
-3.30739677e-01 5.55442929e-01 1.60032916e+00 1.47510007e-01
-1.72313049e-01 6.62919581e-02 -4.51876521e-01 1.11423790e+00
6.58834219e-01 -4.93210256e-01 -4.38581288e-01 -7.50575721e-01
1.20885953e-01 -1.70980543e-01 9.18859780e-01 -8.10940385e-01
2.35754535e-01 -3.41403008e-01 8.35620821e-01 -7.71797657e-01
7.98867702e-01 -9.64231491e-01 1.88114971e-01 5.75958341e-02
4.23067901e-03 -1.60718784e-01 3.18630546e-01 9.72143292e-01
-6.56184405e-02 -2.30882809e-01 9.74698126e-01 -7.36927986e-01
-6.98670387e-01 3.54777098e-01 -1.40083939e-01 2.53824275e-02
9.19360101e-01 -6.50344074e-01 2.11801101e-02 -3.92996103e-01
-1.49443591e+00 -1.61714524e-01 9.07813191e-01 3.28308046e-01
1.00032640e+00 -1.18551707e+00 -5.13950944e-01 5.78491867e-01
2.65195638e-01 3.10324818e-01 2.08545610e-01 6.25061393e-01
-2.44971141e-01 1.48202732e-01 -3.63298774e-01 -1.05025303e+00
-8.69335890e-01 6.71652377e-01 3.21291894e-01 1.14791729e-01
-1.03931212e+00 1.00389290e+00 9.23947930e-01 -5.64172745e-01
3.67559008e-02 -4.11168694e-01 1.78064659e-01 -2.26723894e-01
1.20917305e-01 1.20856531e-01 3.77346613e-02 -6.50630593e-01
-2.23601758e-01 1.08592737e+00 3.79738621e-02 -1.47797629e-01
1.47049701e+00 -2.42752969e-01 -4.67967568e-03 7.51467288e-01
1.02206349e+00 3.50316614e-01 -1.71840596e+00 -3.42298895e-01
6.81315828e-03 -9.46120679e-01 -1.02051854e-01 -6.89588487e-01
-7.75190592e-01 1.16147375e+00 2.66316503e-01 1.16712190e-01
9.34901476e-01 5.80537856e-01 3.32346171e-01 4.18135226e-01
6.90290451e-01 -4.08784926e-01 3.74706924e-01 5.32855809e-01
9.91607368e-01 -1.19182956e+00 1.54737815e-01 -4.19427902e-01
-4.80322570e-01 1.30800819e+00 4.21124309e-01 -3.00034314e-01
7.65597284e-01 1.60573080e-01 -1.68948144e-01 -5.64462364e-01
-6.37702882e-01 -4.24500257e-01 4.69034255e-01 9.22741354e-01
-4.81546000e-02 -9.26301628e-02 5.00185728e-01 2.96961457e-01
1.62978843e-01 -3.25003177e-01 5.69941461e-01 9.32076931e-01
-6.03036404e-01 -1.04981291e+00 -3.76345038e-01 7.98341036e-02
4.36220244e-02 1.95205703e-01 -4.72110599e-01 1.00044620e+00
3.46044242e-01 5.35266519e-01 4.69681341e-03 8.17522556e-02
4.54947472e-01 1.83031574e-01 1.16866195e+00 -1.17614102e+00
-1.67320110e-02 3.91754299e-01 -3.69391710e-01 -3.92932951e-01
-5.93521237e-01 -9.75405872e-01 -1.11083436e+00 8.94241557e-02
-1.88213289e-01 -2.52761662e-01 3.32633704e-01 1.01401842e+00
2.90851116e-01 1.85465202e-01 4.09017384e-01 -1.21534371e+00
-3.84903580e-01 -5.52827716e-01 -7.21499205e-01 4.17124063e-01
5.72858572e-01 -1.02642286e+00 -4.80638891e-01 1.20695718e-01] | [8.501273155212402, -3.1302289962768555] |
09853cf0-a933-4102-a26d-b766d9922732 | gtlo-a-generalized-and-non-linear-multi | 2204.04988 | null | https://arxiv.org/abs/2204.04988v1 | https://arxiv.org/pdf/2204.04988v1.pdf | gTLO: A Generalized and Non-linear Multi-Objective Deep Reinforcement Learning Approach | In real-world decision optimization, often multiple competing objectives must be taken into account. Following classical reinforcement learning, these objectives have to be combined into a single reward function. In contrast, multi-objective reinforcement learning (MORL) methods learn from vectors of per-objective rewards instead. In the case of multi-policy MORL, sets of decision policies for various preferences regarding the conflicting objectives are optimized. This is especially important when target preferences are not known during training or when preferences change dynamically during application. While it is, in general, straightforward to extend a single-objective reinforcement learning method for MORL based on linear scalarization, solutions that are reachable by these methods are limited to convex regions of the Pareto front. Non-linear MORL methods like Thresholded Lexicographic Ordering (TLO) are designed to overcome this limitation. Generalized MORL methods utilize function approximation to generalize across objective preferences and thereby implicitly learn multiple policies in a data-efficient manner, even for complex decision problems with high-dimensional or continuous state spaces. In this work, we propose \textit{generalized Thresholded Lexicographic Ordering} (gTLO), a novel method that aims to combine non-linear MORL with the advantages of generalized MORL. We introduce a deep reinforcement learning realization of the algorithm and present promising results on a standard benchmark for non-linear MORL and a real-world application from the domain of manufacturing process control. | ['Johannes Dornheim'] | 2022-04-11 | null | null | null | null | ['multi-objective-reinforcement-learning', 'deep-sea-treasure-image-version'] | ['methodology', 'playing-games'] | [ 3.41962390e-02 -6.64780140e-02 -5.05946398e-01 -2.36777142e-01
-9.09294128e-01 -5.44239283e-01 2.86702663e-01 6.49202645e-01
-7.15146661e-01 1.27868998e+00 -1.55751914e-01 -1.53352201e-01
-8.05268645e-01 -7.41819143e-01 -6.97054148e-01 -8.65646482e-01
-1.46440327e-01 8.37240279e-01 -2.05109015e-01 -4.37853128e-01
3.39129359e-01 4.99927849e-01 -1.46207690e+00 1.04548223e-01
1.05219162e+00 1.07892942e+00 3.99456531e-01 3.92077714e-01
5.74116558e-02 4.61581022e-01 -6.92018330e-01 1.31033897e-01
2.96063304e-01 -2.10210741e-01 -6.11921608e-01 1.59888819e-01
-3.02623883e-02 -1.53649986e-01 3.91359448e-01 8.63280833e-01
6.13535047e-01 5.92615664e-01 7.11207390e-01 -1.25924098e+00
-3.64340425e-01 7.36100733e-01 -4.60267961e-01 -2.21277177e-01
1.73960507e-01 3.01168501e-01 1.14421046e+00 -3.85931313e-01
3.07711899e-01 1.38029814e+00 9.84968096e-02 4.34147000e-01
-1.52273953e+00 -6.31268248e-02 6.34695470e-01 2.07310781e-01
-8.26007307e-01 7.40848482e-02 6.48836911e-01 -2.69374639e-01
8.39310110e-01 1.78536102e-01 6.03189647e-01 9.56267893e-01
7.28892982e-01 9.48000252e-01 1.51874757e+00 -2.12210804e-01
7.35989571e-01 -2.65762792e-03 -1.92311764e-01 4.34942186e-01
2.65469968e-01 4.10776585e-01 -2.53538132e-01 -1.12183996e-01
4.56443429e-01 -5.98286241e-02 7.40118176e-02 -8.75544190e-01
-1.19431365e+00 1.09425080e+00 2.30750740e-01 2.07727447e-01
-7.30400085e-01 2.42014199e-01 4.68233943e-01 6.83497965e-01
1.78488538e-01 1.00981927e+00 -6.46005511e-01 -8.78723040e-02
-6.55829966e-01 4.97142434e-01 6.32287621e-01 5.77020407e-01
8.83901834e-01 4.80607152e-01 -6.59741461e-01 6.88794017e-01
1.33809373e-01 4.25265402e-01 3.92328978e-01 -1.01505983e+00
2.68393368e-01 4.46894646e-01 5.78719854e-01 -4.64064389e-01
-5.80234110e-01 -5.86455882e-01 -7.56564558e-01 9.66412246e-01
3.81668746e-01 -4.72186148e-01 -8.16729426e-01 1.67836916e+00
2.48778716e-01 -4.30397153e-01 3.91133100e-01 9.41256821e-01
-1.04623146e-01 5.46132326e-01 -8.38201959e-03 -6.42497182e-01
7.97580063e-01 -6.55726731e-01 -7.69777000e-01 -1.02364160e-01
4.17843431e-01 -4.88491684e-01 1.08626151e+00 6.90417707e-01
-1.06027305e+00 -4.31938708e-01 -1.12659264e+00 7.37180293e-01
-4.13831919e-01 1.01959556e-01 3.87869716e-01 5.05216300e-01
-9.09028709e-01 9.24889863e-01 -5.33419609e-01 1.84832156e-01
1.90365478e-01 8.04097414e-01 9.88828987e-02 5.64394146e-02
-1.25868738e+00 1.13636005e+00 7.24241972e-01 1.39628112e-01
-1.08928251e+00 -4.79635864e-01 -8.24872553e-01 2.56895363e-01
1.08906293e+00 -2.76462078e-01 1.33788550e+00 -1.11497545e+00
-2.05715895e+00 5.57880662e-02 4.49132979e-01 -4.26446080e-01
6.07859135e-01 -3.22022438e-01 -2.15758860e-01 -1.24001965e-01
-1.31290168e-01 4.15358275e-01 1.07677758e+00 -1.34890199e+00
-7.44281292e-01 -8.65179077e-02 4.75287914e-01 3.07236552e-01
-1.88567698e-01 -3.36480319e-01 3.53772104e-01 -4.26130444e-01
-7.01869249e-01 -7.19415128e-01 -7.42162466e-01 -5.86561322e-01
-2.89114207e-01 -4.66467440e-01 7.03837156e-01 -1.66932240e-01
1.10533035e+00 -1.67691183e+00 6.78813040e-01 2.79864669e-01
-1.93954244e-01 2.77450413e-01 -4.32921678e-01 5.67831814e-01
2.30922773e-01 -1.20033346e-01 -3.44995499e-01 -1.15497537e-01
4.56488758e-01 5.23194313e-01 5.59083074e-02 3.94085258e-01
4.43986148e-01 7.84695625e-01 -1.21111786e+00 -3.18932742e-01
4.34179932e-01 -7.51613975e-02 -6.10429108e-01 6.71647638e-02
-7.91144669e-01 5.54688454e-01 -7.34443307e-01 6.20644510e-01
3.00187558e-01 6.68189079e-02 4.41730618e-01 1.45069599e-01
-3.50016087e-01 -1.58779591e-01 -1.51019859e+00 1.36112642e+00
-7.98573434e-01 4.28838953e-02 1.40832931e-01 -1.52189302e+00
1.00102675e+00 3.01647514e-01 1.00293970e+00 -6.54665351e-01
2.57325381e-01 3.24329048e-01 4.10870165e-02 -1.70913056e-01
4.26098585e-01 -2.60880172e-01 -3.66168916e-01 2.87754953e-01
9.87963304e-02 -2.94413924e-01 6.44485116e-01 -6.11670613e-01
7.96343267e-01 3.04641932e-01 5.72521806e-01 -5.64678967e-01
6.50942981e-01 -9.69174355e-02 9.17520106e-01 8.52860510e-01
-6.15061522e-02 -6.18473347e-03 7.88448870e-01 -2.38372087e-01
-7.66353011e-01 -9.26522791e-01 1.66694358e-01 9.66739535e-01
2.00946406e-01 -3.97643596e-02 -3.52034241e-01 -9.50896144e-01
3.51179004e-01 9.38681901e-01 -3.98753077e-01 -1.90735042e-01
-6.43893003e-01 -7.40215600e-01 -1.34890959e-01 2.55308986e-01
1.98793337e-01 -1.24169588e+00 -1.07866764e+00 8.93450737e-01
4.21837747e-01 -8.08578134e-01 -3.52172107e-01 6.42505586e-01
-6.98527813e-01 -9.57244456e-01 -7.53437042e-01 -4.24826413e-01
3.94002795e-01 -4.65710849e-01 1.03277016e+00 -5.30712366e-01
-1.03004903e-01 5.63737452e-01 -2.10270703e-01 -4.72570628e-01
-4.88975108e-01 9.64154154e-02 2.89806843e-01 2.26040259e-01
-1.10983312e-01 -2.93361425e-01 -4.64710504e-01 3.60438079e-01
-9.91990387e-01 -4.14982378e-01 8.79501760e-01 1.22776151e+00
8.38876963e-01 1.42587364e-01 1.11730337e+00 -6.34509921e-01
9.00205910e-01 -2.87715971e-01 -1.19561601e+00 4.98117447e-01
-8.36901367e-01 6.62309289e-01 1.20652616e+00 -7.60505795e-01
-8.57182026e-01 5.84101751e-02 5.49246743e-02 -5.10334134e-01
-8.11322778e-02 7.24453688e-01 -1.85329169e-01 9.98731554e-02
2.75469273e-01 5.81379533e-02 1.58298045e-01 -1.98992327e-01
1.95086539e-01 2.89710224e-01 -2.20797881e-02 -1.07203257e+00
4.74174857e-01 -7.69791380e-02 2.71238625e-01 -4.83555913e-01
-6.68670654e-01 -1.34385571e-01 -3.31278920e-01 -4.81569767e-01
7.20731974e-01 -3.24789196e-01 -1.18145132e+00 1.65442646e-01
-7.67195582e-01 -6.64958715e-01 -8.24487627e-01 6.71182871e-01
-1.28027403e+00 -3.41646075e-02 -9.18076336e-02 -1.03145194e+00
-3.88532691e-02 -1.52973187e+00 7.68235922e-01 3.43806028e-01
3.98456380e-02 -9.58222270e-01 8.81270170e-02 -1.84016123e-01
4.22385335e-01 5.73468149e-01 1.12773383e+00 -4.89865631e-01
-3.84080350e-01 1.82513565e-01 4.84097272e-01 5.70243359e-01
3.82249475e-01 -1.82026312e-01 -3.36647272e-01 -8.00007164e-01
-3.68456319e-02 -7.03368843e-01 5.50759256e-01 5.98628342e-01
1.11981249e+00 -3.52651834e-01 8.23675767e-02 4.09223400e-02
1.74869478e+00 6.79426312e-01 1.32722765e-01 4.68409657e-01
2.27072895e-01 5.65203249e-01 1.20976114e+00 8.40286314e-01
3.73221114e-02 7.19731987e-01 9.15481329e-01 1.14176661e-01
6.21445656e-01 2.05138132e-01 7.10522354e-01 3.51544648e-01
-1.81033775e-01 -4.09843951e-01 -6.00878298e-01 4.20785546e-01
-2.16072345e+00 -8.65298808e-01 4.60704654e-01 2.44269562e+00
7.02925146e-01 3.01548362e-01 3.05913359e-01 1.32948011e-01
6.43177330e-01 1.70005873e-01 -1.14334917e+00 -9.86522913e-01
-1.30153388e-01 2.43481904e-01 7.17024684e-01 6.20103419e-01
-1.08648765e+00 5.86255550e-01 5.52197456e+00 8.78639936e-01
-1.29935884e+00 -1.02378748e-01 5.10828257e-01 -3.16213876e-01
-2.08483219e-01 -2.71477103e-01 -7.63650358e-01 2.44380951e-01
7.82069862e-01 -3.26331854e-01 7.12420464e-01 7.88672864e-01
4.89583969e-01 -2.15585649e-01 -1.10983336e+00 7.47767866e-01
-4.92656350e-01 -1.16222084e+00 -3.30863267e-01 1.82123378e-01
9.91409004e-01 -2.17495978e-01 2.86211520e-01 6.84298754e-01
5.83168268e-01 -9.91641581e-01 6.16999388e-01 4.77241635e-01
4.91818547e-01 -1.24938285e+00 5.55499375e-01 3.28118742e-01
-9.83487666e-01 -7.03452349e-01 -2.98650503e-01 1.44556358e-01
2.32126877e-01 5.07656634e-01 -5.13751090e-01 9.36640084e-01
2.33491212e-01 6.91941023e-01 6.28371909e-02 8.62550855e-01
-6.79637566e-02 1.98976502e-01 -2.25276783e-01 -4.98936653e-01
8.34947348e-01 -3.38431239e-01 7.92717218e-01 7.21078694e-01
4.07935292e-01 -4.49342787e-01 6.81003273e-01 7.14177251e-01
2.89152563e-01 1.32047191e-01 -4.77002203e-01 -1.89166158e-01
-8.29295218e-02 1.12781417e+00 -4.96729314e-01 1.32485367e-02
-2.36163199e-01 5.66960454e-01 3.46331805e-01 3.28841537e-01
-8.36604416e-01 -1.26372874e-01 9.20193970e-01 -2.34713629e-01
5.38895130e-01 -3.76201034e-01 1.51958808e-01 -7.58051157e-01
-3.37690711e-02 -1.09395516e+00 4.02905673e-01 -5.06093167e-02
-1.35170317e+00 3.50730419e-01 1.95978820e-01 -1.41025686e+00
-6.36376262e-01 -7.99475133e-01 -3.36411893e-01 5.97398102e-01
-1.83828676e+00 -6.51699185e-01 5.16536951e-01 6.08085096e-01
8.27620089e-01 -3.47932011e-01 6.98180795e-01 -5.47406496e-03
-5.32724380e-01 3.09270054e-01 7.34902740e-01 -6.66964829e-01
7.10899770e-01 -1.72091842e+00 -5.07727623e-01 2.62450337e-01
-3.39358151e-01 -2.38660350e-02 7.88946211e-01 -3.82113129e-01
-1.63202989e+00 -1.08790803e+00 1.88539594e-01 3.66161346e-01
6.61012650e-01 -9.72942263e-02 -5.77958345e-01 1.55415252e-01
3.80778998e-01 -1.18354291e-01 2.88579643e-01 2.92797945e-02
3.66223991e-01 -4.72729474e-01 -1.16358268e+00 6.02691114e-01
3.65917414e-01 7.29311258e-02 -2.90064812e-01 2.59987742e-01
4.21114087e-01 -2.52016276e-01 -1.17726731e+00 6.44437075e-01
2.76830137e-01 -7.12572813e-01 8.04743409e-01 -7.69494712e-01
3.64008784e-01 -3.94067734e-01 -1.43276185e-01 -2.03761005e+00
-2.00141147e-01 -8.81894290e-01 -2.41237268e-01 8.17489862e-01
2.13593602e-01 -7.16237485e-01 4.62409645e-01 1.00601546e-01
-2.21746385e-01 -1.28189445e+00 -1.18086386e+00 -1.24637091e+00
2.41149113e-01 7.80025870e-02 6.65003657e-01 6.55608892e-01
-9.49194059e-02 2.15403721e-01 -5.84688902e-01 8.63785073e-02
6.85629964e-01 4.24594313e-01 2.60050178e-01 -1.04700446e+00
-6.84843302e-01 -6.79135859e-01 -8.50245729e-03 -7.09996462e-01
2.98843443e-01 -6.44936800e-01 3.81215721e-01 -1.54990172e+00
-4.15141702e-01 -5.94037831e-01 -9.04523432e-01 4.54884171e-01
2.88044661e-03 -5.60312986e-01 3.59289050e-01 -3.84775370e-01
-7.75479972e-01 1.00276160e+00 1.66420913e+00 -5.01579046e-01
-4.63663638e-01 3.31571162e-01 -4.50225204e-01 3.91993672e-01
1.13014174e+00 -3.46196502e-01 -6.75238252e-01 -7.19427168e-02
1.46308497e-01 6.07241929e-01 -8.20335522e-02 -9.44689453e-01
-1.62441567e-01 -9.25685465e-01 1.27228484e-01 -5.88635385e-01
3.97870064e-01 -8.71033072e-01 -1.48798779e-01 6.38593793e-01
-4.60852832e-01 2.26628482e-01 1.51498392e-01 6.02881014e-01
-2.92387754e-01 -3.47890228e-01 9.48985636e-01 -1.37547582e-01
-6.89847231e-01 2.81333596e-01 -7.07577944e-01 -8.02199543e-02
1.29938829e+00 9.31859016e-02 2.66261697e-02 -1.51836216e-01
-9.51584697e-01 7.82510221e-01 4.17576730e-02 4.74295020e-01
6.20895445e-01 -1.28349745e+00 -7.61226892e-01 -2.46731699e-01
-9.54679921e-02 -5.78938872e-02 8.27974603e-02 7.59431660e-01
1.65163592e-01 3.95786494e-01 -5.65672457e-01 -4.60969299e-01
-9.62298691e-01 7.89834261e-01 5.02734303e-01 -9.80350256e-01
-1.30617201e-01 1.11705370e-01 -6.42569214e-02 -4.08294559e-01
6.28100261e-02 -4.82075363e-01 -3.61625552e-01 3.91479850e-01
1.31441295e-01 4.24625009e-01 7.08694160e-02 -6.43175021e-02
-1.31579533e-01 3.66248310e-01 3.43226157e-02 -2.13036597e-01
1.41974771e+00 5.17363474e-02 2.52600044e-01 6.15475178e-01
8.35637212e-01 -2.22970217e-01 -1.59103537e+00 -1.93868149e-02
2.35960379e-01 -1.15016885e-01 3.52364331e-02 -9.05183256e-01
-1.01614654e+00 6.27301693e-01 6.44786596e-01 3.16771299e-01
1.24309516e+00 -5.06228566e-01 3.18744868e-01 7.16030836e-01
5.18889308e-01 -1.64592910e+00 3.18789244e-01 7.73715675e-01
1.11657870e+00 -1.18414629e+00 -1.68011606e-01 2.59882033e-01
-7.79981256e-01 1.34561563e+00 6.83127940e-01 -2.56376892e-01
3.84572178e-01 2.62088984e-01 -7.01357499e-02 2.56543130e-01
-8.29585791e-01 -5.36930621e-01 9.91519839e-02 4.45853114e-01
-3.53565402e-02 1.85099691e-01 -5.81701040e-01 2.15457067e-01
3.59223306e-01 -9.97656956e-02 4.13192540e-01 1.31187487e+00
-4.17820126e-01 -1.65389836e+00 -4.88403797e-01 4.83986974e-01
-3.00103724e-01 3.57885957e-01 1.25653997e-01 8.34299326e-01
-4.12316136e-02 9.62868154e-01 -2.10262880e-01 -9.81783569e-02
4.82270867e-01 -1.14206895e-01 6.98276103e-01 -5.47300637e-01
-8.86093915e-01 2.06314623e-01 8.33353102e-02 -6.76285028e-01
-4.23344404e-01 -7.52633810e-01 -1.19359696e+00 1.23742133e-01
-9.97235104e-02 1.84540465e-01 6.02467775e-01 1.10315895e+00
1.77193984e-01 1.01551926e+00 9.29838419e-01 -8.88834953e-01
-1.18080187e+00 -5.48680842e-01 -6.66898549e-01 4.87624556e-02
4.91999090e-01 -9.77781713e-01 4.24569398e-02 -5.89082003e-01] | [4.39596700668335, 2.42025089263916] |
d62e2227-cbad-48c4-b686-a5ab8baea0e7 | machine-translation-of-low-resource-indo | 2108.03739 | null | https://arxiv.org/abs/2108.03739v2 | https://arxiv.org/pdf/2108.03739v2.pdf | Machine Translation of Low-Resource Indo-European Languages | In this work, we investigate methods for the challenging task of translating between low-resource language pairs that exhibit some level of similarity. In particular, we consider the utility of transfer learning for translating between several Indo-European low-resource languages from the Germanic and Romance language families. In particular, we build two main classes of transfer-based systems to study how relatedness can benefit the translation performance. The primary system fine-tunes a model pre-trained on a related language pair and the contrastive system fine-tunes one pre-trained on an unrelated language pair. Our experiments show that although relatedness is not necessary for transfer learning to work, it does benefit model performance. | ['Muhammad Abdul-Mageed', 'Wei-Rui Chen'] | 2021-08-08 | null | https://aclanthology.org/2021.wmt-1.41 | https://aclanthology.org/2021.wmt-1.41.pdf | wmt-emnlp-2021-11 | ['low-resource-neural-machine-translation'] | ['natural-language-processing'] | [ 5.99834733e-02 -4.81756479e-02 -1.02970719e-01 -4.36495155e-01
-1.07568705e+00 -7.62822270e-01 7.92731047e-01 -1.17469430e-01
-5.04970431e-01 1.09506226e+00 3.45064551e-01 -6.45117640e-01
1.45326123e-01 -7.41778135e-01 -7.28024065e-01 -2.75909841e-01
2.63623267e-01 8.49256277e-01 7.36551285e-02 -9.05174971e-01
-1.55852839e-01 3.07151914e-01 -7.75117159e-01 5.21774411e-01
1.18485844e+00 1.56808823e-01 3.07560205e-01 4.30749714e-01
-5.72330505e-03 2.83346057e-01 -6.26948059e-01 -7.69789636e-01
4.28480089e-01 -8.82365346e-01 -1.22996736e+00 -5.27934253e-01
2.85438389e-01 4.27801907e-02 1.16889477e-02 8.52899253e-01
5.82381248e-01 -3.88533734e-02 9.35256958e-01 -9.73321855e-01
-9.77059841e-01 9.95500207e-01 -8.17936361e-02 4.01005775e-01
4.80682373e-01 -5.49350344e-02 1.02737868e+00 -8.40615511e-01
7.54179955e-01 1.35501814e+00 7.06894875e-01 6.42440498e-01
-1.11963773e+00 -6.43362999e-01 -1.43673345e-01 -4.07678783e-02
-1.39960837e+00 -5.24989843e-01 5.45922637e-01 -3.66068631e-01
1.15758669e+00 5.98031096e-02 4.94218439e-01 9.12514329e-01
2.86779553e-01 4.73942906e-01 1.38297093e+00 -7.53261089e-01
-2.78107464e-01 3.35499913e-01 -3.50216419e-01 4.49534863e-01
2.59691128e-03 2.57074594e-01 -4.48153287e-01 -1.06442645e-01
5.46578526e-01 -6.39842749e-01 -1.33106008e-01 -5.23775108e-02
-1.65851128e+00 8.69497061e-01 4.62157220e-01 1.02523828e+00
-8.55228826e-02 -2.22698838e-01 4.48613375e-01 1.05116129e+00
6.87892795e-01 8.88959825e-01 -7.92370081e-01 -2.38594383e-01
-8.41649532e-01 -8.52624178e-02 8.40733111e-01 1.13166165e+00
7.80310631e-01 -1.40245885e-01 -1.76821947e-01 8.89993191e-01
1.02920584e-01 4.07976687e-01 7.35974133e-01 -3.89838636e-01
8.25987577e-01 2.07192481e-01 -5.65861836e-02 -3.96078914e-01
-9.34494063e-02 -2.52047271e-01 -5.28217673e-01 -3.47875774e-01
5.78588963e-01 -3.42808425e-01 -4.90532249e-01 1.96530342e+00
4.38335724e-02 -2.26826891e-01 5.67121327e-01 5.53046763e-01
7.45118618e-01 6.68869257e-01 9.77858976e-02 -8.71268958e-02
1.20213366e+00 -1.36623931e+00 -2.70042270e-01 -2.46183529e-01
9.85830247e-01 -1.49102259e+00 1.34581041e+00 -3.33788127e-01
-1.39328051e+00 -6.82566822e-01 -9.02974427e-01 -1.90482035e-01
-4.95740086e-01 1.03368245e-01 2.87842631e-01 4.80124950e-01
-1.28573847e+00 8.23528588e-01 -5.09312749e-01 -9.87426639e-01
-1.65287897e-01 2.43032724e-01 -5.74537516e-01 -1.37045801e-01
-1.53705788e+00 1.45187509e+00 5.21831155e-01 -3.78738642e-01
-4.76379007e-01 -6.29594982e-01 -5.58180690e-01 -7.44656026e-02
-4.06508833e-01 -9.54178631e-01 1.35423565e+00 -1.51015103e+00
-1.85216975e+00 1.34548366e+00 1.05537139e-01 -6.50514439e-02
7.24059403e-01 -4.42273840e-02 -4.59654331e-01 -2.50071377e-01
2.21610308e-01 7.20068753e-01 5.90667129e-01 -9.83022630e-01
-5.34824312e-01 6.22686511e-03 1.85191438e-01 5.03227651e-01
-4.08464909e-01 6.62033379e-01 -2.71095008e-01 -8.70476186e-01
-3.62849385e-01 -1.09672737e+00 1.11230031e-01 -5.16517580e-01
-5.99758960e-02 -2.40315273e-01 2.07266614e-01 -8.05543602e-01
7.46427715e-01 -1.97012198e+00 3.83573383e-01 -5.94096109e-02
-4.39758092e-01 3.73896688e-01 -7.78253376e-01 8.25690567e-01
-7.81023279e-02 2.30695724e-01 -8.76110047e-02 -8.93134400e-02
-2.01004386e-01 1.38664767e-01 -2.29491755e-01 2.27685124e-01
5.64607084e-01 1.19832504e+00 -1.14343059e+00 -4.00874674e-01
-3.74922693e-01 3.46856385e-01 -5.64837456e-01 4.40629661e-01
8.53746533e-02 9.13162589e-01 -3.20067555e-01 4.10393894e-01
2.99113661e-01 9.59010273e-02 3.88422370e-01 -4.05746512e-02
-2.55782902e-02 9.67370391e-01 -3.35881710e-01 1.81144476e+00
-9.14317131e-01 5.14508069e-01 -1.33188635e-01 -9.19710219e-01
9.97291803e-01 4.42333132e-01 1.59478068e-01 -7.72595048e-01
2.22303852e-01 7.44562626e-01 6.61882281e-01 -2.00603217e-01
6.85525477e-01 -6.03182733e-01 -1.84401572e-01 7.97234535e-01
2.05391273e-01 -6.77852392e-01 2.81583905e-01 -2.58124202e-01
9.96495187e-01 3.83333683e-01 4.90995884e-01 -7.37374723e-01
7.38134861e-01 1.95904821e-01 4.85261112e-01 2.82765538e-01
-1.54567093e-01 4.76851761e-01 9.06612650e-02 -1.38307810e-01
-1.28265357e+00 -1.23473382e+00 9.47164446e-02 1.57738554e+00
-8.27635750e-02 -4.14172411e-01 -7.99504399e-01 -8.24531317e-01
-2.28413880e-01 7.13217020e-01 -4.24013108e-01 -3.13491374e-01
-9.24442708e-01 -4.30742949e-01 8.54370117e-01 5.17322838e-01
3.18819642e-01 -1.05093849e+00 -9.70016718e-02 5.89457043e-02
-4.95903075e-01 -1.03536677e+00 -7.53478706e-01 4.64966148e-02
-8.42607617e-01 -5.61663866e-01 -1.07567656e+00 -1.39498842e+00
5.52274644e-01 2.43587792e-01 1.81341386e+00 1.68217197e-01
3.98169041e-01 2.76474208e-01 -6.06153846e-01 -3.53119612e-01
-1.06073201e+00 8.46993625e-01 1.68743700e-01 -4.80636597e-01
5.40545225e-01 -4.73740727e-01 -6.86446158e-03 5.43789029e-01
-5.54064870e-01 1.28665520e-02 5.73308766e-01 8.37509453e-01
2.61103809e-01 -6.29511595e-01 6.34799957e-01 -8.09375823e-01
9.01113033e-01 -5.19076943e-01 -2.84257770e-01 6.55712605e-01
-1.92302912e-01 9.93277803e-02 8.72170806e-01 -5.39021432e-01
-9.25702751e-01 -1.13541842e-01 -1.27427980e-01 1.00317765e-02
1.73286065e-01 5.02045453e-01 -1.69248804e-01 -1.60599779e-02
8.24186623e-01 3.97204049e-02 -2.73470312e-01 -4.52009529e-01
6.46211445e-01 8.97152901e-01 3.42930704e-01 -1.10102296e+00
1.09181106e+00 -1.60545588e-01 -5.03674328e-01 -5.39153636e-01
-4.07028109e-01 -2.18450949e-01 -1.03397751e+00 2.26955906e-01
8.04150999e-01 -1.00422096e+00 8.43932033e-02 7.47367442e-02
-1.36237729e+00 -6.94172740e-01 -3.92710656e-01 6.38673723e-01
-8.30765724e-01 -6.36349097e-02 -7.02110708e-01 1.38590753e-01
-4.16212857e-01 -1.20785737e+00 1.02003920e+00 -7.64321014e-02
-4.72081661e-01 -1.36432266e+00 5.42994916e-01 1.24988556e-01
6.52578712e-01 -2.51935244e-01 1.23911464e+00 -1.02651632e+00
-9.94082093e-02 2.84017086e-01 -1.19932041e-01 3.65013838e-01
5.93661964e-01 -2.60690212e-01 -5.20106614e-01 -6.78844988e-01
-2.41969660e-01 -5.25991619e-01 3.53495091e-01 -2.51916379e-01
9.42079201e-02 -2.41865311e-02 -2.07122147e-01 5.45808971e-01
1.06169713e+00 7.45331421e-02 5.32073438e-01 4.12773550e-01
4.69084591e-01 7.66791344e-01 6.76854908e-01 -3.83814156e-01
5.27405977e-01 8.26581657e-01 -4.28958386e-01 -4.19938147e-01
-4.25067902e-01 -3.07374626e-01 7.24814415e-01 1.68001688e+00
-2.95064777e-01 -9.09985304e-02 -1.13105857e+00 6.87627017e-01
-1.61730957e+00 -5.16382813e-01 2.27458850e-01 2.08727288e+00
1.42306411e+00 -2.45911956e-01 2.47092739e-01 -4.44587320e-01
7.60066867e-01 -2.04935595e-01 -7.85202682e-02 -9.68298316e-01
-1.37061507e-01 5.62696576e-01 1.63635269e-01 6.29271150e-01
-7.87305534e-01 1.66630578e+00 7.14064360e+00 7.22218215e-01
-1.23495936e+00 4.07222241e-01 4.13148731e-01 1.70091569e-01
-4.64904785e-01 1.27069846e-01 -7.14914382e-01 2.14274004e-01
1.21710622e+00 -7.32154965e-01 5.62178016e-01 4.42478329e-01
-7.63434023e-02 3.98773193e-01 -1.69053471e+00 5.28172135e-01
1.26174778e-01 -8.58765960e-01 1.82582453e-01 -1.39986828e-01
9.83026147e-01 2.63072044e-01 -1.53093770e-01 5.86054862e-01
6.08953059e-01 -1.03707838e+00 6.07733428e-01 1.02960028e-01
9.25056040e-01 -7.03560412e-01 7.01131642e-01 2.16094121e-01
-9.54986989e-01 5.38260877e-01 -4.54051942e-01 -2.61039078e-01
1.03621468e-01 1.79060459e-01 -1.01412499e+00 7.85096765e-01
5.22534013e-01 6.55381203e-01 -5.48975468e-01 6.36933267e-01
-5.40949404e-01 5.93459785e-01 -2.02150382e-02 -5.45969175e-04
2.39895135e-01 -5.37181556e-01 4.01579767e-01 1.48274684e+00
5.68499207e-01 -2.28609845e-01 2.29399487e-01 6.23679340e-01
-2.95486450e-01 7.29241312e-01 -8.85346174e-01 -1.04957804e-01
2.74985999e-01 1.07772493e+00 -4.46966976e-01 -4.88982767e-01
-5.99368155e-01 1.24341440e+00 7.50915766e-01 7.29883835e-02
-6.54964864e-01 -4.21228677e-01 6.74066722e-01 1.45677164e-01
1.88447133e-01 -3.44201356e-01 -5.70318475e-02 -1.33686197e+00
-1.50839999e-01 -1.21870804e+00 6.41005635e-02 -6.50101602e-01
-1.70589745e+00 9.80308831e-01 -5.65563254e-02 -1.40680969e+00
-7.32241035e-01 -4.23619986e-01 -6.13870680e-01 1.43504131e+00
-1.66500163e+00 -1.58539665e+00 3.00049871e-01 7.48427391e-01
4.73215163e-01 -5.42733312e-01 1.07680488e+00 3.75591725e-01
-5.60921244e-02 9.40664351e-01 3.04041982e-01 3.61527264e-01
1.29634070e+00 -9.77863014e-01 9.34632003e-01 6.80397272e-01
2.66999096e-01 8.02577198e-01 4.78990614e-01 -5.80116689e-01
-1.05537355e+00 -1.20924139e+00 1.65293574e+00 -4.36640501e-01
7.28552222e-01 -3.34847093e-01 -8.68984878e-01 8.49989533e-01
7.28480518e-01 -1.78836465e-01 8.75859261e-01 3.96019429e-01
-5.49070716e-01 2.51811832e-01 -1.08802617e+00 7.31662214e-01
1.13733268e+00 -9.40839946e-01 -1.19705820e+00 5.01627028e-01
7.69587576e-01 -2.81387627e-01 -1.16360378e+00 2.69249052e-01
4.42039758e-01 -4.82429981e-01 7.78382421e-01 -1.12697148e+00
7.01316833e-01 1.11395326e-02 -2.18473658e-01 -2.11749649e+00
-4.62488711e-01 -7.02150643e-01 7.13014960e-01 1.53592515e+00
7.30508149e-01 -8.34737122e-01 1.93446651e-01 1.26681983e-01
-5.66279218e-02 -3.79788548e-01 -8.63286734e-01 -1.40403926e+00
1.21992612e+00 1.80010960e-01 6.93508387e-01 1.41553867e+00
2.19082177e-01 9.83622074e-01 -6.25132620e-02 -2.90179402e-01
-2.07645312e-01 1.77709863e-01 9.05824304e-01 -1.03733492e+00
-5.15105188e-01 -4.32723850e-01 -1.75655097e-01 -7.39915550e-01
6.74944282e-01 -1.54824710e+00 1.42141476e-01 -1.29386294e+00
1.61014467e-01 -7.19543576e-01 -1.75329626e-01 5.69522738e-01
-4.44308698e-01 4.72475499e-01 4.32127059e-01 4.62731659e-01
-2.08464205e-01 4.42792416e-01 1.21451569e+00 -1.80458561e-01
-4.14058089e-01 -8.62225424e-03 -7.14654803e-01 5.35802960e-01
1.00153029e+00 -6.00116491e-01 -2.89809585e-01 -9.36185896e-01
2.60663986e-01 -2.37241060e-01 -4.40945446e-01 -5.67663491e-01
-1.68873802e-01 -1.83393538e-01 -3.86350676e-02 1.60904810e-01
2.00455599e-02 -4.63652909e-01 -7.28916749e-02 5.69480658e-01
-2.72589624e-01 7.22525358e-01 3.94923389e-01 -2.16856882e-01
-4.32150990e-01 -2.39266962e-01 8.09821188e-01 -3.13987792e-01
-2.33284980e-01 -1.16731562e-01 -4.46129233e-01 4.34482098e-01
7.62974739e-01 1.78478375e-01 -2.16663957e-01 -2.73872793e-01
-2.91342050e-01 -1.56122804e-01 7.92588592e-01 8.20121527e-01
-1.98411266e-03 -1.66648281e+00 -1.32466388e+00 1.14397332e-01
3.47718477e-01 -7.08028972e-01 -6.74778521e-01 6.81598246e-01
-4.93759871e-01 5.43426514e-01 -6.76049054e-01 -3.90909940e-01
-1.21836162e+00 4.46359098e-01 2.94642419e-01 -5.19520462e-01
-8.11350569e-02 6.52316630e-01 1.07534789e-01 -1.13165498e+00
-5.12931585e-01 -1.88708290e-01 -5.37689915e-03 -4.71482240e-02
1.99917004e-01 1.14039406e-01 2.93950498e-01 -1.05960786e+00
-4.07789260e-01 7.84981132e-01 -2.24943295e-01 -4.74157333e-01
1.17411613e+00 -7.61691853e-02 -3.73980999e-01 4.65914369e-01
1.21948564e+00 2.83959240e-01 -5.00853598e-01 -4.95012730e-01
7.68609792e-02 -2.36692160e-01 -3.73278260e-01 -9.04165447e-01
-7.21167207e-01 8.40617836e-01 2.32425958e-01 -1.32796317e-01
1.08659673e+00 1.06745912e-02 8.30520570e-01 5.72973847e-01
5.22994876e-01 -1.10281432e+00 -1.43276781e-01 1.21432364e+00
9.35424387e-01 -1.19206524e+00 -2.10157394e-01 -4.42479938e-01
-5.53138852e-01 9.44491148e-01 3.51751208e-01 -1.47021055e-01
3.86740029e-01 1.30983219e-01 3.94757181e-01 3.78777474e-01
-6.52931571e-01 -3.07245821e-01 4.94533092e-01 8.79016221e-01
1.14185393e+00 1.96572661e-01 -8.22939456e-01 4.69001114e-01
-7.36143768e-01 -9.50974692e-03 2.27792740e-01 8.99934769e-01
2.93254368e-02 -1.67895555e+00 -3.09426665e-01 -2.77440846e-02
-3.14971864e-01 -5.35917997e-01 -9.27998006e-01 1.00902343e+00
8.39396715e-02 9.00265396e-01 1.63805932e-01 -3.58281314e-01
3.78478169e-01 1.67243719e-01 9.69546735e-01 -9.46309209e-01
-1.23202693e+00 -1.09260999e-01 2.54726827e-01 -1.31159872e-01
-4.34443116e-01 -5.64122856e-01 -8.69444251e-01 -3.98462087e-01
-1.71964705e-01 3.45774859e-01 5.22320211e-01 1.12102842e+00
2.64358610e-01 3.03862393e-01 6.75857544e-01 -7.10052192e-01
-5.93676388e-01 -1.23283696e+00 -1.23325102e-01 4.81773913e-01
-7.70295113e-02 -2.17784420e-01 -5.72552830e-02 2.24284902e-01] | [11.438313484191895, 10.240743637084961] |
5f1e1cf5-523b-44bd-ac74-84a38310a918 | mixture-dense-regression-for-object-detection | 1912.00821 | null | https://arxiv.org/abs/1912.00821v2 | https://arxiv.org/pdf/1912.00821v2.pdf | Mixture Dense Regression for Object Detection and Human Pose Estimation | Mixture models are well-established learning approaches that, in computer vision, have mostly been applied to inverse or ill-defined problems. However, they are general-purpose divide-and-conquer techniques, splitting the input space into relatively homogeneous subsets in a data-driven manner. Not only ill-defined but also well-defined complex problems should benefit from them. To this end, we devise a framework for spatial regression using mixture density networks. We realize the framework for object detection and human pose estimation. For both tasks, a mixture model yields higher accuracy and divides the input space into interpretable modes. For object detection, mixture components focus on object scale, with the distribution of components closely following that of ground truth the object scale. This practically alleviates the need for multi-scale testing, providing a superior speed-accuracy trade-off. For human pose estimation, a mixture model divides the data based on viewpoint and uncertainty -- namely, front and back views, with back view imposing higher uncertainty. We conduct experiments on the MS COCO dataset and do not face any mode collapse. | ['Ali Varamesh', 'Tinne Tuytelaars'] | 2019-12-02 | mixture-dense-regression-for-object-detection-1 | http://openaccess.thecvf.com/content_CVPR_2020/html/Varamesh_Mixture_Dense_Regression_for_Object_Detection_and_Human_Pose_Estimation_CVPR_2020_paper.html | http://openaccess.thecvf.com/content_CVPR_2020/papers/Varamesh_Mixture_Dense_Regression_for_Object_Detection_and_Human_Pose_Estimation_CVPR_2020_paper.pdf | cvpr-2020-6 | ['dense-object-detection'] | ['computer-vision'] | [-1.48359880e-01 -1.19568668e-02 -1.89782605e-01 -4.19480294e-01
-6.98332071e-01 -4.03959394e-01 4.00585264e-01 -3.29132438e-01
-5.83773971e-01 4.29811746e-01 -1.81421146e-01 -6.96401000e-02
-2.24671215e-01 -5.56075275e-01 -5.61880648e-01 -9.17451084e-01
3.38880241e-01 8.80690932e-01 4.11457688e-01 1.08123660e-01
1.27176762e-01 4.39943343e-01 -1.56215668e+00 1.05658136e-01
8.56896460e-01 9.11158025e-01 2.77717561e-01 4.50713038e-01
1.65601164e-01 4.07196552e-01 -5.70649385e-01 -3.53117794e-01
2.96798110e-01 -1.60363346e-01 -3.07013780e-01 4.87979650e-01
3.61867398e-01 -2.13985637e-01 -1.12798102e-01 1.10508823e+00
4.58536625e-01 -4.14889213e-03 1.13768077e+00 -1.45326734e+00
-4.14701492e-01 2.07963273e-01 -1.10391235e+00 -1.45355701e-01
1.38993720e-02 2.84877121e-02 7.45534718e-01 -9.81792450e-01
3.09285790e-01 1.40630472e+00 5.84553599e-01 3.94616067e-01
-1.38195562e+00 -4.60786492e-01 4.12017256e-01 2.38256320e-01
-1.58716452e+00 -1.40347064e-01 8.75459373e-01 -7.61487603e-01
2.33394638e-01 3.81728649e-01 3.91499311e-01 9.17517245e-01
1.84557736e-01 9.50210214e-01 1.21358967e+00 -3.15764695e-01
3.36933285e-01 4.54483718e-01 -1.12927139e-01 4.40812618e-01
5.39398253e-01 -1.00798402e-02 -2.56608993e-01 -2.65421886e-02
1.06894755e+00 1.98927075e-01 -3.68195951e-01 -1.04729593e+00
-1.11695111e+00 7.15490282e-01 3.43026578e-01 1.34434417e-01
-2.18980446e-01 -1.44698516e-01 -6.12955093e-02 -1.38139576e-01
5.63234508e-01 1.88864931e-01 -3.93890828e-01 4.17189986e-01
-1.01619291e+00 2.82421380e-01 5.09813786e-01 9.74848986e-01
6.98529005e-01 -2.96842337e-01 -1.05912723e-01 7.60255396e-01
6.66454613e-01 5.40899217e-01 3.58348548e-01 -8.84284735e-01
3.30935448e-01 5.81982851e-01 2.51434714e-01 -1.02509439e+00
-5.13522029e-01 -5.77122033e-01 -1.03435683e+00 5.70185363e-01
8.15501451e-01 -1.34556055e-01 -8.97860229e-01 1.80193019e+00
6.39429152e-01 -2.06847414e-01 -2.86581665e-01 1.15877581e+00
5.30349433e-01 2.97149420e-01 -2.52697945e-01 -1.58091486e-01
1.50580478e+00 -9.69754934e-01 -5.67361593e-01 -4.43511337e-01
2.25823000e-01 -5.99634528e-01 1.02012432e+00 6.91509128e-01
-8.66617560e-01 -8.28484714e-01 -1.01676798e+00 1.17192782e-01
-1.42567798e-01 5.10289013e-01 3.02572876e-01 7.35035360e-01
-7.04159141e-01 4.90264952e-01 -8.32873166e-01 -4.40093391e-02
3.22849333e-01 2.96353728e-01 -3.57385129e-01 7.99264479e-03
-6.69789732e-01 1.12679434e+00 3.91994417e-01 2.64910758e-01
-5.17270446e-01 -4.22626376e-01 -8.30113351e-01 -2.23890617e-01
6.62905812e-01 -7.33389020e-01 1.05055559e+00 -7.05004573e-01
-1.38882291e+00 9.20581579e-01 8.66185203e-02 -1.50134578e-01
8.46768916e-01 -3.83918136e-01 -1.62226304e-01 -3.87074016e-02
1.97202995e-01 7.12476552e-01 1.23238099e+00 -1.39454496e+00
-6.70345545e-01 -8.83262873e-01 -1.73091665e-01 3.63386989e-01
-9.40296724e-02 -2.15193003e-01 -6.80569470e-01 -5.98804653e-01
6.37372315e-01 -1.14282215e+00 -4.49556798e-01 1.54249579e-01
-4.28862512e-01 -1.37862816e-01 7.10149288e-01 -5.30863166e-01
1.08999407e+00 -2.10508585e+00 4.76697832e-01 5.77038452e-02
4.70277488e-01 -1.73230410e-01 1.68266907e-01 -1.33661121e-01
-7.48362113e-03 -2.05501676e-01 -2.58634925e-01 -4.68553126e-01
1.87289938e-02 -3.25960778e-02 -2.47239284e-02 8.82694006e-01
1.18751027e-01 6.01460040e-01 -5.21962762e-01 -6.63833261e-01
3.03780109e-01 3.59182924e-01 -5.40389061e-01 3.23228598e-01
-1.83707297e-01 6.88288212e-01 -3.18808168e-01 6.09923601e-01
1.03786969e+00 -2.25036591e-01 3.40428650e-02 -3.30225945e-01
-3.78400786e-03 -3.16155821e-01 -1.63604009e+00 1.51289940e+00
-2.04581112e-01 4.16537195e-01 2.94892728e-01 -8.96572769e-01
8.21416497e-01 6.83503076e-02 4.25887436e-01 -1.32570013e-01
1.45992637e-01 2.18949124e-01 8.86204317e-02 -4.53834623e-01
2.67381251e-01 -2.85097748e-01 -5.07931411e-02 1.57070592e-01
6.65303916e-02 -2.86092997e-01 1.23034410e-01 -1.70949638e-01
4.90278006e-01 4.89060789e-01 3.50648552e-01 -2.67649323e-01
3.36537570e-01 -1.64511085e-01 4.83315468e-01 5.39694905e-01
-1.83184803e-01 1.11591899e+00 5.40010035e-01 -1.68610677e-01
-6.98137403e-01 -1.19002438e+00 -2.13155299e-01 1.00332093e+00
5.07864416e-01 -1.53656676e-01 -9.41613257e-01 -7.66062140e-01
-2.25463007e-02 5.54875851e-01 -6.29210651e-01 -7.78063983e-02
-4.98275310e-01 -8.65942657e-01 1.40020609e-01 6.79012179e-01
4.54355270e-01 -7.95167327e-01 -7.95963943e-01 -7.94503912e-02
-3.80830944e-01 -1.02849627e+00 -5.80506206e-01 2.11308792e-01
-8.19101512e-01 -1.22935009e+00 -1.18540132e+00 -5.06627142e-01
7.31863737e-01 3.47066253e-01 1.04340935e+00 -5.75488031e-01
-3.33213776e-01 3.99364769e-01 -6.87754080e-02 -4.87936407e-01
-1.25433654e-01 -2.17382789e-01 3.10063779e-01 3.01219702e-01
2.81705678e-01 -5.19601285e-01 -6.11276150e-01 7.74974167e-01
-7.58190155e-01 -1.54124340e-02 8.10360134e-01 6.29491746e-01
7.21628726e-01 1.64085716e-01 1.07375108e-01 -4.99592841e-01
3.12473565e-01 -2.88229316e-01 -6.96394861e-01 2.75727689e-01
-5.10113299e-01 3.24722864e-02 1.66588247e-01 -9.42372322e-01
-1.00063562e+00 3.56732219e-01 2.47583538e-01 -6.67860925e-01
-3.22080374e-01 2.15130389e-01 -5.37926793e-01 1.67974636e-01
7.66124070e-01 5.89530654e-02 1.79493099e-01 -6.17420912e-01
5.41552126e-01 6.48669600e-01 4.74616975e-01 -5.13882101e-01
7.78059781e-01 6.32967591e-01 -1.03322705e-02 -9.10334468e-01
-8.11240375e-01 -6.09374464e-01 -8.90339553e-01 -2.71941721e-01
1.04235935e+00 -8.38963270e-01 -5.80454826e-01 4.29102868e-01
-1.06438565e+00 1.68621272e-03 -2.79746115e-01 6.84543967e-01
-6.96902573e-01 3.71358573e-01 -2.72065729e-01 -1.00314343e+00
2.17781752e-01 -1.18345129e+00 1.25255489e+00 2.81769723e-01
-2.08489016e-01 -6.81548178e-01 -4.39298190e-02 4.30773139e-01
-2.29540952e-02 4.24828082e-02 4.87499863e-01 -6.13705575e-01
-5.12449265e-01 -2.20582709e-01 -2.44437620e-01 2.71650642e-01
6.39820993e-02 -1.33138910e-01 -1.05433893e+00 -3.33692998e-01
4.48220700e-01 -1.79328263e-01 8.90963316e-01 7.47725666e-01
1.12114537e+00 -3.28607112e-02 -4.57917154e-01 6.04692996e-01
1.00132811e+00 1.66474422e-03 2.32037112e-01 2.52526909e-01
7.76746571e-01 1.02592039e+00 7.56173134e-01 3.43801498e-01
3.39710772e-01 8.03031862e-01 4.93714720e-01 -2.52715677e-01
6.09968118e-02 -1.49139553e-01 1.70161232e-01 1.22494116e-01
-1.99806020e-01 -1.15979835e-01 -7.68150389e-01 1.01849616e-01
-2.11324549e+00 -7.78945744e-01 -2.07320228e-01 2.37122679e+00
5.92846632e-01 2.38189712e-01 4.69928384e-01 1.30266890e-01
9.16207850e-01 1.48688885e-03 -6.26477659e-01 5.70342302e-01
9.03469622e-02 -4.67514724e-01 3.90829980e-01 3.84890050e-01
-1.35955667e+00 5.64539135e-01 5.69680119e+00 1.12043953e+00
-1.04204631e+00 2.05385108e-02 7.13804543e-01 -3.27112563e-02
9.48667750e-02 -2.39786997e-01 -9.43988323e-01 4.34480131e-01
2.88312107e-01 3.02834660e-01 5.25415465e-02 1.20701241e+00
-8.17834660e-02 -4.21211988e-01 -1.37740970e+00 1.35013545e+00
8.52455348e-02 -8.11433911e-01 -3.92694592e-01 2.23030567e-01
4.29025918e-01 -3.09177667e-01 2.41321266e-01 4.01772350e-01
-1.61316156e-01 -9.73900676e-01 1.11932123e+00 3.96211654e-01
6.52353287e-01 -5.90868175e-01 6.33925438e-01 1.00213897e+00
-1.06416821e+00 -4.43781950e-02 -4.40119654e-01 1.49383783e-01
1.00279838e-01 7.00802743e-01 -4.03903872e-01 3.82578313e-01
8.27544630e-01 1.59575865e-01 -7.60690391e-01 9.25042689e-01
-1.21477827e-01 3.62917781e-02 -5.23123324e-01 6.89585060e-02
-2.73722112e-01 -5.39240837e-01 5.80538988e-01 1.07574809e+00
9.45292711e-02 -1.01627730e-01 3.37114990e-01 1.06676006e+00
4.72984016e-01 -9.38924551e-02 -4.36532468e-01 4.26372558e-01
2.30778322e-01 1.32472110e+00 -1.02807498e+00 -2.21527442e-01
-1.92711428e-01 8.99227381e-01 2.37014323e-01 3.98130864e-01
-1.07437122e+00 -4.15767133e-02 2.29895756e-01 5.01848996e-01
3.25670898e-01 -3.10433835e-01 -3.43309075e-01 -1.28011572e+00
2.91526943e-01 -9.21653807e-01 3.00840080e-01 -5.26743591e-01
-1.28958976e+00 6.44795299e-01 5.93519747e-01 -1.51618230e+00
-3.34459960e-01 -9.18107510e-01 -2.22678185e-01 7.19088197e-01
-9.99038100e-01 -1.16530013e+00 -2.73535877e-01 4.02848601e-01
5.61409950e-01 -3.26357745e-02 3.39043736e-01 2.45132029e-01
-5.43233514e-01 3.59164774e-01 -6.93882108e-02 7.18194097e-02
7.73801684e-01 -1.26664579e+00 -6.40031695e-02 5.93815029e-01
3.03812236e-01 5.28435469e-01 8.52817178e-01 -4.84680027e-01
-9.46441233e-01 -8.02785039e-01 3.38067621e-01 -7.30783343e-01
2.58227617e-01 -5.48944652e-01 -8.93162429e-01 4.46706504e-01
-2.56726950e-01 2.01072499e-01 5.05198836e-01 3.16237003e-01
-3.24415177e-01 9.26619745e-04 -1.09831095e+00 5.67417204e-01
8.04060161e-01 -2.41190851e-01 -5.27425826e-01 1.27020299e-01
3.46289963e-01 -4.90389854e-01 -4.68856961e-01 7.22645521e-01
5.97392082e-01 -1.38457453e+00 1.08865416e+00 -3.82482857e-01
2.54529268e-01 -5.32094061e-01 -2.81556010e-01 -1.21219301e+00
-5.80315888e-01 -2.47490287e-01 -3.68013382e-01 1.02517867e+00
3.34245443e-01 -3.82278144e-01 9.29162085e-01 5.16185284e-01
2.38150924e-01 -9.56641138e-01 -8.55164409e-01 -8.61589849e-01
1.21774198e-02 -4.00589257e-01 1.21097900e-01 6.65165365e-01
-3.81658524e-01 5.85278928e-01 -3.03521156e-01 2.75148183e-01
9.39386308e-01 2.44171873e-01 8.53251278e-01 -1.34100711e+00
-6.42194867e-01 -5.86319864e-01 -1.67331234e-01 -1.33479786e+00
-2.29162723e-01 -2.43655831e-01 3.29022139e-01 -1.26336694e+00
2.63819158e-01 -3.68947715e-01 9.11872238e-02 -2.69956551e-02
-4.01749939e-01 2.17231125e-01 9.54362676e-02 4.65415448e-01
-5.75993121e-01 5.60013175e-01 1.13150358e+00 -4.27249260e-02
-2.23089427e-01 4.58639234e-01 -5.40098190e-01 1.29530048e+00
7.04150140e-01 -3.71385276e-01 -5.67341089e-01 -2.30860665e-01
-1.01978749e-01 7.18732849e-02 4.12300974e-01 -1.01421893e+00
9.49613750e-02 -4.23710048e-01 6.98828816e-01 -7.38402069e-01
5.53157568e-01 -1.00031030e+00 9.90604609e-02 3.49731028e-01
-1.41288444e-01 -2.65921265e-01 -1.11283869e-01 7.08136976e-01
-9.10833180e-02 -2.73639232e-01 1.13316822e+00 -1.03781730e-01
-3.34064662e-01 4.07247692e-01 -9.62931290e-02 2.51538977e-02
1.16916668e+00 -4.52727824e-01 2.15783432e-01 -4.20386851e-01
-9.28318918e-01 9.67768505e-02 4.11796361e-01 3.98650765e-01
4.84541535e-01 -1.25765574e+00 -6.67503417e-01 3.65071088e-01
1.80690661e-01 4.32008356e-01 3.25480253e-01 1.02952504e+00
-2.18198270e-01 1.96262076e-01 -9.22047533e-03 -1.07728338e+00
-1.27842963e+00 8.32724392e-01 4.94293004e-01 -2.01060414e-01
-4.16725248e-01 9.55273092e-01 8.29528511e-01 -6.05634570e-01
4.53540504e-01 -3.19558471e-01 -2.64553964e-01 2.46486530e-01
4.35254991e-01 4.72548336e-01 -1.55252948e-01 -5.95267117e-01
-3.95236135e-01 7.34188139e-01 2.61720866e-01 -3.74899536e-01
1.02718008e+00 -2.14554712e-01 8.48958194e-02 5.70904791e-01
9.23406243e-01 1.46470666e-01 -1.58341300e+00 -1.45321473e-01
-3.10684014e-02 -4.17971402e-01 -8.35366547e-02 -5.53779066e-01
-7.82359242e-01 1.00237405e+00 6.96778834e-01 2.94632345e-01
8.84233177e-01 3.12334895e-01 5.70247360e-02 1.33334979e-01
3.56618971e-01 -1.16038013e+00 2.87912369e-01 8.95496532e-02
1.12493920e+00 -1.50615859e+00 1.23464890e-01 -6.71372473e-01
-8.02601337e-01 9.55011308e-01 1.04164457e+00 -1.27291486e-01
6.44466996e-01 3.78462523e-01 -4.74820808e-02 8.98557832e-04
-3.33244771e-01 -1.44821018e-01 7.59094954e-01 8.37796152e-01
2.30036631e-01 1.93295822e-01 -1.92349404e-02 8.88856888e-01
1.36694862e-02 -2.68097967e-01 8.70954804e-03 6.87158108e-01
-5.95777154e-01 -8.03982019e-01 -1.02747035e+00 2.12637365e-01
-5.57182580e-02 2.67535388e-01 -1.87500685e-01 9.39054728e-01
4.52585757e-01 7.60414243e-01 1.16319098e-01 -3.27961475e-01
4.61341470e-01 -1.86916992e-01 8.53742480e-01 -5.96188903e-01
1.35194868e-01 6.66919231e-01 -2.51902103e-01 -3.92709643e-01
-1.88067555e-01 -5.23845136e-01 -1.01247203e+00 7.56496936e-02
-6.74671710e-01 8.05551261e-02 5.49546599e-01 8.38942766e-01
1.35855719e-01 4.71350670e-01 4.47265625e-01 -1.29558003e+00
-1.02512801e+00 -1.11029935e+00 -7.43846357e-01 1.04041487e-01
2.39943624e-01 -9.01725292e-01 -4.23350066e-01 -1.60699729e-02] | [7.296097755432129, -1.5730516910552979] |
66f571b5-1496-41f8-b461-812f96b75aef | feasibility-of-colon-cancer-detection-in | 1812.01464 | null | http://arxiv.org/abs/1812.01464v2 | http://arxiv.org/pdf/1812.01464v2.pdf | Feasibility of Colon Cancer Detection in Confocal Laser Microscopy Images Using Convolution Neural Networks | Histological evaluation of tissue samples is a typical approach to identify
colorectal cancer metastases in the peritoneum. For immediate assessment,
reliable and real-time in-vivo imaging would be required. For example,
intraoperative confocal laser microscopy has been shown to be suitable for
distinguishing organs and also malignant and benign tissue. So far, the
analysis is done by human experts. We investigate the feasibility of automatic
colon cancer classification from confocal laser microscopy images using deep
learning models. We overcome very small dataset sizes through transfer learning
with state-of-the-art architectures. We achieve an accuracy of 89.1% for cancer
detection in the peritoneum which indicates viability as an intraoperative
decision support system. | ['Daniel Drömann', 'Lukas Wittig', 'Nils Gessert', 'Alexander Schlaefer', 'Tobias Keck', 'David B. Ellebrecht'] | 2018-12-04 | null | null | null | null | ['colon-cancer-detection-in-confocal-laser'] | ['medical'] | [-1.48313060e-01 -5.02765961e-02 -8.27171803e-02 2.36943010e-02
-7.63486803e-01 -4.75159973e-01 1.57491982e-01 6.20743990e-01
-1.04177988e+00 7.60268927e-01 -2.32621208e-01 -7.95688272e-01
4.21229869e-01 -5.95281243e-01 -3.70605916e-01 -1.00389588e+00
-1.65772527e-01 4.09257323e-01 3.69328819e-02 5.32505289e-02
1.01187415e-01 7.88523376e-01 -7.28774846e-01 5.54565191e-01
6.94014490e-01 8.93739343e-01 3.99410158e-01 9.46868896e-01
-3.51586891e-03 6.73097968e-01 -1.22277975e-01 -7.67389219e-03
1.29208893e-01 -1.52067661e-01 -9.45504248e-01 -2.48793647e-01
1.79483011e-01 -4.45770293e-01 6.24384470e-02 9.94001389e-01
7.19276369e-01 -5.46069205e-01 9.32577848e-01 -4.09759849e-01
-4.26759422e-02 1.70455217e-01 -5.22046804e-01 4.00419384e-01
8.62887204e-02 1.80234000e-01 3.11791688e-01 -7.38432884e-01
5.15100479e-01 3.19417804e-01 8.08881342e-01 6.20470047e-01
-1.16254961e+00 -5.70844710e-01 -5.02796531e-01 -9.18178633e-02
-1.11089492e+00 -2.47882292e-01 1.69719666e-01 -8.49348545e-01
7.28001654e-01 2.26639628e-01 1.00259292e+00 7.72192121e-01
1.25722122e+00 6.79648817e-01 1.25852275e+00 -5.03878415e-01
9.16387215e-02 1.65858135e-01 -3.74723822e-01 1.09593987e+00
4.47312742e-01 4.16221827e-01 2.17284150e-02 -9.25102308e-02
1.05292988e+00 2.59509236e-01 -3.90302032e-01 -4.99527872e-01
-1.50883937e+00 6.94594562e-01 8.16804647e-01 6.52701020e-01
-4.41612422e-01 2.73750693e-01 7.76178241e-01 3.32154363e-01
2.34430701e-01 5.68120182e-01 -2.43756384e-01 3.21112603e-01
-7.77798474e-01 -5.66867530e-01 7.33768106e-01 5.03575742e-01
2.05950454e-01 -5.45323908e-01 1.83756456e-01 4.30596411e-01
2.72393376e-01 1.34143040e-01 9.18747306e-01 -6.95591271e-01
-2.64664799e-01 5.11071801e-01 1.12626597e-01 -3.88402373e-01
-8.30675662e-01 -5.73296010e-01 -1.39363170e+00 6.16250753e-01
5.85131764e-01 -1.45839438e-01 -7.40336716e-01 9.06839371e-01
1.38757452e-01 -1.26251712e-01 7.96080753e-02 9.57302153e-01
8.32840204e-01 -4.93014641e-02 2.47835964e-01 2.74213322e-04
1.34974432e+00 -9.74232972e-01 -3.97221327e-01 6.77798539e-02
1.29298115e+00 -6.36876523e-01 5.29744029e-01 4.13466722e-01
-7.72547185e-01 -4.43708599e-02 -8.08199346e-01 1.42646013e-02
-2.50886828e-01 5.62597752e-01 9.60732639e-01 5.83054781e-01
-1.07717180e+00 3.88706535e-01 -1.34999835e+00 -5.92330515e-01
5.11128306e-01 6.31405234e-01 -1.11149681e+00 -3.53206471e-02
-6.57677352e-01 1.02381516e+00 2.55648345e-01 2.20473140e-01
-1.00191772e+00 -6.35194540e-01 -7.27413356e-01 -9.08710957e-02
-2.47829154e-01 -1.16148043e+00 1.21054280e+00 -7.07811117e-01
-1.52809322e+00 1.58334279e+00 8.22762866e-03 -6.16934717e-01
7.14754701e-01 2.24390715e-01 5.96204437e-02 3.61370683e-01
8.01916700e-03 4.54855710e-01 3.15869749e-01 -9.88780618e-01
-8.37561190e-01 -3.51464868e-01 -1.45958170e-01 -9.89672448e-03
-4.85903509e-02 -1.07812107e-01 2.01003775e-02 -1.64898947e-01
5.97113185e-02 -1.06210613e+00 -6.98029876e-01 5.85352600e-01
-2.15509146e-01 3.24166268e-01 8.49342197e-02 -5.92030168e-01
7.24004209e-01 -1.97596359e+00 -2.54531533e-01 1.54807135e-01
4.94147032e-01 1.63817644e-01 2.36175627e-01 2.60008760e-02
6.07344434e-02 1.68717340e-01 3.09017837e-01 -9.33543742e-02
-5.48976481e-01 -2.51262456e-01 5.23395956e-01 1.20208812e+00
-9.41829458e-02 1.06820714e+00 -1.14098477e+00 -8.53751242e-01
4.63509947e-01 3.41597319e-01 -4.10778642e-01 1.90365329e-01
4.44149405e-01 8.95779192e-01 -2.39269003e-01 8.95860136e-01
2.71511257e-01 -5.64762414e-01 2.75148541e-01 -1.85443908e-01
-1.71448320e-01 -2.82531738e-01 -2.97894418e-01 1.96812022e+00
-6.85742438e-01 6.24114990e-01 5.58262825e-01 -9.52857435e-01
4.96449679e-01 5.80295861e-01 6.31511152e-01 -7.38326669e-01
3.99660468e-01 6.70422077e-01 2.54365534e-01 -7.07660139e-01
-1.10794276e-01 -7.97160327e-01 1.63915604e-01 1.96310252e-01
5.97353131e-02 8.24970938e-03 1.44287087e-02 -2.82681733e-01
1.07906997e+00 -3.05143982e-01 8.27871919e-01 -7.88072526e-01
7.37090588e-01 3.64341766e-01 1.02363460e-01 4.98289913e-01
-5.40096104e-01 4.42929804e-01 2.31874898e-01 -9.34209645e-01
-7.46204317e-01 -9.74909008e-01 -5.74268699e-01 5.11095583e-01
1.51987851e-01 4.48510766e-01 -4.81597632e-01 -7.08322883e-01
7.14993253e-02 -6.28994033e-02 -1.01328969e+00 1.06027104e-01
-4.29388225e-01 -7.14810133e-01 3.39284211e-01 6.09242916e-01
1.67591691e-01 -6.62149370e-01 -1.04603148e+00 3.50188106e-01
-1.17761612e-01 -8.73860896e-01 6.59933966e-03 5.33898830e-01
-1.12104237e+00 -1.52218449e+00 -1.32130122e+00 -1.17468381e+00
1.15746999e+00 2.87560791e-01 1.15453565e+00 5.09241521e-01
-8.33831787e-01 9.26810354e-02 -4.10015322e-02 -5.41627109e-01
-7.63644457e-01 7.97152966e-02 -7.83580393e-02 -3.45271945e-01
3.40054899e-01 -6.85943812e-02 -1.00511479e+00 1.65518090e-01
-4.56970602e-01 7.30108917e-02 1.23466790e+00 1.20045269e+00
7.65277326e-01 -2.31564537e-01 1.03528552e-01 -9.85574901e-01
4.61591601e-01 -3.29535097e-01 -3.81789058e-01 -5.39855659e-02
-2.74065346e-01 -3.24905932e-01 7.25275815e-01 -3.28714959e-02
-5.91234744e-01 1.94499671e-01 -1.60751417e-01 -2.02782169e-01
-4.89702195e-01 8.67780685e-01 9.35748577e-01 -5.74055791e-01
