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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
f2c9d8b5-d814-47e5-aca9-39f739faf324 | generating-representative-samples-for-few | 2205.02918 | null | https://arxiv.org/abs/2205.02918v1 | https://arxiv.org/pdf/2205.02918v1.pdf | Generating Representative Samples for Few-Shot Classification | Few-shot learning (FSL) aims to learn new categories with a few visual samples per class. Few-shot class representations are often biased due to data scarcity. To mitigate this issue, we propose to generate visual samples based on semantic embeddings using a conditional variational autoencoder (CVAE) model. We train this CVAE model on base classes and use it to generate features for novel classes. More importantly, we guide this VAE to strictly generate representative samples by removing non-representative samples from the base training set when training the CVAE model. We show that this training scheme enhances the representativeness of the generated samples and therefore, improves the few-shot classification results. Experimental results show that our method improves three FSL baseline methods by substantial margins, achieving state-of-the-art few-shot classification performance on miniImageNet and tieredImageNet datasets for both 1-shot and 5-shot settings. Code is available at: https://github.com/cvlab-stonybrook/fsl-rsvae. | ['Hieu Le', 'Jingyi Xu'] | 2022-05-05 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Xu_Generating_Representative_Samples_for_Few-Shot_Classification_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Xu_Generating_Representative_Samples_for_Few-Shot_Classification_CVPR_2022_paper.pdf | cvpr-2022-1 | ['classification'] | ['methodology'] | [ 4.92600687e-02 1.73734143e-01 -3.67077827e-01 -4.15332228e-01
-8.63498211e-01 -9.27938595e-02 7.99735665e-01 -2.02607527e-01
-2.97148913e-01 7.23960400e-01 3.40089172e-01 3.72038603e-01
2.43641078e-01 -9.70603883e-01 -8.69979441e-01 -6.34765387e-01
3.88176769e-01 3.18843395e-01 2.40840316e-01 -8.37986693e-02
2.50283186e-03 1.30672351e-01 -2.06907272e+00 3.75829786e-01
6.79453254e-01 8.73006821e-01 2.05302030e-01 6.10615015e-01
-2.27453504e-02 8.47827554e-01 -4.80578005e-01 -2.57228076e-01
2.62080878e-01 -6.38238072e-01 -5.88243961e-01 3.09356540e-01
3.87941122e-01 -5.24549007e-01 -3.93744290e-01 1.04123724e+00
4.91762847e-01 5.58799386e-01 1.03134668e+00 -1.56679547e+00
-1.07287085e+00 4.95203286e-01 -5.71837604e-01 3.31723809e-01
-1.04502723e-01 3.65010709e-01 9.73790050e-01 -1.32719767e+00
1.01943576e+00 1.10720766e+00 5.22722721e-01 1.04442942e+00
-1.27665055e+00 -6.86998963e-01 -8.45486149e-02 3.48106265e-01
-1.39110696e+00 -6.74424589e-01 7.31751859e-01 -5.09340227e-01
6.16264403e-01 -1.96762756e-01 6.85090363e-01 1.42968524e+00
-9.10225958e-02 1.07185674e+00 9.41683471e-01 -3.99457097e-01
6.98737741e-01 3.64316881e-01 2.72041321e-01 5.68453610e-01
1.88614011e-01 1.75633822e-02 -4.53466982e-01 9.00004804e-03
4.87558305e-01 4.37226683e-01 -3.34586829e-01 -6.96359694e-01
-6.78980708e-01 1.37703478e+00 6.28614783e-01 2.94504285e-01
-3.64872992e-01 2.01036841e-01 3.53814304e-01 1.97820389e-03
7.14794219e-01 2.59262383e-01 1.21677116e-01 4.99190986e-02
-9.51851130e-01 1.54215068e-01 4.96051341e-01 9.62681890e-01
8.91022086e-01 2.30805010e-01 -6.21745944e-01 1.17245710e+00
-1.37914777e-01 3.51498991e-01 7.36130297e-01 -1.04254091e+00
-1.75749198e-01 1.72309190e-01 -1.07572608e-01 -4.89396483e-01
6.18677102e-02 -3.13606471e-01 -5.34180582e-01 3.40727895e-01
-7.46577755e-02 8.78669471e-02 -1.41403019e+00 1.66840219e+00
3.36605012e-01 3.14611167e-01 1.37154933e-03 9.14927483e-01
1.16484535e+00 8.03483665e-01 2.97621340e-01 -9.06660035e-02
1.16355228e+00 -1.11661971e+00 -5.23055613e-01 -2.35289022e-01
3.93664062e-01 -1.80981159e-01 1.34146297e+00 5.56856347e-03
-7.36292422e-01 -6.27296090e-01 -1.15039968e+00 -1.17417328e-01
-4.10614818e-01 -2.13819630e-02 4.51831043e-01 4.82704431e-01
-7.53416896e-01 7.87957430e-01 -8.24706316e-01 -4.86891031e-01
1.16160309e+00 -4.10774380e-01 -2.62381047e-01 -5.15428483e-01
-1.05027509e+00 5.97745895e-01 3.68178904e-01 -6.74710572e-01
-1.30257237e+00 -1.12425315e+00 -1.06666315e+00 1.90919131e-01
2.18349233e-01 -5.60548604e-01 1.29192805e+00 -7.27092505e-01
-1.24623621e+00 8.84548664e-01 -1.03647061e-01 -5.14971375e-01
4.85949397e-01 1.08794242e-01 -1.84828296e-01 4.32919830e-01
3.45184863e-01 1.10202849e+00 1.03295100e+00 -1.43686080e+00
-4.69769299e-01 -6.84424639e-02 -1.61654755e-01 -2.16145627e-02
-3.87533218e-01 -3.96107763e-01 -2.20565662e-01 -5.50702929e-01
-4.59498525e-01 -6.95187628e-01 -5.07432707e-02 2.58079559e-01
-8.35880786e-02 -4.61721510e-01 6.47771239e-01 -3.98198187e-01
7.93779731e-01 -2.29862809e+00 -1.32155791e-01 -3.13120514e-01
4.80338126e-01 3.87798578e-01 -3.15472037e-01 2.18737751e-01
1.35091357e-02 -1.38656184e-01 -3.75182509e-01 -5.29046118e-01
9.13375393e-02 2.99239635e-01 -3.47667962e-01 4.54423457e-01
3.72689635e-01 1.01915205e+00 -1.14928818e+00 -4.35310632e-01
4.26059604e-01 6.99444413e-01 -6.60675764e-01 2.44297758e-01
-2.33773813e-01 2.88571101e-02 -6.19458482e-02 6.89467013e-01
6.19215548e-01 -3.89130771e-01 -2.22379565e-01 -9.48516279e-02
2.61088371e-01 -3.05531532e-01 -7.35056043e-01 1.63315690e+00
-4.57599431e-01 6.84467673e-01 -5.84776402e-01 -1.00734615e+00
8.12820792e-01 4.57214080e-02 1.13896780e-01 -4.73637015e-01
4.64740485e-01 -7.64102116e-02 -2.50529289e-01 -2.75162548e-01
3.11022490e-01 -4.89560604e-01 1.77838668e-01 4.43304420e-01
7.50641763e-01 -7.49264210e-02 4.25032079e-01 4.01160926e-01
7.93995023e-01 1.14794396e-01 4.36421603e-01 -4.86387871e-02
-1.68500781e-01 -2.02899843e-01 7.02763498e-01 9.25008535e-01
-5.79889238e-01 1.00516117e+00 2.61746705e-01 -2.56700784e-01
-1.29788184e+00 -1.31539416e+00 -1.64264888e-01 1.19150591e+00
-5.76869324e-02 -2.77415752e-01 -7.73530602e-01 -6.54190481e-01
5.90283014e-02 1.21441734e+00 -1.04637778e+00 -6.73834860e-01
1.17432147e-01 -5.79682231e-01 1.11651085e-01 5.89304328e-01
3.81662637e-01 -1.19952929e+00 -7.56532133e-01 1.18952163e-01
-2.39989832e-02 -1.00546098e+00 -1.92584261e-01 1.91958416e-02
-5.19942999e-01 -1.02769470e+00 -1.10627151e+00 -6.48417473e-01
6.86986268e-01 4.83810991e-01 9.54019129e-01 -2.19050199e-01
-7.59883940e-01 3.96309733e-01 -7.29556322e-01 -5.67092538e-01
-3.45807552e-01 -1.42111957e-01 1.72189735e-02 1.61621302e-01
7.58465052e-01 -5.64872384e-01 -7.48069286e-01 -1.84339583e-01
-9.30962563e-01 5.97458705e-02 3.24647039e-01 1.08117247e+00
6.73049569e-01 -3.94578665e-01 7.36345887e-01 -1.04380167e+00
3.99164289e-01 -8.11163247e-01 -2.21568674e-01 1.30728111e-01
-6.04351401e-01 -9.11579803e-02 7.37124503e-01 -6.03213131e-01
-1.05673432e+00 -2.52497643e-01 2.22463030e-02 -1.12646675e+00
-3.16735864e-01 1.17032817e-02 2.33732313e-01 1.28893733e-01
1.09174275e+00 4.14002806e-01 8.17292556e-02 -3.63148063e-01
7.04323292e-01 8.48794639e-01 3.33541483e-01 -2.33602822e-01
7.37850308e-01 7.40647137e-01 -4.48545307e-01 -9.91089702e-01
-1.09351075e+00 -5.39476395e-01 -4.83442277e-01 -5.01987815e-01
8.57784450e-01 -1.05763507e+00 3.45683433e-02 3.25804591e-01
-6.72914863e-01 -4.43851084e-01 -8.21422040e-01 3.48917484e-01
-7.52230942e-01 6.46842644e-02 -5.47901452e-01 -7.31003582e-01
-5.44993699e-01 -8.79420638e-01 9.87025738e-01 5.28542578e-01
-1.74306855e-01 -6.65400743e-01 1.60932064e-01 1.73566207e-01
2.97829509e-01 2.89082021e-01 6.43306792e-01 -6.30752027e-01
-2.56593943e-01 -2.48080999e-01 -2.88213879e-01 4.76570070e-01
-2.53264382e-02 -1.45823210e-01 -1.32585704e+00 -3.78177404e-01
-2.54837990e-01 -9.16572213e-01 1.33313847e+00 4.81365442e-01
1.23837626e+00 1.56707298e-02 -2.18840376e-01 5.97576857e-01
1.57672513e+00 -1.15463197e-01 5.44975519e-01 3.44186500e-02
6.31979823e-01 3.42558831e-01 5.58063626e-01 7.84451127e-01
2.18794167e-01 3.28388393e-01 1.22524604e-01 2.06713021e-01
-6.56041980e-01 -4.39823538e-01 9.64571387e-02 5.19270957e-01
5.89054525e-02 2.73870286e-02 -6.76847696e-01 9.41037953e-01
-1.69976223e+00 -1.27035129e+00 4.51738209e-01 1.93116188e+00
8.77380490e-01 -4.94416952e-02 2.18375847e-01 -3.06743793e-02
8.37288558e-01 3.83722633e-01 -8.48807395e-01 -9.31023806e-03
1.34343028e-01 3.93206686e-01 2.43126437e-01 2.23821238e-01
-9.96376395e-01 1.12308097e+00 5.27462864e+00 9.86545503e-01
-9.52602506e-01 5.15021145e-01 7.82285631e-01 -5.78547299e-01
-2.24360153e-01 -2.07488418e-01 -7.63456345e-01 6.02372289e-01
8.39474380e-01 -4.38250273e-01 4.74259913e-01 1.26301110e+00
-1.50578916e-01 1.69910714e-02 -9.20513213e-01 1.11094093e+00
5.37473857e-01 -1.57273400e+00 1.81643113e-01 -1.49548084e-01
1.06052458e+00 1.94848612e-01 1.51920661e-01 8.44247997e-01
4.05293792e-01 -7.29139507e-01 7.10930824e-01 5.28765619e-01
9.84307170e-01 -7.72861183e-01 3.41182530e-01 2.03150377e-01
-8.39659512e-01 -1.28440112e-01 -8.99429202e-01 8.58970582e-02
2.32384987e-02 5.62257230e-01 -7.40943849e-01 8.32477733e-02
7.35037386e-01 8.82188499e-01 -6.21198714e-01 1.03724945e+00
-2.89449126e-01 6.52944863e-01 -1.23900678e-02 -1.21673383e-01
1.99848309e-01 3.13900597e-03 4.40091610e-01 8.64555001e-01
2.41455257e-01 2.38273412e-01 2.39057615e-02 1.23588991e+00
-3.63176614e-01 -1.28549844e-01 -6.07926548e-01 -1.76386878e-01
6.71924114e-01 1.22685874e+00 -7.28020966e-01 -6.85525179e-01
-4.32296813e-01 1.02128744e+00 5.85055292e-01 4.21159923e-01
-8.05472136e-01 -6.73821688e-01 6.90186381e-01 6.85645118e-02
6.37750030e-01 4.67877984e-01 6.83965608e-02 -1.43289375e+00
-2.58804739e-01 -5.93384683e-01 4.06726718e-01 -9.81738567e-01
-1.53281701e+00 4.94622469e-01 1.09221697e-01 -1.24433577e+00
-2.74942428e-01 -3.91380668e-01 -7.62185633e-01 4.56763297e-01
-1.38150334e+00 -1.18583226e+00 -5.71712434e-01 3.57241809e-01
9.29730773e-01 -2.80282944e-01 6.88101530e-01 8.81963074e-02
-4.31422800e-01 6.28576577e-01 3.39523971e-01 1.43991053e-01
6.55546129e-01 -1.07104552e+00 3.50370258e-01 8.55469584e-01
1.77501157e-01 3.06537062e-01 8.27498317e-01 -5.03707647e-01
-7.49287903e-01 -1.26676679e+00 5.33587277e-01 -4.42732722e-01
5.41614413e-01 -4.63279843e-01 -1.00573540e+00 6.97523892e-01
1.53105110e-01 5.18296123e-01 9.12202835e-01 -1.71649039e-01
-5.49586892e-01 1.64644271e-01 -1.27968407e+00 6.04353607e-01
1.04333961e+00 -6.28573477e-01 -8.01679492e-01 3.14400703e-01
7.83846676e-01 1.58655182e-01 -6.92365050e-01 1.63818493e-01
4.59207714e-01 -8.85578275e-01 9.02471423e-01 -7.00962007e-01
8.32854152e-01 -8.41197595e-02 -3.24758351e-01 -1.75657785e+00
-5.53337157e-01 3.31791490e-02 -4.87583280e-01 1.06287992e+00
5.72275668e-02 -4.23683733e-01 7.95243561e-01 3.79268199e-01
-6.72047362e-02 -6.91989005e-01 -6.76846206e-01 -9.44168091e-01
1.74749076e-01 -1.13381833e-01 4.60600466e-01 8.97518635e-01
-7.18048215e-02 3.00648689e-01 -4.74453747e-01 -3.57586920e-01
1.19100082e+00 2.20378071e-01 7.70878434e-01 -1.11927938e+00
-3.51942033e-01 -3.61917257e-01 -4.18531179e-01 -4.10636634e-01
3.47211480e-01 -9.93706822e-01 2.16295525e-01 -1.61063600e+00
6.18309140e-01 2.54509617e-02 -3.49652410e-01 5.05568206e-01
-3.90665114e-01 6.75387859e-01 4.84063447e-01 1.15322754e-01
-7.47460067e-01 1.06033587e+00 9.92205739e-01 -2.49094725e-01
-2.55264398e-02 -2.31750727e-01 -6.88779831e-01 4.51844305e-01
8.35408509e-01 -4.91518855e-01 -5.75530946e-01 5.50556704e-02
-4.87906158e-01 -3.89714479e-01 3.50108147e-01 -1.23217273e+00
-1.40542597e-01 -8.38308930e-02 7.68000185e-01 -2.94438213e-01
5.07492006e-01 -1.76845446e-01 -1.78587601e-01 5.32708704e-01
-3.69175971e-01 -9.90267396e-01 -1.62643954e-01 7.51109242e-01
-4.05276604e-02 -4.48663056e-01 1.26058292e+00 -2.65268952e-01
-1.13308895e+00 6.78338230e-01 -2.66365469e-01 2.88974792e-01
1.18028593e+00 -2.13777676e-01 -5.36894202e-01 -3.93874496e-01
-8.46616924e-01 1.33200586e-01 6.20121539e-01 5.93427598e-01
7.65209496e-01 -1.59752882e+00 -6.44497871e-01 2.22612441e-01
7.76800811e-01 -2.94539213e-01 6.27464771e-01 5.07250369e-01
-3.17395806e-01 -1.57446079e-02 -5.20346940e-01 -4.07324493e-01
-9.72992301e-01 7.37876117e-01 1.60457671e-01 3.07582617e-01
-9.05832827e-01 9.97647345e-01 2.97336370e-01 -2.68605173e-01
5.28032631e-02 1.86541572e-01 -2.20621020e-01 2.86962926e-01
9.12936985e-01 3.86374861e-01 -3.31051618e-01 -4.69624370e-01
-9.42209214e-02 1.23214006e-01 -2.45459139e-01 2.99519841e-02
1.68108737e+00 2.05977336e-02 5.40175140e-01 7.17518747e-01
1.43704629e+00 -6.10486150e-01 -1.73246908e+00 -4.00920182e-01
-4.45575744e-01 -6.52008057e-01 1.76955193e-01 -4.66278076e-01
-9.91291285e-01 1.05344856e+00 7.36342072e-01 -1.01264760e-01
7.38494873e-01 2.33914942e-01 8.37991953e-01 1.11591458e-01
3.81002158e-01 -1.23650801e+00 3.29449356e-01 2.09480494e-01
7.24012554e-01 -1.47832227e+00 -1.30967885e-01 -4.65291217e-02
-9.39335108e-01 7.58366287e-01 7.08545089e-01 -5.57310164e-01
6.94254518e-01 -3.04283559e-01 -8.02973732e-02 -1.54750139e-01
-8.01538229e-01 -5.05772650e-01 2.20932573e-01 9.42946970e-01
9.15626362e-02 1.98041722e-01 -3.48516107e-02 7.75844336e-01
-4.96522598e-02 2.45595053e-01 8.02103221e-01 9.80147064e-01
-7.05313683e-01 -5.29972732e-01 9.15324781e-03 8.48196328e-01
-7.07453266e-02 -9.43923742e-02 -9.75107476e-02 6.87362432e-01
-1.18600652e-01 7.05215037e-01 3.20169866e-01 -2.66840786e-01
8.72997120e-02 4.06995475e-01 5.58736861e-01 -8.88608754e-01
2.22618490e-01 -1.44428283e-01 -2.59659261e-01 -5.09634852e-01
-1.76093042e-01 -5.87627888e-01 -9.73294258e-01 -2.71333992e-01
-8.32544193e-02 5.20141199e-02 3.42592716e-01 7.51626253e-01
4.66429293e-01 5.90829611e-01 6.60556734e-01 -1.15283942e+00
-7.36284792e-01 -1.18241191e+00 -8.34127605e-01 7.87578762e-01
2.03182817e-01 -1.03902698e+00 -6.42544329e-01 7.58833215e-02] | [9.94788646697998, 2.8009345531463623] |
e369aebc-d442-48ef-8677-c5d398dbc9fb | kpt-keyword-guided-pre-training-for-grounded | 2212.01739 | null | https://arxiv.org/abs/2212.01739v1 | https://arxiv.org/pdf/2212.01739v1.pdf | KPT: Keyword-guided Pre-training for Grounded Dialog Generation | Incorporating external knowledge into the response generation process is essential to building more helpful and reliable dialog agents. However, collecting knowledge-grounded conversations is often costly, calling for a better pre-trained model for grounded dialog generation that generalizes well w.r.t. different types of knowledge. In this work, we propose KPT (Keyword-guided Pre-Training), a novel self-supervised pre-training method for grounded dialog generation without relying on extra knowledge annotation. Specifically, we use a pre-trained language model to extract the most uncertain tokens in the dialog as keywords. With these keywords, we construct two kinds of knowledge and pre-train a knowledge-grounded response generation model, aiming at handling two different scenarios: (1) the knowledge should be faithfully grounded; (2) it can be selectively used. For the former, the grounding knowledge consists of keywords extracted from the response. For the latter, the grounding knowledge is additionally augmented with keywords extracted from other utterances in the same dialog. Since the knowledge is extracted from the dialog itself, KPT can be easily performed on a large volume and variety of dialogue data. We considered three data sources (open-domain, task-oriented, conversational QA) with a total of 2.5M dialogues. We conduct extensive experiments on various few-shot knowledge-grounded generation tasks, including grounding on dialog acts, knowledge graphs, persona descriptions, and Wikipedia passages. Our comprehensive experiments and analyses demonstrate that KPT consistently outperforms state-of-the-art methods on these tasks with diverse grounding knowledge. | ['Minlie Huang', 'Xiaoyan Zhu', 'Qun Liu', 'Xin Jiang', 'Yitong Li', 'Yasheng Wang', 'Zheng Zhang', 'Fei Mi', 'Qi Zhu'] | 2022-12-04 | null | null | null | null | ['response-generation'] | ['natural-language-processing'] | [ 8.12604502e-02 8.83760333e-01 -4.69888970e-02 -4.53843355e-01
-1.01448441e+00 -6.74304426e-01 8.04534733e-01 1.04549013e-01
-3.55460227e-01 1.20693982e+00 7.07813919e-01 -2.04617649e-01
1.59579381e-01 -9.36855257e-01 -3.58557820e-01 -3.78368914e-01
3.98514211e-01 1.16264760e+00 3.56745571e-01 -9.57468927e-01
6.90059587e-02 -1.65430203e-01 -9.05599058e-01 6.02664948e-01
1.18593347e+00 7.58290827e-01 3.53263676e-01 4.56018120e-01
-6.78896010e-01 1.03849900e+00 -9.23673928e-01 -9.45567489e-01
-9.22927335e-02 -7.43476391e-01 -1.63995421e+00 2.32866213e-01
-3.16358954e-01 -4.01862353e-01 -3.28568876e-01 7.28977323e-01
4.64385271e-01 5.24801254e-01 5.58962762e-01 -1.17434263e+00
-6.31223381e-01 1.37033939e+00 2.50230819e-01 -1.69531107e-01
7.87199199e-01 3.32834125e-01 9.75073397e-01 -8.56703222e-01
7.55582154e-01 1.58843172e+00 4.04974252e-01 1.11765337e+00
-8.44721437e-01 -2.40101039e-01 1.33011356e-01 6.55328333e-02
-1.09773207e+00 -4.87898022e-01 6.39592528e-01 -2.76614934e-01
1.00060081e+00 8.46492127e-02 3.26340675e-01 1.46975493e+00
-4.37390059e-01 7.94325709e-01 9.55179870e-01 -4.87189710e-01
1.39613032e-01 6.18165195e-01 4.05304134e-01 6.96582556e-01
-2.93467820e-01 -2.75919259e-01 -7.54821301e-01 -5.74681580e-01
3.28677982e-01 -4.44776297e-01 -4.71935332e-01 3.17092419e-01
-1.29479349e+00 1.09113216e+00 1.75594062e-01 2.14248270e-01
-3.66885096e-01 -5.18325686e-01 4.02353734e-01 2.19785556e-01
3.70893568e-01 7.00921357e-01 -5.57363927e-01 -3.38523597e-01
-3.07508439e-01 4.50136840e-01 1.45013440e+00 1.33782959e+00
9.38344061e-01 -2.01219201e-01 -7.96075284e-01 1.05171108e+00
2.01189801e-01 4.12267804e-01 6.81895554e-01 -8.48998070e-01
9.17182922e-01 1.00829184e+00 4.93963569e-01 -6.06185436e-01
-3.36853683e-01 2.72153378e-01 -6.51447356e-01 -6.06163263e-01
6.09519243e-01 -7.55032480e-01 -5.66939175e-01 1.77205789e+00
4.89342839e-01 -1.11531742e-01 7.76755989e-01 6.89037204e-01
1.47033572e+00 9.06374812e-01 3.46259810e-02 -3.93239349e-01
1.51652682e+00 -1.11948419e+00 -9.59606528e-01 -2.15676621e-01
6.90167427e-01 -4.62964445e-01 1.23603868e+00 2.88873957e-03
-8.39606822e-01 -3.99384499e-01 -6.31985724e-01 -5.12802750e-02
-4.63236690e-01 -6.57509267e-02 3.58025014e-01 4.47864473e-01
-7.80520320e-01 2.63087749e-01 -1.86319411e-01 -3.55300307e-01
-2.44304284e-01 -3.56725343e-02 -6.70002773e-02 -5.36185168e-02
-1.94437969e+00 1.05267322e+00 7.41262376e-01 1.24670893e-01
-1.00563312e+00 -4.03889924e-01 -9.95159030e-01 1.21688461e-02
8.99183810e-01 -8.09157312e-01 1.74168110e+00 -4.89195406e-01
-1.98719931e+00 6.29678011e-01 -1.55383214e-01 -4.13393527e-01
4.15515065e-01 -2.03431740e-01 -2.84149975e-01 9.19521451e-02
1.69773892e-01 5.82564712e-01 4.82192606e-01 -1.45485246e+00
-5.34976780e-01 -6.43297508e-02 5.80632389e-01 5.29586911e-01
-4.02081311e-01 -1.01292394e-01 -3.64403188e-01 -2.75239140e-01
-3.61401528e-01 -1.01597679e+00 -2.84367919e-01 -9.64118719e-01
-9.10773993e-01 -7.38747418e-01 4.60407764e-01 -7.23978281e-01
1.21958983e+00 -1.66871560e+00 2.58063823e-01 -9.28304940e-02
1.13805421e-01 3.69778633e-01 -2.27759965e-02 9.57698464e-01
7.07221806e-01 9.62563008e-02 -1.68511316e-01 -3.22018355e-01
1.67854354e-01 3.85135651e-01 -6.32323980e-01 -5.28286099e-01
1.94825158e-01 9.54679966e-01 -1.18871057e+00 -7.43084371e-01
-6.55283108e-02 -5.38200252e-02 -2.88894564e-01 8.16340268e-01
-6.94440603e-01 6.72058582e-01 -7.99792349e-01 2.84524649e-01
6.39016554e-02 -2.47245058e-01 2.46751443e-01 -1.98873237e-01
1.86761871e-01 7.39306509e-01 -8.82159352e-01 1.73160505e+00
-6.82624876e-01 1.80607349e-01 -8.45630020e-02 -4.94097382e-01
1.05053174e+00 7.44286835e-01 1.98259950e-02 -1.79026499e-01
2.09064513e-01 5.44402609e-03 -1.05702788e-01 -6.66541100e-01
9.34748650e-01 -2.91536480e-01 -5.12838006e-01 6.30093694e-01
4.73320127e-01 -3.93887043e-01 3.07586610e-01 6.74756885e-01
9.18048322e-01 -2.11867794e-01 2.90656954e-01 7.56859258e-02
6.38444543e-01 4.56974626e-01 4.38276619e-01 7.55502701e-01
3.33157810e-03 1.43714070e-01 4.27843899e-01 -9.02317651e-03
-3.92372400e-01 -7.69565403e-01 3.10297132e-01 1.26417196e+00
2.56712288e-01 -7.37633169e-01 -9.84263957e-01 -9.32739258e-01
-3.50479484e-01 1.03440607e+00 -3.65176082e-01 -3.19442749e-01
-5.58401406e-01 -4.01539266e-01 8.86060297e-01 4.23820853e-01
8.30861390e-01 -1.48204947e+00 -1.51969030e-01 4.62128162e-01
-9.31917489e-01 -1.40582120e+00 -5.37904441e-01 -1.23231262e-01
-4.89916652e-01 -1.01609278e+00 -5.11911154e-01 -7.34241605e-01
4.92045403e-01 9.69009772e-02 1.35083592e+00 8.18595141e-02
4.19299722e-01 5.36282003e-01 -8.03162932e-01 -3.22604835e-01
-9.82334971e-01 2.50655234e-01 -1.84056647e-02 -2.26483755e-02
4.62292194e-01 -4.17524576e-02 -4.21049334e-02 4.59845603e-01
-5.38839996e-01 2.20360443e-01 2.66604692e-01 1.20393407e+00
1.01554014e-01 8.30501784e-03 1.09899831e+00 -1.02440357e+00
1.39511693e+00 -5.22243679e-01 -1.85895383e-01 6.38275981e-01
-1.89219251e-01 1.83061644e-01 7.51896143e-01 -5.14626086e-01
-1.75240886e+00 -2.07541659e-01 -2.38366768e-01 4.60174978e-02
-1.65597290e-01 7.46576726e-01 -3.60507190e-01 4.22922343e-01
9.23045874e-01 2.07886979e-01 -1.48845643e-01 -3.60272229e-01
8.08284461e-01 1.10802007e+00 5.40882111e-01 -1.16117787e+00
5.91127396e-01 -6.35056198e-02 -6.72543526e-01 -8.12873363e-01
-1.13441944e+00 -5.13251185e-01 -5.14394879e-01 -2.25725263e-01
8.26126456e-01 -6.60865009e-01 -7.07708418e-01 3.86200368e-01
-1.49379575e+00 -7.46802568e-01 -3.16407263e-01 2.53325254e-01
-5.29505432e-01 2.99574912e-01 -7.62545288e-01 -9.90872562e-01
-6.83924794e-01 -9.47916448e-01 8.20941567e-01 4.25258517e-01
-4.28473085e-01 -1.12026608e+00 2.10905463e-01 7.87193954e-01
1.13837674e-01 -1.12282939e-01 9.89270151e-01 -1.38920116e+00
-2.20842838e-01 9.94134396e-02 1.03067033e-01 2.39461884e-01
3.72679770e-01 -4.08054441e-01 -9.95072305e-01 4.24329527e-02
9.74005237e-02 -1.02828741e+00 3.91213208e-01 -2.63663143e-01
5.53543150e-01 -9.33381557e-01 -1.89619631e-01 -2.44329438e-01
6.26803041e-01 2.64669240e-01 4.40331817e-01 -1.01457164e-01
5.29272258e-01 1.18390644e+00 9.36851263e-01 4.50472623e-01
7.58022308e-01 8.30511093e-01 -3.36251408e-02 2.16796070e-01
9.76459756e-02 -5.87840736e-01 3.79918903e-01 9.69591916e-01
-3.73811796e-02 -4.34185326e-01 -8.85109663e-01 6.78551674e-01
-2.03104067e+00 -1.01669300e+00 1.48723334e-01 1.98494661e+00
1.71498907e+00 9.93890129e-03 2.24994749e-01 -3.41773748e-01
8.44217181e-01 -5.90745807e-02 -4.45078850e-01 -1.77354097e-01
4.57087345e-02 -1.01420514e-01 -2.95782536e-01 8.30379009e-01
-6.20871067e-01 1.34330678e+00 5.01571465e+00 8.13578010e-01
-6.07877374e-01 2.40007818e-01 4.03402060e-01 4.39163625e-01
-3.85558158e-01 9.81258005e-02 -1.23710024e+00 3.88205498e-01
9.80761349e-01 -6.44396901e-01 2.85470545e-01 9.43572521e-01
-1.07383206e-01 1.27496660e-01 -1.12873459e+00 5.56475818e-01
-3.44273858e-02 -1.28093469e+00 3.20029676e-01 -2.40857586e-01
6.23928964e-01 -3.72453868e-01 -5.33001184e-01 9.21622515e-01
7.88289011e-01 -8.40271175e-01 4.14333552e-01 2.55129218e-01
5.83478153e-01 -4.74211276e-01 9.74511921e-01 7.96074808e-01
-1.00203037e+00 1.20272674e-01 -3.20190907e-01 2.09138572e-01
6.27971828e-01 2.70189673e-01 -1.52883196e+00 9.29266334e-01
2.96415091e-01 1.58955395e-01 -1.11741796e-01 3.38726968e-01
-5.83203495e-01 5.73350012e-01 -7.44461715e-02 -3.19191158e-01
3.55168909e-01 -6.88501596e-02 4.63562965e-01 1.29699421e+00
-4.30143364e-02 7.05649674e-01 5.52616298e-01 9.75895882e-01
-2.60327548e-01 2.27766708e-01 -6.17747247e-01 -2.20704973e-01
8.71220171e-01 1.35779786e+00 -3.53257924e-01 -7.19592869e-01
-3.34508806e-01 8.39083970e-01 3.79540473e-01 3.36563021e-01
-5.07187545e-01 -4.76563066e-01 2.34237194e-01 -2.76344985e-01
-1.22696847e-01 -4.49494645e-02 1.99491322e-01 -1.17587137e+00
-2.24395305e-01 -9.86141205e-01 5.46807051e-01 -7.19178975e-01
-1.68415880e+00 1.07230377e+00 3.98312956e-01 -8.60918343e-01
-1.00240719e+00 -4.50676858e-01 -6.56559825e-01 9.26366866e-01
-1.24896026e+00 -1.12439811e+00 -5.86048543e-01 8.61871243e-01
8.59995246e-01 -2.78568506e-01 1.11385167e+00 -1.50481269e-01
-3.80927444e-01 5.04066229e-01 -5.80625713e-01 4.84611332e-01
8.05305004e-01 -1.26959908e+00 2.66609728e-01 5.33604622e-01
-4.99889590e-02 9.16619122e-01 6.29876256e-01 -9.18406069e-01
-1.17637217e+00 -1.07742774e+00 1.05192995e+00 -6.58820450e-01
6.14075065e-01 -2.66822010e-01 -1.18085504e+00 7.32794940e-01
4.87274557e-01 -7.24100828e-01 9.12139773e-01 3.81991237e-01
-1.66821703e-01 3.36416274e-01 -1.10637963e+00 6.82516515e-01
9.06491876e-01 -4.86070901e-01 -1.23539782e+00 5.57022691e-01
1.16607940e+00 -7.48635173e-01 -9.33571577e-01 2.46033907e-01
-1.95550174e-03 -4.88459140e-01 7.62505174e-01 -8.40950668e-01
1.89925060e-01 -5.70673347e-02 -8.17330033e-02 -1.61427069e+00
2.21655801e-01 -1.09470809e+00 -2.55289286e-01 1.69372118e+00
6.57178164e-01 -4.93395746e-01 4.36932445e-01 1.01499796e+00
-5.07506371e-01 -6.82995021e-01 -7.74817646e-01 -8.17850351e-01
-6.89798640e-03 6.83217123e-02 8.32696378e-01 1.02160454e+00
7.74606287e-01 1.11412168e+00 -4.97483253e-01 -8.89662653e-02
6.41825050e-02 1.78189665e-01 1.08632898e+00 -1.09061396e+00
-1.98826790e-01 2.96469945e-02 3.11208457e-01 -1.45614243e+00
5.28070986e-01 -6.32574737e-01 5.21331549e-01 -1.74953854e+00
8.15333724e-02 -5.89040577e-01 3.90646160e-01 8.54814649e-01
-6.26654387e-01 -4.85773474e-01 5.39203137e-02 1.15713425e-01
-5.16731501e-01 9.75479662e-01 1.23381293e+00 -1.30157188e-01
-6.98319852e-01 1.95878506e-01 -8.11708272e-01 5.91803551e-01
8.09519947e-01 -2.76922733e-01 -8.69934738e-01 -6.36270791e-02
-3.40386890e-02 5.21070778e-01 2.19928637e-01 -4.49481517e-01
4.55506593e-01 -4.71159369e-01 -4.57923293e-01 -4.38275307e-01
6.94224775e-01 -3.94271910e-01 -2.96779394e-01 2.52036694e-02
-5.32642901e-01 -1.81443810e-01 1.71280518e-01 4.63244379e-01
-3.22635174e-01 -6.68236434e-01 3.99832278e-01 -5.23327768e-01
-7.16448367e-01 2.02269889e-02 -3.76448542e-01 6.69051170e-01
6.20789528e-01 1.45583227e-01 -7.23060489e-01 -7.73290873e-01
-5.84508896e-01 5.32764077e-01 8.59000683e-02 4.62626338e-01
6.02026701e-01 -1.25606382e+00 -7.62283206e-01 -3.40588242e-01
3.28282505e-01 2.09498331e-01 2.98408329e-01 4.20477748e-01
2.51461267e-01 5.54676235e-01 4.01698612e-02 -2.70858139e-01
-1.02729523e+00 4.58685279e-01 2.76188463e-01 -5.01833916e-01
-4.44657266e-01 8.05734694e-01 7.84875602e-02 -7.37747610e-01
2.58856326e-01 -1.19518518e-01 -5.02238333e-01 1.62870586e-01
6.23052418e-01 1.77861959e-01 -6.51228502e-02 -6.37191176e-01
-4.40373570e-02 -4.18797657e-02 -2.36608014e-01 -5.59904218e-01
8.19559515e-01 -1.85064018e-01 -4.97828983e-02 5.15970767e-01
5.07191181e-01 5.50028794e-02 -9.91736650e-01 -7.75281131e-01
4.19240631e-02 -6.97323233e-02 -4.98830646e-01 -1.06636512e+00
-5.90541780e-01 5.93400419e-01 -3.65604401e-01 4.84330803e-01
6.59442127e-01 2.94603497e-01 9.79136050e-01 9.51023161e-01
7.24202275e-01 -1.13811111e+00 5.47121108e-01 1.01189339e+00
1.10331869e+00 -1.25684679e+00 -3.70575249e-01 -6.67031646e-01
-1.36884809e+00 9.68834043e-01 1.12022614e+00 6.52197480e-01
1.66763842e-01 -1.69658512e-01 2.93875664e-01 -3.64996970e-01
-8.00092280e-01 -3.42444122e-01 2.75727898e-01 6.77413046e-01
2.15934440e-01 -4.93448563e-02 -1.40778854e-01 1.19501412e+00
-4.76388961e-01 -2.25182056e-01 4.90674466e-01 8.66386831e-01
-5.46545088e-01 -1.11820042e+00 -7.19992295e-02 1.83718100e-01
1.02656856e-01 -3.09602976e-01 -1.01354015e+00 6.33350909e-01
-3.45725507e-01 1.69347119e+00 -5.74087262e-01 -6.12121642e-01
4.60595161e-01 5.41877031e-01 1.75896809e-01 -1.15836728e+00
-9.43586588e-01 -4.85483319e-01 9.05101001e-01 -1.30131826e-01
-4.01758820e-01 -1.83699757e-01 -1.45508540e+00 -2.12347344e-01
-6.72179818e-01 8.85103762e-01 1.66831970e-01 1.26774466e+00
2.68205673e-01 2.38441154e-01 6.02627575e-01 -4.75911349e-01
-8.82600129e-01 -1.46407306e+00 -1.81684509e-01 5.63823283e-01
-1.63461104e-01 -7.46734023e-01 -2.73320377e-01 2.01117828e-01] | [12.534274101257324, 8.103514671325684] |
31d82806-4c5f-44f1-ba7a-102826ec3197 | robust-human-detection-under-visual | 2307.03623 | null | https://arxiv.org/abs/2307.03623v1 | https://arxiv.org/pdf/2307.03623v1.pdf | Robust Human Detection under Visual Degradation via Thermal and mmWave Radar Fusion | The majority of human detection methods rely on the sensor using visible lights (e.g., RGB cameras) but such sensors are limited in scenarios with degraded vision conditions. In this paper, we present a multimodal human detection system that combines portable thermal cameras and single-chip mmWave radars. To mitigate the noisy detection features caused by the low contrast of thermal cameras and the multi-path noise of radar point clouds, we propose a Bayesian feature extractor and a novel uncertainty-guided fusion method that surpasses a variety of competing methods, either single-modal or multi-modal. We evaluate the proposed method on real-world data collection and demonstrate that our approach outperforms the state-of-the-art methods by a large margin. | ['Chris Xiaoxuan Lu', 'John Stankovic', 'Peize Li', 'Qiyue Xia', 'Kaiwen Cai'] | 2023-07-07 | null | null | null | null | ['human-detection'] | ['computer-vision'] | [ 3.21295291e-01 -5.05886257e-01 5.62975526e-01 -2.65747607e-01
-8.88787985e-01 -6.10321522e-01 5.95248640e-01 -6.24422610e-01
-6.45310163e-01 5.57536721e-01 -1.86088711e-01 1.66774929e-01
5.13676740e-02 -5.70631802e-01 -2.40420103e-01 -8.75983834e-01
6.55199885e-01 7.86549971e-02 5.64554155e-01 4.96348068e-02
-2.98623499e-02 2.10329875e-01 -1.73137665e+00 4.46983725e-02
5.79181015e-01 1.29254174e+00 1.81569889e-01 7.40685582e-01
3.34951729e-01 1.26108006e-01 -6.10452652e-01 -3.91746432e-01
6.37722671e-01 -5.91108128e-02 6.03390813e-01 1.06526859e-01
7.58246541e-01 -6.09789670e-01 -4.87326324e-01 1.04442084e+00
5.58773100e-01 -2.49832988e-01 6.73625708e-01 -1.20850003e+00
-3.97052705e-01 -2.39078239e-01 -9.13363934e-01 2.43923049e-02
6.43484533e-01 5.01201630e-01 4.58831280e-01 -1.08112037e+00
2.27905475e-02 1.09002852e+00 7.56875157e-01 4.66561615e-01
-9.94382262e-01 -8.29703927e-01 -3.32253069e-01 1.79534525e-01
-1.51467085e+00 -5.18998086e-01 7.05147266e-01 -4.32955295e-01
6.98749304e-01 9.49824154e-02 5.13023138e-01 1.28707862e+00
5.76865077e-01 3.01036805e-01 1.62849748e+00 -4.25311655e-01
2.57993698e-01 3.56444679e-02 1.89010993e-01 8.61408889e-01
9.49145675e-01 7.15144634e-01 -1.01637590e+00 -4.07157153e-01
4.20375258e-01 2.64510125e-01 -3.87679130e-01 -3.18641573e-01
-9.65628505e-01 4.68699872e-01 9.20735300e-02 -8.92383978e-02
-5.57886779e-01 3.07650328e-01 -3.78349066e-01 -2.00537875e-01
2.41591521e-02 1.01796210e-01 1.04850441e-01 2.35997885e-01
-1.27961147e+00 -1.32452875e-01 4.57608610e-01 9.05825198e-01
7.72055209e-01 1.43990545e-02 -1.27642632e-01 1.78342775e-01
1.00144076e+00 1.56584895e+00 -1.14355251e-01 -6.70475125e-01
2.37600774e-01 3.83886933e-01 4.89743978e-01 -8.14124882e-01
-4.78639603e-01 -2.84110934e-01 -7.97386348e-01 6.28499448e-01
3.58783901e-01 -4.07514572e-01 -1.23093343e+00 1.14643693e+00
1.60789534e-01 2.61677533e-01 1.95952505e-01 1.28189516e+00
8.00580442e-01 3.45934182e-01 -4.15253252e-01 -2.84791499e-01
1.43256950e+00 -2.73137242e-01 -7.00089812e-01 -6.38851345e-01
-4.68674719e-01 -9.52861428e-01 2.18801990e-01 9.33595955e-01
-4.99133289e-01 -5.05879581e-01 -1.36369896e+00 5.46537042e-01
5.04997931e-02 3.84771019e-01 3.34687680e-01 1.32443047e+00
-7.10446179e-01 -1.43282777e-02 -1.05098152e+00 -6.35812521e-01
1.24990970e-01 7.08553120e-02 -4.99128178e-02 -4.47444826e-01
-8.60932589e-01 1.02077460e+00 -1.44236563e-02 3.26181024e-01
-7.21327066e-01 -3.27576429e-01 -4.87459093e-01 -3.40252846e-01
5.40779173e-01 -8.86024535e-01 7.89664447e-01 -1.93524018e-01
-1.59692323e+00 5.35849452e-01 1.38201369e-02 -4.00948972e-01
4.47108805e-01 -6.69556499e-01 -4.73819196e-01 5.36040127e-01
-4.10197675e-01 2.10580438e-01 1.26490211e+00 -1.30193889e+00
-6.98246360e-01 -6.99489832e-01 -3.45398873e-01 4.95932996e-02
-2.23656604e-03 -2.14969404e-02 -2.79184341e-01 2.76107341e-02
3.85038495e-01 -1.08055019e+00 -1.21944487e-01 2.26895586e-02
-4.76020873e-01 3.65978360e-01 8.57678235e-01 -1.90055817e-01
6.85852945e-01 -1.97277772e+00 -2.89136142e-01 2.44474024e-01
1.65881962e-01 1.29543737e-01 2.10079387e-01 4.24496084e-01
7.26068139e-01 -5.16232312e-01 -3.51061374e-01 -4.39070851e-01
1.28614128e-01 1.19475788e-02 -2.29433373e-01 9.67201471e-01
-1.67888582e-01 5.19475281e-01 -5.06400943e-01 -3.72442007e-01
7.51008391e-01 7.68555522e-01 3.68992418e-01 8.42301100e-02
7.99401924e-02 4.39433157e-01 -3.01028252e-01 1.12976360e+00
1.01215303e+00 2.19036639e-01 -1.73657745e-01 -4.47604209e-01
-3.77673030e-01 -4.46022600e-01 -1.55175722e+00 1.49676335e+00
-7.22873732e-02 5.93661189e-01 3.92240405e-01 -2.14912131e-01
1.09442377e+00 1.44853979e-01 1.90994918e-01 -5.36646187e-01
5.91322221e-02 1.85034290e-01 -3.77855927e-01 -3.73011887e-01
6.32789135e-01 -3.93530339e-01 -1.43158406e-01 1.77785650e-01
-5.25964200e-02 -1.17662907e-01 -3.09689850e-01 5.62751517e-02
1.42739463e+00 1.69209585e-01 1.50429741e-01 1.54268727e-01
1.56571805e-01 -1.54833049e-01 7.44449139e-01 1.05702865e+00
-4.33879226e-01 8.77238512e-01 -4.48076457e-01 4.94093783e-02
-6.16647065e-01 -1.49424171e+00 3.58295441e-03 4.96687233e-01
5.12722373e-01 -2.86079139e-01 -2.43661255e-01 -1.87181234e-01
1.15000263e-01 5.01725912e-01 -1.39011264e-01 -4.82839495e-02
-8.26155171e-02 -1.13430905e+00 6.66709960e-01 5.38627267e-01
6.83774829e-01 -2.29123145e-01 -1.67717671e+00 -8.26168209e-02
-9.05061215e-02 -1.60697973e+00 2.24503770e-01 5.51317856e-02
-4.88294780e-01 -1.11957419e+00 -4.70211953e-01 2.13941559e-01
3.34195822e-01 7.75887072e-01 8.22062969e-01 -4.62843895e-01
-6.41940594e-01 1.03145385e+00 -3.77983630e-01 -5.70909321e-01
3.15621465e-01 -7.65150428e-01 3.81604970e-01 2.82723695e-01
8.28184187e-01 -3.08109164e-01 -7.75040984e-01 2.06992164e-01
-4.42659140e-01 -2.65748948e-01 1.05112767e+00 5.31156242e-01
2.35570461e-01 2.37018000e-02 -1.42982781e-01 -1.92947432e-01
1.14062577e-01 6.70957472e-03 -1.12067807e+00 1.74973935e-01
-6.77086651e-01 1.15980739e-02 7.47108757e-02 -1.56732902e-01
-1.29914892e+00 6.95444882e-01 5.94494104e-01 -6.30005360e-01
-3.93561393e-01 2.98087656e-01 -1.91454336e-01 -4.71911907e-01
8.46306026e-01 2.79566973e-01 -3.03993016e-01 -2.69305915e-01
3.26270461e-01 7.92319179e-01 9.19118702e-01 -9.63174775e-02
1.33628476e+00 1.02612829e+00 3.29564720e-01 -1.17925990e+00
-5.45912623e-01 -8.18004429e-01 -6.07609570e-01 -6.55438364e-01
9.71452236e-01 -1.06027889e+00 -6.59299314e-01 5.98354697e-01
-1.16403365e+00 3.91686976e-01 2.73051023e-01 1.02662683e+00
-2.70272672e-01 5.33429623e-01 -2.89257616e-01 -1.65009046e+00
-2.57174015e-01 -7.50560582e-01 1.34940481e+00 5.01212478e-01
3.15868676e-01 -1.30724952e-01 3.26566726e-01 6.02795660e-01
4.12320048e-01 4.48788613e-01 -1.77861795e-01 3.60451676e-02
-1.05228674e+00 -3.53134096e-01 -1.60100237e-01 1.11143298e-01
-1.74108297e-01 1.11729980e-01 -1.28343225e+00 -2.78920382e-01
9.41276848e-02 -1.54368177e-01 1.18705499e+00 5.27878284e-01
-1.26347085e-03 5.38327098e-01 -6.29325390e-01 6.36289358e-01
1.80407834e+00 4.15543374e-03 6.72526002e-01 1.47806376e-01
5.47961533e-01 3.85446399e-01 7.75768161e-01 7.41384268e-01
1.86605334e-01 4.85804766e-01 5.44491172e-01 2.22008616e-01
8.68034512e-02 2.54586875e-01 5.62271595e-01 1.33985057e-01
-1.91848576e-01 -2.19264299e-01 -1.00674641e+00 2.19077006e-01
-1.96177721e+00 -1.02135134e+00 -3.32264066e-01 2.45805192e+00
-6.70133084e-02 1.32027432e-01 -1.02509230e-01 -1.32140458e-01
7.13702857e-01 6.12972416e-02 -3.74238521e-01 4.08824831e-01
-3.32560092e-01 1.48074090e-01 1.06937480e+00 2.15870380e-01
-1.11563659e+00 4.43518013e-01 6.44361258e+00 1.83765844e-01
-6.68002963e-01 1.55519903e-01 -4.04258281e-01 -2.09820986e-01
1.91922814e-01 3.79048064e-02 -1.01947057e+00 3.74819130e-01
8.19017291e-01 3.02183539e-01 9.08287391e-02 2.94215798e-01
-1.43087432e-01 -7.55820930e-01 -9.35491204e-01 1.47731447e+00
5.09497762e-01 -5.87619841e-01 -6.34271085e-01 2.78837293e-01
3.19437921e-01 2.11745635e-01 1.77795693e-01 -1.14127748e-01
3.00695091e-01 -7.97415614e-01 5.31430185e-01 1.17487872e+00
4.14891511e-01 -6.30192637e-01 8.71230185e-01 4.06299561e-01
-1.10046220e+00 -7.07458854e-02 -4.42013711e-01 -9.00607705e-02
2.86721975e-01 1.23915768e+00 -3.16779137e-01 6.97707891e-01
9.20186639e-01 1.97417483e-01 -5.08489430e-01 1.19251990e+00
-3.69270563e-01 5.05203843e-01 -8.24034035e-01 -1.61395669e-02
-5.41338623e-02 -3.24395537e-01 8.60617936e-01 1.19152153e+00
8.37229967e-01 4.31404859e-01 3.00320476e-01 9.57086444e-01
5.43986201e-01 -8.10876310e-01 -7.98069060e-01 3.35915238e-01
5.35619915e-01 1.68381846e+00 -4.43287641e-01 1.60509665e-02
-8.43342721e-01 8.39857519e-01 -4.67228860e-01 4.72793370e-01
-8.00600171e-01 -2.17858419e-01 4.09593761e-01 -6.25678822e-02
6.10292077e-01 -7.90189803e-01 -1.67599350e-01 -1.20658684e+00
2.55715132e-01 -3.43759090e-01 1.68275282e-01 -1.13918233e+00
-1.42968512e+00 4.62132037e-01 7.02479258e-02 -1.47898126e+00
-4.98694219e-02 -7.73885727e-01 -2.40153387e-01 7.80734241e-01
-1.67793870e+00 -1.34998345e+00 -7.83347726e-01 5.90009868e-01
-1.42630771e-01 -2.69919783e-01 4.81829822e-01 6.59336802e-03
-3.26137930e-01 1.71056628e-01 2.38639146e-01 -2.60835793e-02
1.13359141e+00 -8.86446714e-01 -1.20177113e-01 1.38134944e+00
3.82304601e-02 4.73930717e-01 9.29623723e-01 -8.06638420e-01
-1.95303893e+00 -7.20925570e-01 7.96954557e-02 -5.47067702e-01
4.53276336e-01 -5.66178501e-01 -3.03339750e-01 4.00140882e-01
4.67591077e-01 1.34398207e-01 7.65794933e-01 -7.71525651e-02
-4.35490489e-01 -3.09200197e-01 -1.33030736e+00 1.55979916e-01
5.78596592e-01 -3.58500779e-01 -8.65489841e-01 -6.43278956e-02
3.85095030e-02 -2.37119406e-01 -3.47745776e-01 6.65455461e-01
8.98424745e-01 -1.26334226e+00 9.05317783e-01 4.53289300e-01
-3.81624907e-01 -7.46598542e-01 -6.87861562e-01 -1.11892068e+00
-3.95164400e-01 -5.58811009e-01 -2.55123883e-01 1.08410060e+00
-6.24923408e-02 -7.00179935e-01 7.56103098e-01 6.12069368e-01
9.10317302e-02 2.29222685e-01 -1.27571619e+00 -1.04542029e+00
-8.19883466e-01 -5.18041015e-01 -3.26352641e-02 3.09983402e-01
-1.60576820e-01 4.25087541e-01 -6.43596649e-01 8.64402413e-01
1.73378909e+00 2.06975922e-01 5.63585818e-01 -1.48392713e+00
-4.46549118e-01 2.67201215e-01 -6.88003302e-01 -6.64414585e-01
-3.77635390e-01 -6.87404275e-02 5.69804490e-01 -1.38159549e+00
4.35555667e-01 1.53368190e-01 -3.16811919e-01 -3.35000223e-04
-1.44990847e-01 4.87960309e-01 3.02117318e-01 2.15320662e-01
-7.77276814e-01 4.79031086e-01 3.16107631e-01 -1.11477986e-01
2.68023480e-02 -1.22731149e-01 -3.30237925e-01 9.23609436e-01
3.79246265e-01 -4.82891053e-01 5.70272375e-03 -3.89098197e-01
3.02614897e-01 7.07739517e-02 9.81979489e-01 -1.84832525e+00
1.00752044e+00 -7.07790181e-02 7.32811749e-01 -1.02844846e+00
8.42154741e-01 -1.14862227e+00 1.57219946e-01 5.02478182e-01
3.78927469e-01 -1.22435436e-01 -4.63879108e-02 1.18411016e+00
1.63108096e-01 -1.20357694e-02 8.51571679e-01 -1.87921897e-01
-7.07151949e-01 -1.98966101e-01 -6.42008960e-01 -4.58935171e-01
9.10189807e-01 -4.01028097e-01 -8.50420713e-01 -3.31995606e-01
-2.76844829e-01 2.62423027e-02 4.69347566e-01 9.11294296e-02
1.05748010e+00 -1.10432172e+00 -8.01817715e-01 7.15206116e-02
3.89789969e-01 -4.76038098e-01 5.25900349e-02 9.88492131e-01
-3.55060734e-02 4.52924311e-01 -2.83147097e-01 -9.24609840e-01
-1.44120646e+00 3.58813316e-01 2.82331139e-01 3.18015069e-01
-5.29156208e-01 6.20535433e-01 -8.71821493e-02 -3.03082149e-02
4.97860685e-02 -1.09478973e-01 2.10616007e-01 -3.23078334e-01
7.12298691e-01 5.53798735e-01 -5.08750081e-02 -7.10961103e-01
-8.96887600e-01 9.59796309e-01 3.64827991e-01 -6.88135326e-01
9.67508078e-01 -3.93309474e-01 1.39441833e-01 5.72445095e-01
2.98654079e-01 -1.81462597e-02 -1.39628685e+00 -4.09263045e-01
-3.32778960e-01 -6.77359104e-01 3.87630045e-01 -9.01993096e-01
-8.06856871e-01 7.89840102e-01 1.35512567e+00 -1.30894408e-01
1.33223212e+00 -1.53648660e-01 6.07712269e-01 6.75644934e-01
8.82228196e-01 -1.26367629e+00 1.31091014e-01 1.95888162e-01
2.45658010e-01 -1.41468859e+00 5.73426485e-01 -4.01242554e-01
-5.13313830e-01 1.01753008e+00 4.87049878e-01 -1.58003271e-01
4.78760004e-01 6.61778808e-01 2.00873047e-01 -2.80152678e-01
-7.49249697e-01 -8.24798048e-01 2.28199959e-01 9.34314668e-01
-9.26007479e-02 -5.98333366e-02 4.95375246e-02 4.84771311e-01
2.61928856e-01 1.15258992e-01 6.16471350e-01 1.15330780e+00
-9.31842148e-01 -4.89272863e-01 -1.34039867e+00 3.09975147e-01
-1.68642238e-01 -3.02551985e-02 -6.65137589e-01 5.43063521e-01
1.72834784e-01 1.69339597e+00 -1.55555353e-01 -8.14436793e-01
3.83412451e-01 5.01466729e-02 8.84588778e-01 -3.20946932e-01
-2.51844227e-01 5.02710521e-01 -8.38273838e-02 -6.81250870e-01
-8.56103122e-01 -9.23360169e-01 -7.51039207e-01 1.07846633e-01
-6.84400558e-01 -5.24592996e-01 8.12765896e-01 9.91546035e-01
3.03188801e-01 2.73747236e-01 3.68441612e-01 -8.28109860e-01
-6.02003276e-01 -9.10981238e-01 -8.56840849e-01 -1.12129174e-01
3.97108912e-01 -9.21664000e-01 -4.84525472e-01 -9.01398659e-02] | [7.909421443939209, -1.4531307220458984] |
7983a384-1b92-46a7-bf55-4a3c31296f67 | learning-robust-visual-semantic-embeddings | 1703.05908 | null | http://arxiv.org/abs/1703.05908v2 | http://arxiv.org/pdf/1703.05908v2.pdf | Learning Robust Visual-Semantic Embeddings | Many of the existing methods for learning joint embedding of images and text
use only supervised information from paired images and its textual attributes.