9.09877121e-01 1.29608382e-02 5.53709209e-01 4.60684858e-02
-4.76123929e-01 6.09604895e-01 4.53064024e-01 -3.13716322e-01
1.44786209e-01 9.20342565e-01 4.22025144e-01 -7.21198618e-01
-5.21480203e-01 -1.07769489e+00 -8.08341265e-01 -1.35574341e-01
8.72448027e-01 -1.25494647e+00 -7.66946971e-01 3.73867154e-01
-8.19144726e-01 -5.87781131e-01 3.55928019e-02 8.83963168e-01
-5.16076684e-01 3.85598451e-01 -1.19415057e+00 -4.20100272e-01
-6.67873919e-01 -1.10478675e+00 1.08702636e+00 -2.57910918e-02
-3.22800279e-01 -1.47403169e+00 2.16334477e-01 -1.19208820e-01
5.96648276e-01 6.50439501e-01 8.12451899e-01 -4.21757996e-01
-3.20184886e-01 -7.54521549e-01 -2.27295667e-01 -7.12510571e-02
2.24240482e-01 6.28269911e-02 -8.02187085e-01 -6.84717238e-01
-1.25395432e-01 -2.78335631e-01 9.67712641e-01 7.51145661e-01
1.07328320e+00 1.86864048e-01 -9.64686275e-01 7.92298555e-01
1.67545545e+00 3.29102278e-02 5.05237997e-01 5.71522713e-01
2.31492206e-01 4.51904446e-01 5.45156062e-01 2.55005598e-01
7.75905624e-02 2.13485043e-02 6.50588870e-01 -6.96986198e-01
8.70370045e-02 1.98286816e-01 -2.44744509e-01 4.87177908e-01
-3.23507607e-01 7.17355695e-04 -1.28972638e+00 7.97760367e-01
-1.26281798e+00 -5.95246315e-01 -1.35810897e-01 2.27596283e+00
7.54973531e-01 -1.19167663e-01 -2.40982831e-01 -8.57966766e-02
4.57449526e-01 -6.32741868e-01 -2.03786269e-02 -2.25438662e-02
4.30663645e-01 2.62891240e-02 6.62252307e-01 4.95510995e-01
-1.31991756e+00 3.78208339e-01 7.05463028e+00 3.43187511e-01
-1.67988551e+00 -6.51710108e-02 8.44198883e-01 2.16171548e-01
2.39436522e-01 -3.66789639e-01 -3.40440810e-01 3.66666287e-01
5.53602993e-01 1.40054775e-02 -1.62357315e-01 5.50573409e-01
3.87623608e-02 -6.29490256e-01 -1.39696729e+00 9.33351159e-01
-2.85019606e-01 -1.61523426e+00 -2.70002544e-01 4.08215493e-01
5.65045536e-01 2.11989865e-01 -4.34537344e-02 1.45071447e-01
-1.93360731e-01 -1.31363845e+00 -8.79831091e-02 6.19471371e-01
1.43623126e+00 -4.33548391e-01 1.49490011e+00 5.71915805e-01
-8.37194562e-01 -1.84223277e-03 -3.81673962e-01 1.08191781e-01
-3.60002190e-01 5.47245324e-01 -1.42727304e+00 2.44589597e-01
5.47369540e-01 6.11365855e-01 -3.26488256e-01 1.30104661e+00
1.82166025e-01 1.30308941e-01 -2.90528893e-01 -1.41342625e-01
2.49553680e-01 6.11353181e-02 -2.17193123e-02 1.54831469e+00
4.74773675e-01 -1.45076960e-01 8.96969587e-02 6.06549442e-01
1.39475584e-01 2.68518031e-01 -5.88409781e-01 1.07969798e-01
-6.03425615e-02 1.58961725e+00 -9.36344087e-01 -1.42618760e-01
-3.69788945e-01 8.25179517e-01 1.72718927e-01 -1.48524269e-02
-3.56643856e-01 -5.04028983e-02 1.13217272e-01 3.91045451e-01
-2.40297616e-01 -3.28622237e-02 -2.20864177e-01 -1.15253067e+00
-5.12678444e-01 -3.04833889e-01 3.63266796e-01 -4.05276746e-01
-1.12650752e+00 4.83993590e-01 -6.37389243e-01 -1.61883342e+00
-1.13218307e-01 -1.19377649e+00 -7.57025063e-01 9.05507267e-01
-1.93867004e+00 -9.86504138e-01 -7.93566585e-01 3.68721247e-01
1.24781169e-01 -6.54012710e-02 1.47730029e+00 8.70505795e-02
-2.08256155e-01 3.58429581e-01 2.00863361e-01 4.66959864e-01
8.54666650e-01 -1.65145135e+00 -2.49400318e-01 2.35489696e-01
-4.94347066e-01 7.16825366e-01 4.02588546e-01 -2.75329530e-01
-1.35962534e+00 -9.33772564e-01 6.01681113e-01 -2.00201526e-01
6.40700161e-01 2.34239668e-01 -4.35878158e-01 6.13069713e-01
2.57097751e-01 5.94264507e-01 1.41258931e+00 -1.41567811e-01
2.07914531e-01 1.06081106e-01 -1.24291658e+00 5.72047472e-01
2.05812290e-01 -3.78910094e-01 -8.63054097e-02 4.19341087e-01
-2.82429665e-01 -7.99956977e-01 -1.24847603e+00 5.87656796e-01
8.17479312e-01 -1.11246729e+00 7.28935063e-01 -4.15545732e-01
5.15464544e-01 -9.02484953e-02 3.44027042e-01 -1.36593783e+00
-4.53818083e-01 -1.88798636e-01 2.99353004e-01 3.23548838e-02
4.39613342e-01 -5.88366508e-01 1.20483911e+00 2.20893145e-01
-1.34477645e-01 -9.53605115e-01 -8.94060969e-01 -4.20472801e-01
5.07503033e-01 3.83838087e-01 -1.88117504e-01 9.90601718e-01
5.03476322e-01 9.10089724e-03 4.32501733e-01 1.69909690e-02
5.55804253e-01 3.00816238e-01 6.63759828e-01 -1.32625961e+00
9.75516513e-02 -6.35320663e-01 -6.48044586e-01 -6.10335648e-01
4.16169800e-02 -1.07594836e+00 -1.12156883e-01 -1.85882080e+00
4.98612881e-01 -4.89609420e-01 -6.36965513e-01 2.30268881e-01
-1.26454145e-01 4.01007444e-01 -3.65344733e-01 3.31736624e-01
-5.11429608e-01 -1.28528237e-01 1.57974470e+00 -2.23240018e-01
2.43545726e-01 1.03629015e-01 -4.49252993e-01 6.60275757e-01
9.32739377e-01 -2.93654054e-01 3.75805408e-01 -1.28353417e-01
1.21691853e-01 3.32274169e-01 4.17298704e-01 -1.16781127e+00
4.97624338e-01 9.05521140e-02 7.87736654e-01 -2.98780859e-01
9.08665080e-03 -1.06355107e+00 2.00686548e-02 1.25350046e+00
-2.39863515e-01 -3.59514326e-01 2.84374475e-01 4.45540011e-01
-5.50269306e-01 -9.57193673e-02 1.07584262e+00 -8.64171445e-01
-5.64881802e-01 2.85977870e-01 -8.50473821e-01 -4.80193645e-01
1.33110154e+00 -4.34363812e-01 -2.78212994e-01 5.54565825e-02
-8.71856570e-01 2.12905005e-01 8.74609172e-01 -4.41184431e-01
6.38263822e-01 -1.02802956e+00 -1.04058754e+00 1.56835854e-01
3.86804521e-01 2.94481695e-01 3.33890140e-01 1.75877917e+00
-1.43747807e+00 6.61686003e-01 -5.19191384e-01 -8.03053141e-01
-1.21262085e+00 6.06794298e-01 1.07558036e+00 -6.14771545e-01
-4.82761055e-01 9.66323316e-01 2.63555110e-01 -3.74132425e-01
-1.62943825e-01 -5.67894816e-01 -3.62663776e-01 -2.75360316e-01
3.39982241e-01 -3.27391364e-02 2.40783915e-01 -2.18866870e-01
-3.33751857e-01 4.41416711e-01 -2.58346349e-01 4.49645281e-01
9.88719761e-01 4.53236438e-02 -3.39932442e-01 2.89894283e-01
1.32517469e+00 -2.80333329e-02 -8.09715152e-01 -1.88019872e-02
-2.46071905e-01 -4.00084943e-01 2.85548061e-01 -8.06364417e-01
-9.34721470e-01 1.10481334e+00 7.48922348e-01 6.32734001e-02
9.20718610e-01 -2.43574008e-01 4.44700450e-01 4.96104419e-01
7.85905898e-01 -6.55197799e-01 -1.80952832e-01 2.14192167e-01
6.92436337e-01 -1.84888506e+00 1.43524542e-01 -4.88884300e-01
-3.64378005e-01 1.53638172e+00 4.16302770e-01 -3.82421702e-01
6.88585877e-01 6.84618950e-01 6.06913388e-01 -1.47959724e-01
-6.93938315e-01 1.09961271e-01 6.35462478e-02 4.14759934e-01
1.00950062e+00 2.97789574e-01 -2.38575861e-01 2.33130857e-01
2.18922392e-01 3.29268605e-01 5.56540787e-01 1.06658554e+00
-4.23639596e-01 -6.34584904e-01 4.97767814e-02 6.09931707e-01
-1.14134049e+00 -1.03002883e-01 -8.14107507e-02 9.34497476e-01
-2.97229767e-01 4.64425862e-01 -1.02630205e-01 8.64234492e-02
-5.90263791e-02 -3.22620839e-01 6.17658198e-01 -7.27069318e-01
-8.26442361e-01 1.74419492e-01 -2.00151086e-01 -4.42254454e-01
-4.48213935e-01 -3.37064713e-01 -1.31812537e+00 4.70036305e-02
-3.49931449e-01 1.99176297e-01 6.56173706e-01 6.04345202e-01
1.27450272e-01 5.36037624e-01 4.46904749e-01 -9.01849508e-01
-3.99648130e-01 -1.06385922e+00 -8.27194095e-01 4.03180569e-02
7.24048436e-01 -1.93891853e-01 -4.71319765e-01 1.38514504e-01] | [15.052409172058105, -2.9366087913513184] |
f7c24821-f8ba-4ae2-b40c-6ffeb5207e24 | gcn-based-linkage-prediction-for-face | 2107.02477 | null | https://arxiv.org/abs/2107.02477v3 | https://arxiv.org/pdf/2107.02477v3.pdf | A Linkage-based Doubly Imbalanced Graph Learning Framework for Face Clustering | In recent years, benefiting from the expressive power of Graph Convolutional Networks (GCNs), significant breakthroughs have been made in face clustering area. However, rare attention has been paid to GCN-based clustering on imbalanced data. Although imbalance problem has been extensively studied, the impact of imbalanced data on GCN- based linkage prediction task is quite different, which would cause problems in two aspects: imbalanced linkage labels and biased graph representations. The former is similar to that in classic image classification task, but the latter is a particular problem in GCN-based clustering via linkage prediction. Significantly biased graph representations in training can cause catastrophic over-fitting of a GCN model. To tackle these challenges, we propose a linkage-based doubly imbalanced graph learning framework for face clustering. In this framework, we evaluate the feasibility of those existing methods for imbalanced image classification problem on GCNs, and present a new method to alleviate the imbalanced labels and also augment graph representations using a Reverse-Imbalance Weighted Sampling (RIWS) strategy. With the RIWS strategy, probability-based class balancing weights could ensure the overall distribution of positive and negative samples; in addition, weighted random sampling provides diverse subgraph structures, which effectively alleviates the over-fitting problem and improves the representation ability of GCNs. Extensive experiments on series of imbalanced benchmark datasets synthesized from MS-Celeb-1M and DeepFashion demonstrate the effectiveness and generality of our proposed method. Our implementation and the synthesized datasets will be openly available on https://github.com/espectre/GCNs_on_imbalanced_datasets. | ['Xingjian Chen', 'Qijie Shen', 'Rong Du', 'Fangyi Zhang', 'Huafeng Yang'] | 2021-07-06 | null | null | null | null | ['face-clustering'] | ['computer-vision'] | [-7.41672069e-02 1.13157041e-01 -3.72007698e-01 -4.43900019e-01
-2.06916139e-01 -6.48258701e-02 6.18291087e-02 -9.36801806e-02
3.66701722e-01 4.51693058e-01 4.22839820e-02 -1.67516228e-02
-2.90725112e-01 -1.04003954e+00 -5.33181250e-01 -7.96483040e-01
2.55921613e-02 4.81540829e-01 -2.06379667e-02 -1.90742701e-01
2.97547709e-02 3.78406525e-01 -1.51362264e+00 2.95832336e-01
9.73844230e-01 9.02144253e-01 -3.42391372e-01 1.01750337e-01
-3.18429768e-01 6.42015100e-01 -6.65943444e-01 -5.78733027e-01
2.27471679e-01 -4.81441230e-01 -5.13160706e-01 2.91718632e-01
4.38117027e-01 1.07892849e-01 -6.96601987e-01 1.07390106e+00
8.82609725e-01 -2.41270036e-01 7.12024808e-01 -1.89411581e+00
-6.02710605e-01 5.14319658e-01 -1.31830657e+00 1.17500946e-01
1.15126587e-01 4.05235849e-02 6.97333932e-01 -5.56570172e-01
4.70797479e-01 1.58426523e+00 8.14914227e-01 4.92643237e-01
-1.01026773e+00 -1.30420399e+00 2.00470433e-01 3.00252765e-01
-1.47057664e+00 -3.06315213e-01 1.07879400e+00 -3.75559896e-01
4.84679252e-01 1.37167320e-01 7.90758908e-01 9.31057751e-01
-7.63957128e-02 6.83777213e-01 7.88326204e-01 -1.35050993e-02
-7.48123527e-02 -2.95195132e-01 9.79040787e-02 6.36811972e-01
7.90339112e-01 -3.06236625e-01 -5.29283345e-01 -7.07668439e-02
5.63759387e-01 1.26409590e-01 -3.62326145e-01 -5.92135489e-01
-7.43820190e-01 6.42948627e-01 7.69165754e-01 6.72498569e-02
2.58628149e-02 1.29698068e-01 6.68471813e-01 1.93287954e-01
8.37420940e-01 -1.76693052e-01 -1.60400227e-01 3.86612952e-01
-8.12335253e-01 1.56673655e-01 5.35202742e-01 9.19831991e-01
6.61553502e-01 -6.45392647e-05 -1.29967362e-01 1.00779462e+00
3.44961971e-01 3.71639222e-01 4.97768492e-01 -4.58866388e-01
7.05184579e-01 1.13450193e+00 -6.01176322e-01 -1.83972549e+00
-5.84698200e-01 -6.62630737e-01 -1.42017448e+00 -1.63828611e-01
3.34563553e-01 1.12456389e-01 -9.30287004e-01 1.78547323e+00
5.80264091e-01 3.88685346e-01 -3.33536685e-01 9.37089682e-01
1.15801394e+00 4.86999154e-01 -2.03731265e-02 -3.46115679e-01
1.10164201e+00 -7.00316668e-01 -8.49362910e-01 2.83705834e-02
7.20598221e-01 -5.96814275e-01 8.76545906e-01 2.95162857e-01
-6.69184148e-01 -4.74742711e-01 -9.41840470e-01 1.15909174e-01
-1.90198049e-01 -8.11040252e-02 7.92809188e-01 7.49571800e-01
-9.58441675e-01 4.35426593e-01 -5.56991994e-01 -2.78782129e-01
1.04614973e+00 5.51679611e-01 -4.33017820e-01 -5.01265347e-01
-1.01507914e+00 1.29493698e-01 3.53241324e-01 3.26084793e-01
-5.65513909e-01 -7.04624951e-01 -7.68350244e-01 -5.02451472e-02
3.13171238e-01 -4.30886507e-01 6.04815483e-01 -1.12942684e+00
-8.33044052e-01 9.80136395e-01 4.43408117e-02 3.56174968e-02
5.36111832e-01 3.52228999e-01 -5.40577531e-01 -2.17331499e-02
8.24980363e-02 6.53023005e-01 7.06298172e-01 -1.19227171e+00
-4.81549278e-02 -9.58231807e-01 -2.96027482e-01 1.72100320e-01
-5.14608622e-01 -2.70328969e-01 -5.98813415e-01 -7.64088392e-01
4.95380223e-01 -7.95576274e-01 6.97764978e-02 -6.33502975e-02
-7.75461197e-01 -3.26753706e-01 9.47686970e-01 -4.30170298e-01
1.35043752e+00 -2.10588193e+00 -5.93067668e-02 2.82792866e-01
4.92221415e-01 3.50283474e-01 -3.96315873e-01 2.57061660e-01
-5.59856176e-01 3.55862528e-01 -1.91076443e-01 -1.19241200e-01
-1.99866325e-01 -6.21110462e-02 -3.81750688e-02 5.69815576e-01
2.78304189e-01 7.45945275e-01 -7.15640604e-01 -6.41899943e-01
-1.33625373e-01 5.16634524e-01 -7.11902499e-01 2.51615286e-01
-5.72592653e-02 3.78470331e-01 -2.77733564e-01 8.44500422e-01
1.45291603e+00 -2.56262243e-01 5.74557304e-01 -5.52303612e-01
5.07881284e-01 -5.50579932e-03 -1.39050591e+00 1.40221250e+00
1.51908426e-02 1.79896533e-01 -3.81992795e-02 -1.36843252e+00
1.08941436e+00 -2.36664917e-02 5.24997473e-01 -8.47820044e-01
1.43612981e-01 7.64930099e-02 7.82839730e-02 -5.31220198e-01
7.16309696e-02 -2.35643685e-02 1.85541078e-01 4.14304346e-01
8.10930580e-02 1.79359347e-01 1.53923556e-01 3.91224504e-01
9.47410226e-01 7.27487206e-02 1.06711844e-02 -2.86473840e-01
4.62520540e-01 -2.29009628e-01 1.01455152e+00 6.65687174e-02
-3.07048887e-01 7.21449494e-01 1.11447453e+00 -4.83822554e-01
-5.99145889e-01 -7.40486324e-01 -1.19721197e-01 8.23880494e-01
3.47833574e-01 -6.43055737e-01 -8.84880364e-01 -9.19215560e-01
2.40221053e-01 -1.29663587e-01 -5.58161557e-01 -4.68122333e-01
-4.62648958e-01 -1.50766408e+00 6.08605206e-01 3.39632601e-01
6.27655149e-01 -8.31846416e-01 3.38234991e-01 -1.80111498e-01
-2.85001695e-01 -7.82185733e-01 -2.94434935e-01 -2.87670583e-01
-8.99151921e-01 -1.74308479e+00 -4.42941964e-01 -8.63022029e-01
9.89585400e-01 6.37356520e-01 1.32072353e+00 6.04102790e-01
-5.20915270e-01 1.71670746e-02 -1.95575759e-01 -3.13578665e-01
1.95546150e-02 9.14160758e-02 1.04692228e-01 1.38936639e-01
6.07251883e-01 -6.36990905e-01 -9.49870288e-01 4.86278206e-01
-1.04855967e+00 3.17021683e-02 2.56588072e-01 7.26916015e-01
5.28008699e-01 2.97360003e-01 1.04091501e+00 -1.21146715e+00
4.79662418e-01 -7.16627419e-01 -4.70175683e-01 6.08552806e-02
-5.53788066e-01 -3.33746970e-01 5.12392104e-01 -2.15809554e-01
-7.69275129e-01 -1.48502022e-01 -1.79053515e-01 -4.38038737e-01
-1.85012408e-02 2.58659452e-01 -7.88730204e-01 -1.56063423e-01
3.24330717e-01 -1.61766768e-01 3.25228184e-01 -2.76031256e-01
1.40683144e-01 8.05380881e-01 5.79573214e-03 -4.44883585e-01
5.60653865e-01 4.98722672e-01 9.97300521e-02 -5.07666230e-01
-7.34141827e-01 -2.77911037e-01 -3.14029247e-01 -3.06197137e-01
4.30138081e-01 -1.15959442e+00 -9.71321940e-01 8.71236622e-01
-8.20962906e-01 -1.40000626e-01 1.85506389e-01 9.55199376e-02
-2.53847331e-01 5.26106477e-01 -5.53096652e-01 -5.30240238e-01
-2.65162766e-01 -1.16663110e+00 1.00251198e+00 3.43488395e-01
1.48313761e-01 -6.86482728e-01 -1.17180973e-01 7.57978261e-01
1.91051811e-01 4.67655927e-01 1.07151377e+00 -5.57182074e-01
-5.56183755e-01 -3.51377502e-02 -5.56617856e-01 2.31836393e-01
2.69375622e-01 9.56009105e-02 -9.35231388e-01 -5.90329468e-01
-2.13106588e-01 -4.95840043e-01 9.68320906e-01 3.13198060e-01
1.65185916e+00 -2.36704081e-01 -6.42236531e-01 7.55185366e-01
1.46823657e+00 -1.15483426e-01 7.69791365e-01 -1.67080313e-02
1.14640570e+00 7.71770358e-01 5.44461787e-01 3.58682513e-01
6.00282788e-01 6.26852095e-01 7.77954280e-01 -2.28679866e-01
-4.02111381e-01 -2.37980351e-01 -3.54016176e-03 9.49216008e-01
1.50282130e-01 -3.88967395e-01 -9.40022588e-01 3.94621342e-01
-1.75802505e+00 -6.57845378e-01 -3.71916175e-01 2.08057165e+00
8.26398373e-01 -7.75202066e-02 2.43759364e-01 3.61836016e-01
1.22156787e+00 2.70869613e-01 -7.14111328e-01 1.03471868e-01
-1.88523054e-01 -1.97427675e-01 1.20618798e-01 2.55022570e-02
-9.97507513e-01 7.21592903e-01 5.06975365e+00 1.11797082e+00
-1.07639599e+00 1.81029625e-02 1.28251898e+00 -1.65137827e-01
-3.20989519e-01 -2.79152662e-01 -7.83010542e-01 8.58444393e-01
3.83670151e-01 3.34261097e-02 2.42180705e-01 6.32363558e-01
5.60833700e-02 1.97077990e-01 -7.55810916e-01 1.30478442e+00
2.77922720e-01 -1.04999363e+00 3.21653068e-01 1.75780088e-01
7.82425821e-01 -2.32559800e-01 1.60620093e-01 2.80847371e-01
-1.38785513e-02 -1.23973227e+00 1.41609296e-01 1.88186482e-01
9.02138233e-01 -1.11116540e+00 8.34445477e-01 1.00666597e-01
-1.16898489e+00 -7.64939412e-02 -7.94122517e-01 3.82585190e-02
-4.43084359e-01 1.28771830e+00 -5.79758704e-01 7.64808953e-01
8.61839652e-01 9.15454626e-01 -7.86978483e-01 9.76730168e-01
-2.69820429e-02 5.99294245e-01 -1.28352270e-01 4.29341376e-01
-3.07173133e-01 -4.33025688e-01 2.20489204e-01 9.21489000e-01
2.10551739e-01 1.55595941e-02 1.80832259e-02 7.06678092e-01
-5.05240858e-01 3.73581290e-01 -7.22250879e-01 1.06848702e-01
4.42814380e-01 1.48662925e+00 -8.73010457e-01 -3.51765356e-03
-1.74272656e-01 5.54022670e-01 4.96649861e-01 2.37496018e-01
-9.16950285e-01 -3.50898981e-01 7.08293021e-01 4.24826890e-01
1.50404587e-01 2.73458898e-01 -1.36044115e-01 -1.19061720e+00
5.03374711e-02 -1.19096601e+00 6.98441446e-01 -4.75265861e-01
-1.58632433e+00 5.95010757e-01 -3.32644969e-01 -1.14236522e+00
3.79541308e-01 -4.31183308e-01 -6.72425210e-01 4.96931136e-01
-1.56166863e+00 -1.21251965e+00 -8.10618281e-01 5.18355191e-01
4.18324247e-02 -7.27939755e-02 3.69107962e-01 6.89221919e-01
-1.12061667e+00 8.01500618e-01 -1.34415999e-01 2.89911926e-01
9.67381299e-01 -8.73263717e-01 1.12660810e-01 5.64508259e-01
-2.36772358e-01 4.17454422e-01 2.53316760e-01 -7.96893358e-01
-1.29303277e+00 -1.52032495e+00 3.84822905e-01 -4.68157716e-02
2.15049788e-01 -5.87551236e-01 -1.02585220e+00 5.12776673e-01
-7.48327151e-02 4.09956187e-01 8.83466244e-01 1.19142279e-01
-6.59452736e-01 -6.54497921e-01 -1.34833181e+00 3.84160757e-01
1.40164304e+00 -8.68345052e-02 -3.31403613e-02 6.85614407e-01
6.79499686e-01 -3.63942504e-01 -7.71236539e-01 8.57531250e-01
2.40444601e-01 -1.30671966e+00 9.63945150e-01 -5.92421293e-01
4.04564589e-01 -5.72174430e-01 1.21505357e-01 -1.32256055e+00
-2.80963540e-01 -1.43393710e-01 -4.96050864e-02 1.61482656e+00
-8.24442059e-02 -6.32874668e-01 1.07632065e+00 9.90772694e-02
1.13478623e-01 -1.05515432e+00 -7.99211621e-01 -5.26232243e-01
6.20452240e-02 -1.22137487e-01 1.05282557e+00 1.14958858e+00
-2.21105322e-01 2.40373492e-01 -1.26825035e-01 1.42891183e-01
8.32404256e-01 2.28297427e-01 8.41115832e-01 -1.37043321e+00
1.68613806e-01 -4.27431822e-01 -8.31770658e-01 -5.37183344e-01
3.24230433e-01 -1.22630334e+00 -4.21322048e-01 -1.41790700e+00
4.61539775e-01 -6.80349648e-01 -2.10553199e-01 4.52028334e-01
-3.71690959e-01 7.93267667e-01 9.98464972e-02 1.44072682e-01
-8.13011646e-01 6.75639391e-01 1.29411232e+00 -2.71041423e-01
1.03533067e-01 -2.09691018e-01 -1.00122738e+00 5.68973064e-01
9.88154948e-01 -4.40764964e-01 -6.47295415e-01 -2.75265515e-01
5.32356977e-01 -2.14765340e-01 2.74883091e-01 -1.04389727e+00
1.35561273e-01 2.39257038e-01 6.48417354e-01 -6.36422634e-01
5.68304881e-02 -6.39165163e-01 2.54521310e-01 4.17074829e-01
1.09419763e-01 -3.11147738e-02 1.62233174e-01 7.89285481e-01
-3.58266473e-01 3.29407185e-01 9.22172189e-01 -3.37136071e-03
-3.06802511e-01 7.62369037e-01 2.31600657e-01 2.34467328e-01
1.12939548e+00 -1.76258519e-01 -6.02713048e-01 -2.73854345e-01
-4.24347192e-01 4.74120915e-01 5.88793874e-01 4.16129887e-01
6.08271062e-01 -1.58996689e+00 -6.82564437e-01 4.55395997e-01
1.94291219e-01 2.80788749e-01 6.07297122e-01 9.35053825e-01
-5.27510405e-01 1.85848605e-02 -2.14619234e-01 -6.96684480e-01
-1.19429839e+00 6.01922750e-01 4.42241341e-01 -2.09491864e-01
-2.55747974e-01 8.38210881e-01 3.95664454e-01 -8.52599084e-01
3.32958490e-01 2.26108924e-01 -3.78154308e-01 3.88877094e-01
3.73337805e-01 4.88127053e-01 2.97767192e-01 -6.59314990e-01
-5.89774013e-01 6.79047227e-01 -1.27627552e-01 8.50239515e-01
1.33588529e+00 -1.46405891e-01 -7.01102674e-01 9.42934826e-02
1.11089516e+00 -2.05417886e-01 -8.00963640e-01 1.12819694e-01
-2.41858959e-01 -5.52902281e-01 -1.85701549e-01 -5.42494059e-01
-1.89481950e+00 9.28349316e-01 5.84143758e-01 2.01261044e-01
1.31247425e+00 -1.81610912e-01 7.68317878e-01 -1.31146595e-01
2.29964584e-01 -8.60913217e-01 3.37613195e-01 4.27934863e-02
7.02372789e-01 -1.32573593e+00 7.72453994e-02 -9.87487137e-01
-4.52702165e-01 9.67322111e-01 1.18753266e+00 -1.34848461e-01
7.89908528e-01 8.40182155e-02 4.47524413e-02 -5.49294829e-01
-4.62504387e-01 1.24609753e-01 -2.48180870e-02 7.03645289e-01
5.87806225e-01 5.14759198e-02 -3.06960464e-01 7.01215327e-01
-2.74204239e-02 -1.72567479e-02 4.78516817e-01 4.27259654e-01
-4.67347614e-02 -1.05146182e+00 -3.73933703e-01 7.21580327e-01
-3.70442897e-01 6.90938681e-02 -3.85689974e-01 6.67115808e-01
4.18369800e-01 9.59220231e-01 2.49091879e-01 -5.08069575e-01
1.75379127e-01 -1.96643412e-01 3.84382457e-01 -4.65974361e-01
-2.75379121e-01 -1.59360662e-01 -2.18831941e-01 -7.54370928e-01
-4.40085351e-01 -3.25743616e-01 -1.15081763e+00 -5.96135914e-01
-4.76488143e-01 9.91041586e-02 4.36479777e-01 5.08885860e-01
7.93516755e-01 6.20497465e-01 8.48300219e-01 -6.07683480e-01
-1.93810701e-01 -9.49081957e-01 -8.45766664e-01 6.36911571e-01
-8.51487815e-02 -7.96560287e-01 -5.03330350e-01 -4.65849459e-01] | [7.325470924377441, 5.9987897872924805] |
06fe86a4-760d-4ed7-bb69-14f88333051d | asrtrans-at-semeval-2022-task-5-transformer | null | null | https://aclanthology.org/2022.semeval-1.82 | https://aclanthology.org/2022.semeval-1.82.pdf | ASRtrans at SemEval-2022 Task 5: Transformer-based Models for Meme Classification | Women are frequently targeted online with hate speech and misogyny using tweets, memes, and other forms of communication. This paper describes our system for Task 5 of SemEval-2022: Multimedia Automatic Misogyny Identification (MAMI). We participated in both the sub-tasks, where we used transformer-based architecture to combine features of images and text. We explore models with multi-modal pre-training (VisualBERT) and text-based pre-training (MMBT) while drawing comparative results. We also show how additional training with task-related external data can improve the model performance. We achieved sizable improvements over baseline models and the official evaluation ranked our system 3^{rd} out of 83 teams on the binary classification task (Sub-task A) with an F1 score of 0.761, and 7^{th} out of 48 teams on the multi-label classification task (Sub-task B) with an F1 score of 0.705. | ['Arjun Rao', 'Ailneni Rakshitha Rao'] | null | null | null | null | semeval-naacl-2022-7 | ['meme-classification'] | ['natural-language-processing'] | [ 1.38423458e-01 2.60968149e-01 -1.15618911e-02 -2.92065233e-01
-9.44756150e-01 -5.26383817e-01 1.13153458e+00 1.44127175e-01
-6.73447013e-01 3.86349022e-01 1.01626374e-01 -4.38346015e-03
3.42801541e-01 -3.11523110e-01 -4.63160396e-01 -3.87066185e-01
2.20220283e-01 5.92907131e-01 1.68792754e-01 -1.87209159e-01
2.88155437e-01 -1.47872837e-02 -1.12838078e+00 9.60497975e-01
4.33497459e-01 1.02478456e+00 -4.61849153e-01 1.02296829e+00
1.79259673e-01 1.05567777e+00 -8.18738580e-01 -1.18827343e+00
-3.04275960e-01 -1.90946475e-01 -1.05271041e+00 -1.83318436e-01
1.14857101e+00 -3.16931039e-01 -2.84071237e-01 1.09348238e+00
7.67907858e-01 -9.43455324e-02 1.11544490e+00 -1.28183961e+00
-7.33312845e-01 6.06780887e-01 -8.97753119e-01 2.03807399e-01
7.68633932e-02 -1.76638022e-01 6.70237362e-01 -1.16837478e+00
6.53014719e-01 1.70224476e+00 9.91730332e-01 9.51725423e-01
-1.02348673e+00 -1.07271647e+00 -3.91539693e-01 3.27940404e-01
-1.33870363e+00 -6.84137642e-01 4.07267153e-01 -7.36834526e-01
7.70309091e-01 3.31377238e-01 1.74245797e-02 1.68059015e+00
-2.31454126e-03 9.54429805e-01 1.33668721e+00 -4.00871873e-01
-4.52834368e-01 6.23806894e-01 1.31628051e-01 8.51689398e-01
-6.57523274e-02 -2.64343947e-01 -6.25523984e-01 -1.04835957e-01
1.40537605e-01 -3.84223610e-01 4.27613318e-01 5.21971762e-01
-8.63839686e-01 1.16001177e+00 1.10144243e-01 5.70163727e-01
2.60109901e-01 2.27630541e-01 9.12867665e-01 1.73107088e-01
9.48929608e-01 3.16939950e-01 1.71547920e-01 -1.49191573e-01
-1.21334243e+00 2.25541845e-01 4.55011696e-01 6.73362315e-01
4.79716986e-01 -7.96228424e-02 -5.76710105e-01 1.45089090e+00
1.86890773e-02 8.24244678e-01 4.69610780e-01 -1.02370119e+00
6.12184584e-01 1.80967629e-01 -7.00077638e-02 -1.34773135e+00
-6.55825377e-01 -1.95223495e-01 -7.05928385e-01 8.61731693e-02
6.15211904e-01 -2.93463171e-01 -1.01159954e+00 1.50892091e+00
-6.72573224e-03 5.88418879e-02 -3.06125343e-01 4.30014372e-01
1.28743649e+00 6.99497581e-01 4.65142578e-01 -4.43386920e-02
1.49892282e+00 -1.18092620e+00 -8.02789629e-01 -1.99487239e-01
1.06307507e+00 -9.58125055e-01 8.97949815e-01 4.49558437e-01
-1.00319099e+00 -4.38193291e-01 -8.45960319e-01 -1.23657107e-01
-6.30081713e-01 3.31740469e-01 1.71898305e-02 1.08327270e+00
-1.10238516e+00 4.20550376e-01 -3.35625350e-01 -6.51882350e-01
6.04792655e-01 1.65866196e-01 -5.63669503e-01 7.28612542e-02
-1.10114956e+00 1.20950496e+00 5.74915111e-02 -1.50623754e-01
-1.25185823e+00 -5.81650138e-01 -7.31254995e-01 -4.00312275e-01
1.44259334e-01 7.53500164e-02 1.19019914e+00 -8.86261404e-01
-9.36806202e-01 1.73675013e+00 5.72687872e-02 -3.44517440e-01
6.77015007e-01 -3.03997010e-01 -6.14078522e-01 3.64352494e-01
3.04332674e-01 1.02380407e+00 1.04517424e+00 -1.20819426e+00
-7.08690464e-01 -3.15411955e-01 1.85500353e-03 1.31694600e-01
-9.33799326e-01 8.90598476e-01 -2.31291398e-01 -5.03181398e-01
-6.31182492e-01 -1.10321116e+00 5.12649059e-01 -4.28132772e-01
-6.24336064e-01 -2.71929562e-01 1.21396697e+00 -1.18128157e+00
1.26407552e+00 -2.11194634e+00 9.61898081e-03 -3.55429575e-02
5.92521846e-01 5.38249791e-01 -9.28042307e-02 2.29476109e-01
-5.64191937e-02 6.42881870e-01 1.14122637e-01 -1.02265060e+00
-3.19470726e-02 -1.76798776e-01 1.02076113e-01 5.74342966e-01
9.33762938e-02 7.40714788e-01 -8.73482525e-01 -7.12263644e-01
2.71618622e-03 4.47206765e-01 -2.14352071e-01 4.79500331e-02
1.27387375e-01 2.28001773e-01 -1.93239283e-02 6.05372429e-01
4.93111432e-01 -2.25434154e-01 -1.35345936e-01 -2.33543098e-01
1.12218075e-01 -3.03690225e-01 -4.41906512e-01 1.12140453e+00
-4.91212040e-01 1.22858465e+00 2.19709188e-01 -6.83898687e-01
7.16684520e-01 4.77799147e-01 2.12431654e-01 -8.12668204e-01
4.48615909e-01 2.50450969e-01 -1.20180123e-01 -6.01034343e-01
5.70304632e-01 -1.26746267e-01 -2.60616481e-01 3.21871608e-01
3.54283541e-01 2.63202369e-01 2.01444909e-01 3.25014263e-01
1.08534539e+00 -1.02906823e-01 -8.50044936e-02 -3.10628057e-01
4.92500842e-01 -1.57284185e-01 -3.47799510e-02 8.46219897e-01
-4.69881475e-01 8.11605632e-01 6.74616098e-01 -5.44042468e-01
-1.20077825e+00 -4.55704719e-01 -1.48811251e-01 1.64100993e+00
-1.72353461e-01 -5.32045186e-01 -8.62350106e-01 -9.75862145e-01
-2.24850520e-01 7.34809458e-01 -1.19054544e+00 -2.02166457e-02
-5.38196862e-01 -9.69709575e-01 1.35004044e+00 1.34494662e-01
6.44744277e-01 -1.04073894e+00 -3.10652763e-01 -1.69551209e-01
-5.69586575e-01 -1.31684768e+00 -3.59551162e-01 -7.40491748e-02
-7.08680600e-02 -7.66842902e-01 -1.00939345e+00 -5.43357670e-01
3.32733095e-01 3.97646315e-02 8.86519074e-01 2.65809596e-01
-3.00414741e-01 1.28471181e-01 -4.84347999e-01 -4.34740275e-01
-6.61826372e-01 2.10079789e-01 -8.21048096e-02 2.53061473e-01
4.10806537e-01 -9.55406427e-02 -1.52129099e-01 3.76856506e-01
-7.28582144e-01 2.27179021e-01 6.81814849e-02 8.32259476e-01
-1.60942078e-01 -3.71077836e-01 2.76548028e-01 -1.32168448e+00
2.48460218e-01 -5.75499475e-01 -8.06677341e-02 1.67840481e-01
-2.87010252e-01 -6.43369079e-01 1.56867847e-01 -6.30934000e-01
-8.68894875e-01 -3.12925100e-01 -9.91152152e-02 -6.45623744e-01
-5.57344966e-02 1.08284146e-01 1.84408814e-01 -1.25263751e-01
8.83420527e-01 -2.48973846e-01 -8.10802951e-02 -4.53225762e-01
4.94946204e-02 1.01935887e+00 5.45562506e-01 -4.10819322e-01
7.44294941e-01 3.08848113e-01 -1.62897721e-01 -9.60165381e-01
-1.23773408e+00 -4.57550734e-01 -4.73982453e-01 -6.10922754e-01
1.34091115e+00 -9.83746767e-01 -8.06578338e-01 9.35600340e-01
-1.33465779e+00 -4.75524396e-01 6.12356961e-01 3.46429013e-02
-1.45851955e-01 2.83588588e-01 -9.14767385e-01 -1.09612048e+00
-3.89630765e-01 -9.40721154e-01 1.24717700e+00 9.77705792e-03
-5.16087234e-01 -1.14787734e+00 -7.43556619e-02 1.03951037e+00
3.82892489e-01 5.20176768e-01 7.92520106e-01 -1.20096755e+00
4.54162747e-01 -2.23079324e-01 -7.01515019e-01 3.73993695e-01
-2.92847663e-01 1.50419161e-01 -1.59541118e+00 -3.87166739e-01
-3.56571764e-01 -9.39289749e-01 1.04891145e+00 9.86330137e-02
1.08540750e+00 -3.71067494e-01 -4.71745968e-01 1.96275815e-01
1.06701255e+00 -1.14146419e-01 6.55915320e-01 3.59673351e-01
1.12055647e+00 9.59101319e-01 4.27949756e-01 3.89000118e-01
4.61517185e-01 1.00798059e+00 3.96984041e-01 -1.56633437e-01
-4.51559812e-01 -1.99362904e-01 4.76876885e-01 4.89341706e-01
-7.46410787e-02 -3.09310049e-01 -1.32246828e+00 7.68881619e-01
-1.75981057e+00 -1.14820588e+00 -4.33318764e-01 1.79059339e+00
7.81315982e-01 1.37503939e-02 6.22317791e-01 -3.97948064e-02
1.08457959e+00 2.07276523e-01 1.17770210e-01 -7.70023108e-01
-3.88053171e-02 -1.00501820e-01 6.04808629e-01 6.56616509e-01
-1.67149496e+00 1.18276417e+00 6.23350811e+00 1.52194023e+00
-1.01559961e+00 7.60280550e-01 1.11084199e+00 -1.67181104e-01
1.10249877e-01 -5.45582771e-01 -1.14930511e+00 7.42919624e-01
1.48906195e+00 3.29710662e-01 7.81926960e-02 4.95548517e-01
-2.70524442e-01 -8.24334025e-02 -9.18597043e-01 1.08522928e+00
8.00695658e-01 -1.20789170e+00 -1.76834211e-01 2.17563033e-01
7.33459711e-01 1.84604600e-02 3.92149895e-01 5.65963507e-01
1.94695070e-01 -1.45553696e+00 9.70497727e-01 1.94989696e-01
1.13649178e+00 -5.86755633e-01 9.09765482e-01 2.99181134e-01
-5.89338601e-01 -1.02011502e-01 -1.98184401e-01 7.34979138e-02
-1.61365524e-01 2.47216403e-01 -9.02575016e-01 3.56936574e-01
9.73340631e-01 7.19870985e-01 -1.07789135e+00 6.16653740e-01
9.38969851e-03 8.25796604e-01 -2.66755581e-01 -1.47651747e-01
3.80180031e-01 3.83347213e-01 3.49375486e-01 1.91280174e+00
6.81674331e-02 -3.88448149e-01 -9.05188452e-03 2.41380095e-01
-2.51075894e-01 1.17026992e-01 -6.04424238e-01 -1.62782624e-01
3.55348796e-01 1.61396372e+00 -6.60710990e-01 -3.98501575e-01
-3.93561125e-01 1.01957726e+00 3.22349399e-01 7.67373713e-03
-9.98854876e-01 -3.35558742e-01 -8.32417235e-03 1.54564932e-01
-3.16912271e-02 2.56867588e-01 -4.10492480e-01 -8.15225422e-01
-5.54291189e-01 -9.71469402e-01 5.46031415e-01 -6.46518111e-01
-1.30406892e+00 5.08729935e-01 1.99928984e-01 -8.62432897e-01
-2.75363982e-01 -6.97177231e-01 -3.65016043e-01 6.33204758e-01
-1.17509925e+00 -1.80666375e+00 -1.78863913e-01 3.65081966e-01
3.44267607e-01 -4.99879241e-01 9.10962343e-01 7.40917444e-01
-6.79012775e-01 1.17530644e+00 -4.21989374e-02 5.44308066e-01
1.18400681e+00 -1.10411429e+00 -7.34561011e-02 4.85519469e-01
-6.92300275e-02 -6.69337437e-02 7.30585098e-01 -6.44850969e-01
-5.55429220e-01 -1.23539102e+00 1.23498452e+00 -9.58993018e-01
8.65113199e-01 -5.86148143e-01 -5.64201713e-01 6.80449009e-01
3.87226224e-01 -2.78420269e-01 9.27384496e-01 2.35397443e-01
-7.76173472e-01 2.04731286e-01 -1.36188722e+00 4.31778461e-01
8.00277412e-01 -6.72743022e-01 -1.32857054e-01 6.98310614e-01
3.60978276e-01 -5.49601197e-01 -8.81382287e-01 2.64071256e-01
7.24385381e-01 -7.89095044e-01 7.71534979e-01 -5.74476302e-01
1.19453108e+00 2.51834393e-01 -2.75997847e-01 -9.19446826e-01
-2.32569501e-01 -4.13877428e-01 1.05288200e-01 1.62366152e+00
6.32933676e-01 -3.67932743e-03 5.88394284e-01 3.47165585e-01
4.70477417e-02 -3.02806169e-01 -1.00230038e+00 -5.54251671e-01
4.72119331e-01 -2.58755356e-01 -1.78750873e-01 1.28370976e+00
-9.58253443e-02 6.42370045e-01 -1.06956685e+00 -1.62949026e-01
6.30926847e-01 -5.82171917e-01 7.39529431e-01 -9.79795754e-01
3.71426828e-02 -5.53592324e-01 -4.15355861e-01 -1.25278607e-01
3.74880463e-01 -9.52788889e-01 -2.73823589e-01 -1.06944299e+00
8.91818106e-01 1.69815812e-02 -4.44184206e-02 8.79887283e-01
-1.46098226e-01 1.18890965e+00 5.77130377e-01 2.88223416e-01
-1.02836204e+00 4.33029868e-02 9.72469032e-01 -4.02564496e-01
5.12745321e-01 -2.87721366e-01 -6.48627460e-01 7.87694991e-01
5.92106998e-01 -6.95087790e-01 -4.07700688e-02 -4.92281318e-01
2.48468861e-01 -2.21269667e-01 3.80508125e-01 -8.17979872e-01
-2.01933160e-01 1.31730169e-01 5.55934906e-01 -6.12709820e-01
7.41058230e-01 -4.19545144e-01 -1.37711003e-01 4.87185687e-01
-4.87763405e-01 -1.05666816e-01 3.30152631e-01 6.55132830e-02
2.36517447e-03 -6.06608927e-01 1.04444826e+00 -3.75856794e-02
-4.50789094e-01 9.55117401e-04 -5.79496443e-01 2.99660832e-01
9.39328730e-01 -8.65599066e-02 -9.30221140e-01 -5.55275202e-01
-8.42111588e-01 2.42678374e-01 2.47343004e-01 4.67579663e-01
2.73973644e-01 -1.39086306e+00 -1.13204384e+00 -3.13391954e-01
2.21095413e-01 -8.85601640e-01 1.29470453e-01 8.65293145e-01
-4.93854880e-01 2.18331248e-01 -4.89025861e-01 -5.08783877e-01
-1.76862764e+00 6.08425327e-02 2.40841940e-01 -4.82194006e-01
-1.88621003e-02 1.18090129e+00 1.62841380e-01 -5.19745827e-01
2.56084859e-01 7.65794396e-01 -6.86190963e-01 4.79504526e-01
9.39172506e-01 6.95081055e-01 -8.59816149e-02 -1.33240032e+00
-4.45470989e-01 4.42252994e-01 -4.63501126e-01 -3.54410559e-01
1.14537692e+00 -9.83684883e-02 -2.24574283e-01 5.32854497e-01
1.59422386e+00 -4.69972529e-02 -6.12475216e-01 -1.12363718e-01
3.00640576e-02 -3.85603994e-01 -4.20091264e-02 -1.05948806e+00
-7.91638076e-01 1.18891096e+00 7.17710972e-01 3.02510828e-01
5.31173527e-01 1.71027761e-02 5.87819755e-01 1.58662319e-01
-8.41671079e-02 -1.33854949e+00 5.92557549e-01 7.36991048e-01
7.76742578e-01 -1.47228336e+00 -8.11640173e-02 -3.56044143e-01
-9.43060756e-01 9.29219067e-01 7.93151140e-01 1.54742435e-01
4.21482861e-01 -1.48332790e-02 2.85818279e-01 -3.88583899e-01
-4.68479931e-01 -3.39396223e-02 4.44308788e-01 6.54273748e-01
5.88605404e-01 1.60540968e-01 -2.78528243e-01 4.86129701e-01
-2.55787283e-01 -4.23552185e-01 4.32760209e-01 4.81143713e-01
-3.63996446e-01 -7.37020373e-01 -4.58182842e-01 4.59563076e-01
-1.11861992e+00 -2.58278642e-02 -6.94520414e-01 7.90683568e-01
3.32034767e-01 1.17301512e+00 5.72871193e-02 -8.28833818e-01
8.39251000e-03 2.08283782e-01 4.32774872e-01 -6.64262831e-01
-9.46660519e-01 1.18365951e-01 8.69885027e-01 -3.51078242e-01
-4.65106577e-01 -6.55928195e-01 -6.90962374e-01 -6.98536575e-01
-1.40581623e-01 -2.43658036e-01 7.48579264e-01 1.01043522e+00
-1.40237257e-01 2.36688584e-01 4.65131611e-01 -8.80851805e-01
-2.44647801e-01 -1.39633822e+00 -4.36154515e-01 7.19710231e-01
2.19047189e-01 -6.99790835e-01 -1.77848279e-01 1.44369557e-01] | [8.577978134155273, 10.61656379699707] |
97fca98e-e96b-4659-81ed-33b5d339a9e5 | the-nlms-algorithm-with-time-variant-optimum | 1411.4834 | null | http://arxiv.org/abs/1411.4834v1 | http://arxiv.org/pdf/1411.4834v1.pdf | The NLMS algorithm with time-variant optimum stepsize derived from a Bayesian network perspective | In this article, we derive a new stepsize adaptation for the normalized least
mean square algorithm (NLMS) by describing the task of linear acoustic echo
cancellation from a Bayesian network perspective. Similar to the well-known
Kalman filter equations, we model the acoustic wave propagation from the
loudspeaker to the microphone by a latent state vector and define a linear
observation equation (to model the relation between the state vector and the
observation) as well as a linear process equation (to model the temporal
progress of the state vector). Based on additional assumptions on the
statistics of the random variables in observation and process equation, we
apply the expectation-maximization (EM) algorithm to derive an NLMS-like filter
adaptation. By exploiting the conditional independence rules for Bayesian
networks, we reveal that the resulting EM-NLMS algorithm has a stepsize update
equivalent to the optimal-stepsize calculation proposed by Yamamoto and
Kitayama in 1982, which has been adopted in many textbooks. As main difference,
the instantaneous stepsize value is estimated in the M step of the EM algorithm
(instead of being approximated by artificially extending the acoustic echo
path). The EM-NLMS algorithm is experimentally verified for synthesized
scenarios with both, white noise and male speech as input signal. | ['Walter Kellermann', 'Roland Maas', 'Christian Huemmer'] | 2014-11-18 | null | null | null | null | ['acoustic-echo-cancellation', 'acoustic-echo-cancellation'] | ['medical', 'speech'] | [ 3.37621421e-01 1.35629117e-01 4.00768071e-01 -1.63705930e-01
-3.02102000e-01 -3.11243117e-01 5.60494602e-01 -3.41730297e-01
-5.97470820e-01 3.78575593e-01 1.69721887e-01 -2.65948713e-01
-4.03945893e-01 -3.83859277e-01 -4.27041650e-01 -9.70750809e-01
2.43308619e-02 2.44930964e-02 1.55531600e-01 5.95998615e-02
1.10006660e-01 3.07821780e-01 -1.07447433e+00 -4.96375680e-01
5.66859543e-01 7.52975941e-01 3.20022196e-01 1.14487886e+00
-1.74055308e-01 5.44628978e-01 -6.13522232e-01 -2.34683976e-01
1.90521151e-01 -7.49123096e-01 -2.18490779e-01 7.10618198e-02
4.52914834e-02 -1.76372170e-01 -6.17587388e-01 1.21470857e+00
5.92908144e-01 5.29077768e-01 7.57906497e-01 -7.05667675e-01
-4.46943231e-02 5.00578821e-01 -2.21847251e-01 1.08215243e-01
2.94549584e-01 -3.09749633e-01 4.87195015e-01 -8.36827457e-01
1.91454858e-01 1.31213033e+00 6.75451517e-01 5.27818501e-01
-1.07322001e+00 -7.30632305e-01 9.60353110e-03 2.36684546e-01
-1.38829863e+00 -6.77492559e-01 6.66471541e-01 -4.23180580e-01
5.52316368e-01 2.48977304e-01 5.47153115e-01 8.12345207e-01
3.43622655e-01 5.78623712e-01 1.01870620e+00 -8.22747350e-01
3.98757815e-01 8.32633898e-02 4.66803044e-01 4.27336603e-01
6.51031211e-02 4.72629994e-01 -7.91962981e-01 -2.21128494e-01
6.40783906e-01 -2.91339397e-01 -2.80171543e-01 -2.65170366e-01
-8.07254493e-01 4.12427634e-01 -2.32199013e-01 3.19580793e-01
-4.96876419e-01 2.70039350e-01 -1.58572286e-01 3.30726296e-01
1.60734862e-01 -4.06713970e-02 -2.80588776e-01 4.04490903e-03
-1.07094574e+00 1.09158702e-01 1.28010488e+00 7.55911231e-01
5.12898147e-01 7.28526354e-01 1.08223267e-01 5.84631026e-01
1.00072491e+00 1.10581684e+00 4.22052890e-01 -8.66391897e-01
1.65916264e-01 -3.63110363e-01 2.73423433e-01 -9.25827503e-01
-3.39941859e-01 -1.04065180e+00 -9.42862034e-01 1.55987233e-01
4.97975320e-01 -5.95238686e-01 -8.20659697e-01 1.82595301e+00
3.10484827e-01 7.45868206e-01 2.05165610e-01 6.05424106e-01
3.19819391e-01 9.61643815e-01 -2.27421790e-01 -9.37935114e-01
8.94430339e-01 -5.61611295e-01 -1.47724628e+00 -2.68384993e-01
2.27891073e-01 -7.95546889e-01 2.87640274e-01 7.12961853e-01
-1.07232988e+00 -5.55717826e-01 -1.13606656e+00 7.44704306e-01
-1.88289396e-02 -1.48020715e-01 1.58653542e-01 8.41749370e-01
-8.49206269e-01 4.63310301e-01 -1.03433967e+00 -1.45708978e-01
-6.84512019e-01 2.23087624e-01 -1.25172287e-01 2.76321501e-01
-1.15545511e+00 9.04363155e-01 3.03320382e-02 7.17395902e-01
-9.02819753e-01 -4.34817314e-01 -6.27935767e-01 2.18791515e-02
2.59691030e-01 -5.00611246e-01 1.54399896e+00 -7.00261474e-01
-2.26757216e+00 -1.15937665e-01 -7.00828850e-01 -2.56140649e-01
2.17367962e-01 -3.82073641e-01 -6.94920540e-01 -2.05860943e-01
-3.17053080e-01 -2.96178013e-01 1.18460310e+00 -1.21648026e+00
-4.77167577e-01 -3.77988480e-02 -5.72753310e-01 2.08813682e-01
-8.97513181e-02 -5.25652915e-02 -3.80402714e-01 -5.23810625e-01
7.68752873e-01 -8.31204474e-01 -5.28661907e-01 -4.93147075e-01
-3.15307170e-01 2.95399964e-01 3.26437175e-01 -8.10731888e-01
1.57992351e+00 -2.35694337e+00 2.13068172e-01 6.48869574e-01
-1.81412399e-01 2.08115578e-01 8.37733969e-02 6.63257778e-01
-1.55950114e-01 -4.79869962e-01 -3.54190856e-01 -4.08824235e-01
1.55433430e-03 1.63128316e-01 -4.16150272e-01 6.53367281e-01
-5.73960245e-01 2.34825909e-01 -8.11769068e-01 -1.80352345e-01
2.92198151e-01 5.21381557e-01 -3.22118580e-01 4.53636616e-01
3.48620355e-01 5.90603888e-01 -2.45039731e-01 -9.06915888e-02
6.97477341e-01 3.30491602e-01 2.03562066e-01 -2.49763086e-01
-4.32661384e-01 7.71482289e-02 -2.04696131e+00 1.31150615e+00
-5.22983909e-01 6.97793663e-01 6.51478529e-01 -8.67133737e-01
7.75786638e-01 8.39545310e-01 1.24102376e-01 -2.39863873e-01
6.27242252e-02 9.20545086e-02 2.49735251e-01 -4.50188518e-01
1.96305037e-01 -4.04125601e-01 1.70977250e-01 3.61581057e-01
3.16627681e-01 -1.50966987e-01 3.36665735e-02 1.12236708e-01
8.21364701e-01 -4.33660112e-02 5.09436727e-01 -1.42801195e-01
7.58743703e-01 -6.21313810e-01 6.42114162e-01 1.13191640e+00
9.20051592e-04 1.47207573e-01 -4.59613204e-02 1.65466040e-01
-5.89219511e-01 -1.34081709e+00 -1.22046404e-01 8.20692956e-01
-1.30651638e-01 -2.32402265e-01 -7.10646629e-01 1.52347431e-01
-4.04732257e-01 9.37448144e-01 -1.91837579e-01 -1.31781921e-01
-5.16710818e-01 -7.42344320e-01 3.48882854e-01 5.61994798e-02
2.49455363e-01 -5.45719206e-01 -8.88901651e-02 6.01821065e-01
1.32326605e-02 -9.55957651e-01 -2.79614210e-01 3.49247605e-01
-7.53990114e-01 -4.43261683e-01 -6.45350277e-01 -4.50413078e-01
6.43281639e-01 -9.30174440e-02 4.14364040e-01 -4.67705309e-01
6.92264140e-02 7.58591294e-01 2.52229255e-02 -6.42737746e-01
-6.88505352e-01 -5.56791365e-01 5.21761239e-01 4.04600471e-01
2.93420758e-02 -7.17035711e-01 -3.58917445e-01 2.03996703e-01
-6.70273185e-01 -5.37098199e-02 6.20688856e-01 5.36730409e-01
2.07832962e-01 3.78965408e-01 2.84423262e-01 -3.36800009e-01
6.35978520e-01 -2.04966784e-01 -9.43985641e-01 8.27973485e-02
-6.35091245e-01 2.64172375e-01 3.14997882e-01 -6.43803477e-01
-1.52637792e+00 1.47514552e-01 -3.77555788e-01 4.84703332e-02
-2.89349675e-01 4.33726788e-01 -2.79418886e-01 -5.48046827e-02
2.63096213e-01 5.85384905e-01 -1.26925677e-01 -6.74298465e-01
5.41989863e-01 7.02293217e-01 8.09096098e-01 -2.16061249e-01
1.09390271e+00 3.11870635e-01 2.73204237e-01 -1.18577671e+00
-6.12980187e-01 -6.01871431e-01 -6.12036765e-01 -3.93339157e-01
8.29522848e-01 -6.92491770e-01 -8.38671744e-01 7.74882376e-01
-1.17354035e+00 1.24434363e-02 -1.49150833e-01 1.30751586e+00
-4.56764072e-01 4.96418446e-01 -5.97119570e-01 -1.57065117e+00
-8.11097957e-03 -6.41053855e-01 3.04725558e-01 2.02267408e-01
-1.10380001e-01 -1.16668010e+00 4.78247643e-01 -3.42874706e-01
4.56431061e-01 -4.37302023e-01 5.18262327e-01 -4.63280767e-01
-4.16567266e-01 -4.48902190e-01 4.20893222e-01 5.69426537e-01
-1.02605648e-01 -1.29609495e-01 -9.25333858e-01 -1.08299196e-01
7.78162777e-01 6.91406190e-01 5.93259037e-01 7.48648047e-01
1.49306849e-01 -1.73065200e-01 -3.45337600e-01 5.16259193e-01
1.32035565e+00 6.07790530e-01 3.06425095e-01 -1.48402944e-01
3.46417844e-01 5.23505867e-01 2.30066150e-01 4.82319534e-01
-3.18467170e-02 4.53949898e-01 8.29009712e-02 1.49722517e-01
1.00640304e-01 -2.01560348e-01 5.56243539e-01 1.72425163e+00
2.15139925e-01 -3.73877585e-01 -5.72803497e-01 2.80865103e-01
-1.61944568e+00 -9.01181519e-01 -3.52451950e-01 2.49754977e+00
5.68497777e-01 1.80265568e-02 -4.73488301e-01 2.74802327e-01
6.01872444e-01 -2.16720393e-03 -2.11528420e-01 -3.96772265e-01
-3.62109323e-03 7.04591861e-03 6.81986034e-01 1.25257099e+00
-7.25468755e-01 5.61554551e-01 7.55260611e+00 8.27892780e-01
-8.70705068e-01 2.63013482e-01 -3.25040728e-01 1.59393802e-01
-6.05089217e-02 1.90808967e-01 -1.02399874e+00 3.17621291e-01
1.23785460e+00 -1.46273464e-01 5.14279842e-01 2.60279447e-01
6.18489206e-01 -2.53876776e-01 -8.06363404e-01 9.09334958e-01
1.92832425e-02 -4.88550007e-01 -2.95229435e-01 -2.85322163e-02
5.15922308e-01 -3.29121739e-01 -3.61702815e-02 4.54505645e-02
3.79085511e-01 -5.10555208e-01 7.66102254e-01 1.07049787e+00
7.62070790e-02 -4.07757103e-01 5.56472957e-01 6.89308465e-01
-1.14539135e+00 -2.12588921e-01 -2.38282099e-01 -3.98365408e-01
7.92165697e-01 9.36384380e-01 -7.15530455e-01 4.87512767e-01
2.98790455e-01 1.31984666e-01 2.87786782e-01 1.29654551e+00
-5.16310513e-01 1.12842870e+00 -6.02285445e-01 -6.35362193e-02
7.07120374e-02 -5.01170456e-01 1.09285998e+00 1.32014084e+00
6.23919606e-01 1.75418437e-01 -3.18925083e-01 6.30249858e-01
5.56228817e-01 5.45164235e-02 -1.94767103e-01 1.53154001e-01
4.75044161e-01 1.02217829e+00 -2.98140317e-01 -2.37965360e-01
-3.55178744e-01 7.61073112e-01 -4.78113204e-01 9.48839903e-01
-3.68407726e-01 -4.86664832e-01 4.17996585e-01 -2.39737824e-01
2.71608442e-01 -5.76600432e-01 8.17965791e-02 -7.83936977e-01
-4.41798836e-01 -4.62658346e-01 -1.15146011e-01 -7.07617640e-01
-8.71933997e-01 4.28217262e-01 2.85033882e-01 -1.16369987e+00
-6.87047005e-01 -5.21983683e-01 -5.91579974e-01 1.23785686e+00
-1.00781512e+00 -5.05505025e-01 2.45319188e-01 3.76656085e-01
3.27354699e-01 -1.77393690e-01 7.96041250e-01 2.11682111e-01
-6.23882174e-01 2.75478154e-01 7.46598959e-01 -6.22929931e-02
5.34638166e-01 -1.07040417e+00 1.73575833e-01 1.20114994e+00
8.48585516e-02 9.66601491e-01 1.44306874e+00 -6.08823955e-01
-1.32112741e+00 -4.35338825e-01 1.04960144e+00 -1.09775431e-01
7.77112782e-01 -3.71084303e-01 -7.11382031e-01 5.60364366e-01
3.63090694e-01 -5.55524170e-01 6.57063901e-01 -1.01637445e-01
8.58894885e-02 -2.00293034e-01 -6.27546430e-01 5.08140683e-01
6.40602171e-01 -4.32440609e-01 -8.00427318e-01 -2.21038871e-02
4.33857083e-01 -3.31938833e-01 -6.20441556e-01 2.68433362e-01
7.70892739e-01 -6.90597892e-01 7.94855893e-01 -1.48947611e-01
-5.15212953e-01 -5.75921178e-01 -3.04125965e-01 -1.41478062e+00
-5.90789676e-01 -1.05348837e+00 -3.50133538e-01 1.25677323e+00
3.96890193e-01 -7.42510200e-01 3.61261606e-01 3.44695866e-01
-1.17412418e-01 -1.60741732e-01 -1.00808680e+00 -8.09594631e-01
-2.43643746e-01 -8.76785517e-01 9.90089029e-02 2.36973003e-01
4.35602032e-02 3.23890060e-01 -7.68110394e-01 8.06632936e-01
1.02920163e+00 -4.04102534e-01 5.44356644e-01 -1.08594477e+00
-8.01173747e-01 -2.44321778e-01 -1.46544710e-01 -1.73327518e+00
1.15624115e-01 -2.65146822e-01 6.21235847e-01 -1.41315401e+00
-1.36909604e-01 7.01352432e-02 -4.74278748e-01 -3.23517323e-01
-7.22041279e-02 -2.64172524e-01 2.19280884e-01 -1.61443781e-02
-2.24829495e-01 3.91745716e-01 9.14590359e-01 2.83375800e-01
-3.08673918e-01 7.09824145e-01 -1.32670939e-01 1.03874052e+00
2.99226344e-01 -7.41461754e-01 -3.99768680e-01 -2.75724679e-01
3.11413229e-01 4.59164679e-01 5.36536276e-02 -1.10026646e+00
7.25654185e-01 9.16463062e-02 6.96696341e-02 -7.10916817e-01
6.59877717e-01 -9.39564466e-01 5.11663795e-01 5.83485126e-01
-3.95264030e-01 -2.77380109e-01 -5.22102714e-02 8.20455551e-01
-2.25391835e-01 -7.28549659e-01 4.86859858e-01 8.26588646e-02
-4.84829277e-01 -9.75992307e-02 -1.22461033e+00 -5.44326603e-01
3.73232543e-01 -2.58181654e-02 3.03291649e-01 -9.48194981e-01
-1.21851361e+00 -1.80724069e-01 -3.30595165e-01 -9.41127166e-02
4.84295547e-01 -9.90469038e-01 -5.82012296e-01 3.99633199e-01
-5.95667839e-01 -4.98185724e-01 3.90942216e-01 9.95224774e-01
-7.65404329e-02 4.98131484e-01 4.43523020e-01 -5.31282544e-01
-1.19863832e+00 2.75283456e-01 5.21854460e-01 -8.28920901e-02
-1.63942248e-01 8.93957794e-01 2.79603928e-01 -4.71643358e-01
5.02071083e-01 -9.93110090e-02 -2.37802818e-01 -2.85852671e-01
6.63440943e-01 7.77059495e-01 -1.89956397e-01 -6.09287381e-01
-1.27016589e-01 8.80915165e-01 4.64285851e-01 -9.61472511e-01
8.22813749e-01 -7.91986942e-01 -1.68213680e-01 9.86258924e-01
9.74778891e-01 4.22694713e-01 -9.58770752e-01 -5.48175812e-01
-2.47978698e-03 -1.40992005e-03 2.97538877e-01 -6.07915521e-01
-5.48909783e-01 8.62074971e-01 9.86101151e-01 5.58005311e-02
9.70490336e-01 -3.42737466e-01 3.53987664e-01 4.96537358e-01
2.58106679e-01 -8.33171606e-01 -3.75963300e-01 8.00826132e-01
5.57778418e-01 -5.16953766e-01 -2.22986564e-01 -2.91930318e-01
-6.10286891e-02 1.04903698e+00 -2.97715105e-02 5.14870360e-02
1.28295410e+00 7.22375214e-01 3.44040036e-01 3.75752181e-01
-6.00735784e-01 -2.89656557e-02 3.23589921e-01 6.38421059e-01
3.34894091e-01 -1.62213165e-02 -3.73134285e-01 5.95829725e-01
-1.57174617e-01 -1.32906049e-01 5.51614165e-01 6.32759869e-01
-7.52410173e-01 -1.04194140e+00 -8.17586124e-01 -9.07185953e-03
-3.84966105e-01 -2.01345742e-01 1.95525065e-01 2.46046886e-01
-1.46521673e-01 1.36866283e+00 2.07187589e-02 -3.01145345e-01
3.99093330e-01 3.12119931e-01 2.94671595e-01 -2.92473763e-01
-7.72123858e-02 7.14360356e-01 -1.41996816e-01 -2.54988819e-01
-3.27415079e-01 -7.69357681e-01 -1.22050750e+00 -6.33427349e-04
-6.45914733e-01 5.67979038e-01 1.14308763e+00 1.22113204e+00
-1.20185606e-01 6.87520325e-01 4.97407407e-01 -8.27470005e-01
-8.13696623e-01 -1.26150787e+00 -8.40826750e-01 -2.40253806e-01
4.34377551e-01 -3.15289438e-01 -8.80116224e-01 6.43206686e-02] | [15.206494331359863, 5.694552898406982] |
1d1fc88f-f68c-4a3a-8964-cf05fd1c520b | overview-of-the-nlp-tea-2015-shared-task-for | null | null | https://aclanthology.org/W15-4401 | https://aclanthology.org/W15-4401.pdf | Overview of the NLP-TEA 2015 Shared Task for Chinese Grammatical Error Diagnosis | null | ['Li-Ping Chang', 'Liang-Chih Yu', 'Lung-Hao Lee'] | 2015-07-01 | null | null | null | ws-2015-7 | ['grammatical-error-detection'] | ['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] | [-7.274957656860352, 3.84256649017334] |
1a35bb12-fd98-402b-b13a-0aebbc5c47ad | openbrand-open-brand-value-extraction-from | null | null | https://aclanthology.org/2022.ecnlp-1.19 | https://aclanthology.org/2022.ecnlp-1.19.pdf | OpenBrand: Open Brand Value Extraction from Product Descriptions | Extracting attribute-value information from unstructured product descriptions continue to be of a vital importance in e-commerce applications. One of the most important product attributes is the brand which highly influences costumers’ purchasing behaviour. Thus, it is crucial to accurately extract brand information dealing with the main challenge of discovering new brand names. Under the open world assumption, several approaches have adopted deep learning models to extract attribute-values using sequence tagging paradigm. However, they did not employ finer grained data representations such as character level embeddings which improve generalizability. In this paper, we introduce OpenBrand, a novel approach for discovering brand names. OpenBrand is a BiLSTM-CRF-Attention model with embeddings at different granularities. Such embeddings are learned using CNN and LSTM architectures to provide more accurate representations. We further propose a new dataset for brand value extraction, with a very challenging task on zero-shot extraction. We have tested our approach, through extensive experiments, and shown that it outperforms state-of-the-art models in brand name discovery. | ['Johann Gamper', 'Mouna Kacimi', 'Kassem Sabeh'] | null | openbrand-open-brand-value-extraction-from-1 | https://aclanthology.org/2022.ecnlp-1.19/ | https://aclanthology.org/2022.ecnlp-1.19/ | acl-2022-5 | ['attribute-value-extraction'] | ['natural-language-processing'] | [ 3.39927152e-02 7.82148987e-02 -6.26922965e-01 -6.27154231e-01