Taking advantage of the recent success of unsupervised learning in deep neural
networks, we propose an end-to-end learning framework that is able to extract
more robust multi-modal representations across domains. The proposed method
combines representation learning models (i.e., auto-encoders) together with
cross-domain learning criteria (i.e., Maximum Mean Discrepancy loss) to learn
joint embeddings for semantic and visual features. A novel technique of
unsupervised-data adaptation inference is introduced to construct more
comprehensive embeddings for both labeled and unlabeled data. We evaluate our
method on Animals with Attributes and Caltech-UCSD Birds 200-2011 dataset with
a wide range of applications, including zero and few-shot image recognition and
retrieval, from inductive to transductive settings. Empirically, we show that
our framework improves over the current state of the art on many of the
considered tasks. | ['Liang-Kang Huang', 'Yao-Hung Hubert Tsai', 'Ruslan Salakhutdinov'] | 2017-03-17 | learning-robust-visual-semantic-embeddings-1 | http://openaccess.thecvf.com/content_iccv_2017/html/Tsai_Learning_Robust_Visual-Semantic_ICCV_2017_paper.html | http://openaccess.thecvf.com/content_ICCV_2017/papers/Tsai_Learning_Robust_Visual-Semantic_ICCV_2017_paper.pdf | iccv-2017-10 | ['generalized-few-shot-learning'] | ['methodology'] | [ 2.61257738e-01 -2.36235023e-01 -4.97439772e-01 -7.93733597e-01
-9.71672535e-01 -5.45484185e-01 1.01086879e+00 3.93397212e-01
-7.79831767e-01 6.13175511e-01 2.66916305e-01 1.89220041e-01
-2.76546568e-01 -6.43819511e-01 -7.63238847e-01 -5.52636206e-01
3.15111838e-02 6.82406485e-01 -1.09482259e-01 -1.08838202e-02
8.26787949e-02 2.21352309e-01 -1.80272186e+00 3.58223349e-01
5.58862329e-01 1.14616048e+00 -2.37644315e-02 5.20235121e-01
-1.10908359e-01 8.01009536e-01 -1.32289708e-01 -4.99248743e-01
1.81259707e-01 -3.11621755e-01 -8.19281161e-01 2.87940860e-01
6.82193696e-01 -4.46699977e-01 -3.78515154e-01 7.88898468e-01
5.12288272e-01 2.95892686e-01 1.44315302e+00 -1.35863888e+00
-1.31245458e+00 4.00430918e-01 -5.09739697e-01 2.82619912e-02
1.06240343e-02 -3.18262056e-02 1.45992517e+00 -1.32057798e+00
6.97595000e-01 1.34781790e+00 5.15812159e-01 5.38220763e-01
-1.61832213e+00 -4.65224981e-01 -1.00223958e-01 3.26585799e-01
-1.38046801e+00 -5.17249525e-01 6.39302075e-01 -6.19551837e-01
8.80348742e-01 -3.49287271e-01 8.70373324e-02 1.28565454e+00
-8.59167501e-02 9.21206355e-01 1.08184314e+00 -6.99481070e-01
2.53540009e-01 4.77793306e-01 -4.07328606e-02 6.10284626e-01
1.18910760e-01 1.94277331e-01 -5.94023108e-01 -1.78597987e-01
4.40288395e-01 2.70632297e-01 2.59962291e-01 -1.03998661e+00
-1.24296296e+00 1.20266902e+00 5.60276270e-01 1.77556932e-01
-1.95121884e-01 1.18010245e-01 6.19221747e-01 4.48894858e-01
7.21201956e-01 3.18060786e-01 -4.68946010e-01 2.88426012e-01
-7.65934467e-01 2.48097256e-02 6.17056727e-01 1.08453953e+00
1.03891873e+00 -4.34179343e-02 -1.83749542e-01 1.28404844e+00
4.38488007e-01 5.37331820e-01 6.09621704e-01 -7.13986993e-01
9.05019939e-02 4.09234077e-01 -9.68828946e-02 -8.22221458e-01
-7.68687725e-02 -9.55820456e-03 -7.44200289e-01 1.28503859e-01
1.61847129e-01 1.16374165e-01 -9.64490831e-01 1.89178777e+00
1.14646055e-01 1.60799250e-01 4.09787565e-01 6.69157028e-01
8.69243264e-01 6.08216465e-01 3.41960311e-01 3.00680250e-02
1.13736451e+00 -9.63730037e-01 -6.63024366e-01 -3.52821618e-01
7.85748065e-01 -5.89421749e-01 9.62758243e-01 1.35197625e-01
-8.32712889e-01 -6.49934709e-01 -1.15432644e+00 -3.29410940e-01
-9.02999878e-01 3.58550578e-01 6.20997071e-01 2.31347471e-01
-8.16829324e-01 4.34776187e-01 -5.03570497e-01 -8.58315885e-01
5.70819378e-01 3.32977712e-01 -7.84621835e-01 -4.42613065e-01
-1.11277425e+00 9.05432463e-01 5.35138488e-01 -3.27565283e-01
-1.09349453e+00 -5.77292502e-01 -1.06769502e+00 -1.39529496e-01
1.69066921e-01 -7.73949265e-01 9.78134215e-01 -1.23213923e+00
-1.36521637e+00 1.35278285e+00 3.57227236e-01 -6.47908032e-01
1.03342667e-01 -3.69137853e-01 -2.25213066e-01 3.46459955e-01
9.31546912e-02 1.13064468e+00 1.12738216e+00 -1.20345664e+00
-4.49346155e-01 -6.19835913e-01 2.19268797e-04 2.06165269e-01
-1.08747232e+00 -1.76661968e-01 -1.26510635e-01 -6.33220971e-01
-4.06793147e-01 -8.87391031e-01 -3.06546278e-02 4.22201842e-01
-1.19281769e-01 -4.84259963e-01 7.30687380e-01 -3.18865269e-01
6.66242242e-01 -2.32513833e+00 5.99640489e-01 -1.79452598e-01
-4.13602171e-03 2.36428216e-01 -4.72297907e-01 5.72466731e-01
-1.31004065e-01 -1.97004512e-01 -3.57778788e-01 -7.51498401e-01
3.01816642e-01 6.33554816e-01 -3.02137226e-01 6.21656239e-01
5.26071012e-01 8.59053671e-01 -8.60474527e-01 -6.04926825e-01
3.38058323e-01 4.44668472e-01 -5.58159351e-01 5.00112176e-01
-1.96674347e-01 1.92562453e-02 -2.46054038e-01 6.09037459e-01
5.10933399e-01 -1.52773380e-01 1.57885272e-02 -1.33121833e-01
2.40051463e-01 -1.10928401e-01 -8.88009012e-01 2.26449442e+00
-6.26907408e-01 6.09270871e-01 -2.51437753e-01 -1.54252446e+00
1.04423583e+00 2.13587984e-01 3.49278867e-01 -4.67724681e-01
2.04309806e-01 2.16876432e-01 -5.30645728e-01 -5.79918981e-01
4.02457207e-01 -3.05582762e-01 -3.15369010e-01 3.68434817e-01
1.03212535e+00 -8.31660628e-02 1.02924496e-01 2.30690733e-01
8.83825600e-01 2.30858877e-01 4.74677503e-01 -2.20613331e-01
4.78285968e-01 -1.66767389e-01 1.76270947e-01 6.13548875e-01
-1.32093042e-01 5.91529906e-01 1.65089652e-01 -3.26615214e-01
-1.32592738e+00 -1.18854713e+00 -1.96435392e-01 1.75881040e+00
-1.02223113e-01 -2.35841095e-01 -4.54999387e-01 -7.63270795e-01
3.88000876e-01 6.81748807e-01 -1.01531041e+00 -5.24455130e-01
1.86619848e-01 -4.55910146e-01 5.50842941e-01 8.29767764e-01
7.86695555e-02 -8.97761226e-01 -1.55579612e-01 7.46254344e-03
1.61746919e-01 -1.33832693e+00 -2.48658136e-01 3.84426117e-01
-8.48347425e-01 -7.25594342e-01 -7.90673375e-01 -1.08335829e+00
5.53326845e-01 1.48338705e-01 1.17188013e+00 -3.88816655e-01
-5.92236698e-01 8.90709460e-01 -5.22705793e-01 -3.33940804e-01
-2.89568543e-01 -6.87075108e-02 2.02280462e-01 2.44141430e-01
7.31900156e-01 -4.59808737e-01 -2.79550374e-01 1.68024823e-01
-1.30960941e+00 -3.67685407e-01 8.57044399e-01 1.39043546e+00
7.10367739e-01 -4.96961355e-01 7.50764489e-01 -8.91943336e-01
4.56341088e-01 -6.71204209e-01 -4.25774574e-01 4.05760854e-01
-5.89525461e-01 3.50704193e-01 5.48914015e-01 -5.50779939e-01
-8.77865016e-01 1.72723815e-01 2.57987589e-01 -8.10212374e-01
-4.25132632e-01 4.86044019e-01 7.25283846e-03 1.63444906e-01
9.23653483e-01 1.64146334e-01 1.05541013e-01 -4.95968789e-01
9.76842403e-01 7.85194695e-01 5.51170588e-01 -5.51755548e-01
6.79749429e-01 6.15706205e-01 -6.58721849e-02 -8.61070096e-01
-9.97107029e-01 -7.79074430e-01 -9.44982529e-01 1.25734046e-01
9.89273369e-01 -1.10220861e+00 -1.90565437e-01 2.78409243e-01
-1.05300570e+00 -2.73718778e-02 -4.43242610e-01 7.61394262e-01
-9.27242279e-01 2.06873074e-01 -4.54038531e-01 -5.95162034e-01
-1.57081783e-01 -7.61979222e-01 1.22970569e+00 -2.28715315e-02
1.09718055e-01 -1.24259722e+00 3.98631930e-01 2.13397726e-01
2.31715202e-01 1.33766085e-01 1.06897509e+00 -1.15246904e+00
-2.44737372e-01 -2.39422113e-01 -3.70778799e-01 8.52494657e-01
1.69797912e-02 -1.63490679e-02 -1.16091096e+00 -4.44707692e-01
-4.32965100e-01 -1.25447559e+00 1.31016970e+00 2.58077770e-01
1.02893412e+00 -8.59907568e-02 -2.40779012e-01 4.70522285e-01
1.73379183e+00 -3.14923853e-01 5.44721246e-01 1.69928938e-01
5.66244423e-01 7.99634576e-01 7.37512410e-01 6.27259851e-01
3.03614527e-01 5.94751537e-01 3.54528129e-01 5.70234284e-02
-5.52919619e-02 -2.82376647e-01 3.88273388e-01 7.19295323e-01
1.81164831e-01 -2.94191360e-01 -6.67942107e-01 9.26453829e-01
-1.86758244e+00 -8.59203637e-01 5.61815739e-01 2.19555807e+00
7.94427156e-01 -1.07122764e-01 1.39665961e-01 -8.43186155e-02
6.92173004e-01 1.45762369e-01 -5.76192677e-01 -4.46698159e-01
-4.17890847e-02 2.57136911e-01 3.54216754e-01 1.22808181e-01
-1.40419888e+00 9.14257407e-01 6.19743347e+00 8.54036272e-01
-6.89923108e-01 1.65131211e-01 3.98822904e-01 1.08117133e-01
-1.93577990e-01 -1.62213936e-01 -5.64892173e-01 -1.00543760e-02
9.06703472e-01 -3.66572328e-02 2.71459073e-01 9.91052091e-01
-5.89090168e-01 2.40112290e-01 -1.62687671e+00 1.00292790e+00
5.99029899e-01 -1.14945483e+00 2.91206867e-01 -1.88219562e-01
8.26054692e-01 1.76363826e-01 4.88529652e-01 5.42292535e-01
3.51486564e-01 -1.01566601e+00 3.10533643e-01 5.79150021e-01
9.81257498e-01 -7.37600684e-01 7.68189549e-01 1.76991716e-01
-8.36303949e-01 -1.96551517e-01 -6.68114066e-01 2.32122838e-01
-3.27416837e-01 2.84861594e-01 -8.72714937e-01 5.06265998e-01
5.59803069e-01 1.34334981e+00 -6.10931277e-01 7.89683998e-01
-7.38296583e-02 3.43286812e-01 -1.63475841e-01 -1.15393296e-01
3.93323988e-01 -7.89406151e-02 3.18475068e-01 1.44267237e+00
7.22309947e-02 -2.99115956e-01 2.00484052e-01 8.38094532e-01
-4.81641322e-01 2.71368533e-01 -1.14064646e+00 -4.49081719e-01
3.69953394e-01 1.21127677e+00 -3.32807928e-01 -4.46042925e-01
-6.64216638e-01 9.19977486e-01 5.91900885e-01 2.95418173e-01
-5.53271949e-01 -5.06319642e-01 5.96745253e-01 -2.70642519e-01
6.70483828e-01 -1.51066259e-01 1.07855931e-01 -1.07561123e+00
-2.95960456e-01 -5.68610966e-01 6.43158555e-01 -6.94061995e-01
-2.05391240e+00 3.65381807e-01 2.92344689e-01 -1.40880251e+00
-6.03566885e-01 -8.23450863e-01 -1.40458882e-01 4.23810810e-01
-1.81833851e+00 -1.45934820e+00 -2.39598248e-02 7.27405787e-01
5.98005533e-01 -5.60546100e-01 1.19992733e+00 3.26942831e-01
-1.68075845e-01 7.38164067e-01 7.38689184e-01 3.35564792e-01
1.22797060e+00 -1.33847320e+00 -6.52638003e-02 3.16606045e-01
5.14328539e-01 3.72448653e-01 5.18862545e-01 8.61727297e-02
-1.45497990e+00 -1.16500890e+00 6.04510248e-01 -4.82637316e-01
8.21439385e-01 -5.26566565e-01 -8.42425704e-01 7.77160823e-01
4.10116583e-01 4.18266058e-01 1.02924228e+00 2.56054580e-01
-9.48552549e-01 -2.87849009e-01 -1.09759390e+00 2.29567453e-01
7.32828438e-01 -8.99844944e-01 -7.66289532e-01 4.10203159e-01
6.45366848e-01 1.96510375e-01 -9.46153998e-01 3.80905598e-01
6.10137165e-01 -6.71276867e-01 1.11999583e+00 -9.63257968e-01
7.63463259e-01 7.04373345e-02 -7.09572136e-01 -1.46017122e+00
-3.04760516e-01 -1.11738201e-02 -3.59271616e-02 1.34307194e+00
3.57408077e-01 -4.36630934e-01 2.86352903e-01 1.59048468e-01
7.99205378e-02 -5.56760967e-01 -8.06599796e-01 -8.49565268e-01
3.46393853e-01 -1.13125809e-01 5.61671145e-02 1.05577731e+00
-1.07078351e-01 7.19224870e-01 -5.93494773e-01 -1.17501572e-01
8.25449347e-01 1.66262239e-01 7.37716198e-01 -1.34645224e+00
-2.41468623e-01 -9.07157362e-02 -8.84192646e-01 -8.51230860e-01
6.46826327e-01 -1.08604908e+00 2.14013487e-01 -1.38162386e+00
4.33835417e-01 -5.40281534e-02 -7.04645216e-01 4.96345878e-01
1.79946363e-01 4.07322377e-01 6.80142194e-02 1.17130205e-01
-1.02925527e+00 9.25928473e-01 8.31932425e-01 -3.60602647e-01
3.04779500e-01 -4.70545381e-01 -4.44973826e-01 6.89852118e-01
5.15542269e-01 -6.64667964e-01 -4.49288189e-01 -6.07489347e-01
2.87311599e-02 -3.09720963e-01 5.54781675e-01 -7.77708888e-01
1.18634969e-01 7.43240342e-02 4.75019157e-01 -4.36456442e-01
6.40936732e-01 -9.17949855e-01 -5.17829299e-01 9.24751461e-02
-9.61168349e-01 -2.85057932e-01 1.02630116e-01 1.08251739e+00
-3.98701787e-01 -3.65179300e-01 9.75300372e-01 4.93203178e-02
-9.29667115e-01 4.26552355e-01 -1.67575449e-01 2.30583578e-01
1.02008641e+00 1.41101584e-01 -1.83463439e-01 -2.65727818e-01
-8.05624723e-01 1.35092080e-01 3.68083417e-01 6.52607262e-01
8.04706752e-01 -1.76139760e+00 -8.84041846e-01 1.01871125e-01
1.00515985e+00 -4.29962844e-01 4.36087139e-02 5.60910404e-01
1.33220740e-02 2.89717108e-01 -4.94893104e-01 -8.68180871e-01
-1.10833585e+00 8.90209019e-01 -8.89226496e-02 -1.85413212e-01
-1.94704965e-01 8.15824151e-01 1.84066162e-01 -6.43920362e-01
3.50532442e-01 -1.89585879e-01 -3.20379823e-01 3.64216536e-01
4.82705384e-01 6.95185438e-02 -1.35858357e-01 -7.48037934e-01
-2.26362452e-01 5.29587805e-01 -4.10707116e-01 -7.03754202e-02
1.52358341e+00 -1.04178317e-01 4.22898754e-02 9.43780184e-01
1.83773279e+00 -7.17840374e-01 -1.05490613e+00 -9.45609868e-01
-5.77750169e-02 -4.94266212e-01 8.90622884e-02 -5.78406394e-01
-6.91761792e-01 1.27858651e+00 8.36055160e-01 -1.56923719e-02
9.85319614e-01 2.87561983e-01 4.98225093e-01 8.48180234e-01
1.47987545e-01 -1.36853302e+00 5.65835893e-01 3.89883041e-01
7.41437197e-01 -1.63474238e+00 6.04992658e-02 1.13359511e-01
-8.36716413e-01 1.08752859e+00 4.44431305e-01 -3.81884217e-01
6.82368755e-01 -1.75890297e-01 -2.76249319e-01 -2.44260114e-03
-9.73664105e-01 -4.51596141e-01 4.16232377e-01 7.72475004e-01
3.78221005e-01 9.06515494e-03 -9.03570950e-02 4.28139985e-01
5.47502577e-01 3.70602794e-02 1.05846621e-01 9.28905964e-01
-3.76029819e-01 -9.49012578e-01 3.39105092e-02 4.23043817e-01
-2.06959367e-01 -7.93242529e-02 -4.19995040e-01 7.06659257e-01
-2.88739540e-02 7.76155531e-01 1.57980487e-01 -2.78279424e-01
1.92444652e-01 2.71682292e-01 6.17371321e-01 -7.68801808e-01
-4.09899652e-02 -1.16321541e-01 -7.22689405e-02 -2.67881185e-01
-9.97276008e-01 -5.12704790e-01 -9.00058150e-01 4.08037081e-02
-3.85028690e-01 -8.73388499e-02 6.05116308e-01 1.02205694e+00
2.73877949e-01 2.52385229e-01 8.39331031e-01 -9.42924201e-01
-9.43553627e-01 -1.10524976e+00 -7.81606317e-01 8.41919720e-01
2.70151526e-01 -1.00110674e+00 -4.02983695e-01 3.32189649e-01] | [10.049399375915527, 2.427971601486206] |
909793d0-dccd-444d-baef-6fc743b5687c | text-conditional-contextualized-avatars-for | 2304.07410 | null | https://arxiv.org/abs/2304.07410v1 | https://arxiv.org/pdf/2304.07410v1.pdf | Text-Conditional Contextualized Avatars For Zero-Shot Personalization | Recent large-scale text-to-image generation models have made significant improvements in the quality, realism, and diversity of the synthesized images and enable users to control the created content through language. However, the personalization aspect of these generative models is still challenging and under-explored. In this work, we propose a pipeline that enables personalization of image generation with avatars capturing a user's identity in a delightful way. Our pipeline is zero-shot, avatar texture and style agnostic, and does not require training on the avatar at all - it is scalable to millions of users who can generate a scene with their avatar. To render the avatar in a pose faithful to the given text prompt, we propose a novel text-to-3D pose diffusion model trained on a curated large-scale dataset of in-the-wild human poses improving the performance of the SOTA text-to-motion models significantly. We show, for the first time, how to leverage large-scale image datasets to learn human 3D pose parameters and overcome the limitations of motion capture datasets. | ['Sonal Gupta', 'Devi Parikh', 'Guan Pang', 'Akbar Shah', 'Thomas Hayes', 'Samaneh Azadi'] | 2023-04-14 | null | null | null | null | ['text-to-3d'] | ['computer-vision'] | [ 1.49360299e-01 2.78281808e-01 3.37166071e-01 -3.22120100e-01
-7.06005871e-01 -8.91935110e-01 7.65902996e-01 -7.48117447e-01
-1.57142282e-01 4.11659271e-01 5.10697663e-01 3.88319314e-01
4.96386349e-01 -6.63486302e-01 -8.41928363e-01 -3.39110494e-01
4.76361662e-01 9.33620095e-01 8.60748067e-02 -4.45919752e-01
-9.60080475e-02 4.34364498e-01 -1.44012725e+00 3.70736688e-01
3.58947486e-01 7.23926902e-01 9.77748930e-02 1.13529420e+00
4.10554558e-01 8.51874948e-01 -6.21958435e-01 -5.44835985e-01
4.13550496e-01 -6.40805542e-01 -6.36729240e-01 4.04330969e-01
9.51180100e-01 -7.72552848e-01 -4.12509948e-01 5.24190247e-01
8.31516087e-01 1.67798236e-01 5.26393294e-01 -1.29323733e+00
-8.61985147e-01 4.38709378e-01 -5.44175446e-01 -4.52274889e-01
6.16884768e-01 5.39816558e-01 7.48087049e-01 -6.40644252e-01
1.32886255e+00 1.39982867e+00 7.10744500e-01 1.18020594e+00
-1.46303022e+00 -6.05516315e-01 -9.21696723e-02 -2.74430811e-01
-1.34689391e+00 -5.89267552e-01 6.44050717e-01 -5.57358205e-01
6.51796997e-01 2.74874687e-01 1.16124499e+00 2.00319505e+00
-8.54123980e-02 8.43611956e-01 6.78796828e-01 -1.29604697e-01
1.51339680e-01 -6.83988631e-02 -8.40844214e-01 7.85635591e-01
-1.96705371e-01 -8.13538060e-02 -7.66230762e-01 -1.15933239e-01
1.37919819e+00 -2.79297084e-01 5.41241700e-03 -8.52790356e-01
-1.45873272e+00 8.69436681e-01 2.06718922e-01 -6.51081353e-02
-3.06099564e-01 7.91887164e-01 3.81247140e-02 -2.16939017e-01
3.40376318e-01 7.07120180e-01 -2.68642515e-01 -4.29901898e-01
-8.80926609e-01 8.61482441e-01 6.36978865e-01 1.19109249e+00
3.65841538e-01 2.16809139e-01 -4.92771029e-01 4.49554175e-01
8.42431933e-02 8.56779993e-01 2.66573280e-01 -1.48339963e+00
5.82352057e-02 2.43314415e-01 2.67313361e-01 -1.08344352e+00
-2.34176368e-01 -1.70537844e-01 -5.42521060e-01 4.11344886e-01
2.73702323e-01 -2.71461040e-01 -1.00025320e+00 1.87818646e+00
5.70545733e-01 2.69637723e-02 -2.31855616e-01 1.03630638e+00
7.49725401e-01 5.79822004e-01 1.32162347e-01 3.64748120e-01
1.22092402e+00 -1.09394979e+00 -5.17135084e-01 -3.82925510e-01
2.04287916e-01 -9.15810883e-01 1.22577298e+00 2.62977630e-01
-1.37721503e+00 -6.11173511e-01 -6.95943415e-01 -3.23994577e-01
2.06823438e-01 8.61417875e-03 6.17837489e-01 6.73851609e-01
-1.04359925e+00 4.72913563e-01 -8.72873604e-01 -6.00080490e-01
4.39108700e-01 2.38941610e-02 -5.73662043e-01 2.48728767e-01
-8.25833917e-01 7.20496833e-01 3.69750075e-02 -3.30394059e-01
-1.24576473e+00 -8.86358798e-01 -8.17733169e-01 -3.29299569e-01
2.78441221e-01 -1.33865798e+00 1.42823005e+00 -1.08788359e+00
-1.78194153e+00 8.55988920e-01 1.88132808e-01 -1.48683444e-01
1.16503739e+00 -4.18639809e-01 1.15349233e-01 2.18618274e-01
2.38609999e-01 1.37321520e+00 1.10193801e+00 -1.43275380e+00
-3.57430547e-01 -2.07540035e-01 5.26698132e-04 3.45209241e-01
-2.45846421e-01 -2.12943867e-01 -6.92301750e-01 -9.39183176e-01
-2.85138965e-01 -1.31655550e+00 -3.12168986e-01 3.36222708e-01
-2.84101069e-01 3.23309809e-01 1.00017238e+00 -5.83890676e-01
6.97012365e-01 -1.97357273e+00 4.76861328e-01 -1.01461373e-01
3.13365340e-01 1.96477681e-01 -3.31590414e-01 4.56433624e-01
2.54351676e-01 2.31440756e-02 7.78126344e-02 -8.04291308e-01
1.37181282e-01 1.17772110e-01 -1.98072597e-01 1.83681071e-01
-3.31066661e-02 1.32144928e+00 -9.41835523e-01 -5.91233432e-01
2.35019431e-01 8.76052678e-01 -1.00988424e+00 5.80048501e-01
-5.58445513e-01 9.61799562e-01 -3.73263180e-01 2.84150451e-01
3.33245397e-01 -3.00491810e-01 -2.87288148e-02 -3.24652761e-01
1.63074806e-01 -3.68739605e-01 -8.94033968e-01 2.33747935e+00
-4.06723559e-01 6.99267983e-01 -9.55208242e-02 3.15847136e-02
7.12766826e-01 3.43659490e-01 6.75443470e-01 -2.48600274e-01
1.93179592e-01 -8.92231539e-02 -4.79691505e-01 -6.43691957e-01
8.85505438e-01 -1.76497042e-01 -3.85326535e-01 6.06377363e-01
1.04075804e-01 -6.63236439e-01 4.29170169e-02 4.65566546e-01
8.37686539e-01 8.54394972e-01 -9.18618739e-02 1.27635866e-01
-1.95403710e-01 7.53656402e-02 1.74140453e-01 6.38247609e-01
8.80404711e-02 1.38652229e+00 1.76111162e-01 -5.25898993e-01
-1.66748035e+00 -1.23509705e+00 4.68424320e-01 1.13061774e+00
-8.02604407e-02 -4.85139340e-01 -1.02081168e+00 -3.75607014e-01
-2.49775827e-01 7.45783925e-01 -8.54052246e-01 -1.00333311e-01
-5.79448283e-01 -2.41019264e-01 8.63301933e-01 4.92420346e-01
5.75965941e-01 -9.16092396e-01 -9.48838413e-01 1.01260312e-01
-5.58815777e-01 -1.37253642e+00 -1.01830447e+00 -7.62933135e-01
-5.15473664e-01 -5.35791695e-01 -1.17587888e+00 -5.39572656e-01
6.42771304e-01 -1.09408339e-02 1.15600967e+00 -3.48749250e-01
-5.24778903e-01 5.89166999e-01 -3.82128149e-01 -2.83525288e-01
-7.04602897e-01 9.86409783e-02 -8.62107649e-02 2.15824485e-01
-3.72336626e-01 -6.39978230e-01 -8.71103525e-01 2.85866141e-01
-7.82161117e-01 6.43229723e-01 2.96007276e-01 5.51465213e-01
4.10790563e-01 -5.49602032e-01 1.30802274e-01 -8.29932213e-01
4.30315435e-01 -4.53031883e-02 -1.62961319e-01 6.98088184e-02
-2.06178918e-01 6.51863441e-02 4.00185317e-01 -8.44307423e-01
-1.22680688e+00 4.50578809e-01 -1.41268492e-01 -6.39376402e-01
-6.94303662e-02 -2.75882274e-01 -1.64264306e-01 -5.43664061e-02
1.00334501e+00 -2.75979247e-02 -8.44701156e-02 -2.63290226e-01
9.82505560e-01 2.85939902e-01 9.53261733e-01 -8.00620258e-01
1.04069495e+00 6.53535426e-01 -1.42185818e-02 -5.67420721e-01
-6.44692123e-01 5.05204313e-02 -6.96389914e-01 -3.99126559e-01
1.26787603e+00 -1.08081555e+00 -8.35505188e-01 6.10801935e-01
-1.22786450e+00 -7.16869056e-01 -5.62640905e-01 9.84120294e-02
-9.99106050e-01 2.48607725e-01 -5.80568016e-01 -5.03814101e-01
-4.75998402e-01 -1.08066225e+00 1.50749254e+00 9.78634059e-02
-8.41882527e-01 -7.83994079e-01 2.85970956e-01 6.18913949e-01
4.80762690e-01 7.46180952e-01 2.31859282e-01 1.02449343e-01
-6.49765491e-01 -2.65326321e-01 8.88344198e-02 -3.26112323e-02
-6.09810464e-02 7.51657933e-02 -8.50406945e-01 -2.16011450e-01
-4.88344461e-01 -5.98106086e-01 2.99227655e-01 1.22419141e-01
7.24899590e-01 -5.31206429e-01 -5.25354147e-02 8.60253572e-01
8.82587790e-01 -3.03356349e-01 6.65726304e-01 1.13759495e-01
1.17907941e+00 4.51202661e-01 3.19169849e-01 7.52912939e-01
5.62527299e-01 9.68308389e-01 2.71285862e-01 -1.02721699e-01
-5.03905535e-01 -9.10126150e-01 1.38587072e-01 1.96667343e-01
-4.77733910e-01 -4.19435352e-01 -5.01640379e-01 4.94203955e-01
-1.82371509e+00 -1.13096118e+00 8.61029699e-02 1.86579943e+00
9.80140507e-01 -3.11262280e-01 3.27352703e-01 -5.93689501e-01
4.64519680e-01 2.03203246e-01 -4.96784776e-01 -1.94064856e-01
-1.38020903e-01 2.45769080e-02 3.55930805e-01 4.21981543e-01
-8.63996387e-01 1.35642016e+00 6.25013638e+00 6.09545290e-01
-9.97590721e-01 1.77684575e-01 5.53058147e-01 -5.13032436e-01
-4.79544699e-01 -1.28449872e-02 -4.31981415e-01 2.57537141e-02
5.20714462e-01 2.49046944e-02 4.71898556e-01 9.48753238e-01
3.16879332e-01 1.68463752e-01 -1.14170289e+00 1.19915974e+00
3.51225674e-01 -1.50024843e+00 3.70656163e-01 1.35193625e-02
1.16176379e+00 -3.73836726e-01 4.11333799e-01 -8.43227580e-02
5.87162495e-01 -1.05888939e+00 1.41140437e+00 7.00370610e-01
1.13971233e+00 -6.68145835e-01 8.77215937e-02 2.17747629e-01
-8.59799266e-01 3.45115572e-01 -1.18939513e-02 1.46020189e-01
7.08165646e-01 -7.95082748e-02 -8.80256295e-01 9.40912217e-02
7.84445345e-01 4.74556327e-01 -6.89960122e-01 3.80627155e-01
-1.50364682e-01 8.80422518e-02 -1.28575802e-01 4.39332286e-03
8.60799551e-02 8.04966390e-02 5.69574177e-01 1.07755446e+00
5.56724131e-01 1.11139417e-01 7.30279237e-02 1.14767349e+00
-6.94440454e-02 -9.83338356e-02 -5.14959395e-01 -2.74982333e-01
1.02347545e-01 1.35805428e+00 -4.97996330e-01 -2.97431469e-01
2.39917442e-01 1.68636048e+00 2.40764096e-01 2.76333600e-01
-9.92408395e-01 3.20081934e-02 7.02243745e-01 4.46376145e-01
2.51147300e-01 -5.11557043e-01 -2.27084830e-01 -1.24775457e+00
-2.90853918e-01 -1.05664229e+00 -2.11991556e-02 -1.35397041e+00
-1.10818946e+00 7.99887359e-01 1.11573420e-01 -9.34870005e-01
-7.44183540e-01 -2.18269035e-01 -2.54483670e-01 6.63014412e-01
-2.81536937e-01 -1.89126301e+00 -6.63141727e-01 7.04133153e-01
6.01730883e-01 -2.60214388e-01 7.75602102e-01 6.76341802e-02
-1.03391439e-01 6.60982907e-01 -4.18308675e-01 1.65509567e-01
1.01315331e+00 -8.79867375e-01 1.00082779e+00 7.24854112e-01
2.27684125e-01 4.92343783e-01 1.06584930e+00 -7.70152092e-01
-1.48191059e+00 -1.06028926e+00 4.70326215e-01 -1.11914873e+00
3.91186982e-01 -6.47457182e-01 -2.34484136e-01 9.08139586e-01
2.44109452e-01 -5.63500598e-02 3.78373444e-01 -3.29968691e-01
-4.30347145e-01 1.46875054e-01 -1.02371025e+00 1.12447023e+00
1.41077411e+00 -3.98496240e-01 -2.51868904e-01 1.87912941e-01
8.37509096e-01 -8.04485917e-01 -8.10592055e-01 1.18043616e-01
9.08097386e-01 -9.84553277e-01 1.23574710e+00 -6.68626428e-01
7.09621429e-01 -3.31851363e-01 -3.25796343e-02 -1.21695757e+00
-4.73449439e-01 -1.12225771e+00 5.86925633e-02 1.22183299e+00
2.57615894e-01 4.88839857e-03 1.08823431e+00 9.52723980e-01
5.60128987e-02 -1.84169546e-01 -5.72722971e-01 -4.36035484e-01
-4.52057347e-02 -4.05242890e-01 7.62739599e-01 7.36496687e-01
-3.11421543e-01 5.21006823e-01 -1.12569594e+00 -1.35891765e-01
7.18474329e-01 -1.03438012e-01 1.57948768e+00 -8.04632425e-01
-7.00846672e-01 -1.38022915e-01 -3.28786105e-01 -1.12309253e+00
-1.13528989e-01 -5.96691668e-01 2.70708185e-02 -1.53284132e+00
2.73463309e-01 -2.41732657e-01 7.39169478e-01 3.60234022e-01
-9.19408947e-02 8.10614705e-01 5.12274325e-01 2.54007876e-01
-6.28507435e-01 7.24620044e-01 1.72926533e+00 1.44341020e-02
-2.51722813e-01 -2.24950537e-01 -6.32606268e-01 7.19317317e-01
5.35594702e-01 -2.83514053e-01 -6.29908085e-01 -6.82895124e-01
4.81934190e-01 9.77888480e-02 7.49911010e-01 -1.08165586e+00
6.28540963e-02 -3.48814279e-01 6.93405509e-01 -3.11394244e-01
7.60399759e-01 -4.82888252e-01 9.25707757e-01 1.77968040e-01
-4.72749054e-01 6.03011623e-02 -3.35766934e-02 4.97903645e-01
5.20376444e-01 2.41422072e-01 7.72901773e-01 -4.62191939e-01
-6.78242564e-01 6.63275242e-01 -1.55012205e-01 2.63999671e-01
9.61569071e-01 -7.61593357e-02 -7.76348189e-02 -1.04593170e+00
-6.00657105e-01 -1.57373995e-01 9.57453549e-01 7.94726253e-01
5.56528747e-01 -1.50395215e+00 -9.48486447e-01 7.29406253e-02
3.00456494e-01 1.10474207e-01 4.78361011e-01 2.25182295e-01
-9.91047084e-01 -1.06880590e-01 -4.82508451e-01 -7.19882429e-01
-1.14336753e+00 5.67896783e-01 2.75248528e-01 1.08876005e-01
-8.38279665e-01 8.83704722e-01 4.79557633e-01 -4.13840026e-01
-1.25564575e-01 2.26339638e-01 3.32307011e-01 -2.73890018e-01
5.20327806e-01 2.71642566e-01 -4.96190339e-01 -1.06964195e+00
2.82609239e-02 7.45201468e-01 2.07167566e-01 -8.60832989e-01
1.17884851e+00 -2.67670244e-01 1.93231657e-01 2.98591703e-01
1.04172838e+00 1.99951291e-01 -1.95942676e+00 1.42789796e-01
-8.88514459e-01 -7.94110537e-01 -4.45165366e-01 -7.18366325e-01
-1.04818285e+00 6.53040349e-01 3.47459257e-01 -4.69045281e-01
6.63398623e-01 1.50442749e-01 1.21100390e+00 2.42352054e-01
6.33471131e-01 -1.10434270e+00 8.11358392e-01 2.71571070e-01
1.20094550e+00 -9.47473824e-01 -2.19249234e-01 -1.52471527e-01
-1.17034149e+00 9.15774226e-01 7.09908664e-01 -6.14758097e-02
1.39348537e-01 3.12427044e-01 3.79716069e-01 6.37296736e-02
-5.45621097e-01 1.91116050e-01 4.30204421e-01 9.78192151e-01
3.26575458e-01 1.73657998e-01 2.29055032e-01 2.98489422e-01
-7.71169782e-01 1.67512614e-02 5.78807771e-01 5.85469604e-01
-6.79140612e-02 -1.05900097e+00 -4.53326106e-01 -2.93321043e-01
-8.15290436e-02 1.02005512e-01 -5.26156545e-01 5.87805212e-01
1.03896789e-01 8.53198051e-01 -7.96034485e-02 -5.51767945e-01
4.55574453e-01 -2.66021013e-01 8.13290894e-01 -4.35375541e-01
-4.65093464e-01 1.74497157e-01 1.07631750e-01 -7.08705425e-01
-2.41229385e-01 -7.42156625e-01 -1.04734027e+00 -7.18110919e-01
3.38613987e-01 -3.85283411e-01 5.24677932e-01 5.31462371e-01
5.37784874e-01 3.59286010e-01 4.37493950e-01 -1.48531175e+00
-1.61187485e-01 -7.90264070e-01 -2.12714151e-01 9.32691872e-01
-5.94422035e-03 -3.70555431e-01 1.26390904e-01 5.12852967e-01] | [12.01846981048584, -0.663482666015625] |
6277d512-00c7-4df3-8934-8c1d4e9248ca | estimating-and-detecting-random-processes-on | 2211.07884 | null | https://arxiv.org/abs/2211.07884v1 | https://arxiv.org/pdf/2211.07884v1.pdf | Estimating and detecting random processes on the unit circle | The problem of detecting a sinusoidal signal with randomly varying frequency has a long history. It is one of the core problems in signal processing, arising in many applications including, for example, underwater acoustic frequency line tracking, demodulation of FM radio communications, laser phase drift in optical communications and, recently, continuous gravitational wave astronomy. In this paper we describe a Markov Chain Monte Carlo based procedure to compute a specific detection posterior density. We demonstrate via simulation that our approach results in an up to $25$ percent higher detection rate than Hidden Markov Model based solutions, which are generally considered to be the leading techniques for these problems. | ['A. Melatos', 'B. Moran', 'R. J. Evans', 'S. Suvorova', 'Changrong Liu'] | 2022-11-15 | null | null | null | null | ['astronomy'] | ['miscellaneous'] | [ 2.88833857e-01 -2.72990048e-01 3.66012812e-01 -5.06738424e-02
-8.22813272e-01 -4.04750437e-01 4.64720935e-01 -4.85242270e-02
-7.87554622e-01 9.14992332e-01 -3.28529686e-01 -4.33492541e-01
-2.62297571e-01 -4.81671423e-01 -3.16139162e-01 -9.53022718e-01
-5.60392916e-01 4.74031061e-01 4.15128857e-01 1.30637527e-01
4.53367829e-01 5.12818694e-01 -1.38252258e+00 -6.45517826e-01
3.48392069e-01 8.02630782e-01 1.01498753e-01 9.84180152e-01
1.10784985e-01 2.96736926e-01 -7.09454477e-01 2.20957790e-02
-1.34515643e-01 -5.58771789e-01 -3.10967058e-01 -3.59785140e-01
-3.27451587e-01 -2.23999545e-01 -1.07055493e-01 1.36388826e+00
5.82580149e-01 2.44354289e-02 7.29210079e-01 -6.87319398e-01
4.84087080e-01 7.23495483e-01 -7.38939822e-01 4.18823481e-01
6.10752702e-01 4.71156202e-02 4.40638244e-01 -5.54538548e-01
1.37718409e-01 1.13133180e+00 6.53634846e-01 3.72934878e-01
-1.09604681e+00 -7.92384148e-01 -6.45329773e-01 1.22643828e-01
-1.44701707e+00 -5.49352527e-01 4.66793984e-01 -4.34564203e-01
7.14978933e-01 7.79531524e-02 5.58433115e-01 5.73893189e-01
6.02134764e-01 4.42911893e-01 1.17231297e+00 -7.91301608e-01
4.13345486e-01 -2.42544591e-01 2.63737980e-02 4.15212780e-01
7.05722332e-01 5.05836725e-01 -6.52177274e-01 -4.84889895e-01
4.64386195e-01 -5.88423967e-01 -5.63562274e-01 5.93066849e-02
-8.94341946e-01 9.80017304e-01 -1.97513789e-01 1.55728668e-01
-1.78442553e-01 7.76460111e-01 -8.69496018e-02 3.29056740e-01
1.75966248e-01 3.95344347e-01 1.24344444e-02 -4.75125492e-01
-1.06050980e+00 3.34073573e-01 1.07366097e+00 7.13944554e-01
6.97163463e-01 3.94688755e-01 4.41372782e-01 3.45868990e-02
1.01053429e+00 1.09422147e+00 3.70732427e-01 -6.90007865e-01
-2.18664587e-01 -4.90658164e-01 5.98094463e-01 -4.06160414e-01
-5.40256679e-01 -4.07193720e-01 -5.32398582e-01 1.82601810e-01
4.08450425e-01 -7.18147099e-01 -7.18755364e-01 1.40864742e+00
1.76058814e-01 6.50813758e-01 2.91506439e-01 6.59321189e-01
3.49373907e-01 7.35139012e-01 -3.00663054e-01 -7.07009792e-01
1.40299034e+00 7.55630210e-02 -9.56565499e-01 -4.58590001e-01
1.00502625e-01 -1.32159793e+00 -2.37089828e-01 6.60109282e-01
-8.72076571e-01 -1.26298219e-02 -1.27447951e+00 6.62051499e-01
5.26232958e-01 -5.87380767e-01 5.98101020e-01 1.10263658e+00
-6.73402250e-01 6.42642677e-01 -1.09678578e+00 -2.30551530e-02
-8.15369189e-02 2.18272477e-01 4.14859235e-01 6.41293675e-02
-1.25617993e+00 7.83297360e-01 1.97426841e-01 3.95248421e-02
-8.30294788e-01 -3.19001257e-01 -6.92349195e-01 -8.46187621e-02
-1.78482626e-02 -3.00129920e-01 1.60670805e+00 -2.20983088e-01
-1.59043419e+00 4.54908997e-01 -2.81935215e-01 -7.35521734e-01
1.71483457e-01 -9.49329212e-02 -5.82594931e-01 1.35004371e-01
-6.39570132e-02 -1.00113049e-01 1.11865604e+00 -8.05362165e-01
-6.16808176e-01 2.74863672e-02 -6.12223685e-01 -1.08969755e-01
5.49204230e-01 1.80728942e-01 4.84031327e-02 -2.38967076e-01
4.32182968e-01 -1.08276284e+00 -5.45955122e-01 -4.55767661e-01
-1.46140352e-01 -1.35174289e-01 5.04389167e-01 -1.54411092e-01
8.26678693e-01 -2.22078586e+00 -8.19895342e-02 4.82724011e-01
-2.38251284e-01 7.90931657e-03 5.58587670e-01 6.59422517e-01
2.79469013e-01 -3.01548451e-01 -3.99946183e-01 -2.12724835e-01
-1.04580514e-01 3.71404588e-02 -3.74506295e-01 9.19057429e-01
6.03004582e-02 1.47944063e-01 -1.08957052e+00 -1.54908106e-01
1.07510179e-01 2.99529761e-01 -2.90125430e-01 -9.80723929e-03
-4.19443399e-02 6.54703021e-01 -3.58353615e-01 2.09730014e-01
5.83187878e-01 -1.04641981e-01 1.91635653e-01 1.93991974e-01
-3.89562756e-01 1.45122886e-01 -1.49301004e+00 1.47879052e+00
-4.71276492e-02 1.01872706e+00 3.15570503e-01 -8.15302432e-01
8.42503190e-01 4.96941626e-01 3.47528934e-01 -1.86445251e-01
3.92409742e-01 5.53417444e-01 3.97582144e-01 -5.31411946e-01
7.49078631e-01 -7.97583699e-01 -2.58295536e-01 6.83004141e-01
-7.74935447e-03 -4.38930064e-01 6.99492693e-02 1.88406259e-02
1.05071473e+00 -1.77899167e-01 4.51931477e-01 -3.71632993e-01
2.19979703e-01 -2.15848461e-01 4.77299899e-01 9.72053111e-01
-2.62322724e-01 4.77704465e-01 2.11508587e-01 1.35353208e-01
-6.23557687e-01 -8.12171340e-01 -4.55773085e-01 -3.27286310e-02
3.19893271e-01 -2.78701791e-05 -5.46404183e-01 4.01873142e-01
-1.37331814e-01 4.29506958e-01 1.65494561e-01 -8.53088349e-02
-6.55634284e-01 -1.16497970e+00 6.44100487e-01 3.49803641e-03
3.56457412e-01 -8.91835749e-01 -8.59093130e-01 6.03268802e-01
-1.10536916e-02 -1.13192368e+00 2.25932062e-01 3.85998607e-01
-7.14281976e-01 -7.70005763e-01 -6.58876598e-01 -4.97596741e-01
1.93783954e-01 7.97472671e-02 8.30325544e-01 1.22328348e-01
-6.13313437e-01 3.26203376e-01 -4.03222471e-01 -7.10902750e-01
-7.10943103e-01 -4.26432073e-01 5.98871112e-01 -9.31740105e-02
4.85609740e-01 -3.98461848e-01 -6.23975456e-01 1.16166152e-01
-6.13027453e-01 -6.40857816e-01 5.73316813e-01 4.54842329e-01
1.38709679e-01 2.80234009e-01 4.51324284e-01 -7.66042173e-01
5.54156125e-01 -4.67575043e-01 -1.12237954e+00 -2.88173586e-01
-4.04036403e-01 8.73105898e-02 6.15502298e-02 -2.25516707e-01
-6.59139514e-01 2.61808008e-01 -5.19327164e-01 1.85788184e-01
8.94687232e-03 8.43428791e-01 3.00341189e-01 -4.74373668e-01
8.19001853e-01 3.46085489e-01 1.37147218e-01 -1.62195370e-01
-2.01000035e-01 6.43531501e-01 5.07283151e-01 -2.00565845e-01
1.22492182e+00 5.24223983e-01 3.72999072e-01 -1.37636244e+00
-4.90431875e-01 -6.74449086e-01 7.51621872e-02 -2.49345124e-01
7.47258902e-01 -8.49047482e-01 -8.32268417e-01 6.35398507e-01
-1.07998896e+00 2.11631395e-02 5.06967269e-02 1.16067445e+00
-5.53083777e-01 6.57512188e-01 -4.82895792e-01 -1.50575030e+00
1.58925336e-02 -8.99822593e-01 6.12767100e-01 6.88357949e-01
-1.58527046e-01 -9.18003917e-01 4.09405410e-01 -3.57878625e-01
5.07213712e-01 2.74898201e-01 3.96600634e-01 3.08126006e-02
-8.48755479e-01 -4.85380203e-01 2.59164780e-01 -1.43800035e-01
-8.80395472e-02 1.52448732e-02 -9.91774023e-01 -5.22604525e-01
4.35028642e-01 1.90972477e-01 8.51324677e-01 8.24517488e-01
1.51093647e-01 3.92443836e-01 -5.72103024e-01 2.44645149e-01
1.62545264e+00 3.84019256e-01 7.00471580e-01 -9.98747945e-02
-6.49306178e-02 2.38196552e-01 8.05989265e-01 6.99661195e-01
-2.69642681e-01 4.72846270e-01 3.43779534e-01 4.03573185e-01
1.43428057e-01 2.61394233e-01 2.64376044e-01 8.13743949e-01
2.09187821e-01 -3.07923943e-01 -8.29651475e-01 4.29878354e-01
-1.49176407e+00 -1.19455540e+00 -6.09156609e-01 2.34523296e+00
5.91432631e-01 3.38063717e-01 -1.54860362e-01 4.40089136e-01
7.41359472e-01 -2.46288210e-01 -2.24316716e-01 -1.39733061e-01
3.28996450e-01 4.03346688e-01 1.11953926e+00 8.02058160e-01
-8.22107553e-01 5.03665507e-01 7.15993357e+00 5.40305793e-01
-9.67940450e-01 -6.22189865e-02 -3.77475411e-01 3.55411112e-01
-1.13279648e-01 2.11703524e-01 -1.11823118e+00 5.75188100e-01
1.18445706e+00 -4.52371955e-01 5.96774817e-02 2.09025353e-01
3.46868664e-01 -8.41243386e-01 -5.88240862e-01 8.72881174e-01
-2.26973742e-01 -9.26812232e-01 -4.00406063e-01 2.25726679e-01
4.36100274e-01 1.64386109e-02 -5.96264564e-02 -6.32339939e-02
2.61415035e-01 -6.57771468e-01 4.91410553e-01 5.04865825e-01
6.78190231e-01 -8.83104980e-01 8.12997520e-01 7.00119138e-01
-9.90501940e-01 2.05642298e-01 -4.00336623e-01 -4.36970353e-01
9.09381330e-01 1.01680708e+00 -1.00261283e+00 5.81390738e-01
6.42854750e-01 7.63147026e-02 4.26858515e-01 1.83630371e+00
-4.06059414e-01 1.28925085e+00 -8.13985348e-01 -6.42290831e-01
3.01492587e-02 -9.93079543e-02 9.51586425e-01 9.69347537e-01
7.12709725e-01 2.44784757e-01 -1.07279323e-01 1.41646326e-01
2.02149957e-01 -4.14927065e-01 -5.45638561e-01 -2.48805508e-02
6.70399547e-01 8.24325085e-01 -9.04720128e-01 6.57368870e-03
-2.47373417e-01 3.77985537e-01 -5.53477228e-01 3.99024487e-01
-7.44863808e-01 -8.91533077e-01 5.99344492e-01 1.07092343e-01
5.17916143e-01 -8.23113024e-01 1.95967272e-01 -8.28459024e-01
-6.45667315e-01 -3.36273789e-01 1.56396460e-02 -2.79582113e-01
-8.56685936e-01 2.51208961e-01 2.49751255e-01 -1.59340477e+00
-7.51858592e-01 -5.91053247e-01 -5.66077411e-01 1.23387253e+00
-1.78017187e+00 -5.11879921e-02 -3.33057716e-02 2.63935238e-01
1.17508516e-01 -1.62588164e-01 9.93956804e-01 3.14073622e-01
-1.20792918e-01 4.07491289e-02 5.08470953e-01 2.21322760e-01
4.48218822e-01 -1.11993301e+00 4.38017339e-01 1.17638135e+00
2.55640984e-01 6.56945050e-01 1.60747969e+00 -7.02053845e-01
-1.51238573e+00 -4.42754894e-01 5.51806867e-01 1.23407029e-01
8.75368953e-01 -4.49004993e-02 -6.24171615e-01 3.07307541e-01
1.45004749e-01 -2.06102639e-01 8.26166987e-01 -2.91032374e-01
4.33800966e-01 2.09656969e-01 -1.15008140e+00 8.61449763e-02
2.58176267e-01 -1.66035578e-01 -3.97623569e-01 4.03922915e-01
1.60109207e-01 -6.95292830e-01 -5.41400850e-01 3.48929256e-01
6.08183920e-01 -7.77375638e-01 5.93747556e-01 2.61450469e-01
-2.99365640e-01 -7.76422977e-01 9.90847349e-02 -1.27700937e+00
-1.61675259e-01 -1.40020120e+00 -3.08277253e-02 8.68629992e-01
3.55883300e-01 -8.68874729e-01 9.16282833e-01 -2.44525552e-01
1.84923988e-02 2.14940578e-01 -1.54651999e+00 -8.61126244e-01
-1.38513535e-01 -5.89804888e-01 1.61958877e-02 3.73124063e-01
-1.30851597e-01 2.29615912e-01 -4.73291665e-01 7.16478050e-01
1.33002102e+00 5.77847436e-02 5.69781482e-01 -1.37079716e+00
-8.12977612e-01 -1.50836825e-01 -6.27810538e-01 -1.54353833e+00
-3.82977098e-01 -2.83543020e-01 8.13050389e-01 -1.36805153e+00
-2.95026720e-01 -4.23928589e-01 3.48148122e-02 -3.75268459e-01
-4.08493653e-02 5.67075312e-01 -3.31286430e-01 1.25256151e-01
-6.05651550e-02 1.36024952e-01 5.65509617e-01 2.88919151e-01
2.18262807e-01 6.45795882e-01 -1.80388272e-01 8.39694917e-01
5.81057250e-01 -1.01820970e+00 4.22935411e-02 -2.23421931e-01
6.20354116e-01 2.87236869e-01 5.23485281e-02 -1.43658209e+00
5.39425373e-01 2.48264242e-02 -2.99319066e-03 -5.52285254e-01
4.15936649e-01 -6.35580182e-01 3.22289497e-01 1.17001665e+00
3.82591784e-01 -2.38890216e-01 7.58345947e-02 9.41879749e-01
-2.27623656e-01 -1.07569146e+00 1.00449753e+00 -1.86558276e-01
-6.26132071e-01 -4.76003513e-02 -1.18132401e+00 -1.06770962e-01
8.12922835e-01 5.75099587e-02 -2.12263122e-01 -7.92532206e-01
-5.69446027e-01 1.49129614e-01 7.83776790e-02 -1.32689148e-01
6.07998788e-01 -7.41053760e-01 -8.30403984e-01 4.42514494e-02
-2.66811967e-01 -8.37562233e-02 5.38302325e-02 8.35710943e-01
-8.04655373e-01 3.11087638e-01 4.02341127e-01 -9.42269444e-01
-1.18354249e+00 -5.94650470e-02 3.36459517e-01 -2.51010191e-02
-4.58544374e-01 1.16004562e+00 -5.30263901e-01 5.16300797e-01
-8.23719054e-02 -3.05882487e-02 -1.78595096e-01 -2.02979654e-01
7.28248000e-01 3.15663755e-01 -1.77070826e-01 -3.15182030e-01
-3.58203977e-01 7.48744488e-01 8.12540203e-02 -6.98716342e-01
8.12838614e-01 -2.69538790e-01 8.52680355e-02 7.83423364e-01
7.13724434e-01 4.48541433e-01 -1.04324210e+00 -2.08701581e-01
3.30333799e-01 -2.99610913e-01 1.37655586e-01 -3.21666747e-01
-3.09672296e-01 7.44934559e-01 8.31072688e-01 7.98053563e-01
7.30264187e-01 -7.48635083e-02 6.20743752e-01 4.82317686e-01
1.03866541e+00 -6.86647356e-01 -2.84612447e-01 5.03930986e-01
2.14396268e-01 -9.89665806e-01 4.68382090e-01 -3.11725646e-01
1.54005766e-01 1.17851520e+00 -2.71203667e-01 -3.71617079e-01
1.09787047e+00 8.46608698e-01 1.61887780e-01 1.25682624e-02
-4.17689562e-01 -3.17576170e-01 -2.86164165e-01 4.15831715e-01
3.48546326e-01 6.26944080e-02 -4.17451322e-01 -2.22122326e-01
-2.00580135e-01 -1.02350064e-01 1.19882011e+00 1.23561013e+00
-1.09539950e+00 -1.13549387e+00 -6.64245963e-01 4.42402661e-01
-7.80273259e-01 -1.30064979e-01 3.72423768e-01 4.34492916e-01
-4.30378228e-01 1.12035835e+00 -2.59298012e-02 2.59648804e-02
3.70945595e-02 -2.00247150e-02 6.45329952e-01 -7.37028122e-01
3.40268649e-02 3.95641774e-01 1.47827387e-01 -2.00006608e-02
-7.92691171e-01 -1.24848080e+00 -1.54712260e+00 -1.49480775e-01
-7.55916417e-01 7.57431567e-01 1.00795114e+00 1.08185101e+00
-3.03049058e-01 3.67893487e-01 7.82831788e-01 -8.27388108e-01
-7.03427553e-01 -1.12802625e+00 -9.65578139e-01 -4.43989366e-01
5.98863840e-01 -7.20037222e-01 -1.04242861e+00 -3.20212334e-01] | [6.756054401397705, 3.6839001178741455] |
0dd426ef-2ea9-4a20-af79-89e63590c8b2 | leveraging-information-bottleneck-for | 2110.01280 | null | https://arxiv.org/abs/2110.01280v1 | https://arxiv.org/pdf/2110.01280v1.pdf | Leveraging Information Bottleneck for Scientific Document Summarization | This paper presents an unsupervised extractive approach to summarize scientific long documents based on the Information Bottleneck principle. Inspired by previous work which uses the Information Bottleneck principle for sentence compression, we extend it to document level summarization with two separate steps. In the first step, we use signal(s) as queries to retrieve the key content from the source document. Then, a pre-trained language model conducts further sentence search and edit to return the final extracted summaries. Importantly, our work can be flexibly extended to a multi-view framework by different signals. Automatic evaluation on three scientific document datasets verifies the effectiveness of the proposed framework. The further human evaluation suggests that the extracted summaries cover more content aspects than previous systems. | ['Shirui Pan', 'Lan Du', 'Yuan Jin', 'Huan Yee Koh', 'Ming Liu', 'Jiaxin Ju'] | 2021-10-04 | null | https://aclanthology.org/2021.findings-emnlp.345 | https://aclanthology.org/2021.findings-emnlp.345.pdf | findings-emnlp-2021-11 | ['sentence-compression', 'scientific-article-summarization'] | ['natural-language-processing', 'natural-language-processing'] | [ 6.50038779e-01 1.64658770e-01 -2.45399162e-01 -1.94072977e-01
-1.07500041e+00 -5.40440261e-01 5.49627364e-01 5.65124810e-01
-3.22075635e-01 8.76272142e-01 9.86099243e-01 -8.70553181e-02
-8.41663480e-02 -6.46292448e-01 -4.07962203e-01 -4.83521134e-01
1.41297251e-01 1.69254988e-01 2.88296580e-01 -1.64644718e-01
1.14063203e+00 2.40647942e-01 -1.16756356e+00 4.31127459e-01
1.10169625e+00 3.48565072e-01 6.05504453e-01 1.13663125e+00
-6.22528315e-01 7.74282217e-01 -1.11406600e+00 -3.58573586e-01
-2.88771361e-01 -6.74979031e-01 -1.00056303e+00 7.74577856e-02
2.20438942e-01 -3.78313363e-01 -2.13118926e-01 1.03154564e+00
8.32841396e-01 1.50177523e-01 6.61741912e-01 -4.90877241e-01
-4.52921808e-01 1.05673373e+00 -7.24209070e-01 3.97234589e-01
7.13553548e-01 -2.99906313e-01 9.89381135e-01 -7.57517636e-01
7.04719186e-01 1.32131624e+00 2.98310429e-01 3.51938993e-01
-9.31501746e-01 -2.93764383e-01 1.69559091e-01 2.63059624e-02
-8.76810730e-01 -7.20346272e-01 9.40286160e-01 1.51782197e-05
1.09490168e+00 5.73917329e-01 6.17972195e-01 8.44948053e-01
4.87135679e-01 9.84457791e-01 6.25733554e-01 -5.84843278e-01
2.53283232e-01 8.99363533e-02 5.46338618e-01 4.64126438e-01
5.59839606e-01 -5.98221064e-01 -8.92554462e-01 -2.91220248e-01
7.42847752e-03 -1.92247242e-01 -2.99478769e-01 2.21259654e-01
-1.28791845e+00 6.67473197e-01 -2.43925929e-01 3.83932292e-01
-4.37651277e-01 4.19004597e-02 9.66722131e-01 4.11266655e-01
5.65145314e-01 5.18773139e-01 -1.07274354e-02 -1.49000317e-01
-1.51614511e+00 1.67676225e-01 1.01918590e+00 1.15930152e+00
1.73917159e-01 7.37594366e-02 -6.93844736e-01 6.45736098e-01
5.22860810e-02 4.30044472e-01 6.24090254e-01 -1.08414507e+00
7.04195023e-01 3.67760152e-01 -5.10323606e-02 -1.09193134e+00
-7.65810907e-02 -6.08332813e-01 -9.33241010e-01 -6.19250476e-01
-3.85918736e-01 -7.61829764e-02 -4.38833356e-01 1.18717396e+00
-3.56644802e-02 -2.92827487e-01 4.53330994e-01 3.76872689e-01
1.20909357e+00 9.31635797e-01 -3.10195148e-01 -8.27769637e-01
1.36365104e+00 -1.07516861e+00 -1.17869687e+00 1.50073498e-01
3.19285780e-01 -9.46264446e-01 7.22536922e-01 5.51834762e-01
-1.41919076e+00 -4.68644291e-01 -1.37153864e+00 -1.85964480e-01
-1.07104853e-01 3.70225787e-01 4.84001726e-01 4.29633141e-01
-1.20014250e+00 4.98711467e-01 -5.96127748e-01 -5.04863679e-01
2.49784753e-01 8.11812803e-02 -3.12684178e-02 7.46521458e-04
-1.04341257e+00 4.78792489e-01 6.97369516e-01 -1.51186556e-01
-7.23241210e-01 -3.32430720e-01 -5.62927008e-01 4.22079265e-01
4.39335316e-01 -1.00103974e+00 1.14957654e+00 -3.19828898e-01
-1.64763904e+00 3.49908262e-01 -6.33914709e-01 -5.39702952e-01
2.79075652e-01 -3.89361441e-01 -6.02033809e-02 8.27608049e-01
2.38932610e-01 2.86744475e-01 6.85755312e-01 -1.18217444e+00
-4.90038365e-01 -2.09837124e-01 -1.09240964e-01 3.03278685e-01
-6.87693834e-01 2.00703874e-01 -6.89609647e-01 -8.17317843e-01
3.91512066e-02 -3.43226582e-01 5.04265167e-02 -6.43691063e-01
-9.10929263e-01 -2.68764228e-01 6.66825652e-01 -7.80922115e-01
1.83851862e+00 -1.88031387e+00 3.59164864e-01 -7.50035644e-02
2.82197088e-01 2.69401837e-02 -2.41002098e-01 1.05637860e+00
3.07368219e-01 3.11169654e-01 -3.33558768e-01 -4.71483529e-01
-2.23213255e-01 -3.11405748e-01 -6.71566904e-01 4.92147841e-02
-1.37347952e-01 7.24260151e-01 -1.02281594e+00 -9.46631193e-01
-2.80891299e-01 3.06696668e-02 -3.79250675e-01 9.61133465e-02
-4.55966145e-02 7.16257170e-02 -6.80841386e-01 4.46099371e-01
4.83641148e-01 1.70717146e-02 -2.68614143e-02 -1.89811528e-01
-2.26769909e-01 4.08403069e-01 -7.74044275e-01 1.93745744e+00
-1.68384716e-01 6.91604376e-01 7.03300291e-04 -1.10516906e+00
9.10642207e-01 3.00226271e-01 4.77672279e-01 -2.78394669e-01
-6.18649833e-02 1.33537084e-01 -2.68032908e-01 -7.76756048e-01
1.09037983e+00 8.03096667e-02 -1.98130712e-01 6.84809268e-01
1.87684163e-01 -3.57052565e-01 7.25140095e-01 1.11921155e+00
1.09653091e+00 1.31548569e-01 4.77015465e-01 -2.79280216e-01
8.51392329e-01 -1.14869261e-02 3.11090291e-01 1.02839077e+00
-9.74261668e-03 6.85138702e-01 7.00068831e-01 1.70401469e-01
-9.20187294e-01 -8.43897879e-01 7.01037571e-02 8.11193526e-01
7.53611373e-03 -9.71509635e-01 -8.78628552e-01 -7.65788555e-01
-2.33785644e-01 1.00691974e+00 -1.04622014e-01 -1.58505246e-01
-3.78856391e-01 -5.71828246e-01 6.94660604e-01 2.62977868e-01
4.61953521e-01 -1.05135810e+00 -7.84472644e-01 1.98621601e-01
-6.24133646e-01 -8.36848080e-01 -8.12287569e-01 -1.23421073e-01
-1.23738813e+00 -5.90643823e-01 -7.18926728e-01 -6.19533360e-01
6.09275103e-01 5.39385855e-01 7.89344549e-01 -4.49542329e-02
-1.17323957e-01 5.02506196e-01 -4.89220023e-01 -6.23163223e-01
-6.45299792e-01 2.69560635e-01 -6.20287545e-02 -3.67142469e-01
1.68183789e-01 -5.93078375e-01 -5.48717499e-01 -5.79119384e-01
-1.07330739e+00 8.34432244e-02 8.43297422e-01 5.64245760e-01
4.39616323e-01 1.60444930e-01 1.06603813e+00 -1.00878072e+00
1.49541926e+00 -2.58753061e-01 1.89020429e-02 3.94728512e-01
-6.69865370e-01 3.73595387e-01 7.94952512e-01 -3.22143510e-02
-1.30390787e+00 -3.11968088e-01 -3.61739583e-02 7.92449787e-02
1.29475594e-01 9.00968134e-01 -1.13762245e-02 5.78439951e-01
4.13527757e-01 8.73903990e-01 -1.05492197e-01 -4.74718332e-01
3.73283595e-01 9.70657229e-01 5.04446149e-01 -3.01103532e-01
4.87300217e-01 4.62341428e-01 -1.48690194e-01 -1.07129133e+00
-7.72514343e-01 -5.32358825e-01 -6.09033406e-01 -2.95104921e-01
6.93063676e-01 -6.10442996e-01 -6.11902773e-01 5.68922386e-02
-1.54516172e+00 6.51118755e-01 -3.17738801e-01 5.54701805e-01
-5.78944445e-01 1.01938677e+00 -6.37588501e-01 -9.26761508e-01
-1.17268813e+00 -7.68954158e-01 1.18227530e+00 1.74398705e-01
-2.91999102e-01 -7.64234006e-01 1.65194571e-01 3.10747981e-01
4.39687252e-01 -5.54787479e-02 9.86084819e-01 -8.76358151e-01
-3.70636046e-01 -3.68477970e-01 6.63665906e-02 1.86395660e-01
2.00940594e-01 1.27814487e-01 -6.98145032e-01 -3.43084723e-01
3.97751063e-01 -2.42746651e-01 1.46238935e+00 4.66547579e-01
9.29232359e-01 -5.77964187e-01 -4.29052830e-01 2.41330951e-01
1.31287801e+00 1.19908206e-01 5.81791222e-01 2.49796063e-01
2.67038584e-01 6.56942129e-01 4.97044533e-01 5.60640991e-01
2.49070629e-01 5.45549989e-02 -8.20894390e-02 3.33638102e-01
-6.06714897e-02 -3.82267326e-01 5.93617082e-01 1.80477524e+00
1.40798002e-01 -5.19532979e-01 -2.59523660e-01 5.96949160e-01
-1.67329526e+00 -1.38590527e+00 6.72643110e-02 2.01532149e+00
9.77173090e-01 3.69038701e-01 -1.39637515e-01 1.55104294e-01
5.54838121e-01 5.14541984e-01 -5.07186830e-01 -7.15617836e-01
-9.29953977e-02 -2.11108461e-01 2.72739053e-01 5.02343833e-01
-7.27328539e-01 7.73793697e-01 6.74610996e+00 9.34161067e-01
-9.14228141e-01 -2.36870080e-01 2.86600441e-01 -4.43914890e-01
-7.00780213e-01 1.00087143e-01 -7.97778249e-01 3.97873878e-01
1.04241562e+00 -9.06901181e-01 9.87855345e-02 3.43888402e-01
5.18401802e-01 -4.64739054e-01 -8.79569948e-01 6.63659692e-01
5.29349566e-01 -1.48314691e+00 6.29166067e-01 -1.88247278e-01
5.32071531e-01 -3.90055388e-01 -3.06949109e-01 2.04298526e-01
-1.05397925e-01 -5.95667422e-01 5.46940684e-01 1.06433153e+00
5.95045269e-01 -9.78873730e-01 6.81777298e-01 7.40155160e-01
-9.98164952e-01 1.49435177e-01 -6.44455850e-01 1.53722718e-01
3.08192492e-01 8.77752900e-01 -6.68990672e-01 1.18923402e+00
2.68654823e-01 7.89614677e-01 -6.20547295e-01 8.79957199e-01
-7.53525868e-02 7.36976981e-01 -5.05459011e-02 -3.64569217e-01
-6.53339252e-02 -3.08877975e-01 1.13194597e+00 1.60298586e+00
4.25157309e-01 -2.30094623e-02 -1.21100709e-01 6.27815306e-01
-2.60945797e-01 4.20264035e-01 -8.00906360e-01 -2.44332626e-01
5.92993915e-01 1.21967721e+00 -8.44129860e-01 -7.50940502e-01
-1.89047277e-01 1.16168928e+00 1.36027679e-01 3.70246977e-01
-3.13074350e-01 -1.03134906e+00 -4.97650921e-01 -3.11290622e-01
4.00329977e-01 -1.27821952e-01 -3.53670448e-01 -1.34348202e+00
2.07500756e-01 -7.75575340e-01 1.88077316e-01 -6.85597897e-01
-7.97828913e-01 5.14440060e-01 1.44364953e-01 -1.16084516e+00
-2.69804239e-01 3.63379419e-01 -9.69017804e-01 6.25458896e-01
-1.37619436e+00 -7.58265615e-01 1.46264017e-01 6.29172921e-02
9.94769275e-01 -2.81875074e-01 8.50723207e-01 -1.41377434e-01
-4.26593006e-01 2.77571470e-01 5.30411184e-01 -1.97407871e-01
7.65661538e-01 -1.21446538e+00 2.23413482e-01 1.14265132e+00
-2.68672276e-02 9.83042777e-01 9.40757394e-01 -8.36374938e-01
-1.63841403e+00 -7.86036909e-01 1.25083113e+00 -9.97052938e-02
3.09137017e-01 -2.99942374e-01 -7.43392825e-01 3.50348949e-01
1.04625452e+00 -1.05502009e+00 7.76705325e-01 -2.66438305e-01
1.09113947e-01 -1.29523531e-01 -9.18199003e-01 5.53850830e-01
8.50912869e-01 -2.49675199e-01 -1.01715291e+00 2.74800003e-01
1.15172815e+00 -1.35559320e-01 -8.22224438e-01 2.66762227e-01
4.43894655e-01 -7.49271095e-01 6.89658582e-01 -4.43811983e-01
9.04046416e-01 -1.91836819e-01 2.74960436e-02 -1.27509022e+00
-1.45727828e-01 -7.57012546e-01 -4.29741532e-01 1.65312660e+00
2.60426909e-01 -2.13512361e-01 3.56821358e-01 3.36300209e-02
-2.09142193e-01 -6.91619098e-01 -7.97289789e-01 -5.70019245e-01
-9.45885032e-02 -1.59012258e-01 3.47866952e-01 4.03014779e-01
4.94930297e-01 9.08756852e-01 -4.20845300e-01 -3.31607997e-01
7.12315083e-01 4.79214519e-01 6.05812728e-01 -1.08654451e+00
-1.60747916e-01 -5.64648867e-01 1.73373267e-01 -1.37651122e+00
1.76083688e-02 -1.16264594e+00 1.34517057e-02 -2.10470390e+00
8.03374112e-01 4.25226808e-01 -2.30153948e-02 -1.73461393e-01
-3.52461755e-01 -3.75668705e-01 2.35472947e-01 3.52391124e-01
-9.37714458e-01 8.94913435e-01 1.29067552e+00 -3.50118190e-01
-3.17125827e-01 1.48470432e-01 -1.26513529e+00 3.84317905e-01
8.20261419e-01 -5.15282214e-01 -6.30460203e-01 -2.43423119e-01
2.05400661e-01 4.06483620e-01 -4.01299685e-01 -8.58809769e-01
6.45432234e-01 2.02991851e-02 1.63410515e-01 -1.30068505e+00
-1.30072132e-01 -4.17221189e-01 -4.45848227e-01 6.09419525e-01
-8.35308373e-01 1.62514105e-01 2.17371374e-01 8.21337700e-01
-3.98489326e-01 -7.18596220e-01 3.23291183e-01 -2.51460761e-01
6.52454644e-02 -1.25445724e-01 -6.72611654e-01 7.28131160e-02
6.66604519e-01 -1.61607385e-01 -3.59240472e-01 -6.09775066e-01
-3.69422019e-01 4.24366385e-01 6.86743334e-02 1.95280209e-01
9.97291803e-01 -9.27225649e-01 -1.06490827e+00 -6.65476248e-02
-1.35436118e-01 -4.75030653e-02 1.34164855e-01 8.67656946e-01
-4.25890595e-01 8.43861759e-01 1.87389553e-01 -3.76972973e-01
-1.31355488e+00 6.40446365e-01 -2.79477656e-01 -5.22244990e-01
-8.10355127e-01 3.54893178e-01 -2.84873635e-01 5.77928722e-02
3.07385236e-01 -2.14417264e-01 -6.57010913e-01 2.99303979e-01
7.97279418e-01 6.41997576e-01 -5.53139187e-02 -2.50382364e-01
-1.42830402e-01 3.72155696e-01 -4.86137241e-01 -4.14998502e-01
1.54123962e+00 -4.68263388e-01 -5.65864563e-01 5.73316097e-01
1.16899967e+00 6.46371245e-01 -4.87588823e-01 -1.78803459e-01
1.98665008e-01 -1.15576960e-01 -5.66781275e-02 -5.95991135e-01
-6.33319855e-01 8.00744414e-01 -7.81445801e-02 4.46116596e-01
1.26022696e+00 -2.08345413e-01 7.35696197e-01 6.96920395e-01
-5.24188504e-02 -1.33373010e+00 8.76559168e-02 5.70857346e-01
1.07813966e+00 -7.56315172e-01 5.00062227e-01 -1.79428652e-01
-5.88626206e-01 1.42848933e+00 2.04424039e-01 3.67198884e-02
3.36186022e-01 2.83154339e-01 -3.98556322e-01 -2.92069912e-01
-1.06507921e+00 2.81362325e-01 7.10598603e-02 1.91093698e-01
7.21789360e-01 -2.70921737e-01 -1.17651343e+00 6.99093521e-01
-3.22084427e-01 1.31775409e-01 1.06936097e+00 9.86954689e-01
-9.46674109e-01 -8.92875195e-01 -3.12952548e-01 7.16989160e-01
-7.27246761e-01 -2.31558204e-01 -6.02119923e-01 1.50730640e-01
-6.62599146e-01 1.27063417e+00 -2.72610992e-01 -1.91161215e-01
4.11030680e-01 2.39780709e-01 2.80231982e-01 -7.44176328e-01
-6.37713432e-01 4.49709386e-01 4.88079526e-02 -2.27615625e-01
-6.09274328e-01 -6.30598366e-01 -1.35295212e+00 -1.69537723e-01
-3.07068974e-01 5.44496536e-01 5.91203749e-01 7.28467524e-01
6.57505870e-01 9.15704846e-01 7.44607568e-01 -5.54366529e-01
-8.12904954e-01 -1.27255714e+00 -2.70987928e-01 1.15009360e-01
4.32119906e-01 1.43729329e-01 -4.21419144e-01 2.32420534e-01] | [12.535895347595215, 9.5143461227417] |
49e0c918-e52e-4904-b8b3-110259e243af | generating-code-with-the-help-of-retrieved | 2104.05310 | null | https://arxiv.org/abs/2104.05310v2 | https://arxiv.org/pdf/2104.05310v2.pdf | Generating Code with the Help of Retrieved Template Functions and Stack Overflow Answers | We approach the important challenge of code autocompletion as an open-domain task, in which a sequence-to-sequence code generator model is enhanced with the ability to attend to reference code snippets supplied by a semantic code search engine. In this work, we present a novel framework to precisely retrieve template functions as well as intent-snippet pairs and effectively train such a retrieval-guided code generator. To demonstrate the effectiveness of our model designs, we perform extensive experiments with CodeSearchNet which contains template functions and CoNaLa which contains Stack Overflow intent-snippet pairs. We also investigate different retrieval models, including Elasticsearch, DPR, and our fusion representation search model, which currently holds the number one spot on the CodeSearchNet leaderboard. We observe improvements by leveraging multiple database elements and further gain from retrieving diverse data points by using Maximal Marginal Relevance. Overall, we see a 4% improvement to cross-entropy loss, a 15% improvement to edit distance, and a 44% improvement to BLEU score when retrieving template functions. We see subtler improvements of 2%, 11%, and 6% respectively when retrieving Stack Overflow intent-snippet pairs. We also create a novel Stack Overflow-Function Alignment dataset, which consists of 150K tuples of functions and Stack Overflow intent-snippet pairs that are of help in writing the associated function, of which 1.7K are manually curated. | ['Changran Hu', 'Neel Sundaresan', 'Mikhail Breslav', 'Chen Wu', 'Dawn Drain'] | 2021-04-12 | null | null | null | null | ['code-search', 'code-search'] | ['computer-code', 'computer-vision'] | [ 2.84606785e-01 -3.10855433e-02 -1.49611384e-01 -1.56472683e-01
-1.38925970e+00 -9.75181460e-01 3.03571850e-01 8.87821391e-02
-1.07975580e-01 4.87922788e-01 3.11952621e-01 -5.65537572e-01
-1.46931201e-01 -6.03710949e-01 -1.02604234e+00 -2.07967967e-01
-5.65494038e-02 3.40934932e-01 2.15318814e-01 -4.20932651e-01
6.58019364e-01 -1.95018455e-01 -1.69224405e+00 6.00107253e-01
1.31726015e+00 5.94411671e-01 6.84225738e-01 8.40150416e-01
-4.24296796e-01 6.92447901e-01 -8.79912376e-01 -7.49546289e-01
2.17579693e-01 -3.49240661e-01 -7.80760348e-01 -7.51073897e-01
4.60348845e-01 -1.69353381e-01 -2.01152280e-01 1.14656317e+00
6.06890976e-01 -1.87059790e-01 3.51093918e-01 -1.23500097e+00
-8.62667739e-01 8.79867971e-01 -3.47453117e-01 2.14781389e-01
1.00791144e+00 2.44338155e-01 1.31006992e+00 -1.03416634e+00
6.90092504e-01 8.90002191e-01 8.22313666e-01 5.91420293e-01
-9.84524250e-01 -6.93354964e-01 -4.14808869e-01 1.65232837e-01
-1.38006151e+00 -4.65361178e-01 2.16268882e-01 -4.50400144e-01
1.60444617e+00 3.90319318e-01 6.24648444e-02 9.41366851e-01
1.22531550e-02 7.82744586e-01 2.09113240e-01 -4.69765455e-01
7.92581290e-02 -1.16318196e-01 1.71269998e-01 7.70447731e-01
2.82290131e-01 2.08240543e-02 -2.95686603e-01 -6.23892665e-01
1.71618178e-01 2.34848186e-01 -4.15536672e-01 -2.54780799e-01
-1.02201486e+00 6.18204355e-01 2.03259304e-01 5.72877526e-02
1.71480238e-01 3.20655346e-01 6.41312361e-01 5.60517311e-01
-2.14430600e-01 7.38667965e-01 -6.44222081e-01 -6.54620051e-01
-9.49262857e-01 5.38106799e-01 1.06573439e+00 1.76539969e+00
9.96199489e-01 -2.92171687e-01 -4.56774056e-01 1.01532876e+00
1.96581483e-01 8.11211824e-01 9.13475275e-01 -7.72838533e-01
1.11923003e+00 9.02906835e-01 -5.22187762e-02 -6.55960977e-01
7.13653862e-02 -3.59539270e-01 -1.15059882e-01 -3.28062207e-01
-7.86838960e-03 4.28603776e-02 -5.18684506e-01 1.58728397e+00
-3.25643867e-01 -1.46312676e-02 2.05346555e-01 4.32492435e-01
5.93046546e-01 3.65762979e-01 -3.79654557e-01 2.21798778e-01
1.37636864e+00 -1.32500386e+00 -3.49941671e-01 -4.07453924e-01
1.01246619e+00 -1.03115427e+00 1.36808193e+00 -3.59105412e-03
-1.06592524e+00 -3.32139581e-01 -1.13394463e+00 -8.56751576e-02
-2.49649674e-01 6.81420416e-02 3.20192575e-01 6.32449746e-01
-1.21766102e+00 6.40871346e-01 -5.49488127e-01 -2.87447721e-01
1.15465462e-01 1.69302642e-01 -2.20817044e-01 -2.12891638e-01
-7.97978938e-01 6.21787846e-01 4.64230835e-01 -5.48896849e-01
-7.62834251e-01 -1.25318718e+00 -9.25681412e-01 5.55009604e-01
4.14820492e-01 -7.94448078e-01 1.41616380e+00 -5.39978743e-01
-8.81519675e-01 6.87324464e-01 -2.88347006e-01 -4.17457789e-01
1.00842454e-01 -1.57775939e-01 -5.30326426e-01 -8.87817889e-02
3.46231371e-01 1.98218182e-01 4.99769121e-01 -7.81398058e-01
-6.48733497e-01 -2.41948739e-01 1.20470181e-01 -2.03099828e-02
-3.42078716e-01 3.51359904e-01 -6.53805494e-01 -8.35683107e-01
-4.27852839e-01 -9.21960056e-01 1.53932855e-01 -3.69917989e-01
-2.60422438e-01 -2.28427984e-02 5.34852982e-01 -9.49371338e-01
1.82931125e+00 -2.32375836e+00 -2.27176528e-02 1.82451770e-01
2.02452958e-01 3.70492280e-01 -4.66319054e-01 6.46123886e-01
-2.14720711e-01 3.32311064e-01 -5.43540537e-01 7.09608719e-02
3.35092932e-01 -3.75178047e-02 -5.75337648e-01 -2.08823562e-01
2.46143401e-01 1.21276605e+00 -1.12408173e+00 -1.10207714e-01
-4.52165008e-01 3.43647115e-02 -1.22754550e+00 2.96018064e-01
-5.41229069e-01 -4.24607486e-01 -3.49544644e-01 7.83734858e-01
4.58410591e-01 -4.40268993e-01 9.76646617e-02 3.19078624e-01
2.59760857e-01 5.89419127e-01 -7.48796523e-01 2.02257466e+00
-7.92757630e-01 4.15893197e-01 -1.96206108e-01 -4.40357357e-01
1.14612758e+00 1.33192003e-01 2.58949939e-02 -9.34072614e-01
-5.52012444e-01 6.25270247e-01 -2.13105202e-01 -5.22793174e-01
8.81441593e-01 5.33440888e-01 -3.52983028e-01 6.18725002e-01
4.29229401e-02 9.86337662e-02 4.24213350e-01 4.03084993e-01
2.05509663e+00 1.88546345e-01 1.13162801e-01 -2.22387031e-01
6.87918544e-01 -6.88943081e-03 2.34098405e-01 9.70262110e-01
2.77461469e-01 7.03142107e-01 5.67929685e-01 -1.33241281e-01
-1.21733022e+00 -9.07718599e-01 7.84396231e-02 1.22696340e+00
1.44936368e-02 -1.07892239e+00 -8.96505713e-01 -8.92011225e-01
1.09525122e-01 8.69734406e-01 -2.02595845e-01 -5.01600385e-01
-9.88066375e-01 -5.50757229e-01 1.08549690e+00 6.42825544e-01
3.42082053e-01 -8.25116992e-01 -4.19216484e-01 1.31830156e-01
-3.30534935e-01 -7.14139044e-01 -1.22356915e+00 2.06775457e-01
-6.73779607e-01 -1.28803217e+00 -6.88471675e-01 -9.36056376e-01
7.79826283e-01 6.54995441e-02 1.67975688e+00 5.03181994e-01
-2.26057321e-01 2.68864423e-01 -5.69012284e-01 1.86516926e-01
-9.16397333e-01 4.25540119e-01 -4.97099042e-01 -7.71461904e-01
5.09847820e-01 -5.95587552e-01 -5.60953557e-01 6.19804502e-01
-1.03503656e+00 -4.05965477e-01 5.90755045e-01 9.62826967e-01
1.59608260e-01 -4.18657750e-01 4.70266372e-01 -8.13410819e-01
7.95504630e-01 -8.32591772e-01 -7.30857313e-01 6.33884192e-01
-8.25024426e-01 4.96209353e-01 7.86293566e-01 -3.68729346e-02
-9.40951645e-01 -6.21659644e-02 -2.68986911e-01 -4.13189054e-01
2.68705308e-01 5.61941803e-01 -4.72223684e-02 1.26669146e-02
1.10554230e+00 4.02171463e-01 1.05989166e-01 -5.94430029e-01
4.11449432e-01 8.84017110e-01 7.40704179e-01 -9.60502088e-01
8.21600378e-01 -2.82848537e-01 -6.91923499e-01 6.57650009e-02
-1.03756748e-01 -6.79042339e-01 -3.16018276e-02 5.19171596e-01
4.54799861e-01 -9.96953964e-01 -5.66946864e-01 5.11628576e-02
-1.11893618e+00 -6.98737726e-02 -1.63393438e-01 -1.82973102e-01
-6.88280404e-01 6.14117563e-01 -4.63866144e-01 -2.67791033e-01
-6.73679829e-01 -1.50093293e+00 1.34126425e+00 -4.00378555e-02
-3.01356465e-01 -6.44387305e-01 2.47406319e-01 4.22198683e-01
6.59174562e-01 -3.09153736e-01 1.36941218e+00 -1.10472655e+00
-9.94214118e-01 -8.54456648e-02 -2.38322631e-01 2.37789318e-01
1.37690306e-02 -1.73626468e-01 -5.61518312e-01 -4.40583050e-01
-3.54863375e-01 -1.68950543e-01 7.76291549e-01 -4.75365996e-01
1.14905119e+00 -3.65333825e-01 -4.84329641e-01 7.95445561e-01
1.54412305e+00 5.05211413e-01 8.29525948e-01 2.98714727e-01
4.16088372e-01 1.18811525e-01 4.03259069e-01 5.00426888e-01
4.11743015e-01 7.52377629e-01 4.28938866e-01 6.21323884e-01
-3.12248796e-01 -5.76479435e-01 5.52009940e-01 1.14862359e+00
4.97110128e-01 -1.20895140e-01 -1.11235094e+00 6.97265983e-01
-1.67904222e+00 -8.41349959e-01 2.40384609e-01 2.41674709e+00
1.10369813e+00 -1.66440740e-01 -3.52887399e-02 -1.60887182e-01
7.35816240e-01 -2.99222559e-01 -5.18507361e-01 -3.04662883e-01
2.28948500e-02 5.04032075e-01 5.50017059e-01 3.80092978e-01
-5.64615726e-01 6.76623702e-01 6.20906878e+00 1.01169300e+00
-6.12044573e-01 -1.58015624e-01 1.14920564e-01 1.18441820e-01
-8.17956686e-01 2.59351611e-01 -1.05301535e+00 9.94265497e-01
1.18460751e+00 -6.22789681e-01 9.80453372e-01 1.23110831e+00
-5.48196614e-01 1.24121264e-01 -1.54085028e+00 9.80484843e-01
3.57496381e-01 -1.48889530e+00 -3.33120562e-02 -2.03399792e-01
7.09438264e-01 2.13798255e-01 -1.87251598e-01 8.31954956e-01
4.61152494e-01 -8.78783762e-01 5.54620922e-01 5.54305911e-01
9.69189465e-01 -5.79683602e-01 6.08326912e-01 1.16845548e-01
-1.20514214e+00 -3.97805631e-01 -2.80812144e-01 2.46974111e-01
-2.97756106e-01 3.05365235e-01 -1.08179522e+00 6.67362273e-01
6.01421416e-01 6.70984447e-01 -8.22064102e-01 1.39035165e+00
-9.05598253e-02 1.87235638e-01 -7.96893686e-02 -2.55763948e-01
-1.69052899e-01 1.12208903e-01 4.36782062e-01 1.34625518e+00
7.89347470e-01 -4.68211561e-01 -1.75695077e-01 1.31057930e+00
-4.14922118e-01 -7.22644478e-02 -7.40312397e-01 -1.03996210e-01
9.34761345e-01 8.79331708e-01 -2.06796885e-01 -4.39060539e-01
-5.71951151e-01 1.01082575e+00 2.83650607e-01 3.21370363e-01
-9.31552112e-01 -1.10881770e+00 9.34820116e-01 -1.84550971e-01
7.19636023e-01 1.51966438e-01 7.03004077e-02 -1.23855507e+00
6.30343616e-01 -1.42713070e+00 3.61868680e-01 -8.69169652e-01
-1.04654908e+00 6.03270411e-01 1.73272943e-04 -1.09142518e+00
-7.45141447e-01 -3.77834946e-01 -5.81747472e-01 1.03202271e+00
-1.30496848e+00 -4.30345297e-01 -2.80328870e-01 3.62884879e-01
5.74370384e-01 -5.60922205e-01 8.15035343e-01 8.89064014e-01
-1.85440093e-01 1.17669654e+00 4.70990628e-01 1.87217414e-01
5.91895759e-01 -1.26000929e+00 1.12247765e+00 7.51056433e-01
-9.37516838e-02 1.33952963e+00 4.13449705e-01 -7.23092139e-01
-1.91950274e+00 -1.14494014e+00 9.97847259e-01 -8.82302880e-01
7.50037909e-01 -4.99540359e-01 -9.29357946e-01 6.85304761e-01
1.22837201e-02 -9.93035510e-02 5.16604781e-01 -2.11209416e-01
-7.07708597e-01 -1.39151206e-02 -1.06997812e+00 5.83563626e-01
1.30117214e+00 -8.21099699e-01 -7.91021347e-01 2.51506656e-01
1.24696910e+00 -7.09397435e-01 -8.29618156e-01 2.98877239e-01
3.87294888e-01 -8.02578807e-01 1.19718492e+00 -4.10674125e-01
7.08467901e-01 -3.05437863e-01 -3.71329695e-01 -1.21478379e+00
-1.47029469e-02 -7.80181110e-01 -1.53729752e-01 1.13097394e+00
7.73730993e-01 -4.78579104e-01 7.19840348e-01 4.88131076e-01
-5.83648682e-01 -5.00137210e-01 -6.19505882e-01 -1.05251384e+00
-3.77860456e-03 -3.33988339e-01 1.07656884e+00 6.66293025e-01
4.07023489e-01 2.43402094e-01 5.02150767e-02 -1.58628047e-01
1.34281352e-01 1.95128113e-01 7.64322340e-01 -6.01637423e-01
-8.14399004e-01 -6.25056148e-01 -1.49272740e-01 -1.26949739e+00
1.96732700e-01 -1.57040095e+00 -2.57002879e-02 -1.02386606e+00
4.05795306e-01 -3.61655265e-01 -4.61847484e-02 4.69587564e-01
-3.27495307e-01 -2.53598988e-01 1.49422690e-01 1.39203608e-01
-6.92531407e-01 4.35988218e-01 6.55709982e-01 -2.57983983e-01
-1.43858371e-02 -1.47679508e-01 -9.43584919e-01 1.86742097e-01
3.87242317e-01 -5.45221925e-01 -5.65299273e-01 -7.30657578e-01
6.98824108e-01 4.76061583e-01 1.73215732e-01 -9.51812625e-01
3.88716698e-01 3.72888684e-01 -2.71229357e-01 -2.97367841e-01
-1.29219055e-01 -6.83958888e-01 2.27942511e-01 3.67785573e-01
-5.34079194e-01 6.74974024e-01 2.79635012e-01 5.58787048e-01
-2.50814617e-01 -8.85166109e-01 1.36866316e-01 -3.52425903e-01
-6.16741955e-01 -8.22318718e-02 -9.67930257e-03 5.90604007e-01
6.80679321e-01 -2.23232061e-01 -8.16108942e-01 3.84049267e-02
-1.46138415e-01 3.35334927e-01 6.78442955e-01 7.95076787e-01
6.02786183e-01 -1.24615395e+00 -4.73145932e-01 4.48301643e-01
6.42658889e-01 -3.69980186e-01 -1.79775745e-01 4.51388985e-01
-6.04751587e-01 6.06660068e-01 -2.51288689e-03 -3.06438893e-01
-1.18495655e+00 6.57361269e-01 2.39760932e-02 -2.37567008e-01
-4.08877969e-01 8.34727764e-01 -4.99196462e-02 -9.52095151e-01
3.36270511e-01 -6.29725695e-01 3.40112805e-01 -4.85302091e-01
6.96095109e-01 5.49137175e-01 6.04609370e-01 -1.34780072e-02
-5.19145191e-01 4.09083843e-01 -2.44727433e-01 3.72771114e-01
9.88375306e-01 2.96646774e-01 -3.50730866e-01 -3.37090790e-01
1.59918046e+00 2.86739469e-01 -8.46455038e-01 -3.22072864e-01
6.18309438e-01 -6.12142503e-01 -5.58090210e-01 -1.09372580e+00
-9.98093486e-01 4.22978044e-01 3.18747431e-01 -7.37738982e-02
1.07694376e+00 3.94112878e-02 1.03199911e+00 8.03775370e-01
6.18338048e-01 -6.68123543e-01 2.09088206e-01 7.75298238e-01
8.34198833e-01 -9.26499784e-01 -5.66008568e-01 -6.63496032e-02
-2.28872478e-01 1.01589072e+00 6.54114425e-01 2.98324935e-02
1.65941805e-01 6.15699172e-01 -4.19202983e-01 -1.46241069e-01
-1.03029311e+00 3.53565887e-02 7.49220401e-02 5.18729508e-01
5.14377356e-01 -3.34545970e-01 -1.30371779e-01 7.75779366e-01
-3.57806742e-01 -4.06839559e-03 3.84547055e-01 1.07736599e+00
-4.06896353e-01 -1.42892838e+00 -2.40530074e-01 7.22338736e-01
-3.93826157e-01 -8.31471205e-01 -9.63485241e-02 4.70628202e-01
-3.17121655e-01 5.96150279e-01 -9.40155610e-03 -5.70055008e-01
4.67374682e-01 3.42902571e-01 1.05462298e-01 -7.81744361e-01
-9.38970268e-01 -4.16822106e-01 1.97178751e-01 -7.58689046e-01
4.12126422e-01 -4.67975050e-01 -1.30991232e+00 -1.92767173e-01
-3.77529085e-01 3.46829504e-01 6.14572167e-01 3.96828949e-01
1.06007004e+00 6.16491675e-01 4.52592134e-01 -8.23699757e-02
-1.04224324e+00 -6.74827099e-01 7.10035488e-02 4.62874830e-01
3.37845951e-01 -3.38499099e-01 -3.58559817e-01 7.72755519e-02] | [7.599555492401123, 8.021174430847168] |
d7747213-d629-44c9-86fe-de1e2af282a2 | zero-and-few-shot-semantic-parsing-with | 2306.00824 | null | https://arxiv.org/abs/2306.00824v1 | https://arxiv.org/pdf/2306.00824v1.pdf | Zero and Few-shot Semantic Parsing with Ambiguous Inputs | Despite the ubiquity of ambiguity in natural language, it is often ignored or deliberately removed in semantic parsing tasks, which generally assume that a given surface form has only one correct logical form. We attempt to address this shortcoming by introducing AmP, a framework, dataset, and challenge for parsing with linguistic ambiguity. We define templates and generate data for five well-documented linguistic ambiguities. Using AmP, we investigate how several few-shot semantic parsing systems handle ambiguity, introducing three new metrics. We find that large pre-trained models perform poorly at capturing the distribution of possible meanings without deliberate instruction. However, models are able to capture distribution well when ambiguity is attested in their inputs. These results motivate a call for ambiguity to be explicitly included in semantic parsing, and promotes considering the distribution of possible outputs when evaluating semantic parsing systems. | ['Benjamin Van Durme', 'Kyle Rawlins', 'Elias Stengel-Eskin'] | 2023-06-01 | null | null | null | null | ['semantic-parsing'] | ['natural-language-processing'] | [ 4.75578487e-01 5.29898345e-01 1.15167983e-01 -7.75696516e-01
-9.36200082e-01 -1.03747118e+00 6.80345416e-01 3.87864798e-01
-2.52100706e-01 6.25564814e-01 4.65756863e-01 -7.50783324e-01
-4.68197605e-03 -9.47966814e-01 -6.90408170e-01 5.06569333e-02
5.54385483e-01 7.44968355e-01 2.91472375e-01 -4.31063056e-01
3.16995859e-01 6.65099248e-02 -1.84980142e+00 4.30309802e-01
1.10024047e+00 5.85625947e-01 2.70648450e-01 4.18455303e-01
-1.07467473e+00 6.91939831e-01 -1.00973785e+00 -4.97325003e-01
-4.51656021e-02 -4.74070221e-01 -1.28104329e+00 -1.37848884e-01
6.27802193e-01 -9.98687744e-02 4.94452715e-01 1.28647721e+00
2.72878379e-01 -5.07884435e-02 4.74271923e-01 -1.09521806e+00
-5.24716496e-01 1.03894711e+00 3.79425824e-01 3.23109865e-01
7.71815002e-01 2.11161450e-01 1.32528782e+00 -6.57926857e-01
8.16729963e-01 1.62768209e+00 5.70318162e-01 8.66424859e-01
-1.24547124e+00 -3.06392491e-01 3.83356512e-01 -5.34565523e-02
-8.05626214e-01 -4.71733630e-01 6.07454062e-01 -3.22497696e-01
1.41693449e+00 7.74195120e-02 1.87428251e-01 1.16991317e+00
-5.08646071e-02 6.43381417e-01 1.21797764e+00 -8.57130051e-01
5.25118351e-01 -2.66710788e-01 8.07906747e-01 5.24917901e-01
6.23820364e-01 -2.11386636e-01 -4.72884238e-01 -1.51197791e-01
3.86412024e-01 -5.49816906e-01 1.96026206e-01 -7.02524781e-02
-7.56969094e-01 7.62413204e-01 -1.22056179e-01 4.05937999e-01
6.12443388e-02 9.30739641e-02 4.97263283e-01 4.54510242e-01
1.77675605e-01 9.27816093e-01 -6.32644773e-01 -5.14113188e-01
-3.30526978e-01 4.34625715e-01 9.78654563e-01 1.15616250e+00
7.61000872e-01 -9.93230492e-02 1.18018046e-01 8.67808580e-01
9.85840857e-02 1.77839205e-01 4.79006946e-01 -1.38427222e+00
5.19833386e-01 8.83112967e-01 1.96977139e-01 -4.54264015e-01
-4.94101465e-01 -3.42211388e-02 2.99025059e-01 6.75978214e-02
9.01868343e-01 -5.61289117e-02 -9.02880371e-01 2.12695670e+00
1.94862634e-01 -2.66269058e-01 5.28370023e-01 6.25014722e-01
7.86596179e-01 3.06932211e-01 7.83346295e-01 7.09404796e-02
1.79035664e+00 -5.05373716e-01 -7.05497921e-01 -1.09415352e+00
9.84010994e-01 -7.27894723e-01 1.61803651e+00 3.73806417e-01
-1.21730638e+00 -3.26874137e-01 -1.00624955e+00 -3.49722624e-01
-4.62434679e-01 -4.60710168e-01 9.30492938e-01 7.38706231e-01
-9.26842928e-01 5.60879409e-01 -6.21440411e-01 -4.38166112e-01
6.78888187e-02 1.96879990e-02 -2.22534746e-01 -5.22216499e-01
-1.38942170e+00 1.14930725e+00 4.41223681e-01 -2.39641413e-01
-4.02168036e-01 -5.67636013e-01 -1.37042332e+00 1.13556623e-01
6.50684237e-01 -8.47164333e-01 1.72546184e+00 -1.18646038e+00
-1.11963069e+00 1.01559615e+00 -4.64167744e-01 -1.88120082e-01
7.29607418e-02 -3.08036864e-01 -3.43268156e-01 -1.03445552e-01
4.75598663e-01 8.03589344e-01 4.15335119e-01 -1.18993437e+00
-5.93501568e-01 -5.17065167e-01 6.48123622e-01 2.82615453e-01
2.97010183e-01 3.54312897e-01 -6.99152984e-03 -4.66748655e-01
4.37468290e-01 -5.71709394e-01 -1.62209004e-01 -5.08393943e-01
-2.49630064e-01 -6.30014598e-01 1.94599926e-01 -3.89854074e-01
1.04301548e+00 -2.06426358e+00 -4.13328744e-02 -3.10984254e-01
-1.79801822e-01 4.43711355e-02 -3.57745886e-01 3.64577681e-01
1.08820900e-01 6.79616749e-01 -4.27493185e-01 -3.94277483e-01
4.61972505e-01 9.05560434e-01 -5.14303982e-01 -4.12789524e-01
6.11527979e-01 8.69892478e-01 -1.11502445e+00 -2.97982901e-01
1.75491065e-01 2.03231187e-03 -6.73566163e-01 2.10221201e-01
-8.32549512e-01 9.59448218e-02 -4.72674310e-01 8.43878686e-01
4.79211539e-01 -1.31001517e-01 4.59165215e-01 2.59433568e-01
5.72414622e-02 1.24353254e+00 -1.20255768e+00 1.81083238e+00
-4.19141144e-01 1.22415505e-01 1.38329372e-01 -7.14539468e-01
6.34887576e-01 1.11459337e-01 -2.95093715e-01 -6.99903190e-01
8.81526843e-02 5.52255929e-01 1.39898524e-01 -2.41279766e-01
5.57099819e-01 -4.91344362e-01 -6.02566183e-01 4.05472249e-01
1.84613287e-01 -4.07878369e-01 3.31907809e-01 2.20626164e-02
1.23432446e+00 2.27522254e-01 3.62991869e-01 -5.22773147e-01
2.80587524e-01 5.67292154e-01 7.26500213e-01 9.24585700e-01
-3.02283376e-01 5.20382285e-01 7.97101676e-01 -5.24098337e-01
-7.01290250e-01 -1.15538478e+00 -2.22368941e-01 1.37214696e+00
2.67231047e-01 -5.01280129e-01 -1.05701113e+00 -8.79600883e-01
-2.43323892e-01 1.33810747e+00 -2.32084930e-01 3.92878018e-02
-7.87164569e-01 -5.40341914e-01 5.90242684e-01 4.88705575e-01
1.71113983e-02 -1.31281364e+00 -8.38437021e-01 5.32146156e-01
-3.45050901e-01 -1.32267237e+00 1.12932600e-01 5.06799757e-01
-8.37761998e-01 -1.30255735e+00 2.24385232e-01 -8.38393092e-01
4.89543736e-01 -2.28888132e-02 1.68731511e+00 3.20518911e-01
-1.55414060e-01 5.85484326e-01 -4.29223329e-01 -5.80671132e-01
-8.24923933e-01 -7.09697902e-02 -4.26872849e-01 -9.13198113e-01
9.70414937e-01 -3.53451192e-01 -1.34097129e-01 1.50141180e-01
-9.85395133e-01 -6.03959896e-02 1.73012391e-01 7.11075008e-01
4.26945180e-01 -2.15521649e-01 5.72708011e-01 -1.17571127e+00
8.42305779e-01 -4.61365283e-01 -5.03904641e-01 2.97544330e-01
-5.64587474e-01 5.06356895e-01 6.61315203e-01 6.36670217e-02
-1.18945670e+00 -1.80712447e-01 -3.37816000e-01 2.62381345e-01
-6.88186646e-01 3.76860201e-01 -4.80345488e-01 3.74633968e-01
5.83257198e-01 -1.07438885e-01 -5.82041442e-02 -4.22423273e-01
4.46235716e-01 1.55121863e-01 5.17656207e-01 -1.48238349e+00
2.89608270e-01 -1.39235024e-04 -2.57879972e-01 -5.37898719e-01
-1.20289707e+00 -3.03814948e-01 -3.76557112e-01 3.79611433e-01
8.34529221e-01 -8.88475955e-01 6.54765544e-03 1.89081639e-01
-1.48441923e+00 -2.75298536e-01 -4.48704213e-01 1.70398112e-02
-6.55500710e-01 4.90173280e-01 -5.55606127e-01 -5.73884130e-01
-5.72336689e-02 -1.26852632e+00 1.26363862e+00 1.02259412e-01
-7.96753943e-01 -1.04889655e+00 -1.39596954e-01 3.65868300e-01
4.41544801e-01 -1.32645309e-01 1.59599721e+00 -1.14283657e+00
-5.24209976e-01 3.16847652e-01 -1.34481490e-01 1.43570170e-01
1.29754007e-01 -2.81471282e-01 -1.04193902e+00 3.59418750e-01
1.00041687e-01 -4.82623249e-01 5.12305737e-01 1.36722684e-01
8.23322594e-01 -2.15422973e-01 4.20491174e-02 2.01301351e-01
1.42436361e+00 1.18375190e-01 5.14967620e-01 5.00992298e-01
2.70718813e-01 1.04402661e+00 6.03011310e-01 1.32319762e-03
6.97262287e-01 2.86583394e-01 4.03725386e-01 5.94213307e-01
-9.44280028e-02 -3.78091514e-01 3.16130668e-01 4.67105776e-01
4.53647524e-01 -2.33118653e-01 -1.23311210e+00 5.75485229e-01
-1.56941164e+00 -6.00175560e-01 -2.13843603e-02 1.95592642e+00
1.06875396e+00 4.75127369e-01 -3.66714090e-01 -6.23091906e-02
5.80812931e-01 1.40657157e-01 -3.01818430e-01 -8.84642124e-01
-1.50622234e-01 7.13070810e-01 8.65492001e-02 9.41713870e-01
-8.81407440e-01 1.47323537e+00 6.91698647e+00 3.96114707e-01
-7.72514462e-01 -3.04965600e-02 4.82120067e-01 4.14501011e-01
-8.35688353e-01 4.17189449e-01 -1.05848134e+00 2.75497764e-01
1.00432730e+00 -1.93881482e-01 4.28579599e-01 8.84816289e-01
-1.40756845e-01 -1.64132074e-01 -1.44798720e+00 4.24908280e-01
-1.60921618e-01 -1.04453468e+00 3.40890735e-01 -4.78356183e-01
2.26576984e-01 -1.84406370e-01 -4.12243158e-01 6.88441575e-01
7.00519741e-01 -1.23781180e+00 7.76681423e-01 7.72499740e-02
3.90569746e-01 -4.71206903e-01 5.68793654e-01 3.33929151e-01
-8.70389104e-01 -5.56888022e-02 -3.94098550e-01 -6.46305323e-01
2.81577885e-01 3.71706247e-01 -6.66125298e-01 2.71859527e-01
2.63996631e-01 2.81216443e-01 -6.26473904e-01 3.75568122e-01
-7.38681912e-01 5.47975242e-01 -3.23156446e-01 -5.37329726e-02
3.93819720e-01 -2.30036601e-02 4.89788562e-01 1.10690713e+00
2.72904962e-01 2.41010979e-01 3.58092338e-01 1.10322857e+00
2.46049970e-01 -1.32769886e-02 -6.86406434e-01 -8.38275328e-02
9.21398699e-01 8.33775997e-01 -6.69700921e-01 -4.56341982e-01
-4.95085657e-01 7.21502841e-01 4.47031289e-01 2.55539685e-01
-4.08595592e-01 -1.35976234e-02 1.01325858e+00 -1.22888267e-01
-6.65868819e-02 -2.36092001e-01 -7.48477578e-01 -1.08699536e+00
2.97842085e-01 -1.05999041e+00 6.00420177e-01 -9.73816991e-01
-1.47053993e+00 3.99822950e-01 1.65339202e-01 -5.43443978e-01
-6.52032912e-01 -1.07412279e+00 -7.01326966e-01 9.67568517e-01
-1.66305399e+00 -8.52701068e-01 -7.03608543e-02 1.04993386e-02
9.03974831e-01 3.42359722e-01 1.13592279e+00 -2.58288532e-01
-2.95940548e-01 2.70566136e-01 -8.04075241e-01 -1.67155266e-01
5.64796090e-01 -1.73449767e+00 8.35637391e-01 8.82969975e-01
7.68309692e-03 1.05240464e+00 1.03343260e+00 -6.20750546e-01
-1.34742987e+00 -9.05713677e-01 1.36664021e+00 -8.87755275e-01
6.81209385e-01 -2.33542487e-01 -1.18868124e+00 8.52989316e-01
1.64869383e-01 -1.44930422e-01 6.11114383e-01 3.18447351e-01
-6.92132592e-01 5.07859290e-01 -1.21771002e+00 6.03848219e-01
1.47400630e+00 -6.72984719e-01 -1.56383443e+00 4.68590260e-02
1.08139038e+00 -4.17159319e-01 -5.68120539e-01 3.61550182e-01
1.17084160e-01 -9.22993779e-01 8.00951838e-01 -8.68210912e-01
6.14121735e-01 -4.83871512e-02 -4.09135729e-01 -1.37239385e+00
7.75274262e-02 -4.23107564e-01 3.46613497e-01 1.16284084e+00
6.12629116e-01 -7.26027310e-01 6.03939354e-01 1.21307480e+00
-5.94132125e-01 -2.66526937e-01 -1.05193853e+00 -7.42648005e-01
6.04654312e-01 -8.20517957e-01 7.91103721e-01 1.03562796e+00
2.09639907e-01 5.79663932e-01 5.11980832e-01 2.20850766e-01
4.35900450e-01 6.22086413e-02 4.44354862e-01 -1.39723945e+00
-1.65240794e-01 -5.58143914e-01 -1.69341668e-01 -7.80990958e-01
4.78607714e-01 -8.13870549e-01 2.27257982e-01 -1.70788598e+00
-1.30316094e-01 -7.04077125e-01 -8.37763697e-02 6.90418482e-01
-4.20846939e-01 -4.20902014e-01 -6.61156001e-03 -1.77561447e-01
-5.64633548e-01 1.19467191e-01 8.99248004e-01 1.25971884e-01
8.04364383e-02 -5.84153056e-01 -1.15432119e+00 1.04839301e+00
7.79354811e-01 -2.71914750e-01 -4.84137446e-01 -8.83386433e-01
4.11437809e-01 -2.87378859e-02 2.90628821e-01 -9.01814044e-01
-1.14575818e-01 -4.85826224e-01 -2.03505829e-01 -7.28531256e-02
4.54075672e-02 -6.24647260e-01 -1.03570506e-01 1.00807622e-01
-4.96459007e-01 4.15770084e-01 4.96494532e-01 2.06100985e-01
-2.92061269e-01 -7.49482512e-01 3.57578009e-01 -5.67201436e-01
-1.30196667e+00 -3.69167566e-01 -3.55619967e-01 7.28146613e-01
5.69375634e-01 -2.16846943e-01 -7.24262476e-01 1.60331041e-01
-8.43755543e-01 1.94746107e-01 7.85384297e-01 7.18801141e-01
3.23038995e-01 -8.99122536e-01 -2.37482116e-01 3.48257333e-01
4.49573338e-01 3.13430190e-01 -9.40334350e-02 4.12305295e-02
-4.57736909e-01 2.93589860e-01 -8.90439972e-02 -4.44770455e-01
-1.07824707e+00 3.30596685e-01 3.84573519e-01 -1.94646657e-01
-5.62268853e-01 6.29267514e-01 1.47415236e-01 -8.10547709e-01
3.91419902e-02 -8.24523747e-01 -1.26823470e-01 -2.06906483e-01
4.84756887e-01 -2.22958878e-01 1.90696120e-01 -2.64449835e-01
-3.45143557e-01 4.10776854e-01 1.08874775e-01 -8.22720677e-02
1.01878881e+00 -6.39881268e-02 -1.97376519e-01 5.39535642e-01
6.09455943e-01 1.23614743e-01 -9.24649358e-01 -1.45263121e-01
7.39630103e-01 -1.90277338e-01 -4.56982940e-01 -1.10988891e+00
-1.22023366e-01 8.91478479e-01 5.85037470e-02 3.25848252e-01
7.41382062e-01 3.05354387e-01 6.67936802e-01 6.46312237e-01
7.32898295e-01 -1.06429434e+00 -2.41769835e-01 1.20090461e+00
5.30365586e-01 -1.14443421e+00 -4.79367703e-01 -9.16372657e-01
-4.00770277e-01 1.04884434e+00 9.70409691e-01 7.29178563e-02
-5.87964756e-03 3.45589310e-01 2.66178042e-01 -3.44556212e-01
-1.00494707e+00 -3.02023530e-01 -1.81374073e-01 5.91792166e-01
6.95903778e-01 2.07407817e-01 -5.24989247e-01 1.06682456e+00
-5.87888896e-01 -3.40993643e-01 6.78251386e-01 1.47097039e+00
-8.89749825e-01 -1.55365789e+00 -2.72816658e-01 1.76971421e-01
-6.34679615e-01 -4.36285794e-01 -6.71653509e-01 8.03977609e-01
3.28816064e-02 9.85458374e-01 3.22958946e-01 2.54666656e-01
4.01663840e-01 8.87281895e-01 5.82869530e-01 -1.28136206e+00
-5.54203093e-01 -4.76645142e-01 8.21869016e-01 -7.09478498e-01
-5.10984585e-02 -5.54190576e-01 -1.63036633e+00 2.07827881e-01
8.65482688e-02 2.50513613e-01 6.75861955e-01 1.38754797e+00
4.60462421e-01 2.92435437e-01 -9.03081447e-02 -2.33607695e-01
-8.86274338e-01 -7.25975811e-01 -4.53731790e-02 7.80113101e-01
-2.95819733e-02 -6.33030474e-01 -6.01847708e-01 -1.13487504e-01] | [10.376376152038574, 8.9683198928833] |
65e08aa3-4dc2-44a0-9316-355fab785dff | playing-carcassonne-with-monte-carlo-tree | 2009.12974 | null | https://arxiv.org/abs/2009.12974v2 | https://arxiv.org/pdf/2009.12974v2.pdf | Playing Carcassonne with Monte Carlo Tree Search | Monte Carlo Tree Search (MCTS) is a relatively new sampling method with multiple variants in the literature. They can be applied to a wide variety of challenging domains including board games, video games, and energy-based problems to mention a few. In this work, we explore the use of the vanilla MCTS and the MCTS with Rapid Action Value Estimation (MCTS-RAVE) in the game of Carcassonne, a stochastic game with a deceptive scoring system where limited research has been conducted. We compare the strengths of the MCTS-based methods with the Star2.5 algorithm, previously reported to yield competitive results in the game of Carcassonne when a domain-specific heuristic is used to evaluate the game states. We analyse the particularities of the strategies adopted by the algorithms when they share a common reward system. The MCTS-based methods consistently outperformed the Star2.5 algorithm given their ability to find and follow long-term strategies, with the vanilla MCTS exhibiting a more robust game-play than the MCTS-RAVE. | ['Anger Fernando Kuri Morales', 'Fred Valdez Ameneyro', 'Edgar Galvan'] | 2020-09-27 | null | null | null | null | ['board-games'] | ['playing-games'] | [-2.44391784e-02 -3.07638347e-01 -1.85461212e-02 3.20752710e-01
-7.75959313e-01 -6.43338263e-01 5.67063928e-01 -2.42036134e-01
-6.54898942e-01 1.12781036e+00 -1.54554725e-01 -4.24825341e-01
-5.64596713e-01 -7.19724774e-01 -2.18225464e-01 -8.44621480e-01
-2.15687931e-01 7.25349069e-01 8.75521004e-01 -6.80480480e-01
4.66328800e-01 1.62094310e-01 -1.57758737e+00 -1.08393870e-01
7.49084890e-01 9.28840399e-01 1.57394201e-01 9.12081182e-01
4.49369818e-01 1.20791352e+00 -5.66778481e-01 -6.14823043e-01
4.84258294e-01 -8.31179202e-01 -8.13140869e-01 -3.97754490e-01
-4.39178199e-01 -3.85643482e-01 -2.90193379e-01 6.94807053e-01
9.11220670e-01 5.64428270e-01 4.98480916e-01 -1.43582070e+00
4.36758995e-01 3.91495228e-01 -6.72180593e-01 5.08433878e-01
6.65688574e-01 5.55797815e-01 1.03807020e+00 -1.06341064e-01
7.04348326e-01 9.27867472e-01 5.18608153e-01 6.20820582e-01
-1.06977355e+00 -6.73046529e-01 -2.64245152e-01 7.70992637e-01
-1.46696043e+00 -1.51370391e-01 3.86941075e-01 -3.46313864e-02
9.91952896e-01 4.90117103e-01 1.34021389e+00 8.49966824e-01
4.00644839e-01 9.17219043e-01 1.54394209e+00 -2.80424029e-01
1.06350315e+00 -4.07892853e-01 -6.94823802e-01 5.52055538e-01
1.69823557e-01 7.81257689e-01 -7.47362852e-01 -3.88778239e-01
3.94257903e-01 -6.62191451e-01 1.85673267e-01 -6.58228636e-01
-5.59399664e-01 1.16520083e+00 8.51140469e-02 6.90138713e-02
-6.89337611e-01 6.84721231e-01 4.49759156e-01 1.14232615e-01
3.62886190e-01 6.87710345e-01 -1.52016088e-01 -1.03375471e+00
-1.58212411e+00 1.08195901e+00 7.84760237e-01 3.73563081e-01
1.17907509e-01 2.31647089e-01 -4.96022671e-01 3.03467363e-01
2.64959514e-01 -1.29907489e-01 3.31818372e-01 -1.25436914e+00
1.82888612e-01 -6.17189473e-03 3.25180680e-01 -3.57482702e-01
-3.87054175e-01 -2.93868333e-01 -2.18319312e-01 8.60882640e-01
5.90148628e-01 -2.25004107e-01 -5.51181853e-01 1.56166971e+00
4.59665269e-01 3.51597488e-01 -4.35838522e-03 7.18881130e-01
2.80425489e-01 4.44034010e-01 -8.80246609e-02 -3.19002479e-01
9.06695843e-01 -1.00350761e+00 -4.97176349e-01 -2.02173561e-01
5.37900746e-01 -5.23942649e-01 6.88836932e-01 8.75416636e-01
-1.55761552e+00 6.13234714e-02 -1.14764309e+00 6.90261126e-01
-7.78698102e-02 -6.66093230e-01 7.07184553e-01 1.28689313e+00
-1.25600135e+00 8.74887109e-01 -8.12608004e-01 -2.00916067e-01
4.90841806e-01 3.79142106e-01 3.18414897e-01 -1.91822141e-01
-1.23784566e+00 1.28581989e+00 3.82626534e-01 -2.71931797e-01
-1.43849945e+00 -2.26438820e-01 -4.78569686e-01 -1.66834495e-03
1.07462263e+00 -4.05877352e-01 1.63941717e+00 -5.24987638e-01
-2.03210783e+00 5.01080573e-01 3.84376317e-01 -7.89128244e-01
1.10423434e+00 2.56424308e-01 1.08402893e-02 3.83180417e-02
1.01469181e-01 4.37935114e-01 4.35898423e-01 -6.66231453e-01
-6.47834599e-01 9.50285122e-02 3.88572901e-01 7.16329575e-01
5.26029229e-01 1.89248309e-01 -2.81017870e-02 -4.10846353e-01
-3.71238858e-01 -1.01463258e+00 -7.37422705e-01 -4.19012487e-01
-2.69759268e-01 -1.35913864e-01 1.41678631e-01 -1.12169690e-01
1.42457390e+00 -1.64105165e+00 3.98845106e-01 1.85200900e-01
1.47268698e-02 1.52758032e-01 1.40933350e-01 8.53557587e-01
3.64938915e-01 -1.47055471e-02 -8.10736269e-02 8.21157638e-03
1.84628382e-01 4.07295108e-01 4.04098928e-01 4.50976670e-01
-3.05070370e-01 8.11662793e-01 -1.21722412e+00 -4.72479314e-01
1.84978276e-01 -2.80144542e-01 -7.24584281e-01 8.13432112e-02
-3.06043625e-01 6.55849054e-02 -5.58028162e-01 4.99865294e-01
3.09788555e-01 3.05612296e-01 2.19915166e-01 6.11079395e-01
-2.19554201e-01 2.00661853e-01 -1.37401259e+00 1.43542075e+00
-1.46440059e-01 3.74610752e-01 1.23096846e-01 -8.20613921e-01
4.25446659e-01 2.44081199e-01 5.44437289e-01 -6.83545232e-01
2.11770535e-01 3.78561199e-01 3.57128292e-01 -1.84591651e-01
8.81841660e-01 -5.03109574e-01 -3.45299840e-01 4.90084738e-01
-8.58259797e-02 -7.32067883e-01 6.83978498e-01 3.57896805e-01
1.67349982e+00 4.92323160e-01 7.36286342e-01 -2.29234949e-01
1.72377393e-01 3.26537013e-01 3.11205238e-01 1.20608783e+00
-6.17108345e-01 3.72286737e-01 6.85578346e-01 -1.73699751e-01
-8.30465913e-01 -1.05171275e+00 3.21640074e-01 1.16130388e+00
3.77643615e-01 -5.89748859e-01 -7.69645810e-01 -4.29719895e-01
-2.62733340e-01 1.00245619e+00 -6.36443377e-01 -3.52180630e-01
-1.50401279e-01 -9.05012488e-01 5.29251456e-01 1.52125463e-01
5.52963138e-01 -1.11344314e+00 -1.41943908e+00 6.02331698e-01
-9.34145674e-02 -7.41125405e-01 -2.08946019e-01 3.94572675e-01
-5.72869897e-01 -1.21369183e+00 -5.36812365e-01 1.86024606e-01
-1.57518148e-01 -1.21584572e-01 1.26153243e+00 3.74965593e-02
-2.64220208e-01 3.83128583e-01 -6.84144855e-01 -3.52864653e-01
-5.91461062e-01 -1.26492968e-02 -7.29392841e-02 -5.48193991e-01
1.09872296e-01 -4.63581800e-01 -6.03451610e-01 4.51634198e-01
-7.78305829e-01 1.29668429e-01 3.07708323e-01 9.86340463e-01
2.53431916e-01 3.92723709e-01 2.40636319e-01 -5.67947388e-01
9.58466530e-01 -6.98652685e-01 -7.01175213e-01 -1.77832708e-01
-6.64880037e-01 4.10626829e-02 2.97969699e-01 -3.65936607e-01
-7.57928312e-01 -8.51298645e-02 -3.13736588e-01 -2.16550708e-01
3.40013415e-01 3.99714082e-01 1.81483999e-01 -3.58838350e-01
6.69602394e-01 3.25497538e-01 -2.47454420e-01 -1.34066030e-01
2.19731748e-01 1.45909339e-01 2.36230090e-01 -6.48974121e-01
6.60313129e-01 1.57636374e-01 2.45571792e-01 -1.36485830e-01
-3.96035403e-01 -2.46660784e-01 -1.20596692e-03 -8.44375968e-01
7.47781217e-01 -4.51261699e-01 -9.43744123e-01 7.21716046e-01
-5.88805020e-01 -6.52169287e-01 -7.01269150e-01 3.89751345e-01
-1.16901648e+00 2.24448353e-01 -5.24819493e-01 -1.31053352e+00
8.96038339e-02 -1.27016890e+00 7.23751247e-01 3.75925899e-01
-1.70501396e-01 -5.85842192e-01 4.67919260e-01 4.08139586e-01
4.00099814e-01 6.47417188e-01 2.76105434e-01 -4.70292389e-01
-4.44453895e-01 5.02059087e-02 4.85628933e-01 -2.27792591e-01
-4.44848388e-01 -5.01772040e-04 -3.94855231e-01 -5.05603790e-01
-9.79897827e-02 -4.91675675e-01 5.27722597e-01 5.70964098e-01
8.06614816e-01 -5.07708341e-02 -5.39699346e-02 2.41772562e-01
1.63656747e+00 5.85106671e-01 8.23593795e-01 9.40085411e-01
-1.05661862e-01 2.28524402e-01 9.44993675e-01 8.24058890e-01
2.66603231e-01 1.16035044e+00 1.02816367e+00 2.80238956e-01
2.68793464e-01 -1.95399687e-01 5.53597987e-01 4.20844287e-01
-4.85381514e-01 -4.05718267e-01 -6.17250025e-01 4.93190825e-01
-1.97246575e+00 -1.26854861e+00 -3.43161412e-02 2.22901058e+00
7.31894433e-01 5.70234835e-01 7.00352788e-01 4.12283450e-01
6.01423144e-01 3.10839474e-01 -6.32047176e-01 -9.84063685e-01
1.81132451e-01 7.00258136e-01 7.72579610e-01 2.65998721e-01
-5.25979996e-01 1.00237632e+00 7.08519077e+00 1.63340998e+00
-2.55817592e-01 3.58051181e-01 5.33383608e-01 -8.48236322e-01
8.33972394e-02 1.44217983e-01 -2.58648604e-01 8.07800531e-01
1.15129578e+00 -4.77164626e-01 9.69723940e-01 7.39403129e-01
5.04662216e-01 -1.01894534e+00 -7.27908969e-01 5.71732640e-01
-3.34348559e-01 -1.26376724e+00 -6.42719626e-01 3.25651765e-01
8.85888338e-01 8.08739513e-02 -1.24059722e-01 5.26370168e-01
1.21294677e+00 -1.15368927e+00 1.21022081e+00 2.22321898e-01
6.14144504e-01 -1.08173156e+00 8.25659096e-01 4.37281847e-01
-1.19519210e+00 -1.19337544e-01 -1.36406779e-01 -5.69580019e-01
3.92772377e-01 1.04670435e-01 -4.16030288e-01 7.37792492e-01
8.17248940e-01 1.93490286e-03 -2.40294784e-01 1.50798464e+00
-3.35011929e-01 8.84555638e-01 -4.84791785e-01 -6.27633035e-01
9.74072874e-01 -3.28155518e-01 8.55654061e-01 5.58945239e-01
3.65454346e-01 9.66051668e-02 1.43251076e-01 6.55744731e-01
4.61072922e-01 -7.55536780e-02 -1.96644306e-01 1.98407218e-01
5.63770771e-01 9.64228511e-01 -9.17559505e-01 -1.39597952e-01
1.84803233e-01 8.57750893e-01 5.85034601e-02 -1.42594278e-01
-1.03932369e+00 -9.10931081e-02 5.63477039e-01 2.92927865e-02
4.97818321e-01 7.95764732e-04 -2.00550989e-04 -6.35527730e-01
-3.99831593e-01 -1.06195450e+00 5.06620526e-01 -1.13065696e+00
-7.89635777e-01 4.48371917e-01 5.05418897e-01 -1.10951293e+00
-8.17869842e-01 -2.60737062e-01 -9.12983537e-01 6.00401998e-01
-1.19241464e+00 -2.88205266e-01 4.26842943e-02 5.97364247e-01
5.97140849e-01 -2.42161393e-01 4.51966077e-01 -1.52750000e-01
-4.30345923e-01 3.05033565e-01 6.07341170e-01 -6.90386057e-01
-2.05355883e-02 -1.33405578e+00 4.71516788e-01 8.65696967e-01
-6.38991073e-02 -3.42958212e-01 1.22223842e+00 -5.08359015e-01
-1.18989623e+00 -4.43986505e-01 8.15065950e-02 -9.23472568e-02
6.57730341e-01 -2.27471307e-01 -1.29661396e-01 1.90030772e-03
2.30486095e-01 -2.32043251e-01 3.41929823e-01 -2.63570487e-01
3.43594283e-01 3.85614425e-01 -1.60153580e+00 8.13501060e-01
1.15767300e+00 -9.47044864e-02 -2.48283088e-01 1.28472626e-01
4.28479770e-03 -7.80417264e-01 -3.41196686e-01 -8.78439173e-02
5.02694905e-01 -1.68155229e+00 7.10764468e-01 -4.22132373e-01
3.77924085e-01 -2.39902213e-01 -2.73884255e-02 -1.61468732e+00
-3.28829259e-01 -1.06265819e+00 -1.73291415e-01 6.89157009e-01
1.01257376e-01 -5.16735077e-01 1.28226256e+00 1.55604005e-01
3.31812680e-01 -1.13959146e+00 -1.74928927e+00 -1.13035536e+00
1.86193109e-01 -5.31659245e-01 6.15520656e-01 2.72844672e-01
4.21989895e-02 -1.07525796e-01 -7.05846071e-01 -5.01585424e-01
6.75827086e-01 -3.12490314e-01 5.30961990e-01 -9.70930338e-01
-8.25273931e-01 -6.63541555e-01 -5.37961066e-01 -6.71465635e-01
-3.58523786e-01 -4.86485064e-01 1.86229184e-01 -1.46096802e+00
3.33387643e-01 -2.66005456e-01 -1.89504966e-01 1.04405649e-01
-2.74055302e-01 3.43067944e-01 4.36151385e-01 -1.10377461e-01
-9.35905814e-01 5.84849179e-01 9.68746305e-01 1.94527790e-01
-7.63610229e-02 2.37527177e-01 -5.05719066e-01 4.83687997e-01
8.47249329e-01 -8.74933243e-01 -4.11713421e-01 4.35986906e-01
6.14302874e-01 3.42964798e-01 9.24684331e-02 -1.26164472e+00
2.94341266e-01 -3.45060378e-01 -2.45964557e-01 -4.26432848e-01
5.23197889e-01 -4.85898226e-01 6.40713930e-01 1.04736137e+00
-8.55101123e-02 1.97627589e-01 1.45848364e-01 7.48999059e-01
1.81281447e-01 -8.50322902e-01 8.60362172e-01 -4.12816495e-01
-7.68122256e-01 1.48629442e-01 -1.23580849e+00 4.47049975e-01
1.55281377e+00 -8.80129874e-01 1.69565901e-01 -7.51403809e-01
-4.78128970e-01 3.47340137e-01 5.31166911e-01 -7.37580955e-02
2.96705633e-01 -1.26644897e+00 -7.48328090e-01 -4.91893947e-01
-2.40884751e-01 -4.25482750e-01 2.92922080e-01 6.92111075e-01
-7.43611038e-01 5.77493571e-03 -4.52057034e-01 -1.51282847e-01
-1.30663407e+00 3.31544667e-01 7.37575054e-01 -1.17268395e+00
-2.12157890e-01 7.37545490e-01 -3.48007202e-01 -2.18263775e-01
-9.27013084e-02 1.90310374e-01 3.54324691e-02 -1.43640682e-01
1.33149266e-01 9.71575320e-01 -4.30285074e-02 -4.36208725e-01
-4.80538785e-01 2.26151180e-02 3.85599554e-01 -6.56674325e-01
1.39594591e+00 2.37762481e-02 3.77273142e-01 9.17300284e-02
3.23720545e-01 -1.48223460e-01 -1.45124292e+00 1.85543716e-01
-6.46197200e-02 -6.30096972e-01 4.67574745e-01 -1.09190488e+00
-9.87569749e-01 3.44398916e-01 4.66500700e-01 4.42288041e-01
1.16491389e+00 -3.76166284e-01 5.41411221e-01 -1.92680776e-01
1.06675959e+00 -1.36833012e+00 -3.67816053e-02 4.08582985e-01
2.96520203e-01 -6.49183273e-01 3.36180806e-01 6.45170361e-02
-8.83286417e-01 7.54934013e-01 3.94975573e-01 -1.96324185e-01
1.11259118e-01 2.46429414e-01 -5.53829849e-01 -3.37951809e-01
-1.09639323e+00 -6.76511586e-01 -2.91082829e-01 6.85636222e-01
-3.49139035e-01 1.06093258e-01 -8.07506084e-01 4.36759502e-01
-3.61534685e-01 1.83668777e-01 9.87921298e-01 1.22270977e+00
-3.40452850e-01 -1.20836186e+00 -3.38871777e-01 5.80781102e-01
-5.53560376e-01 -2.62057930e-01 -4.30662036e-01 9.71044242e-01
2.57986337e-02 1.26691842e+00 -3.03067863e-01 -3.92529219e-01
2.47052640e-01 -2.04048902e-01 7.46485531e-01 -3.33706260e-01
-1.20989692e+00 -3.37698102e-01 4.47847277e-01 -9.30723190e-01
-3.68164957e-01 -1.08855736e+00 -7.37468541e-01 -8.51701975e-01
-6.49994254e-01 5.11247694e-01 5.52371442e-01 1.03632641e+00
-5.40720187e-02 6.49078310e-01 6.52839005e-01 -1.07851994e+00
-6.92141593e-01 -5.36765099e-01 -1.06438744e+00 -1.88791350e-01
-3.41043234e-01 -1.13304889e+00 -2.64757603e-01 -9.17569935e-01] | [3.523635149002075, 1.5250855684280396] |
28f13b68-c5c6-4f2d-97fc-d2d43f0b6143 | curaj-iiitdwd-lt-edi-acl-2022-hope-speech | null | null | https://aclanthology.org/2022.ltedi-1.25 | https://aclanthology.org/2022.ltedi-1.25.pdf | CURAJ_IIITDWD@LT-EDI-ACL 2022: Hope Speech Detection in English YouTube Comments using Deep Learning Techniques | Hope Speech are positive terms that help to promote or criticise a point of view without hurting the user’s or community’s feelings. Non-Hope Speech, on the other side, includes expressions that are harsh, ridiculing, or demotivating. The goal of this article is to find the hope speech comments in a YouTube dataset. The datasets were created as part of the “LT-EDI-ACL 2022: Hope Speech Detection for Equality, Diversity, and Inclusion” shared task. The shared task dataset was proposed in Malayalam, Tamil, English, Spanish, and Kannada languages. In this paper, we worked at English-language YouTube comments. We employed several deep learning based models such as DNN (dense or fully connected neural network), CNN (Convolutional Neural Network), Bi-LSTM (Bidirectional Long Short Term Memory Network), and GRU(Gated Recurrent Unit) to identify the hopeful comments. We also used Stacked LSTM-CNN and Stacked LSTM-LSTM network to train the model. The best macro average F1-score 0.67 for development dataset was obtained using the DNN model. The macro average F1-score of 0.67 was achieved for the classification done on the test data as well. | ['Sunil Saumya', 'Ankit Mishra', 'Vanshita Jha'] | null | null | null | null | ltedi-acl-2022-5 | ['hope-speech-detection'] | ['natural-language-processing'] | [-5.24592340e-01 3.11301589e-01 -2.07942843e-01 -2.04544038e-01