-7.72920310e-01 -6.54258072e-01 4.04766917e-01 4.99608129e-01
-5.32400370e-01 6.38361633e-01 3.70086491e-01 -2.08577095e-03
-1.74623691e-02 -1.08531976e+00 -5.69397926e-01 -5.47349036e-01
-1.17797472e-01 6.60642147e-01 -1.88166693e-01 -4.44277078e-01
3.56227696e-01 1.91596627e-01 -1.59224474e+00 3.37745041e-01
4.85951900e-01 1.40184331e+00 1.06343608e-02 2.55780607e-01
-7.20733285e-01 6.06996715e-01 -3.73109132e-01 -1.04582429e+00
3.38100791e-01 7.42503554e-02 -8.84109795e-01 7.15132058e-02
7.85252526e-02 -3.67582023e-01 -1.78816877e-02 1.29263973e+00
3.26477766e-01 5.55164032e-02 5.27844191e-01 -9.81961370e-01
-1.19620156e+00 1.21586931e+00 -3.58420819e-01 5.99385947e-02
1.07161619e-01 -6.92096949e-02 1.74706924e+00 -5.31214595e-01
5.76566756e-01 1.08550012e+00 6.36497259e-01 3.26735735e-01
-1.21548557e+00 -5.93047321e-01 4.24743444e-01 3.69369566e-01
-1.26505589e+00 -2.18838722e-01 7.41690636e-01 -3.20803285e-01
1.06351995e+00 -6.02835417e-02 7.19263196e-01 1.39263260e+00
1.14833698e-01 1.03126884e+00 8.87497842e-01 -2.56475896e-01
1.78152427e-01 4.00016069e-01 1.86128095e-01 2.86808580e-01
4.31358933e-01 -2.08354909e-02 -2.49725014e-01 6.51079789e-02
7.13918924e-01 2.62533754e-01 3.33084404e-01 -3.25694352e-01
-1.23903084e+00 1.22165322e+00 5.54604769e-01 5.86262226e-01
-6.22979462e-01 2.08537504e-01 4.85566407e-01 1.92775548e-01
5.89868844e-01 6.88641965e-01 -8.07674289e-01 -2.82494873e-01
-5.74481010e-01 9.15005431e-02 9.69782352e-01 1.29404879e+00
5.69643319e-01 -1.29066214e-01 -5.91779016e-02 8.45374525e-01
4.12529945e-01 1.27952769e-01 6.51838541e-01 -4.35380220e-01
1.29855111e-01 3.91487271e-01 1.04506396e-01 -7.59014308e-01
-3.35461497e-01 -9.50257421e-01 -5.50890982e-01 -2.45282963e-01
2.25637510e-01 1.62933081e-01 -8.97350132e-01 1.50196886e+00
4.12827618e-02 2.41977312e-02 2.84114573e-02 7.07085252e-01
8.83618832e-01 4.24824536e-01 5.34072101e-01 1.21821143e-01
1.86056936e+00 -8.77980113e-01 -9.21407938e-01 -2.58363843e-01
6.46038234e-01 -9.38562274e-01 7.08603621e-01 4.19034332e-01
-7.44380236e-01 -5.80574632e-01 -1.08439028e+00 -4.35076654e-02
-1.01921654e+00 -3.37522089e-01 1.02750015e+00 5.23613811e-01
-4.94325995e-01 7.54247129e-01 -3.44539464e-01 -2.75061309e-01
6.77271605e-01 3.78244817e-01 -1.23441733e-01 -1.29186109e-01
-1.67832530e+00 9.75917518e-01 5.81343532e-01 3.78563739e-02
-5.88600576e-01 -3.76523495e-01 -1.07775080e+00 3.03246647e-01
4.82500315e-01 -6.51346982e-01 1.58605254e+00 -9.57092524e-01
-1.16562235e+00 7.61306167e-01 1.85389727e-01 -6.81858540e-01
-1.79725394e-01 -2.74610430e-01 -8.84752393e-01 -3.99343699e-01
3.85680944e-01 7.18119323e-01 7.81134486e-01 -9.85795379e-01
-7.92419732e-01 -4.13574219e-01 1.51550069e-01 -2.83661246e-01
-4.21581537e-01 -7.38115311e-02 -5.22150770e-02 -6.41344368e-01
-1.30415186e-01 -6.91013277e-01 -3.80996764e-01 -4.97903556e-01
-4.80043977e-01 -6.14030898e-01 -4.80659045e-02 -5.36649048e-01
1.18898165e+00 -2.20174646e+00 -3.80631983e-01 1.88261960e-02
3.25207651e-01 2.28182241e-01 -3.65021408e-01 4.23068345e-01
4.50099669e-02 2.36229464e-01 6.10895902e-02 -1.23579912e-01
6.17715180e-01 3.30603093e-01 -1.75832972e-01 -5.45185544e-02
5.86709797e-01 1.24596548e+00 -9.87577558e-01 -4.17150140e-01
-3.73104922e-02 6.40857100e-01 -3.90562296e-01 7.17393607e-02
-5.93389034e-01 -3.60052921e-02 -6.38763487e-01 9.02673185e-01
7.96059608e-01 -2.37454221e-01 1.00513987e-01 -5.22966087e-01
-6.77141994e-02 3.22920740e-01 -1.03912091e+00 1.69182098e+00
-5.52350700e-01 1.79003373e-01 -4.32872653e-01 -1.04417145e+00
1.08199203e+00 1.71574697e-01 5.07658541e-01 -1.00085473e+00
4.81099725e-01 2.53838480e-01 -4.40305546e-02 -5.00193417e-01
8.62683952e-01 -4.46045578e-01 -5.33892453e-01 3.83268995e-03
3.35530758e-01 4.72381979e-01 2.75342703e-01 -9.30065066e-02
8.10466588e-01 3.17547470e-01 6.02447212e-01 -5.33323288e-02
2.58850753e-01 7.82037824e-02 6.06603920e-01 4.28174198e-01
-2.97782570e-01 2.89392352e-01 4.26103622e-01 -7.57876635e-01
-1.06827641e+00 -9.81703401e-01 -3.86488408e-01 1.22437263e+00
3.12780663e-02 -4.04517531e-01 -5.45652688e-01 -6.29409730e-01
1.73917606e-01 1.04014897e+00 -6.51997089e-01 -9.44545567e-02
-3.23752403e-01 -5.90966523e-01 1.73066571e-01 6.59833133e-01
4.05231148e-01 -1.13873410e+00 -1.59119174e-01 8.64576995e-01
-6.13025576e-02 -1.29267526e+00 -4.68322307e-01 6.13760829e-01
-6.45746589e-01 -7.75651455e-01 -5.74258447e-01 -1.03040075e+00
1.42355725e-01 -1.63891643e-01 1.56346178e+00 -4.89481598e-01
-1.54495269e-01 -1.29534706e-01 -6.18600726e-01 -4.21643078e-01
-1.58896893e-01 6.28165662e-01 -3.63406867e-01 3.33327651e-01
1.35841477e+00 -4.63817894e-01 -4.47800070e-01 1.90182701e-01
-7.28748024e-01 -2.99848944e-01 1.13984215e+00 7.04278290e-01
8.06034029e-01 7.92562664e-02 8.97552133e-01 -1.25250661e+00
6.65893137e-01 -6.34348333e-01 -3.91129673e-01 -4.26389165e-02
-8.69367301e-01 5.71966469e-01 5.41710258e-01 -4.58760470e-01
-8.39742243e-01 1.06019758e-01 -8.03259432e-01 1.32509664e-01
-3.81908774e-01 6.06568992e-01 -3.62725616e-01 4.65047479e-01
6.66570663e-02 1.67513072e-01 -2.81480908e-01 -8.96788538e-01
7.75414467e-01 8.61197472e-01 3.59110683e-01 -2.23878995e-01
4.91025358e-01 1.30185515e-01 -2.72629499e-01 -5.08802474e-01
-1.36192167e+00 -7.51994550e-01 -8.23437631e-01 3.09552103e-01
1.05175805e+00 -9.69669044e-01 -7.99445748e-01 2.71799173e-02
-1.07596624e+00 4.09568489e-01 -4.18003201e-01 4.91062969e-01
-4.66496557e-01 -1.05796708e-02 -7.66510308e-01 -5.77436447e-01
-4.14402455e-01 -9.33691859e-01 9.69608486e-01 1.70470104e-01
-3.39777112e-01 -9.20732021e-01 3.50375958e-02 2.81357020e-01
7.61530936e-01 3.00978035e-01 1.08772671e+00 -1.22489822e+00
-6.50633216e-01 -3.16035777e-01 -2.75660098e-01 2.27331355e-01
1.95370153e-01 -6.06266499e-01 -1.06486404e+00 4.81543392e-02
-1.72111869e-01 -1.84157062e-02 1.07882929e+00 3.36801142e-01
8.46346438e-01 -3.07103992e-01 -2.44310215e-01 4.36954379e-01
1.54324853e+00 3.20231885e-01 7.73329794e-01 8.09383690e-01
7.45766640e-01 4.70660836e-01 7.47001708e-01 4.28558648e-01
5.81490815e-01 5.49103916e-01 6.65204167e-01 -5.78345023e-02
1.14904404e-01 -3.31790388e-01 1.86463669e-01 8.57767105e-01
7.00863004e-02 1.63689610e-02 -2.02569351e-01 7.14592934e-01
-1.73356986e+00 -9.80380177e-01 -2.91135907e-02 1.96620190e+00
7.28348792e-01 4.58962142e-01 2.81220019e-01 -8.56017321e-02
7.36627936e-01 9.66607630e-02 -6.70536578e-01 -7.77524889e-01
1.07907681e-02 4.04616147e-01 6.68399990e-01 4.43973467e-02
-1.33934999e+00 1.02417147e+00 5.38177776e+00 7.40104914e-01
-5.92896819e-01 3.05939853e-01 5.42815506e-01 1.04930729e-01
-5.61918795e-01 -1.50215551e-01 -1.16042614e+00 5.21395683e-01
1.13568354e+00 5.02142720e-02 1.33409500e-01 1.17651236e+00
-3.66333067e-01 3.53953034e-01 -1.24148047e+00 9.94433403e-01
8.36099684e-02 -1.09463608e+00 -5.11952080e-02 3.61214310e-01
4.03170526e-01 -1.21693328e-01 2.00050082e-02 7.13566899e-01
4.85294968e-01 -1.00722063e+00 6.08312309e-01 3.87458146e-01
4.36844140e-01 -1.02595186e+00 1.26501811e+00 -1.64721638e-01
-1.41851127e+00 -8.93915296e-02 -5.76989830e-01 9.80337113e-02
4.97950792e-01 7.71945238e-01 -7.50089645e-01 3.73702645e-01
5.85557401e-01 1.08327544e+00 -5.03165960e-01 8.36246252e-01
-1.24958113e-01 2.38482237e-01 8.29913616e-02 -3.20955336e-01
5.13783872e-01 -9.25754905e-02 -1.54325128e-01 1.18297160e+00
3.25251549e-01 -2.03731686e-01 8.14751387e-02 1.01286101e+00
-3.42838287e-01 1.89081609e-01 -5.77118814e-01 -6.61309898e-01
2.11922303e-01 1.39153695e+00 -6.88126624e-01 -4.02912080e-01
-8.97729278e-01 8.89203310e-01 1.72561720e-01 2.88571771e-02
-8.19299698e-01 -6.32650614e-01 1.11635530e+00 2.42621079e-01
9.53290641e-01 -1.42525703e-01 -8.80418196e-02 -1.00511551e+00
-1.94978550e-01 -6.76833391e-01 4.65004414e-01 -2.71063209e-01
-1.83356667e+00 8.32040668e-01 -4.18000966e-01 -1.42852676e+00
-3.21849793e-01 -6.71416521e-01 9.45546478e-02 6.49653137e-01
-1.93268168e+00 -1.28044379e+00 3.54979783e-02 2.43346080e-01
6.34722829e-01 -8.38873982e-02 9.89842355e-01 5.67970991e-01
-2.41655245e-01 5.44178486e-01 1.30836427e-01 4.95382190e-01
5.17205715e-01 -1.20270288e+00 8.48413527e-01 2.77491361e-01
4.02229011e-01 6.48324251e-01 4.77702826e-01 -4.94827420e-01
-1.37086034e+00 -1.21153486e+00 1.33073092e+00 -3.89164120e-01
8.42049181e-01 -6.24206483e-01 -7.57685244e-01 8.24561059e-01
4.71747637e-01 -2.09423721e-01 1.11823511e+00 4.52922493e-01
-5.66626966e-01 -2.77082086e-01 -1.02598369e+00 1.47184521e-01
1.05957937e+00 -3.85828227e-01 -6.82895482e-01 8.88895318e-02
1.11711562e+00 3.11399847e-01 -1.42193079e+00 -3.38300951e-02
6.24840736e-01 -7.24785447e-01 1.07389939e+00 -5.20936370e-01
3.48825514e-01 1.63049418e-02 -3.12000960e-01 -1.35414660e+00
-7.16524661e-01 -1.45333067e-01 -8.90927240e-02 1.55218220e+00
6.01573884e-01 -7.56840706e-01 5.92719913e-01 4.62947816e-01
6.47299513e-02 -7.23309457e-01 -5.74300826e-01 -7.93435812e-01
-2.20802631e-02 -2.80497372e-01 1.17553627e+00 8.98763597e-01
-1.01867810e-01 8.64928782e-01 -3.97101194e-01 -1.26424685e-01
6.93992078e-01 4.33543473e-01 1.80720046e-01 -1.64145577e+00
-3.00491542e-01 -4.38363731e-01 -6.83145583e-01 -1.04612386e+00
1.61577821e-01 -9.27992582e-01 -4.76738811e-02 -1.76829422e+00
2.98511565e-01 -2.71607399e-01 -8.49072397e-01 4.08359915e-01
2.17593536e-01 2.32220143e-01 6.97416961e-02 8.58907681e-03
-7.44292438e-01 4.33006912e-01 1.26132286e+00 -5.53085923e-01
1.26636952e-01 -7.75338486e-02 -1.22671461e+00 3.24646354e-01
9.51729476e-01 -6.14464879e-01 -6.28262460e-02 -3.81499559e-01
6.25670135e-01 -3.90821904e-01 -1.06848851e-01 -5.95542312e-01
-9.04056579e-02 2.91070435e-02 5.32895923e-01 -5.24507403e-01
2.94642419e-01 -8.82829070e-01 2.57542878e-02 2.86263674e-01
-3.68792862e-01 1.39688671e-01 1.27123967e-02 4.70363885e-01
-4.01703864e-01 -4.26676720e-01 3.64414692e-01 -4.44055110e-01
-1.36533165e+00 4.13459837e-01 -3.02155972e-01 -1.06772371e-01
8.19914103e-01 -4.36507016e-02 -8.43359604e-02 -8.64373520e-02
-9.42309082e-01 -2.45907921e-02 7.49497265e-02 8.32655251e-01
5.67444921e-01 -1.62294757e+00 -7.10124254e-01 2.93459624e-01
5.37567198e-01 -4.46053147e-01 -1.87196210e-01 4.04576540e-01
1.23950504e-02 5.42972803e-01 -2.98548520e-01 -2.89672881e-01
-7.50856936e-01 1.13333774e+00 -1.61336392e-01 -3.65252614e-01
-4.37750399e-01 6.85650647e-01 -7.19748735e-02 -3.10249329e-01
6.58887625e-02 -5.30376971e-01 -7.39139974e-01 5.59097886e-01
4.80289727e-01 -1.83352619e-01 1.97560370e-01 -6.91677630e-01
-2.67387033e-01 6.59094810e-01 -4.96606082e-01 3.13426107e-01
1.65630662e+00 -1.18032277e-01 3.93208228e-02 3.02181184e-01
1.59958458e+00 -4.32443500e-01 -1.01469362e+00 -5.23030877e-01
5.71388602e-01 -5.50068498e-01 1.65567055e-01 -6.89093590e-01
-1.08079064e+00 7.84185052e-01 7.82260656e-01 4.54064786e-01
8.65536511e-01 2.90740013e-01 1.47870624e+00 2.37147555e-01
5.80655754e-01 -1.20506620e+00 -6.41276035e-03 3.25432420e-01
3.14963430e-01 -1.38680077e+00 -3.64815056e-01 -3.60462576e-01
-6.55640662e-01 1.08603621e+00 2.27313191e-01 8.04718211e-02
6.80561006e-01 1.36610091e-01 1.17930815e-01 -3.06242406e-01
-4.40470099e-01 -1.00958717e+00 9.80656147e-02 5.50025940e-01
7.16929615e-01 3.50064486e-01 -1.45508215e-01 1.04153037e+00
-2.40250319e-01 -3.43340682e-03 3.24826300e-01 7.77914882e-01
-5.61185062e-01 -1.51045609e+00 1.07991576e-01 5.26854217e-01
-6.31232560e-01 -5.14123440e-01 -8.00783262e-02 6.78798735e-01
3.87931436e-01 8.68271887e-01 1.57405496e-01 -4.74977106e-01
4.97991085e-01 2.46602744e-01 4.28173333e-01 -6.05249345e-01
-5.81196368e-01 6.02021441e-02 3.42261970e-01 -3.74072045e-01
-3.87939006e-01 -5.02424419e-01 -1.04329097e+00 -3.99782479e-01
-3.29156280e-01 3.00428450e-01 8.76150370e-01 8.51525962e-01
3.78876626e-01 7.04643548e-01 5.91442406e-01 -4.56930935e-01
-8.49801898e-01 -9.65055346e-01 -8.67933273e-01 7.39138961e-01
2.02810705e-01 -6.99091136e-01 -1.73178148e-02 2.19024904e-02] | [9.962224960327148, 6.193742275238037] |
81365175-a7b0-41d9-91fa-d4163644f78c | learning-the-loss-functions-in-a | 2003.09124 | null | https://arxiv.org/abs/2003.09124v1 | https://arxiv.org/pdf/2003.09124v1.pdf | Learning the Loss Functions in a Discriminative Space for Video Restoration | With more advanced deep network architectures and learning schemes such as GANs, the performance of video restoration algorithms has greatly improved recently. Meanwhile, the loss functions for optimizing deep neural networks remain relatively unchanged. To this end, we propose a new framework for building effective loss functions by learning a discriminative space specific to a video restoration task. Our framework is similar to GANs in that we iteratively train two networks - a generator and a loss network. The generator learns to restore videos in a supervised fashion, by following ground truth features through the feature matching in the discriminative space learned by the loss network. In addition, we also introduce a new relation loss in order to maintain the temporal consistency in output videos. Experiments on video superresolution and deblurring show that our method generates visually more pleasing videos with better quantitative perceptual metric values than the other state-of-the-art methods. | ['Seonghyeon Nam', 'Younghyun Jo', 'Seoung Wug Oh', 'Seon Joo Kim', 'Peter Vajda', 'Jaeyeon Kang'] | 2020-03-20 | null | null | null | null | ['video-restoration'] | ['computer-vision'] | [ 2.22097546e-01 -2.87377119e-01 -9.79125723e-02 -3.66815537e-01
-7.97229886e-01 -2.21620753e-01 4.44884092e-01 -7.25039184e-01
1.10615075e-01 9.02604699e-01 6.83081567e-01 2.89596140e-01
3.19892261e-03 -9.48230386e-01 -9.89974380e-01 -8.49828124e-01
5.90768605e-02 -1.01111658e-01 -6.57099783e-02 -9.69056040e-02
3.00640464e-02 3.76650959e-01 -1.32482898e+00 3.46825719e-01
1.05541992e+00 1.15289378e+00 2.80731887e-01 3.25746179e-01
3.69016588e-01 1.20839989e+00 -3.40597421e-01 -2.94604748e-01
3.91332984e-01 -8.58616173e-01 -5.98920584e-01 3.44836593e-01
6.15776837e-01 -8.11709285e-01 -1.19491744e+00 1.11867106e+00
4.63002652e-01 3.55416119e-01 5.32675683e-01 -9.82267201e-01
-1.11652017e+00 4.67525661e-01 -4.55774367e-01 1.18110366e-02
3.45635265e-01 2.18110114e-01 9.06493783e-01 -8.44330370e-01
5.24052024e-01 1.34984016e+00 7.10491121e-01 7.31390119e-01
-1.38916385e+00 -6.24416053e-01 9.96030197e-02 4.91551429e-01
-1.22422802e+00 -4.94210094e-01 1.01793301e+00 -3.91169161e-01
4.36121017e-01 6.44683838e-02 5.70241928e-01 1.29716480e+00
1.08439110e-01 8.24379027e-01 7.39819467e-01 -1.12177260e-01
6.76994622e-02 -3.69019240e-01 -5.81787169e-01 7.42283702e-01
-7.79114068e-02 2.14379564e-01 -6.39450312e-01 2.50958204e-01
1.26706743e+00 2.24526420e-01 -8.07006121e-01 -5.95990479e-01
-1.02762532e+00 8.71376991e-01 6.21947706e-01 3.55213583e-01
-5.44879854e-01 3.99200410e-01 3.45418632e-01 3.35442603e-01
7.26876616e-01 8.66507143e-02 -1.71366021e-01 1.56113356e-01
-1.12118101e+00 1.70353621e-01 3.01572084e-01 6.86597526e-01
7.39126205e-01 4.99085575e-01 -5.27061641e-01 1.00540125e+00
2.44915351e-01 1.70693114e-01 5.43981731e-01 -1.40915203e+00
4.19329911e-01 1.13526613e-01 1.14382282e-01 -1.17831683e+00
1.61979586e-01 -5.59046090e-01 -1.35024548e+00 2.56026655e-01
-2.65649483e-02 1.30221516e-01 -8.73603046e-01 2.08834672e+00
-6.95027187e-02 6.04378462e-01 -1.44222692e-01 1.07995725e+00
5.05026758e-01 8.08008850e-01 -2.55172849e-01 -2.95518339e-01
7.35268414e-01 -1.13924909e+00 -9.94084418e-01 -3.78951319e-02
-5.54618835e-02 -5.72261989e-01 1.09791613e+00 4.13596332e-01
-1.33172584e+00 -8.94210994e-01 -1.12327743e+00 -1.97266996e-01
3.77572060e-01 1.80020958e-01 4.39843297e-01 3.91494066e-01
-1.42484212e+00 9.25213575e-01 -8.29551458e-01 -7.34062269e-02
6.71651602e-01 9.58054513e-03 -3.25254709e-01 -4.27414536e-01
-1.09834123e+00 6.05341613e-01 1.38202921e-01 1.99979722e-01
-1.52390885e+00 -7.35269427e-01 -1.04677856e+00 1.73100427e-01
6.57228604e-02 -1.07342684e+00 9.83278275e-01 -1.19845808e+00
-1.84752154e+00 7.61197388e-01 -1.07992087e-02 -4.44657028e-01
6.39699876e-01 -4.33191478e-01 -4.10176992e-01 2.53054053e-01
1.76009402e-01 6.06244326e-01 1.15097332e+00 -1.49719799e+00
-4.68998849e-01 -4.41301540e-02 7.22441375e-02 -1.15710767e-02
-4.02262837e-01 -1.46772578e-01 -4.84296948e-01 -1.19613850e+00
2.16126218e-02 -4.58257616e-01 -1.02298692e-01 3.05379480e-01
-3.18433017e-01 1.64280072e-01 1.00259113e+00 -1.15589249e+00
1.09495735e+00 -2.19110394e+00 6.76628768e-01 -1.11694269e-01
3.08250248e-01 1.05969079e-01 -5.03806233e-01 2.70048797e-01
-2.03674555e-01 -6.24797866e-02 -4.49221283e-01 -4.93256301e-01
-6.03478663e-02 3.33497137e-01 -5.98140299e-01 5.55251241e-01
5.01389354e-02 8.00089300e-01 -9.72572386e-01 -1.83582097e-01
2.01077044e-01 8.48336220e-01 -8.58746111e-01 6.11355186e-01
-1.13844924e-01 7.33228803e-01 -3.22915047e-01 5.53394258e-01
7.23665059e-01 -3.30700040e-01 1.44157723e-01 -4.84530509e-01
1.39697284e-01 1.32170200e-01 -8.22739661e-01 2.14204383e+00
-4.80392814e-01 8.29402089e-01 2.47869790e-01 -1.50230694e+00
8.83893967e-01 3.12571079e-01 6.06076837e-01 -7.75065541e-01
-9.62822661e-02 5.95684685e-02 -5.64154983e-01 -5.98362923e-01
3.78980398e-01 -2.43739903e-01 3.60354125e-01 1.92483962e-01
2.33762339e-01 3.42221558e-02 7.73311481e-02 1.29353225e-01
1.02880323e+00 4.36007291e-01 -1.94512457e-01 -1.86535399e-02
5.83589435e-01 -5.94382107e-01 1.00288653e+00 4.92012531e-01
6.75434917e-02 1.04879808e+00 2.87693799e-01 -5.17646015e-01
-1.26766276e+00 -1.14659560e+00 3.60410959e-02 7.08165050e-01
3.43158901e-01 -2.82110453e-01 -8.94404233e-01 -6.07099712e-01
-2.40464732e-01 5.15120149e-01 -4.85023141e-01 -4.14114654e-01
-8.53382766e-01 -5.05805731e-01 2.79171258e-01 6.47955477e-01
8.71724844e-01 -1.01747739e+00 -3.02943890e-03 2.83460885e-01
-6.21560335e-01 -1.16811931e+00 -7.43895292e-01 -4.11417872e-01
-1.10929179e+00 -9.09678400e-01 -9.55109060e-01 -1.06168711e+00
7.13408887e-01 3.03740978e-01 1.22804058e+00 2.06758648e-01
-6.47283047e-02 2.55360305e-01 -3.89065832e-01 5.50972402e-01
-3.63200724e-01 -3.14932734e-01 -9.34844911e-02 3.35272193e-01
-2.96455681e-01 -1.01255703e+00 -9.77080464e-01 3.65167439e-01
-1.25176227e+00 1.65697932e-01 4.05175209e-01 1.03444386e+00
7.14737058e-01 2.61372298e-01 5.14732957e-01 -4.24499691e-01
4.50484812e-01 -3.75513107e-01 -4.69741911e-01 9.40332487e-02
-3.95133197e-01 7.07641318e-02 8.63872886e-01 -2.40356654e-01
-1.15142190e+00 -6.01319522e-02 -1.62951723e-01 -7.41706908e-01
6.51288927e-02 3.16139787e-01 -4.25864756e-01 -1.99956715e-01
4.24326777e-01 5.49862564e-01 9.35824662e-02 -8.04900408e-01
3.69618326e-01 2.46926770e-01 8.14036012e-01 -5.55820763e-01
1.08352625e+00 5.44097066e-01 -1.87064737e-01 -3.32406640e-01
-8.93830299e-01 1.76618397e-01 -2.86385894e-01 -3.11777711e-01
7.32014954e-01 -1.13164544e+00 -5.38213968e-01 7.58198261e-01
-1.15992415e+00 -5.24066746e-01 -4.83747661e-01 4.94621098e-01
-9.02255177e-01 5.96394122e-01 -1.02423525e+00 -2.26019695e-01
-1.91527501e-01 -1.14776313e+00 1.00799406e+00 1.82206079e-01
3.83275658e-01 -9.34908092e-01 1.57648072e-01 4.09035236e-01
6.16286039e-01 3.72444540e-01 8.49557400e-01 2.11450174e-01
-9.24870014e-01 2.52449006e-01 -3.68749887e-01 9.66170073e-01
3.66492808e-01 -3.08479875e-01 -8.18639874e-01 -6.19982362e-01
2.65068263e-01 -3.02681953e-01 1.26696455e+00 6.35347903e-01
1.77009571e+00 -6.98009849e-01 -7.02445060e-02 1.15671921e+00
1.43154860e+00 4.43501435e-02 1.15396821e+00 3.16377670e-01
7.48454988e-01 2.68536776e-01 2.46709034e-01 2.65925825e-01
2.81445414e-01 7.74553776e-01 5.97152948e-01 -1.87023371e-01
-5.51989496e-01 -5.29738367e-01 7.55092561e-01 7.74256527e-01
-2.24548161e-01 -3.07077527e-01 -2.96021938e-01 5.21826088e-01
-1.97136390e+00 -1.18114817e+00 4.91767526e-01 2.09172058e+00
1.06789374e+00 -3.55748445e-01 -9.12970454e-02 2.31647659e-02
8.68022442e-01 5.01889825e-01 -5.51442027e-01 2.79703379e-01
-3.23853046e-01 8.63109902e-02 1.13178357e-01 5.75406015e-01
-1.05406976e+00 8.43035400e-01 6.53237057e+00 9.05348718e-01
-1.13024342e+00 1.86061680e-01 9.09478426e-01 -7.17884004e-02
-5.51197171e-01 -7.57010803e-02 -2.11795673e-01 5.64754844e-01
3.85623276e-01 9.33883619e-03 8.21832061e-01 6.20995045e-01
4.65942025e-01 4.77577984e-01 -1.07743835e+00 1.19402182e+00
2.95796901e-01 -1.64856005e+00 3.63207906e-01 -3.91631760e-02
1.06608295e+00 -2.54614741e-01 2.01370254e-01 8.18068236e-02
2.49871418e-01 -1.15834665e+00 8.22085679e-01 8.56508434e-01
9.35436487e-01 -7.27587044e-01 5.53624809e-01 -4.48741987e-02
-9.84446466e-01 -9.21321064e-02 -4.67393011e-01 1.46670461e-01
5.22715211e-01 7.41856456e-01 -1.75830200e-02 7.30268180e-01
7.16199398e-01 1.33054161e+00 -2.20026001e-01 1.16624534e+00
-5.61967373e-01 4.99383211e-01 3.40114802e-01 8.02595079e-01
6.81661740e-02 -5.33993602e-01 6.35705471e-01 9.27290678e-01
6.13542557e-01 -6.64140582e-02 1.06227785e-01 9.80798900e-01
-5.67123890e-01 -2.29182273e-01 -4.42374766e-01 1.97697416e-01
2.64503449e-01 1.12440050e+00 -1.60343230e-01 -2.30437055e-01
-3.22153777e-01 1.32985282e+00 1.88610137e-01 7.75395513e-01
-9.41260874e-01 -1.70316771e-01 9.05385911e-01 2.34953120e-01
4.86722708e-01 -1.80777609e-01 3.66340503e-02 -1.55480516e+00
2.91418463e-01 -1.04322493e+00 1.42737582e-01 -9.54008818e-01
-1.56810808e+00 7.15273798e-01 -3.97881150e-01 -1.39362156e+00
-3.55653644e-01 -4.17565942e-01 -6.12745583e-01 6.83419824e-01
-1.74713600e+00 -1.06863868e+00 -5.73865891e-01 9.00485933e-01
5.94059348e-01 -3.71320099e-01 4.39223677e-01 5.91382086e-01
-5.61896682e-01 5.13145149e-01 3.08858901e-01 2.51174122e-01
6.83632851e-01 -9.71679866e-01 1.08615160e-01 1.21035409e+00
4.13829796e-02 3.16118091e-01 6.51924312e-01 -4.10999030e-01
-1.21727109e+00 -1.24673605e+00 3.59206885e-01 1.41669258e-01
4.70265388e-01 -1.27995580e-01 -9.93219197e-01 7.08805799e-01
4.93775219e-01 1.85581565e-01 2.64483482e-01 -3.61063927e-01
-5.54570377e-01 -5.32524705e-01 -1.29481936e+00 4.10245031e-01
1.31683123e+00 -7.62516320e-01 -2.86346257e-01 3.90992284e-01
8.85918975e-01 -3.56503576e-01 -7.67776132e-01 5.21105587e-01
3.56031328e-01 -1.12389052e+00 1.24177051e+00 -5.08156657e-01
9.68720853e-01 -4.48389918e-01 -3.87075186e-01 -1.49687743e+00
-6.73036218e-01 -7.61636436e-01 -4.06305820e-01 1.22831333e+00
-1.56028539e-01 -4.80988950e-01 6.42217815e-01 7.12506771e-02
-3.73916030e-01 -6.12083614e-01 -8.59296143e-01 -8.02602470e-01
-1.46479130e-01 5.90981133e-02 5.74002445e-01 9.15149629e-01
-6.19016469e-01 9.33958888e-02 -9.16532040e-01 2.76170094e-02
9.14448917e-01 1.63747638e-01 5.23228526e-01 -8.18239868e-01
-4.35390353e-01 -3.98751348e-01 -4.90478337e-01 -1.51977026e+00
4.13022429e-01 -7.60848820e-01 8.40051696e-02 -1.54994285e+00
5.38218796e-01 -1.43658757e-01 -4.41356003e-01 3.87023568e-01
-4.53019738e-02 4.92696732e-01 2.13961210e-02 2.91229039e-01
-4.57796246e-01 1.22830725e+00 1.59612536e+00 -2.16855034e-01
9.68627259e-02 -2.62350053e-01 -7.73791790e-01 5.93174040e-01
5.26029766e-01 -2.59419620e-01 -3.94061625e-01 -8.32400918e-01
-4.85396981e-02 1.81592986e-01 6.91889048e-01 -9.44026291e-01
3.37892100e-02 -1.08085729e-01 5.55395305e-01 -2.82555878e-01
2.53671288e-01 -6.13617182e-01 2.61570066e-01 2.20687911e-01
-4.06225234e-01 -1.64302096e-01 -8.62438828e-02 6.61460400e-01
-6.76632941e-01 1.28235251e-01 1.03429735e+00 5.62120378e-02
-6.06484234e-01 6.71078444e-01 -8.34140554e-02 4.32080999e-02
6.63114488e-01 -1.36857718e-01 -2.08163008e-01 -7.52977550e-01
-6.20907366e-01 -2.29452271e-02 6.30313516e-01 4.91564214e-01
8.75375152e-01 -1.75188744e+00 -9.37052667e-01 2.37354949e-01
-4.25072491e-01 -7.81954974e-02 5.33555508e-01 5.77072740e-01
-5.38410306e-01 7.71678239e-02 -5.76804101e-01 -4.05087918e-01
-7.77437508e-01 4.81675267e-01 5.30981421e-01 -2.03784853e-01
-8.86059701e-01 6.41249835e-01 6.10938072e-01 -1.74645975e-01
3.35537463e-01 -4.16943803e-02 -7.83190653e-02 -3.19668233e-01
6.84010983e-01 3.58525276e-01 -1.54243872e-01 -5.12877524e-01
-4.69008982e-02 5.60170531e-01 5.20581864e-02 1.04219526e-01
1.75384557e+00 -3.06484401e-01 -3.48706782e-01 -1.39648929e-01
1.32151532e+00 -5.00283353e-02 -1.82204413e+00 -3.76423925e-01
-4.35034752e-01 -9.79378343e-01 2.22642645e-01 -5.78025460e-01
-1.87618399e+00 6.10534310e-01 6.20896757e-01 7.33469650e-02
1.73485613e+00 -1.22056834e-01 1.04385686e+00 -1.94280520e-01
2.31639430e-01 -8.22070241e-01 5.34949481e-01 2.20810965e-01
1.34840596e+00 -9.77049947e-01 -9.65110287e-02 -2.35312551e-01
-3.03835362e-01 1.05379748e+00 5.82345963e-01 -4.15141761e-01
3.67956609e-01 4.14474048e-02 -2.12549269e-01 1.13157466e-01
-4.63675410e-01 1.92462713e-01 3.51555943e-01 7.31317520e-01
4.73253220e-01 -1.90293595e-01 -2.43787438e-01 5.17678320e-01
7.61445761e-02 2.47086212e-01 3.54277074e-01 3.55494797e-01
-2.85697877e-01 -1.28023088e+00 -1.82641879e-01 1.21315196e-01
-4.67317849e-01 -1.08972847e-01 1.47323206e-01 4.19697285e-01
-7.20623434e-02 7.82326341e-01 -4.52458225e-02 -4.33415949e-01
1.93636343e-01 -4.20060009e-01 6.17274284e-01 -2.88384885e-01
-6.85286298e-02 1.18153065e-01 -1.54534936e-01 -1.05614376e+00
-6.25516653e-01 -4.11604345e-01 -8.32613647e-01 -3.89700681e-01
7.89756924e-02 -3.13000418e-02 1.94414034e-01 7.99482524e-01
4.58737820e-01 6.63493395e-01 9.90878165e-01 -1.14033926e+00
-6.09184265e-01 -7.75075436e-01 -6.93539381e-01 5.86471200e-01
6.14089370e-01 -5.85713089e-01 -3.19257021e-01 4.44416851e-01] | [11.272512435913086, -1.911670446395874] |
ceaea01f-93b0-469b-849d-643c53cd141c | a-kernel-based-quantum-random-forest-for | 2210.02355 | null | https://arxiv.org/abs/2210.02355v2 | https://arxiv.org/pdf/2210.02355v2.pdf | A kernel-based quantum random forest for improved classification | The emergence of Quantum Machine Learning (QML) to enhance traditional classical learning methods has seen various limitations to its realisation. There is therefore an imperative to develop quantum models with unique model hypotheses to attain expressional and computational advantage. In this work we extend the linear quantum support vector machine (QSVM) with kernel function computed through quantum kernel estimation (QKE), to form a decision tree classifier constructed from a decision directed acyclic graph of QSVM nodes - the ensemble of which we term the quantum random forest (QRF). To limit overfitting, we further extend the model to employ a low-rank Nystr\"{o}m approximation to the kernel matrix. We provide generalisation error bounds on the model and theoretical guarantees to limit errors due to finite sampling on the Nystr\"{o}m-QKE strategy. In doing so, we show that we can achieve lower sampling complexity when compared to QKE. We numerically illustrate the effect of varying model hyperparameters and finally demonstrate that the QRF is able obtain superior performance over QSVMs, while also requiring fewer kernel estimations. | ['Lloyd C. L. Hollenberg', 'Charles D. Hill', 'Maiyuren Srikumar'] | 2022-10-05 | null | null | null | null | ['classification'] | ['methodology'] | [ 5.75470924e-01 3.72096598e-01 -1.51384979e-01 -1.89799026e-01
-9.42650735e-01 -4.36874449e-01 5.91171563e-01 -1.70407280e-01
-4.10255581e-01 1.02076817e+00 -4.44122255e-01 -7.73253322e-01
-5.00235498e-01 -9.75315213e-01 -5.52569032e-01 -1.03541160e+00
-2.40034491e-01 3.46648097e-01 7.52287135e-02 -1.66251093e-01
4.21932727e-01 5.20972013e-01 -1.19237256e+00 1.05394600e-02
9.56946671e-01 8.01740229e-01 -2.55929977e-01 9.48718369e-01
4.65902120e-01 7.14503765e-01 -2.73923576e-01 -3.92811477e-01
3.55603576e-01 -6.11844480e-01 -1.08020115e+00 -2.72267133e-01
3.38944972e-01 -2.61927903e-01 -7.06798851e-01 1.25642169e+00
4.28407460e-01 1.16767593e-01 1.16615057e+00 -1.17504692e+00
-2.87719190e-01 3.85328352e-01 -6.56233728e-02 8.46908316e-02
1.43981263e-01 -1.30293190e-01 1.22401905e+00 -4.88129467e-01
6.64271832e-01 8.54959011e-01 6.11277878e-01 4.59435225e-01
-1.90676963e+00 -7.71016359e-01 -1.04239595e+00 6.87257946e-01
-1.78783596e+00 -4.53363270e-01 2.91868418e-01 -2.96878785e-01
1.01235926e+00 2.82184392e-01 2.68282056e-01 8.03640723e-01
4.22358900e-01 3.29256147e-01 1.80917764e+00 -1.08720839e+00
5.40764213e-01 2.08982900e-01 3.57360899e-01 1.01865780e+00
2.20458299e-01 5.43755651e-01 -5.80012023e-01 -6.96445525e-01
7.75558293e-01 -5.48658311e-01 -7.81738162e-02 -5.08133650e-01
-9.53134835e-01 1.17795944e+00 3.44701350e-01 -6.05163313e-02
-1.43087432e-01 4.54394847e-01 1.42755032e-01 5.43406665e-01
-1.79506245e-03 3.48698258e-01 -2.89592415e-01 1.15427256e-01
-1.01060951e+00 7.25786388e-02 9.87230301e-01 7.10433304e-01
1.12250185e+00 -2.02386096e-01 1.62036136e-01 3.45905334e-01
8.45253021e-02 6.80122674e-01 -1.77092105e-01 -9.97465670e-01
5.80783486e-02 3.70125324e-02 1.61326706e-01 -3.60289663e-01
-5.16167879e-01 -5.17058253e-01 -8.68869126e-01 3.47488135e-01
4.20886070e-01 4.96494323e-02 -6.78430080e-01 1.40620434e+00
3.46963257e-01 1.55465141e-01 4.72374797e-01 5.31741440e-01
1.80421650e-01 5.79172671e-01 -1.62435576e-01 -3.22966725e-01
1.01244473e+00 -4.49118257e-01 -2.82798052e-01 2.60359466e-01
1.49078107e+00 -5.98584473e-01 5.40213108e-01 6.89271510e-01
-2.84464419e-01 -2.72490799e-01 -1.37184417e+00 3.10871214e-01
-2.80306228e-02 1.25120446e-01 9.23088670e-01 1.21330464e+00
-8.29487145e-01 1.27880895e+00 -7.74317324e-01 -1.63382694e-01
1.31955758e-01 8.83818924e-01 -5.44684291e-01 -8.57274514e-03
-1.36338329e+00 1.15870881e+00 7.55657434e-01 -4.73052636e-02
-5.50442398e-01 -2.71907006e-03 -4.90503848e-01 -3.22824538e-01
2.78214723e-01 -7.05010176e-01 1.11762261e+00 -5.86775661e-01
-1.51489258e+00 6.04926169e-01 -1.39679119e-01 -7.22531319e-01
6.88069463e-02 2.10653469e-01 -3.67839366e-01 4.65798110e-01
-1.79188028e-02 2.23526791e-01 8.50341499e-01 -8.74388278e-01
-3.28451335e-01 -3.98993909e-01 8.59920774e-03 -2.30864152e-01
1.09910846e-01 -1.45041242e-01 3.95668447e-01 2.01247543e-01
5.44864297e-01 -1.46986723e+00 -3.40771228e-01 -7.37461388e-01
-5.18438280e-01 -3.53072852e-01 3.85967970e-01 -3.30817580e-01
1.33807385e+00 -1.90163779e+00 9.08817872e-02 6.58030808e-01
2.60317922e-01 6.87272996e-02 2.20407739e-01 9.23184097e-01
2.58267373e-02 -1.35462299e-01 -1.98494121e-01 1.69740930e-01
8.68091732e-02 4.82893854e-01 -3.50426555e-01 9.44770277e-01
1.30929202e-01 7.62315333e-01 -7.48484135e-01 -5.61323166e-01
1.37595206e-01 1.55832633e-01 -4.96013880e-01 -4.66584772e-01
4.75112051e-02 2.80807912e-01 -4.47941422e-01 5.04598796e-01
7.60221660e-01 -5.03391087e-01 3.12017918e-01 -1.32828355e-01
7.79908299e-02 4.02833909e-01 -1.36967158e+00 1.20415914e+00
-1.87667131e-01 6.01667821e-01 -5.32213487e-02 -1.17627418e+00
6.16760731e-01 9.10097212e-02 -3.78304720e-02 -5.61461806e-01
-6.75292406e-03 5.74611545e-01 3.79565448e-01 -9.89218354e-02
3.54057044e-01 -7.12660909e-01 -1.06402330e-01 2.71427691e-01
4.51305091e-01 2.69761253e-02 -2.02753440e-01 3.91773403e-01
1.13723660e+00 2.83524871e-01 4.96816337e-01 -4.70663369e-01
5.30614614e-01 3.43871936e-02 1.22675978e-01 1.22718430e+00
-3.48322213e-01 -3.30821574e-02 5.69667995e-01 -9.11694914e-02
-9.32860851e-01 -1.18132317e+00 -8.81063998e-01 7.79934168e-01
-5.44291269e-03 -7.29960978e-01 -7.24397898e-01 -4.57717985e-01
-5.46245612e-02 1.10180843e+00 -6.70984328e-01 -3.23351443e-01
-1.69428542e-01 -1.09841061e+00 8.10140848e-01 5.23248352e-02
3.70485097e-01 -4.59523082e-01 -3.67504627e-01 1.64582208e-01
1.74562231e-01 -1.02739525e+00 4.23191339e-01 8.27941597e-01
-9.30234969e-01 -8.91393065e-01 -1.63281396e-01 -2.68802792e-01
3.43801796e-01 -8.27432796e-02 5.15455246e-01 -3.59426469e-01
-1.48564920e-01 2.98046529e-01 -4.03229773e-01 -4.03388999e-02
-9.39697862e-01 2.01756701e-01 5.15196323e-01 -1.98877320e-01
5.14645875e-01 -5.61255753e-01 -5.25853872e-01 7.75211528e-02
-8.83921504e-01 1.43322930e-01 8.80351663e-01 1.17959499e+00
2.21380621e-01 9.00425017e-02 3.42135698e-01 -7.97805250e-01
7.71846026e-02 -1.23750247e-01 -6.22892618e-01 2.46810108e-01
-9.76808965e-01 5.87905467e-01 6.60401762e-01 1.31631372e-02
-5.50814152e-01 -3.84715125e-02 1.05492547e-02 -4.30121049e-02
-3.57097760e-02 4.98428881e-01 3.66935372e-01 -7.09351122e-01
1.07677591e+00 4.08033580e-01 -1.05245739e-01 -1.51217505e-01
5.05229056e-01 1.12632811e+00 1.94914281e-01 -6.51101887e-01
8.91812444e-01 4.31635767e-01 1.04551172e+00 -1.34939253e+00
-9.47804928e-01 -3.93791735e-01 -9.51707006e-01 4.14880831e-03
6.99048281e-01 -7.51738071e-01 -7.67822266e-01 -1.49282301e-02
-7.47309744e-01 -2.17984512e-01 2.70147324e-02 9.49349105e-01
-8.79311085e-01 6.86034203e-01 -8.00704002e-01 -1.35417378e+00
-2.16839418e-01 -7.89325118e-01 9.34658825e-01 -1.73754711e-02
4.10759216e-03 -8.97508204e-01 1.20516017e-01 4.90499645e-01
5.12243807e-02 1.34775206e-01 1.14993167e+00 -6.57398343e-01
-7.30394661e-01 -3.74155164e-01 -2.56812990e-01 5.05038559e-01
-4.98870581e-01 -9.93587002e-02 -1.11325872e+00 -5.80393791e-01
-1.96650416e-01 -5.06995142e-01 9.91893530e-01 2.05128714e-02
4.25596714e-01 -3.64948958e-02 -3.72252673e-01 5.76100707e-01
1.56255591e+00 -2.07738459e-01 6.75613999e-01 5.28629601e-01
3.24891150e-01 4.06518519e-01 4.78880227e-01 2.63533264e-01
9.62615907e-02 6.10794365e-01 8.80791098e-02 4.16979969e-01
4.54556972e-01 -1.60863519e-01 4.85526234e-01 9.73328590e-01
-3.25934768e-01 4.75921601e-01 -7.77975857e-01 -1.96560845e-01
-1.66492403e+00 -9.06405687e-01 -5.01796007e-01 2.66004610e+00
6.04462087e-01 2.62871891e-01 1.26518104e-02 2.25156531e-01
3.53178382e-01 -3.32182556e-01 -2.01918125e-01 -5.39968073e-01
-3.12305596e-02 8.59881580e-01 1.03118491e+00 5.67359984e-01
-1.09236467e+00 9.99131203e-01 6.10363340e+00 1.39990449e+00
-8.03299785e-01 2.56747097e-01 -4.10067802e-03 3.62117350e-01
-9.11969692e-02 8.69118333e-01 -9.00215745e-01 -1.37424292e-02
1.53913808e+00 7.36475140e-02 5.43555081e-01 5.57242095e-01
-1.38460457e-01 -4.04235095e-01 -9.50016141e-01 8.67202580e-01
-3.30486834e-01 -1.22409678e+00 -1.71412006e-01 6.07943892e-01
5.52750587e-01 1.78898633e-01 -2.79811621e-01 6.07398033e-01
9.17712972e-02 -1.18056548e+00 3.63669187e-01 4.26846266e-01
1.13415015e+00 -8.76785934e-01 6.42847657e-01 7.98111618e-01
-8.01191211e-01 8.60133767e-03 -6.81394458e-01 -2.28057697e-01
-4.02539998e-01 2.88139552e-01 -1.09461641e+00 1.04689777e+00
4.69007418e-02 6.48300871e-02 -4.91223454e-01 5.89308977e-01
-1.22996762e-01 9.74550545e-01 -5.97480416e-01 -4.12238926e-01
3.88410926e-01 -7.93797612e-01 3.26422781e-01 1.02528274e+00
1.76418439e-01 1.53945491e-01 -5.43588549e-02 5.93052387e-01
4.05201405e-01 2.39592001e-01 -5.95161319e-01 -1.71586856e-01
3.06923985e-01 1.15657246e+00 -7.42796779e-01 -2.50264466e-01
-3.41956764e-01 1.31368101e+00 3.59658450e-01 9.87919122e-02
-6.56843722e-01 -5.64109385e-01 1.79380119e-01 -1.68293476e-01
2.37874627e-01 -3.92257482e-01 4.70022522e-02 -1.40957403e+00
-3.90372068e-01 -5.83889961e-01 1.10401727e-01 -5.12752175e-01
-1.07411480e+00 2.11933777e-01 8.10807794e-02 -1.06405091e+00
-3.78830165e-01 -8.96278739e-01 -2.43109427e-02 9.92442012e-01
-1.03753889e+00 -1.13256824e+00 5.47192752e-01 2.87106335e-01
-6.43467426e-01 8.92519802e-02 1.44352710e+00 -1.14492945e-01
-4.32719976e-01 6.68102026e-01 8.68449509e-01 -5.75509444e-02
4.02318358e-01 -1.42409575e+00 4.10853997e-02 5.27625084e-01
4.37824219e-01 8.75631571e-01 9.94519651e-01 -5.70258558e-01
-1.64657295e+00 -6.46385312e-01 7.70592630e-01 -6.19077981e-01
1.02541125e+00 -4.21078682e-01 -4.47488964e-01 6.49417579e-01
-2.22209573e-01 -4.33166251e-02 1.20183539e+00 2.90614992e-01
-4.78387803e-01 1.14892691e-01 -1.06896996e+00 5.01279771e-01
7.12701976e-01 -1.14643538e+00 -4.06560212e-01 4.61506546e-01
1.69420362e-01 3.71403322e-02 -1.14262795e+00 5.08331478e-01
7.29523540e-01 -1.11394060e+00 7.33421624e-01 -7.53471851e-01
2.55562346e-02 -4.03262556e-01 -5.73979795e-01 -8.95174503e-01
-2.74755299e-01 -9.75792646e-01 -1.55873626e-01 5.23282290e-01
3.94485831e-01 -9.25077319e-01 1.01380074e+00 3.08808804e-01
2.91056275e-01 -7.74926305e-01 -1.69446445e+00 -9.51364160e-01
4.67114091e-01 -7.00730741e-01 -2.09545508e-01 8.02696586e-01
2.37772837e-01 6.63145244e-01 -4.41009700e-01 3.57646167e-01
1.21848989e+00 2.72192955e-01 8.09195280e-01 -1.18668246e+00
-8.28159511e-01 -4.12558578e-02 -8.75655591e-01 -1.03175616e+00
1.05512060e-01 -1.31899285e+00 -4.38617796e-01 -9.13488328e-01
3.71027708e-01 -3.83347839e-01 -3.11626703e-01 -9.38674435e-02
9.21350569e-02 2.59566754e-01 -1.27011444e-02 4.00695741e-01
-6.54996634e-01 3.58008534e-01 9.30712938e-01 4.12204206e-01
9.88476500e-02 1.32995874e-01 -1.32309362e-01 5.72926700e-01
5.95410824e-01 -6.77568018e-01 6.09615631e-02 6.50253832e-01
5.53752720e-01 4.02487427e-01 5.79618335e-01 -1.09313786e+00
1.02065533e-01 7.72312433e-02 5.13695955e-01 -4.30029154e-01
3.12496215e-01 -1.31579027e-01 9.72632617e-02 7.47883499e-01
-1.36236519e-01 -6.19095683e-01 -2.09577784e-01 7.93375671e-01
2.38064587e-01 -8.00474644e-01 8.52968752e-01 7.68860653e-02
-4.15936232e-01 -9.25829727e-03 -4.50009227e-01 -4.76736635e-01
8.76014829e-01 -2.16636017e-01 -1.26594707e-01 -1.73807427e-01
-9.70256329e-01 -2.07557157e-01 6.65162742e-01 -5.77992380e-01
2.32616633e-01 -1.10562360e+00 -4.51078773e-01 9.06724706e-02
2.58315772e-01 -6.34068310e-01 1.92628875e-01 1.06010067e+00
-7.98912406e-01 7.78448403e-01 1.40854716e-01 -5.41474819e-01
-1.04500771e+00 3.77361149e-01 4.30579871e-01 -3.40827674e-01
-3.63654137e-01 7.55267143e-01 -3.56832564e-01 -6.45550728e-01
-3.40632081e-01 9.73811373e-02 3.45682472e-01 -1.52286634e-01
1.80745453e-01 6.29969358e-01 5.99727444e-02 -7.96900332e-01
-2.57185817e-01 2.68619180e-01 6.36923611e-02 -3.93229157e-01
8.68827343e-01 8.90850797e-02 -3.44210535e-01 4.48423207e-01
1.25136089e+00 1.79718480e-01 -8.00690353e-01 -3.78376454e-01
2.41263613e-01 -8.96295160e-02 1.75201893e-01 -3.74806792e-01
2.05594465e-01 8.77017021e-01 7.22502053e-01 4.72693294e-01
5.51538050e-01 7.09464997e-02 5.51389933e-01 9.98850584e-01
1.04698753e+00 -1.14331889e+00 -6.19656205e-01 4.71661657e-01
2.45857239e-02 -1.10693085e+00 2.74771810e-01 -4.10638869e-01
-1.36477500e-01 1.62034273e+00 -1.69959396e-01 -3.44206452e-01
5.84584594e-01 -2.73377270e-01 -2.62275338e-01 -1.14839993e-01
-5.72116911e-01 -5.07069349e-01 2.05711320e-01 4.06524152e-01
1.90860018e-01 5.22345006e-01 -2.89396673e-01 2.66193837e-01
-5.43154716e-01 -2.58336738e-02 7.30356038e-01 8.83922160e-01
-5.96713841e-01 -1.35722172e+00 -3.97703528e-01 7.71505952e-01
-5.02365649e-01 -2.93399572e-01 -1.62548363e-01 8.57038200e-01
2.52762046e-02 1.03191507e+00 -5.24958014e-01 -7.99530327e-01
-1.58428669e-01 6.24214530e-01 1.23105609e+00 -5.10163248e-01
-2.40010265e-02 -4.54630293e-02 2.46542066e-01 -1.09866343e-01
-2.43340537e-01 -6.29256129e-01 -1.23014915e+00 -4.43081915e-01
-1.14780521e+00 4.34478819e-01 6.39821172e-01 1.20060146e+00
5.71173131e-02 -9.61258188e-02 7.07344651e-01 -5.87793410e-01
-1.66266894e+00 -1.09503675e+00 -1.15093875e+00 -8.04692581e-02
2.69173354e-01 -5.99434614e-01 -7.08236694e-01 -5.73799849e-01] | [5.574934005737305, 4.953494071960449] |
4c93f5ad-88ad-4459-b0cd-3a5576861c60 | can-autism-be-diagnosed-with-ai | 2206.02787 | null | https://arxiv.org/abs/2206.02787v1 | https://arxiv.org/pdf/2206.02787v1.pdf | Can autism be diagnosed with AI? | Radiomics with deep learning models have become popular in computer-aided diagnosis and have outperformed human experts on many clinical tasks. Specifically, radiomic models based on artificial intelligence (AI) are using medical data (i.e., images, molecular data, clinical variables, etc.) for predicting clinical tasks like Autism Spectrum Disorder (ASD). In this review, we summarized and discussed the radiomic techniques used for ASD analysis. Currently, the limited radiomic work of ASD is related to variation of morphological features of brain thickness that is different from texture analysis. These techniques are based on imaging shape features that can be used with predictive models for predicting ASD. This review explores the progress of ASD-based radiomics with a brief description of ASD and the current non-invasive technique used to classify between ASD and Healthy Control (HC) subjects. With AI, new radiomic models using the deep learning techniques will be also described. To consider the texture analysis with deep CNNs, more investigations are suggested to be integrated with additional validation steps on various MRI sites. | ['Tamim Niazi', 'Christian Desrosiers', 'Camel Tanougast', 'Idowu Paul Okuwobi', 'Yujie Li', 'Qizong Lu', 'Jiali Li', 'Ahmad Chaddad'] | 2022-06-05 | null | null | null | null | ['texture-classification'] | ['computer-vision'] | [ 2.48846129e-01 1.20286323e-01 -1.11578658e-01 -3.95341903e-01
-1.10386282e-01 2.14212433e-01 1.99061036e-01 4.01608735e-01
-2.80223012e-01 5.25073111e-01 8.55980441e-02 1.25030696e-01
-3.37414056e-01 -8.93405557e-01 -2.10914358e-01 -8.33014071e-01
-3.45989376e-01 9.54144537e-01 1.50046319e-01 -1.63905174e-01
9.48013142e-02 4.47815359e-01 -1.45355618e+00 6.80162728e-01
9.02104259e-01 1.14502478e+00 2.94970930e-01 4.08000439e-01
-3.27136308e-01 6.06320262e-01 -4.06799138e-01 8.37938339e-02
-3.01804215e-01 -3.28251213e-01 -8.96945834e-01 -2.74544209e-01
3.28699648e-01 -5.49980998e-01 1.14660349e-03 8.47553790e-01
7.74001658e-01 -3.61680090e-01 1.17452633e+00 -6.74525678e-01
-1.03044784e+00 5.01006126e-01 -6.43597186e-01 6.87723100e-01
3.56526405e-01 -3.46410245e-01 1.42383784e-01 -7.56549478e-01
4.08966690e-01 1.22792196e+00 7.82525659e-01 8.30152869e-01
-7.61115015e-01 -5.75112641e-01 -2.19727501e-01 6.68475270e-01
-8.70219409e-01 -1.81282103e-01 3.62500638e-01 -1.01462400e+00
1.06804180e+00 7.65324458e-02 1.13283837e+00 1.13284576e+00
6.89141154e-01 4.48989153e-01 1.29966831e+00 -5.40134311e-01
3.05587143e-01 -4.70225215e-01 -2.25682128e-02 1.14746344e+00
2.93638110e-01 1.57128870e-01 -5.05530894e-01 -1.32983938e-01
6.33498132e-01 6.87563941e-02 1.34495184e-01 -1.54678732e-01
-1.06551623e+00 8.99697900e-01 2.79202580e-01 8.34370017e-01
-2.95251429e-01 1.13057345e-01 6.71468258e-01 2.69514680e-01
9.92967963e-01 3.62734109e-01 -7.04770684e-01 1.89500690e-01
-9.21539605e-01 7.52239600e-02 -5.16517274e-03 2.59146363e-01
3.52202468e-02 3.85134101e-01 -6.74749240e-02 1.55532813e+00
4.28084970e-01 6.88361704e-01 1.09644938e+00 -5.70641637e-01
4.18633670e-02 6.55146062e-01 -7.77098894e-01 -8.58493388e-01
-1.10739052e+00 -5.70749119e-03 -1.17605233e+00 3.31641555e-01
1.57189935e-01 -1.62986666e-01 -1.06166458e+00 1.17053926e+00
3.38065952e-01 1.32310614e-02 -4.12655652e-01 8.47711921e-01
1.37968493e+00 -2.58231759e-02 1.32237166e-01 -6.66659176e-02
1.56695020e+00 -7.51291931e-01 -7.02910304e-01 7.27234408e-02
1.04984999e+00 -5.56131899e-01 4.02626455e-01 6.18654728e-01
-9.50802565e-01 -5.65456338e-02 -1.03943372e+00 7.07643405e-02
-7.81857729e-01 2.81586885e-01 8.38852763e-01 7.60838687e-01
-1.22482634e+00 6.89408243e-01 -1.16453731e+00 -8.79841685e-01
7.92384088e-01 7.93194294e-01 -6.51997745e-01 5.23325279e-02
-9.91114438e-01 1.15418601e+00 6.45344108e-02 -2.01725543e-01
-7.46016085e-01 -8.88516724e-01 -7.45483637e-01 -6.35498762e-01
-3.32802087e-01 -9.47113931e-01 1.11407804e+00 -7.30094492e-01
-1.71207070e+00 1.33285880e+00 3.32318187e-01 -5.79022706e-01
2.40107954e-01 -8.22172165e-02 -4.92433429e-01 6.80595458e-01
1.40563518e-01 8.44118237e-01 7.28329003e-01 -4.32414174e-01
-3.19665194e-01 -1.13783324e+00 -1.82015523e-01 -2.35848069e-01
-3.27048928e-01 4.49390203e-01 4.42441911e-01 -8.79491687e-01
5.01748204e-01 -8.76292765e-01 -1.05830878e-01 5.33290565e-01
-1.67507276e-01 -6.65206984e-02 7.89566696e-01 -8.21032047e-01
6.64611399e-01 -1.58411026e+00 1.87521577e-01 -2.44377419e-01
8.38021934e-01 3.75013590e-01 2.98498701e-02 3.17155868e-02
-4.49014306e-01 2.79751599e-01 -2.24253088e-01 1.51227757e-01
-5.63437521e-01 -4.31894928e-01 5.80765426e-01 9.93279338e-01
9.02267620e-02 9.44739878e-01 -6.47993386e-01 -4.09327835e-01
5.38029134e-01 5.19851625e-01 -3.82027060e-01 2.75867116e-02
-8.72119442e-02 8.57378125e-01 -7.98938394e-01 9.14553404e-01
6.36930585e-01 -1.68740004e-01 -1.45311609e-01 -9.11180675e-02
6.78472146e-02 -1.77926093e-01 -4.98787425e-02 1.62810087e+00
-4.76237744e-01 4.61665273e-01 5.55950254e-02 -1.43765485e+00
9.12938058e-01 6.63263798e-01 1.00133157e+00 -9.49795246e-01
5.23010194e-01 3.91336501e-01 4.35212195e-01 -9.96409953e-01
-6.15752876e-01 -2.11533010e-01 5.87474525e-01 4.09484446e-01
2.30063513e-01 -1.69141635e-01 -1.99537620e-01 -5.42699695e-01
1.17236137e+00 -4.33254726e-02 2.96304137e-01 -4.12863910e-01
6.77527368e-01 -1.72590509e-01 1.22432716e-01 1.58050060e-01
-3.89513642e-01 7.33104944e-01 2.86154002e-01 -9.65528071e-01
-1.21532500e+00 -6.11900449e-01 -5.98547876e-01 7.59436667e-01
-5.25005758e-01 1.46504328e-01 -1.06035531e+00 -2.88346201e-01
-2.30046913e-01 -2.08103940e-01 -1.22220075e+00 -2.29016975e-01
-2.27766857e-01 -1.36380446e+00 6.31443858e-01 3.87605906e-01
5.16876698e-01 -1.02616835e+00 -7.34979689e-01 2.45582223e-01
4.23745155e-01 -9.37873304e-01 4.40135568e-01 2.02534497e-01
-1.22655165e+00 -1.20669234e+00 -1.20849073e+00 -8.75423431e-01
4.90819871e-01 -1.90988347e-01 6.69866562e-01 2.55936176e-01
-8.52771699e-01 4.37912971e-01 -8.00655782e-01 -8.56581926e-01
-3.13375622e-01 4.46805023e-02 3.55342031e-01 -4.18957740e-01
3.95094484e-01 -8.28107595e-01 -7.64055610e-01 -1.74560249e-01
-6.53295755e-01 2.26995483e-01 5.26871622e-01 8.47156823e-01
5.91572344e-01 -2.06816956e-01 7.82647848e-01 -7.09011495e-01
4.17027712e-01 -5.65016150e-01 -7.08387047e-02 1.46793857e-01
-5.68001270e-01 -2.66032279e-01 4.98542041e-01 -3.46250474e-01
-6.67332709e-01 -3.15579027e-01 -2.68668592e-01 -3.20729375e-01
-6.00751519e-01 4.16735291e-01 4.25468296e-01 -4.34456617e-01
7.23337054e-01 -1.52988240e-01 4.39611018e-01 -6.02620006e-01
-1.46109968e-01 6.04253054e-01 -1.38877138e-01 -5.33310950e-01
-5.45309708e-02 6.83183372e-01 2.25036278e-01 -9.29021597e-01
-7.14659393e-01 -9.98187140e-02 -9.14627314e-01 -3.28937978e-01
1.35664082e+00 -3.84700626e-01 -3.81324500e-01 9.40401554e-01
-1.08187258e+00 -4.24189389e-01 1.67358801e-01 8.18945825e-01
-7.71639109e-01 2.36630321e-01 -7.56545305e-01 -4.77221519e-01
-1.06541789e+00 -1.49746048e+00 1.22659242e+00 7.60481320e-03
-8.07242021e-02 -1.28544414e+00 4.80715990e-01 5.63073993e-01
3.96754593e-01 5.37991226e-01 1.51392078e+00 -7.47478008e-01
3.49499732e-01 -2.22983714e-02 -2.47728631e-01 8.89622346e-02
2.79377103e-01 2.05771551e-02 -1.12776315e+00 -8.75226632e-02
1.95085295e-02 -5.61713576e-01 6.76284075e-01 9.05156732e-01
1.58947074e+00 6.06201030e-02 -3.94748926e-01 4.89156038e-01
1.22094786e+00 6.60054088e-01 3.77967656e-01 4.14958894e-01
6.64990306e-01 8.03373754e-01 3.65483165e-01 4.38141793e-01
1.20427430e-01 6.84589803e-01 3.64524215e-01 -3.11180055e-01
-4.78243709e-01 5.77627420e-01 2.13994905e-02 1.18595326e+00
-4.39793438e-01 1.08591616e-01 -1.23255265e+00 2.74603784e-01
-1.54344380e+00 -6.19201541e-01 -3.00632060e-01 1.75117826e+00
4.44065869e-01 -2.96541393e-01 2.13965312e-01 1.94362432e-01
9.31624949e-01 -2.34253466e-01 -6.98167324e-01 -6.75482273e-01
-1.05818838e-01 8.62618983e-01 2.26021126e-01 -1.26004428e-01
-1.08234239e+00 8.19539666e-01 7.29601145e+00 6.80438280e-01
-1.56266987e+00 5.59279621e-01 8.36661518e-01 2.86657736e-02
1.89758852e-01 -7.72738278e-01 3.68431993e-02 3.92340988e-01
1.08881938e+00 4.32666004e-01 4.59593028e-01 3.99173915e-01
4.15446520e-01 -1.55771255e-01 -9.33606327e-01 9.23290908e-01
4.89383079e-02 -1.11987054e+00 -5.45120761e-02 1.03669845e-01
5.43345034e-01 4.90197241e-01 3.93448591e-01 -2.60232598e-01
-1.58207536e-01 -1.45057452e+00 3.53520453e-01 7.11183071e-01
1.52154005e+00 -3.34308416e-01 9.02622283e-01 -2.89307117e-01
-9.44788754e-01 -1.42144784e-01 -5.17733037e-01 9.80439708e-02
-5.86991429e-01 7.70008087e-01 -1.03478444e+00 5.75622916e-01
1.19340575e+00 8.82505834e-01 -6.05158627e-01 1.02097487e+00
2.03585982e-01 3.71493310e-01 3.00207622e-02 -2.10015904e-02
1.03633106e-01 -2.63946354e-01 1.18091613e-01 1.01225734e+00
5.38124084e-01 1.36040404e-01 -8.94882381e-02 9.68936205e-01
5.27607083e-01 7.53696144e-01 -8.53029013e-01 -3.18445325e-01
-8.19116309e-02 1.17525172e+00 -1.01120496e+00 -1.58774540e-01
-7.11793661e-01 4.88436520e-01 2.69351959e-01 -1.77824244e-01
-4.72808510e-01 6.84271976e-02 2.31676444e-01 3.15472662e-01
-3.44963431e-01 4.50547077e-02 -5.21583557e-01 -8.06895316e-01
-5.23300529e-01 -7.92959154e-01 1.95519283e-01 -9.59841132e-01
-1.39471793e+00 5.95984995e-01 -3.20353657e-02 -9.42918062e-01
1.89782500e-01 -9.47485209e-01 -6.09583557e-01 5.22894561e-01
-9.82474029e-01 -1.31174445e+00 -8.68709683e-02 3.57607305e-01
6.55945122e-01 -7.24151731e-01 1.13914037e+00 3.82753223e-01
-7.70147979e-01 1.71423376e-01 1.18413165e-01 2.48394702e-02
5.47793746e-01 -1.13899124e+00 -5.91218658e-02 -1.23480186e-01
-3.89241189e-01 2.34406605e-01 3.52563024e-01 -7.92074680e-01
-6.40144646e-01 -1.04276216e+00 3.21062863e-01 -9.36831627e-03
7.61903286e-01 1.24132425e-01 -7.16221154e-01 2.89163321e-01
5.36588021e-02 1.03736244e-01 1.00506389e+00 -8.24962780e-02
2.57862322e-02 5.32421730e-02 -1.56468225e+00 3.09211671e-01
1.00004721e+00 -3.76312107e-01 -5.58607817e-01 7.90659845e-01
3.51790816e-01 -2.32374936e-01 -1.24989402e+00 8.49025428e-01
7.64790058e-01 -9.57013309e-01 8.65144610e-01 -8.63610566e-01
6.24915838e-01 1.95431650e-01 1.64998934e-01 -1.24951708e+00
-4.06955600e-01 3.75668377e-01 -1.12321198e-01 5.40270567e-01
1.71635032e-01 -6.56527638e-01 7.09195197e-01 1.93211064e-01
-2.28049621e-01 -1.33040226e+00 -1.10889494e+00 -5.38537860e-01
5.78664243e-01 -2.75070816e-01 5.09929240e-01 9.09562528e-01
1.60647422e-01 -4.94896583e-02 1.09335579e-01 -2.12817460e-01
1.16440192e-01 -4.44969594e-01 -1.91752151e-01 -1.68026423e+00
2.10548118e-01 -6.36317492e-01 -1.04058802e+00 3.74323308e-01
3.53978842e-01 -1.01517856e+00 -5.14950156e-01 -1.68238449e+00
3.44458371e-01 -5.41453242e-01 -2.95957625e-01 4.14168954e-01
4.96822804e-01 4.07005876e-01 -4.33591455e-01 2.94824168e-02