-8.84942412e-01 -1.14839636e-01 6.03984058e-01 7.17392936e-02
-1.37070313e-01 6.81034505e-01 1.18224823e+00 -3.46982270e-01
3.30373913e-01 -3.68693173e-01 -1.36189371e-01 -4.64151949e-01
2.08552465e-01 -1.86245933e-01 -5.06848633e-01 -6.27231598e-01
5.03430128e-01 -4.34236936e-02 -1.00705135e+00 5.48005939e-01
5.48661947e-01 6.82066619e-01 -3.11904281e-01 8.66735756e-01
-1.02390707e-01 1.70770657e+00 -5.83603323e-01 -6.12092495e-01
-4.71040696e-01 -5.26237071e-01 -1.07510102e+00 -3.72647375e-01
8.05953145e-03 -3.85290056e-01 -2.95355499e-01 8.95685434e-01
8.74358416e-01 2.43139908e-01 2.64574140e-01 -9.88427758e-01
-1.23477888e+00 1.02661359e+00 -2.03369766e-01 1.49304554e-01
4.51568604e-01 -7.65593164e-03 8.95253837e-01 -1.07099783e+00
6.10169590e-01 1.40372944e+00 7.23652780e-01 7.33136773e-01
-3.98407549e-01 -7.14440286e-01 -4.31107074e-01 2.52217144e-01
-9.74305570e-01 -6.05439603e-01 9.78297412e-01 -3.52396965e-01
1.39223278e+00 3.17346603e-01 8.06218147e-01 1.81850803e+00
4.01672602e-01 8.96013975e-01 8.50606441e-01 -6.76869079e-02
2.23686900e-02 3.15476149e-01 1.52857840e-01 4.08732533e-01
-6.66752458e-01 -2.08176807e-01 -8.22097003e-01 -2.10595861e-01
1.79562390e-01 1.57392696e-02 -1.08002223e-01 5.67263365e-01
-8.86700451e-01 1.30614781e+00 5.27293324e-01 6.69827402e-01
-4.48987484e-01 -1.56525999e-01 9.83175874e-01 5.57233393e-01
1.00670016e+00 2.22505465e-01 -1.60650119e-01 -8.78883421e-01
-8.25658798e-01 -1.00443706e-01 8.14454317e-01 3.18350613e-01
1.19201103e-02 4.64792401e-01 -2.22282663e-01 1.23983002e+00
4.30537105e-01 2.91992813e-01 1.05758095e+00 -4.23209876e-01
2.76054621e-01 5.29653013e-01 -2.29598448e-01 -1.19408762e+00
-2.91939437e-01 -4.30545896e-01 -1.08102596e+00 -5.45640811e-02
-2.34479338e-01 -4.68800813e-01 -6.97865784e-01 1.67754030e+00
-2.09543556e-01 -1.69172391e-01 6.50452137e-01 6.70224488e-01
1.67472708e+00 1.10648060e+00 -1.36415418e-02 -3.86433512e-01
8.78487408e-01 -1.30464876e+00 -1.11576307e+00 1.18740639e-02
1.12520218e+00 -1.15735149e+00 1.13993371e+00 3.72471631e-01
-1.23226821e+00 -1.24051228e-01 -9.25627530e-01 -5.22122502e-01
-5.08336663e-01 -9.84591097e-02 2.14866191e-01 6.88223124e-01
-1.22414505e+00 5.25459588e-01 -2.30151251e-01 -6.93165243e-01
2.97428608e-01 5.85715845e-03 -3.17741036e-01 3.95994604e-01
-1.46088111e+00 9.94003534e-01 -3.94924581e-02 1.56267032e-01
-9.51590717e-01 -2.04929605e-01 -7.98145711e-01 3.02726198e-02
-3.06041658e-01 -6.74865991e-02 1.09169710e+00 -1.34972656e+00
-1.55908322e+00 9.73617196e-01 -1.59258768e-02 -5.96694171e-01
1.73244789e-01 -2.81171024e-01 -6.07468069e-01 -3.17053020e-01
-1.08140662e-01 6.82244778e-01 6.19173884e-01 -5.88057458e-01
1.10250033e-01 -2.46254146e-01 -1.18401490e-01 1.18159153e-01
-8.71238053e-01 5.38815200e-01 5.85100889e-01 -6.02040052e-01
-2.63783038e-01 -6.52993262e-01 1.91281319e-01 -6.05985284e-01
-5.88038862e-01 -5.87986529e-01 1.32487810e+00 -1.17322719e+00
1.38582110e+00 -2.28982878e+00 -1.72081709e-01 -4.92445350e-01
2.31158182e-01 4.41703916e-01 -1.83924422e-01 7.70019054e-01
-2.71963656e-01 6.20028317e-01 3.10754448e-01 -3.43581527e-01
-2.47985378e-01 1.79529831e-01 -2.04531386e-01 3.61792326e-01
1.18111446e-02 7.83930600e-01 -7.55468190e-01 -2.05119535e-01
-2.02053357e-02 8.62401605e-01 -4.44646239e-01 2.78231680e-01
3.66256475e-01 9.90163237e-02 -6.28466979e-02 5.71102500e-01
3.65828037e-01 -9.48381722e-02 -1.97029203e-01 2.01625854e-01
-3.52951109e-01 6.67503178e-01 -3.55882764e-01 1.23034501e+00
-5.56146383e-01 1.04557574e+00 -1.83647618e-01 -6.96558297e-01
1.42677712e+00 8.88901234e-01 5.33991642e-02 -8.56867135e-01
4.28397447e-01 2.41483241e-01 -3.88295688e-02 -8.53546321e-01
7.59440601e-01 -3.84960562e-01 -4.44655120e-02 4.80247200e-01
-6.85517036e-04 2.01674700e-02 -5.01224637e-01 3.29398990e-01
8.03867161e-01 -6.22468054e-01 4.14006233e-01 -1.82173535e-01
4.88347918e-01 -5.71870863e-01 4.84460354e-01 2.18569666e-01
-5.82693934e-01 6.36589587e-01 8.38140368e-01 -5.87139368e-01
-1.12349403e+00 -5.12583017e-01 2.16457531e-01 1.19475698e+00
-6.36581242e-01 -4.58974689e-01 -5.14700055e-01 -2.65664399e-01
-7.30523765e-01 1.10108495e+00 -5.71787238e-01 -3.64166975e-01
-1.03283860e-01 -2.66611934e-01 7.43820190e-01 1.49501547e-01
6.46717250e-01 -1.43300951e+00 -4.32437181e-01 6.20144233e-02
-5.14104366e-01 -8.43761623e-01 -3.14525306e-01 1.13361232e-01
-5.88743210e-01 -7.61963487e-01 -8.92087519e-01 -7.58301198e-01
-2.65414771e-02 -4.42570746e-02 7.73977995e-01 2.05331272e-03
3.12074155e-01 -9.32170302e-02 -5.73556900e-01 -3.59985858e-01
-3.61456066e-01 -4.53970283e-02 -1.14699930e-01 -2.52501309e-01
3.55572999e-01 -4.92215425e-01 -9.86995250e-02 -2.02268526e-01
-4.29662019e-01 1.63018420e-01 -1.18388714e-04 1.10084569e+00
-2.43127838e-01 -3.29704672e-01 7.83841729e-01 -5.23324430e-01
1.30801439e+00 -8.34897101e-01 4.28099722e-01 -5.20725474e-02
-3.53817880e-01 -3.67484987e-01 4.04890805e-01 -3.56439680e-01
-1.01664770e+00 -6.22839451e-01 -7.32618511e-01 -3.74852121e-01
2.01855332e-01 7.55063176e-01 1.80323228e-01 2.32409388e-01
5.52577972e-01 1.44287631e-01 -1.14253670e-01 -4.06484008e-01
1.05937742e-01 1.35096633e+00 2.67151028e-01 2.07262542e-02
-1.55982599e-01 -7.53360689e-02 -6.81309104e-01 -1.10227656e+00
-8.43378425e-01 -2.74166167e-01 -7.33902454e-02 -4.62036967e-01
9.52783942e-01 -1.00094700e+00 -6.99625254e-01 7.33236194e-01
-1.49545896e+00 -2.66961515e-01 7.28640892e-03 3.09082538e-01
-1.68505237e-01 1.16557166e-01 -1.05436456e+00 -1.28877926e+00
-1.06859612e+00 -1.02343559e+00 4.97819602e-01 3.55105907e-01
-6.07970595e-01 -1.23591447e+00 1.56548411e-01 7.82538056e-01
7.26672530e-01 4.10218686e-01 6.89918935e-01 -1.25438237e+00
5.59168637e-01 -1.87201709e-01 -1.57499462e-01 6.96773767e-01
-1.04151852e-01 2.87570119e-01 -1.16135120e+00 8.51783156e-02
4.82365429e-01 -1.06104481e+00 7.18564808e-01 3.37779105e-01
7.45698512e-01 -1.05524707e+00 5.43312192e-01 6.39889315e-02
9.85230863e-01 1.78327262e-01 9.67502594e-01 2.80662298e-01
5.51308513e-01 4.34396029e-01 1.48022503e-01 9.09216881e-01
5.60198128e-01 1.50976196e-01 5.93922019e-01 1.88541654e-02
8.63784701e-02 -4.07756329e-01 9.00070548e-01 1.26949799e+00
1.74877912e-01 -3.44521970e-01 -1.07403684e+00 8.08931530e-01
-1.69843102e+00 -1.17474985e+00 -6.51737571e-01 1.83257604e+00
7.05413818e-01 1.84071019e-01 1.76249608e-01 2.08567977e-01
5.54319263e-01 5.51627636e-01 -1.13947459e-01 -1.66435409e+00
-2.87482530e-01 -2.67194696e-02 -1.36427566e-01 6.82394624e-01
-8.17269683e-01 9.31368470e-01 5.54965067e+00 8.04553747e-01
-1.67257750e+00 5.78676701e-01 1.27575219e+00 -4.05422539e-01
-4.29896206e-01 -3.30392927e-01 -3.81249577e-01 3.14398378e-01
1.60039115e+00 -3.66981715e-01 1.34800002e-01 1.04576564e+00
6.30577683e-01 3.45207192e-02 -5.21420717e-01 1.02213490e+00
3.45531970e-01 -1.08912969e+00 -2.03485280e-01 -1.84745133e-01
6.78664148e-01 4.17138129e-01 4.08004284e-01 6.06792152e-01
1.46197587e-01 -1.34623349e+00 5.95218658e-01 2.36406639e-01
5.71105838e-01 -1.00016654e+00 1.13975191e+00 5.29728234e-01
-4.65562582e-01 -1.22550219e-01 -2.23333865e-01 -7.89881825e-01
1.61858588e-01 5.53183854e-01 -8.55625987e-01 -2.10128332e-04
6.43872559e-01 1.22009432e+00 -2.61254817e-01 4.61204171e-01
-2.82007515e-01 8.16929042e-01 -1.88317634e-02 -7.47637093e-01
8.50964010e-01 3.46398652e-01 5.57323396e-01 1.40496778e+00
4.34013009e-01 -9.31501761e-02 -1.96649224e-01 6.03411913e-01
-3.10618579e-01 4.20860887e-01 -9.76529777e-01 -6.11324668e-01
1.28457649e-02 1.41515672e+00 -1.35947540e-01 -3.17865580e-01
-9.29239169e-02 7.23201096e-01 2.23375618e-01 1.58479288e-01
-6.67615891e-01 -1.77248850e-01 2.67478228e-01 -8.26009661e-02
-2.35122502e-01 1.10331692e-01 -4.50146675e-01 -1.05313706e+00
-4.08224881e-01 -8.03486764e-01 2.92086273e-01 -1.06782997e+00
-1.25424910e+00 7.38741994e-01 -4.61377144e-01 -7.22665429e-01
-3.04389387e-01 -1.74340725e-01 -1.16750944e+00 7.25045741e-01
-1.08611262e+00 -1.22766805e+00 2.42794827e-02 5.99373400e-01
1.00278640e+00 -2.95370162e-01 8.29005778e-01 2.70920783e-01
-6.85374558e-01 4.71363425e-01 -1.06522739e-01 1.77197799e-01
6.22376204e-01 -7.01865852e-01 -4.72767744e-03 6.39195442e-01
-3.43440503e-01 5.65722823e-01 7.99473882e-01 -4.66605783e-01
-9.48995411e-01 -8.37810636e-01 1.86474526e+00 6.27651662e-02
9.06108201e-01 -1.08928755e-01 -7.07906604e-01 5.99745691e-01
9.68529582e-01 -3.96837980e-01 9.77970302e-01 3.98287833e-01
-3.39762866e-01 4.28385466e-01 -1.29298246e+00 4.92200851e-01
4.13867533e-01 -7.77534246e-01 -5.07576823e-01 6.36179686e-01
7.78799891e-01 -2.11387679e-01 -7.27687180e-01 -1.46506438e-02
4.87536222e-01 -1.32198513e+00 4.87200856e-01 -8.30774665e-01
1.28826547e+00 7.93912530e-01 -4.45201606e-01 -1.22887218e+00
-1.97106808e-01 -7.35959709e-01 4.67384197e-02 1.55090833e+00
4.25708443e-01 -4.12020564e-01 4.67748910e-01 6.86160743e-01
-2.52721578e-01 -1.10406923e+00 -1.13954842e+00 -2.05462113e-01
2.95829386e-01 -4.05748367e-01 -4.97991173e-03 1.28425288e+00
6.54847085e-01 1.00275707e+00 -9.81363297e-01 -6.69633448e-01
-1.74956135e-02 -6.28454208e-01 4.56937402e-01 -7.55998373e-01
3.52318525e-01 -6.05645180e-01 -2.41292000e-01 -4.43228930e-01
2.52353311e-01 -7.91490912e-01 -3.97337615e-01 -1.55036449e+00
4.75082308e-01 1.71920627e-01 -2.15707749e-01 7.96541870e-01
3.36610943e-01 -8.25836211e-02 3.05574566e-01 -4.82105576e-02
-4.67701495e-01 8.28288138e-01 1.09402323e+00 -2.17261553e-01
-1.66558102e-01 -1.46697968e-01 -9.91075099e-01 7.25945950e-01
1.09261572e+00 -3.55287433e-01 -3.01816136e-01 -2.22096473e-01
6.41690195e-01 3.46457511e-01 3.36958796e-01 -6.99921787e-01
1.35122135e-01 7.52058253e-02 1.80947427e-02 -7.27047086e-01
6.02225721e-01 -2.00295150e-01 -1.41756386e-01 6.80451989e-01
-7.98319280e-01 2.49284789e-01 4.10882421e-02 -1.34756818e-01
-4.43395257e-01 -2.84931660e-01 8.53805065e-01 -1.97075963e-01
-2.85582185e-01 -1.34798914e-01 -6.07961535e-01 9.86559838e-02
7.22880065e-01 4.91912067e-02 -6.36798084e-01 -1.08631790e+00
-5.87313890e-01 1.15117773e-01 7.15474859e-02 6.83756649e-01
1.09868610e+00 -1.45189559e+00 -9.45899308e-01 -1.11549452e-01
-1.71751633e-01 -6.19780958e-01 4.75247413e-01 1.04994988e+00
-1.87848598e-01 3.95366520e-01 -2.57219404e-01 -1.18550733e-01
-1.64734042e+00 2.49373868e-01 3.24608356e-01 -2.66495466e-01
-4.96061563e-01 1.23497558e+00 -3.42394710e-01 -5.23675084e-01
1.82318300e-01 -6.84754699e-02 -6.44241095e-01 3.05251271e-01
6.43749356e-01 5.78473747e-01 -1.58991277e-01 -1.00146759e+00
-2.85469055e-01 -8.69805142e-02 -9.85575691e-02 -8.90894160e-02
1.66562235e+00 -2.32087653e-02 -3.50558758e-01 8.72793674e-01
1.77893841e+00 -2.70542979e-01 -1.47584125e-01 2.91997373e-01
-2.39277408e-01 -1.60613820e-01 3.97732198e-01 -8.52393448e-01
-1.02759254e+00 1.30432224e+00 5.26211381e-01 4.74789590e-01
7.98708797e-01 -2.70629644e-01 1.22212505e+00 2.29556844e-01
-5.08918345e-01 -1.25674582e+00 4.45824236e-01 1.00006604e+00
1.27771735e+00 -1.27817738e+00 -4.16475922e-01 4.05022919e-01
-1.12448418e+00 1.42310011e+00 4.83488560e-01 5.72832786e-02
6.40138686e-01 2.13058013e-02 2.05388352e-01 -2.98337430e-01
-1.29368973e+00 4.74350482e-01 -1.13466457e-01 2.82935560e-01
1.15940607e+00 1.15609415e-01 -5.96347213e-01 7.17783689e-01
-4.60272551e-01 1.89322248e-01 1.00134659e+00 4.62985605e-01
-6.16500199e-01 -4.29041892e-01 -9.35680866e-02 2.64244646e-01
-8.94210160e-01 -4.12635058e-01 -8.07410121e-01 2.33860537e-01
-1.31310388e-01 1.35990489e+00 -1.21766314e-01 -8.28493416e-01
-1.48911521e-01 1.73812136e-01 -4.18372393e-01 -3.01712424e-01
-1.21423316e+00 1.50051832e-01 6.29942119e-01 -2.58498847e-01
-3.63784224e-01 -4.35756058e-01 -1.09819353e+00 -8.64095867e-01
-1.35710552e-01 1.63543209e-01 7.31405258e-01 9.26633656e-01
2.48397544e-01 3.80222112e-01 6.25800192e-01 -4.55655098e-01
-3.48199487e-01 -1.37542486e+00 -4.64709669e-01 4.49890196e-02
4.02668357e-01 -8.83322954e-02 -4.28931922e-01 -3.00120205e-01] | [9.009891510009766, 10.71493148803711] |
18e7a6ad-5b4a-4acf-8b0c-864d5ba51336 | mapformer-boosting-change-detection-by-using | 2303.17859 | null | https://arxiv.org/abs/2303.17859v1 | https://arxiv.org/pdf/2303.17859v1.pdf | MapFormer: Boosting Change Detection by Using Pre-change Information | Change detection in remote sensing imagery is essential for a variety of applications such as urban planning, disaster management, and climate research. However, existing methods for identifying semantically changed areas overlook the availability of semantic information in the form of existing maps describing features of the earth's surface. In this paper, we leverage this information for change detection in bi-temporal images. We show that the simple integration of the additional information via concatenation of latent representations suffices to significantly outperform state-of-the-art change detection methods. Motivated by this observation, we propose the new task of Conditional Change Detection, where pre-change semantic information is used as input next to bi-temporal images. To fully exploit the extra information, we propose MapFormer, a novel architecture based on a multi-modal feature fusion module that allows for feature processing conditioned on the available semantic information. We further employ a supervised, cross-modal contrastive loss to guide the learning of visual representations. Our approach outperforms existing change detection methods by an absolute 11.7% and 18.4% in terms of binary change IoU on DynamicEarthNet and HRSCD, respectively. Furthermore, we demonstrate the robustness of our approach to the quality of the pre-change semantic information and the absence pre-change imagery. The code will be made publicly available. | ['Matthias Schubert', 'Niklas Strauß', 'Maximilian Bernhard'] | 2023-03-31 | null | null | null | null | ['change-detection'] | ['computer-vision'] | [ 5.59490919e-01 -3.96118969e-01 1.58070236e-01 -4.83678401e-01
-7.58487582e-01 -6.46782219e-01 1.03984952e+00 3.70047778e-01
-5.82713068e-01 6.27379477e-01 2.80946255e-01 -3.15674752e-01
-9.70192254e-02 -1.02539086e+00 -6.00022316e-01 -7.70503998e-01
-2.71434575e-01 -1.21434927e-01 3.51990938e-01 -4.95742500e-01
-7.02563599e-02 6.09633029e-01 -1.83277452e+00 9.45556238e-02
9.29980576e-01 9.85032797e-01 2.99969107e-01 4.89631504e-01
1.76483005e-01 2.41345480e-01 5.88070489e-02 2.90095508e-01
4.22000110e-01 -2.47842833e-01 -6.22983217e-01 5.22827320e-02
5.35423398e-01 -3.87138665e-01 -2.83090621e-01 1.25293481e+00
5.07172465e-01 2.57575661e-01 4.50006455e-01 -8.88517797e-01
-1.27941534e-01 1.78721011e-01 -5.75550377e-01 6.38571382e-01
1.33088127e-01 2.45495632e-01 1.00881112e+00 -9.41049814e-01
7.41380572e-01 1.28459895e+00 6.59200907e-01 -1.61332682e-01
-1.33595169e+00 -4.19176280e-01 5.76069176e-01 3.13034862e-01
-1.34391415e+00 -7.12376356e-01 8.45811009e-01 -6.44626021e-01
9.37981963e-01 3.40817034e-01 4.18118209e-01 5.08756995e-01
1.64391398e-01 5.54181337e-01 1.22502458e+00 -4.16913480e-01
2.78277367e-01 -1.89440683e-01 -2.31129263e-04 5.71525276e-01
1.77973047e-01 1.68266356e-01 -2.99390107e-01 6.08104616e-02
4.53138947e-01 3.55279207e-01 -3.67795587e-01 -3.82616490e-01
-1.16105568e+00 7.66367793e-01 9.64727819e-01 5.53074896e-01
-5.91828823e-01 1.54327646e-01 1.83280945e-01 2.87473083e-01
9.32791114e-01 1.87493518e-01 -4.99515980e-01 1.22103751e-01
-1.16223085e+00 1.86562054e-02 2.01140136e-01 3.62159252e-01
1.21626246e+00 3.51312920e-04 -1.52128488e-02 5.98784864e-01
3.71833712e-01 8.02884281e-01 3.39319438e-01 -4.89089608e-01
4.44872469e-01 6.65769160e-01 2.73593068e-01 -1.44028687e+00
-5.06833613e-01 -4.83401328e-01 -9.11603808e-01 2.25454599e-01
-1.49154877e-02 1.04881994e-01 -1.23722851e+00 1.74612355e+00
5.52611172e-01 1.55832112e-01 -3.55422273e-02 7.98581243e-01
4.88596380e-01 7.24865139e-01 2.47870646e-02 -1.30961984e-01
1.19330990e+00 -4.91812021e-01 -5.40801406e-01 -3.33705634e-01
3.00616294e-01 -3.68496120e-01 8.45964074e-01 -2.17414647e-01
-5.42693496e-01 -4.68038172e-01 -1.05474436e+00 1.52025953e-01
-7.38760293e-01 5.00714295e-02 4.46490407e-01 3.03260833e-01
-1.16222179e+00 5.30248106e-01 -1.25452507e+00 -7.04499006e-01
4.49154139e-01 1.23086981e-01 -4.97307301e-01 -2.42357165e-01
-1.29258645e+00 7.97398746e-01 4.61190343e-01 3.12993824e-01
-7.69779205e-01 -6.42373800e-01 -1.01425850e+00 1.16917565e-02
3.13201904e-01 -5.28512418e-01 8.03741813e-01 -1.11054623e+00
-1.06848049e+00 7.00188875e-01 -2.71327138e-01 -4.15400535e-01
6.19971752e-01 -2.39193097e-01 -4.65904772e-01 4.06331509e-01
1.62484676e-01 7.22560287e-01 9.78649199e-01 -1.22950661e+00
-8.22117090e-01 -4.93043840e-01 1.67542487e-01 3.33018363e-01
-1.69398919e-01 -2.15517581e-01 -2.97890514e-01 -7.57704318e-01
4.35234487e-01 -8.49726140e-01 -3.75218749e-01 1.75640881e-01
-1.01236969e-01 2.90843517e-01 1.00965166e+00 -8.64679933e-01
1.16960168e+00 -2.25567961e+00 -3.65067199e-02 3.81322026e-01
8.57343301e-02 1.26702964e-01 -2.10483283e-01 3.60559255e-01
-3.97548527e-02 6.90647662e-02 -7.41917014e-01 -1.43591687e-01
-9.82623100e-02 1.00353338e-01 -3.88473004e-01 7.85627246e-01
3.36020827e-01 8.07173729e-01 -1.00488079e+00 -1.13133714e-01
5.02931654e-01 5.18540263e-01 -4.82879549e-01 -8.37117285e-02
-7.51197413e-02 7.09373593e-01 -4.14406836e-01 5.25783718e-01
7.92486668e-01 -4.29912135e-02 7.67405517e-03 -1.83362827e-01
-4.92095649e-01 3.35756630e-01 -1.25275505e+00 1.79192340e+00
-3.65768701e-01 5.59179187e-01 -8.20438266e-02 -8.51900756e-01
6.84893429e-01 5.74922301e-02 6.04101121e-01 -8.14135969e-01
-2.54226178e-01 1.33523941e-01 -1.84712738e-01 -1.70771018e-01
5.12078762e-01 -2.14980438e-01 -6.97167963e-02 1.95834324e-01
-3.19895089e-01 -1.11711495e-01 7.85674676e-02 7.66213015e-02
1.18236351e+00 1.48324609e-01 4.97931719e-01 -3.12087327e-01
5.51629782e-01 -2.32127272e-02 4.95126277e-01 7.23322213e-01
-2.85658568e-01 4.29298580e-01 3.59288715e-02 -4.87578362e-01
-7.27793097e-01 -9.98608768e-01 -2.74112850e-01 9.16714072e-01
9.12052915e-02 -1.51786596e-01 -2.26361200e-01 -6.13725483e-01
2.23422989e-01 5.86872697e-01 -6.81454778e-01 -1.18751138e-01
-4.21435028e-01 -1.03049552e+00 2.06710830e-01 2.81969458e-01
8.09007645e-01 -7.63686717e-01 -7.59551167e-01 2.45804787e-01
-3.01438123e-01 -9.10347223e-01 -5.19817546e-02 3.89802195e-02
-9.41782773e-01 -8.49271595e-01 -3.50405186e-01 -2.37595424e-01
6.41932070e-01 5.35130501e-01 7.54470527e-01 -2.87937373e-01
-4.15982336e-01 6.41899347e-01 -4.29897368e-01 4.02809083e-02
-1.27176479e-01 5.10626622e-02 -1.39623761e-01 2.18725264e-01
6.97512701e-02 -8.07704926e-01 -7.53187537e-01 1.39033105e-02
-1.15090442e+00 9.45613608e-02 6.03726923e-01 7.26530254e-01
5.27139485e-01 3.72915596e-01 3.71213853e-01 -5.15469551e-01
-9.32734236e-02 -6.50110364e-01 -5.52930713e-01 1.04754426e-01
-6.21128500e-01 2.17749521e-01 1.04786307e-01 -2.12216675e-02
-1.20141017e+00 2.43020222e-01 3.61116827e-02 -6.47087619e-02
-2.98473358e-01 9.21474159e-01 -2.47054622e-01 -3.20991427e-02
5.62863231e-01 2.57063776e-01 -3.93965930e-01 -5.75227499e-01
4.88181412e-01 5.40486991e-01 6.35672271e-01 -6.34244159e-02
1.02540839e+00 1.01426649e+00 -7.44249821e-02 -8.75422418e-01
-6.32582068e-01 -8.58383119e-01 -8.08302999e-01 -1.97558790e-01
6.71434164e-01 -1.28943539e+00 8.81125554e-02 7.08973229e-01
-7.83570707e-01 -1.71298221e-01 -1.90609127e-01 3.93135548e-01
-3.90696198e-01 4.26331013e-01 -2.01881945e-01 -6.41422808e-01
-2.75368184e-01 -8.16754162e-01 1.22743464e+00 -1.45684704e-01
1.87962055e-01 -1.03560555e+00 2.17890054e-01 -1.12039130e-02
7.03115582e-01 5.72042882e-01 7.49424040e-01 -2.29921907e-01
-6.22029543e-01 -4.39607650e-02 -2.83300221e-01 2.25544542e-01
6.69384718e-01 -3.12268317e-01 -1.06977582e+00 -5.97414136e-01
-2.01318502e-01 4.91422750e-02 1.54856539e+00 3.64329338e-01
7.46132195e-01 -3.11072409e-01 -3.58995020e-01 5.79675138e-01
1.80346215e+00 -1.39446706e-01 6.78910077e-01 5.15363097e-01
7.80199051e-01 4.98564243e-01 5.41906595e-01 7.89729357e-01
5.63891113e-01 7.09330559e-01 5.90838253e-01 -3.26119810e-01
-1.43823445e-01 -6.71896785e-02 4.30027515e-01 5.03819644e-01
2.83665918e-02 2.43148878e-02 -1.23801303e+00 8.24616611e-01
-1.83697295e+00 -1.12571192e+00 7.49490857e-02 2.21540666e+00
8.60904932e-01 -1.21999914e-02 -2.66623527e-01 9.14489925e-02
7.63922095e-01 5.73501945e-01 -6.16867542e-01 3.43107015e-01
-1.87150508e-01 7.66829327e-02 7.60888278e-01 6.83578908e-01
-1.68441355e+00 1.03896213e+00 5.67302942e+00 4.44241107e-01
-1.35837722e+00 1.93482429e-01 3.15529287e-01 1.91957816e-01
-5.72492480e-01 1.13667943e-01 -5.83490312e-01 3.25607151e-01
7.20091879e-01 1.41079679e-01 2.21275106e-01 4.21545058e-01
5.62922060e-01 -4.73973989e-01 -7.91215062e-01 7.85404742e-01
2.04896405e-02 -9.92116570e-01 1.23525552e-01 -1.63838133e-01
8.50487828e-01 3.96895498e-01 -4.70726751e-02 1.54104218e-01
3.58787715e-01 -6.25273645e-01 7.67172396e-01 8.85418355e-01
7.34439611e-01 -5.37706435e-01 6.69842124e-01 1.67406559e-01
-1.47928810e+00 -3.95416915e-02 -3.31355631e-02 -2.30129510e-01
6.84017017e-02 9.20446754e-01 -6.91539824e-01 9.73004699e-01
7.64731109e-01 1.22720969e+00 -8.50673258e-01 1.00521898e+00
-2.20520422e-01 6.30667806e-01 -6.52379334e-01 6.64457500e-01
3.09973598e-01 3.11334468e-02 7.35351801e-01 1.30081415e+00
3.91438484e-01 5.53733855e-02 4.93436694e-01 6.20073915e-01
1.48163348e-01 -4.98463660e-02 -6.44088924e-01 1.00867204e-01
3.07390422e-01 1.06008220e+00 -7.55082011e-01 -3.61111879e-01
-3.84000838e-01 1.06792152e+00 8.17680508e-02 4.99158531e-01
-4.81767863e-01 -3.35871339e-01 7.34965563e-01 -5.08994907e-02
6.77804291e-01 -3.76995742e-01 -9.30101350e-02 -1.36097121e+00
3.26567680e-01 -4.91800100e-01 3.55444491e-01 -5.49139202e-01
-1.05031157e+00 2.95536041e-01 1.93762034e-01 -1.32072985e+00
-2.03357100e-01 -1.76028699e-01 -4.59965080e-01 8.39397311e-01
-2.29855680e+00 -1.35344052e+00 -5.76901436e-01 4.89226550e-01
3.38019133e-01 2.07204878e-01 6.19357944e-01 2.97475606e-01
-4.15081441e-01 4.93642390e-02 4.01910692e-01 -3.41453031e-02
6.84263766e-01 -1.06838262e+00 4.51219022e-01 1.47955477e+00
-6.47109421e-03 2.26319522e-01 7.59440899e-01 -7.03005552e-01
-1.07803047e+00 -1.52474272e+00 8.57678115e-01 3.14284116e-02
7.07613766e-01 -9.26079899e-02 -9.23430800e-01 5.76505959e-01
-1.41241312e-01 2.67695248e-01 3.53157580e-01 -2.96025374e-03
-4.05863136e-01 -4.56794828e-01 -1.06290305e+00 3.76864225e-01
9.33943808e-01 -6.47064686e-01 -6.09452248e-01 -4.49389964e-03
6.43637180e-01 -5.00700250e-02 -5.93248129e-01 6.97198689e-01
2.53193974e-01 -6.78790510e-01 8.53510559e-01 -2.59566844e-01
1.25116065e-01 -7.28387117e-01 -6.08978152e-01 -1.33659041e+00
-5.85245967e-01 -1.82421446e-01 2.71933734e-01 1.05872691e+00
1.10795066e-01 -7.31036186e-01 1.70809641e-01 2.33268157e-01
-2.18611985e-01 -1.21922761e-01 -1.14037371e+00 -6.38528526e-01
-1.14990495e-01 -5.66890597e-01 4.30880278e-01 1.09737885e+00
-3.49660754e-01 7.40896091e-02 -2.61637300e-01 6.96479917e-01
3.54490101e-01 3.05224687e-01 5.75259566e-01 -1.37059712e+00
-3.99384126e-02 -3.41693014e-01 -7.90037632e-01 -8.20656359e-01
1.09582886e-01 -9.89984453e-01 3.02403033e-01 -1.68363595e+00
2.64886469e-01 -5.69332540e-01 -6.35588109e-01 9.70840633e-01
-4.59040552e-01 2.42706016e-01 1.12468779e-01 3.81621361e-01
-5.26914418e-01 9.78071809e-01 6.91890895e-01 -4.14295256e-01
-3.47435474e-01 -3.08551729e-01 -3.60011548e-01 6.62619948e-01
7.44143188e-01 -5.17872453e-01 -1.67472631e-01 -3.58309001e-01
2.11338371e-01 -3.87542754e-01 7.76035190e-01 -1.28476763e+00
3.14310975e-02 -2.41884738e-01 2.62688458e-01 -6.19713545e-01
1.27764985e-01 -9.47510362e-01 1.66776314e-01 5.24614394e-01
-6.07230365e-02 -5.88306226e-02 3.44151855e-01 8.25678229e-01
-4.14961934e-01 2.65120655e-01 7.74700105e-01 2.09457874e-02
-1.23834705e+00 3.27085584e-01 -4.46263492e-01 -2.91235298e-01
8.65687191e-01 1.44427642e-02 -2.01245502e-01 -2.95456916e-01
-6.42153442e-01 1.98909312e-01 5.40738881e-01 3.73028576e-01
4.47909623e-01 -1.25881457e+00 -9.11855400e-01 2.72897184e-01
4.53202695e-01 -2.41628274e-01 4.43150699e-01 9.46150959e-01
-2.94504523e-01 2.99700826e-01 -2.98525572e-01 -7.78367758e-01
-1.05561841e+00 3.48513871e-01 4.26315010e-01 -3.20351213e-01
-7.20985234e-01 2.93818414e-01 2.02165499e-01 -4.57834035e-01
-2.54826605e-01 -5.85902631e-01 -1.51954681e-01 3.69806707e-01
4.80179101e-01 2.64823496e-01 3.56823146e-01 -7.96994627e-01
-6.24357760e-01 4.24992502e-01 5.46965338e-02 -3.39424968e-01
1.56862426e+00 -5.11375964e-01 -1.21244855e-01 7.32694685e-01
1.10303044e+00 -1.33120820e-01 -1.49716341e+00 -6.71484172e-01
5.05286381e-02 -5.74400842e-01 4.58737224e-01 -9.01560128e-01
-9.40564215e-01 7.27006316e-01 1.09991729e+00 -1.54094458e-01
1.45834899e+00 -3.88209000e-02 3.75409037e-01 5.36428094e-01
3.95152062e-01 -9.36081648e-01 -1.64054722e-01 5.29438496e-01
9.72332656e-01 -1.60606992e+00 2.37063915e-01 -2.81969815e-01
-2.66842753e-01 7.48754978e-01 6.25850335e-02 3.57030556e-02
9.66664672e-01 -1.90171823e-01 -5.30990399e-03 -2.11575806e-01
-6.16078734e-01 -7.94955313e-01 4.32376623e-01 3.58177722e-01
5.95145226e-02 2.49218628e-01 -8.00803900e-02 -4.55037411e-03
2.24472418e-01 -1.66326821e-01 2.08267018e-01 1.12318850e+00
-7.38185167e-01 -7.95750022e-01 -3.86922538e-01 3.56079757e-01
-1.69678971e-01 -3.59949887e-01 -4.14445326e-02 6.15151942e-01
1.42331883e-01 9.00744617e-01 2.11660162e-01 -1.63633496e-01
2.74374038e-01 -3.66521589e-02 8.53047073e-02 -4.71180081e-01
-2.06548095e-01 7.94362836e-03 -5.48716709e-02 -6.48916900e-01
-8.51953268e-01 -1.06972146e+00 -9.62891996e-01 -1.43002585e-01
-5.20892181e-02 -2.60278434e-01 6.70295179e-01 1.01776779e+00
5.01497805e-01 3.95376027e-01 7.61871696e-01 -1.03471828e+00
-3.24608952e-01 -1.07116771e+00 -4.87410992e-01 4.23861742e-01
8.33260357e-01 -8.60858858e-01 -4.31418777e-01 2.72246510e-01] | [9.662079811096191, -1.2805256843566895] |
0d32d23d-a2d2-439d-9869-67333c4b6eb1 | availability-adversarial-attack-and | 2301.01832 | null | https://arxiv.org/abs/2301.01832v1 | https://arxiv.org/pdf/2301.01832v1.pdf | Availability Adversarial Attack and Countermeasures for Deep Learning-based Load Forecasting | The forecast of electrical loads is essential for the planning and operation of the power system. Recently, advances in deep learning have enabled more accurate forecasts. However, deep neural networks are prone to adversarial attacks. Although most of the literature focuses on integrity-based attacks, this paper proposes availability-based adversarial attacks, which can be more easily implemented by attackers. For each forecast instance, the availability attack position is optimally solved by mixed-integer reformulation of the artificial neural network. To tackle this attack, an adversarial training algorithm is proposed. In simulation, a realistic load forecasting dataset is considered and the attack performance is compared to the integrity-based attack. Meanwhile, the adversarial training algorithm is shown to significantly improve robustness against availability attacks. All codes are available at https://github.com/xuwkk/AAA_Load_Forecast. | ['Fei Teng', 'Wangkun Xu'] | 2023-01-04 | null | null | null | null | ['load-forecasting'] | ['miscellaneous'] | [-2.49753147e-01 -5.88257983e-02 -1.29090667e-01 -1.49243131e-01
-3.90758842e-01 -8.22123230e-01 3.05618197e-01 6.28677979e-02
1.55688286e-01 8.33752275e-01 -2.95891076e-01 -8.30667436e-01
-1.79132774e-01 -1.15252411e+00 -6.80820227e-01 -9.12383497e-01
-4.41446513e-01 3.82777154e-01 -3.93199503e-01 -2.11689711e-01
8.69003758e-02 7.69964814e-01 -6.92176938e-01 -1.85991779e-01
8.91616404e-01 1.25802445e+00 -5.26872694e-01 5.16548753e-01
2.72610337e-01 8.16738307e-01 -1.18839276e+00 -3.45069885e-01
6.51910305e-01 1.15019884e-02 -6.07654095e-01 -4.71427798e-01
-1.74283803e-01 -8.55822384e-01 -6.53826773e-01 1.29357648e+00
6.71562254e-01 4.93108183e-02 3.96289110e-01 -2.06109190e+00
-3.86578441e-01 9.38774824e-01 -1.76811814e-01 3.23330671e-01
2.35751122e-02 1.63083196e-01 7.81923115e-01 -3.22863132e-01
-3.11305165e-01 7.60290444e-01 3.54075968e-01 3.04325819e-01
-8.75960708e-01 -1.32447100e+00 1.33420393e-01 4.20950890e-01
-1.30164087e+00 -2.34626457e-02 1.10903502e+00 -5.68839349e-02
9.61249828e-01 3.49316150e-01 4.83758956e-01 9.39475715e-01
5.95372558e-01 8.55566740e-01 8.26138020e-01 -9.22436491e-02
4.94205117e-01 -1.98588148e-02 -3.59164774e-01 2.25504078e-02
2.86155015e-01 6.36315227e-01 2.66065337e-02 -4.57344800e-01
3.36386323e-01 8.19124654e-02 -3.33218545e-01 -9.66364797e-03
-6.11157000e-01 9.17141318e-01 5.74895203e-01 1.77970633e-01
-3.85888100e-01 2.62645125e-01 6.88353360e-01 5.91070890e-01
4.71790552e-01 6.08794630e-01 -5.43011308e-01 1.57460436e-01
-8.23877096e-01 2.39527598e-01 9.50723708e-01 7.18546867e-01
1.19432416e-02 1.18466794e+00 2.21172154e-01 1.63424447e-01
1.09078340e-01 7.91980147e-01 1.41283363e-01 -6.44599438e-01
5.37035644e-01 -4.47250418e-02 2.16619283e-01 -1.22044015e+00
-7.19222486e-01 -8.11595261e-01 -1.10607457e+00 5.25902271e-01
2.97854394e-01 -8.09318066e-01 -4.73004758e-01 1.49082148e+00
1.91238880e-01 5.59898138e-01 1.96697369e-01 6.73164487e-01
1.63696960e-01 1.00847518e+00 -2.35139549e-01 -1.30800813e-01
7.38945425e-01 -6.59464717e-01 -8.10798526e-01 3.25960368e-02
3.44644159e-01 -6.61321461e-01 3.09129238e-01 6.23061419e-01
-1.03237152e+00 -2.02108875e-01 -1.45144498e+00 7.43520856e-01
-5.64507246e-01 -3.37238491e-01 3.93027574e-01 1.26192510e+00
-6.16745532e-01 4.90525663e-01 -6.80546045e-01 4.58501518e-01
3.69032443e-01 5.86808264e-01 1.84913963e-01 3.62329513e-01
-2.02367234e+00 1.04430103e+00 5.49216449e-01 3.60824078e-01
-1.24882662e+00 -8.83114755e-01 -7.94709027e-01 3.54402393e-01
4.16239947e-01 -3.26964915e-01 1.18237495e+00 -8.38909090e-01
-1.68436980e+00 -1.64705798e-01 5.56696892e-01 -9.52747047e-01
5.72159946e-01 -8.67104754e-02 -7.96709657e-01 1.66069027e-02
-4.94767189e-01 -2.26699054e-01 1.00924015e+00 -1.00714815e+00
-5.41800380e-01 -8.29828009e-02 4.45903569e-01 -2.09870383e-01
-4.92813379e-01 1.45525575e-01 6.35027766e-01 -9.63537872e-01
-5.67474604e-01 -7.71282673e-01 -4.72237885e-01 -3.83204609e-01
-6.12982392e-01 1.98952973e-01 9.63388741e-01 -8.32701266e-01
1.13901949e+00 -1.92409122e+00 -2.32547700e-01 8.71670127e-01
-1.35128915e-01 5.00931025e-01 1.56924129e-01 7.87671804e-01
-4.86884177e-01 1.15102112e-01 -1.48186550e-01 -9.16459486e-02
4.50032890e-01 2.04875648e-01 -9.97876704e-01 8.48773658e-01
5.19182123e-02 7.84649372e-01 -4.99135584e-01 4.31711227e-01
6.53169334e-01 4.11235750e-01 -2.53920376e-01 2.42586732e-01
-1.26575038e-01 3.67291242e-01 -5.69282234e-01 6.42276406e-01
9.27704453e-01 1.08912311e-01 2.37914607e-01 -2.41805948e-02
3.32446992e-01 -4.42583337e-02 -1.00417423e+00 7.48506129e-01
-7.92074263e-01 5.74442506e-01 7.32571036e-02 -1.41085160e+00
9.17550027e-01 5.72903633e-01 6.96941137e-01 -7.53378332e-01
5.40228128e-01 1.12934694e-01 2.53806859e-01 2.50729054e-01
3.27149332e-01 2.65199933e-02 -3.72184366e-01 6.76203489e-01
-4.99897361e-01 -2.41323873e-01 -1.17504358e-01 1.99377865e-01
9.60989892e-01 -4.24984008e-01 1.25758916e-01 -2.63654679e-01
6.75470233e-01 -3.63775045e-01 6.10807955e-01 5.02157807e-01
-1.02875598e-01 -1.71995103e-01 4.79000211e-01 -7.42951632e-01
-1.02445030e+00 -1.10043979e+00 -2.79552460e-01 5.20142555e-01
1.33272886e-01 -1.27640918e-01 -5.73091745e-01 -6.90134883e-01
2.48586491e-01 1.17429638e+00 -3.29404980e-01 -4.76704657e-01
-4.90420073e-01 -7.27589607e-01 8.35777342e-01 6.94728851e-01
3.90446365e-01 -8.62436056e-01 -3.87467891e-01 3.67249489e-01
2.33326152e-01 -1.01094186e+00 -2.73364037e-01 2.02226147e-01
-1.69854313e-01 -1.00105047e+00 -5.32560706e-01 -1.87384754e-01
7.73013115e-01 -3.72019887e-01 1.03404057e+00 3.33225638e-01
-1.84608638e-01 1.47524387e-01 -2.50329316e-01 -7.65795410e-01
-7.33737946e-01 2.37592440e-02 4.62638050e-01 -1.65882170e-01
-2.77748909e-02 -7.33293414e-01 -3.68627518e-01 4.51077223e-01
-1.00234568e+00 -4.89114285e-01 1.06287077e-01 9.58790243e-01
1.51175931e-01 1.01147175e+00 1.09964752e+00 -6.84434474e-01
7.29354918e-01 -5.49101889e-01 -1.33516765e+00 3.36088613e-02
-7.03086913e-01 -2.59172201e-01 1.48701024e+00 -3.22398633e-01
-6.32056773e-01 -1.98634908e-01 -5.17337799e-01 -5.79668880e-01
9.96151716e-02 6.31384194e-01 -3.94967377e-01 -5.16763628e-01
1.20658867e-01 1.58400297e-01 -1.52795509e-01 4.82667014e-02
1.04226410e-01 5.48316300e-01 3.49849343e-01 -6.05637372e-01
1.39346826e+00 1.24705449e-01 2.31515869e-01 -4.09824163e-01
-5.34780025e-01 4.92015690e-01 -5.12883142e-02 -3.67700607e-01
7.36531839e-02 -6.37867808e-01 -1.16234779e+00 8.22907984e-01
-7.67193615e-01 -3.85938495e-01 -1.14937678e-01 2.00675786e-01
-4.42734361e-01 4.13418174e-01 -7.34916031e-01 -9.14904594e-01
-6.52551591e-01 -1.27227461e+00 1.62046850e-01 1.23306595e-01
5.20709231e-02 -1.17731750e+00 -3.63712370e-01 3.60366739e-02
7.06590474e-01 7.33080864e-01 7.34969139e-01 -1.10079634e+00
-3.81170571e-01 -7.80986190e-01 1.41789794e-01 6.65389419e-01
7.40136672e-03 1.73948497e-01 -7.50157952e-01 -8.05629909e-01
2.17773497e-01 -1.90001756e-01 -1.89750213e-02 3.77325445e-01
1.48109019e+00 -8.53417456e-01 -1.13724984e-01 7.21190095e-01
1.34463668e+00 6.87180758e-01 5.99634290e-01 4.68170851e-01
4.37725097e-01 1.70579642e-01 5.17001033e-01 8.75119328e-01
6.90755621e-02 4.08407778e-01 1.08913052e+00 5.22381999e-03
1.01206267e+00 1.70540974e-01 2.58717954e-01 5.18267632e-01
2.48146266e-01 -6.33583546e-01 -9.45677340e-01 1.65811718e-01
-1.44520295e+00 -9.49130654e-01 4.12186801e-01 2.09304833e+00
3.75350535e-01 6.42457068e-01 2.86025330e-02 6.94629669e-01
5.32502592e-01 3.01183164e-01 -8.09730232e-01 -7.77493119e-01
-2.82404013e-02 2.33117402e-01 8.51873338e-01 5.72316885e-01
-1.14528668e+00 4.27497983e-01 6.17787790e+00 9.64047730e-01
-1.32925093e+00 -1.06335044e-01 8.36894393e-01 -1.92433864e-01
-1.57021388e-01 -3.15880597e-01 -4.79301482e-01 8.78167808e-01
1.19363916e+00 -8.57299387e-01 5.22073686e-01 8.95052135e-01
1.69433057e-01 3.35821390e-01 -8.77857268e-01 4.89836186e-01
-1.64011717e-01 -1.31080616e+00 -2.27313027e-01 1.12864740e-01
7.47684479e-01 -1.67278007e-01 4.30868387e-01 3.50137919e-01
5.76044440e-01 -1.21836245e+00 4.55567688e-01 6.47308081e-02
4.71272111e-01 -1.78332818e+00 1.15390432e+00 3.29728842e-01
-1.09800279e+00 -5.54984272e-01 -1.48547173e-01 -2.54060984e-01
4.30243075e-01 6.25611961e-01 -4.77457076e-01 1.02717030e+00
6.21410072e-01 4.29059893e-01 5.56819923e-02 7.94717073e-01
-4.09722358e-01 7.74363518e-01 -4.77861941e-01 8.05711448e-02
2.46637598e-01 3.50789912e-02 5.54419041e-01 6.15476191e-01
2.27626666e-01 -6.34285957e-02 4.99205261e-01 5.91759562e-01
-1.13839477e-01 -3.39521229e-01 -4.86665279e-01 -4.74974327e-02
7.72379756e-01 1.04469585e+00 -4.19668257e-01 -8.22915360e-02
-1.05183706e-01 5.66822946e-01 -2.63894856e-01 2.81628162e-01
-1.27270186e+00 -5.79071462e-01 8.46964419e-01 -3.10069889e-01
5.46426279e-03 -2.41405204e-01 -3.17844421e-01 -7.57250667e-01
-7.39233345e-02 -1.14531517e+00 4.76737678e-01 -4.86687303e-01
-1.53115237e+00 6.33072436e-01 -1.68358609e-01 -1.28015578e+00
-5.32019258e-01 -4.70827609e-01 -1.02636194e+00 1.06444669e+00
-1.47834635e+00 -7.55892277e-01 1.98999569e-01 5.80081522e-01
3.33198577e-01 -5.07903337e-01 8.48961115e-01 4.37437236e-01
-9.38613057e-01 1.01805103e+00 6.04754865e-01 3.18265915e-01
6.95966184e-02 -1.05166483e+00 7.07578480e-01 1.28128445e+00
-2.32006013e-01 6.86957613e-02 9.60063338e-01 -4.73866016e-01
-1.36448348e+00 -1.09325790e+00 1.00654513e-01 -1.96612105e-02
1.07153440e+00 -2.23922431e-01 -8.47543478e-01 6.42228007e-01
5.56693971e-01 8.72287452e-02 6.16117179e-01 -4.35033083e-01
-1.79552406e-01 -1.01787344e-01 -1.62158036e+00 4.31894332e-01
1.43990010e-01 -4.15656149e-01 -4.07399796e-02 4.73614186e-01
5.27861774e-01 -7.93748856e-01 -1.20382118e+00 4.22128618e-01
2.56652832e-01 -5.86947441e-01 1.15705454e+00 -5.41324019e-01
1.46628216e-01 -2.44103938e-01 3.59210446e-02 -1.77061641e+00
-1.37256026e-01 -8.80742311e-01 -5.80105782e-01 9.01876867e-01
3.29272896e-01 -1.24219990e+00 7.01525569e-01 6.96369231e-01
-8.92392397e-02 -8.61536086e-01 -1.25456810e+00 -1.16032135e+00
5.34712911e-01 -3.81744891e-01 1.32751501e+00 1.25595343e+00
3.50429714e-02 -3.63647878e-01 -7.28672028e-01 1.01751983e+00
5.98359644e-01 2.69529447e-02 5.46660185e-01 -6.68219924e-01
-2.55476177e-01 -4.76518869e-01 -3.85137796e-01 -3.51606518e-01
5.80715895e-01 -6.30267143e-01 -3.04482639e-01 -1.06845391e+00
-8.18844557e-01 -4.42636192e-01 -5.79991400e-01 6.35756910e-01
1.14347361e-01 5.30478321e-02 3.98194343e-01 -2.37242788e-01
-1.04842754e-02 7.35333383e-01 6.88184500e-01 -5.46685159e-01
2.67886847e-01 5.40052831e-01 -1.84829906e-01 4.98972416e-01
1.71079695e+00 -4.31262255e-01 -4.24020112e-01 -2.17948437e-01
2.50350684e-01 3.79661351e-01 9.87924486e-02 -1.15848076e+00
2.03515902e-01 -1.79391518e-01 4.45489824e-01 -7.26639152e-01
1.68711632e-01 -1.46935737e+00 4.26023841e-01 9.83950794e-01
-1.33806720e-01 3.43388051e-01 5.90023994e-01 3.07720244e-01
-8.92791152e-02 -2.13466540e-01 6.35977626e-01 4.98838514e-01
-2.44497702e-01 6.28877401e-01 -6.04401171e-01 -3.16468805e-01
1.29916120e+00 2.28622347e-01 -4.92392898e-01 -7.65846133e-01
-6.24499023e-01 6.85413003e-01 6.94575086e-02 3.97383362e-01
4.51292336e-01 -1.26160407e+00 -6.85593367e-01 1.66928560e-01
-4.83666241e-01 -1.81012109e-01 4.62028414e-01 4.51416224e-01
-4.51699048e-01 3.42863858e-01 -4.20988351e-01 -2.22006775e-02
-8.50270748e-01 9.58002210e-01 7.66805768e-01 -3.94832194e-01
-3.77118051e-01 4.10356849e-01 -1.62663057e-01 -2.94971675e-01
3.36507678e-01 -3.06420568e-02 6.33373559e-02 -3.28402609e-01
6.63050056e-01 6.42566085e-01 1.32861644e-01 -2.81788200e-01
-3.25052977e-01 -2.33057309e-02 -8.48081708e-02 4.40364212e-01
1.13686740e+00 -6.75393222e-03 -1.76723506e-02 -1.82178780e-01
8.48344505e-01 -5.56258187e-02 -9.33815122e-01 -1.30991429e-01
-2.63981462e-01 -6.28267407e-01 2.28832036e-01 -9.28664982e-01
-1.89055443e+00 7.44815886e-01 3.92595619e-01 8.01540911e-01
1.43903112e+00 -9.06202257e-01 1.04940593e+00 3.63960832e-01
5.52542806e-01 -8.04311693e-01 -3.65596265e-01 4.83481795e-01
7.47037828e-01 -9.46683288e-01 5.98989353e-02 -1.78739876e-01
-3.53348047e-01 1.08751917e+00 6.34310246e-01 -3.31708103e-01
9.72739339e-01 1.08543372e+00 3.51748392e-02 2.95733809e-01
-4.79028314e-01 6.19611859e-01 -1.78879604e-01 5.56369483e-01
-1.91561818e-01 2.30288267e-01 9.68354940e-02 7.34286368e-01
-3.11988264e-01 -1.84321523e-01 8.34823191e-01 9.02258396e-01
1.71314642e-01 -1.05405927e+00 -5.31869411e-01 4.46114063e-01
-7.58998811e-01 -1.34616867e-01 2.37467661e-02 4.84485596e-01
-4.23111647e-01 1.04528487e+00 5.82813062e-02 -4.43564564e-01
1.71448052e-01 -3.43129963e-01 -1.08671531e-01 1.01050951e-01
-8.47717643e-01 -4.28972840e-01 -2.97071114e-02 -4.32631761e-01
1.49821863e-01 -2.31722593e-01 -1.16401350e+00 -1.02898681e+00
-4.05449957e-01 3.63652974e-01 5.82735240e-01 7.52761066e-01
1.35865867e-01 8.56658995e-01 1.44383156e+00 -7.71618664e-01
-1.03953886e+00 -4.45862114e-01 -7.58026481e-01 -2.10168183e-01
3.29071641e-01 -6.14814878e-01 -6.91843927e-01 -7.51336575e-01] | [5.464780807495117, 7.405476093292236] |
52b7d307-c9d3-4d5a-935a-690bc1b2acb8 | neuroprim-an-attention-based-model-for | 2210.12453 | null | https://arxiv.org/abs/2210.12453v2 | https://arxiv.org/pdf/2210.12453v2.pdf | NeuroPrim: An Attention-based Model for Solving NP-hard Spanning Tree Problems | Spanning tree problems with specialized constraints can be difficult to solve in real-world scenarios, often requiring intricate algorithmic design and exponential time. Recently, there has been growing interest in end-to-end deep neural networks for solving routing problems. However, such methods typically produce sequences of vertices, which makes it difficult to apply them to general combinatorial optimization problems where the solution set consists of edges, as in various spanning tree problems. In this paper, we propose NeuroPrim, a novel framework for solving various spanning tree problems by defining a Markov Decision Process (MDP) for general combinatorial optimization problems on graphs. Our approach reduces the action and state space using Prim's algorithm and trains the resulting model using REINFORCE. We apply our framework to three difficult problems on Euclidean space: the Degree-constrained Minimum Spanning Tree (DCMST) problem, the Minimum Routing Cost Spanning Tree (MRCST) problem, and the Steiner Tree Problem in graphs (STP). Experimental results on literature instances demonstrate that our model outperforms strong heuristics and achieves small optimality gaps of up to 250 vertices. Additionally, we find that our model has strong generalization ability, with no significant degradation observed on problem instances as large as 1000. Our results suggest that our framework can be effective for solving a wide range of combinatorial optimization problems beyond spanning tree problems. | ['Tiande Guo', 'Congying Han', 'Yuchen Shi'] | 2022-10-22 | null | null | null | null | ['steiner-tree-problem'] | ['graphs'] | [ 4.86361116e-01 2.46978059e-01 -3.13145548e-01 -1.68451160e-01
-4.87136722e-01 -8.38824868e-01 -9.73001644e-02 7.05433637e-02
-2.46828020e-01 7.31902719e-01 -5.57470977e-01 -7.83891559e-01
-7.75153100e-01 -9.82178450e-01 -8.58540535e-01 -6.87038362e-01
-5.52153826e-01 7.19385982e-01 2.33738750e-01 3.61015126e-02
2.05111995e-01 7.62550175e-01 -1.11072218e+00 -1.80278659e-01
8.58103633e-01 1.01628458e+00 4.05772686e-01 4.84930277e-01
-2.30899140e-01 2.42847204e-01 -4.74813700e-01 -3.80622566e-01
5.19138515e-01 1.14455417e-01 -1.14493871e+00 3.09854776e-01
2.64072895e-01 5.73232025e-02 -4.23519284e-01 9.39833879e-01
3.83978456e-01 2.50801425e-02 3.19098741e-01 -1.75662136e+00
-4.24687088e-01 9.13214803e-01 -9.52550173e-01 2.72557169e-01
-5.82282878e-02 1.11199886e-01 1.24105608e+00 -2.37743661e-01
5.70545495e-01 1.21587837e+00 5.87505877e-01 4.92151320e-01
-1.46039212e+00 -4.45967048e-01 7.09994495e-01 5.60725033e-01
-1.28105116e+00 -1.42022565e-01 5.53593278e-01 -1.57123990e-03
1.28378868e+00 1.57110140e-01 6.64535284e-01 1.07255673e+00
3.13968301e-01 8.18058550e-01 7.83309281e-01 -3.18631232e-02
5.86658657e-01 -5.81275702e-01 2.41957158e-01 6.53422058e-01
6.15056574e-01 6.35460019e-02 -1.49611562e-01 2.54833121e-02
6.88988388e-01 -1.80254981e-01 -2.01118793e-02 -6.51779532e-01
-1.00767720e+00 9.73641813e-01 4.52218831e-01 -3.68765630e-02
-2.81595856e-01 6.90166771e-01 2.88102359e-01 2.87034363e-01
-6.80161715e-02 4.50696290e-01 -7.57425368e-01 -8.83750319e-02
-7.59616017e-01 2.47549698e-01 1.22177267e+00 1.41767585e+00
4.49968904e-01 -1.18639313e-01 1.81049705e-01 5.86128354e-01
9.91377160e-02 3.05898070e-01 -3.44589263e-01 -1.23373055e+00
8.65292311e-01 3.63969237e-01 -5.37129678e-03 -1.17118371e+00
-8.03178847e-01 -6.07543349e-01 -9.92581010e-01 -1.80261210e-01
3.93343389e-01 -3.48511815e-01 -1.10202861e+00 1.89957416e+00
3.27004761e-01 4.35884967e-02 -3.28043029e-02 7.89141595e-01
3.87133747e-01 8.74857485e-01 -2.80729711e-01 -3.05843711e-01
1.02439737e+00 -1.04544425e+00 -3.55433375e-01 -6.47139907e-01
6.27000868e-01 -1.87207922e-01 3.68640393e-01 5.40692687e-01
-1.29039764e+00 1.86038539e-01 -9.00273621e-01 2.45015666e-01
-1.30092070e-01 -4.65444744e-01 1.21574330e+00 6.66176617e-01
-1.22749400e+00 6.21895850e-01 -9.79352534e-01 -3.20103377e-01
5.65396130e-01 7.92474806e-01 -4.85043228e-02 -4.37291324e-01
-7.77390838e-01 6.42618477e-01 3.75287384e-01 3.99680853e-01
-1.02577984e+00 -3.42689961e-01 -7.99799144e-01 3.48175555e-01
1.09391356e+00 -8.14502239e-01 1.14146292e+00 -5.16876101e-01
-1.27044141e+00 3.78614873e-01 -2.58356541e-01 -5.63970149e-01
2.95073986e-01 3.73954594e-01 -8.93991590e-02 9.79015231e-02
1.45995080e-01 6.08364761e-01 4.70251560e-01 -1.02941656e+00
-6.32524967e-01 -4.64268833e-01 4.22402471e-01 2.58974601e-02
4.08951789e-02 -1.16654888e-01 -7.21423388e-01 -8.98524225e-02
3.17893028e-01 -1.24267960e+00 -1.06524622e+00 -9.94645357e-02
-7.70182192e-01 -4.25851077e-01 3.51881713e-01 -7.91091919e-02
1.18037415e+00 -1.68322015e+00 6.03950441e-01 5.25285542e-01
3.67976636e-01 -1.74606234e-01 -7.24009633e-01 6.62967563e-01
5.74120358e-02 3.93640041e-01 -3.81374657e-01 -3.61351669e-02
1.71822578e-01 8.02516103e-01 -2.07594745e-02 3.53083253e-01
1.34404823e-01 9.44365859e-01 -8.24099362e-01 -3.91442835e-01
-2.20311210e-01 -8.70911255e-02 -7.04344749e-01 -4.86171037e-01
-5.50633430e-01 3.11418846e-02 -4.94895369e-01 6.40523791e-01
7.32373416e-01 -5.64880550e-01 5.61281383e-01 3.56136501e-01
8.94527286e-02 1.65628806e-01 -1.51027989e+00 1.72520578e+00
-5.42109191e-01 6.78580523e-01 2.88937539e-01 -1.31196105e+00
6.27375305e-01 -3.38973433e-01 6.64837003e-01 -6.01222575e-01
2.14203775e-01 -1.98246375e-01 2.81927854e-01 -4.26295131e-01
3.47037584e-01 3.06484640e-01 -2.43098870e-01 4.46091235e-01
-4.06580210e-01 1.62867527e-03 7.49290049e-01 2.86872864e-01
1.74630928e+00 -3.22250873e-01 -2.74594873e-01 -5.21210097e-02
-1.21752858e-01 4.72696871e-02 9.55090284e-01 7.98660994e-01
-3.72389369e-02 1.84975177e-01 1.03229105e+00 -4.54872996e-01
-7.61010945e-01 -9.84272361e-01 1.26917632e-02 6.37045860e-01
5.29911160e-01 -3.81998986e-01 -7.59209216e-01 -5.55548131e-01
-1.37845865e-02 5.10222554e-01 -4.61638331e-01 5.36124669e-02
-5.46505332e-01 -9.96654630e-01 2.69979239e-01 6.88424170e-01
3.09083134e-01 -7.87952423e-01 -5.91569126e-01 6.62495375e-01
-1.06545322e-01 -1.55891061e+00 -3.57303619e-01 6.60919189e-01
-1.07812297e+00 -1.08192575e+00 -2.49535680e-01 -1.03573775e+00
6.84942245e-01 4.39554006e-01 9.73190069e-01 8.28152150e-02
-5.68748116e-01 2.13637844e-01 -8.30256790e-02 -6.41087815e-02
-1.91126217e-03 5.07229209e-01 -1.07021205e-01 -3.78049254e-01
1.22074284e-01 -7.34021068e-01 -4.74819988e-01 6.44187272e-01
-7.65108466e-01 7.43318200e-02 6.41369224e-01 5.52605033e-01
7.91120827e-01 6.71438277e-01 5.33492744e-01 -5.58317602e-01
5.80407560e-01 -5.67509413e-01 -1.13399267e+00 2.21156448e-01
-5.91282725e-01 2.21444994e-01 5.41209221e-01 -4.69593436e-01
-3.31903100e-01 2.35876918e-01 2.29340941e-01 -3.65578324e-01
5.23322225e-02 7.71037698e-01 -2.80997366e-01 -1.63214013e-01
1.61009550e-01 1.22878477e-01 -2.88155049e-01 -1.67353660e-01
1.85052827e-01 1.34374142e-01 1.60965830e-01 -7.94221342e-01
7.74042666e-01 3.76549065e-01 7.34774411e-01 -5.17842531e-01
-8.33574951e-01 -1.14708446e-01 -3.16392660e-01 1.00010999e-01
5.93951225e-01 -4.64260280e-01 -1.18971360e+00 2.22925588e-01
-1.22141075e+00 -6.26625478e-01 6.64859042e-02 6.16779039e-03
-6.40138447e-01 5.37513852e-01 -6.84190691e-01 -7.82693684e-01
-6.73811361e-02 -1.28059530e+00 8.91394258e-01 1.39676616e-01
1.36882886e-01 -9.06380475e-01 -1.68481812e-01 3.66690993e-01
2.27367863e-01 4.93064761e-01 1.14348269e+00 -2.50014693e-01
-1.31531823e+00 9.29261446e-02 -3.42057735e-01 -1.85001299e-01
-2.29482368e-01 1.33144498e-01 -1.64962426e-01 -3.99476796e-01
-4.94224429e-01 -9.59665328e-03 7.71379590e-01 5.78706086e-01
1.65147889e+00 -2.81664640e-01 -7.02722609e-01 6.74270749e-01
1.60728824e+00 2.80284017e-01 5.39964378e-01 4.61638659e-01
5.38746595e-01 6.08660281e-01 4.36740696e-01 3.85482550e-01
5.65158367e-01 5.29661834e-01 9.87603724e-01 1.27178118e-01
2.91299522e-01 -1.17864106e-02 1.35001853e-01 5.09783089e-01
1.08539194e-01 -7.26053238e-01 -1.11660790e+00 7.76111603e-01
-2.04066372e+00 -8.31520796e-01 -3.59727442e-01 1.94817746e+00
5.20369828e-01 3.24971646e-01 -9.58752632e-02 2.00374842e-01
8.69737327e-01 -1.47329733e-01 -9.64645088e-01 -8.29389751e-01
1.04236975e-01 2.73168743e-01 9.55041885e-01 1.75700590e-01
-8.28155518e-01 1.13271511e+00 6.45017815e+00 7.40014255e-01
-7.94548333e-01 -1.92669168e-01 3.44282985e-01 -4.85222846e-01
-1.00573540e-01 1.32026091e-01 -9.23883379e-01 3.08581412e-01
1.03062534e+00 -3.11667293e-01 1.03264427e+00 7.96125412e-01
2.34015286e-01 -1.27468720e-01 -1.38601756e+00 1.04595327e+00
-3.38806897e-01 -1.39620411e+00 -3.28199446e-01 6.14150524e-01
9.40149844e-01 1.23924010e-01 1.50483087e-01 1.11556821e-01
8.12927604e-01 -1.03271019e+00 3.44505817e-01 -1.54871508e-01
5.20850718e-01 -1.07050371e+00 3.00714880e-01 3.21060985e-01
-1.26436305e+00 -4.56076026e-01 -5.65872908e-01 5.98186292e-02
4.04097438e-01 7.29039848e-01 -6.96795464e-01 5.22536993e-01
8.12090695e-01 4.61691499e-01 -8.12441185e-02 1.51802063e+00
-2.43135050e-01 3.82338613e-01 -8.26630712e-01 -2.18435422e-01
6.84319913e-01 -1.86408505e-01 6.08563066e-01 6.83569133e-01
1.62395731e-01 3.89329307e-02 2.85868913e-01 8.05774450e-01
-1.79923296e-01 -2.21150234e-01 -5.90536058e-01 -2.68703222e-01
6.40433609e-01 1.23780286e+00 -1.20150483e+00 1.83352098e-01
-2.36044228e-01 7.14318693e-01 4.65811670e-01 5.27103126e-01
-1.22479415e+00 -3.86813551e-01 7.06302583e-01 -2.84265608e-01
8.34409297e-01 -5.71819007e-01 -4.62284029e-01 -7.85867035e-01
1.87145486e-01 -7.72928953e-01 3.56706113e-01 -5.23941159e-01
-1.00457323e+00 4.82198268e-01 -8.94779712e-02 -5.94857991e-01
3.31371985e-02 -5.91611922e-01 -5.65528810e-01 4.14629340e-01
-1.50776267e+00 -6.94205880e-01 -1.23143077e-01 5.17234623e-01
4.92252707e-01 2.19479501e-01 3.35206509e-01 2.02344447e-01
-1.11598086e+00 4.73764688e-01 1.39529258e-01 -2.09838867e-01
-1.54500291e-01 -1.13613725e+00 6.94687605e-01 9.45334554e-01
1.70156136e-01 3.85178894e-01 6.89196944e-01 -4.32265788e-01
-2.16018224e+00 -1.17388594e+00 3.76684546e-01 -1.12974934e-01
8.40401232e-01 -3.09119761e-01 -4.63435650e-01 7.44112492e-01
-9.37713832e-02 -2.25631818e-01 4.72487658e-01 3.40959340e-01
-7.64682293e-02 -1.09547749e-02 -1.12282085e+00 7.35968411e-01
1.81334198e+00 2.32859492e-01 -8.75985846e-02 5.98505557e-01
9.48066652e-01 -4.13944662e-01 -7.76305556e-01 2.31100202e-01
2.19328389e-01 -3.63396734e-01 6.47258878e-01 -8.89978111e-01
5.40236473e-01 -2.13046968e-01 -2.24308550e-01 -1.38681614e+00
-5.33370018e-01 -8.99565458e-01 -3.73206913e-01 7.65243649e-01
8.08083832e-01 -6.16080880e-01 1.12105811e+00 6.66844845e-01
-1.92438275e-01 -1.19080567e+00 -1.19014299e+00 -1.18750346e+00
1.66370869e-01 -6.51861012e-01 7.69788802e-01 7.50778913e-01
-3.00947249e-01 4.40290272e-01 1.20223742e-02 5.80069005e-01
8.58783960e-01 5.52980781e-01 6.25988960e-01 -1.34761620e+00
-4.77343738e-01 -5.45039415e-01 -1.43136054e-01 -1.35678804e+00
3.28669161e-01 -1.04482806e+00 -1.67898610e-02 -2.17405200e+00
6.89631701e-02 -7.01999485e-01 4.00949344e-02 6.82966292e-01
4.40361112e-01 -3.21854442e-01 1.84800848e-01 -3.10085386e-01
-9.19463158e-01 5.41816771e-01 1.38302732e+00 -2.97669172e-01
-1.67426273e-01 1.92143053e-01 -8.47987652e-01 5.91736615e-01
9.18375909e-01 -6.74174190e-01 -4.07319725e-01 -9.91760015e-01
7.35914528e-01 3.81897032e-01 8.58088061e-02 -7.89112806e-01
3.82901788e-01 -5.98730505e-01 -4.89889681e-02 -7.46245086e-01
3.09444904e-01 -1.13285947e+00 1.78156078e-01 6.02115989e-01
-1.47857949e-01 3.30149323e-01 2.82402277e-01 8.53684187e-01
3.25561881e-01 -2.31284022e-01 6.28572524e-01 1.76757142e-01
-7.11740673e-01 8.83144021e-01 -5.94758332e-01 3.00468862e-01
1.50806999e+00 -4.76159066e-01 -4.50427353e-01 -2.63185292e-01
-9.88667130e-01 1.06713355e+00 2.16947690e-01 3.96689564e-01
6.92013025e-01 -9.14224446e-01 -4.24583554e-01 -3.20995152e-01
-1.12915486e-01 4.38201964e-01 2.20494360e-01 9.22446251e-01
-4.28291768e-01 4.66062874e-01 -3.85004841e-02 -6.00404561e-01
-1.22343647e+00 9.35777962e-01 2.56335199e-01 -5.52305102e-01
-6.61575615e-01 7.71611273e-01 -6.20453898e-03 -2.04351947e-01
5.21240830e-01 -5.35294235e-01 2.22944528e-01 -1.19769007e-01
1.38814718e-01 5.00125527e-01 -1.24752015e-01 8.14334024e-03
-4.23194945e-01 5.25162578e-01 -2.90419430e-01 1.45651072e-01
1.69421268e+00 -2.43603855e-01 -1.43490940e-01 -1.55871257e-01
9.59556043e-01 -5.15858710e-01 -1.00008619e+00 -1.36258975e-01
1.16798721e-01 -1.39023632e-01 1.70811340e-01 -7.14269459e-01
-1.55317318e+00 7.73521066e-01 1.66229740e-01 4.51431394e-01
1.13127899e+00 8.88411477e-02 1.15749037e+00 9.37720418e-01
9.19577301e-01 -1.05875182e+00 -2.70424634e-01 7.57981300e-01
5.52072704e-01 -7.77830720e-01 -2.41081953e-01 -7.89912105e-01
-2.48408899e-01 9.96238589e-01 6.66985154e-01 -9.29033086e-02
3.78528953e-01 3.82924229e-01 -6.88201785e-01 -1.95068613e-01
-1.32525921e+00 -2.79915839e-01 -4.36157882e-01 6.04111433e-01
-4.68711019e-01 1.39017448e-01 -2.40885988e-01 4.41050082e-01
-1.82019576e-01 -1.46257922e-01 8.41764748e-01 9.51276720e-01
-4.06761229e-01 -1.26026022e+00 -1.28922492e-01 5.63648403e-01
-3.83239865e-01 -1.95580274e-01 -3.75787914e-01 6.31861627e-01
-2.80425161e-01 1.17252326e+00 7.29848519e-02 -2.25419104e-01
1.28513128e-01 -5.37713110e-01 7.71471083e-01 -5.87967634e-01
-3.37698072e-01 -2.49379277e-01 4.64422643e-01 -8.96457374e-01
-4.59830798e-02 -6.33967400e-01 -1.25063431e+00 -6.38429046e-01
-4.04112011e-01 -2.27949824e-02 9.07051563e-01 9.21139717e-01
6.16526604e-01 6.07763946e-01 7.32467532e-01 -5.12159586e-01
-5.74351966e-01 -3.52151364e-01 -4.96791840e-01 -3.82094234e-01
8.23458284e-02 -6.96798980e-01 -2.29367927e-01 -5.40000319e-01] | [5.248636722564697, 2.9196431636810303] |
a5edf616-1bf9-49db-b859-22210c258294 | self-supervised-modality-invariant-and | null | null | https://openreview.net/forum?id=RunqFdkPuS | https://openreview.net/pdf?id=RunqFdkPuS | Self-Supervised Modality-Invariant and Modality-Specific Feature Learning for 3D Objects | While most existing self-supervised 3D feature learning methods mainly focus on point cloud data, this paper explores the inherent multimodal attributes of 3D objects. We propose to jointly learn effective features from different modalities including image, point cloud, and mesh with heterogeneous networks from unlabeled 3D data. Our proposed novel self-supervised model learns two types of distinct features: modality-invariant features and modality-specific features. The modality-invariant features capture high-level semantic information across different modalities with minimum modality discrepancy, while the modality-specific features capture specific characteristics preserved in each modality. These two types of features provide a more comprehensive representation of 3D data. The quality of the learned features is evaluated on different downstream tasks including 3D object recognition, 3D within-modal retrieval, and 3D cross-modal retrieval tasks with three data modalities including image, point cloud, and mesh. Our proposed method significantly outperforms the state-of-the-art self-supervised methods for all three tasks and even achieves comparable performance with the state-of-the-art supervised methods on the ModelNet10 and ModelNet40 datasets. | ['YingLi Tian', 'Bing Li', 'Zhimin Chen', 'Longlong Jing'] | 2021-09-29 | null | null | null | null | ['3d-object-recognition'] | ['computer-vision'] | [-2.29276419e-01 -1.81827024e-01 -5.88673353e-01 -6.05914056e-01
-1.23452830e+00 -6.81582689e-01 6.91948533e-01 4.43135649e-01
-1.29176062e-02 2.05026850e-01 2.60549843e-01 3.71805042e-01
-3.03221822e-01 -6.48729384e-01 -7.53192127e-01 -5.84589422e-01
-2.27068454e-01 5.66764653e-01 2.76524454e-01 5.10003306e-02
1.02370642e-01 8.73633087e-01 -1.89060402e+00 3.39997321e-01
3.50204647e-01 1.53416586e+00 5.53565565e-03 1.10982202e-01
-4.74220693e-01 4.08142507e-01 1.67508021e-01 1.47292301e-01
4.20955867e-01 1.22929528e-01 -7.84878492e-01 4.10371631e-01
8.78978431e-01 -8.78766179e-02 -6.80624664e-01 9.49975193e-01
5.24412215e-01 -8.16121921e-02 1.03326988e+00 -1.41219199e+00
-8.21869314e-01 -6.92148656e-02 -6.59632564e-01 -8.79118741e-02
5.64629972e-01 -5.58303148e-02 9.79723573e-01 -1.32325268e+00
8.49218726e-01 1.35877013e+00 3.91869366e-01 4.03339922e-01
-9.32532012e-01 -5.16997397e-01 1.01426952e-01 1.19849723e-02
-1.58541191e+00 -3.39792222e-01 1.17634141e+00 -4.35484558e-01
8.66644919e-01 -4.74833548e-02 4.35975164e-01 7.89826930e-01
3.05168517e-02 9.95409727e-01 1.15024030e+00 -2.19280899e-01
2.25600917e-02 1.48471624e-01 -8.26406032e-02 9.10150528e-01
-2.59068489e-01 1.11196801e-01 -1.18273985e+00 -3.21692020e-01
7.78174937e-01 5.34074187e-01 1.91267077e-02 -1.02650285e+00
-1.52418494e+00 4.74373072e-01 5.38341224e-01 3.00093651e-01
-3.05120438e-01 -8.56949687e-02 4.56862599e-01 6.00443661e-01
8.65646005e-01 -2.12115794e-01 -6.67152524e-01 1.03773437e-01
-5.37413180e-01 1.07698077e-02 5.30143738e-01 1.37692475e+00
1.18583417e+00 -2.67983705e-01 1.06806234e-01 1.00855386e+00
7.36795068e-01 1.23182845e+00 1.19751953e-01 -6.55635059e-01
4.82568473e-01 1.12922907e+00 -3.19562882e-01 -8.97980809e-01
-4.48238552e-01 -1.95576891e-01 -9.28659558e-01 1.23019613e-01
-3.92707586e-02 4.19212639e-01 -1.16924155e+00 1.67572844e+00
6.01969481e-01 4.60904948e-02 8.83036479e-02 1.02057707e+00
1.74336290e+00 5.44446349e-01 6.98570386e-02 1.38495803e-01
1.00755668e+00 -7.63614118e-01 -3.15557927e-01 1.12697132e-01
4.70556915e-01 -1.00259030e+00 5.93667686e-01 -3.84903967e-01
-1.17496943e+00 -6.65653169e-01 -8.15013945e-01 -1.13398269e-01
-6.36952996e-01 -1.14522755e-01 6.55564427e-01 2.73258090e-01
-9.63689446e-01 3.23901445e-01 -5.91096997e-01 -6.62310123e-01
6.11813009e-01 3.81375402e-01 -1.19213021e+00 -5.31189263e-01
-8.81667733e-01 8.41387987e-01 2.75906194e-02 -3.41762900e-01
-1.07164812e+00 -9.37636793e-01 -1.17898226e+00 -2.03551471e-01
4.39107269e-02 -9.46109474e-01 7.45344102e-01 -6.13818228e-01
-1.20847321e+00 1.67695498e+00 -1.90831110e-01 3.96434605e-01
7.77416229e-02 1.06712610e-01 -4.57984358e-01 6.28431201e-01
2.13734612e-01 9.29110408e-01 1.02997339e+00 -1.70668268e+00
-5.51821232e-01 -8.97373319e-01 -7.48627447e-03 5.64162016e-01
-3.38821620e-01 -3.76905322e-01 -7.39689350e-01 -2.39205733e-01
6.97630346e-01 -1.05240166e+00 2.92853773e-01 6.40753329e-01
-2.40394711e-01 -4.32254195e-01 1.07610834e+00 8.34152326e-02
2.30132118e-01 -2.35861373e+00 3.31705451e-01 4.36096787e-01
2.67105252e-01 -2.38438532e-01 -5.91031849e-01 4.69540536e-01
-9.69678387e-02 -2.99273640e-01 -3.03821526e-02 -6.38202846e-01
1.08111605e-01 2.11329713e-01 3.67085300e-02 8.01337957e-01
3.65762740e-01 1.08632779e+00 -9.29174542e-01 -8.52610826e-01
6.59022748e-01 6.43084288e-01 -1.12893626e-01 2.19202578e-01
-1.26821473e-02 5.36196887e-01 -6.89375699e-01 1.50043726e+00
8.35594893e-01 -5.86890459e-01 -3.49450886e-01 -6.09679818e-01
1.51841715e-01 -1.90025359e-01 -9.14230525e-01 2.57428050e+00
-4.73121315e-01 1.49585515e-01 -4.70524319e-02 -9.31039274e-01
9.86666560e-01 3.54759723e-01 9.98693645e-01 -7.39718139e-01
1.35917395e-01 3.44661653e-01 -7.81117082e-01 -5.93831003e-01
2.08781227e-01 -1.75311580e-01 -2.01487720e-01 3.94148350e-01
6.62829280e-01 -4.07788366e-01 -3.83133262e-01 3.14782560e-01
8.81887317e-01 1.07779168e-01 -2.12059021e-01 -1.83940321e-01
5.26072800e-01 -1.48212209e-01 9.27807987e-02 7.09948182e-01
-1.47605747e-01 8.95047665e-01 -5.45621999e-02 -2.07299054e-01
-8.06495607e-01 -1.55130482e+00 -2.52645493e-01 8.98210466e-01
7.36612618e-01 -3.17765713e-01 2.48930201e-01 -7.99816906e-01
6.03886664e-01 -2.30187684e-01 -7.01069415e-01 -2.60415614e-01
1.12323076e-01 3.49898124e-03 2.16884375e-01 4.98860657e-01
5.82600355e-01 -6.13178313e-01 1.27074555e-01 -3.83470297e-01
6.71570227e-02 -1.26290858e+00 -3.46777797e-01 -1.60916857e-02
-1.16496205e+00 -1.14568782e+00 -8.74570608e-01 -1.14451981e+00
8.66901338e-01 8.00878108e-01 1.20172524e+00 -1.38488179e-02
-2.86134146e-02 1.36401296e+00 -4.80122566e-01 -2.56328136e-01
-5.12238033e-02 1.36106282e-01 2.41322294e-01 1.48701906e-01
3.19196194e-01 -8.12818468e-01 -5.04626036e-01 3.40028971e-01
-1.02839017e+00 -2.73421735e-01 7.33229637e-01 7.12240279e-01
1.13851368e+00 -2.69796491e-01 3.96543920e-01 -4.74799067e-01
-9.94522125e-02 -6.59559071e-01 1.62082631e-02 4.07945216e-01
-2.19341233e-01 -7.16203004e-02 2.17761146e-03 -4.15281385e-01
-7.87236333e-01 1.84921011e-01 2.36726418e-01 -1.09111893e+00
-5.11884928e-01 5.64664841e-01 -4.29826051e-01 -7.29899347e-01
3.06065857e-01 3.64401728e-01 3.05247813e-01 -6.14658594e-01
4.94481415e-01 5.36150575e-01 3.21017742e-01 -7.21858859e-01
1.12313104e+00 1.03355420e+00 4.85098630e-01 -8.54839683e-01
-1.03587651e+00 -8.44703436e-01 -7.98490405e-01 -3.41305673e-01
6.00901306e-01 -1.39159083e+00 -5.43496788e-01 7.41884470e-01
-8.23728204e-01 2.60612279e-01 -4.21003133e-01 5.19886076e-01
-6.90687716e-01 4.23696876e-01 -5.44878602e-01 -3.39326143e-01
-2.55646050e-01 -7.86397099e-01 1.82214463e+00 1.80551067e-01
3.58469993e-01 -1.04823267e+00 1.02104545e-01 3.30337077e-01
2.58941114e-01 2.43356273e-01 9.91260469e-01 -4.82235581e-01
-7.20352232e-01 -4.77589250e-01 -4.91097927e-01 1.95188493e-01
4.11549330e-01 -2.70562679e-01 -9.18184876e-01 -3.45703393e-01
-4.01981294e-01 -8.17393541e-01 8.39384794e-01 2.09342986e-01
9.72766519e-01 4.03957337e-01 -3.75281841e-01 6.32064998e-01
1.48292160e+00 -4.91566092e-01 1.46170080e-01 -4.47252989e-02
8.57565582e-01 4.49694514e-01 5.87103546e-01 3.36947948e-01
6.22771621e-01 3.47010911e-01 7.45476007e-01 -2.61048943e-01
-8.56000260e-02 -3.95210445e-01 -8.97203758e-02 1.11084986e+00
3.96210849e-02 1.69533312e-01 -7.73153901e-01 7.18653500e-01
-1.93166220e+00 -6.90020502e-01 3.22079539e-01 1.89182091e+00
4.57769036e-01 -5.59384190e-02 -2.46121045e-02 -1.95466280e-01
6.53791308e-01 3.65453422e-01 -6.90298080e-01 4.44721073e-01
-5.80872774e-01 -1.30001632e-02 2.87204057e-01 -8.99618790e-02
-1.41616344e+00 5.75236142e-01 5.95299768e+00 7.59257674e-01
-1.02555048e+00 1.08796656e-01 1.59026012e-01 -2.66888648e-01
-5.41385114e-01 -1.86096400e-01 -4.64061260e-01 6.08885139e-02
3.40121657e-01 4.83108237e-02 -1.57894000e-01 8.12044680e-01
-3.12905759e-01 1.07971266e-01 -1.42792642e+00 1.50548196e+00
4.08607990e-01 -1.60200608e+00 4.50749516e-01 -4.36738320e-02
9.65536833e-01 4.43127453e-01 2.11982727e-01 1.02577314e-01
-2.94222951e-01 -7.69994736e-01 6.30869091e-01 8.59061301e-01
1.02108586e+00 -5.76605916e-01 5.77501416e-01 -4.16601337e-02
-1.34005558e+00 2.36646324e-01 -2.43906200e-01 3.56922805e-01
1.47690535e-01 6.62469625e-01 4.03135791e-02 8.92584622e-01
9.00423944e-01 1.47756016e+00 -5.76377630e-01 1.05162120e+00
2.70426601e-01 -3.68444016e-03 -5.97288728e-01 2.73526847e-01
2.84394652e-01 2.53636867e-01 7.96325982e-01 8.07713032e-01
9.73443016e-02 1.07113391e-01 3.92533064e-01 6.73676193e-01
-2.62762904e-01 1.72817558e-01 -8.88676941e-01 -6.16788231e-02
5.57887077e-01 1.22902024e+00 -3.87571037e-01 -1.79569259e-01
-7.86763668e-01 7.25911319e-01 1.72032192e-01 2.59743392e-01
-2.41303816e-01 -4.00399864e-02 7.12199152e-01 -1.51353911e-01
3.02363276e-01 -3.38973045e-01 -1.58562928e-01 -1.31305492e+00
1.71400785e-01 -3.34843040e-01 4.83636737e-01 -1.00536585e+00
-2.20022655e+00 3.92470717e-01 9.81974006e-02 -1.83545983e+00
1.74935758e-01 -6.12087250e-01 -1.71242967e-01 7.60936975e-01
-1.85171497e+00 -1.85515404e+00 -5.04840910e-01 1.27417612e+00
1.37543157e-01 -5.42103231e-01 9.13961053e-01 3.51077259e-01
2.18527317e-01 4.61186260e-01 2.15878710e-01 4.72667813e-03
9.89112556e-01 -8.74405503e-01 -9.35577080e-02 -5.82749099e-02
6.28611743e-02 4.62256074e-01 -4.67462949e-02 -5.06556153e-01
-2.09865999e+00 -9.27756906e-01 7.38923728e-01 -4.33192402e-01
4.66801941e-01 -1.36044040e-01 -7.21868873e-01 1.58888683e-01
-1.15538143e-01 6.70862079e-01 8.76333296e-01 1.78871810e-01
-9.25882459e-01 -2.29419991e-01 -1.27379572e+00 1.86616331e-02
1.34011555e+00 -1.09165764e+00 -5.43220222e-01 5.66861510e-01
7.63133407e-01 -6.21102452e-01 -1.38616538e+00 7.64778316e-01
5.92930794e-01 -7.00029910e-01 1.23532391e+00 -8.10771346e-01
5.62711239e-01 -2.50515819e-01 -7.45799005e-01 -1.03932333e+00
-1.98466882e-01 -2.93882377e-02 -2.75090069e-01 1.15314043e+00
2.80442953e-01 -3.25296491e-01 7.40292668e-01 3.80537063e-01
-1.80624112e-01 -7.23744333e-01 -1.22807133e+00 -7.62662888e-01
5.67960031e-02 -4.33525324e-01 6.51283383e-01 1.21203291e+00
-1.42535776e-01 2.71769762e-01 -1.56688616e-01 2.35020474e-01
9.97885764e-01 7.81640887e-01 6.90591514e-01 -1.33203244e+00
3.11651468e-01 -1.62975267e-01 -1.04149711e+00 -1.02543616e+00
5.95751643e-01 -1.30841446e+00 -3.68025094e-01 -1.37355196e+00
6.38943791e-01 -6.07090056e-01 -5.65597177e-01 6.34398341e-01
3.85684937e-01 5.18224478e-01 3.12394589e-01 4.26408291e-01
-1.07612062e+00 9.97582674e-01 1.47462010e+00 -6.06565773e-01
-9.53566842e-03 -2.13223442e-01 -3.32672387e-01 4.42569286e-01
3.15046698e-01 -2.61695385e-01 -3.39534968e-01 -6.00115955e-01
2.93310340e-02 9.06225070e-02 7.53632247e-01 -7.59263396e-01
3.45133722e-01 -1.23635821e-01 5.96921921e-01 -1.12360907e+00
6.05320394e-01 -1.24912059e+00 -5.84598407e-02 -2.97856092e-01
-2.73069292e-01 -2.15043128e-01 2.32569084e-01 7.80990958e-01
-4.68681902e-01 3.27346474e-01 6.49743795e-01 -1.72377795e-01
-8.96790385e-01 9.41529095e-01 1.35578528e-01 -1.07860617e-01
7.11409986e-01 -2.30206817e-01 -3.25048804e-01 -4.04805511e-01
-8.62688601e-01 3.77054334e-01 7.31546640e-01 9.70707357e-01
1.04145992e+00 -1.92756510e+00 -5.70869267e-01 3.33789557e-01
9.28990185e-01 1.44207880e-01 7.96013296e-01 5.57612240e-01
-2.07619313e-02 4.20718551e-01 -3.53576005e-01 -1.38917875e+00
-1.12922251e+00 3.72940809e-01 3.01980108e-01 2.25595504e-01
-5.81396222e-01 6.52801096e-01 5.60103096e-02 -9.83766556e-01
3.17383081e-01 2.52582818e-01 2.19630018e-01 -1.37470514e-01
8.33501369e-02 -6.31503686e-02 4.33543362e-02 -1.17569935e+00
-8.35260034e-01 1.24662721e+00 3.67168225e-02 1.65492177e-01
1.56927383e+00 -3.72103602e-01 -2.17379153e-01 8.11913908e-01
1.77851450e+00 -5.20207226e-01 -9.84125614e-01 -9.92826343e-01
-4.94466722e-01 -5.55510998e-01 2.29307711e-01 -6.82000995e-01
-1.35831785e+00 7.22425997e-01 9.47434247e-01 -1.37845576e-01
1.09866107e+00 7.88488626e-01 6.73575222e-01 3.90725642e-01
6.22339606e-01 -8.74872506e-01 2.56356865e-01 5.64370751e-01
7.63641477e-01 -1.78883827e+00 1.93263009e-01 -5.89053273e-01
-3.71759564e-01 7.76641071e-01 5.42527616e-01 -1.09751232e-01
1.30517387e+00 -3.16897720e-01 1.49018928e-01 -6.59381211e-01
-6.41107976e-01 -4.31466550e-01 8.61046672e-01 7.88895011e-01
6.04691021e-02 -1.92208737e-01 3.61133426e-01 2.33543232e-01
4.78774816e-01 -1.01746894e-01 -3.27246308e-01 1.36381638e+00
-4.49137725e-02 -9.29703057e-01 -2.48591006e-01 4.66621161e-01
1.00865006e-01 2.05335855e-01 -5.50217211e-01 8.38613153e-01
-1.05255783e-01 7.06592739e-01 1.30937681e-01 -5.11468828e-01
4.02766317e-01 -4.30886783e-02 9.28175330e-01 -4.82309103e-01
-2.30874673e-01 -1.93025619e-02 -3.38644147e-01 -7.11196721e-01
-1.30644989e+00 -6.34066105e-01 -1.00969613e+00 -1.90195620e-01
-1.88265726e-01 -2.03557774e-01 7.46075034e-01 1.03750575e+00
7.33547509e-01 -9.21091735e-02 1.01456499e+00 -1.32087719e+00
-1.65821493e-01 -6.79005206e-01 -9.01245773e-01 8.45187724e-01
4.90828514e-01 -1.21205425e+00 -4.81253982e-01 -1.41412437e-01] | [8.126936912536621, -3.456651449203491] |
191dc3fa-46c1-425f-a23b-f2bbf94de764 | where-is-your-place-visual-place-recognition | 2103.06443 | null | https://arxiv.org/abs/2103.06443v2 | https://arxiv.org/pdf/2103.06443v2.pdf | Where is your place, Visual Place Recognition? | Visual Place Recognition (VPR) is often characterized as being able to recognize the same place despite significant changes in appearance and viewpoint. VPR is a key component of Spatial Artificial Intelligence, enabling robotic platforms and intelligent augmentation platforms such as augmented reality devices to perceive and understand the physical world. In this paper, we observe that there are three "drivers" that impose requirements on spatially intelligent agents and thus VPR systems: 1) the particular agent including its sensors and computational resources, 2) the operating environment of this agent, and 3) the specific task that the artificial agent carries out. In this paper, we characterize and survey key works in the VPR area considering those drivers, including their place representation and place matching choices. We also provide a new definition of VPR based on the visual overlap -- akin to spatial view cells in the brain -- that enables us to find similarities and differences to other research areas in the robotics and computer vision fields. We identify numerous open challenges and suggest areas that require more in-depth attention in future works. | ['Michael Milford', 'Tobias Fischer', 'Sourav Garg'] | 2021-03-11 | null | null | null | null | ['visual-place-recognition'] | ['computer-vision'] | [ 6.10697307e-02 -1.74565226e-01 -1.73976392e-01 -4.64195348e-02
2.52192050e-01 -9.54275906e-01 7.91216075e-01 -2.06277687e-02
-5.29670775e-01 4.24130857e-01 4.19957265e-02 -9.95686203e-02
-1.21046335e-01 -5.43084502e-01 -5.07155061e-01 -6.21759355e-01
-1.20480500e-01 3.06465656e-01 3.09668392e-01 -4.92561847e-01
6.01405084e-01 9.83547628e-01 -1.91624367e+00 -1.46539614e-01
4.51987863e-01 9.03124988e-01 7.26993084e-01 4.55167592e-01
1.79772109e-01 8.51819813e-01 -3.38250577e-01 3.90661806e-01
3.74564141e-01 -2.56061316e-01 -7.26489961e-01 -3.94630432e-02
6.71357512e-02 -7.90644586e-02 -5.81139743e-01 9.89629984e-01
1.71032101e-01 4.46406782e-01 5.34349740e-01 -1.62893748e+00
-9.79618192e-01 1.98675007e-01 -4.56122428e-01 3.73074949e-01
6.69189334e-01 2.71301776e-01 7.20032156e-01 -7.76608288e-01
1.07065535e+00 9.34861958e-01 2.63276279e-01 6.31421387e-01
-9.29662168e-01 5.05176112e-02 4.91133422e-01 5.07859588e-01
-1.56935930e+00 -6.89967453e-01 9.75556552e-01 -6.15926802e-01
1.03547943e+00 3.68363678e-01 8.64888191e-01 1.00556195e+00
1.59564942e-01 6.41645908e-01 1.13232982e+00 -4.06033993e-01
7.47928858e-01 1.19297234e-02 4.78514805e-02 3.90182674e-01
3.49041633e-02 2.32006878e-01 -7.24516749e-01 8.07870999e-02
1.21136677e+00 2.16488406e-01 -4.55115080e-01 -1.08439517e+00
-1.68338180e+00 4.08540100e-01 8.57440174e-01 7.02788889e-01
-5.44184029e-01 2.13414639e-01 -9.39525738e-02 5.38732596e-02
-3.65025073e-01 8.57589722e-01 -7.39843771e-02 -2.67473441e-02
-3.79402101e-01 1.18665226e-01 4.65731919e-01 1.00713420e+00
7.87510991e-01 7.02514723e-02 7.17103556e-02 6.77519619e-01
5.39222240e-01 2.96810120e-01 5.58805168e-01 -1.08703530e+00
9.58063379e-02 6.43058062e-01 3.40647995e-01 -1.19898963e+00
-5.09822667e-01 -2.06820041e-01 -4.02898818e-01 7.41569936e-01
4.69474226e-01 3.07904214e-01 -7.77640879e-01 1.72497368e+00
2.24554390e-01 9.39099491e-02 1.73668444e-01 1.35625041e+00
7.09878623e-01 3.62248689e-01 -7.59484172e-02 2.36617640e-01
1.57989144e+00 -9.84157979e-01 -6.67453289e-01 -7.73175657e-01
4.13135737e-01 -2.17112228e-01 7.82039702e-01 -1.71455547e-01
-9.12535846e-01 -4.32421453e-02 -1.48592246e+00 -2.93060243e-01
-9.04615819e-01 1.60978779e-01 7.34879673e-01 1.92980424e-01
-1.42257357e+00 3.70434135e-01 -6.65690720e-01 -9.61801648e-01
1.86797678e-01 3.38782102e-01 -6.38467669e-01 2.15458795e-01
-7.51198947e-01 1.28627527e+00 2.42041573e-01 2.01882526e-01
-9.24363494e-01 -2.66478986e-01 -1.04336047e+00 -1.53410017e-01
1.09265245e-01 -7.33308613e-01 1.24413133e+00 -7.90979505e-01
-1.46447289e+00 1.18417954e+00 -2.35962346e-01 -3.92730772e-01
2.92907238e-01 2.39234388e-01 -2.78746516e-01 2.61706095e-02
3.00856903e-02 7.87805974e-01 2.82965273e-01 -1.45110059e+00
-7.10066020e-01 -9.69614685e-01 4.17698294e-01 6.65307224e-01
3.83925289e-01 -2.06743553e-02 -3.06216687e-01 -3.33986610e-01
7.00096607e-01 -1.16278982e+00 -3.93765658e-01 3.78237963e-01
-1.04204625e-01 -2.85788104e-02 7.92792439e-01 -1.67120963e-01
4.69577938e-01 -2.33411026e+00 3.74174088e-01 6.16048165e-02
4.35836136e-01 2.97732055e-02 -1.12123527e-01 3.77830863e-01
-2.80230287e-02 -1.70974717e-01 -1.56786833e-02 -1.62933283e-02
1.60586774e-01 1.85240582e-01 -5.51177919e-01 9.42401826e-01
-2.22755238e-01 9.57338035e-01 -9.69646454e-01 -1.59354106e-01
6.72792733e-01 4.53052580e-01 1.86955798e-02 -7.45666027e-02
-1.41923726e-01 7.58857787e-01 -4.91007030e-01 6.66010976e-01
5.32465696e-01 -4.93533071e-03 -4.68638726e-03 2.58388482e-02
-6.98202908e-01 2.70740509e-01 -1.28975129e+00 1.89361036e+00
-2.08888218e-01 9.77963567e-01 2.79844910e-01 -7.51405954e-01
1.05135214e+00 1.29011780e-01 2.95579344e-01 -1.06313002e+00
2.47566685e-01 2.73119897e-01 -1.07677402e-02 -3.63548130e-01
6.99340224e-01 3.52956146e-01 1.25413463e-01 1.46001443e-01
-8.07400793e-02 1.01898596e-01 -1.81785692e-02 -1.66055188e-01
1.08506143e+00 1.76788434e-01 6.45084858e-01 -3.66237849e-01
3.25253785e-01 1.94968060e-01 4.72102046e-01 6.58685267e-01
-7.75474906e-01 5.97890675e-01 5.90981357e-02 -6.66894138e-01
-7.83368230e-01 -1.12583208e+00 -7.27640614e-02 9.76250768e-01
8.65501642e-01 4.36222255e-02 -4.25664008e-01 -9.20328647e-02
-1.14477791e-01 6.64364636e-01 -8.06061625e-01 1.13793246e-01
-5.07835865e-01 -1.31538793e-01 2.29368359e-01 7.52302468e-01
4.36793119e-01 -1.34106445e+00 -1.51576066e+00 -9.81815383e-02
-1.29763484e-02 -1.09365809e+00 -1.09684002e-02 3.31725568e-01
-6.24476671e-01 -8.98018539e-01 -7.11621106e-01 -1.05147541e+00
8.20939362e-01 7.64651179e-01 8.50011289e-01 4.35412377e-02
-4.34400663e-02 6.66891992e-01 -3.57424527e-01 -3.95851493e-01
1.82992399e-01 -1.27614513e-01 2.39616975e-01 -4.22083251e-02
3.33900243e-01 -7.29491055e-01 -6.52514696e-01 4.99328732e-01
-6.16845667e-01 1.17204569e-01 2.82040060e-01 3.77607107e-01
6.59725487e-01 -2.54227579e-01 9.06002298e-02 -2.35315472e-01
2.73947507e-01 -3.77972573e-01 -8.49710584e-01 4.11542684e-01
-5.14874756e-02 -5.43218330e-02 4.18505400e-01 -2.65617251e-01
-6.13327861e-01 4.17703509e-01 2.50379890e-01 -2.10318372e-01
-4.40746605e-01 2.19575807e-01 -3.70382130e-01 -4.67076719e-01
7.59770930e-01 4.59135920e-01 -2.84309536e-01 -1.86992750e-01
4.28171724e-01 4.75270838e-01 6.20591640e-01 -2.06481934e-01
6.99174345e-01 8.35522771e-01 6.64687250e-03 -9.71404016e-01
-1.43042326e-01 -5.51846683e-01 -8.32151294e-01 -3.07853699e-01
9.46566880e-01 -6.33344531e-01 -9.19031441e-01 4.06232506e-01
-1.31476641e+00 -4.06621039e-01 -2.83420980e-01 5.19469142e-01
-9.07016277e-01 -1.43151000e-01 -1.82496101e-01 -9.92713630e-01
2.43210554e-01 -1.34699869e+00 9.99536574e-01 6.02207303e-01
-1.47379205e-01 -9.53019559e-01 3.32756400e-01 7.73298219e-02
4.08162832e-01 1.08593680e-01 4.70880359e-01 -4.55543101e-01
-8.17475617e-01 1.28695562e-01 -1.93726122e-01 -5.70378482e-01
-3.22592780e-02 -2.24291950e-01 -1.10016584e+00 -1.71929132e-02
1.02302739e-02 2.09759727e-01 6.12841606e-01 4.14710939e-01
6.77339256e-01 -1.28289117e-02 -7.67690241e-01 6.14568293e-01
1.49102044e+00 7.78567195e-01 8.10892403e-01 7.99445987e-01
5.44400156e-01 7.70980537e-01 5.93999028e-01 2.12622032e-01
5.52981615e-01 9.83253121e-01 9.36505139e-01 -4.33039200e-03
5.19292317e-02 -2.77817011e-01 6.05770759e-02 3.24443310e-01
-3.28970671e-01 2.06412692e-02 -1.06823742e+00 7.39063203e-01
-2.15731978e+00 -1.04861093e+00 2.00627387e-01 2.20778680e+00
1.32267118e-01 -5.53557634e-01 -6.41896799e-02 3.40071544e-02
8.07068348e-01 3.78560364e-01 -6.57924294e-01 -2.56417125e-01
-2.37596691e-01 -3.89382809e-01 5.08614182e-01 4.45847392e-01
-1.02358186e+00 1.02230585e+00 6.66386080e+00 6.11285865e-02
-1.32080138e+00 -3.29475403e-02 5.72159607e-03 1.55414835e-01
-9.04334188e-02 6.86041117e-02 -4.36744899e-01 2.31107339e-01
2.70444155e-01 -3.43807694e-03 1.10982180e+00 1.06118226e+00
1.57612599e-02 -3.12107831e-01 -1.56580830e+00 1.43083799e+00
1.63302511e-01 -1.38903952e+00 -3.80850732e-01 2.23856479e-01
3.29646558e-01 2.37626344e-01 4.10510778e-01 -1.18345924e-01
2.47946978e-01 -1.17625105e+00 1.10266280e+00 4.15862709e-01
2.07731754e-01 -6.53159916e-02 2.81068325e-01 3.21436405e-01
-1.31409943e+00 -2.96758533e-01 -5.12414277e-01 -3.68038058e-01
6.06811494e-02 -5.60552478e-02 -7.04537928e-01 2.27755859e-01
7.54721224e-01 6.25441909e-01 -5.55894971e-01 1.38151205e+00
-2.25432739e-01 -4.02956247e-01 -2.51447260e-01 -3.36854011e-01
8.35858807e-02 -3.66897434e-01 7.45111763e-01 6.61926270e-01
4.10066307e-01 9.90170091e-02 -7.42749870e-02 1.13326514e+00
3.76204371e-01 -1.01895437e-01 -1.21677995e+00 1.48863897e-01
8.14259589e-01 1.10992444e+00 -1.00490725e+00 1.22863412e-01
-1.58406228e-01 1.04511535e+00 4.06687677e-01 6.08350217e-01
-7.23234951e-01 -2.49433413e-01 1.06921124e+00 5.72839677e-02
1.42401055e-01 -7.87277520e-01 -4.53912705e-01 -1.05286872e+00
1.46490276e-01 -4.80893046e-01 -8.45760107e-02 -1.33624613e+00
-6.79060638e-01 6.43861890e-01 -3.35636646e-01 -1.33506370e+00
-2.19134569e-01 -9.40545857e-01 -3.20276141e-01 6.79979026e-01
-1.57846212e+00 -1.11168694e+00 -4.33029860e-01 6.40702546e-01
3.46845508e-01 -1.56159684e-01 9.32430923e-01 2.22251974e-02
-3.95538181e-01 1.21401876e-01 2.15758686e-04 1.10537671e-01
1.56909779e-01 -1.07766712e+00 4.99251395e-01 8.08846891e-01
5.10521173e-01 8.06752682e-01 7.62989700e-01 -1.90277770e-01
-1.90932298e+00 -7.15906680e-01 7.81487882e-01 -7.59283841e-01
3.53561819e-01 -4.04655516e-01 -4.47222471e-01 1.07923961e+00
1.72557816e-01 1.32513419e-01 4.59605783e-01 -1.76982265e-02
-4.02351797e-01 3.02681290e-02 -1.34749961e+00 1.28859901e+00
1.34023368e+00 -6.66521490e-01 -6.84317112e-01 6.22425042e-02
4.52375889e-01 -5.93719363e-01 -3.11396331e-01 9.56327096e-02
6.30690157e-01 -1.05054402e+00 1.07666647e+00 -3.38329643e-01
-2.08644837e-01 -1.02233601e+00 -5.41160107e-01 -1.39145267e+00
-6.17837429e-01 -2.98813790e-01 1.11911744e-01 9.11110044e-01
2.89844394e-01 -9.74544764e-01 4.42696661e-01 7.87593007e-01
-9.78927463e-02 -3.77446085e-01 -1.15033352e+00 -7.76841223e-01
-3.55617523e-01 -4.16024268e-01 7.11273313e-01 1.18571103e+00
4.06036019e-01 9.78993848e-02 1.07371911e-01 5.37490010e-01
2.96261877e-01 1.22489013e-01 4.50874716e-01 -1.21804893e+00
1.60047978e-01 -5.05358577e-01 -9.65098202e-01 -1.09172988e+00
9.21975747e-02 -8.53066146e-01 2.77446002e-01 -1.90142047e+00
-1.38253361e-01 -5.20738065e-01 -2.70873666e-01 5.81390023e-01
6.54317379e-01 1.95515350e-01 4.69098508e-01 5.10095298e-01
-8.12018931e-01 4.29559171e-01 1.14796114e+00 -8.04350525e-02
-4.28467125e-01 -3.84450287e-01 -6.27250552e-01 8.01667631e-01
8.77295852e-01 3.91684327e-04 -4.40061331e-01 -9.10635650e-01
2.38247931e-01 -7.08430558e-02 7.48768330e-01 -1.32069075e+00
6.92781270e-01 -4.75809515e-01 3.72009784e-01 -3.74388725e-01
6.81093872e-01 -1.22814476e+00 1.90018415e-01 3.41419935e-01
-2.35963359e-01 4.94909257e-01 -9.77660790e-02 3.50611925e-01
5.44333234e-02 -2.72965193e-01 6.58271015e-01 -3.75890046e-01
-1.42977238e+00 -5.28550055e-03 -7.90844798e-01 -2.91172951e-01
1.49416423e+00 -6.14461064e-01 -7.59111524e-01 -1.92194924e-01
-6.67807162e-01 9.34579968e-02 1.13456631e+00 6.25204861e-01
6.56233430e-01 -1.42677248e+00 -1.35667771e-01 3.04957122e-01
4.50052977e-01 -2.59645939e-01 1.86682031e-01 7.22121358e-01
-8.00619602e-01 6.55144334e-01 -6.29978240e-01 -6.08509839e-01
-8.10416460e-01 8.14719439e-01 5.11907458e-01 4.34496760e-01
-5.24595439e-01 5.02520859e-01 4.70847577e-01 -4.82238054e-01
3.23456615e-01 -3.65916848e-01 -5.30232847e-01 -1.34451181e-01
6.39180183e-01 3.28384697e-01 -5.60032390e-02 -1.11805773e+00
-9.23946023e-01 6.98533237e-01 4.04102802e-01 -4.24206793e-01
1.06433785e+00 -3.73117119e-01 -2.27555454e-01 7.34973967e-01
6.62753284e-01 -2.78441757e-01 -1.22535419e+00 -9.19775516e-02
4.16890979e-02 -4.95590419e-01 1.16185471e-02 -7.61890769e-01
-9.20152009e-01 7.63902843e-01 7.83987105e-01 2.16440558e-01
8.66251171e-01 2.43423358e-01 8.70854929e-02 3.73500466e-01
1.10698235e+00 -1.11537778e+00 -1.76382869e-01 6.78684473e-01
1.27785194e+00 -9.69382823e-01 -3.52876097e-01 -2.85804033e-01
-6.90103769e-01 7.83468068e-01 6.15408421e-01 -2.42223024e-01
5.38797796e-01 1.56726241e-01 3.27974468e-01 -2.02912390e-01
-4.55433905e-01 -4.05083537e-01 1.13930933e-01 1.32205951e+00
8.42018649e-02 8.64513516e-02 3.07102893e-02 2.91470647e-01
-3.25057566e-01 -1.26209661e-01 4.52784479e-01 1.11048603e+00
-5.53033054e-01 -5.75521052e-01 -5.10597706e-01 -2.11986870e-01
3.68025184e-01 2.30352461e-01 -5.46762526e-01 5.81895590e-01
6.36942089e-02 1.06331837e+00 3.22287887e-01 -3.97286892e-01
3.88301641e-01 -1.90450385e-01 7.04948306e-01 -4.64715838e-01
-2.23480687e-01 -5.58771074e-01 -2.34679624e-01 -7.54527986e-01
-4.11293775e-01 -7.80151248e-01 -1.45265317e+00 -1.80517197e-01
4.22013216e-02 -2.21181184e-01 1.04696405e+00 8.92908216e-01
5.34176171e-01 3.83246720e-01 2.60442048e-01 -1.31592607e+00
5.30517772e-02 -4.44571614e-01 -8.65859866e-01 5.24973916e-03
4.85819936e-01 -9.97547746e-01 -1.58616737e-01 -5.20344265e-02] | [7.47868537902832, -1.8831449747085571] |
59d626a5-c056-4542-8fb8-04868fc1f21d | predicting-skull-fractures-via-cnn-with | 2208.06756 | null | https://arxiv.org/abs/2208.06756v1 | https://arxiv.org/pdf/2208.06756v1.pdf | Predicting skull fractures via CNN with classification algorithms | Computer Tomography (CT) images have become quite important to diagnose diseases. CT scan slice contains a vast amount of data that may not be properly examined with the requisite precision and speed using normal visual inspection. A computer-assisted skull fracture classification expert system is needed to assist physicians. Convolutional Neural Networks (CNNs) are the most extensively used deep learning models for image categorization since most often time they outperform other models in terms of accuracy and results. The CNN models were then developed and tested, and several convolutional neural network (CNN) architectures were compared. ResNet50, which was used for feature extraction combined with a gradient boosted decision tree machine learning algorithm to act as a classifier for the categorization of skull fractures from brain CT scans into three fracture categories, had the best overall F1-score of 96%, Hamming Score of 95%, Balanced accuracy Score of 94% & ROC AUC curve of 96% for the classification of skull fractures. | ['Moqsadur Rahman', 'Tareque Rahman Ornob', 'Md Moniruzzaman Emon'] | 2022-08-14 | null | null | null | null | ['image-categorization'] | ['computer-vision'] | [-1.04578994e-01 -5.64872734e-02 1.50747895e-01 -3.32108796e-01
-4.68957722e-01 9.84264305e-04 1.01899713e-01 6.23398542e-01
-6.78883195e-01 5.07890940e-01 5.87236807e-02 -4.97308224e-01
-3.47452700e-01 -8.61472726e-01 -1.05032437e-01 -5.54879010e-01
-6.03662968e-01 6.76704288e-01 4.74088669e-01 -2.38322854e-01
1.42541930e-01 7.47785985e-01 -1.35651767e+00 5.30770123e-01
3.83707374e-01 1.22813141e+00 1.83090553e-01 6.79925144e-01
1.90965645e-02 9.74143445e-01 -5.38104951e-01 -1.07356347e-01
1.77825764e-01 1.66504204e-01 -1.04898250e+00 -2.16794237e-01
-2.58622110e-01 -2.19833270e-01 -4.39552993e-01 6.98687553e-01
7.88304687e-01 -1.12293623e-01 7.92638242e-01 -6.14185154e-01
-4.97823954e-01 3.71823370e-01 -4.24207598e-01 9.00896490e-01
7.51166344e-02 1.29556581e-01 5.80520332e-01 -9.90969658e-01
2.83293992e-01 7.12752759e-01 8.61741304e-01 4.15413827e-01
-6.79593861e-01 -8.43733788e-01 -7.82809138e-01 5.12861133e-01
-1.31956196e+00 1.99804679e-01 2.10822016e-01 -6.72517300e-01
9.80029225e-01 -1.85284745e-02 9.64151084e-01 6.13904417e-01
9.09570575e-01 2.25558668e-01 1.07782841e+00 -3.65163326e-01
2.51601726e-01 -2.03831702e-01 2.96166331e-01 1.15607810e+00
2.97773898e-01 2.40867034e-01 -1.13876000e-01 -8.13733935e-02
7.91067421e-01 2.89983273e-01 -2.10804328e-01 1.03873126e-01
-9.41220760e-01 1.03996634e+00 1.02108431e+00 5.98931253e-01
-5.62364995e-01 -1.22506499e-01 7.58758247e-01 3.98836374e-01
-1.51969176e-02 2.94154346e-01 -2.23309547e-01 2.05043793e-01
-7.14230061e-01 4.99711670e-02 2.43070990e-01 3.95517796e-01
-6.65588602e-02 2.55059272e-01 -1.28730580e-01 9.19116914e-01
1.02038935e-01 2.06483185e-01 1.18064046e+00 -1.23079032e-01
8.43804181e-02 8.66581142e-01 -6.37561917e-01 -7.54952013e-01
-1.02021480e+00 -6.14971638e-01 -1.22755063e+00 5.83123624e-01
1.83098033e-01 8.36049393e-02 -1.37749505e+00 8.51151705e-01
-8.95497054e-02 -4.19183075e-01 -2.20458925e-01 9.44048345e-01
1.09590352e+00 1.69490412e-01 1.77416623e-01 2.91132927e-01
1.73414075e+00 -5.12634814e-01 -1.60031661e-01 6.59912303e-02
5.03019214e-01 -5.70490122e-01 6.88294351e-01 5.07546008e-01
-9.29095864e-01 -3.97349507e-01 -1.40447152e+00 2.08824232e-01
-4.21257943e-01 5.54451393e-03 5.40068924e-01 7.57878482e-01
-9.36755419e-01 5.94983697e-01 -1.10853601e+00 -3.77467036e-01
9.59906161e-01 7.26897240e-01 -7.33955264e-01 -4.41441648e-02
-9.24876034e-01 1.28029704e+00 4.33908761e-01 -1.60048425e-01
-9.49106157e-01 -3.97495955e-01 -5.95340848e-01 1.21685527e-01
-3.61773103e-01 -6.30391240e-01 1.12895191e+00 -6.85357869e-01
-1.00952446e+00 9.63027775e-01 4.47295994e-01 -8.10381532e-01
4.66060340e-01 2.05833510e-01 -4.04794216e-01 4.86156374e-01
4.91512641e-02 6.22531533e-01 5.47009408e-01 -6.54674590e-01
-6.84238374e-01 -6.69864595e-01 -3.04170728e-01 -9.13406461e-02
-1.21605299e-01 1.25024080e-01 3.56867075e-01 -4.68430430e-01
5.66228926e-01 -6.71398878e-01 -2.11025223e-01 -5.96691035e-02
-1.40229449e-01 -2.96960831e-01 8.11666608e-01 -7.59986579e-01
8.49383295e-01 -1.81397438e+00 -3.52026016e-01 3.18597376e-01
5.24753809e-01 4.84882385e-01 5.16516984e-01 5.62815592e-02
-8.17619741e-01 6.84534684e-02 -2.52059788e-01 3.58067423e-01
-5.32058656e-01 -3.18940058e-02 4.21629220e-01 6.19735301e-01
1.55224457e-01 6.78052306e-01 -5.71742356e-01 -3.54431421e-01
3.40217143e-01 4.60335642e-01 -1.55402660e-01 1.12767875e-01
5.98311722e-01 3.29704195e-01 -1.88046724e-01 6.96856141e-01
3.71347129e-01 -2.59818882e-01 -2.40687072e-01 -4.58658934e-02
3.49578828e-01 -1.75318331e-01 -7.49531507e-01 1.23591185e+00
-2.18143329e-01 8.34038317e-01 -3.69920671e-01 -9.42869842e-01
9.54746962e-01 6.44191384e-01 4.65614766e-01 -7.53150642e-01
6.89603686e-01 2.77877718e-01 4.09014344e-01 -9.80198741e-01
-2.46889040e-01 -6.10712290e-01 3.40558916e-01 3.47906619e-01
1.02213636e-01 -1.05114318e-02 -1.82578564e-01 -1.33533105e-01
1.33785546e+00 -7.56807327e-01 5.75457990e-01 -4.17456746e-01
3.88429582e-01 5.75721599e-02 9.71631557e-02 2.80768216e-01
-3.25787485e-01 9.86669004e-01 2.12304015e-02 -1.23691893e+00
-1.16139436e+00 -9.65097368e-01 -5.08227944e-01 4.54931289e-01
-3.67284894e-01 1.28022030e-01 -6.71299636e-01 -3.43937248e-01
-1.31936759e-01 2.55492717e-01 -8.64876151e-01 -2.44394794e-01
-4.82188195e-01 -8.83878589e-01 6.03819609e-01 7.18389153e-01
6.69576406e-01 -1.08084238e+00 -1.38545239e+00 9.02692154e-02
2.63887525e-01 -6.03771925e-01 1.73077002e-01 6.05054617e-01
-1.02761495e+00 -1.66338909e+00 -8.17507267e-01 -1.05435646e+00
6.56324625e-01 -9.30264592e-03 7.54994512e-01 5.04463196e-01
-8.22253823e-01 -8.98410156e-02 -3.91770273e-01 -7.08269835e-01
-3.72042954e-01 -2.33684089e-02 7.94522371e-03 -3.63784403e-01
4.03273493e-01 -6.27866030e-01 -9.34684813e-01 -1.33760557e-01
-1.01730525e+00 -1.88506797e-01 8.38334799e-01 1.06326175e+00
1.86548159e-01 1.32042289e-01 5.38886189e-01 -4.87957388e-01
7.11644351e-01 -5.87635159e-01 -7.77603164e-02 -1.21440664e-02
-7.01957166e-01 -2.60085195e-01 5.76357007e-01 2.13569193e-03
-3.05289388e-01 -1.02935292e-01 -5.03019333e-01 -1.61864251e-01
-2.45572075e-01 5.39738774e-01 6.67931974e-01 -1.98144481e-01
1.05219221e+00 8.70999023e-02 1.95333198e-01 -3.31810385e-01
-7.10480928e-01 9.67599511e-01 4.78894532e-01 -1.56912953e-02
2.16804430e-01 4.24811751e-01 1.63741097e-01 -6.69168353e-01
-3.86861920e-01 -5.15270352e-01 -7.56656528e-01 -2.50317872e-01
1.19349420e+00 -4.97371167e-01 -5.64237177e-01 4.31725472e-01
-8.77198696e-01 1.37654543e-01 7.08287731e-02 6.77532554e-01
-1.45662040e-01 2.80540586e-01 -6.48613989e-01 -5.42386174e-01
-8.74395728e-01 -1.35721684e+00 4.15656090e-01 9.89800468e-02
-3.21047723e-01 -6.63853645e-01 -1.80416882e-01 1.69736147e-01
7.20730841e-01 6.30214930e-01 1.45706260e+00 -1.01992357e+00
6.19211346e-02 -6.61732137e-01 -3.87955189e-01 3.95738780e-01
1.07303329e-01 -1.37210131e-01 -9.19465840e-01 -2.68725961e-01
-5.79530969e-02 -1.94062397e-01 8.22899640e-01 6.81986332e-01
1.33717453e+00 2.18594417e-01 -1.84019759e-01 5.46311319e-01
1.63792408e+00 7.67049551e-01 6.60934269e-01 7.69601226e-01
2.02420205e-01 2.19898090e-01 -2.59749562e-01 2.94742256e-01
2.10042939e-01 2.03851208e-01 6.79632306e-01 7.03132106e-03
-2.05371156e-01 3.95250410e-01 -3.86403233e-01 6.37174368e-01
-2.77159691e-01 1.73271254e-01 -1.56816494e+00 5.57873726e-01
-1.14273739e+00 -6.34376466e-01 -2.45072633e-01 1.93304956e+00
2.17978358e-01 3.36279333e-01 2.12333962e-01 9.04517829e-01
7.71077096e-01 -5.46498239e-01 -2.03894407e-01 -3.65858585e-01
1.57930609e-02 9.05410707e-01 4.54709530e-01 -2.31228098e-01
-1.20387304e+00 2.60489017e-01 6.79743433e+00 4.45926160e-01
-1.44963813e+00 3.09891492e-01 6.46608472e-01 8.09881464e-02
4.50462818e-01 -6.08641863e-01 8.14745426e-02 4.29603845e-01
9.58184659e-01 7.61868134e-02 -2.01081887e-01 7.45758593e-01
6.22802451e-02 -3.15879911e-01 -7.85660028e-01 1.07096052e+00
-2.04804465e-01 -1.40451002e+00 1.13559484e-01 -6.18081205e-02
3.04925829e-01 4.07338291e-01 5.30844443e-02 1.68914080e-01
3.03091407e-01 -1.53310120e+00 5.42721450e-01 1.63279384e-01
8.08888912e-01 -1.01589131e+00 1.43651366e+00 1.14400454e-01
-9.50104535e-01 -5.50105989e-01 -4.19044971e-01 -2.71930724e-01
-4.00170796e-02 2.98075140e-01 -1.09866977e+00 3.66992146e-01
1.26960337e+00 1.35922551e-01 -4.28930223e-01 1.56862617e+00
1.75498679e-01 5.66848695e-01 -1.56202853e-01 -1.56765375e-02
5.37649989e-01 4.57349181e-01 2.05485001e-01 1.15002310e+00
3.40009272e-01 3.76598895e-01 -2.34519895e-02 1.69208243e-01
1.40899897e-01 4.43578750e-01 -4.57646966e-01 5.22540212e-01
1.80741251e-01 9.93052244e-01 -1.28993702e+00 -2.29798123e-01
-2.67855972e-01 3.88507783e-01 1.88617304e-01 -3.28396559e-01
-5.75162888e-01 -4.92826730e-01 6.45838901e-02 3.84628505e-01
5.44409528e-02 9.22627449e-02 -6.06352687e-01 -5.80959737e-01
-2.52490252e-01 -6.56499028e-01 6.68411791e-01 -8.86467338e-01
-1.07853734e+00 1.18330812e+00 -1.70936152e-01 -1.18688416e+00
1.14054963e-01 -8.46500516e-01 -6.75024688e-01 5.54585934e-01
-1.26036739e+00 -8.54829967e-01 -5.53416967e-01 6.78948104e-01
5.39905548e-01 -6.03503406e-01 9.52783883e-01 1.98326617e-01
-3.45913559e-01 3.06370795e-01 -3.78577337e-02 7.31755733e-01
1.56909421e-01 -1.09128273e+00 -9.25875157e-02 4.89287615e-01
-4.58555222e-01 2.12746143e-01 4.01649117e-01 -4.03742611e-01
-6.67689204e-01 -1.10808539e+00 6.49307966e-01 5.86715750e-02
2.69445598e-01 4.42931801e-01 -8.62892210e-01 5.18789828e-01
2.49894261e-01 2.71916483e-02 8.14570010e-01 -4.73031372e-01
-3.38165492e-01 1.09515883e-01 -1.53106129e+00 9.34170280e-03
3.76170814e-01 -6.16618954e-02 -8.51297259e-01 3.37544709e-01
1.40753865e-01 -5.52211881e-01 -1.13115275e+00 6.32051408e-01
8.14239800e-01 -1.18514752e+00 1.00587451e+00 -8.61467957e-01
7.42379010e-01 9.96415094e-02 -1.23162955e-01 -1.14505577e+00
-5.75137496e-01 3.86458635e-01 6.23801589e-01 1.60445645e-01
4.32148635e-01 -6.39729917e-01 6.64492011e-01 3.27668846e-01
-3.16768438e-01 -1.21483469e+00 -1.15293407e+00 -6.48024023e-01
2.41211116e-01 -4.82616812e-01 6.66452467e-01 7.69604087e-01
1.37837127e-01 8.62609297e-02 2.69236416e-01 -3.28032598e-02
2.77964294e-01 -3.80871832e-01 1.92177314e-02 -1.54043877e+00
2.26963699e-01 -7.76321352e-01 -1.33719945e+00 1.71286315e-01
-3.81636202e-01 -1.19278479e+00 -3.94862622e-01 -1.65721762e+00
5.05591750e-01 -3.75202596e-01 -5.32703519e-01 4.74095196e-01
3.22732329e-01 3.89578938e-01 -7.34796077e-02 2.47798562e-01
9.74961445e-02 1.45499125e-01 1.06001794e+00 -2.48444483e-01
2.76431620e-01 2.03307748e-01 -3.67141634e-01 7.44966209e-01
9.02483106e-01 -7.11852551e-01 -2.40589246e-01 -4.75530028e-01
-9.98079702e-02 1.58184752e-01 2.91752189e-01 -1.57959771e+00
2.49164075e-01 3.56954485e-01 8.51309121e-01 -5.54533184e-01
1.28193006e-01 -8.14850628e-01 3.19399089e-01 1.11294878e+00
-3.17201376e-01 4.49734271e-01 1.59571707e-01 2.37218320e-01
-4.53574419e-01 -3.51457566e-01 1.29084420e+00 -5.90084076e-01
-4.96369153e-01 5.37249327e-01 -8.15734088e-01 -3.81651103e-01
1.16380787e+00 -6.83085799e-01 3.27488109e-02 -2.69738913e-01
-1.04833937e+00 -1.59531385e-01 1.63031276e-02 9.29303989e-02
1.06389618e+00 -1.17381215e+00 -7.94083416e-01 2.69488484e-01
2.52910405e-01 1.60492603e-02 3.09606582e-01 9.33920979e-01
-1.32723725e+00 3.83468956e-01 -7.84511149e-01 -7.78305113e-01
-1.38529229e+00 2.09654182e-01 7.10087657e-01 -2.03117087e-01
-1.01273525e+00 1.03364694e+00 -3.17089289e-01 -2.13693753e-01
1.42329276e-01 -4.75756586e-01 -5.48954427e-01 -2.35188052e-01
6.93970263e-01 4.47224468e-01 8.04240108e-01 -6.24655783e-01
-3.98221403e-01 5.49240649e-01 -9.07165036e-02 2.28713483e-01
1.79525304e+00 4.56878811e-01 -5.99084832e-02 3.96845862e-02
1.33701324e+00 -9.01325583e-01 -2.95806557e-01 -5.47232404e-02
1.25349343e-01 -2.26871148e-01 3.55453879e-01 -1.01628268e+00
-1.36106288e+00 1.17372000e+00 1.29283130e+00 2.60509193e-01
1.07272375e+00 -1.62430659e-01 8.25192094e-01 3.70017797e-01
3.47217381e-01 -9.01971757e-01 4.02707756e-02 4.17892367e-01
7.49993384e-01 -1.36520755e+00 -8.56415927e-02 1.12052262e-01
-5.04519939e-01 1.78220630e+00 4.16353703e-01 -4.66679543e-01
1.18640924e+00 3.19884300e-01 7.01924190e-02 -8.52728724e-01
-5.71000576e-01 -9.70435590e-02 1.23069301e-01 6.77089334e-01
5.53106010e-01 2.05721110e-01 -1.92557782e-01 6.41405523e-01
-5.00971675e-01 2.56039649e-01 3.22843045e-01 1.10520899e+00
-9.72943783e-01 -3.91107768e-01 -3.87763828e-01 9.44765747e-01
-9.50514734e-01 1.05577089e-01 -1.93004757e-02 1.05270767e+00
1.68228865e-01 8.06239247e-01 4.69859801e-02 -6.88286602e-01
2.19891712e-01 1.57206744e-01 2.98180342e-01 -4.82321441e-01
-9.40061748e-01 -3.91570568e-01 -2.25409538e-01 -9.22445208e-02
-7.46101290e-02 -3.53541881e-01 -1.42909563e+00 -4.11210150e-01
-2.90217549e-01 5.71904406e-02 8.56250405e-01 1.17707610e+00
-1.15209572e-01 8.05882454e-01 4.21444684e-01 -4.70912784e-01
-2.87082285e-01 -1.24304378e+00 -5.69427490e-01 2.60947704e-01
3.24375182e-01 -8.30418468e-01 6.54345676e-02 -1.24793693e-01] | [14.946714401245117, -2.4079439640045166] |
530d3288-5a9c-4fa2-acba-31dc52a8faf7 | seeing-through-deception-a-computational | null | null | https://aclanthology.org/W12-0403 | https://aclanthology.org/W12-0403.pdf | Seeing through Deception: A Computational Approach to Deceit Detection in Written Communication | null | ["Rafael Valencia-Garc{\\'\\i}a", "{\\'A}ngela Almela", 'Pascual Cantos'] | 2012-04-01 | null | null | null | ws-2012-4 | ['deception-detection'] | ['miscellaneous'] | [-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] | [-7.181015491485596, 3.5639233589172363] |
03e9031a-6e4d-4cc2-a790-af9bb4c0f04f | reflection-invariant-and-symmetry-detection | 1705.10768 | null | http://arxiv.org/abs/1705.10768v2 | http://arxiv.org/pdf/1705.10768v2.pdf | Reflection Invariant and Symmetry Detection | Symmetry detection and discrimination are of fundamental meaning in science,
technology, and engineering. This paper introduces reflection invariants and
defines the directional moment to detect symmetry for shape analysis and object
recognition. And it demonstrates that detection of reflection symmetry can be
done in a simple way by solving a trigonometric system derived from the
directional moment, and discrimination of reflection symmetry can be achieved
by application of the reflection invariants in 2D and 3D. Rotation symmetry can
also be determined based on that.The experiments in 2D and 3D, including the
regular triangle, the square, and the five Platonic objects, show that all the
reflection lines or planes can be deterministically found using directional
moments up to order six. This result can be used to simplify the efforts of
symmetry detection in research areas, such as protein structure, model
retrieval, inverse engineering, and machine vision etc. | ['Erbo Li', 'Hua Li'] | 2017-05-30 | null | null | null | null | ['symmetry-detection'] | ['computer-vision'] | [ 4.77854759e-01 -5.09078465e-02 -4.51236665e-02 -5.86202927e-02