-2.01430634e-01 5.09363294e-01 1.58111012e+00 -4.94090289e-01
1.02568343e-01 -1.72290027e-01 -2.38973677e-01 9.86833513e-01
1.26915681e+00 -3.65863442e-01 -5.48283160e-01 -4.29963142e-01
1.75436050e-01 2.03905702e-01 1.04956500e-01 -1.44146860e+00
-1.98545814e-01 9.86293182e-02 7.14790285e-01 -2.32766926e-01
2.56974161e-01 -6.30264699e-01 -6.02612048e-02 8.02277863e-01
-2.32265681e-01 1.48284450e-01 8.67409557e-02 6.11678511e-02
8.67712498e-02 -3.11595142e-01 1.25983894e+00 -3.44432622e-01
-1.04670338e-01 7.74718404e-01 -9.11281705e-01 -2.84592032e-01
9.31358755e-01 -2.92068273e-01 -2.69032687e-01 4.79830466e-02
-1.21308053e+00 -3.36037666e-01 1.45950720e-01 3.63739729e-01
7.66735733e-01 -1.23494995e+00 -5.79186976e-01 -1.77942351e-01
1.38864174e-01 -7.32578561e-02 6.98028982e-01 1.54953837e+00
-1.10353637e+00 8.71539533e-01 -9.80786562e-01 -7.50594079e-01
-1.23836577e+00 6.97462797e-01 7.58338511e-01 -4.63673808e-02
-7.93093026e-01 7.30369687e-01 4.72250342e-01 -5.56984007e-01
2.16667401e-03 -4.29498106e-01 -1.04196298e+00 -3.09221912e-02
5.02178848e-01 6.05792999e-01 6.07128501e-01 -5.64015806e-01
-4.60374296e-01 1.03218007e+00 -1.38210893e-01 8.58738348e-02
1.73814034e+00 -8.92343000e-02 -6.97624147e-01 5.56450009e-01
9.57140326e-01 -6.39357686e-01 -3.33198488e-01 1.54445440e-01
-4.45521325e-02 2.38291830e-01 3.61081809e-01 -1.04793870e+00
-1.76853621e+00 1.18655527e+00 1.52611923e+00 7.42258132e-02
9.33007300e-01 7.22280219e-02 7.56410241e-01 2.58983988e-02
6.73696756e-01 -1.06791449e+00 2.11896047e-01 2.03591600e-01
1.23070920e+00 -1.14763737e+00 1.37550160e-02 -5.06408632e-01
-1.76459566e-01 1.64624679e+00 9.77343261e-01 -1.44453987e-01
1.07420897e+00 2.87173480e-01 1.74663097e-01 -7.83027232e-01
-4.62554097e-01 -9.78308823e-03 2.64533669e-01 1.21491444e+00
9.33967948e-01 2.86629885e-01 -6.35441601e-01 6.10779226e-01
-1.60654336e-01 2.37910170e-02 3.73477221e-01 7.13648617e-01
-5.71353078e-01 -1.08269215e+00 -5.53573966e-01 1.09001184e+00
-8.37179244e-01 -9.33570862e-02 -7.16901839e-01 3.39070171e-01
6.56700850e-01 7.94390619e-01 6.66789431e-03 -5.67701280e-01
-2.84558088e-01 1.85420252e-02 9.43225741e-01 -1.01361489e+00
-3.04234356e-01 -5.84389195e-02 9.86859798e-02 -4.26535964e-01
-6.91884696e-01 -5.62421381e-01 -1.32595348e+00 -2.56608248e-01
-2.60269374e-01 -4.96334434e-01 8.15886199e-01 1.27825677e+00
1.47016242e-01 8.85031462e-01 8.21484253e-02 -6.50719166e-01
4.72800016e-01 -1.32068121e+00 -1.02440846e+00 -4.56394911e-01
1.66961595e-01 -1.28009486e+00 1.60249099e-01 -2.42406860e-01] | [14.129669189453125, -1.6571358442306519] |
2c4e6593-bf31-4645-9ab0-7b1cf159a78b | first-take-all-temporal-order-preserving | 1506.02184 | null | http://arxiv.org/abs/1506.02184v1 | http://arxiv.org/pdf/1506.02184v1.pdf | First-Take-All: Temporal Order-Preserving Hashing for 3D Action Videos | With the prevalence of the commodity depth cameras, the new paradigm of user
interfaces based on 3D motion capturing and recognition have dramatically
changed the way of interactions between human and computers. Human action
recognition, as one of the key components in these devices, plays an important
role to guarantee the quality of user experience. Although the model-driven
methods have achieved huge success, they cannot provide a scalable solution for
efficiently storing, retrieving and recognizing actions in the large-scale
applications. These models are also vulnerable to the temporal translation and
warping, as well as the variations in motion scales and execution rates. To
address these challenges, we propose to treat the 3D human action recognition
as a video-level hashing problem and propose a novel First-Take-All (FTA)
Hashing algorithm capable of hashing the entire video into hash codes of fixed
length. We demonstrate that this FTA algorithm produces a compact
representation of the video invariant to the above mentioned variations,
through which action recognition can be solved by an efficient nearest neighbor
search by the Hamming distance between the FTA hash codes. Experiments on the
public 3D human action datasets shows that the FTA algorithm can reach a
recognition accuracy higher than 80%, with about 15 bits per frame considering
there are 65 frames per video over the datasets. | ['Guo-Jun Qi', 'Jun Ye', 'Hao Hu', 'Kien A. Hua', 'Kai Li'] | 2015-06-06 | null | null | null | null | ['3d-human-action-recognition'] | ['computer-vision'] | [ 1.80061311e-01 -6.49335682e-01 -3.28732371e-01 -6.77101463e-02
-5.88436246e-01 -4.15894777e-01 4.68054235e-01 -9.64114256e-03
-4.90680724e-01 1.57040060e-01 2.67481387e-01 7.51139596e-02
7.17374012e-02 -6.37322426e-01 -4.38528627e-01 -8.45106244e-01
-3.59475315e-01 5.07264808e-02 7.51104355e-01 7.94836134e-02
4.39025730e-01 5.63499212e-01 -1.94280529e+00 2.86215782e-01
9.01288986e-02 1.28302288e+00 7.50287494e-04 8.37884963e-01
1.15348652e-01 5.74128330e-01 -3.83174390e-01 -2.15852439e-01
5.46809852e-01 -2.10325897e-01 -4.76508141e-01 2.93017507e-01
1.62075728e-01 -8.05637658e-01 -9.48119640e-01 8.99309933e-01
5.39267361e-01 7.73103461e-02 2.97479033e-01 -1.52177989e+00
-3.77945065e-01 4.27382104e-02 -6.12011075e-01 2.22904786e-01
8.19306910e-01 1.96575269e-01 6.47604764e-01 -8.52873802e-01
4.56851959e-01 1.29989898e+00 5.64743578e-01 5.39102852e-01
-7.26801515e-01 -5.13450325e-01 -4.21260625e-01 6.48594499e-01
-1.71267903e+00 -3.98270190e-01 3.76710922e-01 -3.45090270e-01
1.07915652e+00 2.10536912e-01 9.08713281e-01 7.69590259e-01
4.30944055e-01 6.85423255e-01 5.73183656e-01 -3.11906576e-01
2.46311516e-01 -2.77838200e-01 1.30685130e-02 6.58336699e-01
3.70527029e-01 -1.46989673e-01 -8.23642731e-01 -3.70958328e-01
8.55184972e-01 4.78077620e-01 -4.51080561e-01 -5.29778719e-01
-1.41000783e+00 6.31030858e-01 2.69934218e-02 1.18060797e-01
-4.22063410e-01 2.62495518e-01 6.23591304e-01 1.18762583e-01
-2.11921319e-01 -2.34564260e-01 -2.53111292e-02 -6.26408339e-01
-6.87939465e-01 2.68132895e-01 4.85158533e-01 1.02919233e+00
6.23355031e-01 -3.01399559e-01 -7.10432678e-02 3.17168236e-01
2.99210370e-01 8.07080507e-01 8.23536456e-01 -8.25934172e-01
4.88070965e-01 7.09516644e-01 2.13050678e-01 -1.45834041e+00
-1.26302987e-01 5.53119004e-01 -8.79485905e-01 2.25927848e-02
2.54407942e-01 5.10639071e-01 -5.86522400e-01 1.37444460e+00
6.24312103e-01 1.92204818e-01 -9.71077159e-02 9.01634634e-01
3.45518023e-01 8.17066252e-01 -1.32381454e-01 -2.06508964e-01
1.49518633e+00 -7.00136662e-01 -6.74135983e-01 9.22803357e-02
7.42699921e-01 -7.58297920e-01 7.82171369e-01 3.62260461e-01
-9.53102589e-01 -7.72454500e-01 -1.31646025e+00 -5.85887805e-02
-1.47198960e-01 -2.13418439e-01 3.44903260e-01 8.19034696e-01
-9.21270370e-01 3.70286077e-01 -1.15132654e+00 -4.56681669e-01
6.58816695e-02 4.70678747e-01 -7.69323945e-01 -2.84026146e-01
-1.11080062e+00 5.89856148e-01 4.59522188e-01 3.34412940e-02
-6.81879878e-01 -4.10679914e-02 -7.94994235e-01 -5.39199449e-02
4.23442312e-02 -2.50059068e-01 9.91237760e-01 -4.46869940e-01
-1.21898913e+00 8.97923410e-01 -2.85304844e-01 -5.92580616e-01
2.91728795e-01 -2.87396252e-01 -4.78928447e-01 4.78937536e-01
-1.45116881e-01 4.06773537e-01 1.03844655e+00 -3.06888372e-01
-8.36592972e-01 -7.10099638e-01 -1.38848955e-02 1.73765108e-01
-6.18077278e-01 -8.13347176e-02 -5.52105010e-01 -3.65354240e-01
2.32372820e-01 -1.22416115e+00 1.35412052e-01 3.17263007e-01
4.66145128e-02 -8.56856108e-02 1.27240884e+00 -3.56993198e-01
1.48792410e+00 -2.51747394e+00 1.99741229e-01 3.79038677e-02
-8.33983943e-02 3.23554605e-01 1.86790884e-01 5.37380576e-01
2.74456382e-01 -2.28107825e-01 -3.45512517e-02 -1.16062686e-01
8.89392421e-02 1.64313018e-01 -2.99618840e-01 7.49936104e-01
-2.20850170e-01 4.93801594e-01 -6.37471139e-01 -6.44231498e-01
3.56026411e-01 7.16415167e-01 -5.42375386e-01 3.88080597e-01
4.90829736e-01 -6.42204508e-02 -5.12961328e-01 8.06565225e-01
6.93173289e-01 -7.21207857e-02 1.36624679e-01 -1.89610675e-01
-1.98130328e-02 -1.41174480e-01 -1.49688077e+00 1.72281623e+00
1.18481569e-01 4.23546821e-01 -2.03387037e-01 -7.98835158e-01
9.17696953e-01 4.33480352e-01 6.52768850e-01 -6.60310566e-01
-4.68878411e-02 3.68296206e-01 -4.46474135e-01 -5.15414715e-01
5.69570780e-01 1.50990739e-01 -3.08805346e-01 5.33129513e-01
-2.02827245e-01 2.79386759e-01 -4.34168540e-02 8.94044787e-02
1.41235018e+00 -1.55249268e-01 6.36272728e-01 1.90257043e-01
8.37481439e-01 -1.34845883e-01 5.64476490e-01 3.85399610e-01
-6.82385802e-01 5.79043031e-01 9.08589065e-02 -8.63853037e-01
-1.08283782e+00 -8.68550181e-01 2.23289635e-02 5.79880953e-01
4.10007924e-01 -5.78104079e-01 -8.40234756e-01 -3.86259556e-01
1.31786451e-01 -1.56371996e-01 -3.34657252e-01 -5.06449461e-01
-6.67755187e-01 -3.55205894e-01 6.77620947e-01 4.58900124e-01
1.00764954e+00 -8.82281423e-01 -1.36408377e+00 1.68919131e-01
-2.32686087e-01 -1.46166730e+00 -7.23821044e-01 -3.20427030e-01
-1.01294434e+00 -1.07928944e+00 -8.16910803e-01 -7.03218699e-01
3.19836795e-01 8.07972610e-01 4.97193277e-01 2.01732203e-01
-5.29075623e-01 5.75474381e-01 -7.03251898e-01 2.39993706e-01
-7.08883703e-02 -2.24093750e-01 5.16871572e-01 3.30260813e-01
7.76952684e-01 -5.05700827e-01 -8.49907875e-01 4.46589381e-01
-1.15190566e+00 -2.88534254e-01 5.07764280e-01 5.38079500e-01
6.06808066e-01 2.85921812e-01 -8.22348520e-02 -4.60422523e-02
6.88740909e-02 -3.26082259e-01 -3.92492026e-01 1.56876817e-01
-3.17213684e-01 1.28306448e-02 3.80502492e-01 -5.71505845e-01
-3.88687134e-01 3.99162084e-01 1.22454382e-01 -5.38201571e-01
-1.04945153e-01 -3.36602777e-02 -3.05426985e-01 -2.80760407e-01
2.96337962e-01 6.34029984e-01 1.67240426e-02 -2.67798156e-01
5.36799058e-02 9.91221488e-01 5.65127015e-01 -1.51842281e-01
8.25039864e-01 6.83695614e-01 2.42963150e-01 -9.98185933e-01
-1.06593378e-01 -6.29720032e-01 -7.26599097e-01 -2.85735607e-01
1.18251824e+00 -9.59312141e-01 -9.61492181e-01 9.26222980e-01
-1.14401913e+00 1.76301137e-01 9.38935727e-02 6.62296772e-01
-8.41039121e-01 8.53342712e-01 -6.00147307e-01 -8.56238127e-01
-3.19738567e-01 -1.34366810e+00 1.19277072e+00 1.27495959e-01
-1.16616242e-01 -4.46844310e-01 8.33081678e-02 3.07802200e-01
1.69573635e-01 3.20462614e-01 8.88763785e-01 -3.10208857e-01
-8.76873314e-01 -5.43429554e-01 2.68684532e-02 1.50996357e-01
2.75200665e-01 -9.29079130e-02 -6.05404973e-01 -4.73098814e-01
1.90092564e-01 -2.54692078e-01 2.76337028e-01 6.26508668e-02
8.50275397e-01 -2.70641267e-01 -1.93433940e-01 5.30955195e-01
1.40622675e+00 4.63847429e-01 1.00169444e+00 4.04912621e-01
5.41021824e-01 1.36149660e-01 8.70435297e-01 9.91414070e-01
2.85589010e-01 9.91286814e-01 4.33555335e-01 5.40165067e-01
1.30343184e-01 -3.31387341e-01 7.07513273e-01 8.78137708e-01
-1.06504604e-01 -2.85901949e-02 -7.39264905e-01 4.17535931e-01
-1.82455194e+00 -1.29244912e+00 1.80459559e-01 2.71204495e+00
4.92303193e-01 1.45160943e-01 2.72451252e-01 7.42841661e-01
9.12161946e-01 3.33541751e-01 -3.92875314e-01 -3.19159359e-01
1.35008901e-01 7.30392039e-02 5.10602713e-01 1.35885358e-01
-1.08204281e+00 5.49334347e-01 6.27761793e+00 6.64280057e-01
-1.03039730e+00 -6.43880814e-02 4.45819907e-02 5.35519011e-02
2.64481187e-01 -1.01346985e-01 -1.00439811e+00 6.39269948e-01
9.91543651e-01 -7.48922378e-02 1.80672228e-01 9.94302690e-01
1.65092453e-01 -1.69948921e-01 -9.86287892e-01 1.55196691e+00
2.17895031e-01 -9.96558547e-01 4.84004319e-01 2.27647468e-01
2.72268683e-01 -4.60530668e-01 -8.46630186e-02 1.63737312e-02
-5.15689790e-01 -7.27659702e-01 7.33080447e-01 2.64186859e-01
7.71142304e-01 -8.01042616e-01 5.70746899e-01 3.57846767e-01
-1.72648549e+00 -2.80552208e-01 -5.96592546e-01 -4.56727147e-02
3.03776294e-01 2.93190330e-01 -3.71050775e-01 1.53778762e-01
9.36715066e-01 6.88822806e-01 -2.65009999e-01 9.71169770e-01
4.11101282e-01 -4.95655090e-02 -2.28776902e-01 -9.55652446e-02
7.92216435e-02 5.82877286e-02 3.49491000e-01 9.26972270e-01
7.74654627e-01 5.86245656e-01 8.35150778e-02 -1.16447834e-02
1.98207289e-01 8.21084231e-02 -8.27112436e-01 -1.74903169e-01
5.07498622e-01 7.22901404e-01 -6.83225751e-01 -1.87947720e-01
-6.14199877e-01 1.49760580e+00 -9.44744051e-02 -1.37087062e-01
-9.29327965e-01 -6.11955285e-01 9.66611087e-01 3.15505475e-01
6.20363057e-01 -7.47861683e-01 4.78275865e-01 -1.12229717e+00
3.38354051e-01 -1.16081858e+00 4.84083235e-01 -4.66247916e-01
-6.39506042e-01 3.28458190e-01 -1.37847200e-01 -1.68768811e+00
-1.47605896e-01 -5.95274568e-01 3.12625766e-02 2.28473529e-01
-1.00259447e+00 -6.61404192e-01 -6.03476346e-01 1.17342651e+00
5.60802817e-01 -4.47537638e-02 9.23636854e-01 5.81395149e-01
-2.92381853e-01 5.19673049e-01 -1.69899676e-03 1.25584707e-01
6.46871388e-01 -5.65480173e-01 4.28133070e-01 6.99988008e-01
1.88573927e-01 4.82561529e-01 5.29879451e-01 -5.39893866e-01
-2.19875908e+00 -5.33660293e-01 9.02582645e-01 -1.97220743e-01
3.93195957e-01 -2.20279261e-01 -8.05002213e-01 5.35898685e-01
-3.08941483e-01 1.99156404e-01 7.03474581e-01 -7.19791591e-01
-5.51272631e-01 -3.13368201e-01 -1.24222505e+00 3.14290434e-01
1.19064438e+00 -7.30851948e-01 -5.12895167e-01 1.01508103e-01
6.10721946e-01 -3.76130044e-01 -9.43192959e-01 3.74892414e-01
8.39171767e-01 -1.33883548e+00 1.04690790e+00 -2.16302425e-01
8.14019740e-02 -5.31394899e-01 -6.68168187e-01 -3.53397012e-01
-2.90135413e-01 -6.48274124e-01 -4.73279238e-01 8.26485813e-01
-3.11686784e-01 -3.44143033e-01 8.32723618e-01 6.67190611e-01
4.05015141e-01 -4.61702883e-01 -1.37040830e+00 -7.95184970e-01
-7.16690004e-01 -1.96401224e-01 6.14655733e-01 4.26785946e-01
2.41741911e-01 -3.10592711e-01 -6.46331608e-01 2.37236410e-01
7.24566579e-01 1.86639186e-02 9.35469627e-01 -8.71599019e-01
-2.90756762e-01 3.15718465e-02 -1.53481567e+00 -1.40588737e+00
-2.58412033e-01 -3.29910576e-01 -3.55891809e-02 -9.83625531e-01
4.18015063e-01 -1.13399267e-01 -1.73953190e-01 1.36162430e-01
1.97332397e-01 4.22880769e-01 1.95645377e-01 6.04472220e-01
-6.76365197e-01 6.22251153e-01 7.92141676e-01 -5.11196405e-02
9.19397697e-02 -1.42956749e-01 1.46788001e-01 5.31808197e-01
2.62941569e-01 -4.25402343e-01 -4.82767016e-01 -4.48583394e-01
-6.61129579e-02 2.37341881e-01 2.90934354e-01 -1.49399316e+00
4.28200155e-01 -1.12116396e-01 2.02598363e-01 -6.54153883e-01
6.13972127e-01 -1.12233126e+00 3.08862358e-01 9.23711777e-01
-5.06090857e-02 6.49892926e-01 -4.96643335e-02 6.98635221e-01
-3.87067199e-01 -2.94239949e-02 7.01588929e-01 -6.29859343e-02
-1.08850718e+00 4.18172091e-01 -4.78504211e-01 -2.20728949e-01
1.34317911e+00 -6.51933968e-01 -3.42498580e-03 -4.07984376e-01
-3.11307311e-01 -2.22455263e-01 7.10372567e-01 4.36306834e-01
9.84037519e-01 -1.76302624e+00 -1.97434947e-01 4.34186757e-01
2.05737695e-01 -3.27116877e-01 4.78039205e-01 6.00110054e-01
-7.98769712e-01 6.43087089e-01 -6.02365434e-01 -6.57379389e-01
-1.59413397e+00 6.42004430e-01 6.75849468e-02 -1.14902802e-01
-6.93982542e-01 3.86779845e-01 -2.32185811e-01 3.03208292e-01
5.50495982e-01 -1.13800541e-01 1.19177096e-01 -1.58821449e-01
1.04219854e+00 5.96723557e-01 -7.89294690e-02 -9.78050232e-01
-5.40320754e-01 1.02686691e+00 1.08743466e-01 -1.70362249e-01
9.18385565e-01 -3.72072756e-01 7.35096037e-02 3.69663268e-01
1.44845605e+00 -3.26601118e-01 -1.02508712e+00 -1.41335368e-01
-1.12326175e-01 -9.68911707e-01 -2.36739039e-01 2.53841430e-01
-8.65696371e-01 9.21528518e-01 8.75791550e-01 1.95850968e-01
1.23311186e+00 -4.48114187e-01 1.57986343e+00 4.15305823e-01
9.69339550e-01 -8.95651996e-01 2.33282954e-01 2.70973712e-01
4.18749571e-01 -1.04977334e+00 1.01509735e-01 -1.57331333e-01
-4.08929199e-01 1.16206145e+00 3.51163685e-01 -1.40263364e-01
7.10077882e-01 7.11439848e-02 -1.81643143e-01 9.63270962e-02
-5.17628729e-01 1.48302317e-01 -3.14757042e-02 5.21211743e-01
-1.38151878e-02 -1.77132741e-01 -3.96453291e-01 1.48176819e-01
2.05866441e-01 1.04886986e-01 3.36404055e-01 1.26938021e+00
-7.26992607e-01 -1.14407194e+00 -5.62254667e-01 -2.22380057e-01
-3.25980425e-01 2.58953214e-01 -8.33343491e-02 6.79760754e-01
-6.85851127e-02 7.54529059e-01 3.84323373e-02 -8.62397909e-01
3.87054205e-01 1.94988772e-01 5.04889667e-01 -2.04266831e-01
-1.47746459e-01 -3.38644058e-01 -5.45714557e-01 -9.98538852e-01
-4.86196816e-01 -6.66737974e-01 -1.36878729e+00 -6.00407958e-01
1.48954779e-01 -4.94839512e-02 5.36873162e-01 7.03648567e-01
5.04097402e-01 -4.16506052e-01 8.89567614e-01 -8.93421113e-01
-8.24776709e-01 -3.94070476e-01 -8.17902565e-01 7.32762575e-01
2.94461399e-01 -7.73231089e-01 -4.81742710e-01 1.30608544e-01] | [8.09731388092041, 0.20388652384281158] |
4c21823f-1e26-4cf7-960f-6c594a6e7a02 | a-framework-for-combining-entity-resolution | 2303.07469 | null | https://arxiv.org/abs/2303.07469v1 | https://arxiv.org/pdf/2303.07469v1.pdf | A Framework for Combining Entity Resolution and Query Answering in Knowledge Bases | We propose a new framework for combining entity resolution and query answering in knowledge bases (KBs) with tuple-generating dependencies (tgds) and equality-generating dependencies (egds) as rules. We define the semantics of the KB in terms of special instances that involve equivalence classes of entities and sets of values. Intuitively, the former collect all entities denoting the same real-world object, while the latter collect all alternative values for an attribute. This approach allows us to both resolve entities and bypass possible inconsistencies in the data. We then design a chase procedure that is tailored to this new framework and has the feature that it never fails; moreover, when the chase procedure terminates, it produces a universal solution, which in turn can be used to obtain the certain answers to conjunctive queries. We finally discuss challenges arising when the chase does not terminate. | ['Federico Scafoglieri', 'Lucian Popa', 'Domenico Lembo', 'Phokion G. Kolaitis', 'Ronald Fagin'] | 2023-03-13 | null | null | null | null | ['entity-resolution'] | ['natural-language-processing'] | [-1.60420910e-01 5.06135106e-01 -2.32418239e-01 -3.86226624e-01
-6.13288403e-01 -8.10969055e-01 3.52404356e-01 5.17590761e-01
-2.13142827e-01 1.24850535e+00 -2.16409937e-02 -3.42465788e-01
-4.50776160e-01 -1.56842160e+00 -5.92884004e-01 -3.26106071e-01
-5.30716591e-02 9.53800976e-01 7.52649963e-01 -3.77425879e-01
-3.15588564e-01 5.51862597e-01 -1.73716521e+00 2.74163246e-01
7.85961926e-01 8.65970671e-01 -1.90167829e-01 2.87701100e-01
-4.37906981e-01 9.04501438e-01 -5.22068560e-01 -7.92597115e-01
2.65332192e-01 7.54021034e-02 -1.27871954e+00 -2.68512249e-01
6.58145025e-02 -1.61920786e-01 1.01630397e-01 1.02993357e+00
1.25675246e-01 1.14057586e-01 2.99722165e-01 -1.59095323e+00
-4.82629567e-01 5.96220791e-01 1.42060399e-01 -8.31831694e-02
8.00915599e-01 -3.68782759e-01 1.57106566e+00 -6.84185326e-01
1.11758077e+00 1.07121551e+00 1.95403069e-01 4.53165442e-01
-1.26723862e+00 -1.46802664e-01 -8.56374204e-02 4.56140488e-01
-1.70665908e+00 -5.49068928e-01 1.31015599e-01 9.41639841e-02
1.11246109e+00 9.33735371e-01 4.08372015e-01 1.63424417e-01
-2.65391350e-01 4.91959363e-01 7.17640460e-01 -6.94227040e-01
5.85955501e-01 5.51945388e-01 4.67837632e-01 4.44330335e-01
8.21865737e-01 -3.30205292e-01 -5.42065263e-01 -5.53022981e-01
4.01455581e-01 -3.64153504e-01 -2.34372437e-01 -6.41013920e-01
-9.10229385e-01 6.84135199e-01 1.13442399e-01 3.19425702e-01
-3.39240909e-01 -1.22483388e-01 1.88188612e-01 4.74942297e-01
-3.39661449e-01 5.97362220e-01 -6.55382097e-01 4.10798132e-01
-2.84553796e-01 7.40447938e-01 1.39316940e+00 1.38665020e+00
1.04974782e+00 -6.58720791e-01 1.52170928e-02 4.52799380e-01
2.03451172e-01 4.74319458e-01 -3.80681366e-01 -1.09769022e+00
2.91573346e-01 9.62018669e-01 8.72944772e-01 -7.73811758e-01
-2.72657186e-01 -6.80648312e-02 -2.56086648e-01 -1.81289211e-01
3.87052774e-01 1.54527292e-01 -4.75717247e-01 2.14338708e+00
8.58911455e-01 -9.27707106e-02 6.61125004e-01 7.37185657e-01
9.27959740e-01 4.34256285e-01 1.39443800e-01 -5.98440230e-01
1.58196175e+00 -3.65108818e-01 -8.80402386e-01 7.05273706e-05
1.02244258e+00 -5.36607876e-02 4.99329001e-01 -1.12844771e-02
-1.22713554e+00 -7.15347826e-02 -9.26434398e-01 -2.54660189e-01
-7.28880644e-01 -5.66841483e-01 6.46151066e-01 4.94505972e-01
-1.12880576e+00 -2.49911882e-02 -4.18064594e-01 -2.36516625e-01
-2.51549244e-01 6.18114352e-01 -6.69370413e-01 -1.95966914e-01
-1.73537230e+00 1.00009441e+00 9.23407376e-01 -3.33657227e-02
-2.81289965e-01 -4.46038574e-01 -8.74026954e-01 2.25862309e-01
1.18198848e+00 -1.04479897e+00 9.81380165e-01 -2.50899702e-01
-7.73352802e-01 1.04557586e+00 -4.23361361e-01 -4.02500272e-01
8.73323679e-02 3.78090888e-03 -8.85843933e-01 4.36363416e-03
4.55784559e-01 3.60116631e-01 2.49766242e-02 -1.24620867e+00
-1.10597575e+00 -7.30287194e-01 7.72880554e-01 2.23257482e-01
2.42555905e-02 8.19965973e-02 -7.56095111e-01 2.45644778e-01
3.91595155e-01 -6.68053210e-01 -1.18826013e-02 -3.18774730e-01
-7.35934734e-01 -3.80845577e-01 3.36707294e-01 -3.59180272e-01
1.65195620e+00 -1.84399307e+00 3.67896467e-01 7.66623318e-01
2.40016207e-01 5.55144660e-02 1.39868364e-01 6.58787370e-01
9.83394980e-02 3.85288894e-01 -3.03686440e-01 1.45213768e-01
1.55409083e-01 8.99160743e-01 -5.64325154e-01 -1.57496989e-01
2.44109601e-01 1.00296772e+00 -6.56160831e-01 -6.26243591e-01
-2.70283192e-01 -2.71239787e-01 -7.03338981e-01 1.77908003e-01
-7.37532675e-01 4.26963717e-02 -6.81597054e-01 6.95615649e-01
8.06344032e-01 -2.16603637e-01 1.05043983e+00 -1.91381961e-01
-1.92787111e-01 3.89559597e-01 -1.96977496e+00 1.16511691e+00
-5.78326620e-02 -3.85174811e-01 1.19059183e-01 -6.46445870e-01
6.69229448e-01 2.13570610e-01 3.04597616e-01 -6.64921463e-01
-3.63292783e-01 5.95838547e-01 -3.11103255e-01 -6.95794106e-01
7.34028518e-01 -1.00260504e-01 -3.93082231e-01 4.17054594e-01
-1.07660636e-01 4.67946008e-02 5.06203949e-01 4.64484990e-01
1.01486421e+00 -9.27328020e-02 7.00284481e-01 -1.88003674e-01
1.05794764e+00 2.30377719e-01 9.85583603e-01 9.30284858e-01
3.35076094e-01 -1.99089244e-01 8.68780375e-01 -4.72162098e-01
-7.34757006e-01 -1.26655769e+00 -1.29470676e-01 8.97626340e-01
4.60624874e-01 -8.75552297e-01 -3.29167366e-01 -7.58233547e-01
2.81611562e-01 8.20956707e-01 -4.50093597e-01 -5.46996063e-03
-5.50602973e-01 -2.53339380e-01 4.67813909e-01 4.71950829e-01
1.90072581e-01 -9.92310464e-01 -7.61809886e-01 2.49070480e-01
-4.59963053e-01 -1.21713042e+00 2.04142302e-01 8.17419440e-02
-5.89212894e-01 -1.37414718e+00 3.26949298e-01 -6.69089139e-01
6.50623381e-01 -1.22000642e-01 1.24969149e+00 4.60851282e-01
1.69295341e-01 2.18616873e-01 -1.96217850e-01 -2.94096172e-01
-2.62244165e-01 7.35210348e-03 -1.08547416e-02 -9.59377587e-02
6.13091409e-01 -3.68922174e-01 -1.36515945e-02 2.29724362e-01
-1.29063129e+00 -2.73483306e-01 -8.23646411e-02 6.30679190e-01
8.79642069e-01 3.21321815e-01 7.38288820e-01 -1.39908135e+00
4.61417526e-01 -6.68671906e-01 -8.35256815e-01 1.04190063e+00
-6.24828577e-01 3.54814976e-01 4.60572869e-01 1.97539225e-01
-1.10204470e+00 -2.23209605e-01 1.12993933e-01 1.43868297e-01
-8.85155946e-02 1.06820226e+00 -7.45068967e-01 3.53887945e-01
5.29723525e-01 -4.53721322e-02 -3.91105026e-01 -4.30806220e-01
6.08985364e-01 4.00271982e-01 8.44381511e-01 -1.07324409e+00
6.22057974e-01 5.30082107e-01 3.36965054e-01 -2.72919655e-01
-7.42947459e-01 -3.94648403e-01 -5.40812552e-01 1.18066460e-01
4.51752365e-01 -6.85206532e-01 -1.01380408e+00 1.22662708e-01
-1.12614071e+00 1.50271714e-01 -6.26672208e-01 7.28038698e-02
-4.71496761e-01 2.31049895e-01 -2.69464105e-01 -9.89111125e-01
4.31774603e-03 -8.62720370e-01 6.67278469e-01 1.31987929e-01
-3.01054329e-01 -7.34674275e-01 1.72088519e-02 1.47343323e-01
1.30731240e-01 2.32530087e-01 1.38754761e+00 -1.20154297e+00
-8.62469554e-01 -2.44654909e-01 -1.56966716e-01 -1.31080687e-01
5.61263524e-02 -1.55946821e-01 -4.90058184e-01 5.17794415e-02
-2.08515480e-01 -1.44094259e-01 1.56543076e-01 -1.48107469e-01
4.32924896e-01 -4.81869847e-01 -5.59503913e-01 1.64279982e-01
1.58538747e+00 2.55186856e-01 9.15916383e-01 4.37353462e-01
2.24800721e-01 7.95875967e-01 7.38585114e-01 2.40202457e-01
8.26515436e-01 9.17753875e-01 3.13660949e-01 3.94562989e-01
4.32376862e-01 -1.35227680e-01 -2.82119870e-01 2.47502774e-01
-2.00702116e-01 -2.88474768e-01 -8.34909558e-01 7.12746561e-01
-2.19284940e+00 -9.84056950e-01 -3.60389650e-01 2.56327558e+00
1.19852149e+00 -1.56859338e-01 6.67056963e-02 1.44163713e-01
6.05667889e-01 -2.58642465e-01 -4.26110536e-01 -4.37658191e-01
-3.32883358e-01 3.82227391e-01 2.29386404e-01 8.44768286e-01
-5.81455171e-01 9.72895741e-01 6.46424580e+00 7.79999346e-02
-5.60386479e-01 -6.52105082e-03 -2.36864462e-01 2.45705262e-01
-8.84179831e-01 5.05720198e-01 -9.73310471e-01 -1.01422407e-01
7.64955819e-01 -4.45355117e-01 3.96364599e-01 6.57337785e-01
-5.14827609e-01 -3.44450206e-01 -1.32837081e+00 3.50215673e-01
-3.65549743e-01 -1.36244464e+00 1.20845839e-01 -3.61070782e-02
4.91793364e-01 -5.38211882e-01 -4.37915325e-01 3.03216577e-01
3.82990479e-01 -5.72307646e-01 7.54774570e-01 5.84042668e-01
7.27049053e-01 -7.70433247e-01 7.56004274e-01 4.18582082e-01
-1.17500412e+00 -3.18232208e-01 -2.98758954e-01 1.46510661e-01
3.28096271e-01 5.27197838e-01 -7.77209163e-01 1.31717718e+00
5.41199088e-01 7.28672594e-02 -1.75429150e-01 8.34326625e-01
-3.40357274e-01 -1.51238278e-01 -6.42990291e-01 3.34281892e-01
-2.91713715e-01 -3.50043118e-01 5.94812393e-01 9.51766789e-01
1.62709340e-01 4.55164343e-01 -4.78388593e-02 8.52817953e-01
-2.97681335e-02 5.42430840e-02 -4.50798750e-01 3.42664599e-01
1.09501517e+00 8.69908154e-01 -3.34257156e-01 -7.60917425e-01
-5.24781644e-01 4.88078594e-01 6.90267563e-01 6.00834966e-01
-5.41531205e-01 -2.85064816e-01 7.07735896e-01 -1.67564005e-02
3.55879128e-01 1.52279586e-01 -6.97560385e-02 -1.32475770e+00
4.37492281e-01 -8.68659616e-01 1.21297538e+00 -7.74294734e-01
-1.00693393e+00 3.56680363e-01 3.67550761e-01 -5.18641233e-01
-5.61930358e-01 -3.29322785e-01 -1.35008842e-01 8.58367980e-01
-1.73780215e+00 -8.57959628e-01 -3.89688499e-02 9.00291026e-01
-4.91502523e-01 5.58211505e-01 1.32938159e+00 2.98606724e-01
-4.61961150e-01 4.01415676e-01 -4.40740913e-01 -2.26009324e-01
5.01683176e-01 -1.17222905e+00 7.70490477e-03 9.30091619e-01
-9.59028602e-02 1.30583489e+00 6.50934875e-01 -8.37906301e-01
-1.49728847e+00 -9.31181908e-01 1.60336304e+00 -5.58954298e-01
4.14542973e-01 -2.02363268e-01 -1.21477222e+00 1.14603043e+00
-2.06359208e-01 1.05991729e-01 6.74090087e-01 3.92273664e-01
-7.93310642e-01 -3.14198315e-01 -1.35918272e+00 4.96440321e-01
1.11959195e+00 -6.36320651e-01 -8.29084098e-01 1.06392808e-01
7.00751841e-01 -5.02879679e-01 -1.14714837e+00 7.14031458e-01
5.67913771e-01 -9.43631768e-01 8.50321174e-01 -1.24961615e+00
-2.03070670e-01 -9.17334437e-01 -5.95302820e-01 -6.23703063e-01
-2.94410676e-01 -3.31895798e-01 -6.06722534e-01 1.09813666e+00
8.33073318e-01 -9.53775167e-01 2.86993682e-01 1.46138656e+00
2.49172896e-02 -5.27843356e-01 -1.06263363e+00 -6.68186486e-01
-3.63839567e-01 -4.15269881e-01 1.31415534e+00 1.02780926e+00
4.76868719e-01 2.68646121e-01 -2.37453029e-01 6.74802780e-01
3.70372802e-01 6.74020112e-01 7.21609116e-01 -1.59782958e+00
-1.96947202e-01 1.67536423e-01 -2.32779205e-01 -5.24178505e-01
7.78319165e-02 -7.34127939e-01 -3.26185107e-01 -1.70819461e+00
1.16155066e-01 -8.95233572e-01 -1.19281396e-01 9.30248320e-01
-6.92834333e-02 -1.24672733e-01 7.60834217e-02 2.58053541e-01
-8.36906016e-01 -2.17572749e-01 8.56909573e-01 1.82704240e-01
-3.43210667e-01 -6.76976591e-02 -9.90078211e-01 3.14813226e-01
4.73601252e-01 -5.09587407e-01 -4.06407326e-01 -1.92372769e-01
1.02159846e+00 4.89557117e-01 3.47014755e-01 -4.51467544e-01
5.69423914e-01 -6.18521333e-01 -3.73547882e-01 -2.91613907e-01
2.08126113e-01 -8.22140574e-01 1.05962491e+00 2.32168391e-01
-3.44796866e-01 -1.83441460e-01 -8.50975886e-02 3.16924840e-01
-2.94831514e-01 -3.26932043e-01 2.45624319e-01 -1.19991422e-01
-9.58125472e-01 3.83584425e-02 1.10057391e-01 4.44188975e-02
1.06485331e+00 -5.45536242e-02 -5.11120915e-01 -1.03459179e-01
-1.03333485e+00 5.81252813e-01 6.26656830e-01 2.15673909e-01
3.41475874e-01 -1.30130696e+00 -5.27949154e-01 1.73107371e-01
4.30235147e-01 4.51904386e-01 5.30484207e-02 6.86696768e-01
-1.67529985e-01 6.33436143e-01 1.82139091e-02 -9.92472619e-02
-1.14669323e+00 8.36641490e-01 2.88533419e-01 -4.68523264e-01
-1.70260176e-01 4.42740619e-01 -4.48057987e-03 -5.27015686e-01
2.87872367e-02 -1.90253109e-01 -2.82676458e-01 6.00922741e-02
6.19927526e-01 2.33764321e-01 3.29570562e-01 -5.48697948e-01
-8.50563765e-01 1.11232847e-02 -7.95839503e-02 -1.80029050e-01
1.14948821e+00 -2.59939432e-01 -7.32682586e-01 2.16665775e-01
5.41652501e-01 4.50825095e-01 -2.18534648e-01 -5.18121302e-01
4.25001204e-01 -3.88672173e-01 -6.10714734e-01 -8.83688390e-01
-7.37202585e-01 -9.72935036e-02 -2.97625780e-01 7.82653749e-01
1.04555297e+00 4.13176715e-01 4.51453596e-01 7.33596027e-01
7.45205164e-01 -9.93535280e-01 -7.85402656e-01 4.61071491e-01
7.40327179e-01 -5.82054019e-01 -1.73723266e-01 -8.88409555e-01
-4.90719110e-01 8.60049427e-01 5.91982603e-01 5.04158735e-01
1.57928944e-01 4.37615305e-01 -1.69738710e-01 -4.16604400e-01
-1.05200529e+00 -5.82988560e-01 -8.23535994e-02 6.51892722e-01
-1.81476891e-01 2.19119847e-01 -6.87220097e-01 7.95757532e-01
-1.72307007e-02 3.31793189e-01 4.96433467e-01 1.31505430e+00
-4.20280010e-01 -1.52989352e+00 -4.31568444e-01 2.09321693e-01
-3.07233125e-01 -6.74112886e-02 -4.76066411e-01 8.89268637e-01
3.29258442e-01 1.00684631e+00 -2.05753502e-02 -1.33407265e-01
6.21363342e-01 4.20389473e-01 4.42220479e-01 -7.30678916e-01
-2.09475711e-01 -3.89670879e-01 7.46406913e-01 -4.19141322e-01
-5.02384245e-01 -4.35862869e-01 -1.72911680e+00 -2.42619544e-01
-6.68715179e-01 7.07325161e-01 1.12208925e-01 9.73946035e-01
5.72537422e-01 -2.70773377e-02 4.05989647e-01 5.89824021e-01
-2.98668802e-01 -2.03739554e-01 -7.38546729e-01 4.49280232e-01
1.77462608e-01 -6.19382679e-01 -4.88827303e-02 3.05538122e-02] | [8.98937702178955, 7.380207061767578] |
41f0bccd-ad8f-498d-9d58-798341f651bd | deep-stochastic-attraction-and-repulsion | 1808.08779 | null | https://arxiv.org/abs/1808.08779v2 | https://arxiv.org/pdf/1808.08779v2.pdf | Stochastic Attraction-Repulsion Embedding for Large Scale Image Localization | This paper tackles the problem of large-scale image-based localization (IBL) where the spatial location of a query image is determined by finding out the most similar reference images in a large database. For solving this problem, a critical task is to learn discriminative image representation that captures informative information relevant for localization. We propose a novel representation learning method having higher location-discriminating power. It provides the following contributions: 1) we represent a place (location) as a set of exemplar images depicting the same landmarks and aim to maximize similarities among intra-place images while minimizing similarities among inter-place images; 2) we model a similarity measure as a probability distribution on L_2-metric distances between intra-place and inter-place image representations; 3) we propose a new Stochastic Attraction and Repulsion Embedding (SARE) loss function minimizing the KL divergence between the learned and the actual probability distributions; 4) we give theoretical comparisons between SARE, triplet ranking and contrastive losses. It provides insights into why SARE is better by analyzing gradients. Our SARE loss is easy to implement and pluggable to any CNN. Experiments show that our proposed method improves the localization performance on standard benchmarks by a large margin. Demonstrating the broad applicability of our method, we obtained the third place out of 209 teams in the 2018 Google Landmark Retrieval Challenge. Our code and model are available at https://github.com/Liumouliu/deepIBL. | ['Yuchao Dai', 'Liu Liu', 'Hongdong Li'] | 2018-08-27 | stochastic-attraction-repulsion-embedding-for | http://openaccess.thecvf.com/content_ICCV_2019/html/Liu_Stochastic_Attraction-Repulsion_Embedding_for_Large_Scale_Image_Localization_ICCV_2019_paper.html | http://openaccess.thecvf.com/content_ICCV_2019/papers/Liu_Stochastic_Attraction-Repulsion_Embedding_for_Large_Scale_Image_Localization_ICCV_2019_paper.pdf | iccv-2019-10 | ['image-based-localization'] | ['computer-vision'] | [-2.59018391e-01 -3.66552502e-01 -4.39590394e-01 -4.45301473e-01
-1.39511514e+00 -7.04078972e-01 6.36417687e-01 3.35639000e-01
-7.15626597e-01 4.57253754e-01 2.88358390e-01 9.04743522e-02
-4.81702715e-01 -4.90852028e-01 -1.05348337e+00 -6.68413222e-01
-2.80377120e-01 1.91729739e-01 3.27790305e-02 8.91276523e-02
5.09489417e-01 6.80788517e-01 -1.39177942e+00 -1.23622119e-01
6.21927202e-01 1.38404167e+00 4.21374410e-01 2.01440915e-01
2.65342057e-01 5.32640159e-01 -3.46783817e-01 -9.26269069e-02
4.83361393e-01 -1.81448400e-01 -8.51426840e-01 -3.91644388e-01
8.09199929e-01 9.47232172e-02 -3.22094083e-01 1.17969978e+00
6.62534654e-01 4.36494142e-01 6.90756619e-01 -1.31388927e+00
-1.12678361e+00 2.01921090e-01 -7.60598004e-01 4.82565016e-01
2.81757563e-01 -2.13580981e-01 1.20777118e+00 -1.21369636e+00
6.10548019e-01 9.92677212e-01 6.89777553e-01 2.65798181e-01
-1.23202991e+00 -5.62584639e-01 1.46207809e-01 2.43591234e-01
-2.05881023e+00 -2.40514755e-01 6.61641896e-01 -2.68560261e-01
6.13553166e-01 3.15151423e-01 2.68799067e-01 8.86901855e-01
1.82281524e-01 8.11276495e-01 8.34098637e-01 -3.03091913e-01
1.55379623e-01 8.43798146e-02 -1.17297601e-02 7.96910942e-01
1.18927181e-01 -1.32343099e-01 -6.82040513e-01 -2.84349769e-01
6.14737391e-01 3.26863796e-01 -3.93014878e-01 -6.93102598e-01
-1.34469640e+00 8.70029807e-01 1.25847244e+00 5.16812027e-01
-3.11974525e-01 5.22105932e-01 5.01705483e-02 -3.53870261e-03
3.68641943e-01 5.06146491e-01 -1.26331016e-01 7.41988122e-02
-8.59519541e-01 1.99616060e-01 3.74500960e-01 8.72764945e-01
1.15250385e+00 -6.28979802e-01 -2.73624420e-01 9.88427818e-01
2.92872816e-01 5.98203242e-01 6.80774212e-01 -9.51410532e-01
4.27521020e-01 4.03429568e-01 1.65869802e-01 -1.42026865e+00
-1.86736047e-01 -4.10968661e-01 -5.48557520e-01 -2.32651770e-01
2.79534727e-01 4.00628120e-01 -5.22776008e-01 1.93546772e+00
-6.54529547e-03 3.13860834e-01 -3.01136076e-01 1.08846021e+00
6.77670419e-01 6.78903818e-01 7.64732212e-02 4.54372436e-01
1.14240074e+00 -1.05092955e+00 -1.57331720e-01 -2.59072840e-01
5.19870937e-01 -6.89028442e-01 1.10785460e+00 -2.02971235e-01
-8.38371158e-01 -3.97291690e-01 -8.91701162e-01 -3.15782577e-01
-6.45471692e-01 3.81066352e-01 6.06702030e-01 2.19451442e-01
-1.40681863e+00 4.44180250e-01 -6.43595517e-01 -4.89096701e-01
3.89522910e-01 4.36449170e-01 -7.16539562e-01 -1.67837560e-01
-9.44494426e-01 7.67642438e-01 1.57452688e-01 -8.76526348e-03
-8.66893053e-01 -7.61736929e-01 -1.06591344e+00 1.07141167e-01
2.29587629e-02 -4.91207421e-01 7.85406113e-01 -8.09060872e-01
-8.49655330e-01 1.07232523e+00 -3.89362425e-01 -5.25052845e-01
2.25825995e-01 -2.61249006e-01 -1.14691690e-01 1.55496657e-01
6.93649411e-01 8.91213536e-01 4.60729182e-01 -1.24898791e+00
-6.79132104e-01 -4.20064718e-01 -1.51910177e-02 2.00119287e-01
-2.34741002e-01 -2.42750183e-01 -7.96011746e-01 -5.90788066e-01
2.49738231e-01 -9.44808185e-01 -1.96410745e-01 2.59176821e-01
-4.41562206e-01 -4.27972883e-01 4.46356744e-01 -3.94876689e-01
1.08905911e+00 -2.33269000e+00 4.39682156e-02 4.65871543e-01
1.36585996e-01 -1.04464330e-01 -4.71684396e-01 5.48901141e-01
1.74146220e-02 7.58622736e-02 -8.70754570e-02 -4.10525918e-01
1.86075762e-01 -1.22300446e-01 -5.53456366e-01 8.50988805e-01
2.60540321e-02 1.20559728e+00 -1.15194631e+00 -3.51066709e-01
3.20472956e-01 7.06334651e-01 -2.84466803e-01 -7.90376812e-02
3.18218470e-01 3.81489068e-01 -5.26612580e-01 8.10637414e-01
6.04135394e-01 -4.69694316e-01 -4.05826330e-01 -2.39221066e-01
-2.44403705e-02 1.90553412e-01 -9.41081882e-01 2.14620829e+00
-7.04310000e-01 6.93741500e-01 -3.57519001e-01 -9.39973533e-01
9.60667551e-01 -3.12522858e-01 4.27659780e-01 -9.36469138e-01
-1.84893280e-01 3.76857549e-01 -5.59186161e-01 -8.58977214e-02
4.67356294e-01 3.55200320e-01 -2.75397629e-01 2.99795240e-01
1.05806075e-01 1.45968914e-01 -2.67319772e-02 2.59513706e-01
1.09248066e+00 -7.70437941e-02 2.03746781e-01 -3.55890840e-01
4.49615628e-01 -3.39914083e-01 3.56184751e-01 9.03068721e-01
-3.31040442e-01 8.34605813e-01 2.55497158e-01 -4.95887846e-01
-6.32012665e-01 -1.30693054e+00 -3.36006701e-01 1.05971730e+00
6.62313104e-01 -4.80987459e-01 -3.18852842e-01 -7.08999276e-01
3.20887446e-01 2.27562383e-01 -9.49648440e-01 -1.73688710e-01
-3.89401704e-01 -4.15156484e-01 5.38417459e-01 5.29934466e-01
5.49243510e-01 -7.72143900e-01 -4.19013917e-01 -2.77441174e-01
-2.58300781e-01 -7.03121841e-01 -9.97530818e-01 6.57642484e-02
-4.91149992e-01 -9.04090524e-01 -9.76081669e-01 -1.00095940e+00
9.49396372e-01 6.20940983e-01 1.04851460e+00 -2.31610946e-02
-5.29936552e-01 8.17989588e-01 -2.64884919e-01 -7.12175071e-02
6.10223770e-01 2.04158932e-01 1.31906554e-01 2.38398224e-01
4.27471071e-01 -4.02934790e-01 -1.16355705e+00 4.45310563e-01
-7.02663302e-01 -5.21186113e-01 5.75789034e-01 8.21232438e-01
1.25800574e+00 -2.58716226e-01 2.79484719e-01 -3.30207646e-01
5.63078582e-01 -7.45969534e-01 -6.84978724e-01 4.71833616e-01
-5.33838212e-01 1.15088075e-01 3.79764110e-01 -1.54799581e-01
-2.90174186e-01 2.36873105e-01 1.71628594e-01 -5.50806403e-01
1.00884542e-01 3.86231840e-01 8.29941258e-02 -5.10760069e-01
4.81052101e-01 4.50357914e-01 -3.07748318e-01 -4.02019233e-01
5.72718322e-01 4.55624163e-01 5.18732071e-01 -7.13106394e-01
7.01862216e-01 6.26381695e-01 -4.79609193e-03 -6.43540204e-01
-9.08865213e-01 -8.72195423e-01 -6.35667443e-01 1.14265248e-01
4.78417128e-01 -1.08753967e+00 -8.48528385e-01 -1.32016823e-01
-8.97865713e-01 -1.78917661e-01 -3.16510916e-01 5.74053586e-01
-4.56390470e-01 3.35224539e-01 -1.92786410e-01 -5.16064703e-01
-2.83635020e-01 -1.16457713e+00 1.53912139e+00 3.59618872e-01
-1.28473639e-01 -1.08790636e+00 3.41496438e-01 1.26373619e-01
4.67649341e-01 2.42562696e-01 4.97864485e-01 -5.29807210e-01
-8.68522525e-01 -4.54451352e-01 -5.16678631e-01 1.53543204e-01
7.82865360e-02 -3.89079690e-01 -8.75798941e-01 -5.25558889e-01
-3.54421943e-01 -4.17347312e-01 1.19684958e+00 5.61627507e-01
1.52250171e+00 -2.24454075e-01 -5.36645472e-01 9.79687333e-01
1.61641550e+00 -1.51226491e-01 4.89881128e-01 6.53246343e-01
5.86676955e-01 3.96926403e-01 7.26560473e-01 2.70795107e-01
5.08954525e-01 9.26957667e-01 4.44012910e-01 6.64852336e-02
-4.23332639e-02 -6.54398382e-01 1.85199663e-01 4.27304208e-01
2.09242776e-01 -1.65923312e-01 -8.47002268e-01 1.08146763e+00
-2.10516977e+00 -9.30053115e-01 4.21011895e-01 2.47863889e+00
4.42247361e-01 -4.30472791e-01 9.26695392e-03 -4.48261172e-01
6.69782341e-01 3.33979219e-01 -4.85412151e-01 1.78949445e-01
-2.07865238e-01 8.99535939e-02 7.85621583e-01 6.23991668e-01
-1.37580860e+00 8.98147643e-01 5.30254316e+00 1.14602482e+00
-1.27746415e+00 1.87694594e-01 7.20297933e-01 -2.72970647e-01
-2.87128329e-01 -1.30607545e-01 -8.25348437e-01 6.71834707e-01
5.94851196e-01 -2.60565192e-01 4.36028361e-01 1.02933729e+00
-1.34080440e-01 -1.72077194e-01 -1.12122583e+00 1.35427272e+00
2.62818843e-01 -1.46795058e+00 1.06606960e-01 7.22370669e-02
7.48714745e-01 5.34428239e-01 6.50045931e-01 4.91487645e-02
-8.19531605e-02 -1.20415747e+00 7.84338415e-01 4.64969397e-01
8.06654513e-01 -7.64830232e-01 6.15239263e-01 1.13771588e-01
-1.39009428e+00 -7.19420835e-02 -6.87021613e-01 2.91908652e-01
-2.28747055e-01 4.14523780e-01 -6.06903195e-01 4.62842345e-01
8.83408964e-01 1.01299512e+00 -8.74310672e-01 1.44891334e+00
-3.03586274e-01 1.88593298e-01 -4.56929505e-01 -6.08405359e-02
5.43398857e-01 -1.56034201e-01 3.69265854e-01 1.28153598e+00
4.38860238e-01 -3.85048777e-01 1.16589688e-01 9.64551449e-01
-3.70710939e-01 2.74330467e-01 -8.75727534e-01 2.81226337e-01
7.60516346e-01 1.30444765e+00 -7.66599834e-01 2.26329103e-01
-1.09584667e-01 1.14660621e+00 7.15130985e-01 4.40201968e-01
-7.52451122e-01 -6.43631935e-01 6.92165971e-01 4.85039875e-02
3.48230213e-01 -6.34356439e-02 -4.50944155e-02 -1.06531930e+00
2.79280782e-01 -2.35019803e-01 3.99597079e-01 -7.04657078e-01
-1.48462331e+00 6.14462435e-01 -1.53650239e-01 -1.33973348e+00
5.73964342e-02 -5.30760169e-01 -5.53570747e-01 1.00813162e+00
-1.73523664e+00 -1.18899834e+00 -3.20927024e-01 7.29122162e-01
2.45715111e-01 -7.11273924e-02 7.69909382e-01 6.37436807e-01
-2.97705233e-01 8.97160470e-01 4.01292682e-01 2.71983027e-01
8.26960206e-01 -1.31262660e+00 2.41031244e-01 6.69768155e-01
7.13192523e-01 1.07921910e+00 2.17418909e-01 -2.49726012e-01
-1.38791418e+00 -1.26461983e+00 1.07932305e+00 -8.00826132e-01
6.69541121e-01 -3.90428692e-01 -5.28936744e-01 7.79232383e-01
-1.03607699e-01 6.14401281e-01 6.54577255e-01 4.03734595e-02
-6.38819516e-01 -3.96890998e-01 -1.10287917e+00 4.02146757e-01
8.66486132e-01 -9.11021352e-01 -2.01282158e-01 5.22203505e-01
3.34654778e-01 -3.06622148e-01 -6.57986104e-01 1.53574809e-01
5.01804769e-01 -8.33124280e-01 1.35935688e+00 -1.93965241e-01
3.15550625e-01 -5.88827968e-01 -4.94632334e-01 -1.19638455e+00
-3.54548246e-01 -3.27838778e-01 2.56276786e-01 1.03003335e+00
3.61958444e-01 -7.80322134e-01 5.07112622e-01 3.63297135e-01
1.56564459e-01 -1.16036963e+00 -1.22099447e+00 -9.96019661e-01
-1.90499723e-02 -2.31591627e-01 4.97036159e-01 7.84275591e-01
-2.24064931e-01 -1.11682275e-02 -2.80231088e-01 3.87122571e-01
6.31106555e-01 3.58864218e-01 6.13394976e-01 -8.35687578e-01
4.24740389e-02 -5.48408687e-01 -8.19309175e-01 -1.29351509e+00
3.57315004e-01 -1.38270509e+00 5.91562465e-02 -1.57821536e+00
4.35534209e-01 -6.75351977e-01 -9.69471693e-01 5.63235462e-01
5.80548197e-02 6.43974304e-01 2.30292082e-01 5.69077790e-01
-1.24297523e+00 6.72029197e-01 7.41019785e-01 -2.85537422e-01
1.13733143e-01 -1.81080624e-02 -7.81531870e-01 4.48887765e-01
4.88114774e-01 -6.00829065e-01 -3.58405799e-01 -5.45809031e-01
2.52231777e-01 -4.75368142e-01 8.16277325e-01 -1.00490713e+00
5.01051009e-01 2.14487970e-01 5.09188235e-01 -5.04116714e-01
3.08621973e-01 -7.91760921e-01 -3.38698328e-02 2.06679806e-01
-5.78325212e-01 1.66312501e-01 2.61147805e-02 7.69014835e-01
-5.02059579e-01 -1.18501119e-01 6.28193855e-01 1.29621029e-01
-1.07229316e+00 5.95081568e-01 3.85333657e-01 1.15213528e-01
1.03808773e+00 1.36780050e-02 -3.17815572e-01 -4.42393571e-01
-3.43804657e-01 2.63877392e-01 6.32710457e-01 5.62910557e-01
7.21073389e-01 -1.69124782e+00 -3.90059412e-01 3.49237546e-02
5.02150714e-01 -2.51438946e-01 1.21678613e-01 9.32844698e-01
-6.26738250e-01 6.25169098e-01 -4.86169718e-02 -6.61340296e-01
-9.31649804e-01 4.24257904e-01 4.15691763e-01 -5.66004366e-02
-4.42537487e-01 1.36892939e+00 4.96282876e-01 -5.05850673e-01
4.27640110e-01 -7.07375035e-02 1.41867965e-01 -1.15910403e-01
6.69949234e-01 2.80995548e-01 5.11827180e-03 -9.99663472e-01
-8.36512268e-01 9.97807622e-01 -5.79675399e-02 1.83649566e-02
1.33806419e+00 -2.01455444e-01 -2.46266901e-01 4.17893469e-01
1.84193361e+00 1.20829642e-01 -1.17040896e+00 -3.71838063e-01
1.19535491e-01 -8.10694456e-01 3.81073020e-02 -6.63428485e-01
-1.08606279e+00 7.57946789e-01 9.09883559e-01 -1.21507652e-01
9.50638473e-01 4.49367195e-01 7.68836439e-01 4.20288533e-01
6.25128329e-01 -8.18578780e-01 1.29054189e-01 3.41275215e-01
1.01245832e+00 -1.48525703e+00 -9.59520787e-02 1.22140698e-01
-5.12499750e-01 7.58470535e-01 3.30742270e-01 -4.62823868e-01
7.96707392e-01 -3.68613869e-01 7.64060579e-03 -4.11861062e-01
-2.58957714e-01 -2.87548095e-01 7.23197460e-01 4.60195988e-01
3.53784621e-01 -1.51787437e-02 -1.24809749e-01 4.69806880e-01
-6.19612671e-02 -2.65840828e-01 -1.75994068e-01 6.32648587e-01
-1.95468321e-01 -8.61966670e-01 -2.20580161e-01 2.41602480e-01
-3.87630343e-01 -1.71641335e-01 -3.93619597e-01 6.23621702e-01
1.43092543e-01 6.07041061e-01 3.92063558e-02 -2.42394313e-01
2.42407203e-01 -3.43668252e-01 3.81370515e-01 -4.77385759e-01
-2.03295201e-01 -2.28622794e-01 -7.62845457e-01 -9.06481922e-01
-2.85719752e-01 -3.81338686e-01 -1.04698622e+00 3.01181758e-03
-1.15095004e-01 3.50486785e-01 7.83952236e-01 4.77619290e-01
7.90951192e-01 -2.38803122e-02 7.51680434e-01 -1.11357129e+00
-3.53727371e-01 -4.96472925e-01 -7.42778659e-01 5.33139765e-01
5.47536969e-01 -7.91236997e-01 -4.72704381e-01 -3.07864070e-01] | [7.828031539916992, -1.8952535390853882] |
b2de16d6-63a1-46f0-9565-5044d5b0a3bb | a-hybrid-pso-ga-for-extractive-text | null | null | https://aclanthology.org/2021.paclic-1.30 | https://aclanthology.org/2021.paclic-1.30.pdf | A Hybrid PSO-GA for Extractive Text Summarization | null | ['Nguyen Thi Hoai', 'Tran Thi Dinh', 'Thi Thu Trang Nguyen Bui Thi-Mai-Anh'] | null | null | https://aclanthology.org/2021.paclic-1.79 | https://aclanthology.org/2021.paclic-1.79.pdf | paclic-2021-11 | ['extractive-document-summarization'] | ['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] | [-7.268165111541748, 3.729079008102417] |
38ae0668-042e-4f1a-8e9d-984d789fc559 | towards-playing-full-moba-games-with-deep-1 | 2011.12692 | null | https://arxiv.org/abs/2011.12692v4 | https://arxiv.org/pdf/2011.12692v4.pdf | Towards Playing Full MOBA Games with Deep Reinforcement Learning | MOBA games, e.g., Honor of Kings, League of Legends, and Dota 2, pose grand challenges to AI systems such as multi-agent, enormous state-action space, complex action control, etc. Developing AI for playing MOBA games has raised much attention accordingly. However, existing work falls short in handling the raw game complexity caused by the explosion of agent combinations, i.e., lineups, when expanding the hero pool in case that OpenAI's Dota AI limits the play to a pool of only 17 heroes. As a result, full MOBA games without restrictions are far from being mastered by any existing AI system. In this paper, we propose a MOBA AI learning paradigm that methodologically enables playing full MOBA games with deep reinforcement learning. Specifically, we develop a combination of novel and existing learning techniques, including curriculum self-play learning, policy distillation, off-policy adaption, multi-head value estimation, and Monte-Carlo tree-search, in training and playing a large pool of heroes, meanwhile addressing the scalability issue skillfully. Tested on Honor of Kings, a popular MOBA game, we show how to build superhuman AI agents that can defeat top esports players. The superiority of our AI is demonstrated by the first large-scale performance test of MOBA AI agent in the literature. | ['Wei Liu', 'Lanxiao Huang', 'Wei Yang', 'Qiang Fu', 'Tengfei Shi', 'Liang Wang', 'Bei Shi', 'Yinyuting Yin', 'Hongsheng Yu', 'Fuhao Qiu', 'Zhao Liu', 'Jia Chen', 'Bo Liu', 'Bo Yuan', 'Sheng Chen', 'Wen Zhang', 'Guibin Chen', 'Deheng Ye'] | 2020-11-25 | towards-playing-full-moba-games-with-deep | http://proceedings.neurips.cc/paper/2020/hash/06d5ae105ea1bea4d800bc96491876e9-Abstract.html | http://proceedings.neurips.cc/paper/2020/file/06d5ae105ea1bea4d800bc96491876e9-Paper.pdf | neurips-2020-12 | ['dota-2'] | ['playing-games'] | [-3.40248108e-01 -9.33082923e-02 -3.27823535e-02 3.18187088e-01
-7.17869699e-01 -7.17300832e-01 5.68821669e-01 -3.43275696e-01
-8.34951758e-01 1.20780838e+00 -2.35272869e-01 -4.58199233e-01
-4.62998182e-01 -8.64120781e-01 -7.52923310e-01 -6.83054328e-01
-3.72718811e-01 1.05735016e+00 5.34591675e-01 -1.05206120e+00
2.86397368e-01 2.24533286e-02 -1.64142764e+00 9.25153494e-02
1.05877531e+00 7.86529064e-01 1.31840199e-01 8.47690821e-01
3.16187918e-01 1.22177982e+00 -9.43864942e-01 -3.87376368e-01
4.86582726e-01 -5.41012764e-01 -8.63449395e-01 -5.03350437e-01
4.98476513e-02 -7.23857462e-01 -3.21561098e-01 7.05367804e-01
6.63414538e-01 2.22592041e-01 3.58764529e-01 -1.63006425e+00
-1.68673508e-02 9.90193427e-01 -4.97275889e-01 2.38895208e-01
1.54776916e-01 8.07703197e-01 1.00852406e+00 6.90279305e-02
5.03465414e-01 1.22692811e+00 6.24121606e-01 8.65098238e-01
-6.52525306e-01 -9.03134108e-01 -6.82062656e-02 4.15814161e-01
-1.02614498e+00 -1.37948513e-01 4.06954408e-01 -2.52571046e-01
1.03618419e+00 2.17565626e-01 1.28834903e+00 1.29705656e+00
3.84861827e-01 1.03506851e+00 1.19010520e+00 -1.99955061e-01
5.59057832e-01 -5.03821135e-01 -3.80563706e-01 8.09976399e-01
1.51444167e-01 6.48487389e-01 -4.30142075e-01 -2.58112222e-01
7.24625468e-01 -4.92336243e-01 5.39047062e-01 -3.42764258e-01
-1.12011707e+00 1.06867123e+00 7.57233649e-02 -4.82520030e-04
-5.61714172e-01 6.68335497e-01 4.91334081e-01 4.86943752e-01
-1.58055231e-01 1.19104373e+00 -4.07657236e-01 -6.51744545e-01
-7.17639804e-01 1.02796137e+00 8.97834599e-01 6.38282239e-01
4.85116065e-01 5.21599829e-01 1.05172768e-01 4.51267630e-01
1.34136856e-01 6.27970815e-01 5.67677796e-01 -1.56802690e+00
5.84158599e-02 4.03016746e-01 1.73568666e-01 -6.55449390e-01
-7.41961598e-01 -2.01665312e-01 -5.45280397e-01 9.01044488e-01
7.78742552e-01 -7.68359900e-01 -6.02377057e-01 1.89600551e+00
4.82654184e-01 2.69695669e-01 3.66923839e-01 8.26029420e-01
4.27326322e-01 7.16221392e-01 1.75030362e-02 -6.34452850e-02
1.52912426e+00 -1.17587066e+00 -3.18465680e-01 -3.87128979e-01
4.45898950e-01 -3.76916528e-01 8.58367622e-01 7.65676022e-01
-1.35291255e+00 -2.90093809e-01 -1.09987354e+00 4.79585767e-01
-3.44596595e-01 -6.34179294e-01 1.19789243e+00 5.32972515e-01
-7.91072667e-01 5.61838329e-01 -9.94539142e-01 -1.70207009e-01
4.83542919e-01 7.14209676e-01 4.90816981e-02 2.81893969e-01
-1.54148507e+00 1.07751071e+00 7.96328366e-01 -3.25833112e-01
-1.73436749e+00 -6.17136121e-01 -4.90787953e-01 1.19735688e-01
1.20655453e+00 -8.49655628e-01 1.60049450e+00 -8.66996169e-01
-2.07471132e+00 5.17700970e-01 9.77120459e-01 -8.24532032e-01
6.17696881e-01 -2.27665886e-01 -1.58868983e-01 -4.43458222e-02
1.03977494e-01 6.95708811e-01 7.19982564e-01 -9.82081890e-01
-1.09587002e+00 -1.66163355e-01 8.20934474e-01 5.30270636e-01
-2.97327548e-01 6.61179945e-02 -6.69726869e-03 -2.99310297e-01
-7.65465498e-01 -1.11690712e+00 -6.04051411e-01 -6.53947055e-01
1.96824986e-02 -5.17947197e-01 4.26062614e-01 -1.82790533e-01
1.23328567e+00 -1.89286411e+00 4.23805863e-01 1.03530541e-01
4.55003440e-01 4.47845578e-01 -2.76892573e-01 5.24562061e-01
4.45587873e-01 -1.39317900e-01 6.48027137e-02 4.81863588e-01
3.03118259e-01 4.82111692e-01 8.56880397e-02 7.98131749e-02
-2.61773378e-01 9.55160499e-01 -1.28482604e+00 -6.84816360e-01
-1.26882747e-01 -1.88345268e-01 -1.01565969e+00 1.90261737e-01