1.00490727e-01 -4.73795384e-01 6.74938440e-01 -2.21103340e-01
-1.06121622e-01 5.05032659e-01 1.25928950e-02 -5.32099128e-01
-4.77831870e-01 -8.42166781e-01 -1.91633239e-01 -7.44712114e-01
-4.83219512e-03 7.07418561e-01 3.47666979e-01 -3.34701687e-01
9.94959950e-01 1.13671958e+00 -1.38082826e+00 1.55497752e-02
4.71186578e-01 8.16208959e-01 -1.40479937e-01 4.21541333e-01
-2.29266342e-02 6.57468429e-03 -2.43741393e-01 4.10893448e-02
3.56992662e-01 -5.46248972e-01 -7.77437210e-01 1.36809960e-01
3.09505984e-02 -1.14576899e-01 3.67787361e-01 9.27274168e-01
4.41290677e-01 -7.34175444e-02 1.13345385e+00 -9.39116120e-01
-5.00755608e-01 -1.87804386e-01 -6.30475163e-01 -3.90431941e-01
9.50508833e-01 -4.74908382e-01 6.93770826e-01 -9.66360509e-01
8.37676108e-01 1.04493117e+00 4.91959393e-01 3.73485267e-01
-7.40721941e-01 -5.02390981e-01 -6.04065597e-01 2.27117598e-01
-1.34985232e+00 -1.32818654e-01 9.68115330e-01 -4.43488806e-01
8.04269910e-01 6.80203378e-01 5.67658901e-01 2.98676461e-01
4.28228080e-01 1.66623875e-01 1.07837737e+00 -8.24937403e-01
-4.27198671e-02 -2.54461855e-01 1.43031523e-01 9.08127069e-01
4.49127495e-01 -1.94526762e-01 8.33760053e-02 -2.23905757e-01
1.07766545e+00 1.11113526e-01 -2.45544493e-01 -6.16161883e-01
-1.24903023e+00 6.02872849e-01 -5.83491884e-02 3.66568893e-01
-1.71029076e-01 -8.31336558e-01 1.95940607e-03 2.02612072e-01
-7.97061920e-02 6.66641235e-01 -2.00191066e-01 -1.43448055e-01
-5.27672708e-01 1.38245359e-01 8.38800251e-01 7.04291701e-01
5.89754105e-01 -1.09851643e-01 3.39031607e-01 6.74970627e-01
2.34041229e-01 7.05204070e-01 3.15357298e-01 -6.96470201e-01
-1.77514255e-02 8.82360876e-01 1.26799801e-02 -1.03997815e+00
-6.58657849e-01 2.96322227e-01 -8.37601602e-01 4.19879466e-01
3.73338997e-01 4.97383088e-01 -7.09924221e-01 7.66651630e-01
3.80960763e-01 -2.38446802e-01 1.16110772e-01 7.52079248e-01
1.02306569e+00 3.66940320e-01 -9.28139687e-01 -3.70325804e-01
1.59639561e+00 -5.09454846e-01 -3.55130643e-01 5.74646413e-01
5.01812041e-01 -1.20696175e+00 3.96957934e-01 5.82312942e-01
-7.48013914e-01 -2.35770389e-01 -1.19582653e+00 1.28529184e-02
-1.55292004e-01 3.03912163e-01 5.97764492e-01 5.63055038e-01
-2.94524729e-01 6.91689074e-01 -7.16053426e-01 -2.65547931e-01
-1.12293310e-01 3.29240203e-01 -5.67418873e-01 2.58956283e-01
-5.82941771e-01 8.71397793e-01 5.45410328e-02 1.64782032e-01
1.17636710e-01 -2.31885850e-01 -5.06197333e-01 -2.77436197e-01
-9.27999318e-02 -2.89172173e-01 6.93228662e-01 -3.87298405e-01
-1.58733428e+00 1.16954803e+00 -4.37766403e-01 3.00483480e-02
4.09761727e-01 3.60100925e-01 -5.66319942e-01 2.22459376e-01
1.21734224e-01 -1.87034085e-01 7.40077376e-01 -8.77721488e-01
-1.64905488e-01 -6.51214778e-01 -1.96079344e-01 -3.80685702e-02
2.31990427e-01 2.87197590e-01 1.49997115e-01 -9.83246341e-02
1.18625164e+00 -9.59294379e-01 8.83029774e-02 -1.28152758e-01
-5.18809795e-01 -3.31764013e-01 1.19488680e+00 -6.94010437e-01
7.61875570e-01 -1.86609447e+00 -1.56867787e-01 8.58870506e-01
1.00191355e-01 1.18758388e-01 1.07882038e-01 6.23884380e-01
-4.72676039e-01 2.21565720e-02 1.64159462e-02 8.60309303e-01
-1.66787028e-01 -1.46855935e-01 -2.56774694e-01 7.66088486e-01
1.66220933e-01 4.59503055e-01 -4.41713929e-01 -2.71367043e-01
1.36064515e-01 5.32812953e-01 -3.86598080e-01 -2.48358980e-01
3.34936589e-01 3.37128162e-01 -7.64801562e-01 6.20065331e-01
1.00191975e+00 -6.93516135e-02 2.44701564e-01 -5.10872424e-01
-5.94646692e-01 4.62961137e-01 -1.44421101e+00 4.66648847e-01
3.53230275e-02 4.49986428e-01 -4.11252528e-01 -1.08668792e+00
1.56911039e+00 3.62546712e-01 7.48288035e-01 -9.72646534e-01
-1.17239952e-02 3.05846244e-01 2.82633573e-01 -3.49120945e-01
3.73533845e-01 7.49248639e-02 3.43134463e-01 6.70269608e-01
-5.83978176e-01 -3.62625897e-01 2.32277229e-01 -4.67419863e-01
3.89884323e-01 -3.69077399e-02 5.79552948e-01 -3.50099117e-01
1.03847075e+00 -1.84334546e-01 4.02903408e-01 1.61660358e-01
3.15240830e-01 5.54702997e-01 6.26482844e-01 -8.36477935e-01
-1.30508971e+00 -1.05949676e+00 -5.59441864e-01 3.40742439e-01
2.55003393e-01 -3.27527463e-01 -4.41794783e-01 -1.03110284e-01
-8.34580585e-02 4.89715338e-02 -1.20903112e-01 -1.96759060e-01
-8.44409466e-01 -8.00778687e-01 1.12078600e-01 9.84866396e-02
6.07180357e-01 -9.62747693e-01 -6.38720036e-01 -2.30123058e-01
7.54315257e-02 -9.17760730e-01 -3.80246900e-02 -1.59708157e-01
-1.08066428e+00 -1.47663057e+00 -7.27688730e-01 -9.53891277e-01
1.03870773e+00 4.98166502e-01 6.10480428e-01 1.66889191e-01
-4.71633703e-01 2.39847422e-01 -2.02408224e-01 -2.67389685e-01
-4.38610554e-01 -3.65603119e-01 2.86714107e-01 -1.42493472e-01
4.42635596e-01 -7.02213585e-01 -3.60826552e-01 1.00528336e+00
-5.61158955e-01 -5.68327419e-02 2.05667973e-01 5.28628170e-01
6.48413360e-01 3.32243204e-01 9.69779268e-02 -3.58240724e-01
6.12868607e-01 4.41400498e-01 -9.06380296e-01 2.93319166e-01
-3.83130312e-01 5.12166619e-01 4.35441583e-01 -1.66236311e-01
-6.56919718e-01 9.97909158e-02 -2.92858742e-02 1.59527898e-01
4.30380516e-02 1.47051454e-01 -1.54066086e-01 -7.56619155e-01
4.92229939e-01 4.99231100e-01 1.81490585e-01 -5.16266108e-01
-2.43984580e-01 7.34303713e-01 2.31817186e-01 -6.91740274e-01
7.37013876e-01 6.56942606e-01 8.92635763e-01 -1.56357789e+00
-1.65364057e-01 -4.30744439e-01 -7.50044048e-01 -1.60970122e-01
6.27147019e-01 -2.52421409e-01 -1.20774472e+00 4.52705652e-01
-1.35363424e+00 5.56358576e-01 1.74229279e-01 7.41249323e-01
-6.53097630e-01 9.28258181e-01 -3.76517102e-02 -7.75906265e-01
-3.06367248e-01 -1.06682169e+00 1.07009280e+00 7.22923651e-02
-1.56520069e-01 -7.88563192e-01 9.52398181e-02 2.06924960e-01
1.32568508e-01 2.10393846e-01 1.23821080e+00 -5.75236201e-01
-6.51976109e-01 -3.17984968e-01 -1.47707596e-01 8.20665359e-02
2.77160108e-01 6.70810580e-01 -4.10895586e-01 2.23047985e-03
2.27834746e-01 2.22644493e-01 6.15160644e-01 1.45483792e-01
8.72981369e-01 -9.07517076e-02 -2.68481702e-01 5.72162569e-01
1.07463002e+00 5.97625315e-01 9.75445628e-01 4.08756822e-01
3.62569481e-01 5.31220198e-01 4.65679646e-01 3.83841991e-01
-9.00357068e-02 7.72451222e-01 1.66385978e-01 1.62679292e-02
1.97528228e-01 5.32431267e-02 5.47162406e-02 8.57790530e-01
-1.07434273e+00 5.26907861e-01 -8.22821319e-01 -2.00582817e-01
-1.45450354e+00 -1.35347664e+00 -4.55467016e-01 2.63704181e+00
3.20043623e-01 1.43000856e-02 8.46829936e-02 6.61950290e-01
6.89630091e-01 -1.58076286e-01 -5.78527562e-02 -6.84552610e-01
-2.46723175e-01 2.00716391e-01 4.06010896e-01 5.90456665e-01
-7.68266857e-01 4.65222776e-01 7.47771645e+00 5.70026994e-01
-1.60197604e+00 -6.45448685e-01 -9.39512178e-02 7.33434558e-01
-3.99283409e-01 2.82413781e-01 -9.34910715e-01 2.27971330e-01
-1.38630152e-01 -1.56020671e-01 1.78433999e-01 6.44649804e-01
1.68043599e-01 -1.81972027e-01 -8.89619052e-01 1.09085894e+00
7.63084693e-03 -1.13935697e+00 3.96663010e-01 2.63745815e-01
7.01416910e-01 -4.23336774e-01 -1.39727429e-01 -4.31484967e-01
-5.99592447e-01 -8.91669989e-01 -1.37691638e-02 6.61039352e-01
5.35180032e-01 -6.96112871e-01 4.73097444e-01 3.08244437e-01
-1.18386817e+00 3.08619827e-01 -6.62741065e-01 -3.05450618e-01
-6.47536069e-02 5.43011308e-01 -1.14545453e+00 5.80009401e-01
1.41490668e-01 3.51262629e-01 -1.77722782e-01 1.04598463e+00
-1.96326822e-01 4.57656048e-02 -5.47573626e-01 -3.62844765e-01
-1.12852417e-01 -1.15422785e+00 6.25917017e-01 7.16638088e-01
3.89752895e-01 5.15033677e-02 -1.74109727e-01 6.25459313e-01
5.75786054e-01 3.67072314e-01 -5.64317107e-01 -6.30703047e-02
3.60537954e-02 9.56988633e-01 -1.12411916e+00 8.25962424e-02
-1.80796504e-01 6.43003702e-01 -5.11645496e-01 2.82734931e-01
-5.52392602e-01 -7.69780993e-01 2.86902815e-01 2.83355176e-01
4.11096513e-01 -6.09466732e-01 -8.36915076e-02 -1.21553087e+00
3.06325287e-01 -6.73634827e-01 -4.29781117e-02 -6.42963409e-01
-5.67167282e-01 3.59174252e-01 6.56153411e-02 -1.29147673e+00
-3.62070471e-01 -1.31974685e+00 -8.94373417e-01 6.57793522e-01
-8.83149445e-01 -7.74275482e-01 -1.19531425e-02 5.42782068e-01
-2.87083983e-01 -3.79621297e-01 8.95768702e-01 3.59531902e-02
-9.83535722e-02 3.31194639e-01 3.87923449e-01 2.11853415e-01
3.87148798e-01 -8.10668766e-01 2.10648894e-01 4.03573006e-01
3.74764092e-02 5.59475958e-01 7.10844994e-01 -5.16550660e-01
-1.49624491e+00 -7.30471313e-02 1.34385717e+00 -1.43084288e-01
3.66126060e-01 -1.42821565e-01 -5.04330218e-01 2.47933924e-01
-5.32843471e-01 -4.05106455e-01 3.39863598e-01 -6.57269359e-02
-3.86108488e-01 -1.10848986e-01 -1.26393843e+00 7.09013045e-01
1.10336161e+00 -5.36934257e-01 -7.53651679e-01 7.43753731e-01
6.66107163e-02 -1.90604270e-01 -9.53471899e-01 7.06594408e-01
1.08435631e+00 -1.10905278e+00 1.09895563e+00 -6.92997217e-01
1.48980930e-01 -4.12024677e-01 5.34468889e-02 -5.28968692e-01
-3.05750042e-01 -6.22345328e-01 6.12255573e-01 5.67222595e-01
2.97829181e-01 -1.18516362e+00 8.21616590e-01 2.40350530e-01
1.45895839e-01 -5.99029362e-01 -1.07414544e+00 -1.03040636e+00
-2.96810448e-01 6.55449703e-02 6.20110691e-01 7.55735576e-01
5.34763075e-02 3.09222043e-01 -4.83566821e-02 3.98749150e-02
4.95972455e-01 8.44551325e-01 8.18667948e-01 -1.61014533e+00
-5.03269769e-02 -3.66581440e-01 -1.17524207e+00 -1.15884089e+00
-1.57837868e-02 -7.27955937e-01 -6.00321949e-01 -1.28149891e+00
-9.51658096e-03 -2.36614332e-01 1.72007620e-01 1.92844808e-01
7.97076404e-01 2.97759622e-01 -1.63478792e-01 2.12509021e-01
-1.24500908e-01 3.39358240e-01 1.83886087e+00 1.22330971e-01
-9.43846852e-02 3.91541362e-01 -3.22631657e-01 1.13252604e+00
8.06200147e-01 -1.97344288e-01 4.98421863e-02 3.47143114e-01
5.73170185e-01 -1.02661461e-01 1.66026682e-01 -8.55494857e-01
3.93856466e-02 -2.97171682e-01 3.93539518e-01 -8.36140454e-01
3.07593077e-01 -8.89109790e-01 1.92323297e-01 5.96478581e-01
3.84582639e-01 2.08354473e-01 -3.97365689e-01 -1.68833107e-01
-1.33419082e-01 -6.68872535e-01 6.81260407e-01 -6.45588040e-02
-3.71070504e-01 3.56251970e-02 -2.21403450e-01 -1.88480705e-01
8.52386534e-01 -8.67719293e-01 -3.18513542e-01 -2.24182174e-01
-6.37474835e-01 -4.09734190e-01 6.19593382e-01 5.34084477e-02
9.24438477e-01 -1.40513694e+00 -5.35708904e-01 9.33215201e-01
1.70011688e-02 -1.14718780e-01 -1.51268899e-01 1.03040564e+00
-1.13665617e+00 6.39226377e-01 -6.60280585e-01 -7.45016515e-01
-1.72881520e+00 3.49621445e-01 3.79816085e-01 -1.66232344e-02
-2.84806162e-01 4.93370704e-02 7.01784566e-02 -4.89569664e-01
-1.46287918e-01 -3.59750450e-01 -3.77653360e-01 -2.60440707e-01
5.03415108e-01 6.92663133e-01 1.65715948e-01 -8.23394239e-01
-5.24267852e-01 1.69061029e+00 1.08530574e-01 1.12603299e-01
1.21540272e+00 4.62255448e-01 -5.63899517e-01 -7.75406957e-02
1.27289069e+00 5.18390059e-01 -3.69736671e-01 1.97984546e-01
-1.85164243e-01 -2.96529502e-01 -7.80623138e-01 -2.46625751e-01
-4.88462240e-01 7.74294794e-01 8.63955468e-02 5.42834878e-01
9.30525839e-01 -9.01463032e-02 3.70406479e-01 9.09251273e-01
3.75198394e-01 -9.24516082e-01 -2.60350227e-01 8.50340843e-01
1.17450309e+00 -7.88789511e-01 3.90716642e-01 -7.38554776e-01
-1.92479733e-02 1.95609260e+00 1.51410207e-01 -4.04136658e-01
8.91671956e-01 1.69917211e-01 -1.34483814e-01 -9.09285098e-02
-1.92785159e-01 7.51590207e-02 7.21118689e-01 5.64292014e-01
6.33690536e-01 6.38217181e-02 -1.02022493e+00 4.33483645e-02
-3.61771941e-01 -1.69924840e-01 5.83317876e-01 7.84725487e-01
-8.95372868e-01 -1.38185358e+00 -7.73486853e-01 1.33207127e-01
-2.12086111e-01 1.97115451e-01 -8.44192505e-01 8.63189340e-01
-3.33273672e-02 3.97070915e-01 2.36664250e-01 -1.94756821e-01
3.81084830e-01 -1.69877484e-01 9.38264787e-01 3.73532213e-02
1.89807370e-01 -1.38839453e-01 7.30856434e-02 -3.63761902e-01
-6.44095480e-01 -3.42904747e-01 -1.43297410e+00 -2.45544896e-01
-3.07718813e-01 4.54749107e-01 8.55157971e-01 8.04046452e-01
1.20222181e-01 -2.47872010e-01 8.74422252e-01 -5.93519151e-01
-6.23357296e-01 -5.84218264e-01 -7.47515261e-01 2.89851427e-01
6.86085373e-02 -7.23053873e-01 -5.48943281e-01 -2.84124762e-01] | [9.177483558654785, -1.7959234714508057] |
ebb9f790-df51-4d9e-9c42-7bd9b7a0c944 | evolutionary-reinforcement-learning-a-survey | 2303.04150 | null | https://arxiv.org/abs/2303.04150v3 | https://arxiv.org/pdf/2303.04150v3.pdf | Evolutionary Reinforcement Learning: A Survey | Reinforcement learning (RL) is a machine learning approach that trains agents to maximize cumulative rewards through interactions with environments. The integration of RL with deep learning has recently resulted in impressive achievements in a wide range of challenging tasks, including board games, arcade games, and robot control. Despite these successes, there remain several crucial challenges, including brittle convergence properties caused by sensitive hyperparameters, difficulties in temporal credit assignment with long time horizons and sparse rewards, a lack of diverse exploration, especially in continuous search space scenarios, difficulties in credit assignment in multi-agent reinforcement learning, and conflicting objectives for rewards. Evolutionary computation (EC), which maintains a population of learning agents, has demonstrated promising performance in addressing these limitations. This article presents a comprehensive survey of state-of-the-art methods for integrating EC into RL, referred to as evolutionary reinforcement learning (EvoRL). We categorize EvoRL methods according to key research fields in RL, including hyperparameter optimization, policy search, exploration, reward shaping, meta-RL, and multi-objective RL. We then discuss future research directions in terms of efficient methods, benchmarks, and scalable platforms. This survey serves as a resource for researchers and practitioners interested in the field of EvoRL, highlighting the important challenges and opportunities for future research. With the help of this survey, researchers and practitioners can develop more efficient methods and tailored benchmarks for EvoRL, further advancing this promising cross-disciplinary research field. | ['Yaochu Jin', 'Ran Cheng', 'Hui Bai'] | 2023-03-07 | null | null | null | null | ['board-games'] | ['playing-games'] | [-2.51934499e-01 -3.10751021e-01 -3.93490762e-01 2.22146705e-01
-5.88508368e-01 -3.92440557e-01 3.43916982e-01 -1.51008070e-02
-8.10619831e-01 1.35112917e+00 -3.10918272e-01 -8.19650665e-03
-5.48608243e-01 -7.07652807e-01 -4.80862916e-01 -1.00214648e+00
-4.85315919e-01 6.23002708e-01 -1.94789786e-02 -6.04279339e-01
5.06973863e-01 4.23158169e-01 -1.91209972e+00 -3.76705766e-01
1.22547317e+00 8.74566972e-01 3.24662030e-01 5.24460077e-01
1.71836153e-01 8.21445644e-01 -9.57277536e-01 -1.83433488e-01
4.16690297e-02 -5.38894773e-01 -5.16630769e-01 -3.42405081e-01
-5.74750483e-01 -9.72916931e-02 -6.66460842e-02 9.54474330e-01
1.02297258e+00 7.05643356e-01 3.72763753e-01 -1.59000683e+00
-7.07370341e-01 4.84499544e-01 -5.59006274e-01 1.16353109e-01
1.98876098e-01 4.77112532e-01 9.94347751e-01 -6.06192887e-01
3.19364607e-01 1.25353324e+00 5.28355718e-01 8.59963179e-01
-6.16325319e-01 -6.57192945e-01 4.17362720e-01 5.28041661e-01
-1.01232946e+00 3.21155265e-02 5.39422631e-01 1.75418556e-01
1.37222469e+00 1.65866017e-01 1.11190486e+00 1.07458496e+00
2.68253028e-01 1.15008795e+00 1.15190113e+00 -4.23795670e-01
8.20330262e-01 -1.64901897e-01 -7.72239208e-01 5.33102512e-01
5.69774732e-02 9.29034591e-01 -5.46322823e-01 -2.21133456e-01
8.67226958e-01 -4.60667580e-01 1.72112986e-01 -4.29028720e-01
-9.88797486e-01 1.31521654e+00 2.94935375e-01 -3.78660072e-04
-5.32872260e-01 4.42108780e-01 4.05161142e-01 3.82035732e-01
6.18088134e-02 1.03487515e+00 -5.09030163e-01 -6.60679579e-01
-4.53662902e-01 6.65154934e-01 7.14187622e-01 7.41593599e-01
3.97342831e-01 6.27679467e-01 -2.42140487e-01 1.06117404e+00
2.44210958e-01 5.40891707e-01 6.82431817e-01 -1.36118960e+00
2.57354379e-01 2.76120007e-01 4.28574860e-01 -5.61706483e-01
-5.54578900e-01 -6.48754954e-01 -6.12289250e-01 8.39309573e-01
-5.58405332e-02 -6.86850309e-01 -3.97732556e-01 1.79234612e+00
4.52451974e-01 7.46203512e-02 4.14968640e-01 9.63649750e-01
4.99454170e-01 6.32381976e-01 -2.02889387e-02 -3.84931147e-01
8.58195782e-01 -1.40667641e+00 -5.31203687e-01 -4.16400284e-01
2.53439397e-01 -4.23592538e-01 1.00227511e+00 5.29965997e-01
-1.38654733e+00 -2.53647685e-01 -1.03457105e+00 6.55566156e-01
-3.17624837e-01 -9.85072553e-02 9.84292567e-01 6.08728945e-01
-1.10567832e+00 8.00878942e-01 -8.59592855e-01 -7.43123963e-02
3.55674982e-01 6.31880581e-01 5.90235651e-01 2.81625807e-01
-1.52100122e+00 1.19033802e+00 4.20232415e-01 1.08036019e-01
-1.04707563e+00 -3.82049203e-01 -6.78070068e-01 -8.24320596e-03
8.03043902e-01 -3.94538701e-01 1.54970610e+00 -9.42558706e-01
-2.16304040e+00 2.58077592e-01 4.95344996e-01 -3.94336730e-01
6.31694674e-01 -1.43789679e-01 -3.93357188e-01 -2.56222874e-01
-1.61573008e-01 7.54796088e-01 5.73125362e-01 -1.02758372e+00
-1.00012863e+00 -1.16578639e-01 1.02482729e-01 8.40500653e-01
-2.10890844e-01 6.58268109e-02 -2.00228795e-01 -6.91828072e-01
-7.17878163e-01 -9.60377276e-01 -7.33162642e-01 -5.16278923e-01
1.15182482e-01 -4.72909629e-01 4.41975206e-01 1.85746431e-01
1.16637027e+00 -1.67294204e+00 4.30038631e-01 9.01007280e-02
-1.03694595e-01 3.56703132e-01 -4.83382702e-01 6.18325174e-01
4.68389452e-01 -1.02372348e-01 -6.10616282e-02 -5.92941307e-02
3.27474654e-01 3.56562644e-01 -1.56798184e-01 1.39271900e-01
1.47960618e-01 1.26190305e+00 -1.43209481e+00 -2.12338611e-01
1.40973195e-01 2.01495424e-01 -4.55016792e-01 2.66055793e-01
-5.67361355e-01 5.30660689e-01 -7.16839314e-01 9.01121616e-01
1.56989306e-01 -1.44043103e-01 1.17049634e-01 7.00227320e-01
-4.51111913e-01 -1.04313307e-01 -1.22995675e+00 1.41700375e+00
-4.51101810e-01 4.35521334e-01 4.51720841e-02 -9.08920944e-01
1.11471760e+00 1.02990568e-01 8.23076785e-01 -1.13698900e+00
1.97445661e-01 4.94515061e-01 -1.42797153e-03 -3.86775434e-01
7.52716362e-01 3.00167948e-01 -7.58785754e-03 6.65521324e-01
-1.45529687e-01 -2.89127618e-01 4.91395265e-01 -4.43611115e-01
9.00605679e-01 3.16773117e-01 2.50846177e-01 5.70280142e-02
4.10000324e-01 9.74219069e-02 6.70462966e-01 9.95240688e-01
-5.22244751e-01 -5.64629138e-02 3.35858405e-01 -5.06982386e-01
-9.21843231e-01 -8.19505155e-01 2.38258898e-01 1.59605253e+00
4.59094226e-01 -5.96965440e-02 -4.15123731e-01 -3.71910185e-01
1.63768291e-01 5.89974821e-01 -7.12930083e-01 -3.76942009e-01
-7.92222440e-01 -1.24006510e+00 5.43758273e-01 5.48883080e-01
5.08347511e-01 -1.91838443e+00 -1.18318021e+00 5.40692449e-01
6.71556294e-02 -5.88319182e-01 -2.18027860e-01 2.86506325e-01
-8.21877539e-01 -1.03851724e+00 -9.07740116e-01 -8.53881478e-01
2.13442728e-01 -3.17279547e-02 1.31887925e+00 2.52718776e-01
-4.13043827e-01 4.51553971e-01 -5.02157629e-01 -5.70570707e-01
-2.92060584e-01 -2.45018750e-02 1.36435956e-01 -5.77856123e-01
1.66653454e-01 -4.12092924e-01 -6.36235118e-01 6.16884470e-01
-6.33302927e-01 -4.72638994e-01 6.74515724e-01 1.20169735e+00
7.23127484e-01 -1.12788845e-02 1.04558420e+00 -2.25521252e-01
1.33659041e+00 -5.63474417e-01 -1.05395222e+00 4.78907019e-01
-9.40995038e-01 2.02769324e-01 5.79287350e-01 -7.83201873e-01
-8.78325164e-01 -5.02715886e-01 -1.45354092e-01 -2.50280887e-01
2.11653635e-01 4.32263523e-01 4.25323159e-01 -4.50792611e-01
6.46742880e-01 3.00186515e-01 1.91553056e-01 4.14176434e-02
2.78545499e-01 3.29902321e-01 1.56024083e-01 -9.10646975e-01
1.69134185e-01 -4.64707687e-02 -4.36308887e-03 -3.30918342e-01
-5.60767770e-01 -2.30408870e-02 1.25061885e-01 -5.26364744e-01
4.01129216e-01 -5.11929750e-01 -1.20697033e+00 6.47578716e-01
-7.20451713e-01 -8.45813870e-01 -6.90460742e-01 5.18770456e-01
-1.10886014e+00 -2.42874771e-02 -5.23861110e-01 -1.11677527e+00
-4.57516044e-01 -1.48275781e+00 4.89904433e-01 8.84062171e-01
1.08630762e-01 -1.14559102e+00 6.35250986e-01 1.91386435e-02
6.89433575e-01 2.95561135e-01 6.76334441e-01 -3.65565896e-01
-3.74492764e-01 3.88849080e-01 3.27657431e-01 2.61720568e-02
-2.59015530e-01 6.52465075e-02 -3.61104101e-01 -6.05610967e-01
-2.82106191e-01 -1.18338287e+00 4.90722239e-01 6.37694359e-01
1.02509785e+00 -1.47503406e-01 -4.75509875e-02 6.81359887e-01
1.27853334e+00 7.76082695e-01 3.72762293e-01 1.08649147e+00
2.77511291e-02 4.07561988e-01 1.23064542e+00 1.07858253e+00
4.54962224e-01 5.54922879e-01 8.97641778e-01 5.33852316e-02
3.64638001e-01 7.29261562e-02 4.70846266e-01 6.54154778e-01
-4.65514630e-01 -3.54705751e-01 -6.89823031e-01 3.14845473e-01
-2.33872628e+00 -9.41007793e-01 5.24658144e-01 2.08671308e+00
9.12455261e-01 -1.54695913e-01 4.64610040e-01 -2.51514018e-01
7.84080625e-01 2.40233205e-02 -1.45335555e+00 -8.36187065e-01
-3.13385010e-01 2.51901835e-01 3.70507181e-01 3.38499427e-01
-9.11618412e-01 1.16535819e+00 6.97765970e+00 1.03021944e+00
-1.09738135e+00 -1.08198494e-01 5.66028476e-01 -4.28736478e-01
4.70130667e-02 -4.34804082e-01 -7.82774150e-01 4.55694944e-01
7.25723743e-01 -2.43240386e-01 1.28714466e+00 1.02587581e+00
1.88125074e-01 -2.46134907e-01 -6.17325544e-01 9.34248507e-01
-1.68283820e-01 -1.41671574e+00 -4.89860445e-01 -5.43902628e-02
1.29738593e+00 4.13846880e-01 5.10235488e-01 8.24438632e-01
1.03531969e+00 -1.30986917e+00 6.32199764e-01 3.02037269e-01
6.06151640e-01 -1.34660137e+00 6.68105125e-01 2.72334874e-01
-1.02509999e+00 -6.80761993e-01 -6.16184711e-01 -8.91849101e-02
-5.32256206e-04 -8.81558433e-02 -3.26399535e-01 4.43867177e-01
8.69257987e-01 5.80168247e-01 -1.20513298e-01 1.40366149e+00
-4.03133392e-01 1.07822686e-01 -1.06826246e-01 -8.39205325e-01
7.31286705e-01 -4.41135794e-01 6.37491524e-01 7.29362607e-01
3.43660772e-01 -1.42721282e-02 1.93782538e-01 8.61560643e-01
2.41940051e-01 3.13286372e-02 -1.08198866e-01 -9.17206481e-02
8.20982337e-01 1.14959574e+00 -6.10172510e-01 1.20762222e-01
-8.85699540e-02 5.35475075e-01 6.00491941e-01 4.11968857e-01
-1.08021867e+00 -4.31246668e-01 6.86249614e-01 -8.60254347e-01
2.47309849e-01 -2.51052588e-01 -1.57425553e-01 -6.83549881e-01
-3.09945315e-01 -1.26911521e+00 4.64123040e-01 -5.00971735e-01
-1.12437129e+00 6.97259367e-01 -1.46224186e-01 -1.14305854e+00
-6.42415881e-01 -4.17868763e-01 -6.35634005e-01 4.07515734e-01
-1.88275540e+00 -5.63294709e-01 -1.73031762e-01 5.91975272e-01
7.39668190e-01 -7.32601762e-01 7.49544859e-01 -3.62053253e-02
-7.52802193e-01 6.35996878e-01 7.83891618e-01 -5.24945378e-01
5.83465755e-01 -1.29515898e+00 1.95992932e-01 1.25954002e-01
-2.50300556e-01 -3.40284891e-02 4.62920994e-01 -4.80413586e-01
-1.58721900e+00 -8.98505092e-01 1.15212515e-01 2.56786887e-02
7.01975524e-01 8.48813653e-02 -4.73540068e-01 2.85105761e-02
2.23549679e-01 -2.43603453e-01 4.85556901e-01 -6.50885887e-03
4.71100152e-01 1.01741381e-01 -9.79735911e-01 8.19482625e-01
8.80032420e-01 2.65313506e-01 -2.82183047e-02 2.17573345e-01
4.42542285e-01 -7.61280715e-01 -7.30224133e-01 3.75891596e-01
6.08247340e-01 -1.06695890e+00 1.04866147e+00 -6.42907858e-01
2.18250960e-01 8.08217973e-02 2.72589892e-01 -1.89457476e+00
-4.03960556e-01 -9.67072964e-01 -4.39700872e-01 6.65564179e-01
2.26458713e-01 -7.10386932e-01 9.63734031e-01 3.96252185e-01
-2.11465269e-01 -1.51485515e+00 -8.56404424e-01 -1.03560996e+00
3.74313563e-01 -1.22944325e-01 8.76604617e-01 6.41637444e-01
-2.84297857e-02 -1.13317803e-01 -5.04971385e-01 -3.05208057e-01
5.14154911e-01 1.91680014e-01 4.60630417e-01 -9.29297030e-01
-4.59694594e-01 -1.02225435e+00 2.72707641e-01 -8.10852766e-01
2.18319178e-01 -3.43698889e-01 2.59159476e-01 -1.40764380e+00
-1.17247827e-01 -9.71043110e-01 -4.94538218e-01 3.39582503e-01
-1.99050784e-01 -2.75619719e-02 3.46747816e-01 2.46371806e-01
-1.13958406e+00 1.27824903e+00 1.71145511e+00 -7.74240196e-02
-6.46141946e-01 2.95100003e-01 -4.58396316e-01 6.15684211e-01
1.31488490e+00 -4.29190516e-01 -5.11921585e-01 -3.99571776e-01
6.07542038e-01 2.46271715e-01 -5.07550016e-02 -8.62890780e-01
3.28837365e-01 -9.52112913e-01 2.03898028e-01 -3.26811492e-01
4.42370802e-01 -3.71140450e-01 -1.19273044e-01 7.12572932e-01
-3.69171590e-01 6.68627977e-01 3.13633442e-01 4.85447854e-01
-1.70880899e-01 -4.80762422e-01 7.63009310e-01 -3.83508623e-01
-9.03224349e-01 3.10665369e-01 -7.34369457e-01 5.02571821e-01
1.35370016e+00 -2.80051887e-01 -3.26996982e-01 -4.33985710e-01
-5.81109583e-01 9.84316111e-01 2.99588710e-01 4.84490484e-01
7.81006694e-01 -1.29937685e+00 -6.66020989e-01 -7.41938278e-02
-3.13288927e-01 -1.51441202e-01 1.15130901e-01 7.19697714e-01
-3.22836667e-01 2.80817568e-01 -6.26646936e-01 -1.54848054e-01
-8.63086402e-01 3.05426806e-01 6.45717680e-01 -7.02630460e-01
-1.55174837e-01 8.96293044e-01 -3.80188555e-01 -6.76601708e-01
6.82116985e-01 1.06917962e-01 -4.77384925e-01 -1.35035396e-01
3.26513618e-01 9.61930513e-01 -1.51301548e-01 -1.70062631e-02
-3.08197647e-01 4.81095016e-01 2.29518592e-01 -2.52807915e-01
1.50187337e+00 1.17101833e-01 9.09896418e-02 6.61443844e-02
3.61215502e-01 -3.73581290e-01 -1.78075433e+00 9.69669744e-02
-1.29408017e-01 -2.01311111e-01 1.91358421e-02 -1.24741459e+00
-1.14076042e+00 5.70200682e-01 5.24665415e-01 1.38973847e-01
1.12437153e+00 -3.20912510e-01 6.54639840e-01 6.37598097e-01
6.52622402e-01 -1.79736757e+00 7.11952209e-01 1.04883480e+00
1.04935455e+00 -1.27742517e+00 -3.74853909e-02 4.59260881e-01
-1.09407878e+00 1.18688142e+00 1.04131293e+00 -2.88707644e-01
2.63572603e-01 2.36406371e-01 -4.19035032e-02 -2.40577292e-02
-1.10487509e+00 -3.57712060e-01 -1.44530177e-01 9.50211346e-01
-3.89547274e-02 3.89817841e-02 -2.29196161e-01 3.06533605e-01
-2.22576827e-01 -1.83705837e-01 2.47201756e-01 1.17179930e+00
-6.23695493e-01 -1.46805763e+00 -4.28756148e-01 1.97790161e-01
-2.83150166e-01 1.27710626e-01 -2.30766937e-01 8.18973422e-01
-8.01707581e-02 1.03951752e+00 -2.49831453e-02 -1.83482036e-01
1.97958246e-01 -7.21030056e-01 6.12770796e-01 -1.75607830e-01
-9.89096582e-01 -1.18923970e-01 -6.21122234e-02 -6.71922386e-01
-3.42963666e-01 -7.39396513e-01 -1.46118009e+00 -3.42676282e-01
-3.28855455e-01 5.10134935e-01 7.54746318e-01 8.00965667e-01
4.47920471e-01 6.47887647e-01 7.96073973e-01 -9.68826473e-01
-1.03811777e+00 -5.01478672e-01 -4.03824121e-01 -3.26502383e-01
8.13994110e-02 -1.16853189e+00 -2.28783600e-02 -8.26884389e-01] | [3.9038450717926025, 1.9781428575515747] |
9649ef67-d3b4-4357-b618-df8b51e48718 | paste-inpaint-and-harmonize-via-denoising | 2306.07596 | null | https://arxiv.org/abs/2306.07596v1 | https://arxiv.org/pdf/2306.07596v1.pdf | Paste, Inpaint and Harmonize via Denoising: Subject-Driven Image Editing with Pre-Trained Diffusion Model | Text-to-image generative models have attracted rising attention for flexible image editing via user-specified descriptions. However, text descriptions alone are not enough to elaborate the details of subjects, often compromising the subjects' identity or requiring additional per-subject fine-tuning. We introduce a new framework called \textit{Paste, Inpaint and Harmonize via Denoising} (PhD), which leverages an exemplar image in addition to text descriptions to specify user intentions. In the pasting step, an off-the-shelf segmentation model is employed to identify a user-specified subject within an exemplar image which is subsequently inserted into a background image to serve as an initialization capturing both scene context and subject identity in one. To guarantee the visual coherence of the generated or edited image, we introduce an inpainting and harmonizing module to guide the pre-trained diffusion model to seamlessly blend the inserted subject into the scene naturally. As we keep the pre-trained diffusion model frozen, we preserve its strong image synthesis ability and text-driven ability, thus achieving high-quality results and flexible editing with diverse texts. In our experiments, we apply PhD to both subject-driven image editing tasks and explore text-driven scene generation given a reference subject. Both quantitative and qualitative comparisons with baseline methods demonstrate that our approach achieves state-of-the-art performance in both tasks. More qualitative results can be found at \url{https://sites.google.com/view/phd-demo-page}. | ['Yusuke Iwasawa', 'Yutaka Matsuo', 'Paul Yoo', 'Jiaxian Guo', 'Xin Zhang'] | 2023-06-13 | null | null | null | null | ['scene-generation'] | ['computer-vision'] | [ 5.31934261e-01 1.00683630e-01 4.96687219e-02 -5.10574758e-01
-8.00071836e-01 -6.62108421e-01 8.19161057e-01 -1.82460546e-01
-1.96568415e-01 3.77156138e-01 2.88878530e-01 1.29871488e-01
2.32196063e-01 -5.11792660e-01 -7.27998912e-01 -5.67290664e-01
6.34873986e-01 3.52754921e-01 1.70282990e-01 -1.45106018e-01
9.00936797e-02 2.84435332e-01 -1.39550507e+00 2.39908695e-01
1.40245557e+00 5.75819135e-01 6.21787310e-01 5.28273821e-01
-5.57324439e-02 4.57916796e-01 -6.23312652e-01 -6.79847002e-01
4.01681483e-01 -8.38844299e-01 -5.32439411e-01 7.33151019e-01
6.95570111e-01 -4.39835995e-01 -2.87464231e-01 1.10837066e+00
5.44680417e-01 3.23798895e-01 5.75220346e-01 -1.00116622e+00
-8.96787763e-01 4.21040833e-01 -9.18190777e-01 -1.71711370e-01
4.99491364e-01 5.81642926e-01 7.49537706e-01 -8.55362654e-01
9.97506857e-01 1.08967209e+00 2.84077883e-01 5.83530307e-01
-1.48037434e+00 -6.11586928e-01 2.76985615e-01 -2.09161386e-01
-1.21043932e+00 -6.55159771e-01 1.03058887e+00 -5.32109022e-01
2.11465940e-01 4.51541156e-01 8.08163106e-01 1.16289449e+00
1.43871354e-02 8.49597692e-01 1.00635159e+00 -3.41658205e-01
4.98716421e-02 2.27115601e-01 -3.51507217e-01 6.61790848e-01
-1.64726093e-01 -4.94915806e-02 -4.51517612e-01 2.31710732e-01
1.07632470e+00 4.92806360e-02 -5.24186075e-01 -4.08445686e-01
-1.31120539e+00 6.44285738e-01 3.74791712e-01 1.71365887e-01
-5.50471425e-01 -1.41298911e-02 9.57961977e-02 -4.73982207e-02
6.81199431e-01 4.28196102e-01 -3.30887549e-02 4.18295413e-02
-1.48334646e+00 5.06167233e-01 4.41040844e-01 1.17893291e+00
7.53459394e-01 1.20364681e-01 -6.87992215e-01 1.05989933e+00
3.43905129e-02 3.98384303e-01 1.31442875e-01 -1.34664774e+00
3.20235789e-01 5.19947588e-01 1.70441523e-01 -8.07294190e-01
2.09051996e-01 -6.22047126e-01 -8.29818785e-01 2.57402211e-01
2.88633078e-01 -5.54405376e-02 -1.17861903e+00 1.71845913e+00
5.32638073e-01 6.22279011e-02 -4.26358610e-01 9.48383868e-01
7.05429792e-01 7.59114802e-01 1.13295093e-01 -1.84894606e-01
1.41406393e+00 -1.18687356e+00 -7.98012853e-01 -4.54799414e-01
7.59489611e-02 -9.64138448e-01 1.49852014e+00 2.63311595e-01
-1.58595610e+00 -6.93220377e-01 -7.63987482e-01 -3.28460902e-01
7.46420920e-02 2.85871655e-01 2.01538131e-01 3.83333802e-01
-9.91858661e-01 3.52795094e-01 -6.97947919e-01 -3.37961078e-01
6.57408714e-01 5.87818678e-03 -1.40411973e-01 -3.47618699e-01
-8.73133302e-01 5.37667572e-01 8.13722834e-02 -6.63059503e-02
-1.08264303e+00 -9.79996800e-01 -9.68577981e-01 -2.05072582e-01
4.95726049e-01 -1.06458569e+00 1.21593213e+00 -1.14659560e+00
-1.49792659e+00 1.16643357e+00 -4.23305273e-01 6.08242350e-03
1.01401317e+00 -2.23418772e-01 -1.03141256e-01 2.02610046e-01
4.49613094e-01 1.07526112e+00 1.21579969e+00 -1.72578096e+00
-4.84047323e-01 -8.69302601e-02 -5.78773431e-02 3.40364397e-01
-6.49607033e-02 1.61438018e-01 -1.25310588e+00 -1.27393222e+00
-2.26902366e-02 -9.07078505e-01 -1.85534149e-01 2.48192027e-01
-7.75259078e-01 2.13838056e-01 8.94777954e-01 -9.11505163e-01
1.29765403e+00 -2.38806677e+00 2.59113014e-01 8.70687142e-02
3.76946092e-01 8.60172883e-02 -2.89394766e-01 2.28530392e-01
2.54656784e-02 2.82355715e-02 -5.88434160e-01 -1.07430029e+00
5.70037514e-02 1.07970893e-01 -3.09068948e-01 2.96139926e-01
1.31722033e-01 1.15143728e+00 -8.37242723e-01 -6.55576110e-01
4.31285024e-01 6.92187965e-01 -6.99169636e-01 3.13284039e-01
-4.28647190e-01 8.56327415e-01 -3.91229510e-01 5.88139594e-01
7.10992873e-01 -2.72564381e-01 -1.44292712e-01 -3.57696295e-01
6.41896874e-02 -3.84960696e-02 -1.26080203e+00 2.01142836e+00
-4.22396064e-01 5.93201637e-01 3.25369447e-01 -3.73795897e-01
6.67162895e-01 1.27524063e-01 4.20640469e-01 -6.70724571e-01
-6.78810757e-03 -2.20329817e-02 -2.59504884e-01 -3.64351153e-01
6.46453261e-01 1.33655384e-01 4.58467044e-02 4.89439130e-01
-1.29759207e-01 -5.28239787e-01 4.54679072e-01 5.11983275e-01
6.01598620e-01 4.55519140e-01 -6.52288795e-02 -1.23825073e-01
4.12356824e-01 -1.33424178e-01 5.37248492e-01 7.04618275e-01
-4.62873615e-02 1.26226950e+00 3.21831942e-01 -2.83357389e-02
-1.18488562e+00 -1.10716093e+00 1.85440611e-02 1.09139836e+00
1.84428692e-01 -3.90208393e-01 -1.23039389e+00 -5.08267045e-01
-3.52954328e-01 9.96146679e-01 -6.95299268e-01 1.69636115e-01
-5.07665098e-01 -3.92870963e-01 1.17624097e-01 2.04429910e-01
7.81552911e-01 -1.17248273e+00 -4.78764772e-01 2.01094553e-01
-4.64615077e-01 -1.07603824e+00 -1.36100578e+00 -3.22583944e-01
-4.49460804e-01 -5.77065229e-01 -1.02270544e+00 -7.78374195e-01
1.09081101e+00 2.34563872e-01 1.07190168e+00 7.83933774e-02
-1.52402624e-01 4.18518275e-01 -2.73212880e-01 -9.04216394e-02
-6.25857055e-01 -6.23815581e-02 -2.94638306e-01 3.66869777e-01
-2.94668645e-01 -7.10589290e-01 -1.01763701e+00 3.39811057e-01
-1.30500913e+00 5.66691637e-01 4.24667060e-01 5.87773383e-01
7.00328708e-01 4.27674316e-02 9.70321298e-02 -9.57145512e-01
6.13708854e-01 -9.42890346e-02 -5.03941298e-01 1.13680072e-01
-3.17961395e-01 -1.87677562e-01 3.78266603e-01 -7.04475284e-01
-1.32420599e+00 2.56620675e-01 2.37641837e-02 -5.59192002e-01
-1.81360886e-01 1.48012653e-01 -5.32210708e-01 1.07354201e-01
4.76585388e-01 4.94281530e-01 -4.57493514e-02 -3.13491434e-01
7.03315079e-01 3.38315070e-01 9.09619987e-01 -6.59022570e-01
9.45069373e-01 4.85335946e-01 -6.26935959e-01 -4.86966908e-01
-9.19439912e-01 -1.97575986e-02 -6.92473769e-01 -3.66619706e-01
9.97395515e-01 -9.37896311e-01 -2.58715808e-01 5.75173438e-01
-1.24376571e+00 -7.65814543e-01 -4.40215260e-01 1.74345169e-02
-5.55170178e-01 1.63528547e-01 -5.15926182e-01 -3.72952044e-01
-2.71545768e-01 -1.44811308e+00 1.35300529e+00 3.87558669e-01
-4.65013623e-01 -9.22924995e-01 -2.54399687e-01 6.54108584e-01
3.69702905e-01 3.77229542e-01 6.34423077e-01 -1.79113165e-01
-8.08941424e-01 -5.97700998e-02 -1.29408211e-01 2.45161161e-01
2.20993072e-01 9.01995897e-02 -8.68974566e-01 -2.31330082e-01
-5.35367196e-03 2.43263841e-02 6.86052561e-01 5.13687909e-01
1.15053010e+00 -3.12375993e-01 -3.08953345e-01 8.86171103e-01
1.11680305e+00 1.66304454e-01 8.77488971e-01 2.20371544e-01
1.03229928e+00 7.29662418e-01 4.70628709e-01 4.82670367e-01
4.29536790e-01 6.54554486e-01 4.46239375e-02 -4.17256236e-01
-5.50908625e-01 -4.91501749e-01 2.74397075e-01 4.40499246e-01
6.59560710e-02 -4.61597294e-01 -5.70985734e-01 5.96406162e-01
-1.60665965e+00 -1.07213485e+00 -5.16389199e-02 2.05388308e+00
1.30802274e+00 -3.39582860e-02 5.56331873e-02 -2.68670589e-01
7.21604645e-01 4.49761987e-01 -6.60096407e-01 1.52687903e-03
-1.58811823e-01 3.39488909e-02 4.20600213e-02 6.96864903e-01
-9.63702917e-01 1.20976293e+00 5.10829973e+00 1.11261272e+00
-1.20148122e+00 2.31546104e-01 9.74723935e-01 -2.84642816e-01
-5.80950499e-01 3.63817587e-02 -5.57489514e-01 6.08112931e-01
2.76747614e-01 -1.55470058e-01 5.15661180e-01 4.41006809e-01
7.77854204e-01 -2.68962145e-01 -9.73665237e-01 9.29739416e-01
1.86328515e-01 -1.31610191e+00 2.11488008e-01 8.46957713e-02
1.11140108e+00 -3.98749262e-01 2.88844019e-01 1.27545251e-02
2.81305820e-01 -7.22607613e-01 1.05474496e+00 5.05133152e-01
1.02118218e+00 -5.52713275e-01 -4.09774557e-02 1.53009966e-01
-1.03473341e+00 3.54134321e-01 7.58939087e-02 4.73996490e-01
5.82902968e-01 7.32662201e-01 -3.13548237e-01 4.46512878e-01
6.59107327e-01 7.42705822e-01 -7.54029810e-01 7.91323602e-01
-5.27726352e-01 4.91297871e-01 -1.29834339e-01 6.45110965e-01
9.07017738e-02 -4.56015050e-01 7.72476494e-01 1.11832213e+00
2.23098576e-01 3.13124776e-01 3.12267601e-01 1.48361647e+00
-1.77397162e-01 4.85690571e-02 -3.17392737e-01 1.09458826e-01
4.19425964e-01 1.37067115e+00 -8.67196202e-01 -4.70994234e-01
-1.77955806e-01 1.66011500e+00 -8.00080597e-02 6.31765187e-01
-1.00366759e+00 -3.73662502e-01 4.88959283e-01 5.05451560e-01
3.70592535e-01 -2.25154623e-01 -4.56334561e-01 -1.15745521e+00
1.48139909e-01 -1.09197414e+00 6.82144389e-02 -1.13788950e+00
-1.12107170e+00 6.77552044e-01 1.44506350e-01 -1.23446655e+00
-1.69051394e-01 1.54764116e-01 -8.36491942e-01 1.05438685e+00
-1.09814692e+00 -1.49600899e+00 -5.34405053e-01 6.87246978e-01
8.73308659e-01 1.35742962e-01 2.21376985e-01 4.21688318e-01
-6.97814524e-01 6.30831838e-01 -1.60971805e-01 4.72524799e-02
1.08667135e+00 -1.10479999e+00 4.27295744e-01 1.21208549e+00
5.37725240e-02 8.08935702e-01 8.07725191e-01 -1.04245925e+00
-1.02451158e+00 -1.22709179e+00 6.42082095e-01 -5.58732867e-01
3.01129609e-01 -4.88134652e-01 -1.02740741e+00 7.85646796e-01
5.56011081e-01 -5.15360758e-02 3.22190613e-01 -4.84047741e-01
-2.55259685e-02 -4.41240519e-02 -1.00030267e+00 1.13091087e+00
1.15944529e+00 -3.53365302e-01 -2.32471123e-01 2.62311935e-01
8.32573414e-01 -8.23313951e-01 -7.11453021e-01 7.70961121e-02
3.06166857e-01 -9.96337712e-01 9.72213268e-01 8.08850378e-02
7.09531903e-01 -5.66469193e-01 2.62678206e-01 -1.20307577e+00
-3.13571423e-01 -1.29273903e+00 2.40592703e-01 1.77590132e+00
2.81056702e-01 -3.73421460e-01 7.18686879e-01 1.05027747e+00
-1.62579641e-01 -6.44602239e-01 -3.57103318e-01 -4.22838777e-01
-1.59080014e-01 -4.65213388e-01 4.82665628e-01 1.02250516e+00
-5.04657686e-01 2.20698193e-01 -5.16717315e-01 5.62546290e-02
7.96850204e-01 1.57812223e-01 9.78808343e-01 -6.34213030e-01
-3.10655087e-01 -6.11697316e-01 2.35632360e-01 -1.22312891e+00
2.17664279e-02 -7.47599900e-01 8.32178220e-02 -1.65784252e+00
2.13481605e-01 -3.18063229e-01 2.73393333e-01 4.49554801e-01
-5.64043045e-01 4.96144146e-01 3.90074581e-01 4.05802518e-01
-5.14662206e-01 6.94244623e-01 1.83915913e+00 -1.33236200e-01
-5.93434393e-01 -6.61693886e-02 -9.53488827e-01 5.84592998e-01
5.27450502e-01 -3.60891134e-01 -5.92911839e-01 -6.35358214e-01
-1.96328893e-01 -1.24255484e-02 5.61947584e-01 -7.94467092e-01
2.46899500e-01 -2.67465800e-01 5.49221098e-01 -3.80060315e-01
2.98023582e-01 -5.60786784e-01 4.96884048e-01 9.01381969e-02
-4.06365126e-01 -9.52413958e-03 1.48623630e-01 3.77451897e-01
-1.56465530e-01 -1.04649879e-01 9.45653439e-01 -1.89373881e-01
-5.07793844e-01 6.17824435e-01 -1.00751020e-01 1.25180423e-01
9.65434492e-01 -5.28072774e-01 -1.36867790e-02 -6.79914892e-01
-7.14158118e-01 3.40150625e-01 9.97069716e-01 4.41163540e-01
5.48856676e-01 -1.16419613e+00 -8.31904113e-01 2.39357486e-01
-6.72788024e-02 2.77963430e-01 6.26769006e-01 8.41282725e-01
-4.26329970e-01 -2.39645749e-01 6.09309645e-04 -6.06405914e-01
-1.19307542e+00 3.97073984e-01 3.05609673e-01 -5.89401871e-02
-7.42697835e-01 7.48440683e-01 7.36503780e-01 -2.56586254e-01
1.96492836e-01 -1.19222008e-01 2.79847860e-01 -1.35703012e-02
4.41764534e-01 2.96046603e-02 -3.67273569e-01 -6.64491832e-01
-6.17418550e-02 7.14266360e-01 -2.62980103e-01 -5.56486785e-01
1.24849665e+00 -5.28054893e-01 -5.73273189e-02 4.27206531e-02
1.01497817e+00 3.64047766e-01 -1.80630100e+00 -3.91686082e-01
-5.81275105e-01 -6.70148909e-01 -4.58539352e-02 -9.60521579e-01
-1.26903248e+00 6.67803586e-01 3.12066823e-01 -1.75494432e-01
1.33544946e+00 1.81258600e-02 9.06323314e-01 -3.77064645e-01
-5.87472282e-02 -9.10648882e-01 2.98277676e-01 1.12297677e-01
1.29930651e+00 -1.10554290e+00 7.37074390e-02 -6.28448904e-01
-1.06788540e+00 6.10467613e-01 6.78454220e-01 4.19677375e-03
2.36331195e-01 2.09916204e-01 1.85547143e-01 -9.53861624e-02
-3.43630314e-01 -5.62968701e-02 5.22563994e-01 6.52253389e-01
3.14908624e-01 -2.14999527e-01 -9.42284614e-02 3.73481750e-01
-4.05669063e-01 1.38751008e-02 4.06147212e-01 7.99478233e-01
-7.76056424e-02 -1.14573824e+00 -4.32589680e-01 2.19435945e-01
-2.78186291e-01 -2.38989964e-01 -3.16147208e-01 6.39693201e-01
2.70628631e-01 7.90893137e-01 -2.22747978e-02 4.80754580e-03
4.27287519e-01 -1.51807010e-01 4.78099167e-01 -6.55287802e-01
-7.10647404e-01 5.49980819e-01 -1.93100721e-01 -6.05209053e-01
-2.46736005e-01 -8.24247956e-01 -1.10619330e+00 -3.16997260e-01
9.45748091e-02 -2.33753607e-01 3.91821533e-01 6.77884936e-01
5.73472381e-01 6.80509984e-01 4.04591054e-01 -1.20541739e+00
6.69376925e-02 -7.52304971e-01 -4.23401028e-01 6.73546791e-01
2.26794705e-01 -3.37410927e-01 -1.27800643e-01 7.63675332e-01] | [11.334070205688477, -0.6742768883705139] |
f7f2b791-69f9-4e5d-a199-70ceae19da72 | counting-guidance-for-high-fidelity-text-to | 2306.17567 | null | https://arxiv.org/abs/2306.17567v1 | https://arxiv.org/pdf/2306.17567v1.pdf | Counting Guidance for High Fidelity Text-to-Image Synthesis | Recently, the quality and performance of text-to-image generation significantly advanced due to the impressive results of diffusion models. However, text-to-image diffusion models still fail to generate high fidelity content with respect to the input prompt. One problem where text-to-diffusion models struggle is generating the exact number of objects specified in the text prompt. E.g. given a prompt "five apples and ten lemons on a table", diffusion-generated images usually contain the wrong number of objects. In this paper, we propose a method to improve diffusion models to focus on producing the correct object count given the input prompt. We adopt a counting network that performs reference-less class-agnostic counting for any given image. We calculate the gradients of the counting network and refine the predicted noise for each step. To handle multiple types of objects in the prompt, we use novel attention map guidance to obtain high-fidelity masks for each object. Finally, we guide the denoising process by the calculated gradients for each object. Through extensive experiments and evaluation, we demonstrate that our proposed guidance method greatly improves the fidelity of diffusion models to object count. | ['Hyung Il Koo', 'Kevin Galim', 'Wonjun Kang'] | 2023-06-30 | null | null | null | null | ['image-generation'] | ['computer-vision'] | [ 4.73860234e-01 -4.05305684e-01 3.91260594e-01 -4.13500041e-01
-7.20995963e-01 -5.04816234e-01 6.36188328e-01 6.81821480e-02
-4.82431293e-01 3.67790639e-01 2.14225575e-01 -4.95336689e-02
1.09899811e-01 -9.75565791e-01 -6.90430760e-01 -6.74091578e-01
5.78469813e-01 4.45046067e-01 2.66588062e-01 5.03954068e-02
5.51361680e-01 3.93218666e-01 -1.33627903e+00 6.68101132e-01
1.02111959e+00 9.98818398e-01 7.52424717e-01 1.08089173e+00
-4.35338795e-01 1.32692122e+00 -1.05184078e+00 -5.16406476e-01
3.16219807e-01 -7.52101004e-01 -4.54612762e-01 3.32423955e-01
7.61285424e-01 -8.62958491e-01 -5.92494667e-01 1.46043932e+00
5.13628364e-01 1.92297235e-01 9.29046988e-01 -9.69554007e-01
-1.34249485e+00 5.60893476e-01 -9.76820529e-01 5.71599066e-01
3.33651543e-01 6.22531772e-01 5.93618393e-01 -8.87328744e-01
6.68670654e-01 1.48871124e+00 3.76958549e-01 7.67410338e-01
-1.05184054e+00 -7.78953373e-01 1.89326122e-01 1.50474027e-01
-1.19535637e+00 -3.53563309e-01 6.37812912e-01 -5.17754376e-01
5.26679218e-01 2.10032046e-01 6.56851053e-01 1.03848803e+00
-1.04646245e-03 8.76212120e-01 1.10364032e+00 -3.09605837e-01
2.33438969e-01 -1.66465826e-02 -1.17784113e-01 8.13751459e-01
1.17524132e-01 -1.24128103e-01 -5.76637208e-01 1.18389741e-01
1.19649982e+00 -3.93688716e-02 -1.26278400e-01 2.64626861e-01
-1.20114529e+00 6.81675553e-01 4.62630212e-01 2.94350356e-01
-6.46316648e-01 4.62309659e-01 2.83831637e-02 -1.40470549e-01
6.96363151e-01 1.96536601e-01 1.00823395e-01 -6.97404370e-02
-1.18357635e+00 3.41007769e-01 4.81506854e-01 9.62183833e-01
5.30890107e-01 1.96692631e-01 -8.59760582e-01 8.69350314e-01
-1.28531351e-03 8.01367104e-01 9.73245054e-02 -1.22768545e+00
7.16158986e-01 4.55924183e-01 2.33661965e-01 -1.37088847e+00
8.13382305e-03 -4.54941839e-01 -1.27081954e+00 2.99399167e-01
7.95564175e-01 -1.86133549e-01 -1.12542355e+00 1.52630818e+00
3.80290151e-01 2.99303263e-01 -2.07557485e-01 1.12400770e+00
6.22976959e-01 1.02063155e+00 1.29557222e-01 -7.59522393e-02
1.25961268e+00 -1.02369058e+00 -8.82160783e-01 -3.23279738e-01
3.38186234e-01 -7.51951277e-01 1.20574582e+00 4.45214748e-01
-1.33069253e+00 -7.04954624e-01 -7.16734827e-01 -2.25159392e-01
3.05654388e-02 4.15758789e-01 2.74580508e-01 5.71387768e-01
-1.00486374e+00 7.52956331e-01 -6.06561422e-01 -7.84559697e-02
6.40211225e-01 5.70020452e-02 1.03271745e-01 -3.70244175e-01
-7.92144239e-01 7.16419458e-01 1.55789495e-01 2.73741223e-02
-1.14189172e+00 -7.52133012e-01 -6.42766237e-01 1.28165722e-01
1.49507761e-01 -8.23978901e-01 1.22149241e+00 -9.91000235e-01
-1.14954925e+00 5.82949042e-01 -2.95807600e-01 -3.54477316e-01
8.08781683e-01 -5.22362255e-02 1.30274734e-02 4.33727324e-01
3.62663686e-01 1.10665107e+00 1.28735018e+00 -1.56051469e+00
-1.00633693e+00 -3.54896188e-01 -1.05138950e-01 1.52165979e-01
-4.87701893e-01 2.01914031e-02 -6.46923661e-01 -9.28370118e-01
5.96390199e-03 -3.99689376e-01 -3.14638555e-01 4.00492668e-01
-4.18564498e-01 -5.87751158e-02 7.31458783e-01 -9.35355186e-01
1.25136316e+00 -2.04358077e+00 -4.97771837e-02 -2.55883578e-02
5.15826762e-01 4.09715772e-02 -4.31107074e-01 2.36385595e-02
1.68335885e-01 2.86876899e-03 -3.88594538e-01 -5.55846453e-01
-2.11732432e-01 1.17331363e-01 -3.00993174e-01 2.44757056e-01
4.00266171e-01 8.53668690e-01 -1.16637099e+00 -6.55379534e-01
1.26236409e-01 6.81527555e-01 -5.67420542e-01 2.13825047e-01
-2.97497600e-01 6.15848541e-01 -2.37727940e-01 3.41107428e-01
9.11815286e-01 -4.49955046e-01 -3.75319034e-01 -3.57831210e-01
1.47586301e-01 -2.64381438e-01 -1.20592022e+00 1.52022672e+00
-4.84509140e-01 5.36801577e-01 -3.31138149e-02 -4.21928644e-01
9.55990374e-01 -9.21110362e-02 3.18149507e-01 -8.05567205e-01
1.23885982e-02 1.39993086e-01 6.87935129e-02 -4.70250636e-01
7.80627668e-01 4.73557003e-02 2.24716797e-01 5.31909227e-01
-4.72372882e-02 -5.66561401e-01 5.94212770e-01 5.70204437e-01
1.15941060e+00 -2.23372668e-01 -2.48776585e-01 1.02203652e-01
3.79520774e-01 1.18628526e-02 2.29248971e-01 1.23969018e+00
4.45928099e-03 1.02968884e+00 4.05831069e-01 -2.84466445e-01
-1.41139376e+00 -8.92183363e-01 2.25890368e-01 7.77519345e-01
5.78950606e-02 -7.01888949e-02 -1.22871852e+00 -6.95579290e-01
-3.36763293e-01 9.61967707e-01 -7.04938471e-01 -6.85526431e-02
-5.28197289e-01 -7.47771263e-01 5.21750510e-01 4.50477332e-01
8.15365076e-01 -1.17589164e+00 -4.12456751e-01 3.64842564e-01
-4.32923883e-01 -1.19637012e+00 -1.16122973e+00 -1.27285570e-01
-8.55375230e-01 -9.10822630e-01 -1.30762076e+00 -7.82112956e-01
1.21158540e+00 4.32400554e-01 9.07491565e-01 3.62485409e-01
-1.93264738e-01 3.42967957e-01 -1.60867468e-01 -1.87639907e-01
-7.84368038e-01 -2.08508670e-01 -2.39778593e-01 2.56816685e-01
2.29993582e-01 -1.70707867e-01 -9.23733413e-01 1.69777408e-01
-1.27098310e+00 2.60884196e-01 4.62007731e-01 7.30446696e-01
5.17449915e-01 3.04767489e-01 3.16907167e-01 -5.87203920e-01
1.19134331e+00 -2.07794368e-01 -5.99433959e-01 1.26324952e-01
-4.36612159e-01 5.33277839e-02 7.22701669e-01 -8.28652322e-01
-1.08395231e+00 1.88656360e-01 -1.90729573e-01 -5.42381883e-01
9.05018747e-02 -6.88847527e-02 1.59227744e-01 1.74895123e-01
7.74283946e-01 6.09139860e-01 -3.03452522e-01 -2.64288247e-01
4.52560812e-01 4.77356225e-01 7.64475763e-01 -6.85778081e-01
7.45251417e-01 6.78246021e-01 -2.80961663e-01 -5.66034436e-01
-1.03659976e+00 -3.02062869e-01 -5.28712928e-01 -4.18905824e-01
1.12106347e+00 -5.93815088e-01 -6.89411759e-01 8.58674109e-01
-1.71056402e+00 -5.58483958e-01 -5.67101181e-01 1.18329249e-01
-4.28287059e-01 5.73742211e-01 -8.37813854e-01 -1.00640237e+00
-4.80833381e-01 -1.16659796e+00 1.11809421e+00 1.92693189e-01
-2.91963876e-03 -8.09473693e-01 -2.41071120e-01 1.51117802e-01
4.29380745e-01 -1.34742245e-01 8.00706625e-01 1.12413475e-02
-6.63021863e-01 -1.10317230e-01 -7.59355664e-01 6.05909050e-01
1.35464847e-01 -9.46735777e-03 -7.11447716e-01 2.57467870e-02
3.10860246e-01 -6.72068149e-02 9.75423694e-01 6.22034967e-01
1.45007479e+00 -5.09846747e-01 -6.55999407e-04 4.53173518e-01
1.30257082e+00 2.15161532e-01 7.32921243e-01 1.20867486e-03
8.59932125e-01 6.41969860e-01 3.79538894e-01 5.97523868e-01
1.69624537e-01 3.63396436e-01 2.98584849e-01 -1.07586056e-01
-6.88056290e-01 -5.21773875e-01 1.18150763e-01 8.69745672e-01
1.36640117e-01 -5.27198792e-01 -6.73809886e-01 7.14133561e-01
-1.47502005e+00 -1.07576418e+00 -4.55844998e-01 1.95643353e+00
9.62790430e-01 1.43682003e-01 -1.47427112e-01 5.49612008e-02
9.53109920e-01 1.04944043e-01 -7.07652628e-01 -5.95919490e-02
-2.43299305e-01 -1.01988800e-01 4.66257632e-01 6.58881664e-01
-5.90040743e-01 9.22662616e-01 6.37133455e+00 1.17548692e+00
-7.01459587e-01 2.27493554e-01 1.11688209e+00 -1.29360184e-01
-3.48637909e-01 -2.52611488e-01 -9.51263189e-01 6.54779196e-01
4.96523112e-01 -8.71932358e-02 6.94242775e-01 5.90276718e-01
4.11317825e-01 -3.78353238e-01 -8.76458764e-01 1.24752128e+00
1.44017220e-01 -1.33361304e+00 5.14705896e-01 -2.39006832e-01
1.12995207e+00 -4.89073992e-01 3.54364306e-01 -5.13232276e-02
4.96351033e-01 -8.30534279e-01 1.14217091e+00 6.02125823e-01
9.06695485e-01 -4.78199154e-01 3.19172978e-01 6.43703401e-01
-9.03300345e-01 -1.10446602e-01 -6.38893425e-01 1.05213217e-01
3.83739978e-01 9.06578779e-01 -5.87358594e-01 -1.44398317e-01
6.96038961e-01 3.98972958e-01 -6.92174554e-01 1.01032519e+00
-3.32654506e-01 4.10341263e-01 -2.57840395e-01 -8.51712078e-02
3.81466329e-01 -1.86916769e-01 3.06921571e-01 1.22346413e+00
6.81874752e-01 3.64070624e-01 -2.07377613e-01 1.38253427e+00
-3.42265815e-01 -4.14184630e-02 -5.24142265e-01 4.20971327e-02
3.12984973e-01 1.19568765e+00 -9.13314939e-01 -8.60045195e-01
-1.42538816e-01 1.48699296e+00 3.55324477e-01 5.04252493e-01
-8.75794232e-01 -1.97252750e-01 1.12325616e-01 3.52081358e-01
2.78292537e-01 -1.94835976e-01 -5.65210342e-01 -9.45307732e-01
7.36854821e-02 -8.84273469e-01 1.30147010e-01 -1.30683887e+00
-1.46892381e+00 5.63715696e-01 -9.93919224e-02 -8.45056832e-01
4.02737260e-02 -3.62084895e-01 -5.68310380e-01 9.55195904e-01
-1.17371821e+00 -7.56869912e-01 -6.63699985e-01 4.55589920e-01
9.74441588e-01 6.11976795e-02 2.40382716e-01 4.76226151e-01
-2.16146067e-01 4.16979283e-01 6.80895969e-02 1.83348596e-01
5.34553409e-01 -1.20824850e+00 7.36176014e-01 9.73502636e-01
1.87538296e-01 4.42586452e-01 6.88327014e-01 -1.13988233e+00
-9.66338992e-01 -1.15960956e+00 6.61865115e-01 -5.65088987e-01
4.97160047e-01 -3.44177723e-01 -9.21557963e-01 2.80329883e-01
2.03700095e-01 -5.52310869e-02 -2.78901935e-01 -7.04022765e-01
-1.77643523e-01 -4.46913345e-03 -1.29294634e+00 7.53179133e-01
1.11359012e+00 -4.02865797e-01 -5.87485507e-02 5.80214858e-01
7.28646576e-01 -5.95921695e-01 -4.13630962e-01 -1.99809462e-01
1.52868718e-01 -9.20971215e-01 8.85221958e-01 -1.35610610e-01
9.79584515e-01 -2.77219921e-01 8.60351250e-02 -1.31000781e+00
-4.07924742e-01 -3.49203408e-01 -1.45333126e-01 1.18931389e+00
1.84968099e-01 1.05986884e-02 8.71650457e-01 6.15291893e-01
1.78814545e-01 -4.18171972e-01 -6.93864107e-01 -4.83152777e-01
3.23607661e-02 -4.94122952e-01 5.82588017e-01 6.60763025e-01
-6.76119447e-01 2.45542973e-01 -5.61204910e-01 6.13729954e-02
8.58914673e-01 -3.28091562e-01 7.07952976e-01 -7.11227655e-01
-3.68801951e-01 -4.71025825e-01 6.36007637e-02 -1.69814169e+00
-2.62331635e-01 -6.42660499e-01 1.79676861e-01 -1.90017903e+00
5.03355980e-01 -3.34500164e-01 2.26556897e-01 7.27581531e-02
-5.96347451e-01 4.17748779e-01 3.78513783e-01 1.99604601e-01
-5.71602941e-01 4.45429146e-01 1.96726167e+00 -7.31934607e-01
-3.37553658e-02 -1.77064836e-01 -6.23694062e-01 6.16854012e-01
7.35305429e-01 -9.23925757e-01 -3.29531431e-01 -1.04965222e+00
1.96695074e-01 2.23018363e-01 5.94639122e-01 -1.12000847e+00
5.40653229e-01 -6.41028360e-02 7.24322140e-01 -7.01551318e-01
3.81906688e-01 -7.50465572e-01 -8.90613049e-02 5.08402526e-01
-4.84248012e-01 1.09666340e-01 5.53893745e-02 6.98209047e-01
-6.54110759e-02 -5.95800877e-01 8.88788283e-01 -3.70204508e-01
-4.09661770e-01 3.58774811e-01 -3.40737402e-01 1.74832925e-01
7.37927914e-01 -3.57393295e-01 -3.41801435e-01 -7.42949367e-01
-5.51877320e-01 6.11052401e-02 2.74745673e-01 3.23352844e-01
9.01142478e-01 -1.34081769e+00 -1.02073860e+00 7.97399580e-02
-2.65302479e-01 1.14101000e-01 5.44416666e-01 5.75773239e-01
-5.30610859e-01 -6.24754429e-02 -4.19263728e-02 -5.43463409e-01
-9.97687280e-01 6.08027160e-01 3.18301916e-01 -2.59662688e-01
-7.33942449e-01 1.06803501e+00 5.44508457e-01 -1.46092698e-01
3.13015550e-01 -5.93504548e-01 -2.06464510e-02 -1.36097699e-01
9.65819120e-01 4.11565661e-01 -2.16726929e-01 -4.21017528e-01
1.69441625e-01 6.47476137e-01 -1.89956829e-01 -4.33179915e-01
1.11691773e+00 -1.41081139e-01 8.56874287e-02 1.24435455e-01
8.77846599e-01 -1.28330559e-01 -1.84688926e+00 -1.29116118e-01
-3.64968151e-01 -8.45279694e-01 1.41257077e-01 -8.69107902e-01
-1.54608977e+00 1.10550559e+00 7.02332854e-01 1.08501963e-01
1.15275407e+00 -2.65137196e-01 1.03256929e+00 1.11153863e-01
5.92966564e-02 -1.09569526e+00 5.74660122e-01 3.26944560e-01
9.80356455e-01 -1.09678090e+00 -1.18055850e-01 -2.95699030e-01
-6.98955297e-01 9.32746947e-01 7.87836373e-01 -1.33950382e-01
1.12779692e-01 5.25512338e-01 9.57297087e-02 -2.38845125e-01
-5.58086157e-01 4.89063524e-02 7.35664070e-02 9.45209980e-01
4.03647535e-02 -4.09310341e-01 -7.20472783e-02 2.65478730e-01
1.08782254e-01 1.52068108e-01 6.23919487e-01 5.02689302e-01
-5.80269992e-01 -6.43632650e-01 -8.88806164e-01 7.54958153e-01
-4.54041779e-01 -4.57059920e-01 -2.31426880e-01 1.68061763e-01
5.40351868e-02 1.11643445e+00 1.72602430e-01 -1.35877103e-01
4.03840154e-01 -3.68499011e-01 6.82867467e-01 -4.88045901e-01
-6.67007923e-01 2.93912785e-03 -3.77761364e-01 -2.07405642e-01
-1.24292992e-01 -4.19686347e-01 -1.19935334e+00 -6.44151568e-01
-3.69580388e-01 -1.70921683e-01 6.47046387e-01 5.16146302e-01
3.40829268e-02 8.02673817e-01 3.24482232e-01 -8.59757483e-01
-6.52761281e-01 -1.03689981e+00 -6.37015343e-01 7.23854542e-01
2.88831275e-02 -2.90094346e-01 -3.87869686e-01 3.93300831e-01] | [11.428325653076172, -0.34015366435050964] |
4e2b1f9b-b35c-4c38-b8b3-f21e56a3b558 | fedhgn-a-federated-framework-for | 2305.09729 | null | https://arxiv.org/abs/2305.09729v1 | https://arxiv.org/pdf/2305.09729v1.pdf | FedHGN: A Federated Framework for Heterogeneous Graph Neural Networks | Heterogeneous graph neural networks (HGNNs) can learn from typed and relational graph data more effectively than conventional GNNs. With larger parameter spaces, HGNNs may require more training data, which is often scarce in real-world applications due to privacy regulations (e.g., GDPR). Federated graph learning (FGL) enables multiple clients to train a GNN collaboratively without sharing their local data. However, existing FGL methods mainly focus on homogeneous GNNs or knowledge graph embeddings; few have considered heterogeneous graphs and HGNNs. In federated heterogeneous graph learning, clients may have private graph schemas. Conventional FL/FGL methods attempting to define a global HGNN model would violate schema privacy. To address these challenges, we propose FedHGN, a novel and general FGL framework for HGNNs. FedHGN adopts schema-weight decoupling to enable schema-agnostic knowledge sharing and employs coefficients alignment to stabilize the training process and improve HGNN performance. With better privacy preservation, FedHGN consistently outperforms local training and conventional FL methods on three widely adopted heterogeneous graph datasets with varying client numbers. The code is available at https://github.com/cynricfu/FedHGN . | ['Irwin King', 'Xinyu Fu'] | 2023-05-16 | null | null | null | null | ['knowledge-graph-embeddings', 'knowledge-graph-embeddings'] | ['graphs', 'methodology'] | [-2.70814568e-01 5.18667698e-01 -5.84631085e-01 -4.45400864e-01
-3.53634000e-01 -8.86899054e-01 2.58785337e-01 5.53354919e-02
-1.58104241e-01 7.70487487e-01 5.03259227e-02 -4.94744182e-01
-2.26848915e-01 -1.29066885e+00 -7.29274035e-01 -4.85513240e-01
1.37795702e-01 5.06403804e-01 7.59312063e-02 -9.88974422e-02
-3.73373389e-01 5.15738845e-01 -1.04172444e+00 2.09553286e-01
7.26180375e-01 7.48151124e-01 -3.01317841e-01 3.32363814e-01
-3.52282047e-01 1.03228951e+00 -2.96680748e-01 -1.27440464e+00
8.37071717e-01 -6.32979572e-02 -7.26195931e-01 -5.11416495e-01
8.02163482e-01 -2.02364475e-01 -8.64192903e-01 1.35778773e+00
5.81248701e-01 7.69317299e-02 -5.52403566e-04 -1.75634813e+00
-1.40592623e+00 1.20393860e+00 -4.78367805e-02 -2.79697359e-01
1.11928113e-01 1.10204764e-01 1.24929667e+00 -2.44527012e-01
9.30182099e-01 1.00459099e+00 8.43725145e-01 8.28279674e-01
-1.18926716e+00 -9.03378069e-01 3.26171219e-01 1.38602644e-01
-1.46509910e+00 -2.44149610e-01 7.44566023e-01 2.42142156e-02
9.49030876e-01 5.00251949e-01 6.79553628e-01 1.18231773e+00
-9.91229564e-02 5.41731119e-01 6.37798250e-01 -9.62437876e-03
1.70568600e-01 2.56517261e-01 1.42021343e-01 7.64784575e-01
9.24374640e-01 -4.91030999e-02 -6.28710926e-01 -4.44097430e-01
6.56273246e-01 2.84779489e-01 -4.34661299e-01 -1.08849406e+00
-6.96112633e-01 9.27671790e-01 7.93074787e-01 1.67236328e-01
-1.99133474e-02 2.35030487e-01 8.26305568e-01 8.36758018e-01
3.71467054e-01 2.51250774e-01 -5.61801791e-01 3.71791065e-01
-1.89521194e-01 3.70460041e-02 1.32281756e+00 1.49547112e+00
1.09229469e+00 -1.01269130e-02 2.27382407e-01 5.64883471e-01
2.36595035e-01 3.24603498e-01 2.03490973e-01 -6.28497541e-01
7.11634040e-01 1.21975374e+00 -6.68289661e-01 -1.20754230e+00
-6.91072792e-02 -3.53921264e-01 -9.92358088e-01 -2.48628601e-01
2.87690967e-01 -2.80844942e-02 -7.60383606e-01 1.75371957e+00
3.86806011e-01 -2.72451877e-03 4.64037210e-01 7.02271342e-01
1.09267521e+00 6.59617186e-02 1.56580266e-02 2.53920138e-01
7.97168195e-01 -1.00561619e+00 -5.43845356e-01 -5.90605736e-02
1.00998509e+00 -4.62271832e-02 9.22055662e-01 -3.45264166e-03
-7.57347524e-01 1.38637215e-01 -7.05047727e-01 -3.05936068e-01
-8.41547430e-01 -4.90472198e-01 1.10136867e+00 1.00253463e+00
-1.62503517e+00 4.53971416e-01 -5.57601333e-01 -8.00803959e-01
8.25855017e-01 6.13596380e-01 -9.24918175e-01 -4.36286598e-01
-1.22736657e+00 3.53398710e-01 7.66753793e-01 5.33107892e-02
-5.41698098e-01 -8.19017708e-01 -9.34176445e-01 1.98612764e-01
7.15632737e-01 -8.99309397e-01 9.19076562e-01 -7.38474309e-01
-1.17208910e+00 8.96163166e-01 5.91304719e-01 -5.45454502e-01
5.87217569e-01 3.05110693e-01 -7.35647321e-01 -9.41006243e-02
-2.35365927e-01 1.54073223e-01 4.79325742e-01 -1.07552207e+00
-2.05063194e-01 -6.80547237e-01 2.09619164e-01 2.02483326e-01
-5.72671056e-01 -2.66936213e-01 -5.06720364e-01 -3.01328719e-01
1.12827681e-01 -8.59988630e-01 -1.53764233e-01 8.29894468e-02
-5.31794012e-01 -1.99038848e-01 1.02795291e+00 -2.18878508e-01
1.23238873e+00 -2.11908627e+00 -8.80112872e-02 6.88939691e-01
9.17850435e-01 5.00382960e-01 -2.26843432e-01 7.16317058e-01
6.60338551e-02 2.70402104e-01 1.39518008e-01 -1.56412855e-01
3.00284922e-01 5.78099906e-01 -3.94751467e-02 5.65549970e-01
-4.61205781e-01 1.32682335e+00 -7.73785412e-01 -4.41295296e-01
-1.34367287e-01 2.71249264e-01 -7.35371590e-01 2.28464752e-01
-2.96662211e-01 -6.13043196e-02 -4.57233906e-01 9.66642201e-01
8.31945777e-01 -6.90853953e-01 8.83780479e-01 -3.90980810e-01
3.81377816e-01 -1.74018368e-01 -1.05604064e+00 1.72358108e+00
-1.08749889e-01 3.03162783e-01 2.08896667e-01 -7.41587996e-01
1.01489830e+00 2.90024161e-01 3.21619332e-01 -5.58795154e-01
1.73713773e-01 1.98702976e-01 -2.28319719e-01 -2.27943003e-01
3.70595008e-01 2.11396933e-01 1.67974213e-03 5.58144689e-01
4.20323461e-01 4.28965122e-01 -1.17519788e-01 5.93932271e-01
1.53449893e+00 -3.72879028e-01 2.75451005e-01 -9.13877785e-02
3.35063636e-01 -3.04883868e-01 6.63491249e-01 9.39247966e-01
-4.42695618e-01 2.57853687e-01 5.36126494e-01 -7.50285149e-01
-7.27138579e-01 -1.16861105e+00 4.75280285e-01 1.32436669e+00
1.09207012e-01 -7.52395689e-01 -5.30742168e-01 -1.16616821e+00
5.44981301e-01 3.59505743e-01 -5.67577839e-01 -4.46062982e-01
-3.76270592e-01 -2.70344764e-01 8.40333641e-01 2.67738402e-01
4.58474606e-01 -1.06241059e+00 3.74517947e-01 8.19560792e-03
2.85744578e-01 -9.77095783e-01 -5.78283250e-01 3.51467505e-02
-5.83012640e-01 -1.39101481e+00 -7.13671669e-02 -7.68012464e-01
8.11401606e-01 3.04021627e-01 1.15418470e+00 2.67909288e-01
-6.36005923e-02 5.91315091e-01 -2.96884000e-01 -1.37393191e-01
-3.85969907e-01 5.72017372e-01 -8.58555138e-02 -4.88980971e-02
6.65517449e-01 -1.01464784e+00 -3.94984424e-01 2.57365495e-01
-1.00285840e+00 -5.04815057e-02 5.08791983e-01 7.08281994e-01
3.95373493e-01 -2.16589838e-01 5.68926871e-01 -2.02058482e+00
7.38978207e-01 -6.19172454e-01 -7.28487670e-01 7.12378383e-01
-1.15041053e+00 -4.96944375e-02 9.68932033e-01 -4.07706171e-01
-7.69105077e-01 -2.95692027e-01 1.58668354e-01 -9.22928631e-01
6.22673407e-02 5.72211266e-01 -5.54757416e-01 -7.50507534e-01
7.71259606e-01 -1.77383013e-02 1.17587060e-01 -3.64295363e-01
6.61387265e-01 3.54493588e-01 3.85596931e-01 -6.50773585e-01
1.00075865e+00 2.69550681e-01 -1.27668113e-01 -9.22006220e-02
-3.60639095e-01 -4.20072198e-01 -2.18234718e-01 -5.53274974e-02
5.21689892e-01 -8.67019773e-01 -1.16876245e+00 3.53022486e-01
-6.69955194e-01 -4.52189058e-01 -2.65433133e-01 2.33102053e-01
-3.28024745e-01 2.50897169e-01 -6.48951650e-01 -3.53944629e-01
-6.66187346e-01 -6.92674518e-01 3.92819703e-01 2.15530545e-01
1.45809874e-01 -1.20175040e+00 1.26816660e-01 4.44154829e-01
8.66674364e-01 2.94001877e-01 9.21752870e-01 -1.26859534e+00
-9.89731014e-01 -1.59385040e-01 -5.04382133e-01 1.05292976e-01
2.88319349e-01 -2.84924895e-01 -8.04547250e-01 -7.07556605e-01
-4.37142938e-01 -5.05661964e-01 3.86052161e-01 -3.07452530e-01
1.07763064e+00 -8.84475052e-01 -2.52788812e-01 1.26978719e+00
1.90240216e+00 5.31336665e-02 3.65916610e-01 2.97073513e-01
1.43764019e+00 3.92981023e-01 -1.49409369e-01 2.76403129e-01
6.52385235e-01 2.06981242e-01 6.67478919e-01 6.12917468e-02
5.57224937e-02 -6.38330400e-01 6.48912266e-02 8.33220482e-01
8.33306983e-02 -5.19749761e-01 -8.21357012e-01 3.32569182e-01
-2.14576411e+00 -7.03423202e-01 1.28691778e-01 1.81843305e+00
5.33409238e-01 -3.70139182e-01 -5.21067530e-02 -6.24459028e-01
8.55865300e-01 4.47841913e-01 -1.03858161e+00 -3.90697002e-01
-4.19196427e-01 1.43001676e-01 1.07479954e+00 6.44111633e-02
-7.03663886e-01 1.11580741e+00 4.72860193e+00 4.41113234e-01
-9.00396585e-01 2.91243255e-01 2.13037089e-01 -1.84803888e-01
-8.42579484e-01 1.56048700e-01 -5.45291960e-01 2.31388994e-02
1.00673068e+00 -8.67441714e-01 1.01153505e+00 1.22060478e+00
-6.69622600e-01 8.28474641e-01 -1.04458010e+00 1.12406564e+00
-1.04237512e-01 -1.63630474e+00 2.79713333e-01 2.10910767e-01
7.29295135e-01 5.67082703e-01 -1.54331122e-02 6.51965082e-01
1.08340693e+00 -9.57589984e-01 2.86218286e-01 3.04013014e-01
7.12154746e-01 -8.21954131e-01 5.81076741e-01 7.09717628e-03
-1.30470443e+00 -2.33458936e-01 -5.82068980e-01 4.36395198e-01
-2.35889614e-01 1.07657462e-01 -8.20378959e-01 1.19525206e+00
7.44999349e-01 7.32620835e-01 -7.76112795e-01 5.45146644e-01
7.47124404e-02 3.84108156e-01 -2.80989677e-01 9.98343229e-02
6.84504956e-02 -5.01520455e-01 2.41889000e-01 7.39915490e-01
2.07783923e-01 1.14768324e-02 7.31256232e-02 8.74239743e-01
-8.20024371e-01 4.16028738e-01 -1.13358104e+00 -5.46627164e-01
7.50724435e-01 1.24872482e+00 -2.57587403e-01 6.39686584e-02
-7.45595396e-01 8.06457222e-01 1.09356737e+00 6.03605092e-01
-5.33865809e-01 -2.65688509e-01 9.87664461e-01 -4.67460640e-02
1.86821863e-01 1.36487782e-01 2.74417371e-01 -1.56959391e+00
6.69277413e-03 -1.12550938e+00 1.15295076e+00 -4.01677877e-01
-1.76205587e+00 8.17955673e-01 -3.72860759e-01 -7.45612323e-01
4.79580872e-02 -4.31731462e-01 -5.04138708e-01 6.63514853e-01
-1.23222613e+00 -1.69009614e+00 -3.32625926e-01 1.28318858e+00
-5.73150754e-01 -5.41743577e-01 9.54723001e-01 4.34158742e-01
-6.25057280e-01 1.33715987e+00 7.88911209e-02 5.06142020e-01
7.41865933e-01 -1.08096564e+00 5.22907794e-01 8.13814521e-01
1.50789782e-01 1.08399403e+00 1.81556880e-01 -9.22193348e-01
-2.01734471e+00 -1.65259993e+00 5.79826176e-01 -2.19957292e-01
8.33139181e-01 -7.71675944e-01 -1.21933734e+00 1.45679712e+00
-6.25732318e-02 7.86435664e-01 9.44898069e-01 3.58281404e-01
-1.04463780e+00 -5.23824215e-01 -1.63645911e+00 6.34847224e-01
1.57032084e+00 -1.03787577e+00 3.40542197e-01 4.64866102e-01
1.28393567e+00 -5.21486163e-01 -1.40790534e+00 1.53460920e-01
5.08043170e-01 -1.04064989e+00 6.95445836e-01 -1.00035942e+00
-4.72598135e-01 -3.01876664e-01 -4.05046612e-01 -9.70166862e-01
-3.22371751e-01 -8.75213981e-01 -3.85483205e-01 1.35192299e+00
3.30322742e-01 -1.51699710e+00 1.45764732e+00 1.11682177e+00
7.61326775e-03 -5.70243239e-01 -7.91006386e-01 -8.08726370e-01
-2.85854369e-01 -1.37565270e-01 1.35580862e+00 1.48735559e+00
-2.23473504e-01 -1.50586978e-01 -2.41582066e-01 5.10601461e-01
9.19705331e-01 1.14555210e-01 1.03941154e+00 -1.11118937e+00
-2.23572731e-01 -2.46205673e-01 -7.84519255e-01 -6.48571923e-02
5.24412096e-01 -1.62943304e+00 -7.58938909e-01 -1.50022233e+00
4.91694808e-02 -5.96940517e-01 -5.98456502e-01 1.11198509e+00
2.94763952e-01 3.04748137e-02 2.03549221e-01 -6.31639501e-03
-8.40841651e-01 3.42566758e-01 9.40453351e-01 -1.30356655e-01
-1.13086782e-01 -8.21860135e-02 -1.16428912e+00 1.24367885e-01
9.73979652e-01 -3.62959236e-01 -9.57564175e-01 -5.00115156e-01
4.79259670e-01 -9.12264660e-02 3.31677586e-01 -6.14730775e-01
7.49097109e-01 -2.27362394e-01 -1.29603103e-01 -8.30786899e-02
-2.92154342e-01 -1.17638147e+00 1.15080774e+00 3.50885466e-02
-2.34656140e-01 1.57701746e-01 -6.72911257e-02 7.86687791e-01
-4.76677269e-02 3.25167388e-01 4.79124814e-01 -2.48566464e-01
-6.68718636e-01 1.04205513e+00 4.40796673e-01 1.40306056e-01
9.64052916e-01 -1.75881252e-01 -9.99495625e-01 -3.34015191e-01
-6.48217916e-01 3.26745957e-01 9.88954723e-01 4.26267415e-01
5.90151429e-01 -1.42265582e+00 -2.82700002e-01 4.89471793e-01
4.96349454e-01 3.14560771e-01 4.24528092e-01 5.60886681e-01
-5.71589112e-01 1.65071353e-01 -1.30202502e-01 -6.46008700e-02
-1.26134717e+00 8.52125287e-01 4.51239318e-01 -3.26355755e-01
-6.53360784e-01 9.36318636e-01 1.92593753e-01 -1.11472082e+00
4.72840130e-01 -7.54693970e-02 2.49026671e-01 -3.30774814e-01
1.45809904e-01 3.20683181e-01 2.07121283e-01 -4.46801931e-01
-2.47099295e-01 -1.41368747e-01 -5.56561768e-01 6.68524563e-01
1.34797299e+00 -2.70233452e-02 -4.98481750e-01 -1.10633127e-01
1.48115182e+00 -5.95426001e-02 -8.67469132e-01 -5.22239745e-01
-2.97091529e-02 -6.58186078e-01 -2.92919099e-01 -6.56116605e-01
-1.62651956e+00 1.18580438e-01 2.36409232e-01 1.92984119e-01
1.13577223e+00 1.41468465e-01 7.63329983e-01 6.49824798e-01
8.06927502e-01 -6.05930626e-01 -5.14195621e-01 1.92221716e-01
4.67650086e-01 -1.06832373e+00 -9.26933438e-02 -4.76539731e-01
-3.94255549e-01 9.37537313e-01 1.06955671e+00 1.72078326e-01
6.92133009e-01 9.14258044e-03 2.46833771e-01 -4.79115367e-01
-8.22463512e-01 1.61230236e-01 6.72837719e-02 8.25993717e-01
-2.07761094e-01 1.45910516e-01 1.83947757e-01 6.60232186e-01
-1.99665621e-01 -3.47479284e-02 2.69611627e-01 8.97033572e-01
2.92013794e-01 -1.47775412e+00 3.83166224e-01 7.01524377e-01
-2.98888773e-01 -3.48520800e-02 -5.88357866e-01 8.92600060e-01
-5.29494509e-02 4.90314811e-01 -3.87806058e-01 -7.08715081e-01
3.95155013e-01 -1.38023347e-01 2.47559190e-01 -6.06667876e-01
-1.03725863e+00 -6.73572838e-01 2.17269897e-01 -9.65126753e-01
-6.98169461e-03 -2.01971039e-01 -1.02732229e+00 -1.05656791e+00
-2.46209323e-01 2.25486025e-01 2.98125565e-01 1.56361625e-01
7.19319224e-01 7.32121617e-02 3.77487630e-01 -9.89796147e-02
-3.90325010e-01 -3.48535866e-01 -1.03278756e+00 6.30987108e-01
4.64076661e-02 -1.21354364e-01 -4.88543063e-01 -5.60098529e-01] | [6.0009236335754395, 6.8851823806762695] |
0d28ed43-b34f-4bfc-a867-c8c01272e300 | inter-species-cell-detection-datasets-on | 2108.08529 | null | https://arxiv.org/abs/2108.08529v1 | https://arxiv.org/pdf/2108.08529v1.pdf | Inter-Species Cell Detection: Datasets on pulmonary hemosiderophages in equine, human and feline specimens | Pulmonary hemorrhage (P-Hem) occurs among multiple species and can have various causes. Cytology of bronchoalveolarlavage fluid (BALF) using a 5-tier scoring system of alveolar macrophages based on their hemosiderin content is considered the most sensitive diagnostic method. We introduce a novel, fully annotated multi-species P-Hem dataset which consists of 74 cytology whole slide images (WSIs) with equine, feline and human samples. To create this high-quality and high-quantity dataset, we developed an annotation pipeline combining human expertise with deep learning and data visualisation techniques. We applied a deep learning-based object detection approach trained on 17 expertly annotated equine WSIs, to the remaining 39 equine, 12 human and 7 feline WSIs. The resulting annotations were semi-automatically screened for errors on multiple types of specialised annotation maps and finally reviewed by a trained pathologists. Our dataset contains a total of 297,383 hemosiderophages classified into five grades. It is one of the largest publicly availableWSIs datasets with respect to the number of annotations, the scanned area and the number of species covered. | ['Christof A. Bertram', 'Katharina Breininger', 'Robert Klopfleisch', 'Andreas Maier', 'Marc Aubreville', 'Jörn Voigt', 'Frauke Wilm', 'Lutz Welker', 'Dorothee Bienzle', 'Jason Stayt', 'Jenny Hill', 'Christian Marzahl'] | 2021-08-19 | null | null | null | null | ['cell-detection'] | ['computer-vision'] | [-4.18242849e-02 -5.37499450e-02 1.95535362e-01 8.16363543e-02
-6.25608742e-01 -6.89261615e-01 4.65969771e-01 5.93699753e-01
-8.46802175e-01 8.52688968e-01 -3.11224461e-01 -2.11118400e-01
3.07306275e-02 -6.34387553e-01 -4.30144787e-01 -7.60143936e-01
3.84539813e-02 1.06441116e+00 8.80020559e-01 3.52187485e-01
2.51199365e-01 1.09831917e+00 -1.49308467e+00 4.04137939e-01
7.16154218e-01 7.39970863e-01 2.87896514e-01 1.22653759e+00
-2.80397773e-01 9.19479966e-01 -6.04489744e-01 -5.42411923e-01
6.30563796e-02 -3.54902655e-01 -7.56334901e-01 -2.13133559e-01
4.83707011e-01 -3.33040237e-01 4.41896230e-01 5.72790444e-01
6.23142362e-01 -4.12145406e-01 1.09411812e+00 -1.19422281e+00
-2.32517198e-01 -7.52347633e-02 -6.04912341e-01 7.30230689e-01
-3.06408137e-01 4.78098452e-01 5.63505530e-01 -9.64638054e-01
1.04309857e+00 1.06437910e+00 9.50661719e-01 5.32665253e-01
-9.31395948e-01 -7.57986188e-01 -1.10446203e+00 3.88081491e-01
-1.16959763e+00 1.40979528e-01 -2.16324627e-01 -1.15767050e+00
9.74923551e-01 2.82459319e-01 9.17073131e-01 3.22551310e-01
1.89473659e-01 3.16303164e-01 1.49911940e+00 -1.62328079e-01
1.79346278e-01 3.45566928e-01 -2.14008093e-01 1.10392082e+00
6.98276818e-01 -2.57570267e-01 -3.55650276e-01 -4.31737125e-01
6.87134445e-01 1.45104736e-01 -1.84807390e-01 -2.42943376e-01
-1.32903087e+00 7.23234653e-01 3.41093838e-01 1.23120941e-01
-1.66131377e-01 -1.98659122e-01 5.40442288e-01 -1.76780894e-01
-3.03636696e-02 5.71593761e-01 -2.34790787e-01 1.53558448e-01
-6.65415287e-01 2.25321278e-02 7.74067760e-01 2.68454850e-01
5.95673919e-01 -2.98308849e-01 -2.11031184e-01 1.01281452e+00
3.83166552e-01 5.76847613e-01 4.68390077e-01 -8.07582438e-01
-2.00349078e-01 9.36202228e-01 8.19040239e-02 -4.46096361e-01
-4.65524524e-01 2.32145876e-01 -6.06026590e-01 6.96782589e-01
5.12678862e-01 1.32573827e-03 -1.08441651e+00 7.92468786e-01
6.95389926e-01 -8.78692195e-02 -3.44102144e-01 8.61101866e-01
1.01597643e+00 4.20442641e-01 4.07506734e-01 1.43052042e-01
1.66497457e+00 -8.76801133e-01 -5.70869803e-01 3.96267086e-01
6.36573732e-01 -7.35241830e-01 8.57840717e-01 4.11693871e-01
-9.09680486e-01 -1.50824115e-02 -1.21613491e+00 -2.92491823e-01
-8.20107758e-01 2.36866474e-01 -3.71223465e-02 2.89761841e-01
-1.05706704e+00 3.33463699e-01 -8.40320110e-01 -8.63263190e-01
9.77113187e-01 2.31600091e-01 -6.57830238e-01 3.85210812e-01
-6.85481012e-01 1.37571537e+00 3.21760654e-01 -2.67927289e-01
-8.05269063e-01 -1.10712981e+00 -2.34230697e-01 -1.94833711e-01
6.89053014e-02 -5.84769189e-01 1.12643600e+00 -1.80592351e-02
-1.04554725e+00 1.62984943e+00 1.76942170e-01 -4.94206786e-01
9.03499722e-01 8.27559605e-02 -1.53383851e-01 8.20500791e-01
6.71177581e-02 8.45070779e-01 1.59708053e-01 -1.22844887e+00
-1.09939325e+00 -2.54146725e-01 -7.15832472e-01 -2.25273743e-01
-2.00508386e-02 2.12779313e-01 -2.32749417e-01 -4.91924107e-01
-7.42994189e-01 -7.27149785e-01 7.23045915e-02 8.53002429e-01
-2.01952621e-01 -4.72506225e-01 9.11665261e-01 -9.04782414e-01
9.76321638e-01 -1.93795109e+00 -2.54755288e-01 2.36446530e-01
7.00605929e-01 8.96087706e-01 1.90154374e-01 4.40397054e-01
3.02002549e-01 4.08130109e-01 -2.84939885e-01 -9.95651856e-02
6.88691884e-02 1.83961362e-01 2.53628343e-01 7.22504199e-01
4.71776158e-01 7.78500259e-01 -8.89112771e-01 -1.30026662e+00
3.53326440e-01 7.46822357e-01 -2.05447704e-01 5.56843400e-01
8.12655017e-02 2.84124941e-01 2.56175734e-02 6.77481592e-01
6.06605768e-01 -5.53418934e-01 3.15589085e-03 -3.37133296e-02
-1.42984539e-01 -3.55518073e-01 -6.88262105e-01 9.27437067e-01
-2.00170487e-01 7.88617611e-01 3.72739136e-01 -4.20390293e-02
6.45119131e-01 1.75675258e-01 2.72945464e-01 1.17933881e-02
1.74356416e-01 6.65252626e-01 1.39337769e-02 -9.61427629e-01
-7.51115009e-02 -4.34476912e-01 6.27350450e-01 4.94276106e-01
1.50948688e-01 -1.65996507e-01 6.49162471e-01 5.01048379e-02
1.25484741e+00 -3.31690103e-01 9.34057534e-01 -5.02694011e-01
6.62367046e-01 3.38508874e-01 2.54108876e-01 2.80395389e-01
-5.43237150e-01 7.29985297e-01 6.69270575e-01 -5.53816319e-01
-1.38647676e+00 -1.02823687e+00 -6.94734514e-01 7.34182358e-01
-3.29348058e-01 1.45601630e-01 -8.01568806e-01 -6.73192501e-01
3.53674650e-01 1.89525560e-01 -9.81998324e-01 4.74260181e-01
-4.75625277e-01 -4.85999018e-01 9.39112306e-01 4.67610300e-01
5.72557569e-01 -1.21545887e+00 -1.16713548e+00 -8.32314715e-02
1.99727133e-01 -4.68252152e-01 -8.00351799e-02 1.52156159e-01
-2.53884196e-01 -1.75495505e+00 -8.81879449e-01 -8.59245479e-01
5.45673013e-01 -2.04719752e-01 8.08568776e-01 4.70783681e-01
-1.29499567e+00 -1.55816302e-01 -2.30822954e-02 -8.27070236e-01
-7.56241024e-01 -2.13230982e-01 -3.78133893e-01 -4.86970603e-01
6.80026054e-01 3.71852592e-02 -6.84792459e-01 2.59724140e-01
-1.19426310e+00 -2.66121149e-01 9.71987903e-01 7.67256856e-01
1.03695202e+00 -4.95517313e-01 5.42751014e-01 -1.22149312e+00
2.15521678e-01 -6.90172553e-01 -4.99876976e-01 3.95489097e-01
-3.80082786e-01 -3.79411131e-01 4.03468698e-01 -1.64013550e-01
-8.54565680e-01 -2.55655676e-01 -1.26193359e-01 -4.36600506e-01
-5.13649702e-01 1.75371245e-01 3.65598947e-01 -2.23407611e-01
8.90485406e-01 -3.57963979e-01 4.02738512e-01 -3.03692490e-01
-1.38277739e-01 9.13962960e-01 7.04024494e-01 2.69482970e-01
4.74606693e-01 7.34910250e-01 3.51017207e-01 -9.98597085e-01
-5.05625904e-01 -1.02473557e+00 -7.50368834e-01 -3.39989007e-01
1.34510696e+00 -5.35129189e-01 -6.96743071e-01 6.58224523e-01
-9.24903810e-01 -6.85503483e-01 -3.06001514e-01 4.13503885e-01
-2.54368752e-01 5.01635432e-01 -7.91898906e-01 -5.69520473e-01
-7.85561323e-01 -8.02144706e-01 1.07889676e+00 5.62370829e-02
-5.67251503e-01 -1.05267191e+00 5.83038151e-01 3.93998593e-01
4.90060985e-01 6.56040430e-01 1.34836900e+00 -1.02451634e+00
-1.55208975e-01 -3.42873096e-01 -4.87966001e-01 1.81236103e-01
-1.34985760e-01 5.77263415e-01 -7.63412595e-01 3.26031670e-02
-1.89633906e-01 -4.84887570e-01 7.29606926e-01 -9.97646078e-02
1.09141338e+00 -2.86105841e-01 -4.64505523e-01 3.00530195e-01
1.70078576e+00 1.39797479e-01 5.46135306e-01 4.09195840e-01
4.71895874e-01 8.09129000e-01 5.81840813e-01 3.66438389e-01
3.34898159e-02 7.67586157e-02 4.11673754e-01 -2.57214755e-01
-3.56889099e-01 2.15683341e-01 -2.13972911e-01 3.53439599e-01
-3.42088550e-01 -5.30936658e-01 -1.37280595e+00 8.69893014e-01
-1.15165567e+00 -7.22757638e-01 -5.88566482e-01 1.97911990e+00
8.84136677e-01 -4.91405249e-01 3.31033409e-01 -1.71325728e-01
9.69518304e-01 -6.64492488e-01 -6.20020151e-01 -2.10604563e-01
-5.90383308e-03 4.40662116e-01 5.49229562e-01 1.26212969e-01
-9.72644806e-01 3.95059973e-01 6.64441395e+00 7.07459390e-01
-1.02445757e+00 2.32120365e-01 4.96109545e-01 -3.70808572e-01
1.87340736e-01 -8.38967264e-01 -8.39058101e-01 3.83443624e-01
9.29740965e-01 -5.61204739e-02 3.52458544e-02 5.93780398e-01
-1.24907188e-01 -2.90290356e-01 -8.78698826e-01 5.79084337e-01
2.48892784e-01 -1.68525863e+00 -4.81396951e-02 2.88107693e-01
3.95586431e-01 5.55711269e-01 -2.65428603e-01 1.53092369e-02
3.48366499e-01 -1.10110986e+00 3.36971134e-01 5.55071175e-01
1.27087510e+00 -5.32821000e-01 1.22131169e+00 9.84569490e-02
-8.76014352e-01 1.58602968e-01 -2.06311211e-01 7.60491192e-01
-3.59627306e-02 1.57785177e-01 -1.48310840e+00 -2.13538319e-01
1.18005157e+00 2.81443447e-01 -1.01870620e+00 1.49822056e+00
-3.32835279e-02 4.67651844e-01 -4.74647403e-01 -3.12891692e-01
-4.58319709e-02 1.17562436e-01 2.12479919e-01 1.72009361e+00
4.49580073e-01 -4.74027134e-02 -3.14845204e-01 7.31634259e-01
1.19770644e-03 4.38915074e-01 -2.90071785e-01 -1.61508694e-01
5.93861878e-01 1.85029685e+00 -1.11647141e+00 -5.52087545e-01
-2.23114237e-01 3.71556044e-01 3.78091216e-01 -1.11692287e-01
-7.36340582e-01 -6.38738871e-01 3.07315022e-01 7.33947277e-01
3.59194070e-01 5.58574677e-01 -8.03222507e-02 -4.63949054e-01
-4.55797613e-01 -4.00757164e-01 8.89247119e-01 -8.69914234e-01
-1.68858683e+00 3.52553010e-01 3.68286520e-02 -1.11575067e+00
5.06912358e-02 -9.60989714e-01 -5.56195319e-01 8.04381907e-01
-1.66397035e+00 -1.24111974e+00 -5.66446424e-01 2.20466122e-01
1.67190716e-01 -1.42624542e-01 7.62504101e-01 1.42760143e-01
-4.27664906e-01 1.32920384e-01 1.19541802e-01 -4.23915200e-02
1.06110072e+00 -1.64358246e+00 -4.34202373e-01 2.05337167e-01
-8.79795015e-01 6.94999754e-01 2.59836853e-01 -7.43390560e-01
-7.79610038e-01 -1.45767379e+00 8.09788465e-01 -4.51314569e-01
8.65619719e-01 3.76297571e-02 -1.25444484e+00 5.64619184e-01
5.61705291e-01 3.32403868e-01 1.50441086e+00 -9.38923538e-01
-9.67303067e-02 4.06240135e-01 -1.81971824e+00 1.48229450e-01
3.94885331e-01 -2.92012125e-01 -5.89957118e-01 6.62340820e-01
-1.63346723e-01 -3.44431549e-01 -1.33528686e+00 2.35913604e-01
5.43763816e-01 -9.62472141e-01 6.18350387e-01 -4.03834134e-01
6.55914724e-01 -6.12083554e-01 5.06666124e-01 -7.30134130e-01
-2.25663841e-01 3.92277092e-02 -1.56453289e-02 8.39038432e-01
2.25004390e-01 -6.96532309e-01 6.21954918e-01 2.01397225e-01
-2.33862579e-01 -9.55550432e-01 -9.36050475e-01 -3.21979731e-01
3.62573326e-01 7.32629895e-01 5.01792550e-01 6.90498829e-01
-3.62839922e-02 9.07885730e-02 3.10779214e-01 -2.56747425e-01
4.71641809e-01 -1.21230274e-01 8.54625344e-01 -1.57888019e+00
4.72877966e-03 -6.53728545e-01 -6.08846486e-01 1.88069865e-01
-2.09306911e-01 -9.09722745e-01 2.07647979e-01 -1.87557590e+00
4.97658134e-01 -1.15562998e-01 1.45357158e-02 6.06762230e-01
-7.02978745e-02 7.67224610e-01 -3.02520543e-01 6.62509382e-01
-5.16863883e-01 -3.84492934e-01 1.19052315e+00 3.49635810e-01
5.27612388e-01 -5.67382991e-01 -2.01787934e-01 8.72578561e-01
7.63796628e-01 -3.82305712e-01 2.07866684e-01 1.27024233e-01
2.14246958e-01 -5.49348354e-01 5.22854388e-01 -1.01076901e+00
2.10506216e-01 -1.19337678e-01 5.23340285e-01 -9.51805353e-01
-7.90118277e-02 -6.53467119e-01 4.38989729e-01 9.33661044e-01
-1.97440714e-01 -2.85312623e-01 1.39476418e-01 6.38417780e-01
-9.48569849e-02 -3.15485835e-01 1.57374191e+00 -3.85554641e-01
-5.80114782e-01 1.82571545e-01 -1.00394905e+00 -1.55254543e-01
1.62533295e+00 -4.05949056e-01 -8.59092474e-01 4.62775767e-01
-4.39502627e-01 3.29404861e-01 9.13460553e-01 -3.21832746e-01
3.06525588e-01 -9.39710021e-01 -8.33719850e-01 1.20422527e-01
4.70426083e-01 4.06575739e-01 3.93317372e-01 1.20757127e+00
-1.71549344e+00 5.14385343e-01 -6.59794569e-01 -7.77822018e-01
-1.48947728e+00 3.56853515e-01 4.67780471e-01 -4.80502665e-01
-4.64305371e-01 9.43384886e-01 9.69339460e-02 -5.82276583e-01
-1.43607641e-02 -1.85956910e-01 -7.38281250e-01 3.39587539e-01
6.01236641e-01 9.42201257e-01 2.10009068e-01 -7.81744421e-01
-3.28285009e-01 3.34647208e-01 -3.64677073e-03 2.56587625e-01
1.30775177e+00 1.63806543e-01 -5.11269569e-01 4.76331383e-01
1.21479452e+00 -6.82678893e-02 -9.58569646e-01 2.39768222e-01
-1.01966396e-01 -4.09803957e-01 -3.23718816e-01 -9.26994503e-01
-8.79912794e-01 1.06376708e+00 7.92932451e-01 1.23916119e-01
6.40234828e-01 1.25061914e-01 6.64339066e-01 2.27176219e-01
4.38326187e-02 -9.85005260e-01 7.91747421e-02 1.30287334e-01
7.93749094e-01 -1.33156693e+00 6.52573854e-02 -4.63252127e-01
-5.49557149e-01 1.17983377e+00 7.52703965e-01 5.22091128e-02
4.54475313e-01 4.42625433e-01 2.84852177e-01 -5.44251800e-01
-8.22623968e-01 2.16971133e-02 3.98423849e-03 7.49746680e-01
5.69688261e-01 -2.91704297e-01 -4.54515398e-01 1.87273845e-01
-7.07467422e-02 3.09597433e-01 3.39086890e-01 1.04140484e+00
-7.37712681e-01 -4.06081796e-01 -2.77272671e-01 9.70096946e-01
-8.05799484e-01 3.48006755e-01 -7.23917484e-01 1.10036802e+00
3.73497218e-01 4.17288333e-01 1.71842873e-01 3.24055791e-01
8.88432339e-02 -9.35847312e-02 1.57512903e-01 -6.65583551e-01
-8.98188293e-01 -4.09233151e-03 8.72687902e-03 -2.05902696e-01
-5.71359813e-01 -6.83961749e-01 -1.62626362e+00 -3.91925722e-01
-1.74097911e-01 1.14404209e-01 4.68380690e-01 7.51897573e-01
1.23428099e-01 2.11351871e-01 -1.66656643e-01 -6.19432151e-01
-7.22852871e-02 -1.20383406e+00 -8.83951008e-01 4.31392878e-01
3.27869594e-01 -6.10774815e-01 -5.95688999e-01 6.55469775e-01] | [15.044086456298828, -3.13150954246521] |
6c97b5aa-7a12-441f-9a6c-65e05759e7c4 | how-human-judgment-impairs-automated | 2003.13316 | null | https://arxiv.org/abs/2003.13316v1 | https://arxiv.org/pdf/2003.13316v1.pdf | How human judgment impairs automated deception detection performance | Background: Deception detection is a prevalent problem for security practitioners. With a need for more large-scale approaches, automated methods using machine learning have gained traction. However, detection performance still implies considerable error rates. Findings from other domains suggest that hybrid human-machine integrations could offer a viable path in deception detection tasks. Method: We collected a corpus of truthful and deceptive answers about participants' autobiographical intentions (n=1640) and tested whether a combination of supervised machine learning and human judgment could improve deception detection accuracy. Human judges were presented with the outcome of the automated credibility judgment of truthful and deceptive statements. They could either fully overrule it (hybrid-overrule condition) or adjust it within a given boundary (hybrid-adjust condition). Results: The data suggest that in neither of the hybrid conditions did the human judgment add a meaningful contribution. Machine learning in isolation identified truth-tellers and liars with an overall accuracy of 69%. Human involvement through hybrid-overrule decisions brought the accuracy back to the chance level. The hybrid-adjust condition did not deception detection performance. The decision-making strategies of humans suggest that the truth bias - the tendency to assume the other is telling the truth - could explain the detrimental effect. Conclusion: The current study does not support the notion that humans can meaningfully add to the deception detection performance of a machine learning system. | ['Bennett Kleinberg', 'Bruno Verschuere'] | 2020-03-30 | null | null | null | null | ['deception-detection'] | ['miscellaneous'] | [ 2.13658303e-01 4.01975572e-01 -3.04728419e-01 -6.21891141e-01
-7.28505731e-01 -6.28331184e-01 7.71277905e-01 3.77958834e-01
-6.53683126e-01 6.99901879e-01 1.69435367e-01 -8.81039262e-01
1.78422153e-01 -3.12321603e-01 -1.82784393e-01 -4.73091990e-01
6.11613154e-01 2.31182456e-01 -1.71163857e-01 -1.68348223e-01
1.07086873e+00 3.63519251e-01 -1.02060437e+00 5.49556673e-01
9.58601594e-01 7.90037155e-01 -8.13830197e-01 5.28942049e-01
4.06168789e-01 1.52262187e+00 -9.97145295e-01 -1.09593272e+00
2.46937498e-01 -6.84968412e-01 -7.25968599e-01 -2.14934796e-01
6.11563623e-01 -8.38692129e-01 -1.73420850e-02 9.64463830e-01
4.48046148e-01 -1.43137440e-01 7.36986279e-01 -1.47923744e+00
-8.79136980e-01 3.70602161e-01 -3.62269938e-01 4.85970348e-01
7.74955451e-01 4.94752556e-01 7.11656749e-01 -7.78541923e-01
3.20557326e-01 1.29621184e+00 8.90865564e-01 6.43980324e-01
-1.25436425e+00 -1.01472104e+00 -4.02810484e-01 4.78148133e-01
-1.08406675e+00 -9.70443130e-01 6.75882399e-01 -7.69846320e-01
7.86906004e-01 2.90889531e-01 5.70746183e-01 1.33621633e+00
4.79156375e-01 9.50212255e-02 1.77289200e+00 -4.58799064e-01
2.45869040e-01 5.99362254e-01 5.22886038e-01 4.35933203e-01
1.00719941e+00 6.30284905e-01 -5.90567291e-01 -7.95288801e-01
1.93939850e-01 -3.56714547e-01 -2.91978389e-01 3.54885370e-01
-7.90953815e-01 1.20408905e+00 3.00067693e-01 3.43360305e-01
-5.15381455e-01 -3.52879256e-01 4.94472682e-01 4.81211990e-01
3.36737871e-01 8.60019684e-01 1.03537910e-01 -3.48356247e-01
-1.11054027e+00 2.44866446e-01 1.22738004e+00 4.23305556e-02
1.50887862e-01 2.04607666e-01 -9.93146971e-02 3.17594379e-01
1.30805656e-01 4.33734477e-01 4.05221552e-01 -9.60560739e-01
2.67984450e-01 5.43199182e-01 6.37617111e-01 -1.84588134e+00
-2.47154951e-01 -1.53249487e-01 -5.80768526e-01 6.15740776e-01
8.68997514e-01 -2.81275362e-01 -3.23790491e-01 1.20928645e+00
1.52956977e-01 -5.55937588e-01 -1.37318298e-01 1.28588688e+00
2.13445619e-01 5.01090407e-01 3.02055806e-01 -3.91839802e-01
1.38050604e+00 -3.33358407e-01 -8.24691534e-01 -5.16848266e-01
7.49862015e-01 -7.41637528e-01 8.04229319e-01 6.61953568e-01
-9.74094689e-01 -2.32914954e-01 -1.39365363e+00 1.08787613e-02
8.69736448e-02 -3.42418134e-01 4.69031572e-01 1.31304169e+00
-6.35611236e-01 3.40368152e-01 -2.41260067e-01 -4.21124883e-02
5.19280255e-01 -2.24682599e-01 -3.21833372e-01 -6.50836602e-02
-1.30501723e+00 1.75756240e+00 1.20412275e-01 4.00653183e-01
-6.50766313e-01 -2.31555983e-01 -7.81455457e-01 -5.85660562e-02
3.82626891e-01 -4.86781597e-01 1.18970871e+00 -1.67371082e+00
-1.04120290e+00 1.22529745e+00 -1.01580396e-01 -6.48194253e-01
6.94097877e-01 -1.72725469e-01 -6.33456409e-01 4.01196480e-01
2.43895739e-01 8.60648528e-02 1.27935922e+00 -1.30619037e+00
-2.83849269e-01 -5.55148661e-01 -2.03320935e-01 3.41391787e-02
8.85512084e-02 5.40429890e-01 1.33311927e+00 -3.77714217e-01
4.65371422e-02 -6.66350603e-01 3.70231241e-01 -1.26369625e-01
-1.81989238e-01 -4.26632464e-01 4.07759339e-01 -8.86262357e-01
1.35679781e+00 -1.93071854e+00 -9.03330266e-01 2.50209033e-01
5.95507562e-01 5.44058442e-01 3.84585857e-01 6.91216528e-01
-8.45275372e-02 7.86431611e-01 6.00117147e-02 1.64567128e-01
1.93466365e-01 -2.01776281e-01 -2.86011159e-01 9.69397187e-01
1.34488419e-02 8.37653458e-01 -7.87819207e-01 -4.81843770e-01
-2.29879573e-01 1.09763868e-01 -1.98848009e-01 -8.07590187e-02
5.41336298e-01 -1.79174691e-01 1.78929307e-02 4.55529034e-01
6.47532582e-01 6.10584989e-02 4.04177345e-02 1.09672457e-01
-1.35451660e-01 6.43867373e-01 -5.97705960e-01 3.23318958e-01
1.88360095e-01 1.03072166e+00 4.22294319e-01 -7.36375749e-01
9.75548327e-01 4.37021524e-01 -6.00444078e-01 -6.43750012e-01
3.52938920e-01 4.93094265e-01 6.68640137e-01 -7.87651420e-01
4.47072476e-01 -9.83254433e-01 -6.46460578e-02 8.36858928e-01
-6.33588433e-01 -2.88631082e-01 -5.82047641e-01 3.72482032e-01
7.70330489e-01 -5.81405818e-01 7.73299932e-01 -2.83903837e-01
4.16959792e-01 4.44551617e-01 6.65765822e-01 1.11496973e+00
-8.57079864e-01 7.43602067e-02 6.73033178e-01 -6.63404286e-01
-8.11481178e-01 -5.88451326e-01 4.38172743e-02 9.20935690e-01
-1.63165987e-01 1.69017404e-01 -7.98760772e-01 -8.34785461e-01
6.28441051e-02 1.66335368e+00 -6.02448165e-01 -5.84334731e-01
-9.37499925e-02 -5.56634367e-01 8.81315649e-01 7.81480372e-02
6.75102711e-01 -7.24468231e-01 -1.10173666e+00 -2.52795517e-01
-3.94869685e-01 -8.24947119e-01 -4.84021634e-01 -4.29935932e-01
-5.96798897e-01 -1.15709186e+00 3.52126025e-02 -1.76805958e-01
5.95743120e-01 4.27857339e-01 7.69213796e-01 8.08116257e-01
1.25100374e-01 3.30676556e-01 -4.06068206e-01 -6.23553991e-01
-9.93413687e-01 -6.85929954e-01 4.09752578e-02 2.28649136e-02
8.20908487e-01 -2.23330125e-01 -4.91222233e-01 2.83436120e-01
-7.54036248e-01 -3.26104820e-01 4.28721488e-01 9.58080769e-01
-7.08171368e-01 -2.43423879e-01 1.02081573e+00 -5.98334908e-01
1.25295734e+00 -5.01045167e-01 -1.26799896e-01 -2.12798584e-02
-1.13479292e+00 -3.79689306e-01 2.38544077e-01 -5.32949269e-01
-1.11711586e+00 -7.45584965e-01 2.82984436e-01 2.17526466e-01
-3.20517212e-01 5.53227484e-01 1.09336004e-01 -1.70829073e-01
1.36722147e+00 -1.65837497e-01 5.02600670e-01 2.37998858e-01
-2.32457623e-01 9.47772264e-01 3.00843328e-01 -3.24278742e-01
7.27813005e-01 5.72366156e-02 -4.83614594e-01 -4.76843178e-01
-1.15088797e+00 -1.18259348e-01 -2.69014508e-01 -3.44950944e-01
4.82290208e-01 -6.25822544e-01 -1.07201552e+00 3.66123617e-01
-1.30805171e+00 -1.38771296e-01 2.76902109e-01 5.67581773e-01
-2.41225287e-02 4.97030348e-01 -5.41348040e-01 -1.52591813e+00
-4.49123502e-01 -6.30321324e-01 2.31051907e-01 7.36103281e-02
-8.49200904e-01 -8.85974586e-01 -1.88186646e-01 1.11335766e+00
4.47295249e-01 3.68841469e-01 7.85524666e-01 -1.66256642e+00
1.89257324e-01 -1.00754237e+00 -1.40955389e-01 4.49173659e-01
-2.04233620e-02 -3.00768167e-01 -9.57499683e-01 -5.73576428e-02
7.23547220e-01 -7.38738477e-01 4.00858879e-01 -1.22854955e-01
2.57823795e-01 -1.07801497e+00 1.57263502e-02 -4.42361385e-01
9.07075584e-01 2.22512588e-01 6.87797070e-01 3.26037973e-01
1.40407160e-01 8.47815692e-01 4.57171172e-01 2.04739809e-01
4.02811021e-01 4.24660921e-01 2.80010760e-01 4.53715593e-01
3.79667610e-01 -3.22992712e-01 6.65048897e-01 -2.03338861e-02
2.00773463e-01 -1.12894528e-01 -1.19961262e+00 2.95034468e-01
-1.35821474e+00 -1.56329620e+00 -5.81301808e-01 2.37267423e+00
6.87511086e-01 4.71066386e-01 8.92767981e-02 5.74326634e-01
9.48648751e-01 3.57959941e-02 -2.56728411e-01 -1.15395737e+00
3.44922021e-02 -3.29594672e-01 1.16711281e-01 7.93228745e-01
-5.16501844e-01 5.46625078e-01 6.48396492e+00 5.49363077e-01
-8.02201748e-01 1.44483238e-01 9.14916039e-01 -8.48904625e-02
-2.88317293e-01 1.40050650e-01 -3.99610043e-01 4.66448873e-01
1.20450771e+00 -4.20239985e-01 -1.48060161e-03 5.12760460e-01
6.58459842e-01 -6.47203386e-01 -1.05887318e+00 8.07267547e-01
6.96241140e-01 -8.52506936e-01 -6.01357520e-02 1.94744855e-01
3.04592997e-01 -8.47756386e-01 -1.14634819e-01 2.01999620e-01
4.42358200e-03 -1.35131145e+00 1.07272816e+00 5.58493972e-01
4.15010184e-01 -4.99738902e-01 1.39233673e+00 1.06704366e+00
3.24437380e-01 -1.32672518e-01 -1.10334763e-02 -1.01977348e+00
2.66573042e-01 6.02556407e-01 -1.36053884e+00 -8.37752502e-03
1.22055635e-01 -2.08834149e-02 -5.74840128e-01 5.85477650e-01
-5.69168508e-01 1.08047342e+00 8.50289688e-03 -3.87337536e-01
1.31051093e-01 1.34336218e-01 7.28628218e-01 1.09126675e+00
-2.01710865e-01 5.97731709e-01 -4.09295887e-01 9.58949327e-01
9.20152664e-02 -1.91568241e-01 -5.63462913e-01 -3.57597589e-01
8.50165427e-01 1.18145919e+00 -4.90977138e-01 -5.45113444e-01
6.05319347e-03 8.46450865e-01 1.32284045e-01 1.49776265e-01
-4.52969670e-01 -2.61906594e-01 1.63268134e-01 4.82567608e-01
-3.69632512e-01 3.46897580e-02 -9.49321628e-01 -8.76904428e-01
-2.78068595e-02 -1.26600003e+00 2.70475239e-01 -9.04102683e-01
-1.43022430e+00 1.60015330e-01 4.83109728e-02 -4.87827986e-01
-3.23885292e-01 -4.47703421e-01 -6.99752569e-01 9.92130458e-01
-8.18712294e-01 -8.03091407e-01 -2.58556396e-01 9.74626616e-02
1.65866211e-01 1.58617482e-01 5.89408994e-01 -3.37864250e-01
-3.65851581e-01 6.77811384e-01 -7.19636083e-01 2.26296216e-01
1.02740204e+00 -8.08224499e-01 -1.40226558e-01 8.65084767e-01
-2.90989488e-01 9.49774623e-01 9.87518549e-01 -9.44372892e-01
-7.71150708e-01 -1.83970965e-02 1.58174634e+00 -8.24786246e-01
7.19256401e-01 -1.18665054e-01 -1.20962095e+00 4.51913953e-01
4.05075431e-01 -4.19522792e-01 9.74856496e-01 -1.35179520e-01
-8.11888874e-01 2.57148862e-01 -1.76906538e+00 4.22675937e-01
5.97269773e-01 -7.23083496e-01 -1.30137634e+00 -2.85939742e-02
8.36797282e-02 4.21632603e-02 -4.06003803e-01 -8.98902118e-02
8.51511598e-01 -1.21432889e+00 5.83618939e-01 -7.88128138e-01
6.02357566e-01 6.69948757e-02 2.33448997e-01 -1.14801919e+00
-3.72878402e-01 -5.92398763e-01 1.08470596e-01 1.14683640e+00
4.10419196e-01 -1.13966930e+00 3.36629510e-01 1.46374810e+00
8.07781518e-02 -3.08816224e-01 -1.05169702e+00 -2.68868506e-01
3.07741016e-01 -3.00376534e-01 1.77424446e-01 1.37917364e+00
6.64413273e-01 3.08337212e-01 -3.90961796e-01 1.55132219e-01
6.05503380e-01 -2.56031722e-01 4.30524290e-01 -1.26841640e+00
1.40416563e-01 -4.71218586e-01 -4.15192544e-01 -3.64026755e-01
2.99152464e-01 -6.15322232e-01 4.96265478e-03 -1.21796119e+00
3.03769767e-01 7.94287100e-02 4.26268578e-01 5.90881824e-01
-5.63201785e-01 -9.34744999e-02 4.81917173e-01 4.26177621e-01
-9.05718058e-02 1.09039135e-01 8.54866743e-01 1.86495513e-01
-8.59900191e-02 -1.66205496e-01 -1.44970214e+00 6.99547052e-01
1.03453827e+00 -6.89351559e-01 -2.07599606e-02 -9.18642581e-02
5.41594148e-01 1.73465207e-01 1.15446579e+00 -4.68969315e-01
6.48549199e-01 -3.54628026e-01 5.57652533e-01 1.19158208e-01
1.61638319e-01 -6.25822008e-01 -7.93049410e-02 6.08198643e-01
-3.97938758e-01 1.16677865e-01 2.34935358e-01 3.34097177e-01
1.57648697e-01 -7.05036819e-01 7.37487137e-01 -7.41595775e-02
1.50020588e-02 -8.79439056e-01 -1.01173139e+00 6.64849058e-02
8.31161737e-01 -6.49435818e-01 -7.48606741e-01 -8.76109183e-01
-3.31530094e-01 -2.22105324e-01 5.44534862e-01 9.60772410e-02
7.69636631e-01 -9.22871232e-01 -9.38807607e-01 -1.79356635e-01
-1.78123400e-01 -9.38040853e-01 1.08290911e-01 1.25997138e+00
-2.85609037e-01 4.47539479e-01 -1.70521483e-01 7.15318173e-02
-1.00625300e+00 4.34265614e-01 5.46420217e-01 2.37261340e-01
-4.87284213e-02 6.35672987e-01 -2.50449896e-01 1.41223952e-01
-2.05282062e-01 4.70191926e-01 -4.61314470e-02 2.99470216e-01
9.32422459e-01 9.14054275e-01 -5.19232303e-02 -8.67587626e-01
-4.30229753e-01 -3.19194853e-01 -2.02641413e-01 -5.35172403e-01
6.82223856e-01 -1.62817955e-01 -2.35184804e-01 5.72797596e-01
5.61872721e-01 2.68181473e-01 -5.60049117e-01 3.05232286e-01
1.40073180e-01 -9.98267770e-01 2.31769279e-01 -1.49650013e+00
-1.80994630e-01 7.15276361e-01 -1.00467160e-01 4.97653842e-01
5.86493552e-01 -3.36442620e-01 2.76353210e-01 2.09206417e-01
2.90491730e-01 -9.72739041e-01 8.71252418e-02 -8.43899995e-02
1.22049534e+00 -1.49690640e+00 2.74666607e-01 -1.24470048e-01
-9.96839345e-01 1.00197256e+00 6.27516389e-01 -3.43329757e-02
2.66265441e-02 -3.15155774e-01 9.54274759e-02 -3.54610771e-01
-7.77514279e-01 4.92068172e-01 2.53074080e-01 5.96113145e-01
4.85680610e-01 1.89445559e-02 -1.07029080e+00 1.15271842e+00
-4.32928711e-01 -8.44345689e-02 8.98225665e-01 7.53047884e-01
-7.18202770e-01 -5.25429785e-01 -1.10490012e+00 5.78363180e-01
-4.76284564e-01 -2.25067168e-01 -1.33679056e+00 8.19065213e-01
5.71547123e-03 1.82827199e+00 -1.59431726e-01 -3.79534543e-01
1.33942172e-01 5.55497825e-01 1.08566366e-01 -3.49074692e-01
-1.12193489e+00 -4.43930596e-01 7.02918530e-01 -3.79273742e-01
-2.02934563e-01 -9.92274225e-01 -7.69382477e-01 -9.00643408e-01
-6.21812999e-01 4.61726040e-01 4.15665478e-01 1.22615659e+00
3.37800145e-01 -4.55633610e-01 3.25061768e-01 -4.94370043e-01
-1.23567879e+00 -1.22135592e+00 -1.84043780e-01 2.10898414e-01
5.12108505e-01 -5.20287931e-01 -1.20886743e+00 -1.42168418e-01] | [8.1876802444458, 10.387429237365723] |
79d256d8-2133-48a6-984f-94b7bbcc2e24 | atypicality-for-heart-rate-variability-using | 1710.07319 | null | http://arxiv.org/abs/1710.07319v1 | http://arxiv.org/pdf/1710.07319v1.pdf | Atypicality for Heart Rate Variability Using a Pattern-Tree Weighting Method | Heart rate variability (HRV) is a vital measure of the autonomic nervous
system functionality and a key indicator of cardiovascular condition. This
paper proposes a novel method, called pattern tree which is an extension of
Willem's context tree to real-valued data, to investigate HRV via an
atypicality framework. In a previous paper atypicality was developed as method
for mining and discovery in "Big Data," which requires a universal approach.