-5.38030982e-01 3.07351500e-01 -6.42828584e-01 5.10818720e-01
1.35132104e-01 -8.44023004e-02 2.01944813e-01 4.20639724e-01
-3.74481112e-01 7.07224384e-02 -1.18574560e+00 1.77028263e+00
-2.80884057e-01 1.66244939e-01 2.24457189e-01 -8.85537088e-01
5.98750055e-01 1.72477886e-01 7.42083013e-01 -8.70700300e-01
4.25764292e-01 3.75764310e-01 7.39800513e-01 -4.13842708e-01
7.42281139e-01 -1.81243956e-01 -6.07998908e-01 4.49798912e-01
-3.63116786e-02 -6.12138629e-01 6.86827779e-01 1.18341856e-01
1.35367620e+00 3.24594766e-01 3.66344988e-01 -3.06770682e-01
2.17754200e-01 6.53224945e-01 6.60526216e-01 1.18389904e+00
-5.17318606e-01 -8.06694999e-02 7.51485407e-01 -6.62883520e-01
-1.22868192e+00 -8.42705369e-01 5.36109447e-01 1.59329891e+00
2.56422520e-01 -5.12784958e-01 -8.52431417e-01 -5.30862987e-01
-5.54164685e-02 6.02085888e-01 -5.34515083e-01 -4.43808883e-01
-9.83559787e-01 -8.54975879e-01 9.20617819e-01 2.83417612e-01
6.85992599e-01 -1.59086061e+00 -1.46836090e+00 4.75160182e-01
7.12783784e-02 -7.04516888e-01 -3.00995767e-01 1.83047861e-01
-3.19005430e-01 -1.08588302e+00 -5.15774190e-01 -7.25689352e-01
-4.36929554e-01 -1.05629399e-01 1.04186368e+00 1.36799542e-02
-2.17536688e-01 2.88038820e-01 -2.62232631e-01 -4.89176929e-01
-6.06932819e-01 2.79169589e-01 2.48686254e-01 -7.25932777e-01
-1.06562518e-01 -7.55584002e-01 -6.22436523e-01 3.38165492e-01
-5.67802787e-01 2.25835577e-01 7.75676847e-01 1.14158297e+00
-3.94046977e-02 1.48930579e-01 7.75866330e-01 -7.59093165e-01
8.41869831e-01 -4.59977895e-01 -8.79058063e-01 1.68697670e-01
-3.59371990e-01 1.82590121e-03 8.51698637e-01 -7.65266716e-01
-7.70694435e-01 -3.11597914e-01 -2.90481001e-02 -1.46052152e-01
2.49121338e-01 3.87881130e-01 1.51741743e-01 -5.69853485e-02
9.41646636e-01 1.96417347e-01 3.00025344e-01 2.08012611e-01
3.28987241e-01 3.82017672e-01 5.36558747e-01 -9.77400362e-01
6.95158064e-01 1.82664022e-03 -4.00444642e-02 -4.35684234e-01
-4.04599309e-01 -1.17285058e-01 -3.26348469e-02 -3.63977402e-01
7.99064934e-01 -8.77082050e-01 -1.71821439e+00 7.60401487e-01
-8.16500366e-01 -8.47687006e-01 -4.61814672e-01 4.66695935e-01
-1.08531010e+00 1.83667261e-02 -7.49520242e-01 -6.46794498e-01
-2.64062822e-01 -1.29718709e+00 4.43614930e-01 5.93087375e-01
-2.74319708e-01 -5.73854506e-01 5.13538182e-01 5.63908815e-01
6.01085722e-01 2.04593450e-01 7.32088625e-01 -6.57393396e-01
-4.51977164e-01 2.43744433e-01 2.52719522e-01 2.23232992e-03
-4.22134399e-01 -2.89265424e-01 -3.27659070e-01 -4.71033365e-01
-4.51103151e-01 -9.98322606e-01 3.02004039e-01 2.84065515e-01
7.71205246e-01 -3.35879296e-01 -1.30502284e-02 4.44325835e-01
1.01502943e+00 8.17918539e-01 4.99185622e-01 9.38828886e-01
4.75625992e-01 1.46138430e-01 6.72190487e-01 9.31175709e-01
4.61310178e-01 6.23651803e-01 8.12487841e-01 2.69659966e-01
8.10057446e-02 -1.69511348e-01 4.56321448e-01 7.35612631e-01
-6.51155412e-01 -2.15268090e-01 -1.07193410e+00 3.66669148e-01
-2.00676990e+00 -1.15290511e+00 4.18390810e-01 1.74813735e+00
1.01074111e+00 3.49106729e-01 7.72018731e-01 -2.45923176e-01
2.80433774e-01 8.40588585e-02 -8.72969508e-01 -4.08059865e-01
1.93840393e-03 2.82495946e-01 5.65735757e-01 4.56757218e-01
-1.05089653e+00 1.48929918e+00 6.41489887e+00 1.32390702e+00
-6.18447840e-01 6.52711373e-03 2.35266939e-01 -1.86600000e-01
1.16915803e-03 6.12786971e-02 -5.95363140e-01 4.08767760e-01
9.90540624e-01 -4.58988428e-01 9.69930768e-01 9.67113793e-01
-9.09887105e-02 -1.74839839e-01 -7.51892328e-01 8.05272043e-01
-1.46342382e-01 -1.33870435e+00 -1.47767216e-01 2.37147838e-01
8.51243436e-01 -8.02228674e-02 1.33362534e-02 1.16798580e+00
1.22081554e+00 -1.12901235e+00 6.84514642e-01 1.26986839e-02
4.01927263e-01 -1.06304264e+00 5.70908427e-01 6.16085768e-01
-8.01768124e-01 -5.61953723e-01 -2.31764808e-01 -4.68187571e-01
-1.05322354e-01 -6.51955962e-01 -6.59342825e-01 4.75031823e-01
6.65328562e-01 1.81983292e-01 -2.38844067e-01 1.06856513e+00
4.14177105e-02 5.61387062e-01 -4.63301539e-01 -5.53402841e-01
8.98403049e-01 -1.66392952e-01 5.65086305e-01 5.65070629e-01
5.82099743e-02 4.72712487e-01 5.21068096e-01 4.59238499e-01
1.24539599e-01 2.85982678e-04 -4.31061864e-01 -9.36929509e-02
3.41665208e-01 1.17643225e+00 -6.26245379e-01 -4.04788703e-01
-1.84315711e-01 5.40320635e-01 4.47172135e-01 1.21193295e-02
-1.10846913e+00 -2.25135401e-01 6.43030345e-01 -9.40300971e-02
1.96044654e-01 -1.63628772e-01 1.39502466e-01 -9.28734839e-01
-5.73350489e-01 -1.75817692e+00 5.47577143e-01 -5.01467705e-01
-1.13863802e+00 5.19232392e-01 2.84617245e-01 -1.01486802e+00
-6.68424129e-01 -5.78028083e-01 -6.68313324e-01 8.89880508e-02
-1.05614328e+00 -1.05902696e+00 -4.30147387e-02 6.64745390e-01
7.03129947e-01 -8.00044358e-01 6.67655587e-01 2.49605671e-01
-5.62556922e-01 6.98455513e-01 2.86868781e-01 -5.80745228e-02
3.41199547e-01 -1.22957909e+00 1.88606903e-01 3.57227951e-01
-2.71388829e-01 8.88083782e-03 8.20897162e-01 -4.10149008e-01
-1.60869229e+00 -2.06612095e-01 -1.47752002e-01 -2.40627393e-01
1.18406916e+00 -2.84564316e-01 -4.69538808e-01 4.95631039e-01
4.52326417e-01 -3.76891345e-01 2.87416816e-01 3.18655670e-01
4.02867384e-02 -1.58925906e-01 -7.29303360e-01 9.61928368e-01
9.12464261e-01 7.73905739e-02 -5.59680104e-01 2.32033953e-01
6.34068847e-01 -6.81096971e-01 -7.74775147e-01 2.73287535e-01
7.74706841e-01 -9.18034077e-01 1.07563114e+00 -1.09308851e+00
6.17189229e-01 -1.23862579e-01 9.88836661e-02 -1.51605392e+00
-2.82504410e-01 -1.04289699e+00 -1.30011931e-01 7.20150292e-01
7.01781586e-02 -3.11414152e-01 1.05671299e+00 2.60315627e-01
-3.38782430e-01 -8.49595904e-01 -1.11528862e+00 -9.66399908e-01
5.63447654e-01 -4.02657129e-02 5.95505595e-01 8.77437651e-01
2.61101723e-01 5.04908085e-01 -9.93832171e-01 -2.04532310e-01
5.35804152e-01 -2.05458086e-02 1.06153929e+00 -9.75205719e-01
-9.69441235e-01 -7.82810748e-01 -1.88930064e-01 -8.14775348e-01
7.82592967e-02 -5.44586420e-01 2.31010199e-01 -1.12670064e+00
1.45141006e-01 -5.78371584e-01 -8.82388428e-02 5.65529108e-01
-9.34201851e-02 -8.57438818e-02 6.28196001e-01 1.42463818e-01
-1.18760931e+00 5.97472370e-01 1.65397465e+00 -2.35436872e-01
-2.96776950e-01 -4.72518764e-02 -5.81349790e-01 9.36814845e-01
1.03746641e+00 -3.45747709e-01 -5.24437368e-01 -2.75265574e-01
5.22691071e-01 5.84869623e-01 1.94198489e-01 -1.31819296e+00
4.29500699e-01 -7.08152592e-01 -1.48440316e-01 -5.84910065e-02
4.19996262e-01 -5.50293565e-01 -1.48479491e-02 1.01316679e+00
-2.01536119e-01 2.61796683e-01 3.68221670e-01 3.15663129e-01
1.11797899e-02 -3.47822607e-01 7.11229444e-01 -5.21634281e-01
-1.06992030e+00 2.37702087e-01 -9.11031425e-01 6.66478276e-01
1.42845571e+00 -9.58799496e-02 -3.57082844e-01 -3.91736358e-01
-4.22441781e-01 8.46183598e-01 1.02051618e-02 1.66661024e-01
2.11953521e-01 -1.06646430e+00 -9.62079883e-01 -1.45310491e-01
-2.02879339e-01 4.48064990e-02 4.84987766e-01 5.40945709e-01
-8.66111159e-01 -1.72426403e-01 -1.02190900e+00 -9.53798592e-02
-1.21072710e+00 5.79719901e-01 4.53612000e-01 -8.79815221e-01
-6.53577805e-01 6.09299839e-01 2.61898935e-01 -4.75082934e-01
1.32155016e-01 2.56557107e-01 -2.63752997e-01 -6.90920949e-02
3.68305564e-01 5.15459001e-01 -5.60933471e-01 1.95216965e-02
-2.65945584e-01 1.74080938e-01 -2.47684717e-01 -3.32359642e-01
1.37319946e+00 3.91016126e-01 2.79112309e-01 1.44154042e-01
2.56798774e-01 -1.46449864e-01 -1.41453063e+00 2.14899257e-01
-1.48447812e-01 -1.07640803e-01 -3.01309109e-01 -1.05562544e+00
-7.98697233e-01 5.11147320e-01 4.29946184e-01 2.24949464e-01
9.19975460e-01 -1.32103309e-01 8.86338890e-01 9.07665133e-01
7.14342117e-01 -1.40624058e+00 5.77970505e-01 9.08930838e-01
5.81216156e-01 -1.20585239e+00 1.31884096e-02 4.61125612e-01
-8.85544598e-01 8.86367798e-01 1.27267647e+00 -4.71075743e-01
1.03061035e-01 3.75118345e-01 -1.26362592e-03 -3.22297841e-01
-1.08576143e+00 -2.65053868e-01 -3.93753946e-01 5.96511841e-01
-2.83410966e-01 1.54644787e-01 -4.07061875e-01 6.48663223e-01
-6.73761249e-01 -1.55665711e-01 6.77579641e-01 9.78173077e-01
-7.54472315e-01 -9.51838553e-01 -4.78232741e-01 1.93149135e-01
-3.43654603e-01 6.03838377e-02 -2.86461487e-02 1.40743959e+00
3.19277883e-01 6.22626364e-01 -5.85913882e-02 -3.65235209e-01
3.16943198e-01 -2.60595232e-01 6.88002467e-01 -1.57908827e-01
-1.02140450e+00 -3.30062471e-02 1.91332117e-01 -5.13783216e-01
-1.08369529e-01 -5.63713431e-01 -1.27103508e+00 -8.57142091e-01
9.86622740e-03 3.23919088e-01 2.58087665e-01 9.24318135e-01
-3.83263901e-02 7.67317295e-01 3.02496642e-01 -5.84631503e-01
-9.72259104e-01 -7.68536091e-01 -5.47581315e-01 3.34918886e-01
-9.47164297e-02 -8.36740792e-01 7.21784681e-02 -5.92035174e-01] | [3.6052610874176025, 1.5689071416854858] |
ae165e0e-b928-4020-b161-3021a57a3b37 | using-the-random-sprays-retinex-algorithm-for | 1310.0307 | null | http://arxiv.org/abs/1310.0307v2 | http://arxiv.org/pdf/1310.0307v2.pdf | Using the Random Sprays Retinex Algorithm for Global Illumination Estimation | In this paper the use of Random Sprays Retinex (RSR) algorithm for global
illumination estimation is proposed and its feasibility tested. Like other
algorithms based on the Retinex model, RSR also provides local illumination
estimation and brightness adjustment for each pixel and it is faster than other
path-wise Retinex algorithms. As the assumption of the uniform illumination
holds in many cases, it should be possible to use the mean of local
illumination estimations of RSR as a global illumination estimation for images
with (assumed) uniform illumination allowing also the accuracy to be easily
measured. Therefore we propose a method for estimating global illumination
estimation based on local RSR results. To our best knowledge this is the first
time that RSR algorithm is used to obtain global illumination estimation. For
our tests we use a publicly available color constancy image database for
testing. The results are presented and discussed and it turns out that the
proposed method outperforms many existing unsupervised color constancy
algorithms. The source code is available at
http://www.fer.unizg.hr/ipg/resources/color_constancy/. | ['Sven Lončarić', 'Nikola Banić'] | 2013-10-01 | null | null | null | null | ['color-constancy'] | ['computer-vision'] | [ 2.47022614e-01 -5.56704223e-01 -2.41602901e-02 -3.50847334e-01
-4.44680214e-01 -5.26091456e-01 4.02641684e-01 -2.23863691e-01
-3.73502940e-01 9.57764089e-01 -2.72452712e-01 -3.82615685e-01
6.11721165e-02 -6.57893538e-01 -4.19479787e-01 -1.05290008e+00
2.73471087e-01 -1.30872028e-02 2.45126501e-01 -2.86013871e-01
5.87314785e-01 8.19625974e-01 -1.91250086e+00 -3.59120280e-01
9.30055141e-01 7.96381235e-01 3.61743510e-01 1.02498639e+00
5.25970384e-02 4.98168468e-01 -6.61377907e-01 7.82765225e-02
4.90612984e-01 -7.50485539e-01 -7.92299449e-01 2.24390045e-01
7.05910146e-01 -4.15107697e-01 3.61489177e-01 1.26610637e+00
6.31701112e-01 4.91561264e-01 6.11883223e-01 -9.89379525e-01
-6.44932091e-01 -1.92482680e-01 -1.02254999e+00 7.06553757e-02
4.82288182e-01 1.23633733e-02 5.71523309e-01 -7.63345659e-01
5.22941530e-01 9.54337120e-01 2.57335842e-01 3.32241952e-01
-1.05524993e+00 -4.61287707e-01 8.39118063e-02 2.33946204e-01
-1.40899861e+00 -5.19581378e-01 8.86259496e-01 1.03771396e-01
5.52122116e-01 7.02256322e-01 5.89159191e-01 3.57665092e-01
6.57872707e-02 4.59032834e-01 2.20837760e+00 -1.01627159e+00
3.00078303e-01 4.19529349e-01 7.87435621e-02 8.46210063e-01
1.95882499e-01 4.80318256e-02 -2.46948376e-01 1.89455956e-01
9.63576972e-01 -2.69880623e-01 -3.99987847e-01 -3.17165181e-02
-9.02384877e-01 3.56317550e-01 6.01879478e-01 3.14083457e-01
-3.27824563e-01 1.74172908e-01 -3.77644375e-02 3.26592922e-01
7.07892060e-01 1.76956281e-01 -2.82917231e-01 4.80773002e-02
-8.21962357e-01 -7.94786885e-02 4.67816204e-01 7.80873060e-01
9.66659904e-01 2.71511585e-01 1.72279254e-01 1.14253259e+00
3.26141506e-01 8.72781694e-01 3.79138023e-01 -1.04325747e+00
-2.18542367e-01 3.08509469e-01 4.28958297e-01 -9.60496902e-01
-1.96100831e-01 -9.12219584e-02 -4.77867097e-01 9.87802029e-01
5.28487325e-01 -6.60954043e-02 -1.09505582e+00 1.36186957e+00
5.41117549e-01 2.62457654e-02 -3.48869190e-02 1.10902882e+00
6.58475578e-01 7.03370988e-01 -1.56132072e-01 -5.48099458e-01
1.19352400e+00 -8.83761406e-01 -1.06682110e+00 3.14044148e-01
1.21239990e-01 -1.35416067e+00 8.49577904e-01 8.00646901e-01
-1.12675011e+00 -6.13374054e-01 -1.05995595e+00 -9.25959870e-02
-7.24563301e-01 3.82675529e-01 6.60494149e-01 1.13933933e+00
-1.43485212e+00 2.63736695e-01 -4.61826086e-01 -7.69311845e-01
-1.54522732e-01 1.35090426e-01 -2.11864322e-01 -1.08303279e-01
-6.23239219e-01 1.01520729e+00 1.91287175e-01 3.94586086e-01
-4.78789330e-01 -1.14995077e-01 -5.58713555e-01 -6.02301538e-01
1.49793819e-01 -6.30126297e-02 8.67695272e-01 -1.57043898e+00
-2.02130628e+00 1.00091684e+00 -6.62755370e-01 1.50020599e-01
6.66847587e-01 -2.24611145e-02 -4.53820229e-01 1.71213999e-01
-3.94322872e-01 4.41310793e-01 9.63283777e-01 -1.82530034e+00
-3.72327834e-01 -3.08549553e-01 -7.84233026e-03 3.18658322e-01
7.21398443e-02 3.28919977e-01 -5.32853127e-01 -2.30347559e-01
3.47843766e-01 -7.52366126e-01 1.26861528e-01 2.77854711e-01
-2.73634344e-01 -5.92707517e-03 8.31709027e-01 -8.61097813e-01
8.39977860e-01 -1.89566541e+00 -3.53721201e-01 5.41061044e-01
-1.82749122e-01 2.90686309e-01 -1.61128104e-01 2.28514060e-01
-3.12833846e-01 -2.18445286e-01 -2.31575027e-01 -2.98885882e-01
-1.42987072e-01 1.11743718e-01 1.27833220e-03 1.06939936e+00
-2.66865224e-01 4.67989892e-01 -9.03517425e-01 -6.87216282e-01
6.18484855e-01 7.17595994e-01 -8.04937724e-03 1.88347563e-01
1.20620370e-01 3.68751675e-01 1.21553093e-01 9.72064435e-01
1.44425225e+00 3.89667034e-01 1.66832954e-02 -3.56236219e-01
-6.81075215e-01 -3.32870215e-01 -1.38977385e+00 1.19701588e+00
-4.67261314e-01 8.53065431e-01 -7.04432093e-03 -7.37964690e-01
1.34892702e+00 2.97945529e-01 2.80551553e-01 -8.35751653e-01
3.50630075e-01 3.33735973e-01 -3.71822953e-01 -2.45995656e-01
7.93742001e-01 -9.20176655e-02 7.36223161e-01 3.41852397e-01
-1.81970343e-01 -5.12219548e-01 3.01351815e-01 -1.28729612e-01
2.73799658e-01 7.05930829e-01 5.35597801e-01 -4.60710257e-01
8.50016415e-01 -1.40765011e-01 2.98371166e-01 5.34459293e-01
-5.28785586e-01 6.45090342e-01 2.63744779e-02 -2.15226248e-01
-8.18206251e-01 -1.04837239e+00 -4.70530868e-01 9.98817503e-01
5.60493112e-01 6.13142848e-02 -6.47000313e-01 -1.96465746e-01
-4.81433779e-01 7.79077768e-01 -6.04356945e-01 5.15632331e-01
-2.21579373e-01 -9.55768764e-01 -9.64777470e-02 -3.20858578e-03
9.59784448e-01 -1.01308286e+00 -3.29055369e-01 -2.88348168e-01
-1.57353997e-01 -8.10554147e-01 -3.21277566e-02 8.72402191e-02
-9.07994032e-01 -1.07697248e+00 -9.95026112e-01 -5.56386054e-01
1.09876013e+00 7.22417593e-01 9.33315635e-01 3.19123894e-01
-5.09377062e-01 6.91767693e-01 -6.15852237e-01 -4.50724959e-01
-3.33831310e-01 -5.33259153e-01 -1.95801303e-01 3.28940116e-02
3.18578154e-01 -1.34585008e-01 -8.85542393e-01 6.09128356e-01
-8.54396284e-01 -4.66448441e-02 2.23482624e-01 3.66754323e-01
8.95172000e-01 2.69319177e-01 1.62434250e-01 -9.55524981e-01
3.37478399e-01 -1.03273235e-01 -1.09610605e+00 3.48780006e-01
-6.22181773e-01 -3.30158621e-01 3.20960850e-01 -3.51244137e-02
-1.52466452e+00 -1.47969902e-01 2.53756344e-02 2.68562436e-02
-4.21541393e-01 -2.74554670e-01 1.09656170e-01 -5.80913365e-01
6.85518980e-01 2.98856407e-01 -1.75266445e-01 -3.65322769e-01
4.60993171e-01 6.28957808e-01 4.61191535e-01 -5.51291704e-01
8.75183702e-01 6.59480751e-01 3.11552256e-01 -1.21350563e+00
-3.62360120e-01 -6.88249886e-01 -6.13872886e-01 -8.10571373e-01
9.75470722e-01 -5.22805989e-01 -6.90676451e-01 7.75008380e-01
-8.15446436e-01 -4.67796475e-01 -5.33724204e-02 2.96688080e-01
-6.24479890e-01 4.89075869e-01 -7.21435770e-02 -1.47570729e+00
-1.55277252e-01 -9.39836860e-01 8.56232643e-01 5.83717763e-01
2.55583823e-01 -1.37639010e+00 1.26347095e-01 2.82917917e-01
6.48199379e-01 3.71399879e-01 2.27898464e-01 4.71043676e-01
-5.72727978e-01 -1.21495344e-01 -4.44800168e-01 5.47794759e-01
5.99880040e-01 6.63988531e-01 -1.28749371e+00 -2.81668186e-01
-8.13841671e-02 -1.95301697e-01 6.95784509e-01 6.00623727e-01
1.12932432e+00 4.87723462e-02 2.28103638e-01 6.33109450e-01
2.30160689e+00 4.74700183e-02 1.10526621e+00 5.69112539e-01
4.37170953e-01 3.95772070e-01 1.03775680e+00 3.38788122e-01
1.08720465e-02 5.66357017e-01 6.48942530e-01 -6.61116183e-01
-4.81285334e-01 2.93895185e-01 3.79212260e-01 4.44644272e-01
-7.35217690e-01 -3.04345608e-01 -4.57984865e-01 3.69809121e-01
-1.42656946e+00 -1.02734625e+00 -4.02366161e-01 2.60165930e+00
8.49170744e-01 -5.32823026e-01 1.07602522e-01 2.74385929e-01
8.57922435e-01 1.67671204e-01 -7.52693191e-02 -1.02154934e+00
-2.33614609e-01 2.40845740e-01 8.70993555e-01 9.91723955e-01
-1.01807797e+00 8.62307847e-01 6.61482954e+00 5.70641875e-01
-1.14094818e+00 9.79516804e-02 7.20026374e-01 2.80100971e-01
-1.12642743e-01 3.83187900e-03 -5.46147346e-01 2.99309224e-01
4.82520908e-01 2.86267281e-01 6.79811656e-01 3.66283655e-01
6.64429426e-01 -9.93983507e-01 -1.79185599e-01 8.49866986e-01
2.48565257e-01 -3.95840615e-01 -1.71753854e-01 -1.95009872e-01
1.14638317e+00 -2.04368696e-01 2.49450997e-01 -3.94745559e-01
2.79498905e-01 -7.57218122e-01 2.82975882e-01 7.25677609e-01
8.98509920e-01 -6.79209232e-01 7.46102214e-01 -1.46157771e-01
-1.12043262e+00 3.29847604e-01 -7.71306515e-01 2.21536204e-01
-2.33989507e-01 5.82358420e-01 -5.70954800e-01 6.21081471e-01
6.00586116e-01 4.62560236e-01 -8.99371803e-01 1.35473740e+00
-4.20035273e-01 4.89127755e-01 -3.54797572e-01 -1.35685042e-01
6.07666224e-02 -6.92171514e-01 4.35618907e-01 1.18909538e+00
3.25343281e-01 -8.07151273e-02 -3.20151985e-01 7.09241033e-01
3.07896644e-01 5.64580262e-01 -4.18508112e-01 3.95461828e-01
1.34585932e-01 1.54856610e+00 -9.27368999e-01 -2.74713159e-01
-3.92165393e-01 1.22040081e+00 -1.28886312e-01 8.25181842e-01
-7.33097613e-01 -5.57992995e-01 3.27430040e-01 -1.34575516e-01
-7.83720240e-02 -2.72810102e-01 -1.65542394e-01 -9.64976013e-01
-3.40582490e-01 -6.65259123e-01 2.15370223e-01 -1.38496315e+00
-8.03221405e-01 3.54302317e-01 9.41666588e-02 -1.32012880e+00
2.45538920e-01 -8.26845109e-01 -7.03669071e-01 1.24021900e+00
-2.10736823e+00 -1.08626235e+00 -8.01071227e-01 8.47474396e-01
4.43210572e-01 1.16105258e-01 7.80228674e-01 -1.38964549e-01
-4.87389535e-01 4.75113451e-01 4.19250220e-01 -3.07943821e-01
1.08044291e+00 -1.58330607e+00 -3.41656655e-01 1.26292300e+00
-3.19069803e-01 6.28591657e-01 9.98677909e-01 -4.29556042e-01
-1.28681946e+00 -7.77345836e-01 5.14453709e-01 -1.96919277e-01
4.24563736e-01 1.05994865e-01 -7.09833145e-01 4.91459042e-01
6.46667302e-01 2.70013288e-02 5.37192822e-01 -1.20332532e-01
-1.80169389e-01 -4.16739702e-01 -1.44129670e+00 5.94048023e-01
5.23125291e-01 -1.93233386e-01 -4.05549966e-02 5.31397045e-01
2.65817177e-02 -3.14422786e-01 -7.74198771e-01 -1.33227319e-01
5.08527875e-01 -1.35956359e+00 7.70113230e-01 2.90033162e-01
3.61333713e-02 -8.30172420e-01 -3.32859933e-01 -1.05210054e+00
-6.66179806e-02 -6.54597402e-01 4.41740006e-01 1.36897290e+00
1.45703867e-01 -1.06868327e+00 3.31686348e-01 4.45139050e-01
3.17687154e-01 -2.14214668e-01 -5.97812831e-01 -7.56467402e-01
-3.02462608e-01 -2.56291121e-01 2.51079768e-01 7.79166698e-01
-4.40279424e-01 -4.19052303e-01 -4.52686757e-01 3.84450704e-01
1.05223370e+00 2.35137224e-01 8.56954157e-01 -7.86696553e-01
-2.20843330e-01 -2.43157431e-01 -4.21475887e-01 -6.20842814e-01
-1.10248290e-01 -3.49355817e-01 1.99188918e-01 -1.69020116e+00
8.11723620e-02 -5.03196120e-01 -5.08790970e-01 4.21937138e-01
-3.48251820e-01 1.01653147e+00 -4.52024862e-02 -6.13273587e-03
-4.11085874e-01 1.68589205e-01 1.68556786e+00 1.31660029e-01
-2.58954376e-01 1.06552847e-01 -4.79528964e-01 4.73765641e-01
1.25895953e+00 9.69517678e-02 -4.45923299e-01 -1.76969230e-01
1.22149378e-01 -5.27797282e-01 3.03038180e-01 -8.78344297e-01
-7.84123316e-02 -3.92694622e-01 6.13623738e-01 -5.80101490e-01
4.84595418e-01 -9.80476856e-01 -3.03117950e-02 1.30247071e-01
6.27024695e-02 -1.37513746e-02 1.17935129e-01 1.65062830e-01
-6.84083253e-02 -3.91289443e-01 1.35253000e+00 -1.48022726e-01
-9.98214543e-01 -7.92690963e-02 -3.37955475e-01 -5.57354510e-01
8.97770643e-01 -5.85760117e-01 -5.87643027e-01 -5.34267604e-01
-2.30134949e-01 -3.71444464e-01 9.47413683e-01 -9.73121747e-02
6.81575537e-01 -1.19967353e+00 -5.90497255e-01 2.95068949e-01
1.74853519e-01 -5.85156500e-01 1.61309332e-01 1.02592266e+00
-1.25326169e+00 -8.75472352e-02 -4.34209406e-01 -4.35336083e-01
-1.66251159e+00 3.68048310e-01 4.82746094e-01 5.29184222e-01
-2.26842970e-01 6.26382351e-01 -1.29305705e-01 -1.60104975e-01
-1.34611219e-01 4.68589440e-02 -2.97165424e-01 -3.54058892e-01
6.86253309e-01 7.00160921e-01 -4.55537215e-02 -8.56240928e-01
-2.03253672e-01 1.20659602e+00 3.38539869e-01 -2.74760842e-01
1.10885322e+00 -5.19729495e-01 -8.23564768e-01 5.21975756e-01
1.21873903e+00 3.51966769e-01 -1.11742163e+00 4.84606512e-02
-6.56204939e-01 -1.11033201e+00 3.80598933e-01 -9.03433859e-01
-1.13848174e+00 6.38917804e-01 1.27691412e+00 2.22719595e-01
1.88373160e+00 -4.66774017e-01 2.24410035e-02 2.11715877e-01
3.00602913e-01 -1.38249683e+00 -1.63479909e-01 1.12288885e-01
8.56549442e-01 -1.33404291e+00 5.75997531e-01 -4.32636410e-01
-5.08326054e-01 1.28533101e+00 4.81009811e-01 -3.15311879e-01
5.16140819e-01 4.45415303e-02 6.50812089e-01 2.01698482e-01
-2.40436316e-01 -6.07546806e-01 2.53203630e-01 8.60411286e-01
7.51692235e-01 9.85737741e-02 -4.75261241e-01 -8.57061565e-01
9.27058756e-02 -2.08422631e-01 7.44710505e-01 6.67570353e-01
-7.90788889e-01 -1.19013107e+00 -9.60229337e-01 -4.25767601e-02
-3.51761490e-01 -2.22298298e-02 -2.82201111e-01 9.53620732e-01
5.30903526e-02 1.23846269e+00 -1.46378130e-01 1.91624716e-01
-1.29031599e-01 -1.26563579e-01 7.94159293e-01 -1.73281319e-02
-1.24090984e-01 2.37382039e-01 -1.13405555e-01 -6.61312044e-01
-9.90781844e-01 -5.89861095e-01 -9.44862247e-01 -5.43088973e-01
-4.79925513e-01 -1.43789938e-02 1.23068404e+00 3.39677066e-01
-4.35242802e-01 3.38155121e-01 9.30010617e-01 -1.03765190e+00
2.99925178e-01 -7.61458814e-01 -1.03036439e+00 2.64225662e-01
3.77658159e-01 -4.78406519e-01 -5.71098387e-01 2.15558454e-01] | [10.4093656539917, -2.5917160511016846] |
003bdc2a-cd44-497d-98f7-e7f735e8fe77 | n15news-a-new-dataset-for-multimodal-news | 2108.13327 | null | https://arxiv.org/abs/2108.13327v4 | https://arxiv.org/pdf/2108.13327v4.pdf | N24News: A New Dataset for Multimodal News Classification | Current news datasets merely focus on text features on the news and rarely leverage the feature of images, excluding numerous essential features for news classification. In this paper, we propose a new dataset, N24News, which is generated from New York Times with 24 categories and contains both text and image information in each news. We use a multitask multimodal method and the experimental results show multimodal news classification performs better than text-only news classification. Depending on the length of the text, the classification accuracy can be increased by up to 8.11%. Our research reveals the relationship between the performance of a multimodal classifier and its sub-classifiers, and also the possible improvements when applying multimodal in news classification. N24News is shown to have great potential to prompt the multimodal news studies. | ['Jie Yang', 'Xiangxie Zhang', 'Xu Shan', 'Zhen Wang'] | 2021-08-30 | null | https://aclanthology.org/2022.lrec-1.729 | https://aclanthology.org/2022.lrec-1.729.pdf | lrec-2022-6 | ['news-classification'] | ['natural-language-processing'] | [-1.12895317e-01 -1.87189415e-01 -4.14981633e-01 -4.69791800e-01
-1.06696725e+00 -6.30552590e-01 1.18561316e+00 1.74307659e-01
-4.18586671e-01 7.68235981e-01 8.82141709e-01 1.39379818e-02
1.84666682e-02 -4.64241922e-01 -6.43270254e-01 -8.33420217e-01
1.36330947e-01 2.83457726e-01 -9.86905172e-02 -3.14231128e-01
6.50426984e-01 -1.71170145e-01 -1.61787665e+00 1.12998497e+00
4.06805187e-01 1.21220517e+00 1.45764634e-01 6.23605311e-01
-4.80434567e-01 6.97722673e-01 -6.31501913e-01 -6.36605859e-01
-5.52276243e-03 -3.09986562e-01 -6.75220430e-01 2.53769845e-01
6.22767746e-01 -5.74224181e-02 -4.14805770e-01 8.11244071e-01
5.96799731e-01 1.02270097e-01 9.11706805e-01 -1.17986071e+00
-8.68731558e-01 9.44478929e-01 -6.52009785e-01 2.72834301e-01
6.70238554e-01 -3.80151004e-01 8.01636994e-01 -8.49815547e-01
6.81869090e-01 1.27064669e+00 4.72676486e-01 1.25110462e-01
-8.03138316e-01 -4.28211123e-01 -1.04062133e-01 2.01699793e-01
-1.00428164e+00 -5.30516267e-01 5.70798934e-01 -5.60708880e-01
6.49682581e-01 2.52561241e-01 5.48366427e-01 1.44335151e+00
6.41233683e-01 9.06416476e-01 1.38642359e+00 -4.25888658e-01
-1.47015497e-01 3.13381284e-01 1.97650969e-01 6.05583966e-01
-1.74402580e-01 -3.63590866e-01 -8.77088130e-01 -2.32607201e-01
2.85679221e-01 2.81072520e-02 -2.16284782e-01 8.60978365e-02
-1.78752804e+00 1.01324856e+00 -1.50205696e-03 2.77083158e-01
-9.68938246e-02 -1.74587637e-01 8.98489833e-01 5.32555282e-01
8.79558563e-01 2.93827802e-01 -5.64757586e-01 -3.49079043e-01
-6.02514863e-01 -1.79957747e-01 6.79544806e-01 7.84364283e-01
3.73879611e-01 -2.68427193e-01 -1.97205082e-01 1.06960809e+00
8.00430253e-02 7.67588913e-01 7.16713369e-01 -5.53503871e-01
6.37939751e-01 6.15606725e-01 -4.71432209e-02 -1.03598964e+00
-6.85192108e-01 -1.86222389e-01 -5.41941345e-01 -3.27974260e-01
2.57605076e-01 -3.77613068e-01 -8.78426790e-01 1.13312924e+00
1.07488215e-01 -2.81180441e-01 3.57013762e-01 6.98167980e-01
1.47611332e+00 9.61241424e-01 -1.71145145e-02 -1.91160515e-01
1.63403690e+00 -1.07832611e+00 -7.95137048e-01 1.12083793e-01
5.91798663e-01 -1.30137062e+00 1.08119774e+00 3.81433189e-01
-7.08480716e-01 -3.04770499e-01 -6.72822237e-01 2.27283314e-01
-7.85172999e-01 7.34329045e-01 5.94601154e-01 6.59020066e-01
-7.29372442e-01 -2.27122530e-01 -8.86586532e-02 -8.15695226e-01
-1.19228989e-01 1.24301128e-01 -7.41329253e-01 -2.27028772e-01
-1.00983822e+00 8.44113171e-01 4.06983763e-01 -3.28874052e-01
-7.32275128e-01 4.47136629e-03 -8.84371042e-01 -2.47403115e-01
3.15951318e-01 -5.83041608e-01 9.19538617e-01 -1.17716289e+00
-1.40882075e+00 7.13841200e-01 -1.10437416e-01 1.83458626e-02
3.39217901e-01 1.33217543e-01 -6.95325315e-01 3.96796644e-01
2.30797380e-01 7.12530494e-01 9.73747075e-01 -1.36888862e+00
-8.68792236e-01 -2.34246105e-01 1.35135591e-01 3.97740334e-01
-6.55798793e-01 2.04207867e-01 -4.42230225e-01 -7.33656943e-01
-3.21896859e-02 -1.05590737e+00 4.41806495e-01 -5.78330815e-01
-4.76698667e-01 -1.59234852e-01 9.26995695e-01 -4.91950154e-01
6.18388474e-01 -2.25830793e+00 4.28269207e-02 -1.18896678e-01
-4.85987626e-02 -3.66818398e-01 -4.52854097e-01 6.73537016e-01
1.39685318e-01 1.06162392e-01 3.14505637e-01 -4.40780729e-01
-1.06036872e-01 1.08754180e-01 -2.48089224e-01 6.08172834e-01
-2.66377807e-01 8.69524419e-01 -2.21495509e-01 -4.60893303e-01
-9.44278166e-02 3.62650812e-01 7.82929584e-02 -2.41828337e-01
-3.54825556e-02 4.33475971e-01 -4.64876115e-01 8.32533717e-01
4.14008558e-01 -2.11239472e-01 -1.41293079e-01 -4.33284670e-01
-2.63950288e-01 -1.27921119e-01 -5.56912959e-01 1.41788507e+00
-1.70718104e-01 8.74305546e-01 -6.31281286e-02 -1.00698793e+00
7.87624061e-01 4.80813920e-01 4.02772456e-01 -8.00482512e-01
4.95652020e-01 -2.89443970e-01 -3.81476074e-01 -8.89598191e-01
7.45208919e-01 -3.95990722e-02 -4.07842934e-01 6.14574432e-01
-4.12540846e-02 2.45081186e-01 2.26212919e-01 2.10642397e-01
5.32259941e-01 -1.80735961e-01 4.06917296e-02 -1.27048999e-01
1.60474434e-01 3.03728938e-01 1.18172336e-02 7.85669088e-01
5.13513498e-02 4.78322148e-01 4.72415179e-01 -3.73056650e-01
-1.08302283e+00 -5.17262757e-01 -3.43275666e-01 1.65047204e+00
2.25068390e-01 -2.84492016e-01 -3.12684804e-01 -9.23672318e-01
-9.42990258e-02 4.35700446e-01 -8.87744784e-01 2.13136181e-01
1.50545970e-01 -1.27624202e+00 6.30826712e-01 2.55444229e-01
5.78218341e-01 -4.93242174e-01 -2.17273701e-02 -2.92377591e-01
-6.84745669e-01 -1.28080595e+00 -4.78948653e-01 -8.24506208e-02
-5.73388219e-01 -8.78610671e-01 -9.98596668e-01 -8.02354813e-01
5.91918528e-01 5.54321468e-01 5.76988637e-01 -3.84707659e-01
1.56385541e-01 1.03630424e+00 -1.04780722e+00 -3.45406324e-01
-2.71695226e-01 2.57521749e-01 -1.21816538e-01 3.41102123e-01
-5.24531864e-02 1.32084385e-01 -2.59384334e-01 4.56615597e-01
-8.64660025e-01 2.74219692e-01 7.93384850e-01 8.91175985e-01
4.09926958e-02 5.38508222e-02 8.10219705e-01 -6.56313002e-01
6.69806957e-01 -9.16358948e-01 6.30342662e-02 5.49017012e-01
-4.56653595e-01 -3.35940540e-01 2.60567605e-01 -6.86790884e-01
-1.28981102e+00 2.30242498e-03 3.13337326e-01 1.61393896e-01
-3.19842368e-01 1.19002414e+00 3.22064519e-01 -1.52964473e-01
4.19751883e-01 4.64946598e-01 1.96540821e-02 -2.55879462e-01
2.15650171e-01 9.07134891e-01 2.92037148e-02 -3.99764836e-01
2.13552624e-01 5.88632345e-01 -2.36416891e-01 -9.90359187e-01
-9.10906672e-01 -5.35671890e-01 -3.06670785e-01 -7.41440892e-01
9.97403860e-01 -1.15284371e+00 -6.46902978e-01 3.27739269e-01
-8.68656635e-01 1.92506224e-01 3.80466312e-01 9.88145709e-01
-2.35255972e-01 4.69261080e-01 -8.08906853e-01 -7.79768169e-01
-5.05004525e-02 -1.05588198e+00 1.11728120e+00 -4.19679023e-02
2.61149585e-01 -1.00249577e+00 -2.73660600e-01 1.01316810e+00
4.02527481e-01 4.29702289e-02 8.94559205e-01 -1.11212242e+00
-1.07588343e-01 -5.06531656e-01 -4.37024325e-01 -1.16496012e-01
-1.26034170e-01 6.35223538e-02 -9.16496456e-01 -1.86847329e-01
-1.57714143e-01 -7.60780215e-01 1.07122314e+00 5.00317991e-01
4.49515909e-01 -2.99458414e-01 -1.96663439e-01 1.70998853e-02
1.14045739e+00 3.77675772e-01 5.08635640e-01 5.82926333e-01
4.68690544e-01 7.47289002e-01 6.32715881e-01 6.43161952e-01
8.19628716e-01 3.28747720e-01 2.67939031e-01 -6.92588538e-02
9.79567766e-02 1.95272863e-01 7.25910187e-01 1.06189609e+00
-1.81841925e-01 -6.05676889e-01 -1.03661466e+00 2.28012741e-01
-1.91787291e+00 -1.11742675e+00 -3.16680223e-01 1.56508183e+00
5.50614059e-01 -2.21323580e-01 9.95447785e-02 -1.62735969e-01
9.34962571e-01 1.65015087e-01 1.73416343e-02 -8.03178549e-02
-6.38650656e-01 -8.71014714e-01 4.01501298e-01 7.88068920e-02
-1.48033786e+00 5.59956551e-01 7.42956066e+00 9.79015827e-01
-1.21590722e+00 2.54583240e-01 7.38623977e-01 -1.33197889e-01
-1.59173399e-01 -1.97847262e-01 -7.80351281e-01 3.63874972e-01
8.72291684e-01 -8.13009664e-02 1.39597014e-01 7.76493788e-01
1.26615927e-01 -6.21277511e-01 -6.79140627e-01 1.06416786e+00
9.64575529e-01 -1.35652161e+00 4.89991993e-01 5.68717681e-02
9.96280015e-01 4.61516641e-02 2.44762614e-01 4.58736211e-01
-3.38868320e-01 -7.30469942e-01 8.04664433e-01 7.30463862e-01
6.04918242e-01 -8.36251497e-01 1.24127591e+00 4.85984027e-01
-9.30131376e-01 -3.34368378e-01 -1.81471303e-01 -8.88053887e-03
4.16216701e-02 2.35286742e-01 -6.97866559e-01 5.96468627e-01
7.29314029e-01 7.87646055e-01 -8.17359149e-01 1.04193974e+00
5.49752414e-01 4.24117744e-01 7.64182359e-02 -3.62724572e-01
2.33970746e-01 -9.07327384e-02 4.96003956e-01 1.46278059e+00
5.52506149e-01 4.94239144e-02 5.40947735e-01 -2.43023992e-01
-4.85373996e-02 6.57134056e-01 -6.14258230e-01 -3.05793315e-01
-6.89612404e-02 1.25721252e+00 -1.07243216e+00 -5.47772944e-01
-8.74123275e-01 8.31723511e-01 -1.56089917e-01 4.24676120e-01
-9.63249803e-01 -2.54917979e-01 -3.00353378e-01 -1.71352282e-01
2.70083636e-01 -2.12668717e-01 -2.78780907e-01 -1.47407949e+00
-3.24735403e-01 -7.83456981e-01 5.49445868e-01 -1.09183371e+00
-1.59116364e+00 7.65111148e-01 1.21780470e-01 -1.22482932e+00
3.82007331e-01 -5.34635782e-01 -2.42406279e-01 1.50368363e-01
-1.28684342e+00 -1.71924925e+00 -1.85744837e-01 7.91900575e-01
8.10322165e-01 -7.48789966e-01 6.96700931e-01 3.22765827e-01
-4.41252589e-01 4.58303899e-01 4.86282200e-01 1.88027501e-01
1.13227057e+00 -7.83278644e-01 -5.68130493e-01 2.23902479e-01
-8.22090507e-02 2.04812184e-01 6.52270973e-01 -6.89167500e-01
-1.66691041e+00 -7.38322377e-01 8.79648268e-01 -3.14082772e-01
8.00197303e-01 3.62212099e-02 -2.51901150e-01 6.38913572e-01
7.74708092e-01 -6.72739089e-01 1.10554159e+00 2.63913393e-01
-4.09055889e-01 -2.66284514e-02 -1.01378632e+00 4.73044574e-01
3.90290260e-01 -6.39605880e-01 -7.99505651e-01 4.76897806e-01
4.72578198e-01 -1.46686330e-01 -1.00512564e+00 2.06474766e-01
6.67262733e-01 -7.44100928e-01 5.89922130e-01 -4.97569948e-01
6.33976579e-01 7.23191649e-02 -4.96667027e-01 -1.30345404e+00
-9.78460684e-02 7.90918767e-02 4.46137160e-01 1.38767505e+00
8.78906727e-01 -7.36598313e-01 4.33873057e-01 3.73685062e-01
-2.50620633e-01 -3.18490446e-01 -9.94046450e-01 -4.14762586e-01
-2.67901361e-01 -4.76866454e-01 -3.86377759e-02 1.56303108e+00
4.56654370e-01 5.90394318e-01 -7.05065072e-01 -9.17310715e-02
1.60933957e-01 1.68355495e-01 7.43927479e-01 -1.03772950e+00
1.63970709e-01 -3.57628614e-01 -4.73219931e-01 -7.96417058e-01
2.56995082e-01 -7.38200486e-01 -1.60322562e-01 -1.58501792e+00
8.16469252e-01 5.63960448e-02 -4.39263023e-02 5.83224177e-01
2.44596183e-01 5.19186437e-01 4.65089828e-02 3.95141602e-01
-8.92434657e-01 6.91634476e-01 1.51211274e+00 -4.16995406e-01
2.12998420e-01 -1.38759777e-01 -5.05466759e-01 7.40049481e-01
7.14105666e-01 -3.53129745e-01 -2.57177293e-01 -3.52698624e-01
4.95806575e-01 3.70713115e-01 1.74695358e-01 -5.59268117e-01
3.34681392e-01 -3.01748723e-01 7.05439150e-01 -7.00864911e-01
6.34005010e-01 -7.44812012e-01 -5.54569997e-02 -1.73055887e-01
-5.84875643e-01 2.92310566e-01 1.70002699e-01 8.09066296e-01
-6.66917682e-01 -1.79846227e-01 2.87395090e-01 -9.84502658e-02
-6.62661254e-01 -7.56645799e-02 -8.35325003e-01 -3.34201336e-01
1.18651664e+00 -6.12511262e-02 -1.01951432e+00 -8.01877022e-01
-9.85622704e-01 1.02665253e-01 2.18649358e-01 5.18992603e-01
6.65589154e-01 -1.31090045e+00 -7.93697655e-01 -2.39181563e-01
4.63381439e-01 -9.50489461e-01 3.50330353e-01 1.25176644e+00
-1.89154491e-01 3.80654126e-01 -2.22136393e-01 -5.30120969e-01
-1.62881505e+00 3.57541144e-01 -1.18250713e-01 4.35765207e-01
-3.07308912e-01 4.15231079e-01 4.45840172e-02 -1.87072173e-01
3.26325029e-01 2.43159775e-02 -8.02242756e-01 9.61339295e-01
5.02767205e-01 3.33594114e-01 -2.08173260e-01 -1.03006899e+00
-1.93142816e-02 6.57349527e-01 -8.83989315e-03 -3.59798729e-01
1.40130913e+00 -6.65314376e-01 -3.26199114e-01 9.38373446e-01
1.07714450e+00 2.16977626e-01 -5.56232572e-01 -1.65502638e-01
-1.17010884e-01 -4.04992849e-01 1.30604327e-01 -1.00869942e+00
-9.75038767e-01 3.77668411e-01 5.69528043e-01 7.26425529e-01
1.02090752e+00 4.49791044e-01 7.13511229e-01 6.74409986e-01
1.20049730e-01 -1.20721567e+00 3.68696243e-01 7.89871037e-01
1.05146587e+00 -1.77646768e+00 1.12520300e-01 -1.01520382e-01
-1.40349829e+00 1.46249521e+00 1.77051097e-01 3.39531839e-01
6.41831398e-01 1.53235674e-01 3.80634248e-01 -2.94855207e-01
-8.40879798e-01 -5.57141043e-02 6.13198876e-01 2.55486518e-01
4.72405225e-01 -6.79904744e-02 -5.11932790e-01 6.10408187e-01
-9.99677181e-03 -4.78260040e-01 6.84072793e-01 9.31193233e-01
-7.18921244e-01 -7.86873698e-01 -7.25863039e-01 5.16703427e-01
-7.09128499e-01 -1.02590904e-01 -4.96426463e-01 6.87017143e-01
1.57881826e-01 1.19826436e+00 -1.81106087e-02 -5.61613619e-01
1.24944612e-01 2.83013970e-01 7.32883811e-02 -3.42200309e-01
-4.20317829e-01 5.26884317e-01 7.24213481e-01 -4.90030646e-02
-1.04940307e+00 -8.49366188e-01 -9.16156888e-01 -3.81298482e-01
-4.24488902e-01 2.62531340e-01 1.18933785e+00 1.07054770e+00
3.05363506e-01 5.78321368e-02 6.36946678e-01 -1.10378671e+00
3.65873165e-02 -1.07980895e+00 -5.23167849e-01 2.13527188e-01
2.99613655e-01 -6.56023741e-01 -5.18534720e-01 4.34534967e-01] | [13.019491195678711, 5.232800483703613] |
619ba071-715e-4587-86a7-687021514d5d | something-old-something-new-grammar-based-ccg | 2109.10044 | null | https://arxiv.org/abs/2109.10044v2 | https://arxiv.org/pdf/2109.10044v2.pdf | Something Old, Something New: Grammar-based CCG Parsing with Transformer Models | This report describes the parsing problem for Combinatory Categorial Grammar (CCG), showing how a combination of Transformer-based neural models and a symbolic CCG grammar can lead to substantial gains over existing approaches. The report also documents a 20-year research program, showing how NLP methods have evolved over this time. The staggering accuracy improvements provided by neural models for CCG parsing can be seen as a reflection of the improvements seen in NLP more generally. The report provides a minimal introduction to CCG and CCG parsing, with many pointers to the relevant literature. It then describes the CCG supertagging problem, and some recent work from Tian et al. (2020) which applies Transformer-based models to supertagging with great effect. I use this existing model to develop a CCG multitagger, which can serve as a front-end to an existing CCG parser. Simply using this new multitagger provides substantial gains in parsing accuracy. I then show how a Transformer-based model from the parsing literature can be combined with the grammar-based CCG parser, setting a new state-of-the-art for the CCGbank parsing task of almost 93% F-score for labelled dependencies, with complete sentence accuracies of over 50%. | ['Stephen Clark'] | 2021-09-21 | null | null | null | null | ['ccg-supertagging'] | ['natural-language-processing'] | [ 5.04419692e-02 9.09770608e-01 -1.35424554e-01 -7.43758738e-01
-1.30647886e+00 -7.19931960e-01 4.11609411e-01 4.40776013e-02
-2.10711751e-02 6.02880597e-01 5.49648762e-01 -1.03862429e+00
1.50556386e-01 -8.54174137e-01 -5.79493403e-01 -4.55077350e-01
-2.42578566e-01 5.82439303e-01 1.31916910e-01 -5.46544969e-01
6.51817620e-02 8.51771310e-02 -1.11588383e+00 6.06365919e-01
4.44421768e-01 6.52314901e-01 3.65623176e-01 8.73600543e-01
-5.03543317e-01 7.47776210e-01 -7.70184577e-01 -5.94569385e-01
-1.65062800e-01 -3.09576988e-01 -1.12537348e+00 -4.15174961e-01
4.73109633e-01 2.70698816e-02 5.75021021e-02 6.43204510e-01
3.86817008e-01 -2.50531524e-01 1.48333237e-01 -8.40263188e-01
-7.03141630e-01 1.33934498e+00 1.10625543e-01 2.62494773e-01
3.22692692e-01 -3.13570023e-01 1.59424412e+00 -2.80042082e-01
5.46292126e-01 1.65912211e+00 9.21894014e-01 9.38751101e-01
-1.01158237e+00 -4.98190939e-01 6.01741850e-01 -3.58041227e-01
-5.07800579e-01 -1.96361735e-01 5.36473274e-01 -4.14541550e-02
1.89821756e+00 2.28709430e-01 5.11956573e-01 7.77246714e-01
4.19632196e-01 8.67338657e-01 1.00337052e+00 -9.48738098e-01
-7.60685578e-02 -4.98170584e-01 6.35661304e-01 7.45512486e-01
2.38294631e-01 2.77705729e-01 -3.11258227e-01 8.24328437e-02
4.99928445e-01 -6.96526468e-01 2.15120628e-01 2.83890754e-01
-8.25759351e-01 1.26425862e+00 4.16533142e-01 6.82774007e-01
8.94842297e-02 5.40282726e-01 7.15147495e-01 3.17284942e-01
6.60838604e-01 5.39124370e-01 -8.54526460e-01 -1.80301189e-01
-4.77802634e-01 3.08428973e-01 1.13525081e+00 1.13001513e+00
3.29400033e-01 2.32982561e-01 6.63034543e-02 1.01349270e+00
5.00650883e-01 2.19779477e-01 3.21028441e-01 -9.90084708e-01
9.72812533e-01 3.67341638e-01 -5.59150696e-01 -6.59683466e-01
-7.75332868e-01 -3.12410146e-01 -1.58675686e-01 -1.71640947e-01
5.99105656e-01 -1.37307361e-01 -9.85870719e-01 2.14188385e+00
-2.00754348e-02 -6.02463722e-01 1.14831567e-01 2.27864131e-01
9.67022061e-01 7.67587125e-01 8.27794075e-01 5.94700966e-03
1.59275234e+00 -5.84924996e-01 -6.16798699e-01 -7.31304884e-01
1.48579836e+00 -4.57550049e-01 9.12177444e-01 3.20352823e-01
-1.19960213e+00 -2.84148365e-01 -9.23051417e-01 -4.02393967e-01
-5.76670110e-01 -1.83946207e-01 1.38564849e+00 1.01970577e+00
-1.50969267e+00 7.18586326e-01 -1.04053342e+00 -6.80528700e-01
2.71985140e-02 6.60148978e-01 -2.66081542e-01 1.97697859e-02
-1.24455583e+00 1.01710069e+00 7.86807179e-01 8.18881765e-02
-2.26110995e-01 -2.63133079e-01 -1.12041676e+00 -1.80648983e-01
2.60415912e-01 -6.24927938e-01 1.76847422e+00 -5.44384480e-01
-1.29825568e+00 1.04461181e+00 -1.86649576e-01 -4.36270952e-01
-2.48589233e-01 -1.70416474e-01 -2.50441283e-01 -1.62835509e-01
4.02081251e-01 6.69154465e-01 5.59932925e-02 -9.40395951e-01
-9.39451575e-01 -5.30717850e-01 2.63869494e-01 2.29244187e-01
2.77638584e-01 5.65831184e-01 -2.15530425e-01 -6.44342422e-01
4.78861421e-01 -7.84575641e-01 -6.17136918e-02 -1.19118643e+00
-3.09196681e-01 -9.10455406e-01 5.48468471e-01 -9.58363295e-01
1.43220329e+00 -1.75975406e+00 -5.31753041e-02 -3.53427440e-01
1.16906926e-01 4.22139280e-02 -2.82092273e-01 6.48451746e-01
-3.67084891e-01 6.05009437e-01 -2.04327762e-01 -5.24228275e-01
2.48469070e-01 7.22781122e-01 -1.51968122e-01 -1.52249664e-01
3.18211019e-01 1.34044862e+00 -8.36040437e-01 -2.15316862e-01
6.21025637e-02 -2.17768233e-02 -5.17091155e-01 -2.34073196e-02
-2.36605868e-01 -9.09642875e-03 -2.41055578e-01 7.33987391e-01
2.81255513e-01 9.61089805e-02 8.00161660e-01 3.41930985e-01
-1.72910556e-01 1.12271488e+00 -3.04490715e-01 1.62254131e+00
-7.25787759e-01 3.41490626e-01 2.18276978e-01 -1.26282907e+00
8.86509120e-01 4.01768088e-01 -2.38262013e-01 -6.12573922e-01
3.87416542e-01 4.98636693e-01 2.76963860e-01 -3.58853266e-02
5.02270341e-01 -4.79848832e-01 -7.79373467e-01 4.36300427e-01
3.17768753e-01 -3.63900721e-01 2.25271717e-01 4.26721781e-01
1.23445678e+00 3.15230668e-01 3.48552763e-01 -5.06803870e-01
9.97084379e-02 1.74029082e-01 5.64661384e-01 9.30274904e-01
1.19352877e-01 4.49003249e-01 6.89972222e-01 -5.58753014e-01
-9.15438890e-01 -6.76304638e-01 1.31309867e-01 1.72086525e+00
-6.60853863e-01 -5.41085958e-01 -1.02121449e+00 -8.83465827e-01
-2.44054452e-01 9.09089684e-01 -7.03409851e-01 1.15080275e-01
-1.36046457e+00 -1.10001373e+00 7.92011380e-01 1.11855519e+00
3.75499964e-01 -1.40280700e+00 -2.39219159e-01 7.47899830e-01
-3.72372538e-01 -1.24421144e+00 1.03923857e-01 9.18377519e-01
-1.26813197e+00 -8.44571948e-01 -1.49767101e-02 -1.37131941e+00
5.26078828e-02 -8.55276138e-02 1.59835684e+00 3.48749101e-01
1.88637018e-01 -2.14127414e-02 -6.72709286e-01 -4.59049225e-01
-9.72374082e-01 4.98989075e-01 -4.55476850e-01 -1.16427386e+00
6.14913821e-01 -3.20657849e-01 2.92196721e-01 -2.06089810e-01
-5.58766782e-01 3.29173878e-02 3.01771611e-01 9.60284293e-01
7.13262036e-02 -3.83291304e-01 8.12362015e-01 -1.38946652e+00
5.90619624e-01 -2.54451185e-01 -4.77932394e-01 1.11041799e-01
-4.41299200e-01 1.50414765e-01 6.06658876e-01 1.44058898e-01
-9.48638618e-01 -2.31486857e-01 -9.28906500e-01 5.26205063e-01
-1.68971211e-01 6.69881701e-01 -5.55479109e-01 1.07946619e-01
3.06397796e-01 -2.75443465e-01 -2.28224307e-01 -6.56317711e-01
5.82019329e-01 5.56984186e-01 6.23934627e-01 -9.85516906e-01
3.36589217e-01 -2.54079759e-01 3.69610228e-02 -4.94353443e-01
-1.04510605e+00 -1.52595475e-01 -6.84840024e-01 3.63742739e-01
1.04495108e+00 -6.86109066e-01 -3.81578624e-01 4.59731340e-01
-1.43036389e+00 -7.54245222e-01 -2.20038760e-02 1.78729534e-01
-5.97793281e-01 5.06777108e-01 -1.22826791e+00 -6.15372479e-01
-4.16299552e-01 -8.30542684e-01 1.17690146e+00 -1.40518486e-01
-1.67992756e-01 -1.51918769e+00 -1.92022040e-01 3.66196126e-01
2.95932025e-01 1.56322792e-01 1.54698539e+00 -1.00722277e+00
-3.22729170e-01 -5.56213297e-02 -1.10766023e-01 3.72226357e-01
-1.89928070e-01 -5.71510315e-01 -8.47927749e-01 -1.03823595e-01
-9.40800980e-02 -1.46716252e-01 6.65768802e-01 5.88864625e-01
6.03722215e-01 -2.56370008e-01 -3.73340338e-01 5.93191564e-01
1.37812471e+00 5.20519614e-01 5.05630970e-01 6.41867697e-01
8.00952137e-01 8.02749574e-01 6.77261770e-01 -4.42797065e-01
7.86817074e-01 4.51581061e-01 4.75279808e-01 1.37931675e-01
-3.36668104e-01 -4.64978725e-01 2.59439528e-01 1.14154196e+00
-8.25075135e-02 -5.08157909e-01 -1.11074328e+00 5.95215201e-01
-1.65266180e+00 -5.35688162e-01 -4.82050091e-01 1.81304324e+00
6.07516110e-01 2.27572992e-01 8.49640518e-02 1.63118467e-01
7.80450940e-01 6.73504546e-02 1.66235611e-01 -1.35841668e+00
-1.25958413e-01 6.15357399e-01 5.64600587e-01 7.42038429e-01
-1.36267090e+00 1.58482587e+00 7.83932495e+00 5.46299517e-01
-6.78631961e-01 2.45324433e-01 5.96412659e-01 5.76643646e-01
-2.54664749e-01 2.27018654e-01 -1.30042648e+00 1.99945822e-01
1.65048587e+00 2.05077618e-01 2.97673702e-01 7.17693865e-01
-1.38131008e-01 7.97717795e-02 -1.09957075e+00 6.57943189e-01
-5.10233417e-02 -1.25900126e+00 -6.35878369e-02 2.54447162e-01
2.71504535e-03 3.58499616e-01 -4.66980129e-01 7.47651339e-01
9.28816140e-01 -9.55015063e-01 7.38807440e-01 -3.84044439e-01
6.43122077e-01 -6.92424297e-01 1.07247400e+00 3.25355917e-01
-1.31078231e+00 -2.22780779e-01 -4.02746826e-01 -4.74130839e-01
2.37073421e-01 1.21438764e-01 -8.23243618e-01 8.26966286e-01
6.83314085e-01 5.43372750e-01 -6.11377001e-01 4.27122921e-01
-6.28881931e-01 1.16567004e+00 -3.88435006e-01 -2.72653461e-01
5.04474640e-01 1.62166730e-02 4.05191183e-01 1.66578853e+00
3.84604111e-02 3.37498665e-01 9.96350199e-02 3.62093359e-01
1.37690261e-01 8.77200961e-02 -7.91381717e-01 -9.39111561e-02
5.41635215e-01 1.02096891e+00 -8.87207747e-01 -4.33357298e-01
-4.67375368e-01 4.09508348e-01 5.39507270e-01 -3.62571105e-02
-4.63752359e-01 -4.91794050e-01 2.95858145e-01 -8.33346322e-02
2.54569501e-01 -3.26965541e-01 -4.87813294e-01 -8.70489478e-01
-2.32451960e-01 -8.93829823e-01 7.72939503e-01 -7.20832169e-01
-1.21895647e+00 8.15660059e-01 3.69450569e-01 -4.97042835e-01
-8.07231069e-01 -1.11492491e+00 -4.92289573e-01 9.46200907e-01
-1.23431706e+00 -1.49752164e+00 2.66164690e-01 5.34378970e-03
5.38388610e-01 -3.80797610e-02 1.12920415e+00 -7.28447437e-02
-3.19358528e-01 5.75816870e-01 -3.56609404e-01 6.03857994e-01
6.83201328e-02 -1.80919874e+00 1.28754544e+00 1.06961238e+00
7.13676736e-02 7.53869891e-01 3.61455411e-01 -5.29105365e-01
-1.14941168e+00 -1.08120322e+00 1.57329023e+00 -9.31173742e-01
7.60268986e-01 -7.91217327e-01 -8.48477542e-01 1.30046678e+00
2.47526929e-01 -3.66586745e-01 4.42837238e-01 7.60309935e-01
-3.64925355e-01 2.57254571e-01 -9.66610134e-01 2.10508227e-01
1.38925469e+00 -3.37047696e-01 -1.11682343e+00 1.58526480e-01
1.16384256e+00 -6.03655994e-01 -8.70226741e-01 1.99151084e-01
4.05183822e-01 -7.31395602e-01 6.76859200e-01 -7.42991626e-01
3.09355497e-01 1.98196843e-01 -3.98493797e-01 -1.38804066e+00
-5.87770283e-01 -5.67471147e-01 4.26008672e-01 1.40596950e+00
5.30245185e-01 -1.04809523e+00 6.15216136e-01 5.63869953e-01
-8.53471100e-01 -4.30694401e-01 -1.06311786e+00 -9.96278405e-01
8.99034083e-01 -9.31259155e-01 6.17542028e-01 5.91420829e-01
2.91294813e-01 7.71857619e-01 1.79989725e-01 -1.52650088e-01
4.29264784e-01 -1.34414285e-01 3.99559259e-01 -1.40412331e+00
-3.02919477e-01 -2.02340782e-01 -5.40033281e-01 -9.94759858e-01
3.13005984e-01 -1.46372616e+00 2.32958451e-01 -1.83177423e+00
-1.35569349e-01 -6.91997409e-01 1.23640507e-01 1.12383962e+00
-8.76954347e-02 1.35245845e-01 5.99535525e-01 -1.48078367e-01
-1.78848729e-01 -2.43900582e-01 7.85213053e-01 1.88113302e-01
-7.78227374e-02 -2.84562141e-01 -1.35216630e+00 7.64873147e-01
9.87428486e-01 -6.78626001e-01 -1.55648559e-01 -8.17390621e-01
3.60987276e-01 1.86002180e-01 2.70287078e-02 -7.34899223e-01
-2.24584967e-01 3.33691418e-01 -1.37790861e-02 -5.28215766e-01
9.19458121e-02 -1.69102326e-01 -2.41590306e-01 2.73299247e-01
-1.86236829e-01 4.66792852e-01 6.14654779e-01 2.03881100e-01
-2.71832645e-01 -3.64843488e-01 4.15111959e-01 -6.59191847e-01
-8.51589739e-01 -2.68984228e-01 -4.89014715e-01 3.08237016e-01
3.39000762e-01 -2.43291974e-01 -6.36797309e-01 -1.44395575e-01
-9.81787920e-01 4.29410208e-03 -1.44728953e-02 4.91083264e-01
1.31873086e-01 -8.74356925e-01 -5.60993314e-01 6.49738386e-02
-1.30149141e-01 -5.52185299e-03 -3.34195316e-01 4.06371891e-01
-4.44687307e-01 9.21547115e-01 1.58574760e-01 -4.30373698e-01
-1.32589459e+00 2.90631473e-01 2.59059876e-01 -7.00155556e-01
-7.75291681e-01 1.12378561e+00 2.85322607e-01 -9.38878417e-01
-6.07634000e-02 -8.48845482e-01 -1.40240908e-01 -2.00330883e-01
2.82665849e-01 -1.77595034e-01 4.99690861e-01 -4.88378167e-01
-6.01640582e-01 5.25464058e-01 -2.86687835e-04 -1.97561964e-01
1.46355033e+00 -8.02436918e-02 -2.58014053e-01 4.80155110e-01
1.01246071e+00 -1.29625618e-01 -4.86278504e-01 2.05360472e-01
3.89923215e-01 1.90754324e-01 -6.58078585e-03 -1.22536850e+00
-6.82716548e-01 1.02777028e+00 -1.99176297e-02 5.76695979e-01
1.07196748e+00 5.50012946e-01 7.83078730e-01 3.50543320e-01
8.10471833e-01 -9.08297360e-01 -5.71933627e-01 1.06147218e+00
6.86826408e-01 -8.61384273e-01 -4.08611596e-01 -8.46428573e-01
-4.67536122e-01 1.27846920e+00 2.06507862e-01 -2.78777152e-01
4.85170037e-01 5.41233063e-01 2.40316033e-01 -3.72173458e-01
-9.44331348e-01 -1.94745764e-01 -3.04426759e-01 8.74305129e-01
8.59813452e-01 5.01072764e-01 -7.30275869e-01 8.11159194e-01
-9.10513878e-01 -2.40067124e-01 4.85134721e-01 1.01200056e+00
-5.67847371e-01 -1.71393466e+00 -3.11261594e-01 3.63428354e-01
-8.57022941e-01 -5.37863076e-01 -3.39358211e-01 1.39165235e+00
-6.05813451e-02 1.30371678e+00 4.98618111e-02 -3.11431885e-01
1.84410736e-01 5.49055040e-01 5.98547757e-01 -1.23122036e+00
-9.60433424e-01 1.29395694e-01 9.68465030e-01 -6.14363015e-01
-3.64845902e-01 -8.27782393e-01 -1.36161482e+00 -1.67890996e-01
-3.50410968e-01 4.36171710e-01 7.87945390e-01 1.05245054e+00
6.46764413e-02 4.90123779e-01 3.39058712e-02 -6.80565536e-01
-3.54526877e-01 -1.20784485e+00 -5.24749994e-01 -9.40139145e-02
-1.21146597e-01 -3.61295372e-01 -3.53122622e-01 -2.49936253e-01] | [10.457274436950684, 9.514153480529785] |
bb33486f-f043-41d5-b46e-e89a573a02fc | a-dataset-for-deep-learning-based-bone | 2306.04579 | null | https://arxiv.org/abs/2306.04579v1 | https://arxiv.org/pdf/2306.04579v1.pdf | A Dataset for Deep Learning-based Bone Structure Analyses in Total Hip Arthroplasty | Total hip arthroplasty (THA) is a widely used surgical procedure in orthopedics. For THA, it is of clinical significance to analyze the bone structure from the CT images, especially to observe the structure of the acetabulum and femoral head, before the surgical procedure. For such bone structure analyses, deep learning technologies are promising but require high-quality labeled data for the learning, while the data labeling is costly. We address this issue and propose an efficient data annotation pipeline for producing a deep learning-oriented dataset. Our pipeline consists of non-learning-based bone extraction (BE) and acetabulum and femoral head segmentation (AFS) and active-learning-based annotation refinement (AAR). For BE we use the classic graph-cut algorithm. For AFS we propose an improved algorithm, including femoral head boundary localization using first-order and second-order gradient regularization, line-based non-maximum suppression, and anatomy prior-based femoral head extraction. For AAR, we refine the algorithm-produced pseudo labels with the help of trained deep models: we measure the uncertainty based on the disagreement between the original pseudo labels and the deep model predictions, and then find out the samples with the largest uncertainty to ask for manual labeling. Using the proposed pipeline, we construct a large-scale bone structure analyses dataset from more than 300 clinical and diverse CT scans. We perform careful manual labeling for the test set of our data. We then benchmark multiple state-of-the art deep learning-based methods of medical image segmentation using the training and test sets of our data. The extensive experimental results validate the efficacy of the proposed data annotation pipeline. The dataset, related codes and models will be publicly available at https://github.com/hitachinsk/THA. | ['Xifu Shang', 'Dong Liu', 'Ziyang Gan', 'Kaidong Zhang'] | 2023-06-07 | null | null | null | null | ['active-learning', 'anatomy', 'active-learning'] | ['methodology', 'miscellaneous', 'natural-language-processing'] | [-1.04986414e-01 6.36895359e-01 -3.35320562e-01 -3.59137803e-01
-1.19806671e+00 -7.59121925e-02 -2.54184455e-02 2.25659043e-01
-5.61633110e-01 6.02741778e-01 1.38505384e-01 -1.80906802e-01
-1.01904839e-01 -6.49206161e-01 -6.12704873e-01 -9.05161679e-01
-1.38418674e-01 1.38181221e+00 7.36125052e-01 1.54965222e-01
1.78053498e-01 5.25453031e-01 -1.16475701e+00 1.66611269e-01
8.53834689e-01 9.99105930e-01 1.23880595e-01 2.22777516e-01
3.03950012e-02 5.41814804e-01 -2.66106337e-01 -2.57168207e-02
4.14584160e-01 -2.08692119e-01 -8.58050644e-01 3.23129922e-01
-3.11663628e-01 -2.47035861e-01 2.45494535e-03 8.39435339e-01
5.49866378e-01 -3.29312235e-01 7.85990596e-01 -7.97435582e-01
3.93668339e-02 5.57972848e-01 -8.71081471e-01 6.00545928e-02
-1.47521198e-01 3.10226738e-01 6.85532331e-01 -9.79369760e-01
7.52382934e-01 6.61130071e-01 5.53187132e-01 5.36988258e-01
-9.77053463e-01 -7.86964655e-01 -4.96941060e-01 5.90643510e-02
-1.46023989e+00 -2.96442837e-01 7.84350038e-01 -7.84137070e-01
3.47049624e-01 9.50772222e-03 1.06282091e+00 6.95573330e-01
6.76450193e-01 9.44149613e-01 9.38579023e-01 -3.98056060e-01
5.60525656e-01 -5.13442457e-01 2.39448100e-01 1.17379904e+00
3.67978334e-01 1.48509756e-01 -9.72518846e-02 -2.30163157e-01
8.90964150e-01 3.29992399e-02 -3.87230098e-01 -6.33006871e-01
-1.02247834e+00 7.86722362e-01 5.98418891e-01 -1.42855709e-02
-6.28669918e-01 -3.92395491e-03 3.47394049e-01 -3.82034630e-01
4.11346704e-01 2.96701908e-01 -3.98618668e-01 2.82507151e-01
-1.17572498e+00 5.91589734e-02 5.59026718e-01 6.47201419e-01
7.15405226e-01 -4.48659807e-01 4.80560027e-02 8.88352990e-01
7.33045995e-01 7.96612799e-02 6.99506223e-01 -8.91827881e-01
-1.23879470e-01 8.02624464e-01 -3.01449090e-01 -4.65135902e-01
-6.08889937e-01 -4.02443588e-01 -8.31436932e-01 4.44499254e-01
4.60313320e-01 -1.32708162e-01 -1.51290739e+00 9.22031879e-01
5.79084158e-01 -9.14738886e-03 -5.69646299e-01 1.07967031e+00
8.25074911e-01 7.89343864e-02 -8.73723179e-02 -4.09736276e-01
1.49788594e+00 -8.74617219e-01 -5.59493542e-01 -1.88896880e-01
9.77159679e-01 -5.98666310e-01 9.28471863e-01 4.73449320e-01
-9.16351557e-01 -1.50505558e-01 -1.03339243e+00 -9.53265876e-02
2.07749233e-01 1.61735520e-01 3.05215210e-01 2.86159962e-01
-7.61550188e-01 3.24439585e-01 -1.55360341e+00 1.54978573e-01
8.43821883e-01 7.46590674e-01 -3.87641251e-01 5.86699396e-02
-8.84100914e-01 5.97378969e-01 4.10768956e-01 1.79234177e-01
-9.76131678e-01 -5.87402284e-01 -8.88184249e-01 -1.78955212e-01
6.03100657e-01 -7.25981891e-01 1.29418182e+00 -6.84175968e-01
-1.50166285e+00 1.19450760e+00 1.09956123e-01 -3.54641050e-01
4.87957478e-01 -3.54737014e-01 2.50693291e-01 1.33706868e-01
2.09490448e-01 5.84317982e-01 5.26095152e-01 -1.26638222e+00
-2.85734713e-01 -6.59118176e-01 -6.52905762e-01 -7.18473941e-02
1.88937798e-01 -1.82236418e-01 -7.31510043e-01 -5.56336284e-01
5.68030119e-01 -1.24668622e+00 -6.80379570e-01 1.33849755e-01
-7.16299772e-01 -7.97553211e-02 5.16011834e-01 -8.83844852e-01
1.13044429e+00 -2.16828299e+00 -6.78684935e-02 4.96364266e-01
5.67695320e-01 -1.11637719e-01 5.21112204e-01 -3.29210430e-01
5.47616594e-02 -5.90874478e-02 -9.08669770e-01 -3.31937402e-01
-5.98895490e-01 2.58493364e-01 4.68455732e-01 6.30613565e-01
-3.92944477e-02 8.72433960e-01 -6.12357736e-01 -1.14362299e+00
2.74215657e-02 1.86915010e-01 -7.25014150e-01 2.47266859e-01
-2.05366775e-01 7.96847939e-01 -4.23322439e-01 7.56412983e-01
3.86462897e-01 -5.48704863e-01 1.31710336e-01 -2.25862324e-01
2.89106071e-01 1.38852403e-01 -9.66268182e-01 1.92736208e+00
-2.20693782e-01 1.22052848e-01 1.35786459e-01 -8.73738050e-01
9.99428630e-01 4.13136482e-01 1.08970857e+00 -4.80715841e-01
3.65380406e-01 6.82754815e-01 4.74163502e-01 -5.77971220e-01
-1.25481248e-01 -1.53810889e-01 8.52223858e-02 4.35965002e-01
7.74748027e-02 -2.92537689e-01 1.10809272e-02 1.60459623e-01
1.07256556e+00 1.84772879e-01 5.81441522e-01 -4.90837723e-01
4.16734099e-01 2.71591932e-01 7.78043389e-01 1.27609968e-01
-2.83280611e-01 9.50364113e-01 3.54800403e-01 -3.93353820e-01
-9.55472827e-01 -9.94730175e-01 -3.04501951e-01 3.38805020e-01
-2.26741377e-02 -2.58184701e-01 -9.33781743e-01 -8.94819677e-01
-1.55805856e-01 3.15495789e-01 -6.73241675e-01 -1.94049463e-01
-9.34117198e-01 -1.00848436e+00 1.06900707e-01 4.63561177e-01
1.83788329e-01 -1.03746188e+00 -6.82445943e-01 2.55009890e-01
-6.75085038e-02 -6.63703859e-01 -1.98665351e-01 5.52744508e-01
-1.22743070e+00 -1.27060127e+00 -7.93638706e-01 -8.45160067e-01
8.43078613e-01 -5.22862494e-01 1.01751351e+00 3.40981722e-01
-5.43194473e-01 -3.22399102e-02 1.29936431e-02 -2.62553662e-01
-2.95969516e-01 2.38890201e-01 -3.23381960e-01 -2.80937165e-01
4.40233722e-02 -2.69107908e-01 -9.27915275e-01 3.84077996e-01
-6.80761755e-01 2.02318192e-01 9.10677671e-01 8.17189038e-01
1.23997819e+00 -1.05638772e-01 2.35658467e-01 -1.24687517e+00
2.85251737e-01 -5.60687184e-01 -4.97395933e-01 8.08343887e-02
-6.73357010e-01 1.96970701e-01 1.96168602e-01 -3.53345610e-02
-1.05959868e+00 5.62425673e-01 -3.95649463e-01 -1.05209507e-01
-3.03098559e-01 6.59352779e-01 -8.69820565e-02 1.44481093e-01
6.90246344e-01 -2.06812814e-01 2.37552822e-01 -4.11425650e-01
1.56707779e-01 6.44022822e-01 5.84885240e-01 -6.41955018e-01
3.39643240e-01 5.89599371e-01 8.11428428e-02 -4.00360376e-01
-7.89895117e-01 -5.34541249e-01 -9.69566941e-01 -4.21837628e-01
1.03444731e+00 -6.23066664e-01 -1.46668285e-01 3.90047669e-01
-9.08468425e-01 -4.66959804e-01 -3.72043163e-01 7.41930425e-01
-6.45239770e-01 4.30225402e-01 -8.56081486e-01 -4.36746866e-01
-7.16209114e-01 -1.83474457e+00 1.27527511e+00 3.52033824e-02
-5.01347363e-01 -6.85142875e-01 3.00547898e-01 6.50330842e-01
-1.12714283e-01 2.36010313e-01 1.05249870e+00 -9.51532185e-01
-5.80901086e-01 -1.89111173e-01 1.47339076e-01 1.08054765e-01
1.86162397e-01 9.15463120e-02 -6.47814929e-01 9.69865322e-02
3.48990947e-01 -2.40051940e-01 7.60384023e-01 9.05714214e-01
1.40466547e+00 1.68279514e-01 -6.33349836e-01 7.65140414e-01
1.26397467e+00 3.62988114e-01 6.91518784e-01 5.29405355e-01
7.28667796e-01 5.13199806e-01 7.52871692e-01 3.82128477e-01
1.41230598e-01 5.62300682e-01 5.28326392e-01 -1.84649631e-01
-2.18735635e-01 1.96981370e-01 -2.57072877e-02 1.09157336e+00
-1.69974536e-01 1.86486408e-01 -1.57655799e+00 5.17196298e-01
-1.54605365e+00 -2.46194433e-02 -4.24771816e-01 2.16598225e+00
1.09194112e+00 5.10208905e-01 -7.66407251e-02 3.82317156e-01
5.74654579e-01 -3.82492185e-01 -5.22903740e-01 3.08034658e-01
4.96700048e-01 6.83599114e-01 5.18668890e-01 6.20649576e-01
-9.66563880e-01 7.33285308e-01 5.28376198e+00 8.53158295e-01
-1.21352100e+00 1.71930969e-01 8.84950936e-01 -7.20595345e-02
1.71763171e-02 5.16094938e-02 -5.62778950e-01 3.90848935e-01
5.99632680e-01 1.96690455e-01 -3.03009480e-01 9.52460229e-01
2.35629469e-01 -5.37133694e-01 -1.10348570e+00 7.11302638e-01
-2.54930526e-01 -1.24886131e+00 -2.39426002e-01 2.08135843e-01
4.05371577e-01 1.11719064e-01 -4.76209551e-01 -4.29127514e-02
3.19977015e-01 -8.70342731e-01 5.22246480e-01 4.41741437e-01
5.91765583e-01 -4.02202666e-01 9.99501407e-01 2.38072932e-01
-1.09009755e+00 1.20421432e-01 7.18618631e-02 3.17823619e-01
3.40244830e-01 1.00315356e+00 -1.03839529e+00 5.00616193e-01
6.30042434e-01 4.10300225e-01 -5.59048593e-01 1.46893764e+00
-3.27680290e-01 7.23286748e-01 -4.59860414e-01 4.08940971e-01
-2.74792343e-01 -2.12793916e-01 3.34999055e-01 6.20996058e-01
1.47365347e-01 3.13984722e-01 3.09203893e-01 7.25900650e-01
-7.97373578e-02 3.18371922e-01 -3.82131450e-02 4.09545064e-01
4.41489041e-01 1.11509681e+00 -1.35297453e+00 -3.10385406e-01
-2.09739476e-01 6.26787663e-01 2.89787412e-01 -8.23021457e-02
-6.51462317e-01 2.39594355e-01 -1.91519968e-02 5.51620603e-01
-2.76030809e-01 -1.60753623e-01 -8.41949046e-01 -7.62645602e-01
-1.23192957e-02 -6.27343833e-01 6.77659392e-01 -6.08218491e-01
-9.82559860e-01 4.52712029e-01 3.53199914e-02 -1.02435136e+00
-2.36024469e-01 -5.77456951e-01 -4.47590739e-01 4.37279105e-01
-9.68615830e-01 -8.15855086e-01 -4.38081443e-01 3.54205072e-01
6.39041543e-01 5.70846982e-02 5.10632455e-01 3.72891039e-01
-6.42310977e-01 2.14393049e-01 -3.00156265e-01 5.39424717e-01
5.88697851e-01 -1.08897960e+00 9.80058610e-02 5.28951049e-01
-1.24546185e-01 3.83316129e-01 4.40371513e-01 -1.16303027e+00
-9.06083465e-01 -8.66995454e-01 2.89457679e-01 -1.18015952e-01
5.39057255e-01 -1.41985357e-01 -9.59869623e-01 8.37089062e-01
-2.81142652e-01 3.05403113e-01 7.01998055e-01 -3.34000707e-01
4.42502171e-01 1.72704265e-01 -1.02826285e+00 2.41168246e-01
8.23937595e-01 6.68417141e-02 -6.52525783e-01 3.14710081e-01
4.59325045e-01 -6.44509792e-01 -1.04846656e+00 1.00955582e+00
4.94622350e-01 -8.16148400e-01 8.36430967e-01 -1.56079009e-01
6.22062802e-01 -1.68659419e-01 2.56767213e-01 -8.40524971e-01
-2.11011827e-01 -1.00373127e-01 -2.22402997e-02 7.15836287e-01
5.24543166e-01 -3.45266491e-01 1.33884907e+00 4.73765612e-01
-7.67025530e-01 -1.44822693e+00 -9.39096451e-01 -3.91380072e-01
2.67683029e-01 -3.22663784e-01 1.74426392e-01 7.99848139e-01
-2.39425421e-01 -3.30925849e-03 1.99712440e-01 8.18220228e-02
7.07604349e-01 -1.31285846e-01 6.47106051e-01 -1.57652962e+00
-2.21391544e-01 -2.87765771e-01 -4.68430489e-01 -7.18738258e-01
-5.09365201e-02 -8.95320117e-01 3.64967108e-01 -1.82775068e+00
2.89431959e-01 -6.01766884e-01 -1.99885190e-01 4.56814080e-01
-1.11322559e-01 2.13788763e-01 -5.29094338e-01 7.48616993e-01
-1.17640041e-01 5.50059736e-01 1.48518014e+00 -1.43926730e-02
-2.57352531e-01 2.98840106e-02 -3.83707955e-02 1.00933099e+00
6.94860756e-01 -7.76457191e-01 -1.92371041e-01 -1.20519377e-01
-6.07485324e-02 9.80739444e-02 2.92176843e-01 -1.11933780e+00
2.42020413e-01 3.24724317e-02 4.18719530e-01 -6.37433946e-01
4.39645983e-02 -9.49925661e-01 2.16638312e-01 1.07212627e+00
-1.18691079e-01 -3.18222076e-01 -1.46181092e-01 2.12537408e-01
-2.83715725e-01 -2.01441392e-01 1.08210039e+00 -4.43926573e-01
-4.39288199e-01 5.90298653e-01 -3.85686338e-01 1.60948038e-01
1.05714703e+00 -2.16880798e-01 5.17884046e-02 9.38753486e-02
-1.23083508e+00 9.20333192e-02 4.70138639e-01 -1.86933130e-02
7.23756671e-01 -1.08510518e+00 -6.88331068e-01 1.44109815e-01
9.93241891e-02 9.45852995e-01 1.17096357e-01 1.32342279e+00
-1.01297629e+00 6.25814199e-02 -1.71141356e-01 -1.18695891e+00
-1.05690753e+00 3.16938668e-01 6.40549660e-01 -4.62987870e-01
-8.31260204e-01 8.95261109e-01 4.18685615e-01 -4.21561509e-01
6.10374808e-02 -4.95650142e-01 -2.18794551e-02 -1.93131998e-01
-1.60894826e-01 1.86715633e-01 4.72427845e-01 -4.96011168e-01
-4.56386238e-01 5.33978581e-01 -1.86185777e-01 -5.45528866e-02
1.24808097e+00 2.04945430e-01 -2.45572284e-01 2.56776601e-01
9.64576542e-01 -9.82194543e-02 -9.88459289e-01 -2.07167313e-01
3.72994155e-01 -1.43660501e-01 3.10476005e-01 -5.80847383e-01
-1.55795062e+00 6.54429197e-01 9.09208417e-01 -4.67309177e-01
9.62634206e-01 3.65607738e-01 9.12667632e-01 -7.70372599e-02
5.17901659e-01 -9.49093699e-01 6.01674523e-03 2.95669027e-02
7.53520429e-01 -1.30901766e+00 3.98564607e-01 -9.32794094e-01
-3.13299805e-01 1.07075191e+00 7.13020146e-01 -1.60324484e-01
1.14951241e+00 6.91541612e-01 3.35698396e-01 -7.48645127e-01
-3.16335499e-01 6.08931072e-02 3.50114703e-01 2.68746912e-01
6.86660528e-01 1.65873751e-01 -6.65987551e-01 7.87330449e-01
-4.02358532e-01 2.43273333e-01 2.65558392e-01 1.18082404e+00
-5.95463514e-01 -9.75778162e-01 -4.71467495e-01 6.53497875e-01
-4.91250426e-01 7.16845393e-02 -4.15729970e-01 9.27416027e-01
7.63203800e-02 4.30837482e-01 8.86481777e-02 -1.22833841e-01
-6.96113333e-02 5.88434190e-02 3.97930026e-01 -9.59729075e-01
-3.73188317e-01 5.52952886e-01 -4.60069329e-02 -4.89501923e-01
-2.44609997e-01 -7.04340816e-01 -1.94033110e+00 4.21964854e-01
-5.54080546e-01 1.15043558e-01 4.12401021e-01 1.26177788e+00
4.66672108e-02 5.73241532e-01 3.62690836e-01 -7.60838389e-01
-1.31160691e-01 -9.19301689e-01 -4.98188823e-01 4.47531968e-01
-3.69760320e-02 -1.00765312e+00 -2.87748635e-01 1.98845625e-01] | [14.367851257324219, -2.195810556411743] |
895593ea-345c-4098-9daf-5cb2843075a3 | uncovering-the-local-hidden-community | 2112.04100 | null | https://arxiv.org/abs/2112.04100v1 | https://arxiv.org/pdf/2112.04100v1.pdf | Uncovering the Local Hidden Community Structure in Social Networks | Hidden community is a useful concept proposed recently for social network analysis. To handle the rapid growth of network scale, in this work, we explore the detection of hidden communities from the local perspective, and propose a new method that detects and boosts each layer iteratively on a subgraph sampled from the original network. We first expand the seed set from a single seed node based on our modified local spectral method and detect an initial dominant local community. Then we temporarily remove the members of this community as well as their connections to other nodes, and detect all the neighborhood communities in the remaining subgraph, including some "broken communities" that only contain a fraction of members in the original network. The local community and neighborhood communities form a dominant layer, and by reducing the edge weights inside these communities, we weaken this layer's structure to reveal the hidden layers. Eventually, we repeat the whole process and all communities containing the seed node can be detected and boosted iteratively. We theoretically show that our method can avoid some situations that a broken community and the local community are regarded as one community in the subgraph, leading to the inaccuracy on detection which can be caused by global hidden community detection methods. Extensive experiments show that our method could significantly outperform the state-of-the-art baselines designed for either global hidden community detection or multiple local community detection. | ['John E. Hopcroft', 'Kun He', 'Boyu Li', 'Meng Wang'] | 2021-12-08 | null | null | null | null | ['local-community-detection'] | ['graphs'] | [ 9.32652950e-02 4.09396142e-01 -1.82981089e-01 4.38008398e-01
-8.46356899e-02 -6.70397997e-01 3.21337223e-01 3.21469188e-01
-1.93923600e-02 4.39490050e-01 6.00187592e-02 1.23116402e-02
1.46799818e-01 -1.14419067e+00 -2.83056051e-01 -9.37945843e-01
-5.68238437e-01 4.51926470e-01 1.07610369e+00 -4.19054404e-02
1.39746545e-02 2.14062244e-01 -9.37751949e-01 2.45931298e-01
6.94898546e-01 7.79029652e-02 2.40556374e-01 3.64294291e-01
-1.73973054e-01 6.02992058e-01 -5.90679288e-01 8.17551911e-02
1.67259216e-01 -5.74232221e-01 -7.68490613e-01 5.81631482e-01
-3.19459774e-02 -1.67030599e-02 -2.94580549e-01 1.33188152e+00
2.68132269e-01 -4.76259410e-01 1.75720587e-01 -1.36541259e+00
-2.03146443e-01 1.17776346e+00 -1.39616787e+00 1.52518883e-01
3.12458754e-01 -1.51489526e-01 1.18431890e+00 -9.03878510e-01
1.03121650e+00 1.29566598e+00 7.43321598e-01 2.92582840e-01
-1.46028578e+00 -1.00033426e+00 6.02612078e-01 -1.98973790e-01
-1.72259641e+00 -1.02436416e-01 9.00877893e-01 -5.47962010e-01
2.67143130e-01 1.98305145e-01 9.23255920e-01 6.45236790e-01
-3.54027331e-01 5.74934244e-01 9.68883634e-01 -4.70632523e-01
7.12105110e-02 2.55300514e-02 3.22690398e-01 9.85339284e-01
8.09231460e-01 -3.82604480e-01 -3.17870855e-01 -5.19200265e-01
6.78911150e-01 1.43531621e-01 -4.36485440e-01 -6.85642362e-01
-1.42190015e+00 1.17093492e+00 9.21385586e-01 6.91282392e-01
-6.89734444e-02 1.02718240e-02 1.51942343e-01 3.84996980e-01
4.50103074e-01 -8.38816166e-02 2.62469593e-02 8.28123510e-01
-1.21840191e+00 -2.56006569e-01 9.09706533e-01 8.49619448e-01
1.19983697e+00 -5.72794497e-01 1.20648421e-01 4.51269835e-01
4.63135660e-01 1.01314239e-01 -1.94709748e-01 -5.23109376e-01
4.28050280e-01 1.37510920e+00 -2.65200615e-01 -1.45649672e+00
-3.76882225e-01 -7.31584847e-01 -1.34081447e+00 -6.45428896e-02
2.31686100e-01 -4.44934890e-02 -6.94075048e-01 1.44853735e+00
6.11622036e-01 3.36466759e-01 -2.52000540e-01 6.77674711e-01
6.60916150e-01 3.66178334e-01 -5.79350352e-01 -5.71084738e-01
1.18730950e+00 -1.04288852e+00 -3.79235923e-01 -1.85530826e-01
4.41158354e-01 -6.39149368e-01 3.79998982e-01 -9.89773422e-02
-8.37220848e-01 -1.94544733e-01 -1.15950406e+00 5.92859924e-01
-5.79412654e-02 -1.58302426e-01 2.75300264e-01 3.29290926e-01
-1.30047846e+00 4.32213128e-01 -7.01963603e-01 -5.79955935e-01
1.89218283e-01 1.60618886e-01 -1.87188700e-01 -2.73855299e-01
-8.97240937e-01 1.12288646e-01 5.44945896e-01 1.04844384e-01
-1.09414792e+00 -8.35033730e-02 -6.18233919e-01 1.41176999e-01
9.02795613e-01 -5.27986884e-01 4.92166519e-01 -9.48441625e-01
-3.93527627e-01 8.76891673e-01 -3.59118164e-01 -1.47993088e-01
4.88457650e-01 6.26918018e-01 -3.87507528e-01 4.01711345e-01
5.58626652e-01 2.83443570e-01 7.05590725e-01 -1.45395970e+00
-4.81446207e-01 -3.59383337e-02 -3.49376816e-03 -8.55095088e-02
-5.09575486e-01 1.68968707e-01 -6.63583338e-01 -3.84881049e-01
9.01957452e-01 -1.08906329e+00 -4.00401682e-01 -8.72774571e-02
-8.70695233e-01 -9.30516869e-02 1.00834823e+00 -2.56903470e-01
1.62929070e+00 -2.12851787e+00 1.50984049e-01 9.15798664e-01
1.37209570e+00 -6.45860732e-02 -2.40246773e-01 6.72392547e-01
-1.09205455e-01 6.23505592e-01 -3.15209389e-01 -3.50036174e-01
-5.18634677e-01 1.17255218e-01 1.09000638e-01 6.76947832e-01
-7.96208978e-02 5.80610275e-01 -1.29117501e+00 -6.66598499e-01
-4.19819832e-01 1.61651745e-01 -5.34162402e-01 -2.87904888e-01
2.48367369e-01 2.58375734e-01 -2.90525019e-01 6.30487204e-01
8.58803093e-01 -9.04846787e-01 9.41850245e-01 2.43296236e-01
-2.15930134e-01 1.84119850e-01 -1.74036980e+00 8.30164433e-01
4.71630216e-01 4.82560515e-01 7.58953452e-01 -8.28086853e-01
8.98245215e-01 3.84386718e-01 5.13913155e-01 2.93375552e-01
-2.46831670e-01 1.36854500e-01 3.43760878e-01 -1.72146019e-02
1.04669422e-01 1.65181056e-01 2.08453074e-01 8.23510826e-01
-3.46363008e-01 7.05182254e-01 4.20622081e-01 8.98239970e-01
1.34915555e+00 -7.10522294e-01 3.22784007e-01 -2.93877810e-01
5.26344836e-01 -4.14904468e-02 6.80000484e-01 8.40123296e-01
-3.47045600e-01 5.35592556e-01 8.98500025e-01 1.99408308e-02
-9.39825714e-01 -9.12688553e-01 4.28470194e-01 1.00648737e+00
2.75330722e-01 -7.80841947e-01 -7.66417265e-01 -5.99705994e-01
9.47340056e-02 -5.88500082e-01 -6.82557940e-01 -7.12418258e-02
-4.82679278e-01 -7.74474680e-01 1.33278653e-01 1.13336317e-01
4.25547421e-01 -9.34076786e-01 1.39607653e-01 3.72193068e-01
-6.35576427e-01 -8.51899564e-01 -7.94676125e-01 -9.10219029e-02
-9.43949163e-01 -1.51138544e+00 -6.97973311e-01 -1.12102735e+00
1.15216184e+00 1.08684456e+00 8.62780035e-01 1.01130712e+00
5.17192744e-02 -2.26325750e-01 -4.54823822e-01 3.36860180e-01
-6.49878740e-01 3.03837240e-01 1.05207544e-02 4.02496815e-01
4.87879850e-02 -6.83505476e-01 -4.35493767e-01 4.30129856e-01
-4.14401680e-01 3.72501612e-02 4.87568229e-01 4.98304635e-01
1.97797209e-01 6.58964694e-01 3.17958444e-01 -8.95129681e-01
6.12751842e-01 -8.41283977e-01 -5.93570769e-01 2.05452800e-01
-5.85487366e-01 -1.93430662e-01 -1.61213763e-02 -5.98853230e-01
-4.22650069e-01 -7.93394670e-02 8.28242183e-01 -3.37169319e-01
3.92097622e-01 1.01664364e+00 -1.12367555e-01 -1.46420732e-01
4.90160286e-01 1.19695768e-01 3.44388671e-02 -6.28875077e-01
-7.78718963e-02 4.18743759e-01 2.12082595e-01 -7.65936896e-02
1.50135887e+00 8.09303045e-01 3.77597054e-03 -6.85654998e-01
-5.19074500e-01 -1.03997350e+00 -1.06089127e+00 -3.15489084e-01
3.91342849e-01 -1.18032873e+00 -4.94920880e-01 3.53478730e-01
-1.21150458e+00 5.40940687e-02 2.79967815e-01 9.49870273e-02
5.99477112e-01 8.39608669e-01 -9.40937400e-01 -9.32906330e-01
-8.52224603e-02 -5.87022305e-01 5.86618841e-01 -1.30384088e-01
-1.00824364e-01 -9.65626001e-01 2.83810914e-01 -3.39964293e-02
1.18207052e-01 4.64894861e-01 6.55913293e-01 -4.15802777e-01
-1.02386749e+00 -2.32796878e-01 -3.85903448e-01 -1.65138274e-01
3.08303267e-01 3.76556695e-01 -5.69221258e-01 -8.57764840e-01
-3.95239204e-01 3.25670481e-01 1.11782897e+00 8.81700516e-02
8.87606218e-02 -5.03207564e-01 -9.67345417e-01 1.79631352e-01
1.37201762e+00 -3.12631041e-01 3.83835077e-01 1.83472887e-01
9.19227481e-01 6.08368039e-01 -1.54222116e-01 2.32422069e-01
1.85385048e-01 1.09041594e-01 4.48742777e-01 -4.95605141e-01
1.00032249e-02 -3.00619185e-01 4.66939151e-01 1.33816993e+00
-2.90215403e-01 -4.56012189e-02 -9.41452980e-01 7.62508571e-01
-2.03713417e+00 -1.16428030e+00 -7.77847052e-01 2.15036988e+00
9.24725235e-01 5.22945344e-01 5.99956930e-01 3.29087794e-01
1.69797921e+00 1.60546958e-01 -9.83116999e-02 3.66917700e-01
-3.72808903e-01 -2.16074914e-01 2.98392415e-01 5.72401762e-01
-8.77127290e-01 8.63520086e-01 6.64239979e+00 3.64502460e-01
-6.39250755e-01 2.45334163e-01 2.46497065e-01 2.40543783e-02
-3.42960000e-01 5.90104699e-01 -8.34182560e-01 2.26455629e-01
2.73917258e-01 -2.94906974e-01 4.67651844e-01 7.12189496e-01
2.15860650e-01 1.66639909e-02 -6.88338578e-01 2.83597291e-01
-1.46159336e-01 -1.28142166e+00 -1.08431041e-01 7.03958035e-01
1.13775361e+00 2.10931942e-01 -6.39971673e-01 -8.54908582e-03
5.59468865e-01 -6.43978655e-01 4.82635260e-01 -1.76688582e-02
5.32894373e-01 -6.41825199e-01 5.22595704e-01 7.67830908e-01
-2.00674057e+00 -8.91243368e-02 -3.80012184e-01 -2.52452403e-01
-1.51497990e-01 1.00719416e+00 -9.16569650e-01 3.48890811e-01
6.80160403e-01 7.55891383e-01 -6.03147626e-01 1.19896090e+00
-2.03839853e-01 6.84410453e-01 -4.37265307e-01 9.55897048e-02
2.02727139e-01 -3.36964875e-01 9.66245055e-01 1.05021787e+00
9.74714905e-02 -7.82284215e-02 8.01243424e-01 1.06130075e+00
-2.36368150e-01 -7.16326311e-02 -6.11531973e-01 -7.17280358e-02
6.90854013e-01 1.39989686e+00 -1.47845578e+00 -5.43058813e-01
-5.45452893e-01 8.47602367e-01 4.82214391e-01 4.95771438e-01
-5.17580509e-01 -2.55693287e-01 1.94493845e-01 8.00435245e-01
5.23843348e-01 -2.53635794e-01 3.14198911e-01 -1.32586718e+00
3.31620798e-02 -8.03711534e-01 5.83243132e-01 -3.65095079e-01
-1.17580354e+00 4.16607320e-01 -1.18693814e-01 -1.19535470e+00
3.16011727e-01 1.09426521e-01 -1.06279266e+00 7.16988087e-01
-1.19512975e+00 -9.94569898e-01 -5.81401825e-01 7.37097621e-01
-1.33392364e-01 1.19579472e-01 4.59384173e-01 2.99817115e-01
-6.24092937e-01 2.54966527e-01 4.82092425e-02 6.55582905e-01
4.34489191e-01 -9.79427516e-01 3.03112000e-01 1.25935423e+00
-1.46400228e-01 1.09839034e+00 2.73948282e-01 -1.15522242e+00
-5.82898557e-01 -9.67460215e-01 1.24341869e+00 1.37994634e-02
1.03053188e+00 -7.91756332e-01 -1.13670576e+00 8.18628490e-01
2.01276273e-01 -3.93165313e-02 2.82162637e-01 2.76470959e-01
-5.08755624e-01 1.88021407e-01 -8.28794658e-01 6.49348378e-01
1.33216333e+00 -5.51691115e-01 -1.87945664e-01 3.32790107e-01
6.86684668e-01 3.02136332e-01 -4.29126352e-01 2.00979695e-01
2.40269169e-01 -9.34169471e-01 7.97879577e-01 -1.20892070e-01
1.42813437e-02 -7.90234864e-01 4.87483263e-01 -1.03292334e+00
-9.49631631e-01 -9.52712595e-01 -8.22066665e-02 1.31394732e+00
2.81670839e-01 -7.44256437e-01 9.22198892e-01 -4.53201264e-01
3.29726875e-01 -1.00978762e-01 -9.50162709e-01 -7.67202735e-01
-2.64302582e-01 4.27870631e-01 4.08016384e-01 1.44544995e+00
3.06792438e-01 6.97500706e-01 -3.05635810e-01 4.32142884e-01
1.12166774e+00 4.11871076e-01 7.16832817e-01 -1.92778361e+00
-1.58129528e-01 -5.49769044e-01 -1.61510989e-01 -8.07612062e-01
-2.67447144e-01 -9.92643654e-01 -1.64062247e-01 -1.52246273e+00
1.04188287e+00 -3.40632737e-01 -2.98155278e-01 6.77165270e-01
-2.90720612e-01 3.67209613e-01 7.64625221e-02 7.11056709e-01
-7.29093671e-01 -1.11717537e-01 1.35380244e+00 -3.20142299e-01
-4.22233850e-01 -4.36494909e-02 -7.62192488e-01 7.67018080e-01
4.50815201e-01 -9.23580348e-01 -1.08882874e-01 2.83919901e-01
3.19263697e-01 -5.21585271e-02 4.16788101e-01 -8.42733502e-01
4.77504402e-01 -3.04347221e-02 1.13965355e-01 -7.29883850e-01
-2.26943910e-01 -6.76688612e-01 3.10189962e-01 1.09416592e+00
8.23021084e-02 -8.39718133e-02 -3.40490192e-01 9.07922208e-01
-4.22119647e-02 -1.06359152e-02 7.91311979e-01 -3.23259920e-01
-2.16809168e-01 2.37958834e-01 -5.70538521e-01 -2.36139804e-01
1.08360839e+00 -5.26999295e-01 -5.07204056e-01 -3.53610128e-01
-1.09119308e+00 6.28328145e-01 7.42351234e-01 1.00192711e-01
4.88147050e-01 -1.25257170e+00 -1.11436093e+00 1.45761907e-01
2.94226438e-01 -2.86525059e-02 -2.62497395e-01 1.21887302e+00
-4.61669981e-01 -7.32439756e-02 2.01933712e-01 -6.91554964e-01
-1.81940377e+00 8.26370478e-01 3.90439034e-01 -5.67083061e-01
-7.75466383e-01 6.23211026e-01 3.37286919e-01 -1.79735452e-01
2.34646499e-01 -2.11023629e-01 -2.94329405e-01 2.78325915e-01
4.59707499e-01 3.57911348e-01 -5.54459155e-01 -7.02901542e-01
-6.73853040e-01 4.80415076e-01 -3.88805978e-02 8.12895298e-02
1.04419625e+00 -3.65287572e-01 -8.37472200e-01 3.00922722e-01
1.14850211e+00 4.94082302e-01 -6.92836165e-01 -5.25712729e-01
2.35536113e-01 -2.68028170e-01 -1.97918311e-01 -9.38860402e-02
-1.40859449e+00 5.08038640e-01 9.85741988e-02 7.56264210e-01
8.81926358e-01 4.53234881e-01 4.24949855e-01 1.24456033e-01
3.55865180e-01 -7.93397248e-01 4.25205946e-01 4.71110433e-01
4.15899724e-01 -8.87302577e-01 3.61467332e-01 -1.11522710e+00
-7.78064951e-02 9.11055088e-01 3.28327209e-01 -1.67303324e-01
8.07993948e-01 8.46682265e-02 -3.97231430e-01 -5.45574963e-01
-6.35474443e-01 -4.33279216e-01 2.97332555e-02 4.61830288e-01
6.32453412e-02 1.82710439e-01 -1.34413138e-01 2.32865512e-01
2.87584037e-01 -2.97027707e-01 8.63082290e-01 9.09501374e-01
-9.79454756e-01 -1.08290899e+00 -6.28998816e-01 8.33217949e-02
-5.36568947e-02 -3.08458030e-01 -1.09316993e+00 8.03602636e-01
2.46115163e-01 1.08832014e+00 7.81060532e-02 -4.80168670e-01
-9.43709686e-02 -1.12085611e-01 6.35061413e-02 -1.13376522e+00
-5.28531790e-01 8.37279618e-01 1.40211597e-01 -1.60409108e-01
-5.61001241e-01 -5.18593967e-01 -1.18537903e+00 -8.53217125e-01
-9.34198797e-01 3.21254075e-01 -2.01504618e-01 5.24161875e-01
1.76361218e-01 3.55092376e-01 8.71797562e-01 -5.29537439e-01
-2.85669379e-02 -9.86537993e-01 -9.22689736e-01 1.50691122e-01
6.70780718e-01 -4.70672250e-01 -9.74947989e-01 -6.34901896e-02] | [6.95444393157959, 5.230868816375732] |
ed01d1e4-5c3f-40b1-ae16-5abf293933ab | optimizing-scoring-function-of-dynamic | 1708.09097 | null | http://arxiv.org/abs/1708.09097v2 | http://arxiv.org/pdf/1708.09097v2.pdf | Optimizing scoring function of dynamic programming of pairwise profile alignment using derivative free neural network | A profile comparison method with position-specific scoring matrix (PSSM) is
one of the most accurate alignment methods. Currently, cosine similarity and
correlation coefficient are used as scoring functions of dynamic programming to
calculate similarity between PSSMs. However, it is unclear that these functions
are optimal for profile alignment methods. At least, by definition, these
functions cannot capture non-linear relationships between profiles. Therefore,
in this study, we attempted to discover a novel scoring function, which was
more suitable for the profile comparison method than the existing ones. Firstly
we implemented a new derivative free neural network by combining the
conventional neural network with evolutionary strategy optimization method.
Next, using the framework, the scoring function was optimized for aligning
remote sequence pairs. Nepal, the pairwise profile aligner with the novel
scoring function significantly improved both alignment sensitivity and
precision, compared to aligners with the existing functions. Nepal improved
alignment quality because of adaptation to remote sequence alignment and
increasing the expressive power of similarity score. The novel scoring function
can be realized using a simple matrix operation and easily incorporated into
other aligners. With our scoring function, the performance of homology
detection and/or multiple sequence alignment for remote homologous sequences
would be further improved. | ['Kazunori D Yamada'] | 2017-08-30 | null | null | null | null | ['multiple-sequence-alignment'] | ['medical'] | [ 6.11759722e-01 -4.87355798e-01 -1.43631682e-01 -3.95136923e-01
-3.08298796e-01 -6.01976931e-01 -4.13218178e-02 3.80404770e-01
-4.99671787e-01 8.72561753e-01 -1.13437667e-01 -2.87803620e-01
-5.22015870e-01 -6.76942647e-01 -2.18212828e-01 -1.01579034e+00
-2.12289661e-01 4.73533034e-01 4.42906320e-01 -6.30215466e-01
6.59904063e-01 6.53651595e-01 -1.58870769e+00 3.22920978e-01
1.17298043e+00 6.16747856e-01 9.49489594e-01 4.05438602e-01
-1.90151781e-01 1.57982558e-02 -6.57344043e-01 -1.31408423e-01
6.18139356e-02 -4.93588060e-01 -5.92940986e-01 -6.64255381e-01
-8.82044528e-03 3.22595499e-02 3.87488782e-01 8.82357359e-01
8.08235109e-01 2.17734292e-01 6.01682186e-01 -9.62159872e-01
-7.86132812e-01 5.22123337e-01 -3.25508565e-01 2.45804768e-02
4.44246531e-01 -9.47497860e-02 9.84756172e-01 -5.87510824e-01
6.08293533e-01 8.40954065e-01 9.93212342e-01 1.91177323e-01
-1.00293410e+00 -4.58413601e-01 -2.66797215e-01 7.69526482e-01
-1.29638648e+00 2.61078537e-01 6.85576677e-01 -2.56936401e-01
1.23610210e+00 5.85866570e-01 9.46229100e-01 7.27965653e-01
2.76510119e-01 3.79435211e-01 9.52818871e-01 -3.45316499e-01
-3.82392555e-02 8.07818547e-02 6.01233579e-02 3.00482005e-01
2.83362926e-04 -1.84540842e-02 -2.31967539e-01 -5.64941406e-01
5.25910139e-01 -2.41992608e-01 -4.51662779e-01 -4.92115408e-01
-1.16607511e+00 9.55580354e-01 3.30652624e-01 6.73370183e-01
-2.90371597e-01 -6.81629419e-01 3.89496416e-01 1.33666933e-01
1.14648849e-01 8.98982406e-01 -5.88560104e-01 -3.39903980e-01
-7.60895133e-01 3.37715186e-02 8.60536158e-01 2.68303633e-01
5.28130233e-01 1.09050326e-01 2.52503067e-01 1.21125424e+00
1.21841274e-01 3.38841319e-01 9.05150831e-01 -8.06249559e-01
7.15037063e-02 8.54312420e-01 3.93935628e-02 -1.38366890e+00
-6.95773482e-01 -1.43611267e-01 -8.05800200e-01 1.37383819e-01
3.82998049e-01 1.56586915e-01 -3.67431551e-01 1.59666133e+00
4.42082167e-01 -1.00642152e-01 1.05040871e-01 8.45640838e-01
5.60175598e-01 8.62868965e-01 -3.06736618e-01 -5.79685390e-01
1.05323899e+00 -8.76240373e-01 -6.47521079e-01 2.82003701e-01
6.21842980e-01 -1.08667123e+00 1.15305889e+00 3.04424435e-01
-8.07379901e-01 -4.92063284e-01 -1.50160360e+00 3.41164678e-01
-3.15485597e-01 -7.60454983e-02 5.69993198e-01 6.91946626e-01
-1.07511854e+00 1.20959151e+00 -6.69797361e-01 -4.22388434e-01
-3.87863576e-01 6.90749288e-01 -6.42661095e-01 6.13917053e-01
-1.54561603e+00 1.41049159e+00 8.94131839e-01 1.50829017e-01
1.38997748e-01 -2.31579795e-01 -5.89194238e-01 1.14747345e-01
-1.10664658e-01 -3.92241061e-01 8.01897764e-01 -9.94657278e-01
-2.02489233e+00 5.13995826e-01 -7.44715631e-02 -1.75723329e-01
-4.96946014e-02 7.72924572e-02 -3.35644394e-01 1.47046998e-01
-2.38518938e-01 2.06432909e-01 6.24180995e-02 -8.43020439e-01
-3.27873498e-01 -1.41050652e-01 -4.33644414e-01 3.01732183e-01
-3.37199956e-01 4.60160255e-01 1.41904652e-01 -2.60134012e-01
1.64018989e-01 -8.26828420e-01 -3.08748543e-01 -3.04346979e-01
3.31093490e-01 -2.37238407e-01 4.85926658e-01 -1.07661474e+00
1.38994634e+00 -1.87081993e+00 2.58394301e-01 6.20186388e-01
-3.50184560e-01 8.91676545e-01 -4.24858063e-01 8.34167957e-01
-3.96687806e-01 -5.20807318e-02 -3.60674620e-01 6.02643549e-01
-1.91188976e-01 2.56625473e-01 2.21683741e-01 3.43369007e-01
9.63941291e-02 5.17379880e-01 -8.40815425e-01 -3.80822361e-01
8.35804343e-02 4.93707031e-01 -4.30622250e-01 2.01065809e-01
8.68260413e-02 2.77077973e-01 -1.47420406e-01 5.13852239e-01
7.82399952e-01 -6.35488108e-02 6.90187037e-01 -2.56292224e-01
-2.35499978e-01 8.78430530e-02 -1.20916271e+00 1.42211485e+00
-8.64409432e-02 3.76226217e-01 -2.76901364e-01 -1.00607479e+00
1.63667393e+00 2.24297002e-01 6.36771619e-01 -4.00990665e-01
-1.13956474e-01 4.28687334e-01 5.79151988e-01 -6.54048443e-01
6.32615745e-01 5.34703694e-02 4.29786891e-01 1.70530960e-01
-2.50901401e-01 1.96042120e-01 2.11180642e-01 -5.12319565e-01
6.97197914e-01 4.68646646e-01 9.35591996e-01 -1.11190081e-01
9.45605993e-01 7.34988647e-03 7.70968914e-01 4.26575094e-01
-1.26564354e-01 3.59855235e-01 1.85453013e-01 -6.29634976e-01
-1.39001620e+00 -8.91606688e-01 -2.70475537e-01 1.04030049e+00
1.30070299e-01 -3.92661780e-01 -6.26570404e-01 -1.07938394e-01
-2.36892194e-01 3.33135724e-01 7.34573975e-03 -1.54721960e-01
-8.56549144e-01 -1.10660303e+00 7.74224043e-01 2.81205863e-01
3.62686545e-01 -1.09829795e+00 -2.42877945e-01 5.63995898e-01
-3.71324807e-01 -3.80941898e-01 -4.30952102e-01 2.23574251e-01
-8.89335155e-01 -9.95683432e-01 -8.25640500e-01 -9.37962651e-01
2.60971040e-01 2.64570981e-01 5.63759565e-01 1.59090087e-01
5.48370518e-02 -5.56549191e-01 -4.87752736e-01 -1.27699897e-01
-7.55610347e-01 2.63428420e-01 2.61938989e-01 -3.49177778e-01
4.65462327e-01 -9.23884571e-01 -4.86727566e-01 7.36795545e-01
-6.53205514e-01 -2.49309555e-01 7.54476845e-01 1.08597219e+00
4.63846415e-01 -2.75316328e-01 6.36285305e-01 -3.35557044e-01
8.06284606e-01 -1.93255782e-01 -4.48705286e-01 4.84057814e-01
-6.76242888e-01 5.46204671e-02 7.41335392e-01 -7.92046487e-01
-1.03225136e+00 1.12054154e-01 -6.99829578e-01 2.98799336e-01
6.47478700e-02 6.05303466e-01 -3.11135113e-01 -3.69550645e-01
7.59059846e-01 5.89961648e-01 5.27648568e-01 -5.38150251e-01
-1.03339992e-01 9.92536962e-01 4.04327303e-01 -5.77265084e-01
2.32910052e-01 -1.39383793e-01 1.52666777e-01 -8.70454788e-01
2.21993793e-02 -6.87825441e-01 -7.44124949e-01 -5.41887805e-02
7.59273648e-01 -3.22563857e-01 -1.17911708e+00 6.15017653e-01
-1.17995322e+00 1.03576988e-01 5.07299662e-01 6.14693284e-01
-6.17277682e-01 1.18223560e+00 -5.01215518e-01 -6.13975227e-01
-5.89103401e-01 -1.26525640e+00 6.02755308e-01 2.53196895e-01
-5.00722766e-01 -7.78039336e-01 5.29452085e-01 1.97899565e-01
6.40861750e-01 2.91465551e-01 1.15483677e+00 -1.04346073e+00
-1.83961373e-02 -1.54536530e-01 1.08911484e-01 5.06057858e-01
2.89693564e-01 3.04315358e-01 -4.47100669e-01 -1.20071851e-01
3.66282538e-02 2.60054767e-01 2.20214903e-01 1.44440979e-01
8.81102502e-01 -4.29669350e-01 -3.89902353e-01 7.39575922e-01
1.24703693e+00 1.04781199e+00 7.38924384e-01 9.17208612e-01
2.80010134e-01 7.16621637e-01 8.20385098e-01 1.53133497e-01
-4.30212542e-03 1.24495065e+00 2.08653465e-01 8.31120312e-02
6.00357831e-01 -4.18882333e-02 3.83145690e-01 1.19683516e+00
-6.10683084e-01 -8.24973509e-02 -8.98208141e-01 3.53482887e-02
-2.05870128e+00 -1.18285429e+00 -2.89244026e-01 2.30924439e+00
1.08648741e+00 -2.96836585e-01 2.91302115e-01 1.33841813e-01
7.94122458e-01 -1.50529653e-01 -3.61295164e-01 -9.53106403e-01
-3.81378502e-01 1.28420800e-01 2.75631696e-01 3.97777110e-01
-8.59435797e-01 5.86502254e-01 6.95915842e+00 9.62728620e-01
-1.16022229e+00 -3.31297278e-01 -4.01665494e-02 1.61966264e-01
-1.02942742e-01 1.28684659e-02 -7.48990715e-01 7.75026441e-01
1.09105241e+00 -1.69887707e-01 2.94005573e-01 1.04826260e+00
2.41019189e-01 1.51950672e-01 -6.48861229e-01 6.79498315e-01
-1.92695893e-02 -1.26350415e+00 1.08345896e-01 -4.62856479e-02
3.52084935e-01 -2.87933588e-01 -3.98617148e-01 -1.42637029e-01
6.70527369e-02 -8.20817530e-01 -1.46662116e-01 4.19720322e-01
7.27154613e-02 -8.99265647e-01 1.27439570e+00 2.80386567e-01
-9.31699157e-01 9.38167050e-02 -6.03759348e-01 -1.01049468e-01
5.52479863e-01 3.40406150e-01 -1.01050711e+00 8.53918076e-01
4.90817904e-01 3.39485645e-01 -2.64835387e-01 1.29013073e+00
-6.72265738e-02 2.92188793e-01 -4.64342713e-01 -5.50021052e-01
9.96156931e-02 -7.68529236e-01 6.90320849e-01 1.29244566e+00
5.64785063e-01 -2.11207420e-01 -3.61383110e-02 2.49815717e-01
8.08235943e-01 7.83210099e-01 -4.85908985e-01 -3.98654863e-02
6.17230773e-01 1.00864947e+00 -5.24476290e-01 -7.77031854e-02
-1.74788147e-01 8.92427802e-01 2.38215476e-01 7.10943863e-02
-8.16582203e-01 -6.81548715e-01 8.70878994e-01 -1.98858529e-01
1.63272262e-01 -2.09724903e-01 -1.63598303e-02 -7.88595676e-01
-5.98200485e-02 -1.18514550e+00 3.22266370e-01 -8.26042652e-01
-1.28585362e+00 6.94199681e-01 4.08814326e-02 -1.25030065e+00
-4.97066975e-01 -8.53008747e-01 -4.68897164e-01 1.09730136e+00
-8.79403293e-01 -9.22982395e-01 -2.39274595e-02 1.40425637e-01
2.35489324e-01 -3.58122349e-01 1.09719718e+00 3.15846831e-01
-3.02777916e-01 5.37411273e-01 6.41146302e-01 -4.39886600e-01
8.24366271e-01 -1.07396257e+00 2.07735851e-01 4.65451270e-01
-2.64890283e-01 9.41648185e-01 6.92487419e-01 -6.83811307e-01
-8.39209437e-01 -3.00515711e-01 1.14003634e+00 6.22481592e-02
6.68069482e-01 7.98422024e-02 -1.39154303e+00 1.54797301e-01
1.47167787e-01 -1.10995853e+00 1.32073820e+00 -9.34054628e-02
-4.78849411e-02 3.09777595e-02 -1.08300698e+00 6.36210501e-01
8.80162597e-01 -1.77679449e-01 -7.71492422e-01 1.39489442e-01
6.50760829e-01 -2.64072090e-01 -1.39494419e+00 6.29653275e-01
1.10774457e+00 -9.09508348e-01 1.39386868e+00 -5.81844568e-01
2.58535057e-01 -7.54280746e-01 -2.12564319e-01 -1.25689352e+00
-7.29806781e-01 -4.05584335e-01 3.17509711e-01 1.06873417e+00
4.75290000e-01 -9.52307284e-01 5.54859877e-01 4.24861833e-02
-2.63921946e-01 -8.53297472e-01 -9.58726287e-01 -1.07037520e+00
-7.66681433e-02 3.14469397e-01 9.66967702e-01 9.45132852e-01
3.31824154e-01 1.51594490e-01 -6.26528919e-01 5.07487319e-02
9.47598293e-02 8.80144536e-02 5.57828128e-01 -1.14558041e+00
-6.90890491e-01 -7.02793956e-01 -6.96683943e-01 -5.82758307e-01
3.63194123e-02 -9.46894169e-01 -7.99204186e-02 -1.37836778e+00
1.06716931e-01 -4.56714556e-02 -3.31075668e-01 4.75378275e-01
-4.78864431e-01 4.18106914e-02 1.49264276e-01 3.20121616e-01
2.29401350e-01 3.21087211e-01 1.07798040e+00 1.39585987e-01
-5.04106343e-01 2.22901806e-01 -3.60594451e-01 6.11956120e-01
1.15581036e+00 -4.15491462e-01 -2.13668361e-01 2.22727343e-01
4.72715706e-01 -1.03402048e-01 -3.54355574e-01 -6.16167545e-01
3.03323455e-02 -5.26925445e-01 3.56507957e-01 -5.98538697e-01
3.43276918e-01 -7.30512679e-01 7.69020498e-01 6.61329865e-01
-5.94594479e-02 5.63337684e-01 1.43587127e-01 2.53479220e-02
-3.60399038e-01 -8.10333908e-01 5.77421188e-01 -7.27196876e-03
-8.34551632e-01 -2.91988701e-01 -5.82064509e-01 -9.29702461e-01
9.13774669e-01 -8.95872831e-01 -1.55364677e-01 -8.56800750e-02
-7.26829886e-01 -1.28303528e-01 6.08468771e-01 2.99903035e-01
5.83669484e-01 -1.29883993e+00 -5.50523281e-01 1.04306616e-01
-4.41551171e-02 -5.27117014e-01 2.76307136e-01 8.89037311e-01
-9.61917162e-01 3.24490964e-01 -9.15793002e-01 -5.97802281e-01
-2.01613784e+00 6.57649159e-01 2.39356652e-01 -3.00566882e-01
-1.04784079e-01 4.26283091e-01 -2.17988566e-01 -9.48747337e-01
-6.06964603e-02 1.24325782e-01 -7.26015568e-01 9.86261219e-02
4.72863466e-01 6.01578832e-01 -3.13775092e-02 -6.75393641e-01
-5.56946814e-01 8.63180995e-01 1.19127043e-01 2.74489433e-01
1.45805919e+00 1.85268056e-02 -5.41102290e-01 1.88112602e-01
1.16395235e+00 -1.50111169e-01 -5.79022527e-01 1.98439106e-01
2.01565042e-01 -4.35690612e-01 -8.36466193e-01 -8.45745087e-01
-3.41362834e-01 5.02882481e-01 6.67861402e-01 -1.06533676e-01
1.20824921e+00 -5.74685693e-01 7.21744001e-01 5.71509719e-01
2.39436522e-01 -9.76599872e-01 -1.56481460e-01 5.49493849e-01
8.05015802e-01 -9.26151395e-01 -1.74973309e-01 -4.15293962e-01
-5.61300516e-01 1.54652631e+00 4.88780141e-01 3.38613659e-01
2.47104708e-02 1.45749763e-01 1.19786911e-01 3.00593436e-01
-3.50383192e-01 1.20082885e-01 3.55641514e-01 9.89255428e-01
6.14265800e-01 -4.09768037e-02 -1.23556852e+00 3.18321764e-01
-4.48094666e-01 1.10438438e-02 3.65519106e-01 6.70120776e-01
-7.57772326e-01 -1.71868265e+00 -4.05240834e-01 9.45030525e-02
-3.29519063e-01 -1.35904670e-01 -5.25919735e-01 6.77534699e-01
1.19278476e-01 6.77286506e-01 -1.48850024e-01 -6.33055747e-01
2.13127255e-01 2.24620834e-01 3.99573743e-01 -6.31659701e-02
-8.18988264e-01 -1.61621273e-01 3.01544815e-01 -3.61708909e-01
-4.81197685e-01 -4.67260867e-01 -9.80518401e-01 -2.96908468e-01
-5.76496840e-01 4.90848750e-01 9.51893032e-01 7.39225388e-01
2.58358836e-01 1.81542281e-02 5.86679459e-01 -6.24333501e-01
-7.42518604e-01 -1.06369185e+00 -3.85955840e-01 2.52381593e-01
-2.61200041e-01 -4.62645411e-01 -1.61804602e-01 -2.39797056e-01] | [4.8131184577941895, 5.272375106811523] |
4a945ae5-3ae7-49c7-ac15-d89a4352ba72 | best-feature-performance-in-codeswitched-hate | null | null | https://openreview.net/forum?id=Skl6peHFwS | https://openreview.net/pdf?id=Skl6peHFwS | Best feature performance in codeswitched hate speech texts | How well can hate speech concept be abstracted in order to inform automatic classification in codeswitched texts by machine learning classifiers? We explore different representations and empirically evaluate their predictiveness using both conventional and deep learning algorithms in identifying hate speech in a ~48k human-annotated dataset that contain mixed languages, a phenomenon common among multilingual speakers. This paper espouses a novel approach to handle this challenge by introducing a hierarchical approach that employs Latent Dirichlet Allocation to generate topic models that feed into another high-level feature set that we acronym PDC. PDC groups similar meaning words in word families during the preprocessing stage for supervised learning models. The high-level PDC features generated are based on Ombui et al, (2019) hate speech annotation framework that is informed by the triangular theory of hate (Stanberg,2003). Results obtained from frequency-based models using the PDC feature on the annotated dataset of ~48k short messages comprising of tweets generated during the 2012 and 2017 Kenyan presidential elections indicate an improvement on classification accuracy in identifying hate speech as compared to the baseline | ['Peter Wagacha', 'Lawrence Muchemi', 'Edward Ombui'] | 2019-09-25 | null | null | null | null | ['topic-models'] | ['natural-language-processing'] | [-6.34598583e-02 1.80905521e-01 -1.61575750e-01 -1.29300535e-01
-4.81767297e-01 -5.99545002e-01 1.26035297e+00 5.65660119e-01
-3.76841605e-01 4.72364992e-01 9.51585829e-01 -2.57283628e-01
6.83090091e-02 -4.73369718e-01 -7.12931827e-02 -7.31006682e-01
5.59239984e-02 2.60741889e-01 -3.22250217e-01 -1.96796104e-01
6.99702084e-01 3.09607834e-01 -1.44847822e+00 5.45267999e-01
8.67001951e-01 1.02210008e-01 -2.39195094e-01 6.69839680e-01
-8.48791003e-02 9.89584267e-01 -1.15871787e+00 -6.19866610e-01
-1.69009671e-01 -1.12936482e-01 -9.84654605e-01 5.00738248e-02
4.43837553e-01 -5.09352744e-01 -3.00020993e-01 9.65055823e-01
5.20491838e-01 -1.61331475e-01 1.31387866e+00 -1.01520324e+00
-1.14609158e+00 8.16383064e-01 -4.00224566e-01 5.95337510e-01
3.98251444e-01 -3.29666585e-02 8.12571406e-01 -8.80915701e-01
5.76948285e-01 1.37214649e+00 8.77228975e-01 4.83738363e-01
-1.21123838e+00 -1.02712703e+00 -3.46953601e-01 1.65180326e-01
-1.43267632e+00 -2.85363346e-01 7.45904207e-01 -1.20107853e+00
1.29398990e+00 4.63487133e-02 5.38930416e-01 1.73435640e+00
1.62357077e-01 8.21261525e-01 1.04543221e+00 -6.07325375e-01
1.37910888e-01 4.99790847e-01 6.18990541e-01 3.73277664e-01
4.80350256e-01 -3.08121055e-01 -7.03377962e-01 -8.25273931e-01
7.16454908e-02 -1.00602105e-01 1.29822418e-01 1.80123478e-01
-9.39831078e-01 1.87514973e+00 -5.86087704e-02 7.93726206e-01
-4.27922845e-01 -2.61571705e-01 8.75286639e-01 -5.74069209e-02
1.11193955e+00 6.64732635e-01 1.02320828e-01 -1.86475277e-01
-1.18621230e+00 3.85928452e-01 7.75934756e-01 4.54553604e-01
4.29437310e-01 1.89127307e-02 -2.53900915e-01 1.01459110e+00
3.00761700e-01 6.62531853e-01 7.96199858e-01 -3.36940020e-01
4.04049575e-01 3.22334051e-01 9.96415243e-02 -1.54837596e+00
-2.45524600e-01 1.19601160e-01 -6.44971967e-01 -3.45982820e-01
1.18353873e-01 -5.21098137e-01 -1.06034398e+00 1.10113168e+00
6.41789511e-02 -2.82760262e-01 -4.75719571e-02 4.83709514e-01
6.28542125e-01 1.15686142e+00 6.14099026e-01 -8.64014253e-02
1.30926323e+00 -5.20257294e-01 -1.15143430e+00 2.33478561e-01
1.05187809e+00 -8.44288826e-01 7.06357062e-01 5.85270107e-01
-3.85966450e-01 -1.85117736e-01 -1.15222704e+00 -2.41454110e-01
-1.06484258e+00 -2.97360808e-01 6.26798928e-01 1.23012245e+00
-8.66621912e-01 1.44497737e-01 -3.02418560e-01 -6.90688193e-01
3.67903352e-01 4.75165807e-02 -4.18639332e-01 1.59273759e-01
-1.67062891e+00 1.31212223e+00 5.17873287e-01 -2.47474596e-01
-1.10673726e+00 -4.89781916e-01 -9.81278539e-01 -3.01195890e-01
-1.73800498e-01 2.78283089e-01 8.66302669e-01 -8.77519488e-01
-1.15941978e+00 1.22821867e+00 8.82770196e-02 -6.22705281e-01
-9.31933448e-02 -3.64667177e-01 -4.23151493e-01 1.03576444e-01
2.37686947e-01 4.03755635e-01 1.33545220e+00 -1.00920880e+00
-3.11722815e-01 -1.95593134e-01 -1.57605946e-01 -1.45657331e-01
-7.40845799e-01 6.22083783e-01 7.42867470e-01 -8.72143984e-01
-7.21140742e-01 -8.65324020e-01 4.72312987e-01 -8.57841790e-01
-5.58014870e-01 -6.25246286e-01 1.22160470e+00 -1.16712308e+00
1.71357334e+00 -2.17801809e+00 1.40663728e-01 9.99310017e-02
5.00917435e-01 7.50771880e-01 3.04849058e-01 9.48261499e-01
-2.07980007e-01 4.34754640e-01 -3.50995570e-01 -3.39256257e-01
3.27533096e-01 9.17528644e-02 -6.39834106e-01 9.24388945e-01
2.30865851e-01 2.88593620e-01 -8.96152854e-01 -2.56177783e-01
2.67028391e-01 8.24168801e-01 -4.30714935e-01 1.61992788e-01
2.31076717e-01 1.75654944e-02 -5.68111092e-02 2.49716520e-01
5.99923193e-01 3.91344965e-01 -7.14975670e-02 4.70302999e-01
-4.57473338e-01 3.89649123e-01 -4.47141796e-01 1.19138122e+00
-1.11337095e-01 1.38905311e+00 -1.87941864e-01 -8.48816454e-01
1.13574934e+00 6.84310734e-01 -9.41652283e-02 -2.73541927e-01
2.37256974e-01 1.48485169e-01 -1.27089709e-01 -6.22704983e-01
6.88359737e-01 -2.61012822e-01 -5.53161919e-01 3.66316766e-01
3.40561241e-01 2.33165803e-03 -1.95109367e-01 2.67910898e-01
9.68833089e-01 -3.51804107e-01 6.29197836e-01 -7.18794346e-01
5.78343749e-01 2.62320429e-01 6.80508614e-02 6.25488400e-01
-4.62061942e-01 3.16291183e-01 8.40860486e-01 -8.17330360e-01
-1.45217788e+00 -4.78948414e-01 -5.43654323e-01 1.47518146e+00
-6.71680808e-01 -4.34860498e-01 -1.11256921e+00 -7.37087488e-01
3.23189981e-02 1.24744344e+00 -1.16172862e+00 -6.96578622e-02
-2.08612263e-01 -9.61734176e-01 1.07383847e+00 -8.93856809e-02
1.73219502e-01 -1.10492647e+00 -7.09201574e-01 -1.90476980e-02
9.27405283e-02 -6.94566846e-01 6.12542033e-02 4.09974813e-01
5.40232472e-02 -8.44840348e-01 -9.82521176e-01 -5.78903079e-01
3.86147678e-01 -6.77968636e-02 6.15149796e-01 1.01013251e-01
-3.84524405e-01 1.16021141e-01 -8.51204753e-01 -8.42330039e-01
-7.58804977e-01 3.20872307e-01 1.14802457e-01 -4.87496816e-02
1.19959211e+00 -2.23829463e-01 -1.38499871e-01 -4.15692657e-01
-1.12317359e+00 -1.94536924e-01 2.23702684e-01 5.99342763e-01
-6.54933572e-01 -4.79133613e-02 4.80709404e-01 -1.12279093e+00
8.08547080e-01 -1.17897797e+00 -2.04057008e-01 -2.35699341e-01
-1.17320932e-01 -1.31223649e-01 3.74248475e-01 -4.58645791e-01
-8.48540127e-01 -2.62792647e-01 -3.56487744e-02 -1.13658033e-01
-5.56387067e-01 3.36403638e-01 3.29737872e-01 4.58033711e-01
8.13291788e-01 6.21516369e-02 -2.07861394e-01 -4.02232140e-01
4.16359782e-01 1.35434031e+00 3.58602628e-02 -3.29636246e-01
1.05281901e+00 1.19328886e-01 -5.30682802e-01 -1.43897295e+00
-9.07891929e-01 -8.25033426e-01 -1.10714734e+00 -1.38183028e-01
1.41664088e+00 -8.45582664e-01 -2.33694270e-01 9.42318916e-01
-1.55966735e+00 -5.19393422e-02 4.33041334e-01 3.38111579e-01
1.79206824e-03 3.60244662e-01 -7.20975637e-01 -1.30139172e+00
-4.87567186e-01 -7.80480802e-01 1.28470612e+00 -3.31563741e-01
-7.61457384e-01 -1.31993496e+00 5.68907142e-01 3.06326747e-01
4.16489929e-01 6.55882597e-01 1.51700139e+00 -1.53611481e+00
4.82716858e-01 -1.29569098e-01 -6.36866167e-02 4.04228508e-01
5.91035783e-02 2.39090666e-01 -1.21068847e+00 -2.45916530e-01
-3.41365747e-02 -5.10925055e-01 4.99317050e-01 -1.72321171e-01
4.57514554e-01 -7.40362763e-01 -1.33723512e-01 -1.54818548e-02
1.40726674e+00 4.70879450e-02 6.41169786e-01 5.79001427e-01
7.51419067e-01 8.83237481e-01 1.70859024e-02 7.48841941e-01
5.14236502e-02 3.85772228e-01 4.76580262e-02 3.51327598e-01
9.36106890e-02 -1.99721217e-01 6.04790211e-01 1.24590516e+00
3.80572945e-01 -8.81040469e-02 -1.68988526e+00 8.97914112e-01
-1.43645728e+00 -1.20016861e+00 -3.01529348e-01 1.85000956e+00
8.66584003e-01 -9.76014510e-02 3.34775805e-01 4.01051134e-01
1.11862767e+00 3.58642936e-01 2.43407801e-01 -1.29379547e+00
1.07440576e-01 2.20337272e-01 4.97424632e-01 6.76202178e-01
-1.37004006e+00 1.12303054e+00 6.39268160e+00 1.01743078e+00
-9.86069262e-01 4.51692045e-01 4.06163633e-01 2.31674954e-01
-1.12914667e-01 -4.84594107e-01 -9.96126473e-01 7.89344132e-01
1.59976804e+00 -2.02703014e-01 1.51967421e-01 6.40907943e-01
6.80185929e-02 -5.68131842e-02 -7.05547392e-01 8.63784194e-01
8.17732036e-01 -1.10786366e+00 3.64720345e-01 4.14264172e-01
8.00342739e-01 -1.49861336e-01 3.96265060e-01 5.54270029e-01
2.80361533e-01 -1.41158366e+00 5.75097561e-01 2.95967966e-01
4.08303738e-01 -1.05801165e+00 9.72475946e-01 5.45296848e-01
-2.56264478e-01 -1.41006961e-01 -5.71946204e-01 -2.78008103e-01
-3.73329461e-01 4.36092168e-01 -1.37283850e+00 -6.64055198e-02
4.70685989e-01 5.17093658e-01 -7.14511633e-01 4.83724058e-01
-1.38516113e-01 9.72359955e-01 5.08878678e-02 -1.15005344e-01
8.61182272e-01 2.68783271e-01 5.42246580e-01 1.90151274e+00
5.85433058e-02 -2.06507102e-01 -2.85656720e-01 6.55560732e-01
2.86461979e-01 2.50065416e-01 -1.18375123e+00 -4.62030083e-01
6.46608174e-01 1.06524050e+00 -6.11380935e-01 -3.28925788e-01
-4.17544395e-01 1.05686235e+00 4.23529685e-01 2.34499685e-02
-8.25728774e-01 -5.99183202e-01 4.51387674e-01 1.32753611e-01
2.11730659e-01 -3.08133692e-01 3.90846506e-02 -7.94814765e-01
-8.01435947e-01 -1.04129553e+00 1.08505562e-01 -5.28296769e-01
-1.58736169e+00 5.17188907e-01 4.24808949e-01 -5.63925743e-01
-3.78565103e-01 -6.28762901e-01 -3.96926939e-01 9.05546486e-01
-9.69847977e-01 -1.41475379e+00 1.31968260e-01 4.57548648e-01
6.58467889e-01 -2.57058412e-01 1.21564126e+00 9.64301676e-02
-7.27313042e-01 3.79755110e-01 3.77279460e-01 7.60743558e-01
6.43451929e-01 -1.39417422e+00 3.27083409e-01 6.82478845e-01
-1.44161105e-01 7.50200331e-01 8.24902117e-01 -8.56759131e-01
-8.47624600e-01 -9.33373809e-01 1.38002670e+00 -9.82208788e-01
1.05659676e+00 -8.25471997e-01 -1.01979232e+00 5.60307920e-01
9.12086844e-01 -5.66890121e-01 1.39068854e+00 1.30500749e-01
-6.96974039e-01 7.90046215e-01 -1.18823111e+00 9.91304219e-02
1.88061118e-01 -9.93968248e-01 -1.43013465e+00 5.14984488e-01
5.88281691e-01 1.87572196e-01 -6.73226893e-01 -3.19714397e-01
3.63960177e-01 -7.05337703e-01 2.47562110e-01 -8.20433915e-01
5.29252470e-01 2.52655864e-01 -2.07097188e-01 -1.28385305e+00
-5.07140160e-01 -8.18060994e-01 -1.15479216e-01 1.46357822e+00
1.62167266e-01 -2.47109070e-01 2.55156606e-01 4.18545216e-01
1.81125104e-01 -1.77965760e-01 -9.78689671e-01 -3.19380969e-01
7.55176604e-01 -1.80700287e-01 1.99858814e-01 1.67933285e+00
3.75502855e-01 5.53961277e-01 -7.09864318e-01 2.87627243e-02
5.32033563e-01 -8.34217846e-01 6.76847219e-01 -1.52432275e+00
5.40956259e-01 -1.85131192e-01 -7.77213633e-01 -8.66263285e-02
7.27346659e-01 -8.52280736e-01 -1.35658607e-01 -1.20251703e+00
5.76032102e-01 1.37609109e-01 1.55169278e-01 3.26492220e-01
2.00032759e-02 2.69636691e-01 2.30957001e-01 3.50267440e-01
-3.35334569e-01 7.12290853e-02 2.84520060e-01 -1.00218698e-01
9.87453535e-02 -8.00437868e-01 -4.78235513e-01 8.66628885e-01
5.95858157e-01 -7.66896844e-01 -1.13917604e-01 -3.49101394e-01
2.41784334e-01 -5.83397031e-01 2.17050433e-01 -9.50054288e-01
-1.33216769e-01 1.18720457e-01 3.92005473e-01 -5.11887968e-01
1.75944358e-01 -5.92759848e-01 -2.52529353e-01 5.41994870e-01
-6.44148350e-01 -7.94964954e-02 2.21729934e-01 3.17075968e-01
-1.02128670e-01 -5.84125519e-01 8.17572534e-01 -4.21620794e-02
-5.20924211e-01 -1.18128069e-01 -1.41209161e+00 -4.09261845e-02
1.10661733e+00 -6.09109029e-02 -7.14935601e-01 -5.06250024e-01
-5.14029860e-01 -4.32426482e-01 2.08996907e-01 5.58150291e-01
1.77786812e-01 -1.30984592e+00 -9.17758226e-01 9.99416336e-02
1.59376353e-01 -5.91345012e-01 -1.22981124e-01 5.05693376e-01
-7.20689535e-01 1.07104778e+00 -3.28175485e-01 -7.99517855e-02
-9.99056160e-01 8.63377094e-01 -1.31491512e-01 -1.31413341e-01
-5.81178784e-01 6.44704938e-01 1.22542761e-01 -1.71376571e-01
1.25175908e-01 2.84534812e-01 -7.09162891e-01 7.66408384e-01
7.66122937e-01 3.94881666e-01 -1.59701586e-01 -1.50483382e+00
-3.83408546e-01 -5.30815162e-02 -4.23347026e-01 -9.57268178e-02
1.30797744e+00 3.17348726e-02 -2.92396784e-01 7.46877253e-01
1.58340740e+00 2.50766009e-01 -3.77229184e-01 5.86109236e-02
5.76175570e-01 -4.61445272e-01 1.71498567e-01 -6.54975057e-01
-2.64131874e-01 1.04235876e+00 5.17988741e-01 7.86262333e-01
2.53169984e-01 -2.42845714e-01 7.94226289e-01 3.22363466e-01
1.41389137e-02 -1.19860101e+00 2.28199735e-02 8.45866323e-01
9.33379173e-01 -9.36195254e-01 3.40830162e-02 -9.53748077e-02
-7.85607398e-01 1.22109270e+00 2.88682789e-01 -3.29943895e-01
6.01098001e-01 -3.77381481e-02 1.14726938e-01 -4.89387691e-01
-5.25391698e-01 7.16577917e-02 1.76276088e-01 9.82986093e-01
8.36010516e-01 1.75148919e-01 -5.26580155e-01 4.62992072e-01
-3.84738326e-01 -6.49974406e-01 7.51971006e-01 7.94534683e-01
-7.85098076e-01 -6.27152383e-01 -6.44901872e-01 1.99563444e-01
-1.01722014e+00 -3.21828157e-01 -1.01120198e+00 9.43145633e-01
3.50830078e-01 9.80177045e-01 2.34024480e-01 -3.87211412e-01
-3.79181117e-01 6.22188151e-01 -1.14095367e-01 -9.97014403e-01
-9.53239501e-01 -1.88017517e-01 3.23092699e-01 2.09412560e-01
-5.79364538e-01 -6.40912056e-01 -6.27269447e-01 -6.59054935e-01
-2.37639338e-01 6.53052032e-01 8.02941024e-01 1.15196347e+00
1.00638792e-01 -6.66043237e-02 5.72950304e-01 -9.29795027e-01
-2.03535333e-01 -1.49089515e+00 -5.77308595e-01 5.76715171e-01
6.08828485e-01 -6.78763926e-01 -6.72286630e-01 -4.11094613e-02] | [8.7759428024292, 10.563959121704102] |
3c943201-4054-4376-b2fb-9ed7b10d1bff | computer-vision-system-for-eye-gaze-tracking | null | null | http://ijmcs.info/ | http://ijmcs.info/ | Computer Vision System for Eye Gaze Tracking | : Eye gaze tracking is a technique use for checking the usability problems in the Human Computer Interaction (HCI). Initially they are
present tracking technology and key elements. Eye gaze tracking technique is based on the behavior of the user when they are looking. It can use
for different kinds of methods i.e. “electro-oculography(EOG), Sceleral Search Coils, infrared oculography(IOG), video oculography(VOG),
different models, probable approaches i.e. “shape based approach, appearance based methods, 2D and 3D models based approach. The eye
tracking applications likes human computer interaction, brain computer interaction, assistive technology, e-learning, psychology investigation,
pilot training assistance, virtual and augmented reality and so on. The Proposed system uses Head pose detection, eye region detection, eye
feature detection, eye vector calibration, Gaze Estimation and eye gaze tracking modules. This system is to design and implement eye gaze
tracking system in desktop environment. | ['Gyankamal J. Chhajed', 'Puja P. Sorate'] | 2017-06-03 | null | null | null | international-journal-of-modern-computer | ['gaze-estimation'] | ['computer-vision'] | [-3.60276341e-01 -1.40201285e-01 2.15499595e-01 1.36643752e-01
6.44822717e-01 -6.06018841e-01 1.19331166e-01 9.26444829e-02
-3.80729556e-01 5.75434506e-01 1.63397491e-01 -8.61145973e-01
-1.52252875e-02 9.47240293e-02 -9.60012078e-02 -3.58121246e-01
3.62388492e-01 -2.41384000e-01 4.11992073e-01 -6.64496869e-02
9.04992759e-01 4.81131285e-01 -2.08391380e+00 -4.98440593e-01
4.00339484e-01 7.58282781e-01 1.43244937e-01 1.20069849e+00
-1.95883922e-02 4.20830935e-01 -1.91206917e-01 6.08568370e-01
-7.95623381e-03 -3.21896285e-01 -4.84823674e-01 1.69046447e-01
1.76901504e-01 -2.25290284e-01 2.89749414e-01 1.04979694e+00
6.74487591e-01 2.02822581e-01 8.40262830e-01 -1.65254676e+00
-6.00072622e-01 -5.60596943e-01 -1.42148113e+00 7.62181401e-01
1.06257772e+00 -6.57862797e-02 -1.53968409e-01 -8.29184353e-01
5.82400203e-01 1.08163166e+00 6.32128298e-01 9.12116528e-01
-6.55895352e-01 -1.09620476e+00 -2.47312829e-01 5.79293728e-01
-1.35558105e+00 -5.19573390e-01 5.84220707e-01 -7.93750823e-01
1.01256192e+00 5.50307214e-01 1.95230573e-01 2.53338397e-01
5.96383572e-01 4.36058253e-01 1.63967526e+00 -1.08797622e+00
-2.60845602e-01 1.01188338e+00 1.08568251e+00 7.60441542e-01
2.61138648e-01 2.39189848e-01 -3.51158470e-01 -1.37702078e-01
2.68154114e-01 -4.60027456e-02 -5.25148094e-01 2.77621988e-02
-6.28643632e-01 3.05373818e-01 -1.05338186e-01 5.56934357e-01
-3.38372141e-01 -3.20457816e-01 1.05738163e-01 2.62019783e-01
4.22975868e-02 -2.43339822e-01 -4.72236186e-01 -2.10629940e-01
-6.90441310e-01 3.11239153e-01 3.66283953e-01 1.12293160e+00
3.12546045e-01 -2.93336540e-01 4.42488771e-03 4.15076166e-01
1.31500363e+00 4.34163392e-01 7.93147624e-01 -3.01015615e-01
-1.25198066e-01 7.93574810e-01 4.28235859e-01 -6.10415041e-01
-7.93036759e-01 6.37783587e-01 2.66451001e-01 1.04307783e+00
1.50982082e-01 -5.11775017e-01 -1.12175679e+00 1.38118088e+00
8.44111681e-01 4.57847118e-02 -3.11198920e-01 7.00492918e-01
1.16328979e+00 1.43774167e-01 3.13299388e-01 -3.54488760e-01
2.27914143e+00 -2.86702842e-01 -1.08014262e+00 1.97166771e-01
6.06979311e-01 -1.40740252e+00 6.51076734e-01 5.71910024e-01
-6.73085451e-01 -3.67555112e-01 -1.18560100e+00 2.14588135e-01
-7.32027173e-01 -4.70235273e-02 9.59920734e-02 1.41158569e+00
-1.30273390e+00 9.29474086e-02 -5.65158844e-01 -1.15291560e+00
-1.76893800e-01 7.94260323e-01 -4.74845022e-01 6.88470781e-01
-5.06142855e-01 1.02151322e+00 9.85614061e-02 -1.85505614e-01
3.26681644e-01 -1.83797404e-01 -6.27019823e-01 -2.97930270e-01
1.08448662e-01 -3.07296723e-01 9.17927682e-01 -1.10534620e+00
-1.45391965e+00 1.40238929e+00 -6.75042570e-01 4.28024195e-02
-1.37416914e-01 -6.70839623e-02 -6.29400015e-01 -1.57592088e-01
-3.88496369e-01 3.51907611e-01 9.77665961e-01 -6.37194216e-01
-7.09362566e-01 -1.06044710e+00 -3.88355315e-01 3.37282509e-01
4.00217652e-01 9.82718766e-01 1.71886519e-01 -1.10977709e-01
1.15869492e-01 -1.28771853e+00 6.37806892e-01 -4.46260542e-01
-3.92121673e-01 -5.41703761e-01 1.56677163e+00 -9.34166610e-01
1.38137949e+00 -1.98545289e+00 -3.55826229e-01 1.97918922e-01
2.92086303e-01 5.32771766e-01 7.56835938e-01 1.33564889e-01
-4.71156061e-01 1.62290707e-01 6.61698639e-01 9.09997895e-02
-1.78400323e-01 -8.62830758e-01 3.95263195e-01 5.32769561e-01
-3.03658277e-01 2.97177643e-01 -3.50861698e-01 -6.58664763e-01
3.65704060e-01 4.40755069e-01 -3.32975164e-02 -1.18441179e-01
5.91307044e-01 7.32966602e-01 -2.33102173e-01 8.19101691e-01
7.64136612e-01 1.36097640e-01 -2.72311926e-01 -3.10558919e-02
-5.23936808e-01 -4.32493120e-01 -1.19615209e+00 8.03662479e-01
-8.12996551e-02 1.31993592e+00 1.41350471e-03 -2.52234340e-02
7.63638258e-01 7.97751069e-01 1.66218802e-02 -5.85060239e-01
9.20676470e-01 1.15920581e-01 1.45220563e-01 -1.09079111e+00
2.51285285e-01 2.14058876e-01 1.01884568e+00 9.10977721e-01
5.60248382e-02 1.02118468e+00 -1.27547249e-01 -2.74484664e-01
4.50085819e-01 6.08837426e-01 8.32083046e-01 -4.13258851e-01
1.15896833e+00 -3.07287067e-01 -1.74232557e-01 -1.19951129e-01
-7.44062304e-01 1.32450417e-01 1.83831081e-01 8.93018022e-02
-7.94055998e-01 -4.07345057e-01 -4.41533983e-01 8.79788458e-01
1.62278131e-01 1.39169455e-01 -1.27302408e+00 -2.92094797e-01
-8.41820687e-02 3.59520078e-01 -6.75409257e-01 2.30707198e-01
-1.68523610e-01 -8.95975083e-02 -6.26087189e-02 1.25689924e-01
2.27474943e-01 -1.31180358e+00 -1.19489944e+00 -3.02117139e-01
8.12009931e-01 -6.62808478e-01 -4.26614553e-01 -5.47513425e-01
-8.73738945e-01 -1.33025086e+00 -8.35855305e-01 -8.13700020e-01
7.22391427e-01 3.00886989e-01 5.40538967e-01 5.42122304e-01
-5.35634637e-01 6.12907887e-01 -3.48043293e-01 -1.05953574e+00
-3.20812576e-02 -5.09665787e-01 5.94667971e-01 -1.48865566e-01
1.40722454e+00 -4.59612548e-01 -8.99498463e-01 2.39356667e-01
-1.06714442e-01 -4.46962386e-01 2.44602889e-01 2.01189220e-01
-1.08563878e-01 -6.08850479e-01 1.88105166e-01 -1.01428545e+00
1.17847085e+00 -3.03247392e-01 -8.00129533e-01 4.30541247e-01
-1.04001832e+00 -1.13177791e-01 -1.84503213e-01 -4.85302955e-01
-6.71312273e-01 -2.30515853e-01 3.91247809e-01 -3.67273182e-01
-7.08323658e-01 3.33475247e-02 7.88669735e-02 -4.64341760e-01
9.52963650e-01 1.74375158e-03 1.66359335e-01 -7.63646781e-01
-3.62628728e-01 1.54882014e+00 2.67722309e-01 2.09314436e-01
3.64287257e-01 -6.18440025e-02 2.28792634e-02 -1.12194204e+00
9.46303606e-02 -9.64379072e-01 -5.62679946e-01 -6.20310485e-01
1.21158314e+00 -3.43143970e-01 -1.57702827e+00 5.32183886e-01
-1.14735067e+00 2.87934095e-01 1.05930483e+00 7.61649013e-01
-4.80386205e-02 2.15178072e-01 1.31702378e-01 -1.80140972e+00
-8.72425258e-01 -1.01810956e+00 8.10456812e-01 9.86337006e-01
-3.97925258e-01 -9.67598617e-01 6.99713156e-02 -2.44939461e-01
2.16706172e-01 3.90420943e-01 5.16888440e-01 -1.35648862e-01
-2.03943536e-01 -3.40892285e-01 -1.35499835e-01 -1.87892914e-01
4.21414852e-01 2.10385740e-01 -1.00371754e+00 2.70707943e-02
8.67658854e-02 2.81783134e-01 -1.36812255e-01 4.80081618e-01
3.44775915e-01 -7.11112544e-02 -7.90422857e-01 3.18006516e-01
1.63531089e+00 1.08743417e+00 8.04456711e-01 4.38629001e-01
5.40600538e-01 6.43800139e-01 6.35725796e-01 2.57081002e-01
1.14087246e-01 6.16621912e-01 -3.14325660e-01 3.31930876e-01
9.70071033e-02 1.97129369e-01 2.01239716e-02 2.50848055e-01
-2.45428845e-01 -1.71505526e-01 -1.14769423e+00 -7.81834051e-02
-1.38731885e+00 -6.89620972e-01 -8.07586551e-01 2.67245173e+00
1.46265760e-01 -1.83886498e-01 4.21413541e-01 5.74989855e-01
1.12050080e+00 -8.34572673e-01 -3.15960437e-01 -9.03451025e-01
7.75016904e-01 5.16278982e-01 7.65082538e-01 5.36074221e-01
-6.49172723e-01 5.38809419e-01 5.04662657e+00 -1.42557472e-01
-1.55433822e+00 4.60770875e-01 6.70457631e-03 2.65349478e-01
1.98217779e-01 -3.13347206e-03 -1.12477493e+00 9.98649478e-01
6.99814081e-01 -2.54607588e-01 3.89301687e-01 4.40108210e-01
4.05873209e-01 -9.07292902e-01 -8.62680554e-01 1.74117577e+00
1.29275903e-01 -6.15280509e-01 -7.78843999e-01 3.02611887e-01
2.37824112e-01 -2.68813044e-01 -8.94533843e-02 -1.05842896e-01
-9.14565325e-01 -1.03258181e+00 3.93634886e-01 6.89528465e-01
7.39807069e-01 -6.31727695e-01 4.43297744e-01 5.73118813e-02
-1.00081480e+00 -9.06022862e-02 1.11607902e-01 -7.52997175e-02
1.63789958e-01 -6.62846506e-01 -7.70003736e-01 -2.18041822e-01
7.96280265e-01 -1.82396859e-01 -7.18955457e-01 1.44504189e+00
3.33711267e-01 2.17195988e-01 -3.68583947e-01 -4.26670015e-01
2.87348535e-02 -2.92948961e-01 6.52037263e-01 6.39581144e-01
4.63271588e-01 5.89973807e-01 -8.30016971e-01 6.83768094e-01
5.78742981e-01 5.42789459e-01 -8.71891975e-01 2.06183463e-01
4.82090622e-01 1.20347404e+00 -8.23136628e-01 -5.63164130e-02
-2.82510877e-01 5.45461416e-01 -5.10372162e-01 4.64754581e-01
-4.03195858e-01 -1.17292833e+00 5.04853606e-01 7.17686474e-01
-7.55790994e-02 -7.24028647e-02 -1.00683689e-01 -4.82259452e-01
-8.79842639e-02 -6.46685719e-01 9.83396173e-02 -1.11901784e+00
-6.10301867e-02 4.95309472e-01 2.58750916e-01 -1.16769636e+00
-3.22600216e-01 -1.07227230e+00 -9.51086342e-01 1.50761902e+00
-9.01764810e-01 -7.42658496e-01 -8.09241235e-01 6.15543425e-01
4.78021532e-01 -4.95215058e-01 4.25332487e-01 3.07183921e-01
-4.54905182e-01 7.52047777e-01 -3.38079870e-01 -3.29822481e-01
8.06028664e-01 -1.33827615e+00 -8.24558511e-02 6.65417612e-01
-3.66627544e-01 9.69954550e-01 1.07281458e+00 -5.85098207e-01
-1.22338617e+00 2.02655867e-01 9.04174685e-01 -8.73909950e-01
1.21165596e-01 1.13752205e-02 -5.92703164e-01 7.18764842e-01
6.75009906e-01 -2.71648288e-01 6.67825699e-01 -1.05899140e-01
2.80388653e-01 3.45233202e-01 -1.34463096e+00 5.17418265e-01
4.45175111e-01 -3.67334813e-01 -3.88843626e-01 1.44647658e-01
4.28160504e-02 -6.84099317e-01 -4.16211873e-01 -1.77983850e-01
9.32033777e-01 -1.13229144e+00 3.01020980e-01 -8.05216968e-01
-5.01533687e-01 -5.22733450e-01 5.63966691e-01 -6.21519685e-01
-1.12666175e-01 -1.08090496e+00 -1.04182549e-01 1.10738051e+00
3.71527761e-01 -9.72194314e-01 6.84296787e-01 1.00848973e+00
2.64307022e-01 -2.66780376e-01 -2.82285899e-01 -1.21620715e-01
-7.75116801e-01 6.75703213e-02 5.73866725e-01 7.30147600e-01
5.47993004e-01 7.15788662e-01 -4.59771864e-02 9.80937108e-02
5.38085938e-01 -8.59033346e-01 7.86685228e-01 -1.71986675e+00
-5.47847059e-03 -2.81062782e-01 -1.03348494e+00 -2.27582052e-01
-1.57959357e-01 -9.79850963e-02 -4.11121577e-01 -1.04608464e+00
-3.07532251e-01 -1.42268524e-01 -6.65014982e-02 -1.75836369e-01
-2.45887399e-01 1.84461191e-01 1.45850345e-01 -8.41346756e-02
1.49169918e-02 -5.58187485e-01 8.17524254e-01 6.28362834e-01
-6.65809453e-01 1.75989177e-02 -5.09290874e-01 3.95916998e-01
4.69469726e-01 -3.03180456e-01 -5.83215773e-01 7.82623440e-02
3.67108494e-01 -4.00022417e-03 1.51361778e-01 -8.73147488e-01
4.35042113e-01 1.38944805e-01 4.75766987e-01 -7.09072232e-01
-1.95876122e-01 -7.68509567e-01 8.42780769e-02 3.41994435e-01
4.32876855e-01 6.36848688e-01 3.25083137e-01 3.29278558e-01
3.32811147e-01 -9.21145260e-01 6.87906384e-01 -1.34610891e-01
-8.26390982e-01 -3.63105275e-02 -5.42502463e-01 -4.02501225e-01
1.39098382e+00 -1.18951428e+00 -2.96139240e-01 -2.00645655e-01
-1.06180286e+00 2.02802122e-01 7.13668466e-01 4.25189316e-01
3.33224863e-01 -1.14105928e+00 -2.31520329e-02 3.24857026e-01
2.31970966e-01 -7.50811636e-01 1.99661851e-01 1.54341173e+00
-1.07861292e+00 7.24154115e-01 -7.65053928e-01 -5.19248009e-01
-2.28113437e+00 8.67034197e-01 3.52064997e-01 5.55789351e-01
-2.07812503e-01 6.69111848e-01 2.44943500e-02 2.87765831e-01
2.41582811e-01 -1.34164140e-01 -1.48835742e+00 -3.60988796e-01
9.63048935e-01 8.51858079e-01 -4.62605758e-03 -1.12327528e+00
-6.61681235e-01 1.16602683e+00 2.51114219e-01 -3.69098872e-01
4.23222601e-01 -6.12244606e-01 6.43638000e-02 4.75066632e-01
1.02840722e+00 1.23574376e-01 -3.41398656e-01 2.23297954e-01
3.71185899e-01 -5.73716760e-01 3.35367084e-01 -6.68642938e-01
-3.13640505e-01 9.83948469e-01 1.74104857e+00 3.38896304e-01
9.53089118e-01 -5.37656486e-01 2.98696309e-01 3.01411599e-01
3.01265746e-01 -1.28985393e+00 -6.66311562e-01 1.37821957e-01
6.67833745e-01 -1.11277950e+00 7.08612427e-02 1.15260251e-01
-5.53415835e-01 1.14095569e+00 8.11657846e-01 -4.44165897e-03
1.57532871e+00 1.14120416e-01 3.01376343e-01 -6.08072221e-01
-7.49895632e-01 -4.94941682e-01 4.97762829e-01 9.68744040e-01
1.14881814e+00 -2.04511702e-01 -1.00121498e+00 6.20214306e-02
-1.36861531e-02 7.41415799e-01 6.47390068e-01 9.28577125e-01
-6.30193055e-01 -6.89130902e-01 -1.20064247e+00 6.07055664e-01
-9.32338953e-01 5.38319163e-02 -3.84692609e-01 8.45005631e-01
1.65858045e-01 1.17910910e+00 3.16309214e-01 -4.23893631e-01
4.13271576e-01 5.00656605e-01 8.67340386e-01 -5.05194783e-01
-6.46712899e-01 -1.16304770e-01 -8.51587057e-02 -2.71738559e-01
-4.24012423e-01 -5.66689610e-01 -9.12530363e-01 -2.61270076e-01
-7.74053335e-01 -9.69365425e-03 1.27281702e+00 1.06549954e+00
3.19317788e-01 -5.69637120e-02 4.34345126e-01 -8.35693479e-01
1.86821058e-01 -1.49269378e+00 -6.88110292e-01 1.65775761e-01
5.71621537e-01 -1.24164128e+00 -8.81358609e-02 2.56309006e-02] | [13.95508861541748, 0.262911856174469] |
a5a5e628-b9e5-4f08-8892-ec9190be7bbd | role-of-language-relatedness-in-multilingual | 2109.10534 | null | https://arxiv.org/abs/2109.10534v1 | https://arxiv.org/pdf/2109.10534v1.pdf | Role of Language Relatedness in Multilingual Fine-tuning of Language Models: A Case Study in Indo-Aryan Languages | We explore the impact of leveraging the relatedness of languages that belong to the same family in NLP models using multilingual fine-tuning. We hypothesize and validate that multilingual fine-tuning of pre-trained language models can yield better performance on downstream NLP applications, compared to models fine-tuned on individual languages. A first of its kind detailed study is presented to track performance change as languages are added to a base language in a graded and greedy (in the sense of best boost of performance) manner; which reveals that careful selection of subset of related languages can significantly improve performance than utilizing all related languages. The Indo-Aryan (IA) language family is chosen for the study, the exact languages being Bengali, Gujarati, Hindi, Marathi, Oriya, Punjabi and Urdu. The script barrier is crossed by simple rule-based transliteration of the text of all languages to Devanagari. Experiments are performed on mBERT, IndicBERT, MuRIL and two RoBERTa-based LMs, the last two being pre-trained by us. Low resource languages, such as Oriya and Punjabi, are found to be the largest beneficiaries of multilingual fine-tuning. Textual Entailment, Entity Classification, Section Title Prediction, tasks of IndicGLUE and POS tagging form our test bed. Compared to monolingual fine tuning we get relative performance improvement of up to 150% in the downstream tasks. The surprise take-away is that for any language there is a particular combination of other languages which yields the best performance, and any additional language is in fact detrimental. | ['Pushpak Bhattacharyya', 'Karthik Sankaranarayanan', 'Samarth Bharadwaj', 'Rudra Murthy V', 'Tejas Indulal Dhamecha'] | 2021-09-22 | null | https://aclanthology.org/2021.emnlp-main.675 | https://aclanthology.org/2021.emnlp-main.675.pdf | emnlp-2021-11 | ['multiple-choice-qa', 'transliteration'] | ['natural-language-processing', 'natural-language-processing'] | [-4.36857522e-01 1.08859867e-01 -2.69039273e-01 -3.23482126e-01
-1.21972787e+00 -1.16129661e+00 6.36766315e-01 2.74012089e-01
-7.71311820e-01 1.10370922e+00 5.88257313e-01 -9.10646081e-01
-1.81369022e-01 -7.26927340e-01 -7.35939384e-01 -4.20940042e-01
8.13989565e-02 6.61107183e-01 9.63632986e-02 -6.96310341e-01
1.21040992e-01 5.07350922e-01 -4.68483716e-01 4.49776024e-01
9.61298704e-01 1.10135987e-01 1.00935459e-01 4.92646366e-01
-9.14767012e-02 7.56326020e-01 -3.19377452e-01 -7.51668692e-01
3.73745233e-01 -2.18402162e-01 -1.17926490e+00 -5.13569593e-01
3.57019037e-01 1.67188019e-01 -1.43096164e-01 5.33976793e-01
5.43197513e-01 -1.09994702e-01 6.61867797e-01 -4.54791546e-01
-5.54660618e-01 1.31296957e+00 -6.00549936e-01 4.80731755e-01
2.72376537e-01 2.51917690e-02 1.22049749e+00 -8.15536022e-01
9.34776425e-01 1.28887200e+00 9.16514456e-01 1.39250159e-01
-1.24315715e+00 -5.82752943e-01 -4.94138077e-02 -1.38532102e-01
-1.49929130e+00 -4.52812046e-01 8.10119659e-02 -3.88714910e-01
1.58483076e+00 7.17517212e-02 -3.05969100e-02 7.37508416e-01
2.88953155e-01 3.74910831e-01 1.39579260e+00 -8.10022354e-01
-3.31998587e-01 6.72295749e-01 -3.06533147e-02 7.12066412e-01
1.97530210e-01 -9.46471989e-02 -3.47839445e-01 -8.54108036e-02
1.45673886e-01 -5.48450351e-01 -4.37498018e-02 1.72366634e-01
-1.24069440e+00 9.22143161e-01 2.37159297e-01 7.79513657e-01
-2.57107556e-01 -2.32760534e-01 6.15950584e-01 5.29739201e-01
4.89720285e-01 7.09859192e-01 -1.16880834e+00 -1.18798271e-01
-9.76131082e-01 4.29486893e-02 1.09532762e+00 9.47942913e-01
8.12556386e-01 -6.97400793e-02 -1.21677987e-01 1.15157747e+00
1.48856983e-01 4.26325053e-01 6.49636388e-01 -6.24141276e-01
9.81639326e-01 4.27269131e-01 -2.70519525e-01 -3.82949471e-01
-3.60324204e-01 -5.06057203e-01 -3.30280602e-01 -3.72971803e-01
7.07731426e-01 -7.16830075e-01 -7.88232207e-01 1.59433782e+00
5.90463020e-02 -6.08200848e-01 5.14563382e-01 3.72399449e-01
4.50921029e-01 9.67875361e-01 2.53666967e-01 -3.18320096e-01
1.40597975e+00 -9.47833538e-01 -1.68117851e-01 -1.10231988e-01
1.14820933e+00 -1.26323903e+00 9.61522162e-01 2.08768278e-01
-1.10324860e+00 -4.58409220e-01 -8.33094358e-01 -1.37371778e-01
-5.51481366e-01 4.65670452e-02 5.26416600e-01 6.98693871e-01
-9.60850596e-01 5.67630768e-01 -5.17455995e-01 -7.92013109e-01
-5.00885919e-02 2.81556606e-01 -6.27919316e-01 -1.84348211e-01
-1.42904460e+00 1.31367183e+00 7.48047054e-01 -3.29178452e-01
-4.59261894e-01 -8.27404797e-01 -7.40536153e-01 -1.16953760e-01
5.84978871e-02 -2.29139760e-01 9.70780432e-01 -7.77553678e-01
-1.13929152e+00 1.08619463e+00 -7.38540441e-02 -3.87329608e-01
5.54833055e-01 -1.07944526e-01 -4.62223023e-01 -3.30006003e-01
4.72374290e-01 4.39033449e-01 2.05638632e-01 -7.35470414e-01
-1.00571728e+00 -1.95867136e-01 9.52035710e-02 2.67483890e-01
-3.04365247e-01 7.13813484e-01 -4.67785865e-01 -5.69932997e-01
-2.33737782e-01 -1.07554924e+00 -1.89155713e-01 -1.27501237e+00
-3.82251590e-01 -3.15437913e-01 2.46688500e-01 -1.12562203e+00
1.61227989e+00 -1.85471749e+00 -6.86770156e-02 3.59421700e-01
-3.44813138e-01 1.23160735e-01 -1.09307282e-01 8.52516055e-01
-2.04641521e-01 3.89209777e-01 1.11043893e-01 1.77979216e-01
-8.95759687e-02 4.05565351e-01 -1.58214137e-01 3.85295212e-01
3.55385542e-01 9.04490411e-01 -6.89013422e-01 -5.49940407e-01
-2.96126038e-01 1.76699281e-01 -3.88965040e-01 -2.42184818e-01
1.19760387e-01 3.96170914e-01 -2.67787993e-01 5.00185072e-01
3.29716384e-01 2.24091709e-01 4.89940524e-01 -5.75322136e-02
-4.67519581e-01 1.00130951e+00 -9.01509345e-01 1.35133755e+00
-8.30672920e-01 3.72162908e-01 -2.48398736e-01 -6.22548223e-01
7.46106863e-01 4.17489350e-01 1.00240491e-01 -7.38878369e-01
-1.49288848e-01 7.83169270e-01 5.92410028e-01 -4.66783196e-01
4.40978318e-01 -3.72672379e-01 -4.63483214e-01 4.84946340e-01
1.26997933e-01 -4.49555255e-02 6.66107357e-01 1.65889740e-01
9.51347232e-01 3.11004996e-01 7.05346763e-01 -7.21628010e-01
6.77215934e-01 3.24240714e-01 6.45277083e-01 5.88936329e-01
-5.10050505e-02 2.45263770e-01 4.42855418e-01 -7.40371421e-02
-1.31718409e+00 -9.01610374e-01 -4.55788851e-01 1.57627106e+00
-6.91146851e-01 -3.78315091e-01 -6.20318115e-01 -9.34540927e-01
-3.05651743e-02 9.41604137e-01 -2.59084195e-01 2.73787886e-01
-1.13101041e+00 -9.96456623e-01 1.08417654e+00 2.09775686e-01
1.71227366e-01 -1.18395317e+00 9.70403031e-02 3.86486530e-01
-2.26494327e-01 -1.15652800e+00 -3.91878545e-01 5.36368012e-01
-7.46872127e-01 -5.59900224e-01 -7.68295705e-01 -1.05515885e+00
2.79579520e-01 -4.28296089e-01 1.23323679e+00 -3.06920499e-01
1.55055404e-01 6.48760051e-02 -3.02447200e-01 -3.83471787e-01
-8.49551737e-01 6.64237022e-01 5.80434725e-02 -4.10914093e-01
5.19237816e-01 -3.88222843e-01 -1.44768180e-02 -1.32885613e-02
-5.03061414e-01 -3.40746701e-01 7.39113092e-01 5.20893276e-01
3.16872418e-01 -4.20709848e-02 6.53200388e-01 -1.44504964e+00
5.61183333e-01 -5.68478703e-01 -2.47908995e-01 3.95219654e-01
-6.47264302e-01 2.52736360e-01 6.19233847e-01 -1.81560099e-01
-1.35769057e+00 -1.31659776e-01 -3.12300384e-01 4.77427036e-01
-3.06896985e-01 8.31291616e-01 -2.52441049e-01 1.43771008e-01
8.57436121e-01 -5.31819128e-02 -6.70954704e-01 -6.50672495e-01
5.55356920e-01 6.43630505e-01 2.71595031e-01 -7.91230738e-01
7.97974169e-01 -6.67959005e-02 -2.98573941e-01 -8.45960975e-01
-7.67531633e-01 -5.27230084e-01 -9.12162662e-01 3.66332591e-01
7.21346498e-01 -1.08470726e+00 -1.51753038e-01 1.06123835e-01
-1.16753805e+00 -4.37543362e-01 -5.16360477e-02 6.23030484e-01
-1.01791240e-01 2.42781967e-01 -1.09925616e+00 -3.43744338e-01
-3.26742262e-01 -9.93102551e-01 6.60954118e-01 -1.28093082e-02
-6.23735666e-01 -1.39920974e+00 2.90962338e-01 4.26969767e-01
1.65228918e-01 -1.14705354e-01 1.49872911e+00 -1.22953403e+00
-1.21611163e-01 6.83463588e-02 -2.16222033e-01 4.57477689e-01
1.16969511e-01 -8.90197083e-02 -6.49983943e-01 -3.36906910e-01
-2.97741205e-01 -3.30175698e-01 5.94424367e-01 1.35911331e-01
3.05852666e-03 -4.05922174e-01 -2.20311731e-01 3.64276499e-01
1.65089607e+00 1.37340352e-01 3.29009593e-01 6.79900050e-01
7.56452382e-01 8.02047431e-01 5.71126938e-01 -2.52114180e-02
6.58883274e-01 4.61087584e-01 -5.03217399e-01 -8.72252584e-02
-8.09486061e-02 -2.57024884e-01 8.02788734e-01 1.21027589e+00
-1.04172260e-01 -3.73818204e-02 -1.32036328e+00 8.32552195e-01
-1.45459437e+00 -5.30919850e-01 -3.66348594e-01 2.09338975e+00
1.35490072e+00 2.30827838e-01 3.23078819e-02 -3.33595544e-01
3.16386670e-01 -2.33004257e-01 1.58450529e-01 -9.15669560e-01
-3.45295399e-01 6.33981824e-01 9.36569333e-01 8.67262602e-01
-9.29403901e-01 1.57344103e+00 5.54870081e+00 1.03066492e+00
-1.20473087e+00 3.86623085e-01 5.71490824e-01 -4.55437303e-02
-2.53078848e-01 3.84463012e-01 -1.32444072e+00 3.14798534e-01
1.46496665e+00 -1.31051570e-01 4.00728077e-01 3.91917557e-01
3.36262137e-01 2.50605661e-02 -1.12702179e+00 3.46383512e-01
-7.56914169e-02 -1.12222290e+00 -1.39777035e-01 1.54425323e-01
9.88397658e-01 8.33579719e-01 -3.23355556e-01 7.54375875e-01
6.97612464e-01 -9.45641279e-01 7.38496363e-01 7.70526156e-02
5.87554455e-01 -1.03370261e+00 8.73171329e-01 6.29576325e-01
-8.49495649e-01 1.21067157e-02 -3.86357635e-01 7.04131648e-02
1.68650821e-01 4.10997152e-01 -1.09738040e+00 7.08273590e-01
6.53823733e-01 2.54744500e-01 -6.93180919e-01 4.81555551e-01
-4.55233306e-01 1.17945838e+00 -3.08154225e-01 1.52368993e-01
6.04382575e-01 -1.80069342e-01 6.18607998e-01 1.99567962e+00
2.34627143e-01 -1.93331927e-01 1.79215476e-01 2.28858903e-01
-2.27867931e-01 8.10612142e-01 -6.66100442e-01 -5.49948998e-02
2.41198584e-01 1.32992399e+00 -5.15031993e-01 -3.42666864e-01
-6.13228619e-01 8.49569917e-01 6.12516165e-01 3.09949934e-01
-6.41051531e-01 -5.24412036e-01 3.06621253e-01 2.28593722e-01
4.22066152e-01 -2.26979390e-01 -3.16025794e-01 -9.33484554e-01
-2.15094760e-01 -1.19679940e+00 6.53249741e-01 -3.42653066e-01
-1.26560593e+00 6.86041653e-01 -7.28725120e-02 -6.01882100e-01
-5.00697672e-01 -6.50563300e-01 -2.65364915e-01 1.44536161e+00
-1.56298757e+00 -1.42659521e+00 7.57610500e-01 3.69238436e-01
4.34202105e-01 -4.38685656e-01 7.73591399e-01 7.27229714e-01
-4.03501838e-01 9.16451752e-01 3.00543457e-01 4.01105970e-01
1.00970936e+00 -1.28156257e+00 2.79066801e-01 9.68047380e-01
4.30421203e-01 8.61822784e-01 5.58486938e-01 -7.27579951e-01
-8.10475588e-01 -1.17399645e+00 1.97228599e+00 -7.15314746e-01
1.12740207e+00 -2.69643843e-01 -8.59420776e-01 9.82647955e-01
5.13733864e-01 -5.49641252e-01 6.73241615e-01 4.99563366e-01
-3.65394026e-01 -2.68750861e-02 -8.19353163e-01 5.40664732e-01
6.47689998e-01 -6.27149284e-01 -7.42779195e-01 4.93255138e-01
6.66812241e-01 -2.53395259e-01 -1.24348783e+00 1.41783148e-01
4.08910215e-01 -6.35370314e-01 7.37637937e-01 -9.67434049e-01
5.90960860e-01 -6.01477548e-02 -2.19769657e-01 -1.50717723e+00
-4.29936916e-01 -6.10770822e-01 5.92511594e-01 1.85300291e+00
1.22645116e+00 -7.12312281e-01 4.04904455e-01 1.30362377e-01
-4.25394066e-02 -5.19689858e-01 -6.11147881e-01 -8.93049002e-01
8.63623619e-01 -4.21713084e-01 2.18946517e-01 1.15815520e+00
6.50121272e-02 1.06237936e+00 -3.26273814e-02 6.55297861e-02
1.77732244e-01 -4.30672653e-02 5.90198874e-01 -8.10406566e-01
-5.88182688e-01 -3.81547719e-01 -6.08072914e-02 -5.47434866e-01
2.65175343e-01 -1.38702333e+00 -8.98059551e-03 -1.43425465e+00
1.64391667e-01 -8.11161876e-01 -2.68795729e-01 6.92772329e-01
-1.13397658e-01 3.70071113e-01 3.29750180e-01 2.05876231e-01
-1.80442885e-01 -2.73644358e-01 8.32609236e-01 1.33370236e-01
-3.72465223e-01 -5.94136901e-02 -1.01389205e+00 8.11129749e-01
7.19351411e-01 -6.92361057e-01 2.89519429e-02 -6.17185354e-01
4.26789254e-01 -4.07220982e-02 -3.63021195e-01 -4.49339211e-01
2.87831537e-02 -2.32485458e-01 3.08721781e-01 -1.69276103e-01
-1.81487545e-01 -5.74073017e-01 -8.06328431e-02 4.94442463e-01
-3.43126893e-01 3.18306863e-01 4.42602485e-01 -1.21021986e-01
-1.82327107e-01 -5.40884018e-01 8.28020871e-01 -4.00089413e-01
-6.71599030e-01 1.64783746e-02 -4.94736373e-01 4.54198927e-01
8.81466091e-01 -4.97770589e-03 9.46504809e-03 -1.19600467e-01
-5.36860764e-01 1.21676117e-01 1.85945004e-01 2.75390625e-01
-2.65487254e-01 -8.95335257e-01 -1.32516408e+00 1.12959649e-02
1.49767235e-01 -4.97621238e-01 -2.34735817e-01 1.07878399e+00
-6.74416244e-01 8.61118913e-01 -8.55921861e-03 -1.70782313e-01
-1.23825610e+00 2.22403720e-01 2.08010212e-01 -1.02353442e+00
-1.41793206e-01 8.65653753e-01 1.73862040e-01 -1.05591881e+00
-2.04554796e-01 -2.35449165e-01 -2.04179615e-01 1.96404412e-01
-5.59119135e-02 3.05910528e-01 4.64822292e-01 -9.42836583e-01
-4.52678800e-01 3.01127076e-01 -3.22113901e-01 -2.84323394e-01
1.32224548e+00 -2.44382054e-01 -4.32006478e-01 4.30397362e-01
1.19299579e+00 9.45763230e-01 -4.58652198e-01 -2.94664472e-01
4.74560559e-01 1.87475905e-01 -1.97004497e-01 -1.24109280e+00
-5.37704289e-01 8.01557124e-01 4.77440916e-02 -1.04849413e-01
7.60625005e-01 1.87905788e-01 7.44528592e-01 3.54284108e-01
3.40480924e-01 -1.20911121e+00 -7.10934937e-01 1.05490232e+00
6.15757704e-01 -9.94438767e-01 -1.29760932e-02 -6.50095046e-02
-7.51975000e-01 8.95435512e-01 3.41588736e-01 -5.76080270e-02
4.96372581e-01 2.43250221e-01 2.87936956e-01 8.48703310e-02
-7.40784824e-01 -2.04279259e-01 2.93052107e-01 2.27511063e-01
1.06198788e+00 2.68754452e-01 -7.20515072e-01 4.97362226e-01
-6.38720214e-01 -4.91565496e-01 3.07476699e-01 6.62850499e-01
-7.67597780e-02 -1.40274251e+00 -3.66194874e-01 4.07292724e-01
-1.19508755e+00 -7.26135910e-01 -2.76138246e-01 1.31871188e+00
4.64070827e-01 6.94935560e-01 -3.88925592e-03 1.53110087e-01
3.04566830e-01 3.47133398e-01 5.00488937e-01 -9.50983405e-01
-1.21746075e+00 2.84139693e-01 6.01460099e-01 4.31534201e-02
1.52570866e-02 -7.10273743e-01 -1.38192773e+00 -4.54207808e-01
-1.92981690e-01 4.51304674e-01 5.58830023e-01 1.22908747e+00
-2.47172290e-03 2.45911896e-01 3.59994739e-01 -1.70356527e-01
-5.65812886e-01 -1.18509221e+00 -4.29124117e-01 2.43702933e-01
-9.36492160e-02 -1.72845237e-02 -1.55652016e-01 1.60681471e-01] | [10.831401824951172, 10.04738998413086] |
1f32563f-07c7-44ed-bf48-8bc8acbf167a | infrared-and-visible-image-fusion-using | 1804.08992 | null | https://arxiv.org/abs/1804.08992v5 | https://arxiv.org/pdf/1804.08992v5.pdf | Infrared and visible image fusion using Latent Low-Rank Representation | Infrared and visible image fusion is an important problem in the field of image fusion which has been applied widely in many fields. To better preserve the useful information from source images, in this paper, we propose a novel image fusion method based on latent low-rank representation(LatLRR) which is simple and effective. Firstly, the source images are decomposed into low-rank parts(global structure) and salient parts(local structure) by LatLRR. Then, the low-rank parts are fused by weighted-average strategy to preserve more contour information. Then, the salient parts are simply fused by sum strategy which is a efficient operation in this fusion framework. Finally, the fused image is obtained by combining the fused low-rank part and the fused salient part. Compared with other fusion methods experimentally, the proposed method has better fusion performance than state-of-the-art fusion methods in both subjective and objective evaluation. The Code of our fusion method is available at https://github.com/hli1221/imagefusion\_Infrared\_visible\_latlrr | ['Xiao-Jun Wu', 'Hui Li'] | 2018-04-24 | null | null | null | null | ['infrared-and-visible-image-fusion'] | ['computer-vision'] | [ 2.22005978e-01 -5.50697446e-01 -7.31895342e-02 -3.18246372e-02
-9.99805689e-01 -1.59904495e-01 2.57072717e-01 5.09711541e-02
-2.14841485e-01 5.38415253e-01 6.31934047e-01 8.92967284e-02
-1.90573037e-01 -7.35337734e-01 -1.02817036e-01 -1.01737201e+00
4.22026634e-01 -3.76386702e-01 3.74591947e-01 -3.56508255e-01
-2.60955351e-03 2.97805369e-01 -1.66185486e+00 2.20786467e-01
1.20053661e+00 1.02529383e+00 4.75122571e-01 3.16938192e-01
-5.09015061e-02 7.58075118e-01 -3.40773642e-01 1.00615531e-01
3.62842977e-01 -6.38673842e-01 -5.01697719e-01 2.60008633e-01
1.01411797e-01 -1.36460260e-01 -3.47372621e-01 1.66690707e+00
5.86052060e-01 4.41419184e-01 4.90589470e-01 -9.34764683e-01
-7.56284475e-01 2.50666767e-01 -1.17585278e+00 4.02572960e-01
4.43413556e-01 -3.18555027e-01 5.91176689e-01 -1.00273931e+00
2.10900247e-01 1.50767684e+00 1.73792616e-01 -1.36439167e-02
-9.56904829e-01 -7.74705470e-01 8.30686092e-02 3.29851627e-01
-1.49014986e+00 -3.75745654e-01 8.90861869e-01 -2.86174387e-01
5.58225773e-02 4.02777642e-01 3.17957878e-01 2.03070581e-01
5.21490932e-01 8.39913130e-01 1.43664527e+00 -4.30528104e-01
-1.30110025e-01 -3.01293969e-01 1.54902652e-01 7.72480249e-01
3.90386313e-01 1.12772159e-01 -4.11576152e-01 -1.07842550e-01
6.51846111e-01 7.07345486e-01 -7.11107314e-01 1.20236784e-01
-1.36325276e+00 5.32370687e-01 7.05992997e-01 5.95986128e-01
-6.20135248e-01 -2.42684484e-01 4.43172865e-02 1.35617435e-01
7.56647766e-01 -3.82146925e-01 4.36981060e-02 5.65683544e-01
-1.02040136e+00 4.81510945e-02 1.19796693e-01 3.09564620e-01
1.05870044e+00 -7.24751428e-02 -3.36445749e-01 1.06949949e+00
7.46642530e-01 6.59740925e-01 4.61348772e-01 -1.09868205e+00
3.19876552e-01 6.52831733e-01 8.08537304e-02 -1.17790985e+00
-1.23077616e-01 -3.12489510e-01 -1.19933152e+00 5.54543376e-01
-4.14692275e-02 -7.00398833e-02 -9.36379611e-01 1.30865514e+00
2.21558884e-01 1.79591283e-01 3.27281445e-01 1.06202960e+00
1.24213743e+00 1.09804308e+00 4.65297662e-02 -5.95334053e-01
1.67258024e+00 -9.44007218e-01 -1.07359540e+00 -6.65707290e-02
-8.00845399e-02 -1.37057972e+00 4.84744370e-01 3.07676047e-01
-1.19559252e+00 -9.09177959e-01 -1.04910409e+00 -1.77720070e-01
-1.01625226e-01 3.43096495e-01 4.14379656e-01 1.57706201e-01
-1.00618422e+00 2.37976998e-01 -3.80358160e-01 -2.67656833e-01
7.04477578e-02 5.04585057e-02 -5.28233767e-01 -5.63731849e-01
-1.23752773e+00 7.27749109e-01 5.11830986e-01 2.65616000e-01
-3.66866231e-01 -2.42954403e-01 -9.06240940e-01 -8.22526682e-03
4.94694173e-01 -6.08687282e-01 9.88040030e-01 -6.20160222e-01
-1.24718761e+00 5.71511507e-01 -5.31503916e-01 1.57465041e-01
-2.64757704e-02 -2.31575519e-01 -5.58830440e-01 3.64439905e-01
2.27055773e-01 4.16924000e-01 8.30913484e-01 -1.49368405e+00
-7.77894735e-01 -2.83905178e-01 -2.67399758e-01 5.33199906e-01
-1.52797541e-02 1.69159979e-01 -5.27094007e-01 -9.18290913e-01
6.74043417e-01 -5.30802608e-01 -2.38958687e-01 -4.31353636e-02
5.71170859e-02 -2.73594618e-01 9.08581674e-01 -1.10821497e+00
1.25461185e+00 -2.60190916e+00 6.38326332e-02 8.13216195e-02
2.28791192e-01 3.21499258e-01 -7.74738342e-02 3.15117031e-01
-2.68062234e-01 -7.21441358e-02 -3.10586244e-01 -1.78670466e-01
-4.46331680e-01 -1.39983431e-01 -2.41771996e-01 5.60136557e-01
-2.82199532e-01 6.52167380e-01 -9.22115982e-01 -8.61470401e-01
6.34051859e-01 6.86380088e-01 2.12236553e-01 2.06726059e-01
3.28957289e-01 6.47294641e-01 -4.95074630e-01 8.68879437e-01
1.05919766e+00 -6.07292689e-02 -3.69670659e-01 -7.37389863e-01
-3.66474897e-01 -3.80652875e-01 -1.29594195e+00 1.56209958e+00
-1.65923968e-01 2.27225542e-01 3.42716247e-01 -7.24380612e-01
1.27495468e+00 4.87897098e-01 6.97538912e-01 -6.09961867e-01
2.72739857e-01 1.56377628e-01 -4.24740732e-01 -2.94817686e-01
5.48376024e-01 -2.84651339e-01 9.83404517e-02 2.02953830e-01
-1.44364983e-01 -6.49648532e-02 2.31314659e-01 2.12430164e-01
5.54251552e-01 9.78097394e-02 6.19652212e-01 -1.47130117e-01
1.02645135e+00 -3.25111836e-01 8.67229581e-01 1.59924924e-01
-2.72682339e-01 7.85458148e-01 -2.20539019e-01 -1.20089605e-01
-5.35350978e-01 -1.12058270e+00 1.17544279e-01 8.37928712e-01
7.00788915e-01 -2.36483067e-01 -2.48779535e-01 -1.74995184e-01
-2.23032132e-01 3.28376025e-01 -2.56838530e-01 -1.45576552e-01
-2.22694054e-01 -6.85098648e-01 -7.94172511e-02 2.27395952e-01
1.11204576e+00 -9.39935386e-01 -1.60792768e-01 8.88519958e-02
-7.02948630e-01 -7.22856462e-01 -6.33156955e-01 -4.14394975e-01
-8.37142527e-01 -8.03941250e-01 -1.17503762e+00 -6.89873636e-01
7.47727633e-01 1.29899883e+00 6.54151976e-01 3.17840248e-01
-7.82873556e-02 3.20022441e-02 -6.66546762e-01 -9.50623304e-02
-5.16474538e-04 -6.29742801e-01 -2.02879488e-01 3.36849034e-01
4.65420866e-03 -2.97035277e-01 -7.64210820e-01 3.44228297e-01
-1.16587245e+00 2.53708988e-01 9.00572777e-01 5.88084221e-01
8.97333384e-01 8.38026643e-01 6.99821711e-02 -1.30731523e-01
5.82560182e-01 -2.41418391e-01 -6.56351805e-01 3.89907420e-01
-1.51667044e-01 -1.18490703e-01 2.27129191e-01 6.18283218e-03
-1.54049814e+00 6.34051766e-03 5.95193431e-02 -5.22925615e-01
-3.86678465e-02 6.79377139e-01 -2.81665683e-01 -1.36601776e-01
2.87221223e-01 3.53388160e-01 9.24658328e-02 -7.92218149e-01
3.37916702e-01 6.69263780e-01 8.68660212e-01 -4.25145060e-01
8.73345077e-01 6.65999591e-01 4.90767285e-02 -6.43333137e-01
-1.08877361e+00 -7.70959437e-01 -3.94086152e-01 -2.87111908e-01
1.10453594e+00 -1.32455790e+00 -5.58643162e-01 5.77263057e-01
-9.04594541e-01 2.66702622e-01 -1.18084177e-01 8.82223010e-01
-7.66605735e-02 7.53617346e-01 -5.72867632e-01 -9.16502357e-01
-5.15902698e-01 -1.13599396e+00 1.06290674e+00 8.46431851e-01
4.54626173e-01 -7.17096329e-01 -3.92144471e-02 5.05597115e-01
2.85164803e-01 3.55085641e-01 3.57132375e-01 1.55091301e-01
-3.96392852e-01 -8.99096765e-03 -5.76391876e-01 4.68687087e-01
5.67074358e-01 -6.16553240e-03 -6.69343531e-01 -3.85500729e-01
2.67293721e-01 5.60449064e-02 1.31643510e+00 5.38795114e-01
8.42664897e-01 -5.45947291e-02 -2.64281064e-01 4.17341024e-01
1.54482007e+00 2.60103077e-01 8.04250240e-01 1.00619279e-01
5.09863913e-01 4.66340005e-01 1.01918101e+00 2.02817157e-01
3.23131889e-01 4.98789161e-01 1.86821058e-01 -6.84078753e-01
-2.14021161e-01 6.47039115e-02 4.96437341e-01 8.94678116e-01
-3.78137022e-01 2.75572669e-02 -6.08488083e-01 2.26682693e-01
-2.17273307e+00 -1.22945631e+00 -1.78795308e-01 2.15568233e+00
6.39012456e-01 -2.64258474e-01 -2.00636476e-01 2.81684935e-01
1.10956430e+00 4.27248478e-01 -5.92091121e-02 1.07589498e-01
-3.19179446e-01 3.73960063e-02 4.48995143e-01 6.28350556e-01
-1.13470817e+00 5.21463394e-01 5.45785379e+00 1.08028626e+00
-8.64828408e-01 2.80379027e-01 4.60327476e-01 3.91715914e-01
-2.84756720e-01 1.60262927e-01 -5.47928274e-01 5.77878118e-01
3.15369964e-01 -3.13828766e-01 2.17863306e-01 2.45960146e-01
2.88381547e-01 -7.02976108e-01 -8.95232931e-02 1.18493032e+00
1.71183452e-01 -1.01967740e+00 -6.51356950e-02 -1.10352570e-02
7.67488122e-01 -2.59804189e-01 4.93456610e-02 -8.37829858e-02
3.76043975e-01 -7.46107936e-01 4.74698007e-01 1.01021504e+00
6.38410330e-01 -8.13594103e-01 1.11972153e+00 2.57861972e-01
-1.81947863e+00 -4.56847548e-02 -4.64638829e-01 3.31930481e-02
3.53022903e-01 1.18067229e+00 2.77898610e-01 1.36804557e+00
8.72306764e-01 8.82325172e-01 -4.94795531e-01 1.16164267e+00
-3.99362594e-01 1.90441146e-01 -1.26274422e-01 7.45745420e-01
-1.37917995e-01 -6.17179573e-01 4.63150263e-01 8.25945139e-01
5.23446202e-01 6.78401589e-01 5.75213790e-01 4.38602954e-01
1.11014284e-01 1.58188447e-01 -3.99620146e-01 1.86626196e-01
2.43040293e-01 1.56762695e+00 -7.82137930e-01 -5.09206951e-01
-5.38181484e-01 8.20292652e-01 -3.43801796e-01 3.80659878e-01
-4.74383712e-01 -3.64864379e-01 1.72493652e-01 -2.82841057e-01
1.05402060e-01 -2.62263417e-01 -5.01692332e-02 -1.33613944e+00
-7.29146376e-02 -7.45204687e-01 7.06352353e-01 -1.18812943e+00
-1.19754064e+00 4.85219747e-01 1.66371807e-01 -1.67236173e+00
3.38261485e-01 -1.95714340e-01 -6.21580958e-01 1.23997355e+00
-1.66961884e+00 -1.20427728e+00 -7.57768035e-01 8.20739210e-01
3.50962579e-01 -7.29873553e-02 5.68206131e-01 1.63917467e-01
-3.12128901e-01 1.69864260e-02 1.89544573e-01 -1.39799491e-01
8.41807127e-01 -6.83740020e-01 -3.29356432e-01 1.24440205e+00
-1.49737582e-01 5.02221286e-01 4.38725889e-01 -8.11651409e-01
-8.50592673e-01 -8.65902603e-01 7.04538822e-01 4.13517207e-01
2.12298438e-01 2.14851111e-01 -1.05483556e+00 1.16905861e-01
5.39827883e-01 2.17129320e-01 5.69522679e-01 -4.01537210e-01
-2.25305736e-01 -3.47028643e-01 -1.26803386e+00 4.70523238e-01
4.79212403e-01 -3.32527995e-01 -7.36585677e-01 4.92493547e-02
6.87571585e-01 -1.90412298e-01 -9.94141459e-01 8.02352726e-01
3.16060215e-01 -1.03295314e+00 1.00912547e+00 6.02685213e-01
1.49468482e-01 -1.26379573e+00 -3.13997120e-01 -1.05839896e+00
-7.49267578e-01 -4.02467012e-01 3.03951949e-02 1.35889626e+00
-2.17907533e-01 -7.13685811e-01 9.77447927e-02 5.30297905e-02
-4.93228287e-02 -4.74470615e-01 -6.55325592e-01 -4.95475531e-01
-5.32932997e-01 3.31336677e-01 3.65935117e-01 6.33450091e-01
-2.34311879e-01 2.41725907e-01 -4.01005358e-01 2.94696480e-01
8.72845292e-01 4.11435366e-01 3.13785374e-01 -1.14526534e+00
3.06781512e-02 -2.54384011e-01 -2.69990116e-01 -7.59135067e-01
-9.18616727e-02 -7.07946122e-01 7.02148527e-02 -2.04767704e+00
5.80727637e-01 2.38682870e-02 -7.51977682e-01 7.07513630e-01
-6.22799456e-01 5.30204475e-01 4.41814274e-01 6.82547927e-01
-5.48693895e-01 5.67713797e-01 1.40759635e+00 -1.63777560e-01
-1.26166493e-01 -1.40613556e-01 -9.40455854e-01 6.62870109e-01
7.80559063e-01 -1.05662614e-01 -1.72319934e-01 -1.42875463e-01
-1.48639575e-01 2.84743726e-01 3.91311646e-01 -1.22205865e+00
3.54655325e-01 -2.99467891e-01 6.09170020e-01 -9.64305520e-01
3.43796402e-01 -8.67138088e-01 4.35171515e-01 4.52284932e-01
1.24533579e-01 -9.50457305e-02 3.87266725e-02 3.05067956e-01
-7.10812211e-01 -1.25571832e-01 9.61347640e-01 -1.96171060e-01
-9.54179645e-01 3.14176291e-01 -1.76637217e-01 -6.16678476e-01
9.34796691e-01 -1.03859700e-01 -5.49148142e-01 -4.98146504e-01
-6.04725957e-01 1.97479084e-01 6.24605060e-01 3.66329163e-01
8.39122117e-01 -1.53949082e+00 -1.01212525e+00 7.38429055e-02
1.22325152e-01 -3.47970240e-02 6.68294609e-01 9.26744938e-01
-3.66056293e-01 5.82342185e-02 -2.48148903e-01 -3.95208389e-01
-1.66383791e+00 6.99568927e-01 -2.95261592e-02 -3.53965729e-01
-5.15679717e-01 5.53944290e-01 5.02090752e-01 1.11283958e-01
-1.87640190e-01 8.28671083e-02 -4.67556298e-01 4.52141128e-02
7.99321413e-01 4.39860106e-01 -2.56107956e-01 -1.25140989e+00
-4.05261040e-01 1.08121133e+00 1.36205316e-01 -1.47791356e-01
1.12651181e+00 -4.86785412e-01 -6.86458170e-01 2.04031631e-01
1.11902571e+00 1.92654699e-01 -9.93221104e-01 -4.60588545e-01
-5.45361757e-01 -7.25546539e-01 3.65677655e-01 -5.37317634e-01
-1.28978348e+00 7.05875278e-01 9.41244304e-01 -6.85947314e-02
1.70511961e+00 4.04203534e-02 6.35940909e-01 -1.54536322e-01
3.81506443e-01 -6.40850246e-01 8.65271464e-02 2.03807503e-01
1.04616880e+00 -1.17274046e+00 6.28122389e-01 -6.31720722e-01
-4.46161658e-01 8.30406487e-01 1.85392290e-01 -1.38690799e-01
7.47817099e-01 -1.71701307e-03 2.74160951e-01 -5.52858189e-02
-3.44857246e-01 -6.61492705e-01 6.24807775e-01 3.77292514e-01
4.48639452e-01 -8.44263732e-02 -6.77436709e-01 2.10926831e-01
3.03861052e-01 1.66747883e-01 1.79948047e-01 9.46728885e-01
-8.69404256e-01 -1.11922216e+00 -1.04587018e+00 2.98216760e-01
-6.44280195e-01 -1.23759862e-02 1.32812917e-01 1.31926477e-01
4.33161229e-01 1.45232964e+00 -2.32033148e-01 -4.96573985e-01
6.13527223e-02 -2.72720367e-01 3.18843037e-01 -4.08196002e-01
-1.21396609e-01 6.14349246e-01 -4.19697464e-01 -5.10699213e-01
-1.05796587e+00 -4.66616780e-01 -1.22794724e+00 -1.31213531e-01
-4.20706689e-01 3.26222003e-01 3.88183564e-01 6.99598491e-01
1.20758124e-01 5.24186850e-01 7.24765778e-01 -1.12635481e+00
1.10294633e-01 -8.41895461e-01 -8.08363438e-01 3.86792451e-01
3.92330229e-01 -7.33299613e-01 -4.05603141e-01 3.39889854e-01] | [10.538492202758789, -1.9445650577545166] |
191d3652-7573-46a6-b48f-53e1e33b2d9a | performance-accuration-method-of-machine | null | null | https://iocscience.org/ejournal/index.php/mantik/article/view/725 | https://iocscience.org/ejournal/index.php/mantik/article/view/725/482 | Performance Accuration Method of Machine Learning for Diabetes Prediction | Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning (ML) techniques allow us to obtain predictively, the dataset we are testing is pima-indian-diabetes with a dataset of 765 raw data with 8 data features and 1 data label we developed a method to achieve the best accuracy from the 5 methods we use with the stages of separation training and testing the dataset, scaling features, parameters evaluation, confusion matrix and we get the accuracy of each method, and the results of the accuracy we get with these 5 methods Gradient-boosting is best with an accuracy score of 0.8, Decision Tree 0.72, Random Forest 0.72, next is Logistic Regression 0.7, and then followed by K-NN method with a score of 0.65. | ['ALI Murtadho', 'Dwi Harini Sulistyawati'] | 2020-05-01 | null | null | null | jurnal-mantik-2020-5 | ['diabetes-prediction', 'automatic-machine-learning-model-selection'] | ['medical', 'methodology'] | [ 7.11980835e-02 1.63778916e-01 -4.66948330e-01 -7.66344130e-01
-4.15942460e-01 -2.99548358e-01 7.39196062e-01 1.46272138e-01
-3.45183253e-01 1.10178685e+00 2.74921060e-02 -7.03904331e-01
-6.63444579e-01 -6.73530936e-01 -3.18860084e-01 -6.78218305e-01
-4.32111919e-01 9.03039753e-01 3.37784104e-02 -1.06260061e-01
4.76222724e-01 4.74129677e-01 -1.56339061e+00 7.26137102e-01
8.19978714e-01 1.28194809e+00 -2.99160540e-01 9.73836124e-01
-1.89779371e-01 1.30769384e+00 -3.34206134e-01 -1.36047795e-01
3.48303586e-01 -2.46115744e-01 -1.04732621e+00 -1.61367968e-01
-9.69476718e-03 -1.79746136e-01 1.96638808e-01 4.12766188e-01
8.53432789e-02 -2.61042297e-01 1.03479183e+00 -1.64089727e+00
-4.48041260e-01 3.33378077e-01 -2.87851721e-01 1.96030252e-02
4.77251917e-01 2.03650892e-02 3.71888638e-01 -7.32451379e-01
4.51392680e-01 1.31630790e+00 1.03007317e+00 5.23902714e-01
-1.35286129e+00 -7.81012893e-01 -4.61339086e-01 4.96760368e-01
-1.26009679e+00 -2.33090639e-01 3.84449393e-01 -8.50045264e-01
1.02365839e+00 3.27327639e-01 8.22805166e-01 5.95791638e-01
2.99560338e-01 5.51614642e-01 1.71016574e+00 -8.54436517e-01
2.97044098e-01 8.53156865e-01 4.63315129e-01 7.00626373e-01
4.50688712e-02 3.40227872e-01 -1.67213336e-01 -2.19260991e-01
3.84467721e-01 2.88847297e-01 4.13057595e-01 -2.39456668e-01
-8.78314435e-01 1.08455479e+00 4.19733554e-01 3.60894412e-01
-4.96573746e-01 -2.01637939e-01 2.69702733e-01 6.96190238e-01
1.00119524e-01 5.76174676e-01 -1.15805483e+00 7.20822439e-02
-8.53924692e-01 3.22918028e-01 8.52075040e-01 4.91407275e-01
7.18761265e-01 -1.51512161e-01 4.77031581e-02 9.55666184e-01
2.68701553e-01 2.72031337e-01 8.35197806e-01 -8.15834224e-01
-7.08606839e-02 1.21363449e+00 1.85030717e-02 -4.65609103e-01
-6.86072290e-01 -1.70728818e-01 -6.69447720e-01 6.92137897e-01
3.95272434e-01 -4.04506594e-01 -1.17382252e+00 1.14606178e+00
5.91641068e-02 -1.27364695e-01 4.24003035e-01 1.86579481e-01
8.34618509e-01 5.69212317e-01 1.95330396e-01 -3.90072137e-01
1.04365480e+00 -9.34338808e-01 -5.47938704e-01 2.16405764e-02
9.67479885e-01 -6.56427026e-01 6.94685161e-01 9.70811605e-01
-6.24095678e-01 -6.94702983e-01 -1.11251378e+00 4.36940551e-01
-7.87335336e-01 1.18498534e-01 1.04510725e+00 8.21507812e-01
-1.14459920e+00 6.29170835e-01 -5.54539919e-01 -3.19023341e-01
3.40857208e-01 6.91938341e-01 -5.93108416e-01 -2.00499967e-01
-8.00275564e-01 1.47118270e+00 7.42508471e-01 -3.84190381e-01
-4.28938687e-01 -5.11245668e-01 -3.02642375e-01 -4.58088100e-01
-1.43943474e-01 -5.45560896e-01 1.04017472e+00 -1.37987661e+00
-1.24981964e+00 7.70241380e-01 -5.39718270e-02 -5.90009868e-01
4.52438951e-01 -1.77595466e-01 -6.63896739e-01 -2.17581168e-01
-5.42581081e-02 7.43627310e-01 2.05306008e-01 -1.33845139e+00
-1.08496141e+00 -6.59220934e-01 -5.06329954e-01 -1.99867666e-01
5.28863352e-03 2.18353167e-01 4.08867240e-01 -1.66851223e-01
2.34612167e-01 -9.50227380e-01 -3.47875297e-01 -4.68632430e-01
5.46038039e-02 -2.92940587e-01 8.17203164e-01 -9.15296137e-01
1.29124832e+00 -1.75633323e+00 -3.24810505e-01 3.92271042e-01
5.98887093e-02 2.58622944e-01 3.31469506e-01 2.20685139e-01
-3.99391532e-01 1.61040783e-01 -9.33998898e-02 4.80387688e-01
-4.85459507e-01 2.70250589e-01 1.49172261e-01 -1.35050684e-01
4.25856680e-01 5.61822593e-01 -5.22723436e-01 -5.82728386e-01
2.67773271e-01 1.85043007e-01 -5.52062213e-01 3.48062813e-01
1.39301732e-01 1.91386804e-01 -4.32445884e-01 8.05823684e-01
5.00210464e-01 -9.05806646e-02 2.08878040e-01 1.47717983e-01
3.54945287e-02 -1.85751349e-01 -1.13843596e+00 1.06155849e+00
-2.82123089e-01 5.36525011e-01 -3.57748955e-01 -1.16422284e+00
1.45300198e+00 2.98821151e-01 5.71303904e-01 -4.35442567e-01
8.58948827e-02 2.46944427e-01 4.56966758e-01 -8.43449354e-01
-3.73655200e-01 -1.57751068e-01 1.91169307e-01 3.47860783e-01
1.24209709e-01 -6.70718029e-02 -7.84869641e-02 5.08229248e-02
1.28569245e+00 2.69981399e-02 6.56392872e-01 -2.88492709e-01
5.02067506e-01 4.95797187e-01 3.60136777e-01 7.10323870e-01
-1.84311196e-01 1.17443688e-01 5.22576988e-01 -1.11426640e+00
-9.98536885e-01 -8.86704385e-01 -5.40397227e-01 1.07113647e+00
-7.96011686e-01 -1.11332722e-01 -5.06862402e-01 -7.38486886e-01
2.58884251e-01 9.42241490e-01 -8.02541494e-01 -6.48497418e-02
-2.66854793e-01 -1.12031698e+00 2.08747819e-01 4.10375863e-01
5.76543868e-01 -1.11671400e+00 -2.31608167e-01 2.93833554e-01
3.42437088e-01 -5.59498847e-01 6.97634637e-01 7.29278803e-01
-1.33642697e+00 -9.15280282e-01 4.33699526e-02 -8.59634042e-01
4.83692527e-01 -4.23695117e-01 1.20787203e+00 1.66852400e-01
-4.78359818e-01 -1.32643953e-01 -4.09736425e-01 -7.58985519e-01
-7.21370637e-01 -1.09101839e-01 3.46349627e-02 -2.83287317e-01
7.87580132e-01 -7.10301936e-01 -4.14657623e-01 1.86337333e-03
-5.48135519e-01 2.22704902e-01 1.01708817e+00 8.75994265e-01
1.93242416e-01 2.96528451e-02 9.11870420e-01 -1.24182677e+00
4.70757395e-01 -6.50426447e-01 -1.61945403e-01 4.21281427e-01
-1.38655317e+00 2.42262464e-02 4.73500937e-01 -3.22446316e-01
-6.37334466e-01 5.30815840e-01 -1.34388562e-02 2.13572979e-01
-4.95352745e-01 3.44626784e-01 6.51425275e-04 -7.20991790e-02
1.10999835e+00 -5.34282364e-02 2.23493919e-01 -6.16990507e-01
1.16824836e-01 1.24596870e+00 3.30139995e-01 -2.74059594e-01
2.47390255e-01 -1.91103995e-01 1.98919768e-03 -2.58793443e-01
-7.37189353e-01 -1.42073527e-01 -9.29426730e-01 -2.00145066e-01
6.91823959e-01 -6.43946826e-01 -5.98217130e-01 3.49277228e-01
-5.59803545e-01 -1.11194186e-01 -6.40243664e-02 7.03763247e-01
-5.15075445e-01 -4.03574377e-01 -4.56488818e-01 -1.09853959e+00
-4.91639405e-01 -8.98886740e-01 3.38710815e-01 1.24118634e-01
-5.07481933e-01 -7.87647605e-01 1.43645272e-01 6.33171797e-01
5.17389596e-01 4.60106492e-01 1.26218951e+00 -1.20344877e+00
-2.11418513e-02 -4.66996402e-01 -2.22929657e-01 5.22157431e-01
2.25444362e-01 1.53345719e-01 -1.06952214e+00 2.00605337e-02
-2.20727071e-01 -3.74126792e-01 6.76054955e-01 3.84871364e-01
1.30197442e+00 -4.23880368e-01 -3.69370818e-01 2.70192504e-01
1.65181279e+00 9.84356105e-01 5.80407560e-01 8.01189482e-01
3.18512589e-01 5.69735110e-01 5.92927873e-01 1.72676072e-01
3.95240963e-01 3.41594666e-01 1.66542437e-02 -1.43609852e-01
1.34677663e-01 3.47565636e-02 2.07765728e-01 5.05733430e-01
-3.09362531e-01 5.02782285e-01 -1.38043296e+00 1.18017800e-01
-1.78910971e+00 -8.77013922e-01 -2.63533890e-01 2.27148533e+00
1.05564201e+00 4.91629958e-01 4.72185969e-01 7.34843969e-01
3.15777421e-01 -7.15757012e-01 -2.85954118e-01 -1.03778982e+00
3.37964416e-01 4.00901586e-01 4.40049171e-01 4.73378062e-01
-1.12608671e+00 8.52867126e-01 8.05739784e+00 5.03809869e-01
-9.65943754e-01 -1.91291004e-01 9.86274362e-01 9.83093232e-02
2.68830568e-01 7.14206398e-02 -7.83022285e-01 5.45896530e-01
1.48034787e+00 1.00672916e-01 6.07529283e-01 9.95989561e-01
-8.62922221e-02 -4.03732777e-01 -1.22734880e+00 7.75791347e-01
-5.59134930e-02 -1.10168779e+00 3.93332802e-02 -7.83529803e-02
7.35270143e-01 -9.31990147e-02 -3.58456582e-01 7.15254724e-01
7.43549705e-01 -1.50339365e+00 1.70850143e-01 7.70236611e-01
5.90837240e-01 -7.88726270e-01 9.60340440e-01 5.17559290e-01
-3.88607025e-01 -6.38553798e-01 -8.43948051e-02 -5.96396983e-01
-6.90260589e-01 8.12617540e-01 -1.58479345e+00 2.42954642e-01
7.96374261e-01 2.88779974e-01 -8.22677612e-01 1.04711258e+00
1.41086355e-01 8.17039847e-01 -2.40926221e-01 -2.97988862e-01
-3.90478410e-02 -9.84138176e-02 -1.00130312e-01 1.30146194e+00
1.52217925e-01 1.71930596e-01 2.49656379e-01 1.52034983e-01
6.17191017e-01 2.87536174e-01 -5.95341742e-01 4.63834256e-01
3.64783198e-01 1.03080606e+00 -5.11380196e-01 -6.91547632e-01
-2.54313201e-01 4.66027707e-01 2.02197984e-01 -7.68489540e-02
-2.39145994e-01 -2.82248706e-01 2.69205626e-02 1.45777941e-01
-1.28448114e-01 2.39852473e-01 -6.44242644e-01 -5.93126535e-01
-3.70436192e-01 -1.12670469e+00 6.49100840e-01 -8.75758111e-01
-1.41206062e+00 5.73121965e-01 -1.54917482e-02 -9.18988585e-01
-6.41273022e-01 -9.02312338e-01 -3.10948402e-01 8.46078157e-01
-9.48782742e-01 -1.10967493e+00 -1.44599959e-01 4.99969214e-01
2.97463834e-01 -7.02789009e-01 1.18013942e+00 1.53360367e-01
-3.60942483e-01 3.41235280e-01 3.21918368e-01 1.35007381e-01
5.89611053e-01 -1.37747931e+00 -2.58797139e-01 -1.18808813e-01
-2.34825954e-01 2.86549091e-01 5.00332415e-01 -4.89464194e-01
-1.01568937e+00 -8.54492366e-01 9.93414104e-01 -7.27257431e-01
4.37883496e-01 1.48180231e-01 -7.58092940e-01 7.24403739e-01
-8.54817629e-02 -1.39743850e-01 1.21086872e+00 4.37045127e-01
-2.11253420e-01 -6.03436828e-01 -1.66406786e+00 -6.45879433e-02
4.54925358e-01 1.62635241e-02 -8.00613403e-01 4.36599135e-01
1.40519798e-01 4.04395498e-02 -1.19155073e+00 6.54099047e-01
1.03784502e+00 -9.47939098e-01 8.17088902e-01 -1.13835025e+00
6.26515448e-01 -2.44930401e-01 -2.36178398e-01 -1.15041089e+00
-5.29027700e-01 1.38570592e-01 1.22499108e-01 1.04393208e+00
9.92288768e-01 -9.04543459e-01 6.68866634e-01 8.73669207e-01
3.39355350e-01 -9.34036613e-01 -5.52664042e-01 -5.30216932e-01
1.46306887e-01 -4.01151568e-01 5.90591371e-01 1.37998533e+00
1.50032341e-01 5.88286757e-01 -2.75626361e-01 -3.63003254e-01
3.53772283e-01 2.88207959e-02 7.52908885e-01 -1.69625616e+00
-2.70589441e-01 -3.84162873e-01 -8.07391107e-01 -9.67377648e-02
-3.24037522e-01 -8.72296751e-01 -3.69753540e-01 -1.52901304e+00
5.72351038e-01 -9.51433361e-01 -6.27766013e-01 9.12254512e-01
-9.32361558e-02 1.19796805e-02 -1.39868975e-01 2.34073341e-01
-2.48230234e-01 -3.55039120e-01 7.94513166e-01 1.55889377e-01
-4.21001405e-01 3.28646451e-01 -7.52793133e-01 8.84258986e-01
1.12466741e+00 -5.53693056e-01 -1.86148152e-01 2.39119247e-01
1.23160221e-01 1.73976824e-01 4.74563465e-02 -1.01292074e+00
-3.03107291e-01 -2.71524400e-01 1.37801564e+00 -5.30477524e-01
1.18870968e-02 -1.03342736e+00 5.47038555e-01 8.92361879e-01
-4.53814238e-01 4.04396318e-02 3.55152562e-02 -7.74703501e-03
2.87183896e-02 -4.96993273e-01 7.77400792e-01 -2.21078575e-01
-7.92088568e-01 -1.72434866e-01 -3.24091166e-01 -3.90963584e-01
1.27742219e+00 -2.63813376e-01 -2.21531868e-01 -7.65292346e-02
-1.27507901e+00 2.16218993e-01 2.15321541e-01 2.87225276e-01
2.08240256e-01 -1.37865877e+00 -9.32015002e-01 4.28621739e-01
3.49649340e-02 -5.01821995e-01 -4.27711695e-01 5.88896155e-01
-7.49670923e-01 2.74017692e-01 -7.56970823e-01 -5.69280624e-01
-1.48801219e+00 5.09997547e-01 2.78253853e-01 -3.70058894e-01
-2.31473222e-01 6.58214748e-01 -5.62964678e-01 -6.19736612e-01
2.21059084e-01 -4.75765467e-02 -8.60964417e-01 1.89107358e-02
6.69779539e-01 6.65143073e-01 2.05135018e-01 -1.82094097e-01
-4.41640496e-01 4.66496736e-01 -9.96880084e-02 -1.52014002e-01
1.51086688e+00 4.75661725e-01 -2.51355857e-01 6.19489610e-01
1.01923573e+00 -4.07107443e-01 -5.48582315e-01 1.65019438e-01
4.16931599e-01 -3.89838964e-01 1.66861206e-01 -1.55763483e+00
-6.86508656e-01 5.60618758e-01 1.32188928e+00 4.36631680e-01
1.24884939e+00 -2.54220992e-01 1.83263630e-01 5.41465521e-01
1.99127421e-01 -1.05283880e+00 -3.55561823e-01 2.23074719e-01
7.45489657e-01 -1.63858616e+00 3.61746401e-01 -4.48000692e-02
-8.46659243e-01 1.29445815e+00 4.92434710e-01 -1.93756282e-01
1.01395190e+00 4.47433740e-01 1.99314013e-01 5.99576011e-02
-1.26145542e+00 1.38402432e-01 3.06729794e-01 7.77187169e-01
5.37846506e-01 4.70809430e-01 -5.40393353e-01 8.19925904e-01
-5.11736631e-01 4.65041190e-01 1.27027417e-02 8.85641634e-01
-7.16831207e-01 -1.13556969e+00 -5.92783034e-01 1.07541406e+00
-6.07134223e-01 1.49998283e-02 -6.77483976e-01 8.59378457e-01
5.38035452e-01 9.97840881e-01 -8.21965858e-02 -7.60721028e-01
2.59262741e-01 5.26176393e-01 4.08558071e-01 -2.54135162e-01
-7.73584127e-01 -3.38348657e-01 3.64219844e-01 -1.67281389e-01
-2.64162183e-01 -7.61009037e-01 -1.18073237e+00 -4.44277078e-01
-2.53345728e-01 3.19538146e-01 1.02842188e+00 8.94660950e-01
9.92643386e-02 2.53905505e-01 1.04849470e+00 -3.66306484e-01
-2.41697356e-01 -1.28383625e+00 -4.07495141e-01 2.05391064e-01
2.77190477e-01 -6.10886812e-01 -5.51152110e-01 2.83006459e-01] | [8.384753227233887, 4.818276882171631] |
ac9a609e-78fa-4f7a-8d96-4f87342dca29 | locate-then-segment-a-strong-pipeline-for | 2103.16284 | null | https://arxiv.org/abs/2103.16284v1 | https://arxiv.org/pdf/2103.16284v1.pdf | Locate then Segment: A Strong Pipeline for Referring Image Segmentation | Referring image segmentation aims to segment the objects referred by a natural language expression. Previous methods usually focus on designing an implicit and recurrent feature interaction mechanism to fuse the visual-linguistic features to directly generate the final segmentation mask without explicitly modeling the localization information of the referent instances. To tackle these problems, we view this task from another perspective by decoupling it into a "Locate-Then-Segment" (LTS) scheme. Given a language expression, people generally first perform attention to the corresponding target image regions, then generate a fine segmentation mask about the object based on its context. The LTS first extracts and fuses both visual and textual features to get a cross-modal representation, then applies a cross-model interaction on the visual-textual features to locate the referred object with position prior, and finally generates the segmentation result with a light-weight segmentation network. Our LTS is simple but surprisingly effective. On three popular benchmark datasets, the LTS outperforms all the previous state-of-the-art methods by a large margin (e.g., +3.2% on RefCOCO+ and +3.4% on RefCOCOg). In addition, our model is more interpretable with explicitly locating the object, which is also proved by visualization experiments. We believe this framework is promising to serve as a strong baseline for referring image segmentation. | ['Tieniu Tan', 'Lei LI', 'Liang Wang', 'Wei Wang', 'Tao Kong', 'Ya Jing'] | 2021-03-30 | null | http://openaccess.thecvf.com//content/CVPR2021/html/Jing_Locate_Then_Segment_A_Strong_Pipeline_for_Referring_Image_Segmentation_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Jing_Locate_Then_Segment_A_Strong_Pipeline_for_Referring_Image_Segmentation_CVPR_2021_paper.pdf | cvpr-2021-1 | ['generalized-referring-expression-segmentation'] | ['computer-vision'] | [ 3.39779407e-01 2.87450701e-01 -4.50582534e-01 -5.04772723e-01
-9.45545018e-01 -7.50475705e-01 6.83274984e-01 2.68588867e-02
-3.61580104e-01 1.84175789e-01 1.09978594e-01 -1.51732385e-01
3.51432383e-01 -5.48753202e-01 -7.82337666e-01 -6.06936753e-01
3.86601657e-01 5.06967962e-01 3.96426409e-01 -1.24491289e-01
3.90150726e-01 4.81331408e-01 -1.28633130e+00 4.55910325e-01
9.77945268e-01 1.08646977e+00 3.49112034e-01 1.11683130e-01
-4.98924851e-01 7.01882482e-01 -6.34699881e-01 -4.63796705e-01
-2.86690835e-02 -5.52456796e-01 -1.15456903e+00 6.64009094e-01
4.23315763e-01 1.22003548e-01 1.10126324e-01 1.20696831e+00
1.72599927e-01 1.86303854e-01 7.54023373e-01 -1.08000636e+00
-9.12312388e-01 7.40622044e-01 -1.06248903e+00 -1.29352614e-01
5.52011669e-01 2.70069391e-01 1.08755410e+00 -9.23531175e-01
7.68603981e-01 1.45365775e+00 2.15716630e-01 5.24551690e-01
-1.38564944e+00 -3.56881768e-01 7.11451948e-01 -2.24446170e-02
-1.34660697e+00 -2.36490101e-01 9.96032715e-01 -5.25803447e-01
5.90246320e-01 4.16663706e-01 6.20116413e-01 9.50748205e-01
-2.07619697e-01 1.31437433e+00 9.49945807e-01 -4.10810351e-01
-9.50968917e-03 1.61861464e-01 2.86151946e-01 7.65059948e-01
-2.98483014e-01 -2.45663315e-01 -1.56744063e-01 1.71220943e-01
6.00159705e-01 -4.28778753e-02 -3.35311562e-01 -1.56439945e-01
-1.22061455e+00 7.35517681e-01 8.65607977e-01 4.97765839e-01
-4.33034867e-01 2.00978383e-01 2.98121810e-01 -3.44829172e-01
5.17140687e-01 2.50679135e-01 -1.83575332e-01 1.82174250e-01
-1.11172271e+00 7.05771223e-02 4.19944137e-01 1.06800020e+00
9.20249283e-01 -2.98631519e-01 -5.89884341e-01 8.60272288e-01
4.20755237e-01 3.68287534e-01 1.78490713e-01 -1.07510221e+00
5.80407560e-01 9.31510389e-01 9.31059048e-02 -1.10622752e+00
-3.13651890e-01 -3.82157981e-01 -6.71556771e-01 1.13175651e-02
4.52434897e-01 -6.22770414e-02 -1.10728574e+00 1.67576122e+00
4.19798225e-01 1.22456299e-02 -2.06395492e-01 1.18819273e+00
1.05765140e+00 8.04106534e-01 3.61200958e-01 2.22593285e-02
1.77267897e+00 -1.29606950e+00 -7.39915013e-01 -6.21466041e-01
4.81074631e-01 -7.81564057e-01 1.36792552e+00 1.47683367e-01
-1.16829908e+00 -6.29247785e-01 -6.91564322e-01 -3.43881726e-01
-3.78877074e-01 5.01401126e-01 5.36092281e-01 1.23978123e-01
-9.88274336e-01 1.57308504e-01 -6.26742959e-01 -3.28846008e-01
6.40724301e-01 1.87661201e-01 -2.46122137e-01 1.26599476e-01
-9.69009340e-01 7.16787279e-01 4.42397952e-01 2.50018954e-01
-5.05302131e-01 -4.91649240e-01 -1.03012609e+00 2.64162105e-02
6.42684877e-01 -7.51633763e-01 1.08882225e+00 -1.39010966e+00
-1.30660915e+00 1.38320291e+00 -5.40437162e-01 -2.23798007e-01
6.39177263e-01 -2.46570066e-01 -9.49343890e-02 3.88355196e-01
4.77626264e-01 1.25345528e+00 8.31585884e-01 -1.80280983e+00
-6.73090339e-01 -2.73772806e-01 2.77075142e-01 1.78079277e-01
1.95673212e-01 1.33392900e-01 -1.30656433e+00 -6.62622035e-01
2.70051569e-01 -8.24522197e-01 -2.19305262e-01 -1.32494373e-02
-9.43402529e-01 -6.45378828e-01 8.93685818e-01 -5.33364117e-01
1.20795166e+00 -2.19572043e+00 3.17059457e-01 6.17287047e-02
2.67284244e-01 2.06905708e-01 -1.13990955e-01 1.58490553e-01
-1.81563526e-01 4.50837225e-01 -3.41739058e-01 -5.43410718e-01
7.75046321e-03 3.88345644e-02 -3.65168989e-01 4.24311280e-01
4.36671197e-01 1.30172276e+00 -8.55894566e-01 -9.42972839e-01
2.42331058e-01 3.95654231e-01 -3.59582543e-01 1.87130094e-01
-4.92280513e-01 5.40554404e-01 -6.78060114e-01 6.84592903e-01
4.96397763e-01 -5.66175520e-01 -9.57465693e-02 -5.56846321e-01
-5.55097833e-02 -1.40088528e-01 -8.54261577e-01 1.72717035e+00
-3.27026963e-01 5.50474703e-01 -5.71756773e-02 -9.62144315e-01
1.01592863e+00 -9.76858661e-02 2.90497810e-01 -7.39795029e-01
2.92443126e-01 -6.46469444e-02 -3.67911190e-01 -8.06485236e-01
2.71194696e-01 1.35374323e-01 -4.72941011e-01 2.66407400e-01
-5.77872172e-02 -1.33252710e-01 2.97405273e-01 3.98726285e-01
4.21707094e-01 4.90852654e-01 1.47984788e-01 -2.11702481e-01
6.95662737e-01 1.33245802e-02 4.23087716e-01 5.61204076e-01
-2.22452998e-01 8.68971586e-01 8.09135377e-01 -2.81082720e-01
-5.92559695e-01 -9.46324527e-01 4.79909070e-02 1.19820046e+00
7.15420306e-01 -2.64035463e-01 -1.04921377e+00 -9.86047328e-01
-1.89770028e-01 1.00135553e+00 -8.19319963e-01 -2.01396625e-02
-5.26421547e-01 -3.45821470e-01 2.71272331e-01 6.66748405e-01
7.39726603e-01 -1.40812612e+00 -5.96398413e-01 -9.26544368e-02
-4.45930034e-01 -1.18807459e+00 -9.30786252e-01 7.45705143e-02
-4.02760267e-01 -8.73224974e-01 -7.60402560e-01 -1.02683306e+00
9.15501058e-01 8.32892060e-02 1.20641017e+00 1.96541950e-01
-5.36156781e-02 2.41696477e-01 -2.43628427e-01 2.26482414e-02
-9.31474715e-02 1.28763229e-01 -5.47532439e-01 3.43295425e-01
2.18414500e-01 -1.99782237e-01 -6.59327269e-01 3.00215006e-01
-8.08989584e-01 4.43714887e-01 4.75109011e-01 6.96402252e-01
9.22760248e-01 -3.74866694e-01 2.23949954e-01 -9.97082412e-01
4.70178574e-01 -3.73496771e-01 -5.13371050e-01 5.83116889e-01
-3.90999578e-02 6.06886558e-02 4.28804487e-01 -5.61333179e-01
-1.04450667e+00 4.09445852e-01 1.05946526e-01 -5.68588316e-01
-3.57924849e-01 4.74911302e-01 -4.50054348e-01 3.06462705e-01
4.38893348e-01 2.09000602e-01 -2.67837703e-01 -3.47586095e-01
8.09329510e-01 4.97155815e-01 6.98998094e-01 -7.86522627e-01
6.17462456e-01 4.63395834e-01 -2.21616134e-01 -5.19031405e-01
-1.30513072e+00 -3.98960173e-01 -8.97723496e-01 -2.63793260e-01
1.29152966e+00 -6.24947608e-01 -8.45371008e-01 1.89063668e-01
-1.48053801e+00 -4.53973949e-01 -1.83120668e-01 -1.07958898e-01
-4.93959486e-01 3.00885767e-01 -3.75316232e-01 -6.72476232e-01
-2.50555575e-01 -1.40297568e+00 1.64617002e+00 4.27619845e-01
-3.35953742e-01 -1.01852715e+00 -5.64483881e-01 5.16859770e-01
3.24157253e-02 4.16516423e-01 8.70419025e-01 -5.68411887e-01
-6.00264788e-01 -1.26988605e-01 -7.46236682e-01 1.87253878e-02
1.15709774e-01 2.73196936e-01 -9.06411290e-01 1.59721449e-01
-2.32908636e-01 -1.10896662e-01 1.00508714e+00 4.02910173e-01
1.51874828e+00 -2.03185230e-01 -7.35575497e-01 5.89760900e-01
1.22290778e+00 2.33454213e-01 4.12458658e-01 5.69243729e-02
9.51466560e-01 8.24663222e-01 7.35401273e-01 8.63920432e-03
4.96563137e-01 8.39170992e-01 5.71527243e-01 -6.08089030e-01
-2.25435883e-01 -3.39108199e-01 1.07878022e-01 3.40687513e-01
1.22582547e-01 -4.38921541e-01 -9.99052823e-01 6.46001041e-01
-2.00028944e+00 -7.98495471e-01 -1.79484829e-01 1.62097752e+00
8.08550477e-01 7.70783871e-02 1.97246000e-01 -1.80050269e-01
9.81943429e-01 2.42605999e-01 -5.31493843e-01 -1.66976348e-01
-7.23309368e-02 -1.49850503e-01 1.65889859e-01 4.97534275e-01
-1.33257234e+00 1.45731723e+00 5.52398491e+00 1.08757806e+00
-1.26603794e+00 -5.82974888e-02 1.03085184e+00 2.27523580e-01
-3.35603237e-01 1.16413221e-01 -8.33941102e-01 4.82666343e-01
2.52780408e-01 -2.11483501e-02 2.81332105e-01 6.86493039e-01
2.92115033e-01 -2.67864436e-01 -1.13347435e+00 1.09020698e+00
1.70039192e-01 -1.24149621e+00 2.73547113e-01 -1.52964666e-01
6.52150571e-01 -3.39949489e-01 2.34783068e-01 1.22557029e-01
7.41798580e-02 -1.14961755e+00 1.19910634e+00 6.84358537e-01
8.45088303e-01 -6.70458913e-01 6.06347799e-01 2.80571997e-01
-1.35472929e+00 1.55905887e-01 5.81292808e-02 3.19248229e-01
3.88482451e-01 3.71750116e-01 -4.60189611e-01 5.93868732e-01
6.52279139e-01 7.85497248e-01 -8.42364907e-01 8.50018203e-01
-5.92077434e-01 5.33698201e-01 -1.94041610e-01 1.94542203e-02
6.53401971e-01 -2.99028516e-01 2.58910239e-01 1.31671953e+00
2.62246691e-02 2.22520772e-02 5.44495106e-01 1.52721894e+00
-1.27578869e-01 2.09868744e-01 -3.26604158e-01 4.19769697e-02
2.94428051e-01 1.48121929e+00 -1.19774640e+00 -4.96788442e-01
-3.05592448e-01 1.10997033e+00 3.94903541e-01 6.64125383e-01
-1.12699533e+00 -4.48896736e-01 5.34791201e-02 9.11209509e-02
4.06424642e-01 9.85194594e-02 -3.45618725e-01 -1.02443755e+00
7.62644932e-02 -6.80785358e-01 2.54489452e-01 -1.08152032e+00
-1.21798086e+00 8.42910826e-01 7.00697973e-02 -1.10166621e+00
-1.92206532e-01 -4.09336299e-01 -7.43780017e-01 9.99077559e-01
-1.23973310e+00 -1.47138751e+00 -3.84145707e-01 4.79184538e-01
7.24890053e-01 3.27712625e-01 4.63269055e-01 9.50659737e-02
-8.17628324e-01 5.02006233e-01 -7.06880212e-01 3.61896604e-01
3.66673052e-01 -1.20839083e+00 2.35501155e-01 8.08139920e-01
3.74750644e-01 7.37835348e-01 5.74706793e-01 -5.69664121e-01
-9.05970395e-01 -1.16403723e+00 8.65818322e-01 -4.95530248e-01
4.73492533e-01 -3.54959071e-01 -1.01778376e+00 7.20929146e-01
4.81007218e-01 2.55812436e-01 1.73793301e-01 -1.31674528e-01
-2.52999574e-01 6.41932487e-02 -9.49343503e-01 8.97533715e-01
1.09777701e+00 -5.28785944e-01 -7.66513050e-01 2.56600559e-01
9.30272400e-01 -5.82618415e-01 -5.32245338e-01 1.76005006e-01
1.40096754e-01 -8.26762438e-01 9.19344723e-01 -4.59930122e-01
5.01151204e-01 -5.13792932e-01 7.04268692e-03 -8.92723024e-01
-1.76248103e-01 -5.56702018e-01 3.88962924e-01 1.75642967e+00
4.76750195e-01 -3.30934078e-01 4.99688119e-01 6.08716846e-01
-1.03376769e-01 -9.34592903e-01 -6.67458415e-01 -4.18651581e-01
-5.68005964e-02 -5.73323250e-01 4.79535669e-01 8.37176323e-01
-3.13010663e-02 5.32920778e-01 -6.72465116e-02 1.11763902e-01
3.44488949e-01 5.35431325e-01 6.14777386e-01 -9.75186050e-01
1.36563912e-01 -7.52119839e-01 -1.01922370e-01 -1.57769239e+00
5.84105790e-01 -1.05722106e+00 1.40445828e-01 -1.68245375e+00
2.95356154e-01 -3.07932466e-01 -1.08153880e-01 8.07816744e-01
-4.43739563e-01 2.38738179e-01 4.25185084e-01 1.07691057e-01
-9.32439744e-01 4.18084443e-01 1.58369958e+00 -3.65657538e-01
-2.69190609e-01 -2.73045786e-02 -9.31747973e-01 9.44501281e-01
6.02961302e-01 -3.69926155e-01 -2.88047493e-01 -3.76051337e-01
-6.61221296e-02 1.44702122e-01 6.16012692e-01 -4.66869473e-01
3.10661793e-01 -2.97155201e-01 3.36938083e-01 -9.07311559e-01
2.16204971e-01 -8.42187464e-01 -7.56852254e-02 1.40358254e-01
-4.68888164e-01 -1.55371860e-01 1.55045375e-01 3.06820720e-01
-3.64235342e-01 -3.01269859e-01 6.90456212e-01 -1.81654751e-01
-7.60956049e-01 2.43043914e-01 -1.27289161e-01 3.21637422e-01
1.16668129e+00 -3.45418811e-01 -1.99795783e-01 -3.28682184e-01
-9.48952258e-01 5.29480457e-01 4.29596424e-01 4.33300346e-01
5.35164297e-01 -1.33543944e+00 -3.39019179e-01 6.52773529e-02
1.84478104e-01 3.03750306e-01 1.38299242e-01 9.88779724e-01
-2.50761032e-01 3.16900849e-01 1.94165215e-01 -1.10916042e+00
-1.22736406e+00 8.13229561e-01 4.57274616e-01 -2.44715810e-02
-6.00842774e-01 1.04404676e+00 7.37214088e-01 -1.29410654e-01
2.79621243e-01 -4.83477145e-01 -3.32245946e-01 3.35022718e-01
1.85695469e-01 -6.81358650e-02 -5.05926013e-01 -1.11431503e+00
-4.72995579e-01 1.16433573e+00 1.56202838e-02 -5.73347248e-02
9.98258352e-01 -2.76133835e-01 -1.85745656e-01 4.34892505e-01
1.27223217e+00 6.60691559e-02 -1.33401728e+00 -2.93223411e-01
1.63589224e-01 -2.78163821e-01 -2.13186905e-01 -8.33095849e-01
-1.54054523e+00 1.03805506e+00 2.08868504e-01 3.22105199e-01
1.21906424e+00 5.67166567e-01 4.97858286e-01 -8.99006352e-02
1.47847995e-01 -7.43939757e-01 3.09047490e-01 3.04240167e-01
1.17759264e+00 -1.17967272e+00 -1.87212661e-01 -7.56016731e-01
-1.05600178e+00 1.01123226e+00 7.77138591e-01 -1.26294985e-01
3.72718513e-01 3.54826590e-03 2.22766951e-01 -3.88740957e-01
-4.00928646e-01 -4.94064391e-01 6.38717771e-01 4.43944246e-01
5.37907243e-01 -1.80541519e-02 -2.15157390e-01 7.70855427e-01
2.37876475e-02 -2.89375573e-01 -6.05848394e-02 4.29420054e-01
-2.65571207e-01 -1.03433371e+00 -3.51852655e-01 8.67486596e-02
-4.57599998e-01 -4.55117151e-02 -6.31929517e-01 9.65583563e-01
3.34087074e-01 9.28055823e-01 9.63785499e-02 -1.65327057e-01
4.61541086e-01 -4.76912744e-02 1.93560228e-01 -6.50245011e-01
-7.50948489e-01 5.74609876e-01 -8.29379708e-02 -7.77456403e-01
-6.43018544e-01 -6.34141684e-01 -1.73144376e+00 2.36609489e-01
-2.98604637e-01 1.82723980e-02 1.45508900e-01 1.04000390e+00
3.11481893e-01 7.70475030e-01 4.47246075e-01 -1.08139098e+00
-5.03511876e-02 -6.45068407e-01 -1.94389820e-01 8.26966345e-01
1.11008070e-01 -6.10269308e-01 -3.48887980e-01 2.92009532e-01] | [10.273747444152832, 1.218767762184143] |
2ac355c5-1fba-46d7-a441-69f5710bf964 | partial-annotation-learning-for-biomedical | 2305.13120 | null | https://arxiv.org/abs/2305.13120v1 | https://arxiv.org/pdf/2305.13120v1.pdf | Partial Annotation Learning for Biomedical Entity Recognition | Motivation: Named Entity Recognition (NER) is a key task to support biomedical research. In Biomedical Named Entity Recognition (BioNER), obtaining high-quality expert annotated data is laborious and expensive, leading to the development of automatic approaches such as distant supervision. However, manually and automatically generated data often suffer from the unlabeled entity problem, whereby many entity annotations are missing, degrading the performance of full annotation NER models. Results: To address this problem, we systematically study the effectiveness of partial annotation learning methods for biomedical entity recognition over different simulated scenarios of missing entity annotations. Furthermore, we propose a TS-PubMedBERT-Partial-CRF partial annotation learning model. We harmonize 15 biomedical NER corpora encompassing five entity types to serve as a gold standard and compare against two commonly used partial annotation learning models, BiLSTM-Partial-CRF and EER-PubMedBERT, and the state-of-the-art full annotation learning BioNER model PubMedBERT tagger. Results show that partial annotation learning-based methods can effectively learn from biomedical corpora with missing entity annotations. Our proposed model outperforms alternatives and, specifically, the PubMedBERT tagger by 38% in F1-score under high missing entity rates. The recall of entity mentions in our model is also competitive with the upper bound on the fully annotated dataset. | ['Zhixiong Zhang', 'Giovanni Colavizza', 'Liangping Ding'] | 2023-05-22 | null | null | null | null | ['named-entity-recognition-ner'] | ['natural-language-processing'] | [ 2.27197334e-02 5.61659157e-01 -4.32788342e-01 -4.59794879e-01
-1.19202721e+00 -3.63278568e-01 1.30015314e-01 6.16705894e-01
-1.03695750e+00 1.42021477e+00 2.00838208e-01 -1.16191588e-01
2.04740703e-01 -4.13973063e-01 -7.95262337e-01 -6.18097484e-01
2.83297926e-01 8.41954172e-01 1.75234675e-01 2.16054350e-01
-1.64717734e-01 3.11051518e-01 -9.26695287e-01 1.97699100e-01
1.10946143e+00 5.49696267e-01 2.05906153e-01 4.38821852e-01
-4.54365224e-01 6.09756589e-01 -7.39611685e-01 -6.30703568e-01
-3.29276472e-01 -2.40253106e-01 -1.05013537e+00 -5.16023517e-01
-2.94639263e-02 1.69006750e-01 1.13357037e-01 1.00209582e+00
8.55431199e-01 -5.73260821e-02 6.12243652e-01 -8.15210223e-01
-4.04726744e-01 8.15931261e-01 -2.40898684e-01 1.14038438e-02
1.69549212e-01 -3.57250333e-01 7.69390583e-01 -8.55937183e-01
1.09707332e+00 6.76319718e-01 1.02621734e+00 9.30293024e-01
-1.04274642e+00 -5.85963249e-01 -2.20089659e-01 -1.81920737e-01
-1.56043220e+00 -4.32773411e-01 8.16124305e-02 -4.99277622e-01
1.09108782e+00 1.30445119e-02 -1.85214374e-02 1.13110709e+00
2.05882818e-01 7.05990374e-01 9.94858623e-01 -4.69271302e-01
2.47182563e-01 3.09275597e-01 3.65816236e-01 7.18468189e-01
4.29373533e-01 -1.61357820e-01 -3.27328593e-01 -4.16301876e-01
4.44105119e-01 -1.28007218e-01 -2.33348712e-01 1.68838710e-01
-1.29522359e+00 3.39063287e-01 1.93087950e-01 6.84131622e-01
-6.19749129e-01 -1.55996561e-01 5.32697022e-01 -1.14027470e-01
6.61975503e-01 5.77018499e-01 -1.04770589e+00 -3.59625220e-02
-1.05619252e+00 -2.50343382e-01 1.17938507e+00 1.13937080e+00
4.97896820e-01 -1.40620872e-01 -4.15683836e-01 8.92973185e-01
1.04512997e-01 2.65905857e-01 6.65962994e-01 -3.65103155e-01
2.03733593e-01 5.87161720e-01 8.18868726e-02 -3.98946643e-01
-7.36112297e-01 -5.67743659e-01 -1.01858282e+00 -5.90068400e-01
3.10997635e-01 -5.08166790e-01 -1.02834833e+00 1.63030863e+00
5.30217350e-01 4.79581892e-01 3.65298718e-01 4.98396337e-01
1.12943363e+00 3.34207952e-01 1.01272964e+00 -3.91097963e-01
1.61240804e+00 -9.01172340e-01 -1.14394200e+00 2.00966462e-01
1.11861253e+00 -6.49861991e-01 3.00167084e-01 -1.75143313e-02
-6.23863399e-01 -3.92784685e-01 -6.09700620e-01 -1.48714930e-02
-7.16607511e-01 5.29946744e-01 3.76798689e-01 6.99149668e-01
-7.69381523e-01 5.30470014e-01 -1.11713517e+00 -3.61955255e-01
3.36080819e-01 6.07227683e-01 -7.23002732e-01 -5.94412684e-02
-1.38569915e+00 1.10846734e+00 8.92991185e-01 1.93561688e-01
-5.91065466e-01 -9.82034147e-01 -8.28548372e-01 9.14317295e-02
2.25139380e-01 -6.62904799e-01 1.12887442e+00 -4.32971716e-01
-9.07453537e-01 9.58363712e-01 -2.07586035e-01 -5.89190423e-01
3.22345734e-01 -3.55619699e-01 -6.80431426e-01 -9.48838294e-02
2.75858402e-01 7.23530412e-01 -2.68750004e-02 -1.06131852e+00
-4.49136138e-01 -2.77178198e-01 -6.39989972e-01 -5.60881086e-02
-1.47376016e-01 1.21911682e-01 -2.14419186e-01 -5.27173579e-01
-4.95592684e-01 -7.41902828e-01 -6.60325050e-01 -5.65037251e-01
-6.10598981e-01 -5.40287137e-01 1.86018661e-01 -9.01721537e-01
1.19049299e+00 -2.05323625e+00 -3.59169066e-01 -1.55422524e-01
1.14213929e-01 4.74879980e-01 -2.92280726e-02 2.98837513e-01
-1.86150625e-01 4.02223289e-01 -2.48554230e-01 -4.11047727e-01
-1.26723140e-01 4.26900387e-01 2.22459078e-01 2.01418787e-01
3.59726638e-01 9.10325348e-01 -1.02507341e+00 -9.84923840e-01
-2.09822938e-01 8.21692944e-01 -2.82781065e-01 2.17170089e-01
-1.36214495e-01 7.08028078e-01 -7.26219058e-01 7.48662412e-01
3.47302884e-01 -5.03679931e-01 4.40479577e-01 -3.75915676e-01
-1.24144658e-01 1.90053269e-01 -9.70429838e-01 1.79701471e+00
-3.22411925e-01 4.66994233e-02 -1.21711791e-01 -8.50054383e-01
9.14880872e-01 9.42733288e-01 7.53622055e-01 -3.25189590e-01
2.75432989e-02 6.41453207e-01 -2.89667666e-01 -5.68127692e-01
4.12158459e-01 -3.55675370e-01 -1.52235210e-01 -1.46580324e-01
5.85595667e-01 7.40091860e-01 1.06729023e-01 9.75079089e-03
1.38601935e+00 3.91441524e-01 8.49960268e-01 -2.01340377e-01
5.47264755e-01 1.60895884e-01 1.05575442e+00 5.60020804e-01
-3.13441187e-01 2.46598423e-01 2.02694178e-01 -2.64401644e-01
-7.48947620e-01 -5.46691477e-01 -5.30037701e-01 9.93996382e-01
-2.77811587e-01 -3.90293866e-01 -9.34928060e-01 -1.18818593e+00
-3.03968012e-01 7.00670123e-01 -3.92112851e-01 1.91780403e-01
-6.03395283e-01 -1.21263719e+00 1.06675446e+00 5.24730802e-01
3.44860554e-01 -1.37503409e+00 -3.54348004e-01 5.57579637e-01
-4.63110656e-01 -1.49069703e+00 -4.07474816e-01 5.84363282e-01
-8.19929123e-01 -1.13864279e+00 -9.82513070e-01 -9.13751304e-01
7.53835022e-01 -7.60859966e-01 1.16898251e+00 -9.06228274e-02
-4.05935556e-01 2.45727137e-01 -3.48102391e-01 -5.54865777e-01
-5.77479303e-01 4.12389576e-01 -4.28513773e-02 -2.89485812e-01
6.83842838e-01 -1.26551554e-01 -4.80173409e-01 2.77973175e-01
-8.38751733e-01 -1.42581940e-01 1.06545949e+00 1.11779439e+00
1.03660619e+00 -3.00835699e-01 1.19422257e+00 -1.49521315e+00
8.75462741e-02 -6.42006159e-01 -2.08148286e-01 6.85906529e-01
-7.33524799e-01 2.42816553e-01 3.56249332e-01 -4.24018055e-01
-1.28235579e+00 5.08883417e-01 -6.43128455e-01 -1.11609951e-01
-5.37392616e-01 7.25325704e-01 -3.85780215e-01 2.07504287e-01
6.46379054e-01 -1.12634934e-02 -6.16115510e-01 -7.70966828e-01
2.03972071e-01 8.85738850e-01 6.84749186e-01 -5.32258451e-01
3.30409594e-02 -6.65921867e-02 -4.53806184e-02 -7.36880004e-01
-1.22387993e+00 -8.32744420e-01 -8.98560643e-01 3.50836486e-01
1.24249148e+00 -1.00076282e+00 -3.60854298e-01 4.01964858e-02
-1.37625015e+00 -8.45965743e-02 -2.77529806e-01 8.05305481e-01
-2.81964511e-01 2.18788609e-01 -9.21778262e-01 -6.13475204e-01
-7.38095403e-01 -9.12853837e-01 1.23278189e+00 2.30669916e-01
-1.85711145e-01 -1.21428061e+00 3.70174706e-01 3.56430769e-01
7.26796091e-02 4.56562757e-01 9.23689067e-01 -1.66230023e+00
-4.16050702e-02 -2.99674690e-01 -1.48043007e-01 1.14380047e-02
3.42545509e-02 -5.97928584e-01 -9.93441403e-01 3.19929719e-02
-2.51766562e-01 -2.15567350e-01 7.91731715e-01 1.17398851e-01
6.61249638e-01 -3.27282876e-01 -9.51061487e-01 1.47164106e-01
1.52036560e+00 1.14951290e-01 5.76780021e-01 1.68999732e-01
6.42173827e-01 5.28456211e-01 6.93401814e-01 2.79794097e-01
2.58054316e-01 4.31243569e-01 -9.01584402e-02 -3.61594677e-01
-6.07868545e-02 -2.99506396e-01 -8.54829997e-02 8.89097631e-01
-5.21971323e-02 -4.38699424e-01 -1.15973866e+00 7.16176927e-01
-1.76110673e+00 -5.69466412e-01 -2.02527672e-01 2.04551148e+00
1.43188667e+00 -1.38826311e-01 -2.06979781e-01 -2.51533777e-01
1.04745710e+00 -5.61837912e-01 -3.63085657e-01 1.16809987e-01
-1.27570229e-02 4.49861199e-01 6.79335058e-01 8.71084780e-02
-1.26296461e+00 9.28220510e-01 5.61963272e+00 7.87016809e-01
-6.23545885e-01 5.52749991e-01 5.40121675e-01 6.91260338e-01
2.33704716e-01 -2.12229729e-01 -1.38973081e+00 3.81806225e-01
1.58600032e+00 1.10357866e-01 -5.44244945e-01 8.38693142e-01
-4.12274711e-02 2.42334887e-01 -1.12452233e+00 7.30071664e-01
5.67643642e-02 -1.18083489e+00 -1.87763289e-01 1.72333747e-01
7.02040911e-01 2.48653159e-01 -6.99775338e-01 5.35420716e-01
4.59362894e-01 -9.04948235e-01 7.53134489e-02 7.71967888e-01
1.02353692e+00 -4.39041644e-01 1.47429490e+00 4.59941357e-01
-1.04894662e+00 3.67403358e-01 -3.55638981e-01 8.45715940e-01
4.27627206e-01 8.69751275e-01 -1.20598769e+00 9.11501825e-01
4.50899035e-01 4.37497795e-01 -2.95538872e-01 1.39659226e+00
-2.90446222e-01 7.57060587e-01 -2.12663248e-01 9.26867276e-02
-1.00302018e-01 4.59797263e-01 2.37968162e-01 1.67823172e+00
3.47968221e-01 3.74160916e-01 4.20467108e-01 6.47863328e-01
-6.52209282e-01 4.50534284e-01 -3.08773279e-01 -3.18734765e-01
5.84642470e-01 1.43095136e+00 -7.90929258e-01 -4.45298582e-01
-3.68374735e-01 9.59538221e-01 4.72364843e-01 1.01745352e-01
-6.68708861e-01 -4.08827424e-01 1.12084866e-01 -1.17741503e-01
1.80386707e-01 2.09493265e-01 1.02217227e-01 -1.03979170e+00
-4.10352677e-01 -5.32993317e-01 7.03287303e-01 -3.51347476e-01
-1.49465907e+00 8.17989230e-01 -1.79999143e-01 -9.91914093e-01
-1.37379065e-01 -5.98331809e-01 -3.98620553e-02 6.64356828e-01
-1.65434957e+00 -1.38891792e+00 2.42209099e-02 2.71095186e-01
3.77367616e-01 9.03019756e-02 1.32048309e+00 8.57324183e-01
-5.97946405e-01 7.47973859e-01 1.87838793e-01 5.87878048e-01
1.25084567e+00 -1.27783513e+00 -1.58773586e-02 2.94032693e-01
1.54218212e-01 7.31131434e-01 4.22156453e-01 -1.09556949e+00
-7.20036983e-01 -1.43346059e+00 1.63621306e+00 -4.29297805e-01
3.75965863e-01 -1.12361692e-01 -1.11423588e+00 7.78019667e-01
1.19586373e-02 3.46769512e-01 1.24884844e+00 1.41675502e-01
-1.67287216e-01 3.69254738e-01 -1.39267373e+00 5.47811687e-02
7.64604211e-01 -2.72704601e-01 -7.06733525e-01 4.10641968e-01
8.06490004e-01 -4.18220907e-01 -1.56784809e+00 6.29404366e-01
3.75213146e-01 -1.42518252e-01 7.60941625e-01 -8.15398574e-01
-2.91848835e-02 -1.17045671e-01 4.76376191e-02 -1.09959602e+00
6.35769516e-02 -3.45340490e-01 8.46973807e-02 1.58661199e+00
7.94241846e-01 -4.61609453e-01 7.36494362e-01 7.34224260e-01
-3.21659178e-01 -5.36639214e-01 -1.05584216e+00 -7.04003632e-01
5.20478887e-03 3.71791460e-02 3.88721734e-01 1.31186235e+00
1.93838581e-01 5.63880324e-01 -1.78848729e-01 2.71418750e-01
4.11686927e-01 -3.97973239e-01 1.37706548e-01 -1.53505456e+00
-1.81889832e-02 1.15570545e-01 -4.04966325e-01 -5.25250316e-01
4.97269243e-01 -1.02252662e+00 2.51870006e-01 -1.63142538e+00
4.30875152e-01 -7.38505840e-01 -6.56654119e-01 1.11472046e+00
-3.62637967e-01 1.80087224e-01 -2.70028353e-01 2.11112440e-01
-9.25960779e-01 2.63154864e-01 8.42878163e-01 5.01324655e-03
-1.56506687e-01 -1.43318459e-01 -3.49279583e-01 7.04258204e-01
5.25256038e-01 -1.03256595e+00 3.00544858e-01 -7.99315646e-02
1.35254428e-01 4.38994944e-01 2.98799314e-02 -8.81654680e-01
5.33303261e-01 1.53960049e-01 4.66416359e-01 -5.30586243e-01
-1.83986947e-01 -8.34190845e-01 2.56508410e-01 3.75652820e-01
-5.26612699e-01 -2.12928146e-01 1.54554546e-01 7.49515772e-01
-2.38441035e-01 -4.65052515e-01 5.19207776e-01 -9.04906541e-02
-4.78971243e-01 1.83665082e-01 -1.37788400e-01 1.72833294e-01
7.69476950e-01 1.97886467e-01 -3.74234349e-01 4.51062500e-01
-1.22840655e+00 2.07471147e-01 -5.66886663e-02 1.51238352e-01
2.30290011e-01 -1.02442098e+00 -7.53984988e-01 -3.02531451e-01
2.07674190e-01 9.41919684e-02 3.18109453e-01 1.09570932e+00
-1.58750594e-01 7.21185207e-01 -2.13588290e-02 -3.52777183e-01
-1.32754827e+00 5.07096946e-01 3.20243418e-01 -9.92566407e-01
-3.45679015e-01 5.58470428e-01 1.48322135e-01 -8.66123796e-01
1.10690482e-01 -6.40261918e-02 -5.98966062e-01 -5.25328740e-02
3.50373179e-01 3.41274112e-01 3.62103343e-01 -5.57361603e-01
-5.04799545e-01 1.02905370e-01 -4.43912260e-02 2.46297628e-01
1.25866532e+00 1.03936233e-01 -4.45057377e-02 2.90290803e-01
9.59965467e-01 -4.29942738e-04 -4.71475512e-01 -2.45676503e-01
8.28065515e-01 3.95803809e-01 -1.70723826e-01 -1.25674093e+00
-7.49228597e-01 5.77223241e-01 7.42828488e-01 -4.37856823e-01
7.78325021e-01 1.91364512e-01 7.35584736e-01 5.59908509e-01
4.62436557e-01 -8.90244007e-01 -4.53694254e-01 3.66726547e-01
2.79414773e-01 -1.28906536e+00 -1.36670485e-01 -5.79821229e-01
-6.41073763e-01 7.84770429e-01 5.81234396e-01 4.45268750e-01
7.09345102e-01 4.86557007e-01 3.67613994e-02 -2.60134667e-01
-5.28474331e-01 -2.83019871e-01 4.66416866e-01 4.83195901e-01
9.93135035e-01 3.45222428e-02 -7.68868506e-01 1.27057743e+00
2.42703184e-01 3.61871451e-01 1.05376534e-01 9.39892828e-01
-1.71741024e-01 -1.44457150e+00 -1.57780796e-02 3.63682061e-01
-1.39744687e+00 -3.62583190e-01 -2.67089069e-01 8.95296633e-01
3.28354180e-01 7.87401974e-01 -3.77960414e-01 2.14666367e-01
3.05906415e-01 5.62137067e-01 1.30458102e-01 -8.43622267e-01
-8.59233260e-01 2.20442370e-01 4.10128981e-01 -1.95802793e-01
-8.54653478e-01 -3.34803224e-01 -1.78081262e+00 5.59756100e-01
-8.38802695e-01 7.98068345e-01 6.49312198e-01 1.16795206e+00
5.51538587e-01 6.88352942e-01 3.12374136e-03 -2.02690929e-01
-3.28194469e-01 -1.19523895e+00 -5.14328182e-01 2.56087989e-01
7.64371781e-03 -4.70814168e-01 -5.38471434e-03 4.13197845e-01] | [8.553701400756836, 8.836126327514648] |
8021dfa2-52b7-4cdf-9285-4a597268cebe | towards-robust-passage-re-ranking-model-by | null | null | https://openreview.net/forum?id=1wAxnkbl-1O | https://openreview.net/pdf?id=1wAxnkbl-1O | Towards Robust Passage Re-Ranking Model by Mitigating Lexical Match Bias | While deep learning models can overcome the limitations of traditional machine learning algorithms that use hand-crafted features, recent studies have shown that these models often achieve high dataset-specific accuracy by exploiting several bias without understanding deeper semantics of intended task. In this paper, we show that neural re-ranking models in information retrieval field are easily deceived by passage and query which exist some lexical match but are semantically irrelevant or less relevant. Then we create a challenging evaluation dataset to expose model's inability in fully capturing contextual semantic information to learn sequence representation for relevance match. After that in order to encourage model to focus on semantic match between passage and query, we explore adversarial removal method to mitigate model's tendency in learning shortcut of lexical match bias in training corpus. Experiments on two benchmarks show that our debiased models can both achieve certain gains on original and challenging test sets compared with baseline models. | ['Anonymous'] | 2021-11-16 | null | null | null | acl-arr-november-2021-11 | ['passage-re-ranking'] | ['natural-language-processing'] | [ 4.07563508e-01 -1.40305623e-01 -9.84393731e-02 -4.47484702e-01
-9.87287760e-01 -8.56629074e-01 8.42621028e-01 1.35943770e-01
-6.87087655e-01 7.85093784e-01 5.09043157e-01 -1.89720199e-01
-1.33211225e-01 -7.62681425e-01 -8.00826132e-01 -1.67829067e-01
1.64318308e-01 2.84310728e-01 4.28561628e-01 -7.19795287e-01
7.40785062e-01 1.37132093e-01 -1.33256042e+00 9.12534177e-01
6.02719307e-01 8.35681975e-01 1.77691821e-02 5.67019463e-01
-1.61127001e-01 1.01291811e+00 -9.76632237e-01 -6.23137057e-01
5.31076550e-01 -5.14222145e-01 -1.31465888e+00 -8.13128293e-01
8.44127119e-01 -5.34115970e-01 -6.68306708e-01 1.15570736e+00
7.99023330e-01 4.41743731e-01 9.77554917e-01 -1.05154526e+00
-1.27582538e+00 5.70342064e-01 -3.42890561e-01 8.20146859e-01
5.56686282e-01 -1.18184194e-01 1.04013348e+00 -8.87141287e-01
5.41777492e-01 1.15214097e+00 5.39606929e-01 9.67750967e-01
-8.17097485e-01 -9.09340978e-01 8.61991122e-02 6.08767152e-01
-1.24579883e+00 -1.54728547e-01 8.56945157e-01 -5.26576899e-02
1.16645265e+00 5.80702782e-01 8.18686932e-02 1.73497295e+00
3.12312633e-01 8.71655405e-01 1.09135878e+00 -3.13818574e-01
-8.76433104e-02 1.03451215e-01 7.31282473e-01 1.70731112e-01
1.03022773e-02 5.53271294e-01 -4.69295710e-01 -2.10455403e-01
1.74560949e-01 2.02109113e-01 -3.66316319e-01 -1.56881452e-01
-8.21051478e-01 7.79060960e-01 6.70162439e-01 1.36033446e-01
-3.18285339e-02 1.02661103e-01 6.31603241e-01 8.99059355e-01
3.10994405e-02 7.57837474e-01 -6.71559215e-01 1.69083089e-01
-7.68820524e-01 4.63768095e-01 5.42255223e-01 8.88188541e-01
5.49694359e-01 -1.97107375e-01 -7.96162367e-01 8.34888518e-01
-4.75175940e-02 3.05067211e-01 1.32410085e+00 -5.28364599e-01
3.18086028e-01 4.76310909e-01 1.40285611e-01 -1.10557413e+00
-1.40838668e-01 -6.71346664e-01 -7.59425223e-01 3.41797434e-02
1.95408896e-01 3.36387455e-01 -9.98963237e-01 1.83660054e+00
-2.04076096e-01 1.67272016e-01 4.65110481e-01 1.02674747e+00
1.04112339e+00 4.03928190e-01 3.22784483e-01 1.58274323e-01
1.37791848e+00 -1.15742803e+00 -4.45826590e-01 -2.95807332e-01
6.89044893e-01 -7.59095013e-01 1.63597524e+00 2.48204291e-01
-9.79134023e-01 -6.80692136e-01 -1.25538218e+00 -3.46367627e-01
-6.32563829e-01 -4.48591113e-01 4.51440066e-01 5.18547356e-01
-8.81734967e-01 7.22603083e-01 -7.15834051e-02 -3.95692706e-01
4.79757845e-01 1.31842762e-01 -2.11536050e-01 -1.48070961e-01
-1.87442935e+00 1.08157635e+00 3.98056358e-01 -1.93948135e-01
-1.37833309e+00 -7.94500172e-01 -4.82572168e-01 2.33602270e-01
1.80382028e-01 -8.52564692e-01 1.27675152e+00 -1.40950131e+00
-1.04133141e+00 1.14891720e+00 2.39871576e-01 -7.27658391e-01
6.36208594e-01 -6.83474183e-01 -6.46724939e-01 6.27062693e-02
-3.70805673e-02 7.25524426e-01 1.04022801e+00 -1.13269198e+00
-3.30775678e-01 -2.44769931e-01 2.51664311e-01 3.33226889e-01
-6.89758062e-01 1.43216727e-02 -5.45868091e-02 -9.75177050e-01
-1.56710714e-01 -6.60742342e-01 4.16537449e-02 -4.37088549e-01
-4.26302910e-01 -3.76881361e-01 7.07256854e-01 -4.88706201e-01
1.18114233e+00 -1.87550676e+00 -3.76663804e-01 -6.52008653e-02
-4.85906452e-02 7.08678782e-01 -5.67200243e-01 7.29088247e-01
-2.37712279e-01 5.35607159e-01 -1.45420292e-02 1.65074855e-01
-1.19889960e-01 4.71410230e-02 -9.52305555e-01 5.53577021e-02
8.68826583e-02 1.17690265e+00 -1.10413253e+00 -2.79293507e-01
-1.42356277e-01 1.95410386e-01 -6.10190332e-01 2.28278607e-01
-1.51665822e-01 1.05029732e-01 -7.05901742e-01 3.25318575e-01
6.05399191e-01 -8.65494832e-02 -3.64864737e-01 -2.22129524e-01
8.43578279e-01 4.76865441e-01 -5.27466178e-01 1.95850587e+00
-2.93920696e-01 4.77641851e-01 -5.81839740e-01 -1.04467618e+00
7.60475576e-01 2.39819854e-01 -1.78431243e-01 -1.12449431e+00
2.17554718e-02 -4.73626256e-02 -8.53129253e-02 -7.02857077e-01
6.49239421e-01 -2.70012349e-01 -1.55587897e-01 5.48279524e-01
-7.90815130e-02 1.01121560e-01 -2.00535744e-01 6.15021348e-01
1.37025309e+00 1.66049093e-01 8.56297240e-02 -3.29911411e-01
8.35028172e-01 5.35460375e-02 2.91538894e-01 1.39223289e+00
-5.66437840e-01 7.79439211e-01 -6.58652559e-03 -7.26617157e-01
-8.69666874e-01 -1.10636497e+00 -3.20615992e-02 1.44028211e+00
5.61946929e-01 -7.12020770e-02 -5.79490423e-01 -1.20243061e+00
-1.18554167e-01 1.08879292e+00 -9.75832939e-01 -1.00538087e+00
-5.06101429e-01 -6.52445793e-01 8.36977601e-01 5.12051940e-01
6.28980875e-01 -1.33078098e+00 -3.85777831e-01 3.03269271e-02
-1.88799515e-01 -6.49858415e-01 -7.07775772e-01 -5.76119572e-02
-7.05605984e-01 -1.08139980e+00 -5.15650213e-01 -8.79558206e-01
4.39241081e-01 3.67025048e-01 1.65675604e+00 5.71740210e-01
-3.17885667e-01 3.40687335e-01 -6.04691327e-01 -4.02708948e-01
-4.15867895e-01 1.04609437e-01 -2.41243303e-01 -4.52140003e-01
9.62360322e-01 -5.16972959e-01 -1.04030728e+00 3.61446172e-01
-1.32348311e+00 -3.43201071e-01 4.06763434e-01 1.09098256e+00
8.33357498e-02 -2.55444676e-01 1.07732975e+00 -9.89318192e-01
1.20168900e+00 -6.86382532e-01 6.85069934e-02 4.70462501e-01
-8.56762052e-01 3.71533394e-01 7.33716190e-01 -5.02954006e-01
-1.05891633e+00 -5.53473771e-01 -4.98521253e-02 -3.78455579e-01
-1.67654380e-01 3.19580406e-01 1.00068741e-01 2.32606441e-01
1.23998928e+00 5.53826272e-01 -3.07451725e-01 -5.63529789e-01
3.00119549e-01 6.28336489e-01 4.65952456e-01 -6.29577994e-01
7.29445219e-01 2.95616537e-01 -3.56104970e-01 3.55950072e-02
-1.38664317e+00 -4.51186538e-01 -2.96179414e-01 1.22375496e-01
4.08980161e-01 -7.73906052e-01 -3.81548375e-01 7.96845555e-02
-9.80588496e-01 9.65902507e-02 -8.25235844e-02 1.70795664e-01
-3.61846954e-01 4.52484041e-01 -6.47087991e-01 -2.42922485e-01
-7.18934894e-01 -6.88557863e-01 7.40903914e-01 6.87779859e-02
-3.40026289e-01 -9.29721057e-01 3.92037183e-01 6.04746878e-01
6.39253736e-01 -6.91138804e-02 1.10357046e+00 -1.46421862e+00
-2.12305367e-01 -5.26740670e-01 -9.40688923e-02 5.38882434e-01
-8.26201290e-02 -4.40625489e-01 -1.21882951e+00 -4.02369142e-01
1.30791932e-01 -8.05209458e-01 1.27443111e+00 -1.70128480e-01
1.34240282e+00 -5.52075863e-01 -2.59615481e-01 3.53304178e-01
1.51126754e+00 -1.88174453e-02 8.97040546e-01 6.36468470e-01
3.28042954e-01 5.54473817e-01 5.80274701e-01 -1.09203368e-01
-1.39826080e-02 4.83124286e-01 5.51348031e-01 3.24448436e-01
-2.88591832e-01 -6.86512351e-01 3.65432233e-01 3.47493857e-01
4.73684192e-01 -4.64879125e-01 -6.69413865e-01 6.69380963e-01
-1.74455976e+00 -1.26944053e+00 1.70914233e-01 2.30916142e+00
1.08675039e+00 2.00319752e-01 -2.78811395e-01 -1.13526963e-01
5.79857349e-01 -6.52966797e-02 -5.95085382e-01 -4.77186799e-01
-2.46298030e-01 1.26002476e-01 2.53338873e-01 4.60214227e-01
-9.79757845e-01 1.18781102e+00 6.62514496e+00 1.14983606e+00
-8.79347324e-01 1.66639328e-01 7.09872663e-01 -3.77461851e-01
-6.88859820e-01 -1.42581820e-01 -6.14867926e-01 3.86571288e-01
8.23814988e-01 -2.13652954e-01 2.63790518e-01 7.16600955e-01
-2.61749268e-01 4.06837493e-01 -1.43622863e+00 8.00977707e-01
3.42116416e-01 -1.12029326e+00 7.09329963e-01 -3.90871644e-01
7.45251834e-01 5.59749231e-02 3.28633517e-01 1.00984097e+00
5.65820456e-01 -1.44604897e+00 4.54544395e-01 5.51426888e-01
2.54288107e-01 -7.03299701e-01 7.78252244e-01 4.90192652e-01
-2.10543230e-01 -1.05717145e-01 -8.65355372e-01 -2.18073159e-01
-4.01131630e-01 3.44335474e-02 -7.38728285e-01 3.06708276e-01
7.73175061e-01 4.16177630e-01 -9.23177600e-01 7.71158576e-01
-2.39560515e-01 5.92943370e-01 1.03410207e-01 -1.67014450e-01
4.01131094e-01 3.81411970e-01 5.69851041e-01 1.16694355e+00
6.29209215e-04 -8.29672366e-02 -1.10787421e-01 6.67827070e-01
-4.30410862e-01 1.62810147e-01 -8.54622781e-01 2.93652833e-01
5.13579547e-01 9.77063537e-01 -1.29646763e-01 -4.57455903e-01
-2.78796822e-01 1.43371868e+00 6.12432361e-01 6.83374941e-01
-7.75932729e-01 -3.67660612e-01 5.41735888e-01 -2.39202511e-02
2.03999057e-02 5.50899446e-01 -9.70489457e-02 -1.18911982e+00
5.13925515e-02 -1.05570233e+00 7.58853793e-01 -9.69325781e-01
-1.65629506e+00 6.79925680e-01 -2.11639285e-01 -1.05842018e+00
-3.25479776e-01 -4.00908828e-01 -6.97927475e-01 8.77153158e-01
-1.78662574e+00 -1.21305382e+00 -1.07889533e-01 9.93235588e-01
8.06103110e-01 -4.59560812e-01 8.95308793e-01 1.13605686e-01
-1.27117932e-01 1.07277441e+00 2.79466957e-01 3.34545881e-01
1.15792954e+00 -1.14682686e+00 2.10671768e-01 7.58776486e-01
4.02658433e-01 1.04344857e+00 8.26155245e-01 -4.96144503e-01
-1.00872099e+00 -9.84601557e-01 7.99450040e-01 -1.10747313e+00
3.00630689e-01 -1.26797184e-02 -1.04653823e+00 7.80276597e-01
4.89357680e-01 2.24441011e-02 9.52026844e-01 -4.57118638e-02
-9.05621350e-01 -4.72140983e-02 -1.39125645e+00 7.05652118e-01
1.11235821e+00 -5.74025393e-01 -1.32641959e+00 3.23171794e-01
8.94483685e-01 3.96073684e-02 -3.61684829e-01 5.90534568e-01
5.20159543e-01 -1.01884258e+00 1.40306950e+00 -1.62413561e+00
6.96771801e-01 -3.66128013e-02 -1.75368682e-01 -1.27480578e+00
-4.54372793e-01 -4.76556927e-01 -1.06507922e-02 9.85151291e-01
3.28928113e-01 -3.46361339e-01 8.10766637e-01 6.33732855e-01
-1.00724429e-01 -4.04572129e-01 -7.47922480e-01 -8.32597435e-01
7.21630991e-01 -1.85144886e-01 6.62862957e-01 1.17184901e+00
9.07003786e-03 7.01507926e-01 -5.25294781e-01 -4.63342629e-02
4.13329780e-01 8.87432396e-02 5.74022174e-01 -9.17455733e-01
-1.62519053e-01 -5.12418628e-01 -3.03823173e-01 -9.38546300e-01
2.77530432e-01 -1.11290348e+00 -2.65027076e-01 -1.07249355e+00
6.56890094e-01 -5.31767085e-02 -1.15410149e+00 3.42537612e-01
-5.46425045e-01 5.11187851e-01 -1.23835802e-01 3.96499127e-01
-8.12708616e-01 4.61605579e-01 1.34866953e+00 -2.36189201e-01
1.91862553e-01 -3.18216145e-01 -1.00975120e+00 5.59803247e-01
8.21808875e-01 -8.05996001e-01 -8.72324347e-01 -6.91577256e-01
6.36601865e-01 -2.21682072e-01 6.46941543e-01 -8.05288374e-01
1.30165204e-01 -7.12657273e-02 5.21230817e-01 -4.04220670e-01
-7.06633925e-02 -8.07774127e-01 -3.21386606e-01 4.92527217e-01
-1.32095575e+00 4.37483704e-03 2.93852687e-01 9.24211681e-01
-3.32487136e-01 -7.12524414e-01 5.08790433e-01 -3.87260973e-01
-8.56794655e-01 2.37670273e-01 -1.59813881e-01 4.49323118e-01
7.23272026e-01 -9.95698869e-02 -5.15394688e-01 -5.29748797e-01
-5.86378157e-01 2.53160775e-01 2.99259216e-01 8.79215300e-01
6.42560184e-01 -1.37446260e+00 -9.36923265e-01 -4.08940986e-02
4.89841998e-01 -5.64268529e-01 4.01382536e-01 9.82171670e-02
-3.84096116e-01 5.47893763e-01 -2.97376513e-01 -8.63287896e-02
-1.18064213e+00 1.14574623e+00 4.28930163e-01 -4.98758465e-01
-3.45974207e-01 1.24900782e+00 4.70812410e-01 -5.19706070e-01
2.80225307e-01 -4.93570380e-02 -3.23377609e-01 -4.12551552e-01
8.21722329e-01 4.97813784e-02 1.79937676e-01 -3.75976890e-01
-3.36060733e-01 2.41524294e-01 -8.15281153e-01 1.27141014e-01
9.00519431e-01 -2.65729457e-01 2.76421458e-01 -1.14698440e-01
1.64145803e+00 -6.55986890e-02 -8.10320199e-01 -4.57288057e-01
1.32024467e-01 -6.00088418e-01 1.08015463e-02 -1.52936745e+00
-8.09767246e-01 1.06330383e+00 9.33289766e-01 6.48203194e-02
1.07191706e+00 -3.43254149e-01 1.01908731e+00 7.98367918e-01
5.67998402e-02 -1.01485085e+00 4.80005145e-01 5.09870172e-01
1.13218665e+00 -1.46186328e+00 -2.98011988e-01 1.14299618e-01
-6.41107678e-01 9.07405376e-01 9.04513955e-01 -4.79606390e-01
5.24174631e-01 -3.20611894e-01 2.21616969e-01 -2.03188524e-01
-9.54170465e-01 2.23181605e-01 4.92813736e-01 6.85082078e-01
4.27723229e-01 -4.24075872e-01 -3.75839353e-01 9.45236862e-01
-2.47816280e-01 -5.25835752e-02 2.52342552e-01 7.78771102e-01
-5.46722651e-01 -1.08107996e+00 -9.61686894e-02 4.59449142e-01
-1.08341348e+00 -7.66604066e-01 -6.89406753e-01 5.61601043e-01
-4.38130468e-01 7.32335508e-01 -7.40965502e-03 -4.68395025e-01
2.71115720e-01 3.64188641e-01 3.72800976e-01 -4.97993290e-01
-1.03186500e+00 -4.59811032e-01 -2.11097393e-02 -6.11851811e-01
-1.84645988e-02 -9.91115496e-02 -8.63343656e-01 -4.16289382e-02
-1.81881323e-01 2.74995059e-01 1.23602971e-01 9.90624130e-01
5.19654930e-01 3.36846292e-01 3.89605761e-01 -5.73304035e-02
-1.34588528e+00 -1.24581063e+00 -2.85230547e-01 1.36799300e+00
4.24240351e-01 -3.22233826e-01 -5.04866540e-01 4.66560572e-02] | [11.307647705078125, 7.6747589111328125] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.