Using the proposed pattern tree as a universal source coder in this framework
led to discovery of arrhythmias and unknown patterns in HRV Holter Monitoring. | ['Anders Høst-Madsen', 'Elyas Sabeti'] | 2017-10-12 | null | null | null | null | ['heart-rate-variability'] | ['medical'] | [ 6.39794394e-02 -4.18906137e-02 -3.10384817e-02 -3.32715571e-01
4.41482008e-01 -4.41798270e-01 -1.01533039e-02 2.64386475e-01
2.00994667e-02 1.03349757e+00 2.15834588e-01 -3.49285126e-01
-5.57739377e-01 -8.76142323e-01 1.07820936e-01 -2.79575288e-01
-4.59843010e-01 1.30581200e-01 -3.43581200e-01 -3.75663579e-01
3.76249850e-01 5.18396914e-01 -1.76423597e+00 1.52949139e-01
6.26600623e-01 9.86316562e-01 -3.93868625e-01 8.83405924e-01
-5.66971637e-02 6.13277018e-01 -9.81494308e-01 3.27571154e-01
1.66254863e-01 -8.88298035e-01 -6.55829966e-01 -2.64952064e-01
-3.03434014e-01 3.21575820e-01 3.86263341e-01 6.94294870e-01
7.30704069e-01 -6.12228848e-02 2.86876798e-01 -1.14817095e+00
-1.02174193e-01 5.37891746e-01 6.75996691e-02 8.16459775e-01
6.24263644e-01 -1.33681297e-01 4.72410470e-01 -2.43021488e-01
3.43571335e-01 6.53928936e-01 1.19314837e+00 2.81776160e-01
-1.40572667e+00 -5.19326270e-01 -5.08473277e-01 3.66209298e-01
-1.56302536e+00 -3.52004431e-02 6.11855745e-01 -4.76041436e-01
1.10733128e+00 8.57583165e-01 1.32622683e+00 8.02472591e-01
7.02214837e-01 -4.99498278e-01 1.37569225e+00 -4.48834360e-01
3.88515532e-01 2.22238809e-01 7.11213708e-01 4.13776010e-01
4.81770068e-01 5.31913996e-01 -3.11843753e-01 -4.77017969e-01
6.10653818e-01 1.92006737e-01 -3.92189056e-01 2.55189538e-01
-1.03313756e+00 5.19476116e-01 -2.05659389e-01 6.44587338e-01
-5.40418208e-01 -3.15697640e-01 6.32719636e-01 9.38844919e-01
1.49588391e-01 4.58494008e-01 -5.93297958e-01 -5.89973390e-01
-7.45112240e-01 1.71959579e-01 1.31731761e+00 7.11344779e-01
3.31876695e-01 2.34507233e-01 -2.58847952e-01 3.21347415e-01
1.76908061e-01 3.21442783e-01 8.43305409e-01 -7.72099733e-01
-2.53351927e-01 9.99373078e-01 -2.22801313e-01 -9.75669444e-01
-1.02288663e+00 -7.26224124e-01 -1.23376417e+00 4.82497215e-02
-7.82530904e-02 -1.16134137e-01 -5.72584987e-01 1.41920233e+00
3.18321109e-01 1.34977028e-01 1.47789270e-01 5.19468367e-01
8.48547399e-01 3.52757156e-01 -1.61688328e-01 -1.00969076e+00
1.52251077e+00 -3.46439555e-02 -1.12632704e+00 6.99506402e-01
5.47420457e-02 -6.37783051e-01 8.38391542e-01 9.14361775e-01
-5.58002710e-01 -8.66203845e-01 -1.11915123e+00 5.14083982e-01
-5.72726786e-01 -5.10625958e-01 3.48976463e-01 1.31287301e+00
-1.01856637e+00 9.83922482e-01 -2.94277191e-01 -8.29103827e-01
1.34739459e-01 3.02216858e-02 5.36928438e-02 6.23777866e-01
-1.28190851e+00 1.05975246e+00 5.82031488e-01 1.12379165e-02
-1.37726977e-01 -5.49143255e-01 -5.62433243e-01 -4.92412522e-02
-1.39573604e-01 -8.97132993e-01 6.44827366e-01 -4.70500171e-01
-1.34263718e+00 7.09807813e-01 -1.03114590e-01 -6.99981570e-01
2.98133314e-01 -1.31097632e-02 -1.19756997e+00 -1.84884861e-01
-2.71627963e-01 -5.18313468e-01 7.71638155e-01 -6.65216565e-01
-9.85071510e-02 -5.28297544e-01 -6.22835875e-01 -3.31835449e-01
2.39070117e-01 -1.71707198e-02 6.25296116e-01 -6.88402295e-01
1.86503664e-01 -5.47936738e-01 -1.16648659e-01 -7.24883437e-01
-2.52482980e-01 -7.27373958e-02 6.90594614e-01 -6.26086533e-01
2.07662916e+00 -1.89126086e+00 -2.04107285e-01 9.32122469e-01
2.78927386e-01 1.40243262e-01 6.56621575e-01 3.99654508e-01
-3.77878219e-01 3.10919493e-01 -4.11513299e-01 4.77917403e-01
-3.68705332e-01 7.95605242e-01 -1.95500284e-01 2.79238373e-01
-1.06689304e-01 5.32052994e-01 -5.78816891e-01 -3.74202520e-01
4.03738141e-01 3.64356011e-01 -1.82665408e-01 1.87285289e-01
3.59736323e-01 8.68311763e-01 5.39313257e-02 7.50238776e-01
4.06763792e-01 3.29158939e-02 1.54741228e-01 -1.06722176e-01
-5.75169265e-01 -1.48373440e-01 -1.28020942e+00 1.07980263e+00
-8.25127438e-02 3.06806028e-01 -5.79021096e-01 -1.11647451e+00
1.61868405e+00 8.67958009e-01 6.11099362e-01 -6.78605497e-01
2.46742353e-01 7.13706464e-02 1.51091397e-01 -8.55174541e-01
9.14551765e-02 -3.40821296e-01 -8.98153782e-02 2.63479948e-01
-8.73170048e-02 2.05033198e-01 2.24122584e-01 -4.70259041e-01
1.21195304e+00 1.44067958e-01 1.39959466e+00 -5.96859217e-01
9.44825947e-01 1.46821411e-02 9.49828267e-01 7.01641083e-01
-4.63374406e-01 4.75424796e-01 6.38421774e-01 -1.08012927e+00
-1.04085457e+00 -8.99984539e-01 -7.55126894e-01 9.72434580e-02
-5.02019346e-01 -7.27150857e-01 -2.89265573e-01 -1.21636242e-01
2.00573597e-02 4.22702432e-01 -6.23126864e-01 -6.33170269e-03
-5.00290513e-01 -9.51455712e-01 7.30970502e-01 5.98189868e-02
5.26121616e-01 -1.44368720e+00 -1.21938050e+00 6.43570125e-01
-1.04204796e-01 -6.57370925e-01 3.43811274e-01 4.55736220e-01
-1.34470189e+00 -1.29802048e+00 -1.29582718e-01 -2.60288030e-01
7.32859522e-02 -5.95624745e-01 1.53631353e+00 1.08870074e-01
-1.07576108e+00 1.31338790e-01 -5.67238688e-01 -7.55019307e-01
-6.42936945e-01 -2.62083530e-01 2.17106178e-01 -1.03707135e-01
5.57593524e-01 -1.31145692e+00 -6.28652215e-01 2.12095171e-01
-4.77432042e-01 -7.11782932e-01 2.73543477e-01 5.41239321e-01
7.66433060e-01 2.39596337e-01 1.16010070e+00 -9.44264829e-01
9.61433828e-01 -6.65409446e-01 -4.99390602e-01 -3.61336060e-02
-1.38520801e+00 -1.20956928e-01 6.80292487e-01 2.00094152e-02
-2.82950312e-01 -2.06415147e-01 -9.70935673e-02 -2.41731226e-01
-3.64898413e-01 3.86204451e-01 1.17915176e-01 1.25472814e-01
9.98866916e-01 4.54230666e-01 2.03331187e-01 -6.53204501e-01
-8.73793438e-02 6.22269273e-01 4.94920462e-01 -1.80026054e-01
2.10026518e-01 -2.61034258e-02 4.02093679e-01 -1.02443922e+00
-2.37603649e-01 -7.26743877e-01 -6.31959200e-01 -2.37550542e-01
8.13421726e-01 -5.19446909e-01 -1.16408038e+00 1.69017594e-02
-9.90599275e-01 3.14703315e-01 -7.34165311e-01 4.29520726e-01
-6.31877899e-01 2.23200962e-01 -6.44778535e-02 -1.18393815e+00
-8.81097317e-01 -1.46947354e-01 2.48410359e-01 1.67586997e-01
-8.92039537e-01 -7.95566380e-01 7.08670795e-01 -1.43836245e-01
6.76827192e-01 1.05716240e+00 9.10036802e-01 -6.45031869e-01
9.10123661e-02 -1.04456268e-01 2.29171738e-01 5.55353880e-01
5.01101673e-01 7.46065974e-02 -9.42553699e-01 1.41561508e-01
7.80669153e-01 4.68907356e-01 2.15349331e-01 2.53456563e-01
1.12149024e+00 -4.10636902e-01 4.44145538e-02 5.83889246e-01
1.80608964e+00 6.96178019e-01 9.65717435e-01 2.63892144e-01
2.75720209e-01 7.02360123e-02 3.40108007e-01 1.00149524e+00
1.33335128e-01 2.83525795e-01 3.82482596e-02 -1.46354418e-02
3.05716008e-01 3.71901274e-01 8.00899640e-02 9.34009969e-01
-8.39565814e-01 5.40253580e-01 -9.19582129e-01 1.57909125e-01
-1.66340685e+00 -1.65895927e+00 -3.82480532e-01 2.29338813e+00
9.11866426e-01 1.94483623e-01 5.69316447e-01 6.98382139e-01
6.53485835e-01 -2.01697618e-01 -2.61605293e-01 -1.29543519e+00
-6.40220121e-02 8.96024704e-01 1.62039757e-01 1.86451957e-01
-7.96042860e-01 4.92386892e-02 7.67100573e+00 -2.50699669e-01
-1.03331447e+00 1.13366529e-01 4.65359688e-02 1.48989409e-01
1.02810383e-01 -9.28754583e-02 -5.40126920e-01 6.74806356e-01
1.50410938e+00 -4.63948995e-01 3.82213950e-01 7.59160519e-01
4.96132284e-01 1.00883253e-01 -9.00510967e-01 1.17495048e+00
-8.22175946e-03 -1.00663662e+00 -2.96156377e-01 4.62653376e-02
2.77796865e-01 -3.05348486e-01 -5.66835999e-01 3.05000186e-01
-5.09581864e-01 -1.04441381e+00 -6.81567714e-02 1.10685635e+00
5.33186197e-01 -7.87163556e-01 1.05552018e+00 6.77342061e-04
-1.35998559e+00 -2.90307701e-01 -2.90381134e-01 -5.39540410e-01
1.16973193e-02 8.53595138e-01 -8.37144017e-01 8.08964670e-01
7.45833874e-01 7.67557502e-01 -4.19612944e-01 1.12891495e+00
2.06833348e-01 6.63302541e-01 -3.18377584e-01 1.89437240e-01
-3.86226684e-01 -1.86671287e-01 8.96556616e-01 1.13968790e+00
4.71318036e-01 1.18959755e-01 1.26163036e-01 9.53368902e-01
6.02147341e-01 2.35223502e-01 -1.08129025e+00 3.00656497e-01
4.66548830e-01 1.14532709e+00 -5.78182161e-01 -4.33450788e-01
-1.26626074e-01 5.28030813e-01 -7.15886056e-01 -1.45395212e-02
-7.81626225e-01 -6.13753855e-01 4.36301053e-01 1.96817338e-01
1.84607990e-02 1.41372785e-01 -9.07477498e-01 -8.75116706e-01
1.36135682e-01 -9.00655925e-01 8.91590238e-01 -3.78232569e-01
-1.09365499e+00 8.71074319e-01 -7.42790923e-02 -1.84708178e+00
-3.09274673e-01 -8.97435918e-02 -9.22945082e-01 1.19869733e+00
-1.15851426e+00 -4.45442915e-01 -5.89415610e-01 1.06676412e+00
1.72153503e-01 -2.92630941e-01 1.60013056e+00 3.10893297e-01
-3.87204200e-01 2.36823782e-01 -2.23127410e-01 -3.09123099e-01
3.33614737e-01 -1.50244248e+00 6.19970113e-02 6.73761845e-01
-1.98635429e-01 6.45816088e-01 9.00365889e-01 -7.45875776e-01
-1.04885983e+00 -9.20656562e-01 1.04328239e+00 -5.67474902e-01
1.38164356e-01 2.82142982e-02 -8.42497706e-01 3.38205725e-01
5.59963435e-02 -1.63444516e-03 1.30055380e+00 2.65770376e-01
4.52371798e-02 -5.58545649e-01 -1.58968437e+00 -2.65868157e-02
8.06155622e-01 -1.69978231e-01 -1.12882817e+00 -3.93152907e-02
4.87023264e-01 2.04628445e-02 -1.67407310e+00 5.34031689e-01
8.91527653e-01 -1.25826716e+00 7.76575148e-01 -2.87940204e-01
-1.86646327e-01 -4.16087210e-01 1.35043985e-03 -9.54985380e-01
-1.39946401e-01 -1.10775471e+00 -5.20797551e-01 9.88715708e-01
1.30006433e-01 -1.07664847e+00 1.96869090e-01 1.77143991e-01
-1.12790719e-01 -5.84143221e-01 -1.08559442e+00 -9.08696532e-01
-6.31735206e-01 -3.92591596e-01 9.31255996e-01 1.13768411e+00
5.07312536e-01 4.51225698e-01 -3.68216157e-01 -1.75602809e-01
8.40793014e-01 2.20910192e-01 3.87702882e-01 -1.89802814e+00
-2.18729362e-01 -2.37641424e-01 -1.11222935e+00 4.61145103e-01
-6.98817670e-01 -7.35825777e-01 -3.93413454e-01 -1.16726029e+00
-3.04553986e-01 -3.52631658e-01 -8.51103485e-01 3.19572389e-01
2.40308732e-01 2.48286113e-01 -1.50139019e-01 1.88044190e-01
4.26467098e-02 2.80148275e-02 7.91683733e-01 3.83267969e-01
-8.35528851e-01 4.92158055e-01 -6.31374359e-01 5.61749578e-01
1.26146996e+00 -5.91900051e-01 -5.00763834e-01 9.72043276e-01
4.01624352e-01 3.33609074e-01 3.11993688e-01 -1.55988038e+00
-2.59421438e-01 4.71667573e-02 6.17241561e-01 -7.42530823e-01
-2.47070208e-01 -1.07215142e+00 8.15169513e-01 1.00394821e+00
1.08098619e-01 6.98502123e-01 1.71239391e-01 4.19164956e-01
-3.42378616e-01 6.38336390e-02 5.83118677e-01 -2.88500756e-01
-5.97316086e-01 -6.54762089e-02 -3.89429361e-01 -1.48033705e-02
1.12311149e+00 -5.75599253e-01 -2.55020350e-01 6.05102405e-02
-1.41058981e+00 -5.78183047e-02 1.19166203e-01 3.34075481e-01
6.69468462e-01 -1.24477768e+00 -6.62521243e-01 5.90161145e-01
9.37059522e-02 -3.81861597e-01 1.36951596e-01 1.06568873e+00
-6.03896081e-01 4.35027808e-01 -9.13102984e-01 -7.19259739e-01
-1.52986526e+00 6.26932859e-01 3.39732885e-01 -1.74450606e-01
-1.11560476e+00 -4.78592403e-02 -8.51312399e-01 -3.22814405e-01
1.05512239e-01 -6.66740656e-01 -8.67997110e-01 1.99245438e-01
6.79569483e-01 7.44559169e-01 1.05595730e-01 -1.71787545e-01
-5.84943056e-01 7.09742248e-01 6.59676671e-01 7.67377913e-02
1.03216398e+00 -3.59131098e-01 -5.46100140e-01 1.16868007e+00
8.14757645e-01 -3.60916257e-01 6.71413392e-02 1.99523538e-01
3.33733857e-01 -1.16585776e-01 -4.17371631e-01 -1.07922852e+00
-4.33040828e-01 6.11973047e-01 1.25480688e+00 8.62687409e-01
1.52583861e+00 -5.30811667e-01 3.13474059e-01 3.92451793e-01
3.01218480e-01 -9.86972213e-01 -6.58129334e-01 3.78433615e-01
1.02484059e+00 -8.47896934e-01 1.70933500e-01 -3.58825885e-02
-4.92005557e-01 1.46862113e+00 1.14931487e-01 4.54811640e-02
1.16382015e+00 1.88773513e-01 2.16623000e-03 -3.87581587e-01
-6.31684184e-01 -2.57441491e-01 -5.34370802e-02 7.43659616e-01
5.89111090e-01 2.76655227e-01 -1.15069103e+00 8.09264481e-01
-5.31393707e-01 7.92209387e-01 5.61825871e-01 1.14268601e+00
-6.84119940e-01 -1.17904842e+00 -4.81020033e-01 6.46022260e-01
-5.93519390e-01 -1.69914722e-01 -7.92914405e-02 6.16214633e-01
6.40580237e-01 9.00619864e-01 8.88750255e-02 -5.53651750e-01
6.83575273e-01 8.06032896e-01 3.70680302e-01 -5.44591069e-01
-1.07998800e+00 -3.52520347e-01 1.46934122e-01 -7.29328990e-01
-5.82794189e-01 -2.58908778e-01 -1.08609831e+00 -3.59745741e-01
3.26051950e-01 3.04857641e-01 6.61498904e-01 7.86811709e-01
4.06032503e-01 8.76839340e-01 1.01521683e+00 8.82333741e-02
-4.53475028e-01 -1.08218002e+00 -9.22186375e-01 2.31684953e-01
5.30419230e-01 -2.55460411e-01 -2.24815682e-01 2.72369593e-01] | [14.13066291809082, 3.147197961807251] |
a94e1d1b-aedf-48c1-8057-6c1adf376db3 | image-smoothing-via-unsupervised-learning | 1811.02804 | null | http://arxiv.org/abs/1811.02804v1 | http://arxiv.org/pdf/1811.02804v1.pdf | Image Smoothing via Unsupervised Learning | Image smoothing represents a fundamental component of many disparate computer
vision and graphics applications. In this paper, we present a unified
unsupervised (label-free) learning framework that facilitates generating
flexible and high-quality smoothing effects by directly learning from data
using deep convolutional neural networks (CNNs). The heart of the design is the
training signal as a novel energy function that includes an edge-preserving
regularizer which helps maintain important yet potentially vulnerable image
structures, and a spatially-adaptive Lp flattening criterion which imposes
different forms of regularization onto different image regions for better
smoothing quality. We implement a diverse set of image smoothing solutions
employing the unified framework targeting various applications such as, image
abstraction, pencil sketching, detail enhancement, texture removal and
content-aware image manipulation, and obtain results comparable with or better
than previous methods. Moreover, our method is extremely fast with a modern GPU
(e.g, 200 fps for 1280x720 images). Our codes and model are released in
https://github.com/fqnchina/ImageSmoothing. | ['Qingnan Fan', 'David Wipf', 'Jiaolong Yang', 'Xin Tong', 'Baoquan Chen'] | 2018-11-07 | null | null | null | null | ['image-smoothing'] | ['computer-vision'] | [ 3.94597858e-01 -1.84511423e-01 -4.30577770e-02 -2.88221687e-01
-3.90183657e-01 -3.60262871e-01 4.37610120e-01 1.85817741e-02
-4.23591614e-01 4.81328428e-01 1.55883655e-01 -3.33944321e-01
2.72358477e-01 -7.79474914e-01 -7.53826678e-01 -6.93618298e-01
1.35790512e-01 -4.49591845e-01 4.30173844e-01 -1.48467675e-01
4.27309334e-01 9.03895557e-01 -1.70087337e+00 3.33158106e-01
1.06768215e+00 1.03202343e+00 2.41552338e-01 8.56679857e-01
-8.55258778e-02 5.73090494e-01 -3.17886561e-01 -2.33717367e-01
3.79467487e-01 -7.75361946e-03 -4.61421847e-01 2.60599524e-01
9.36794221e-01 -4.51296091e-01 -2.62688518e-01 1.15038252e+00
4.67486233e-01 1.93974495e-01 5.01305521e-01 -8.02179754e-01
-8.47638249e-01 -4.54493947e-02 -8.71535420e-01 -1.36442751e-01
-2.06842497e-01 3.30564946e-01 5.49677312e-01 -1.00257719e+00
5.21340430e-01 1.14899266e+00 8.43181849e-01 6.68873250e-01
-1.49337900e+00 -4.68039602e-01 -8.77570957e-02 -1.32267445e-01
-1.15522909e+00 -7.17449188e-01 8.44148397e-01 -2.61426955e-01
4.84870166e-01 4.87973720e-01 3.24819773e-01 6.89305365e-01
4.97837096e-01 6.26280665e-01 9.10251439e-01 -4.91767645e-01
2.18483105e-01 1.08048007e-01 1.20756984e-01 8.43130887e-01
9.40504372e-02 9.28411707e-02 -3.63192499e-01 -2.34180585e-01
1.33340466e+00 7.08161294e-02 -3.43607783e-01 -3.02224219e-01
-8.49384189e-01 6.67657256e-01 4.84089077e-01 -1.64594606e-01
-2.11899191e-01 3.53976309e-01 5.43475211e-01 6.03574421e-03
6.21724308e-01 1.45872027e-01 -4.78385389e-01 3.70023280e-01
-9.36388910e-01 1.79309726e-01 4.23185647e-01 9.88194644e-01
9.11704957e-01 3.27959836e-01 -2.63839960e-01 1.02847779e+00
5.45635931e-02 4.06922489e-01 4.17963080e-02 -1.41933489e+00
2.50651482e-02 3.03816348e-01 2.06899568e-01 -1.14361823e+00
-2.47481883e-01 -3.20648253e-02 -1.25642788e+00 1.01947248e+00
4.04450297e-01 -1.94266096e-01 -1.01415062e+00 1.47660816e+00
3.61005634e-01 1.97837099e-01 -4.08199251e-01 8.34094048e-01
9.46144044e-01 5.98311961e-01 2.30604753e-01 4.71292436e-02
1.46956635e+00 -1.20028615e+00 -6.36653244e-01 -1.24379978e-01
2.98270911e-01 -1.01921201e+00 1.37775123e+00 5.80988526e-01
-1.46829355e+00 -7.93706656e-01 -9.66647625e-01 -8.00507307e-01
-4.33461785e-01 4.38741893e-01 8.26752543e-01 6.92749858e-01
-1.43092525e+00 8.50478649e-01 -6.58739328e-01 -8.67436379e-02
7.16072559e-01 3.02931458e-01 -1.93547428e-01 3.59576121e-02
-7.06392884e-01 4.33051944e-01 -6.76829647e-03 2.64536273e-02
-3.30792159e-01 -9.85503197e-01 -8.64809573e-01 1.70223072e-01
2.63406307e-01 -5.95281184e-01 9.41467404e-01 -1.33797586e+00
-1.98654819e+00 9.77778494e-01 -2.51377642e-01 -2.73159266e-01
4.78685617e-01 -3.95693272e-01 -6.93853647e-02 3.40171605e-02
-2.90393949e-01 8.29682648e-01 1.31596196e+00 -1.32871258e+00
-4.94505256e-01 9.85513534e-03 -2.33020544e-01 3.30867283e-02
-3.04039448e-01 2.13835880e-01 -6.69359326e-01 -1.13950014e+00
-3.08227301e-01 -6.16623580e-01 -3.58275115e-01 7.84828484e-01
-3.56813610e-01 7.98603445e-02 1.02680886e+00 -7.63898790e-01
1.19632220e+00 -2.25561118e+00 -2.49468838e-03 2.25233972e-01
5.25376976e-01 6.98303223e-01 -2.05862805e-01 2.02511892e-01
-1.97638392e-01 6.83614016e-02 -4.87047791e-01 -5.50749242e-01
-2.24654123e-01 -5.07212877e-02 -3.00498933e-01 4.74349529e-01
3.93759668e-01 8.68868887e-01 -6.13096356e-01 -3.07653010e-01
5.13733685e-01 8.64903271e-01 -7.26616561e-01 2.03296155e-01
-1.08235650e-01 5.01587689e-01 -1.93716943e-01 6.50664806e-01
1.23931575e+00 -2.16906294e-01 3.80730480e-02 -4.84169334e-01
-5.81073344e-01 -1.62785813e-01 -1.13410246e+00 1.68510818e+00
-5.03741205e-01 8.87427509e-01 6.46724045e-01 -6.03918433e-01
7.93660581e-01 -5.11524342e-02 1.84441164e-01 -5.20938337e-01
2.29649618e-01 2.42210552e-02 -5.14516115e-01 -1.54827446e-01
6.99995220e-01 3.92923534e-01 3.98520619e-01 1.99285850e-01
-1.55029267e-01 -1.40253425e-01 9.11524147e-02 1.43082574e-01
8.25052321e-01 2.22370084e-02 1.25535354e-01 -7.99548686e-01
6.62802935e-01 -3.89543205e-01 5.27948618e-01 6.41463995e-01
-1.75706729e-01 8.78523469e-01 4.08774704e-01 -7.04153419e-01
-1.38790655e+00 -9.50795352e-01 -2.46730804e-01 1.17877173e+00
1.75352260e-01 -3.95490140e-01 -9.05587971e-01 -2.47294664e-01
-5.49502000e-02 4.55537736e-01 -5.17866850e-01 1.00143321e-01
-7.20241129e-01 -3.96422446e-01 2.29245707e-01 5.53972661e-01
5.95795453e-01 -1.17787755e+00 -4.85469252e-01 5.89462407e-02
4.55573082e-01 -7.66853094e-01 -9.29942846e-01 2.14545019e-02
-8.71396184e-01 -8.44239891e-01 -7.95983553e-01 -1.06186473e+00
9.26935315e-01 4.89026308e-01 1.24074256e+00 5.43252528e-01
-6.19539142e-01 1.78634837e-01 8.12680721e-02 -2.99817860e-01
-1.87464833e-01 -2.88124442e-01 -2.69190401e-01 3.97642814e-02
-1.68510154e-01 -5.21549642e-01 -8.85908306e-01 1.46345884e-01
-1.16577089e+00 3.51178110e-01 3.09520006e-01 8.02460670e-01
6.87398434e-01 -7.83551186e-02 1.90616176e-01 -8.90587389e-01
7.21302450e-01 1.61169760e-03 -9.45134103e-01 9.47889835e-02
-2.80365855e-01 -1.29455090e-01 8.59255314e-01 -5.02437234e-01
-1.37339222e+00 1.37929186e-01 -3.37967694e-01 -3.32311839e-01
-3.78385365e-01 -1.33046672e-01 -6.81052729e-02 -6.38653636e-01
6.79540634e-01 7.28466064e-02 2.73183793e-01 -5.77142000e-01
6.67421937e-01 2.50746816e-01 7.01273203e-01 -6.03035688e-01
6.92441702e-01 8.08877766e-01 5.00850193e-02 -1.22680354e+00
-4.39845681e-01 -1.48043588e-01 -4.51773912e-01 -8.68519619e-02
7.95512319e-01 -6.98369443e-01 -8.97065461e-01 9.41416979e-01
-1.26374602e+00 -8.31139743e-01 -2.04600438e-01 -1.62389472e-01
-4.01112735e-01 5.61483264e-01 -8.86152029e-01 -6.20296001e-01
-7.25123584e-01 -1.13312316e+00 1.08331835e+00 6.81909442e-01
7.27986246e-02 -1.04101408e+00 -1.30699232e-01 -1.08800359e-01
7.63948619e-01 2.87811697e-01 8.38402033e-01 3.15880358e-01
-6.78217769e-01 1.29163399e-01 -7.41068900e-01 5.96647620e-01
2.21918657e-01 4.51709777e-01 -9.43616331e-01 -3.59521240e-01
-1.47106275e-01 -2.76934117e-01 9.98447657e-01 8.20758224e-01
1.79223001e+00 -2.91009396e-01 -1.53049096e-01 1.21735334e+00
1.57334721e+00 1.71317030e-02 1.13102889e+00 4.83870134e-02
8.68607104e-01 4.37642485e-01 6.92548975e-02 2.62533218e-01
2.44881604e-02 5.48166990e-01 1.52255952e-01 -8.46238792e-01
-4.91034627e-01 2.05232978e-01 8.63459706e-02 5.48360109e-01
-1.45963281e-01 -8.18482339e-02 -5.92668355e-01 3.09811831e-01
-1.76396644e+00 -7.14100361e-01 -4.44177568e-01 2.23331213e+00
9.21389759e-01 -6.81839809e-02 -6.41765594e-02 -2.70774901e-01
7.80422270e-01 1.53954417e-01 -5.22970676e-01 -6.76648617e-01
-4.72725295e-02 5.62767982e-01 6.73217058e-01 7.34301090e-01
-1.33795691e+00 1.07388866e+00 6.19729328e+00 1.17915773e+00
-1.31953299e+00 -1.73399016e-01 1.09030652e+00 1.76089808e-01
-1.05354086e-01 -3.11698794e-01 -6.04306221e-01 4.65789557e-01
4.20495063e-01 5.93058839e-02 4.68311638e-01 8.38920951e-01
4.65785295e-01 -2.22132325e-01 -6.02215588e-01 9.24131513e-01
-1.90592587e-01 -1.81061268e+00 -4.61860113e-02 -1.54136181e-01
8.02726507e-01 -1.44228637e-01 2.88520157e-01 -2.65678823e-01
1.92627341e-01 -9.38034952e-01 5.91421485e-01 5.31840801e-01
1.20686865e+00 -7.60365248e-01 2.25015268e-01 -1.16495401e-01
-1.22192609e+00 2.21921965e-01 -5.36570370e-01 1.25786901e-01
1.05078243e-01 8.06320846e-01 2.68743262e-02 2.11799443e-01
8.02466869e-01 6.69222057e-01 -4.46829587e-01 1.01867151e+00
-1.28031626e-01 3.91201288e-01 -1.65657699e-01 3.06047618e-01
4.18405458e-02 -5.50030410e-01 4.17055160e-01 1.42167068e+00
9.98673663e-02 2.65222818e-01 -2.67814402e-03 9.67242360e-01
-2.20410004e-01 1.50942504e-01 -4.30576205e-01 3.62861097e-01
1.52496636e-01 1.55351949e+00 -9.12958920e-01 -4.11683530e-01
-5.64362943e-01 1.07376862e+00 4.66689050e-01 5.85199296e-01
-7.57758141e-01 -7.69802094e-01 9.44224894e-01 8.94767344e-02
2.78601855e-01 -3.58875155e-01 -8.21146488e-01 -1.16535926e+00
-9.62385237e-02 -7.31811404e-01 -8.14392716e-02 -7.48184443e-01
-1.30011213e+00 5.26847959e-01 -4.95540172e-01 -9.78373110e-01
5.39466918e-01 -9.28762913e-01 -1.07234192e+00 1.09526479e+00
-1.78987658e+00 -1.01934385e+00 -5.77940643e-01 5.87282300e-01
6.60066068e-01 2.69064922e-02 6.26586914e-01 3.91774923e-01
-5.23112953e-01 5.23979664e-01 2.78055489e-01 -1.35967359e-02
9.54317808e-01 -1.14463902e+00 7.25188851e-01 9.23993766e-01
-4.61932391e-01 7.78004766e-01 5.12959540e-01 -6.40047073e-01
-1.22908556e+00 -1.19489181e+00 5.09324133e-01 -9.87865031e-02
4.16750968e-01 -4.41862285e-01 -1.36442554e+00 4.14928794e-01
5.45646727e-01 1.87175483e-01 2.79355079e-01 -2.74020344e-01
-2.99557716e-01 -5.33560403e-02 -1.24171209e+00 1.09175336e+00
9.55111623e-01 -1.93130881e-01 3.15150544e-02 2.04795003e-01
5.63423872e-01 -6.11327171e-01 -5.36466181e-01 2.01408491e-01
4.72365737e-01 -1.20177531e+00 1.21830237e+00 -3.86566311e-01
6.23496056e-01 -4.18774396e-01 2.79166937e-01 -1.03198957e+00
-7.17376411e-01 -1.02720153e+00 -5.93243949e-02 1.04428422e+00
4.96757030e-02 -7.35166252e-01 8.92757356e-01 7.93682933e-01
-2.89124817e-01 -7.66981721e-01 -5.63311696e-01 -4.14957851e-01
8.76699984e-02 -4.07141075e-02 3.39675099e-01 8.10842276e-01
-4.45886105e-01 -2.24582970e-01 -5.07357061e-01 1.80870712e-01
1.08248866e+00 -1.98075742e-01 8.16729605e-01 -9.90657151e-01
7.26297051e-02 -7.04329073e-01 4.42477204e-02 -1.21931636e+00
-4.54051327e-03 -5.16379535e-01 4.07094359e-02 -1.17094076e+00
4.30843830e-02 -4.91680264e-01 -3.91295105e-02 4.27273124e-01
-2.92659730e-01 3.85378629e-01 9.36136395e-02 -4.96610180e-02
-2.28100017e-01 4.21257466e-01 1.26816595e+00 6.41416237e-02
-2.84980953e-01 -2.04120651e-01 -5.99722922e-01 1.04648840e+00
1.01651621e+00 -6.89116642e-02 -2.47176439e-01 -7.36172318e-01
-1.95100531e-01 -3.73466730e-01 5.50895631e-01 -8.37541640e-01
1.96829110e-01 -3.48877579e-01 5.48208117e-01 -3.99549991e-01
1.97773561e-01 -6.27825320e-01 3.26778293e-02 3.23381722e-01
-3.49016607e-01 -1.59818426e-01 6.40352786e-01 2.34463274e-01
6.88175038e-02 -1.37156174e-01 1.37210000e+00 -2.87411036e-03
-8.44968319e-01 4.19653058e-01 -2.35094652e-01 -2.22075731e-01
8.04929197e-01 -1.48690835e-01 -4.29045111e-01 -2.27267161e-01
-4.40330833e-01 -9.59069580e-02 8.33390594e-01 1.73062474e-01
7.03788042e-01 -1.42076552e+00 -5.86137891e-01 5.16543329e-01
-2.99964756e-01 1.21689057e-02 4.98606771e-01 4.56844598e-01
-9.50966597e-01 -1.28284013e-02 -4.26512271e-01 -3.68151426e-01
-1.39565074e+00 4.74173248e-01 1.81338623e-01 3.46027128e-02
-9.48383629e-01 9.73609388e-01 5.91934681e-01 -1.05131835e-01
2.70244122e-01 -3.93833548e-01 8.63288492e-02 -5.44628441e-01
9.62945223e-01 5.05524158e-01 -2.23537702e-02 -2.65625387e-01
-7.78422058e-02 6.85008407e-01 -1.68873996e-01 3.21597576e-01
1.29966223e+00 -9.63547826e-02 -2.42507398e-01 -2.58941799e-01
1.20406723e+00 3.20231944e-01 -1.93899608e+00 -1.31506145e-01
-2.85872668e-01 -7.36550391e-01 3.88244003e-01 -4.65806872e-01
-1.34549809e+00 7.74711967e-01 5.27063906e-01 2.33716115e-01
1.28811407e+00 -4.52310681e-01 9.25084710e-01 -9.20722112e-02
3.81493866e-02 -1.23559654e+00 9.43835154e-02 4.13371593e-01
1.11863315e+00 -1.21683776e+00 1.70550033e-01 -6.99131846e-01
-5.10010302e-01 1.22089219e+00 5.11324883e-01 -5.44606805e-01
6.94948018e-01 5.83654284e-01 1.15762450e-01 -1.32951038e-02
-4.06745464e-01 -7.12982714e-02 5.58597088e-01 6.98428214e-01
5.36879361e-01 -9.06965509e-02 -3.81585747e-01 1.90132052e-01
3.88477206e-01 1.26464084e-01 4.90113497e-01 9.76108909e-01
-4.73651975e-01 -1.09654856e+00 -3.36301565e-01 3.97321463e-01
-4.78910208e-01 -3.12024027e-01 1.40772045e-01 4.21730161e-01
1.18285157e-01 5.77226400e-01 1.69862926e-01 8.60063806e-02
2.53178835e-01 -2.96585977e-01 2.40186512e-01 -4.58758771e-01
-5.73448241e-01 2.57426739e-01 -1.42044097e-01 -8.17554712e-01
-2.71833748e-01 -3.24422896e-01 -9.32578564e-01 -6.00956440e-01
-1.20446999e-02 -5.02030730e-01 6.15508735e-01 3.25668395e-01
7.30920017e-01 4.19052601e-01 5.93251288e-01 -1.54737902e+00
6.33260095e-03 -3.51486325e-01 -5.74976802e-01 5.49414515e-01
4.28521633e-01 -4.38539475e-01 -2.92997271e-01 5.62288821e-01] | [10.952305793762207, -1.496046543121338] |
bee84a1b-2ea0-4811-8b24-db06ae1e5ca6 | few-shot-single-view-3-d-object | 2004.06302 | null | https://arxiv.org/abs/2004.06302v2 | https://arxiv.org/pdf/2004.06302v2.pdf | Few-Shot Single-View 3-D Object Reconstruction with Compositional Priors | The impressive performance of deep convolutional neural networks in single-view 3D reconstruction suggests that these models perform non-trivial reasoning about the 3D structure of the output space. However, recent work has challenged this belief, showing that complex encoder-decoder architectures perform similarly to nearest-neighbor baselines or simple linear decoder models that exploit large amounts of per category data in standard benchmarks. On the other hand settings where 3D shape must be inferred for new categories with few examples are more natural and require models that generalize about shapes. In this work we demonstrate experimentally that naive baselines do not apply when the goal is to learn to reconstruct novel objects using very few examples, and that in a \emph{few-shot} learning setting, the network must learn concepts that can be applied to new categories, avoiding rote memorization. To address deficiencies in existing approaches to this problem, we propose three approaches that efficiently integrate a class prior into a 3D reconstruction model, allowing to account for intra-class variability and imposing an implicit compositional structure that the model should learn. Experiments on the popular ShapeNet database demonstrate that our method significantly outperform existing baselines on this task in the few-shot setting. | ['Stavros Tsogkas', 'Mateusz Michalkiewicz', 'Eugene Belilovsky', 'Anders Eriksson', 'Mahsa Baktashmotlagh', 'Sarah Parisot'] | 2020-04-14 | null | https://www.ecva.net/papers/eccv_2020/papers_ECCV/html/5140_ECCV_2020_paper.php | https://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123700613.pdf | eccv-2020-8 | ['single-view-3d-reconstruction'] | ['computer-vision'] | [ 3.31414551e-01 4.88935471e-01 -5.12710959e-02 -9.04945731e-01
-6.72309756e-01 -7.72960424e-01 1.00746202e+00 1.51095530e-02
-3.19395065e-01 2.99971640e-01 5.49085021e-01 -2.02668697e-01
1.81023657e-01 -8.57046247e-01 -1.23157763e+00 -3.63612145e-01
2.11748749e-01 8.17248940e-01 3.11302364e-01 -1.92744672e-01
3.41134071e-01 5.63541591e-01 -1.64899123e+00 6.24533653e-01
8.28990638e-02 7.67509043e-01 1.67175427e-01 5.29792547e-01
-1.45662367e-01 5.99125862e-01 -2.89366215e-01 -5.63947976e-01
4.70735282e-01 -3.27913225e-01 -7.53877401e-01 2.86180109e-01
1.05840337e+00 -7.33965278e-01 -4.09381360e-01 8.51199567e-01
4.93357271e-01 3.49279851e-01 1.07166457e+00 -7.13866949e-01
-1.18804443e+00 4.99626100e-01 -2.13267267e-01 1.23343311e-01
2.83251941e-01 2.72716254e-01 1.18833852e+00 -1.26931071e+00
1.02735198e+00 1.41319609e+00 9.24523473e-01 7.45827258e-01
-1.61983442e+00 -3.73346657e-01 5.18028796e-01 1.69973165e-01
-1.28973055e+00 -6.76732838e-01 6.86515629e-01 -4.82286423e-01
1.44325304e+00 -2.46305168e-02 6.78460181e-01 1.30082119e+00
-5.77805378e-02 8.05516958e-01 7.82335401e-01 -2.14278713e-01
2.86568731e-01 -2.14504209e-04 1.95985194e-02 6.59912884e-01
3.53088528e-01 1.57872424e-01 -5.39182603e-01 1.36011839e-01
6.91527605e-01 3.68736871e-02 -6.00565113e-02 -8.64479363e-01
-1.13401461e+00 7.44300842e-01 5.65101445e-01 1.71788976e-01
-3.88481691e-02 3.34964961e-01 3.45651358e-01 2.81088352e-01
6.36430502e-01 5.30117333e-01 -4.94209290e-01 2.15461090e-01
-1.03098905e+00 4.71270204e-01 8.60518277e-01 1.26542842e+00
6.69892609e-01 1.52103111e-01 -9.45603848e-02 6.45722806e-01
2.19702467e-01 2.71182865e-01 3.39586675e-01 -1.14089179e+00
3.87941986e-01 1.99797735e-01 -1.34892285e-01 -7.11713612e-01
-2.73274362e-01 -3.72281075e-01 -5.04314542e-01 3.17832798e-01
4.10975784e-01 3.50150973e-01 -1.29573774e+00 1.91577888e+00
1.07134677e-01 1.72311753e-01 1.04834951e-01 7.82084286e-01
9.81826007e-01 3.72105867e-01 -1.05082035e-01 2.19753131e-01
1.02594995e+00 -6.88074112e-01 -1.77870259e-01 -4.18253809e-01
3.16014081e-01 -5.98986804e-01 1.18662214e+00 2.63434112e-01
-1.26786721e+00 -6.59244239e-01 -1.10585284e+00 -5.39669812e-01
-4.25805420e-01 -3.21283668e-01 4.36292619e-01 4.71599847e-01
-1.13313890e+00 9.57706988e-01 -7.63771415e-01 -6.72869742e-01
7.75575280e-01 1.16842583e-01 -3.66936624e-01 -3.76172334e-01
-5.59595227e-01 1.10201049e+00 3.30406010e-01 -2.13802174e-01
-1.18604326e+00 -8.52521360e-01 -1.05254233e+00 1.93840340e-02
3.19282800e-01 -1.08728540e+00 1.43595445e+00 -8.74941289e-01
-1.30912840e+00 1.13574135e+00 -9.51173678e-02 -3.91585976e-01
3.30916017e-01 -7.70612732e-02 1.93088010e-01 -1.33242840e-02
3.81307527e-02 1.10493577e+00 8.55295360e-01 -1.61413836e+00
-2.42284238e-01 -3.57906163e-01 2.45749295e-01 3.83162975e-01
2.55625509e-02 -5.93104780e-01 -2.78922945e-01 -6.91124618e-01
4.48163301e-01 -8.93730879e-01 -6.91690072e-02 4.29256052e-01
-2.77572453e-01 -2.53089517e-01 4.97589022e-01 -4.44046915e-01
4.41320032e-01 -2.06184006e+00 3.74848962e-01 -3.34333489e-03
4.11395356e-02 -5.76645620e-02 -3.28648448e-01 2.96099216e-01
-1.87893435e-02 1.08385220e-01 -4.24698889e-01 -5.85133135e-01
3.18576396e-01 5.22555590e-01 -6.35000765e-01 3.65175128e-01
2.65673995e-01 1.03028822e+00 -7.78789639e-01 -3.60454917e-02
1.57142133e-01 4.65380102e-01 -9.74883914e-01 1.80184528e-01
-4.96232599e-01 2.26587474e-01 1.98841795e-01 4.45835948e-01
5.28131902e-01 -3.99198890e-01 1.23828135e-01 -2.52113402e-01
1.82441995e-01 5.36655307e-01 -1.01407766e+00 2.30861616e+00
-3.55723441e-01 5.53240478e-01 -2.41946369e-01 -1.23882329e+00
8.20974767e-01 4.53582034e-02 7.66147673e-02 -5.01693130e-01
-4.31192964e-02 8.65165740e-02 -6.99150115e-02 -3.43672752e-01
3.89390528e-01 -7.11491227e-01 2.07559764e-02 6.75578654e-01
4.40464914e-01 -7.32378900e-01 -1.55048177e-01 1.63688153e-01
9.26360667e-01 6.04698181e-01 1.72204107e-01 -1.04264639e-01
-3.88467796e-02 7.90959597e-03 4.05805916e-01 8.81103516e-01
7.85069466e-02 1.06378424e+00 1.36902273e-01 -7.54590094e-01
-1.44675970e+00 -1.42495239e+00 -7.12201893e-02 9.95051384e-01
1.61796317e-01 -2.32910261e-01 -3.61556172e-01 -6.96594059e-01
1.33339062e-01 1.23205078e+00 -6.56013787e-01 -3.00598890e-01
-5.01261652e-01 -4.10603404e-01 3.30525964e-01 7.20550001e-01
1.65492147e-01 -8.61323178e-01 -6.74894989e-01 2.47246504e-01
3.51519994e-02 -1.06318200e+00 -2.82708347e-01 2.55263776e-01
-1.11825538e+00 -9.26599443e-01 -6.61475420e-01 -9.62886155e-01
7.91741431e-01 3.12198490e-01 1.45626450e+00 -1.11083537e-01
-2.80912846e-01 7.09043860e-01 -1.74099654e-01 -3.72155517e-01
-2.96369821e-01 6.24972023e-02 -2.24174306e-01 -2.75775641e-01
6.45080209e-01 -1.02568102e+00 -5.38856506e-01 -5.29070683e-02
-8.25956643e-01 3.32915246e-01 6.68090940e-01 8.48617971e-01
7.65676141e-01 -5.19681275e-01 5.92987716e-01 -1.06058538e+00
2.47217581e-01 -5.33684433e-01 -2.39857256e-01 1.61356673e-01
-4.38355178e-01 3.96941066e-01 6.02388918e-01 -4.22878534e-01
-1.19775701e+00 2.01395676e-01 -3.01197737e-01 -5.72386622e-01
-4.52054918e-01 2.18142211e-01 -6.90481514e-02 2.48079330e-01
8.23905885e-01 4.38747227e-01 3.07634976e-02 -7.49452055e-01
7.28640258e-01 1.61506698e-01 4.96612281e-01 -6.85839713e-01
8.13743651e-01 7.23998308e-01 -3.15638036e-02 -6.59624517e-01
-1.24073768e+00 -1.23843625e-01 -9.70988572e-01 1.88039362e-01
8.65081131e-01 -1.15787137e+00 -2.17338532e-01 3.66029218e-02
-1.29017127e+00 -4.55519944e-01 -7.92880237e-01 2.99539387e-01
-1.10334468e+00 2.84038752e-01 -4.82079506e-01 -3.53055447e-01
-4.53079306e-02 -8.86204004e-01 1.22720993e+00 -2.08653077e-01
-3.82968009e-01 -9.39622760e-01 -7.58384168e-02 6.78110346e-02
4.75756377e-01 1.18702061e-01 1.29739869e+00 -7.37895727e-01
-8.28555644e-01 1.12691469e-01 -1.90577582e-01 3.02719057e-01
-6.15321957e-02 -5.59626400e-01 -1.08471632e+00 -2.76253134e-01
9.79686826e-02 -6.35581195e-01 1.13253760e+00 1.99016705e-01
1.34665787e+00 -2.87396967e-01 -1.64893791e-01 7.70594358e-01
1.40426946e+00 -1.22393198e-01 4.95670319e-01 -1.73703149e-01
7.12130904e-01 5.05627215e-01 5.79466671e-02 2.70974427e-01
7.40478754e-01 6.38456225e-01 4.77188230e-01 2.80458063e-01
-6.55837774e-01 -5.72584391e-01 1.84339240e-01 8.71380389e-01
-5.35014533e-02 -1.36213722e-02 -7.06147492e-01 7.25482404e-01
-1.62560964e+00 -1.16674829e+00 4.05250758e-01 2.08933711e+00
9.95004296e-01 2.31756121e-01 -2.07470581e-01 -3.35055679e-01
3.56244087e-01 2.38257661e-01 -8.23483765e-01 -3.58755976e-01
-1.00063987e-01 3.84351224e-01 2.79534012e-01 6.10662997e-01
-9.51079786e-01 1.04548526e+00 7.03109217e+00 4.30009216e-01
-7.79101491e-01 1.13229528e-01 4.53238666e-01 -3.36128205e-01
-8.06216955e-01 1.45886704e-01 -8.22964966e-01 -1.52261198e-01
5.93368590e-01 2.27864329e-02 5.20892203e-01 8.03275883e-01
-3.21067691e-01 1.73398539e-01 -1.85109305e+00 9.92524505e-01
6.46285295e-01 -1.38330472e+00 3.58755678e-01 -9.99505371e-02
7.73201168e-01 1.81419030e-01 2.93711871e-01 3.85025084e-01
5.36263824e-01 -1.05950189e+00 9.38095808e-01 8.38491976e-01
8.96075428e-01 -3.90416980e-01 1.33424059e-01 5.26265800e-01
-9.24082398e-01 9.02312547e-02 -7.71372855e-01 -1.25591457e-01
1.49740160e-01 1.87605545e-01 -1.02864683e+00 1.37891605e-01
5.38818657e-01 1.05639243e+00 -6.16280556e-01 8.63535166e-01
-3.47927600e-01 3.16618532e-01 -2.68873215e-01 1.53922573e-01
1.85830370e-01 1.74790323e-01 5.46156585e-01 1.07487643e+00
4.98590857e-01 2.07555875e-01 1.59991514e-02 1.09963155e+00
-2.35101134e-01 -2.24846944e-01 -1.09320199e+00 3.14749897e-01
2.44011909e-01 6.49037182e-01 -6.93426490e-01 -4.26385731e-01
-6.40804291e-01 9.97304142e-01 6.51487589e-01 3.42234492e-01
-4.02890295e-01 -9.36216116e-02 5.64289510e-01 2.61156231e-01
8.71394336e-01 -4.88921791e-01 -5.67550123e-01 -1.36843574e+00
-4.10310365e-02 -8.02587569e-01 3.05766106e-01 -9.96906161e-01
-1.53234375e+00 2.81139582e-01 1.20048642e-01 -1.08698189e+00
-4.54651475e-01 -6.77023053e-01 -2.72639751e-01 5.67187190e-01
-1.52756572e+00 -1.31045747e+00 8.48637021e-04 5.62051117e-01
1.00871515e+00 -2.04935893e-01 1.00810170e+00 1.85650475e-02
2.74227709e-01 5.06776214e-01 -1.19211994e-01 -9.70401317e-02
6.76510870e-01 -1.26274002e+00 7.62693167e-01 7.83743143e-01
7.16558695e-01 6.15412891e-01 6.77583158e-01 -6.76301420e-01
-1.48296690e+00 -9.27523792e-01 9.63342905e-01 -9.25972998e-01
2.36221120e-01 -6.99135721e-01 -1.04457629e+00 1.07254028e+00
1.63305983e-01 2.34625131e-01 7.61351764e-01 2.47592613e-01
-1.05676591e+00 1.61267146e-01 -9.85032678e-01 5.43459058e-01
1.69684172e+00 -7.57892072e-01 -1.14585519e+00 3.42279047e-01
8.22290361e-01 -3.53068441e-01 -7.04463840e-01 3.09074074e-01
7.34821677e-01 -9.36422169e-01 1.24110234e+00 -9.59930599e-01
5.45219362e-01 -1.98528901e-01 -7.64798284e-01 -1.42548549e+00
-3.83911103e-01 -1.72346890e-01 -2.85780132e-01 7.20738709e-01
3.23245466e-01 -9.81318504e-02 7.29439557e-01 6.87006533e-01
-4.51081544e-01 -6.24297321e-01 -8.74825835e-01 -7.14767575e-01
5.76225162e-01 -5.33707619e-01 6.16737723e-01 7.34271348e-01
-4.12868917e-01 5.77372313e-01 -3.32201421e-01 -1.11577243e-01
6.86247408e-01 2.81020164e-01 7.99724221e-01 -1.26356936e+00
-5.51082313e-01 -4.49388474e-01 -4.56111848e-01 -1.36288297e+00
3.71239007e-01 -1.32612741e+00 1.17871001e-01 -1.79308319e+00
3.13566178e-01 -3.26451987e-01 2.49148868e-02 5.45609117e-01
1.61236003e-01 3.32528204e-01 3.96074623e-01 2.24936768e-01
-6.74324811e-01 6.45503938e-01 1.27073050e+00 -3.30816954e-01
9.07682180e-02 -8.40635672e-02 -8.23597491e-01 8.40970576e-01
3.50564122e-01 -4.14476216e-01 -6.72613382e-01 -1.04453170e+00
3.17619234e-01 -2.98658490e-01 5.65289676e-01 -7.89354384e-01
3.24853390e-01 4.43160348e-03 7.49727130e-01 -7.55730391e-01
6.93188131e-01 -8.83682609e-01 1.78790167e-02 2.79264808e-01
-6.32673264e-01 -2.34491289e-01 1.94668233e-01 8.10736299e-01
5.13440296e-02 -2.78831869e-01 7.47510374e-01 -5.10909259e-01
-9.60301816e-01 4.15713251e-01 -1.06024146e-01 2.49475449e-01
7.49424338e-01 -4.79521245e-01 -2.16803476e-01 -4.95211959e-01
-1.22059500e+00 -2.53958881e-01 6.56622708e-01 6.06010199e-01
8.40732694e-01 -1.51049733e+00 -7.33181238e-01 2.99840838e-01
3.41540515e-01 1.58020958e-01 9.23573002e-02 2.86671370e-01
-2.54949272e-01 9.94157493e-02 -3.10452759e-01 -7.80293524e-01
-7.77424634e-01 7.44015038e-01 4.04155731e-01 1.98029175e-01
-9.29053128e-01 1.02546489e+00 4.82441336e-01 -8.00662756e-01
3.47222000e-01 -2.72096455e-01 6.53914437e-02 7.07072839e-02
3.35880727e-01 -1.26244828e-01 9.52875391e-02 -6.65481031e-01
-1.43293142e-01 7.42601275e-01 -1.70770660e-01 -2.13369191e-01
1.79194808e+00 -2.03883305e-01 2.64857262e-01 8.55971515e-01
1.30608749e+00 -3.72500718e-01 -1.67324293e+00 -6.08186185e-01
-5.60396202e-02 -5.57670951e-01 -2.68095225e-01 -8.08425605e-01
-7.15914130e-01 1.29488194e+00 4.18250650e-01 -2.58743852e-01
7.59030163e-01 2.95539558e-01 5.48712075e-01 8.37391496e-01
3.26733470e-01 -1.02152967e+00 3.31198484e-01 8.66464496e-01
1.01688743e+00 -1.21207225e+00 1.03957064e-01 -9.03244093e-02
-6.07945859e-01 1.19749963e+00 6.22833014e-01 -3.71760070e-01
9.02796149e-01 -3.49632837e-02 -3.00747275e-01 -3.21472555e-01
-8.91062677e-01 -2.32498214e-01 4.90408957e-01 7.96485662e-01
3.38938087e-01 -7.32180402e-02 2.12235138e-01 5.64153135e-01
-3.68063956e-01 -2.64303505e-01 5.21686256e-01 8.35849464e-01
-5.50402820e-01 -8.86136770e-01 6.03571683e-02 5.37979782e-01
-3.37415859e-02 -1.57084450e-01 -4.11976308e-01 7.32622147e-01
1.99807927e-01 3.87031108e-01 3.69047731e-01 -1.57529354e-01
4.32237893e-01 5.50013959e-01 8.87996256e-01 -1.18027568e+00
-2.63340652e-01 3.55770974e-03 1.19911516e-02 -5.26830018e-01
-4.82147008e-01 -8.69951487e-01 -1.06425405e+00 -1.93558350e-01
7.07978308e-02 -4.96258110e-01 5.49534142e-01 1.06846619e+00
4.33057457e-01 3.36061209e-01 1.99155107e-01 -1.07400095e+00
-7.64914155e-01 -7.54518747e-01 -3.52897018e-01 7.34723091e-01
2.64203191e-01 -7.35480607e-01 -2.39396632e-01 2.91961789e-01] | [8.467633247375488, -3.1532816886901855] |
5ebbfda1-b4ed-4674-95d1-c763f7900353 | semi-supervised-deep-representation-learning | 1811.04480 | null | http://arxiv.org/abs/1811.04480v1 | http://arxiv.org/pdf/1811.04480v1.pdf | Semi-supervised Deep Representation Learning for Multi-View Problems | While neural networks for learning representation of multi-view data have
been previously proposed as one of the state-of-the-art multi-view dimension
reduction techniques, how to make the representation discriminative with only a
small amount of labeled data is not well-studied. We introduce a
semi-supervised neural network model, named Multi-view Discriminative Neural
Network (MDNN), for multi-view problems. MDNN finds nonlinear view-specific
mappings by projecting samples to a common feature space using multiple coupled
deep networks. It is capable of leveraging both labeled and unlabeled data to
project multi-view data so that samples from different classes are separated
and those from the same class are clustered together. It also uses the
inter-view correlation between views to exploit the available information in
both the labeled and unlabeled data. Extensive experiments conducted on four
datasets demonstrate the effectiveness of the proposed algorithm for multi-view
semi-supervised learning. | ['Lei Zheng', 'Philip S. Yu', 'Sihong Xie', 'Weixiang Shao', 'Vahid Noroozi', 'Sara Bahaadini'] | 2018-11-11 | null | null | null | null | ['learning-representation-of-multi-view-data'] | ['methodology'] | [-1.20076083e-01 -1.02502465e-01 -5.41624844e-01 -7.09979594e-01
-5.74156702e-01 -6.01059675e-01 4.66827631e-01 -5.07371187e-01
5.87004721e-02 3.72498870e-01 4.69712377e-01 3.62397343e-01
-1.89336464e-01 -5.14521182e-01 -3.66320044e-01 -7.83567667e-01
3.52505505e-01 7.50527859e-01 -3.87917995e-01 2.13064745e-01
-1.20355546e-01 5.88482201e-01 -1.43194938e+00 6.16404414e-01
3.56180429e-01 7.76671648e-01 7.34914392e-02 3.56544614e-01
1.62847504e-01 6.19655311e-01 -4.60740775e-02 6.17580265e-02
4.88323659e-01 -5.06796896e-01 -6.10967517e-01 6.07484341e-01
7.38153517e-01 -2.81148970e-01 -4.61472631e-01 1.26053858e+00
4.66771632e-01 1.49318382e-01 7.33326197e-01 -1.24541581e+00
-1.15504348e+00 3.35568160e-01 -8.34365427e-01 8.36467668e-02
-1.67483151e-01 -5.55326819e-01 9.49037969e-01 -1.37445295e+00
7.99262941e-01 1.35616624e+00 3.49677384e-01 6.55064344e-01
-1.35925281e+00 -5.48303664e-01 3.11047912e-01 1.82344317e-01
-1.23809373e+00 -3.15854281e-01 1.20445859e+00 -4.10632163e-01
7.15030313e-01 -5.04747070e-02 6.65850282e-01 1.00617254e+00
2.78317742e-02 1.02728009e+00 1.23724329e+00 -2.66176909e-01
-1.12948008e-01 5.18043697e-01 1.03333429e-01 7.25629151e-01
8.77269059e-02 -1.80795547e-02 -5.33392012e-01 -1.49803117e-01
6.46077335e-01 7.51577020e-01 -2.76669115e-01 -1.28909266e+00
-1.05977654e+00 1.19816399e+00 2.79091954e-01 3.50462884e-01
-2.63885260e-01 -5.49369991e-01 6.57566845e-01 3.13693106e-01
7.10807681e-01 2.66515091e-02 -5.33358753e-01 4.43315893e-01
-7.85259902e-01 -9.93444100e-02 6.38872027e-01 1.03705370e+00
9.59661722e-01 2.73452312e-01 3.05147976e-01 1.03153217e+00
4.19775963e-01 5.23414135e-01 6.36428654e-01 -9.54543412e-01
6.92819297e-01 1.01031578e+00 -3.47211570e-01 -9.67966497e-01
-4.19596851e-01 -3.98537219e-01 -1.36480200e+00 1.58828110e-01
-8.60464424e-02 -2.59653538e-01 -7.83260703e-01 1.89550698e+00
3.64443719e-01 -2.57843938e-02 4.06930447e-01 7.53639340e-01
9.07535434e-01 6.67910874e-01 -4.29533005e-01 -4.74854082e-01
1.02982914e+00 -1.31047201e+00 -5.35710871e-01 -1.54050916e-01
5.75447798e-01 -4.92322922e-01 4.46807861e-01 3.18319261e-01
-9.50148642e-01 -7.58595407e-01 -1.19262040e+00 -1.78763680e-02
-4.32008624e-01 3.30368549e-01 4.85048234e-01 2.83027142e-01
-8.00070703e-01 4.03102160e-01 -7.76708305e-01 -2.82955348e-01
5.24598718e-01 3.87758821e-01 -9.39511001e-01 -5.99511921e-01
-8.23534489e-01 5.26428461e-01 4.97493923e-01 4.60590161e-02
-1.06759930e+00 -4.62958783e-01 -1.01921749e+00 2.67654634e-03
2.46391997e-01 -6.61591589e-01 5.67254364e-01 -1.10714674e+00
-1.07263064e+00 9.58929598e-01 -2.93434918e-01 1.72203541e-01
5.46747819e-03 -3.50424528e-01 -7.81724751e-01 3.31217736e-01
3.03578049e-01 5.91043174e-01 9.77587879e-01 -1.63517427e+00
-5.52262008e-01 -9.41002905e-01 -2.17048019e-01 5.78201890e-01
-6.01076722e-01 -2.18479440e-01 -3.36035937e-01 -4.85961169e-01
5.05042017e-01 -1.17223775e+00 -7.52390996e-02 -1.37532502e-01
-4.81055915e-01 -1.98803797e-01 1.59366405e+00 -5.11030197e-01
8.47054780e-01 -2.05955243e+00 8.07139277e-01 6.54396191e-02
4.46895957e-01 2.97081083e-01 -2.72901505e-01 5.32825589e-01
-5.92464089e-01 -3.16345841e-01 2.02672966e-02 -4.23132807e-01
-3.89093965e-01 2.90067434e-01 -1.76663250e-01 7.97642231e-01
-1.93592027e-01 6.71914279e-01 -7.98097491e-01 -2.86424667e-01
2.90547788e-01 4.78323102e-01 -3.83597702e-01 4.58703935e-01
3.67490321e-01 4.47905600e-01 -3.25072885e-01 6.60555005e-01
8.16882670e-01 -6.48517013e-01 4.00859982e-01 -6.08616292e-01
2.84388036e-01 -3.88033986e-01 -1.37680161e+00 1.97519827e+00
-3.35249633e-01 4.41105276e-01 6.99255317e-02 -1.40922964e+00
7.14347661e-01 4.75965708e-01 7.21110344e-01 -3.07506114e-01
-1.07510120e-01 -1.08223557e-02 -2.91756868e-01 -5.45197785e-01
2.64437199e-01 -5.00507116e-01 -2.53482554e-02 9.12734568e-01
6.14098787e-01 3.71709049e-01 9.71345752e-02 1.83067232e-01
3.45609903e-01 1.43824458e-01 6.58614874e-01 3.41686197e-02
7.19927132e-01 -3.84156853e-01 8.60836983e-01 3.30459565e-01
-1.90691307e-01 8.06083083e-01 8.69511366e-02 -8.01418841e-01
-1.13872123e+00 -1.00020063e+00 6.33713081e-02 1.02171779e+00
3.24257798e-02 -1.48430049e-01 -4.61154014e-01 -1.23891139e+00
-1.40603572e-01 3.81357074e-01 -7.86937177e-01 -2.43391052e-01
-4.60564911e-01 -6.80181861e-01 1.95121281e-02 6.73618853e-01
4.25939292e-01 -8.13137233e-01 -8.79563540e-02 -3.74600440e-02
-1.38368249e-01 -1.08001518e+00 -5.30848622e-01 4.67043310e-01
-1.10793841e+00 -1.15740490e+00 -8.11184466e-01 -1.04698491e+00
1.08234441e+00 8.90390158e-01 1.10270309e+00 -4.53640819e-01
1.06114030e-01 6.32820845e-01 -1.37022659e-01 -4.30536158e-02
-3.68581414e-01 -5.98091222e-02 5.15266836e-01 3.21868598e-01
7.42466033e-01 -8.81643772e-01 -3.30216497e-01 4.24220651e-01
-9.48774695e-01 9.04626176e-02 5.47574401e-01 1.13701296e+00
9.58332956e-01 -1.36404261e-01 5.85148394e-01 -1.22159624e+00
3.20893168e-01 -5.91258883e-01 -3.98353279e-01 4.90028679e-01
-6.75313473e-01 3.63647789e-02 1.08634102e+00 -4.46144938e-01
-9.40397203e-01 3.04575831e-01 4.89590853e-01 -1.11643362e+00
-5.41262090e-01 4.54140127e-01 -4.72942740e-01 6.35453546e-03
3.44642669e-01 5.41681230e-01 1.57156751e-01 -4.77051020e-01
6.40303671e-01 6.17216587e-01 1.56198278e-01 -1.37118921e-01
6.70111716e-01 9.16044474e-01 9.15176570e-02 -5.22371292e-01
-1.29628611e+00 -7.23131478e-01 -1.24639535e+00 -5.23031398e-04
1.04285133e+00 -1.38467062e+00 -1.95054173e-01 3.30687135e-01
-7.88686395e-01 5.02695262e-01 -3.37310761e-01 7.45716512e-01
-5.82971454e-01 5.84561050e-01 -3.98832500e-01 -3.17178607e-01
-3.17924231e-01 -1.06017888e+00 8.89732420e-01 2.13202149e-01
1.76736161e-01 -1.32720804e+00 4.42611903e-01 4.82518017e-01
-1.87186971e-01 -4.44024317e-02 9.64013338e-01 -1.18179739e+00
-2.89062142e-01 -3.70943815e-01 -1.55244783e-01 8.04661095e-01
4.90524054e-01 -4.12064105e-01 -1.03711092e+00 -7.77619720e-01
5.54259360e-01 -7.30676532e-01 8.45368028e-01 4.98703331e-01
1.06813776e+00 -3.25275213e-02 -4.40739840e-01 9.18779552e-01
1.71570349e+00 1.08474545e-01 6.91326186e-02 -3.60748321e-02
1.41091800e+00 6.30897224e-01 3.10678035e-01 3.23852301e-01
4.51255292e-01 3.62250507e-01 4.94988382e-01 -1.51931211e-01
1.70476854e-01 -1.34366602e-01 2.16405421e-01 1.48324418e+00
-1.12338634e-02 -1.72331527e-01 -5.02605319e-01 6.08157158e-01
-1.82295263e+00 -1.10070539e+00 2.75915533e-01 1.91602910e+00
1.11552365e-02 -2.73863435e-01 7.93974251e-02 -2.09904730e-01
8.90183628e-01 4.86658961e-01 -7.97841668e-01 -1.82542577e-01
-1.93643212e-01 -4.24696624e-01 1.73123255e-01 7.69568607e-02
-1.42834485e+00 5.20658016e-01 6.09294701e+00 4.84819800e-01
-1.08373380e+00 5.98467179e-02 7.73813426e-01 -2.98491597e-01
-2.39270657e-01 -1.39077634e-01 -9.19426978e-01 -4.63723056e-02
6.12936199e-01 1.33571029e-01 4.86984670e-01 1.26161611e+00
-5.16373701e-02 3.26524764e-01 -1.45667052e+00 1.26483870e+00
7.99692273e-01 -1.37826478e+00 5.47487855e-01 1.90908670e-01
1.23599684e+00 2.37093866e-01 1.18107587e-01 2.97390968e-01
2.65152067e-01 -7.35156476e-01 1.87701494e-01 4.41731721e-01
6.38737917e-01 -1.09345162e+00 6.01750851e-01 6.25276744e-01
-1.32613397e+00 -1.87331870e-01 -5.93249321e-01 3.70294183e-01
1.56197146e-01 4.13281918e-01 -4.56025958e-01 8.73180985e-01
5.58005035e-01 1.48232591e+00 -5.17087460e-01 2.70952880e-01
1.61314219e-01 1.60833061e-01 9.83220804e-03 5.57193100e-01
3.27834994e-01 -7.05102384e-01 5.40738225e-01 6.31726623e-01
2.32069030e-01 -1.23781852e-01 5.77307642e-01 6.81226432e-01
-1.82632670e-01 -5.97769283e-02 -1.12805307e+00 -8.87630135e-02
2.04232007e-01 1.54853272e+00 -5.60370445e-01 -4.78195846e-01
-9.72903550e-01 9.13309276e-01 7.45214283e-01 4.78636533e-01
-3.33273113e-01 7.77794197e-02 4.57851499e-01 -3.93781662e-01
6.30903423e-01 -9.53321997e-03 -1.57925770e-01 -1.57595384e+00
1.42296189e-02 -9.68062460e-01 6.78639352e-01 -6.25590861e-01
-1.67708111e+00 5.27428031e-01 9.61215347e-02 -1.86299658e+00
-3.48724931e-01 -6.37100160e-01 -2.71011025e-01 9.13684249e-01
-1.21463585e+00 -1.58624995e+00 -5.93763515e-02 8.39188337e-01
6.69230402e-01 -1.00727034e+00 1.17881405e+00 1.72445193e-01
-5.57019532e-01 4.06099111e-01 5.58386028e-01 2.71727204e-01
6.75318718e-01 -1.20111692e+00 3.00362054e-03 8.22353125e-01
4.76813406e-01 7.22664058e-01 6.95028752e-02 -4.43651021e-01
-1.52958715e+00 -1.20469856e+00 3.94504040e-01 -1.22036822e-01
2.85424173e-01 -2.12741315e-01 -1.00880647e+00 9.39400196e-01
4.00352687e-01 5.94173670e-01 1.46564794e+00 2.36355305e-01
-5.65332055e-01 -2.36150861e-01 -9.73629117e-01 1.97252139e-01
7.33924270e-01 -1.01064432e+00 -6.07199430e-01 3.38239461e-01
4.23240662e-01 -1.27163112e-01 -1.05278182e+00 2.31998757e-01
4.50118184e-01 -1.20184433e+00 1.03968668e+00 -9.86304700e-01
6.32053912e-01 -2.29450658e-01 -6.43507838e-01 -1.77624941e+00
-7.09105968e-01 1.06981404e-01 -4.84283954e-01 1.23230612e+00
1.04725063e-01 -3.97638589e-01 9.00978267e-01 1.36631191e-01
-4.82294634e-02 -8.59293699e-01 -7.57502377e-01 -6.19772971e-01
1.94624607e-02 5.19026257e-02 2.05654427e-01 1.43617654e+00
-2.80768573e-01 8.23656917e-01 -8.91320527e-01 3.07477653e-01
9.19674873e-01 8.44316661e-01 6.59792006e-01 -1.27421963e+00
-3.76630783e-01 9.44570079e-02 -3.29839766e-01 -8.20090055e-01
5.20826161e-01 -1.16844893e+00 -3.87228221e-01 -1.54380679e+00
7.87123740e-01 4.52938117e-02 -5.23032665e-01 2.00706914e-01
-6.89456910e-02 1.55708507e-01 1.09056361e-01 4.03040588e-01
-6.78325534e-01 7.24794805e-01 1.38378477e+00 -1.98152572e-01
-1.37672797e-01 -3.33945081e-02 -8.46982718e-01 9.69714999e-01
4.31591004e-01 -6.50509119e-01 -7.56972849e-01 -4.48828071e-01
-5.37525676e-02 3.03599417e-01 -6.48706257e-02 -8.58284771e-01
1.28740355e-01 -2.04456687e-01 8.10948610e-01 -1.11084890e+00
4.01965916e-01 -1.18997216e+00 1.28070250e-01 1.33366078e-01
-3.95453304e-01 1.93694949e-01 -1.03046581e-01 1.05070698e+00
-6.15141094e-01 -6.32676631e-02 9.64285374e-01 -5.37126243e-01
-7.46473908e-01 6.19997978e-01 -3.49072516e-02 1.61599919e-01
1.04209292e+00 -2.12881610e-01 -4.98972759e-02 -3.85996640e-01
-9.63410079e-01 1.25972167e-01 3.27496141e-01 6.05429471e-01
7.44864225e-01 -1.97598743e+00 -4.67093140e-01 4.74773675e-01
4.60339636e-01 6.12897724e-02 6.77477360e-01 4.34778333e-01
1.43257705e-02 4.68630850e-01 -4.42610085e-01 -8.92869294e-01
-1.45874584e+00 1.05396998e+00 2.13708580e-01 -5.68335891e-01
-5.99086225e-01 7.14848518e-01 6.81711555e-01 -1.00005138e+00
-6.33833334e-02 2.74108022e-01 -6.26002848e-01 3.40536922e-01
6.06806636e-01 3.65121901e-01 -2.69954145e-01 -1.06892884e+00
-2.43065968e-01 8.22213650e-01 -5.77973723e-01 1.94597095e-01
1.93947589e+00 -4.68501806e-01 -3.62308532e-01 9.34192896e-01
1.91604567e+00 -3.56209546e-01 -1.19476116e+00 -6.38284624e-01
-5.32624304e-01 -3.36496294e-01 9.08195302e-02 -3.14408392e-01
-1.47634578e+00 9.79600906e-01 8.54830384e-01 6.37486726e-02
1.13802600e+00 -1.46478996e-01 4.88107890e-01 6.54636204e-01
-1.30872400e-02 -1.32108176e+00 4.16381508e-01 5.15675545e-01
7.12846518e-01 -1.57263148e+00 4.16686177e-01 -2.15693191e-01
-8.45861971e-01 1.27720201e+00 8.35595071e-01 -2.97911167e-01
1.03109396e+00 -1.65786311e-01 1.06570356e-01 -5.18007457e-01
-6.44650698e-01 3.05213243e-01 3.94790471e-01 6.71041608e-01
1.75215960e-01 6.10822774e-02 4.85235691e-01 2.89839864e-01
3.73132706e-01 -1.66080728e-01 4.77027357e-01 9.24305260e-01
-4.92312089e-02 -9.76601243e-01 -2.65488088e-01 6.18793547e-01
-3.14309031e-01 1.07083067e-01 -3.52536350e-01 6.92156017e-01
1.23633452e-01 5.29133439e-01 -7.16229081e-02 -5.45825541e-01
8.23779926e-02 2.55853117e-01 3.12769413e-01 -9.16947067e-01
-2.47104004e-01 3.46723318e-01 -2.12209791e-01 -3.77718300e-01
-1.13894427e+00 -7.48663247e-01 -7.63736188e-01 5.09126931e-02
-2.45400280e-01 -9.59253386e-02 2.63388932e-01 1.11901236e+00
4.69531775e-01 3.17109406e-01 1.12647402e+00 -1.07083416e+00
-4.90610152e-01 -8.05542827e-01 -1.14635658e+00 6.15667820e-01
4.16543126e-01 -7.14092672e-01 -3.29935104e-01 1.56716987e-01] | [8.397510528564453, 4.581705093383789] |
cf784f9b-b57c-45e3-81f4-92ae0e2ada70 | decoupling-classifier-for-boosting-few-shot | null | null | https://openreview.net/pdf?id=dVXO3Orjmxk | https://openreview.net/pdf?id=dVXO3Orjmxk | Decoupling Classifier for Boosting Few-shot Object Detection and Instance Segmentation | This paper focus on few-shot object detection~(FSOD) and instance segmentation~(FSIS), which requires a model to quickly adapt to novel classes with a few labeled instances. The existing methods severely suffer from bias classification because of the missing label issue which naturally exists in a few-shot scenario and is first formally proposed by us. Our analysis suggests that the standard classification head of most FSOD or FSIS models needs to be decoupled to mitigate the bias classification. Therefore, we propose an embarrassingly simple but effective method that decouples the standard classifier into two heads. Then, these two individual heads are capable of independently addressing clear positive samples and noisy negative samples which are caused by the missing label. In this way, the model can effectively learn novel classes while mitigating the effects of noisy negative samples. Without bells and whistles, our model without any additional computation cost and parameters consistently outperforms its baseline and state-of-the-art by a large margin on PASCAL VOC and MS-COCO benchmarks for FSOD and FSIS tasks. The code is available at https://github.com/gaobb/DCFS. | ['Chengjie Wang', 'Xi Wang', 'Guannan Jiang', 'Jinxiang Lai', 'Jun Liu', 'Congchong Nie', 'Zhongyi Huang', 'Xiaochen Chen', 'Bin-Bin Gao'] | 2022-09-05 | null | null | null | thirty-sixth-conference-on-neural-information | ['few-shot-object-detection'] | ['computer-vision'] | [ 3.78163069e-01 -1.11791983e-01 -3.72629076e-01 -5.78333080e-01
-9.85644221e-01 -5.42378068e-01 5.17881036e-01 -6.44994155e-02
-4.80034202e-01 7.60623217e-01 -5.14112771e-01 -3.18850614e-02
1.09922938e-01 -5.80869555e-01 -6.73269689e-01 -9.61531162e-01
3.65195394e-01 3.37968379e-01 7.59680569e-01 3.72632109e-02
1.32990420e-01 2.14574531e-01 -1.92511797e+00 2.34110400e-01
9.55439031e-01 9.56275225e-01 2.48314172e-01 5.95339417e-01
-3.77589434e-01 7.44065404e-01 -7.11778879e-01 -3.99579287e-01
2.94947773e-01 -4.07409847e-01 -6.99079633e-01 8.26563239e-02
7.10082889e-01 -2.46677101e-01 -2.36917153e-01 1.32813179e+00
6.55114710e-01 2.67816991e-01 6.75568283e-01 -1.37029052e+00
-3.45156461e-01 5.89921772e-01 -7.48435915e-01 4.67422545e-01
-1.38260916e-01 4.63114500e-01 9.93135214e-01 -1.06557834e+00
7.24056780e-01 1.14793098e+00 5.96287966e-01 9.73279417e-01
-1.25972295e+00 -8.63999844e-01 5.38591802e-01 3.48883420e-01
-1.32689774e+00 -5.32435060e-01 5.75643361e-01 -4.86375600e-01
6.52953207e-01 4.40434992e-01 3.56848657e-01 1.25236011e+00
-9.53205824e-02 1.38757348e+00 1.08203351e+00 -3.23739916e-01
3.96285892e-01 3.83072793e-01 8.09316218e-01 3.74634743e-01
3.02200258e-01 5.38763180e-02 -4.02274549e-01 1.61355048e-01
4.43745293e-02 2.06126869e-01 -3.46339732e-01 -3.11258793e-01
-8.58968258e-01 6.51220024e-01 5.02382755e-01 1.12798601e-01
8.70009661e-02 5.79631627e-02 4.63228405e-01 -4.71008047e-02
5.21765947e-01 1.93534061e-01 -4.51767981e-01 -4.84707467e-02
-9.43910182e-01 3.04872990e-01 5.47776222e-01 1.21559227e+00
6.70371115e-01 1.22511061e-02 -3.68084222e-01 9.63515878e-01
6.61424398e-02 4.49517727e-01 4.30761516e-01 -7.39798784e-01
1.77051648e-01 5.18483877e-01 1.32420555e-01 -4.59024847e-01
-2.31445283e-01 -6.82452738e-01 -4.82758224e-01 1.06887139e-01
2.79416382e-01 -1.30141407e-01 -1.43744421e+00 1.60161340e+00
6.14720047e-01 4.78559703e-01 -1.61044925e-01 7.35588610e-01
1.10804915e+00 6.35953784e-01 1.70092091e-01 -3.25254202e-01
1.03570461e+00 -1.17130947e+00 -8.90286267e-01 -5.57578921e-01
5.17126024e-01 -4.33006674e-01 1.22488225e+00 3.14106107e-01
-7.97130466e-01 -4.26912159e-01 -1.17447436e+00 -2.06442438e-02
-4.55201089e-01 -7.75606632e-02 5.76273561e-01 5.63884676e-01
-4.60163772e-01 4.79770839e-01 -8.09565365e-01 -4.13264066e-01
8.83631408e-01 2.65946865e-01 1.33387327e-01 -3.51803571e-01
-9.72052991e-01 5.13695598e-01 4.66328233e-01 6.81049898e-02
-9.27972853e-01 -6.25800371e-01 -6.88315511e-01 -1.08851314e-01
9.21682656e-01 -2.32712656e-01 1.66000557e+00 -9.14014697e-01
-1.20118201e+00 7.91034997e-01 -3.38626206e-01 -2.91643262e-01
6.76749945e-01 -2.41653785e-01 -3.30198228e-01 4.82799932e-02
3.56373698e-01 8.01022649e-01 9.10114467e-01 -1.46628428e+00
-1.05717468e+00 -2.58332461e-01 -1.06166236e-01 6.56085014e-02
-2.02472642e-01 7.53681455e-03 -5.68459868e-01 -6.59672141e-01
-4.75561544e-02 -8.59933019e-01 -7.02928826e-02 -3.87954488e-02
-5.04955053e-01 -3.75512779e-01 9.47847545e-01 -1.83993071e-01
1.38035822e+00 -2.39274907e+00 -1.37648135e-02 -2.99322277e-01
1.38097122e-01 6.60336971e-01 -1.70322344e-01 5.69398627e-02
1.36911929e-01 1.46993220e-01 -4.53081429e-01 -3.09424639e-01
-1.69340037e-02 1.86120182e-01 -2.90664911e-01 3.69190574e-01
4.14117843e-01 9.29197013e-01 -1.16174769e+00 -3.97708297e-01
2.62195207e-02 1.13982067e-01 -2.24277809e-01 1.87235460e-01
-4.16896135e-01 2.61284202e-01 -3.19341302e-01 1.00391829e+00
7.60511875e-01 -3.11866969e-01 -9.81526449e-02 1.58008590e-01
3.04456372e-02 8.53852481e-02 -1.44676876e+00 1.36588097e+00
1.67109311e-01 3.24071795e-01 -7.61830658e-02 -1.07810128e+00
7.21957862e-01 2.93471990e-03 1.21041737e-01 -4.35489804e-01
3.24453086e-01 3.50261956e-01 -1.49040120e-02 -5.00276506e-01
1.38237849e-01 -2.69916773e-01 -6.42008632e-02 1.51436612e-01
2.87978739e-01 6.03247769e-02 4.03564095e-01 3.03097010e-01
1.06237137e+00 -1.86657626e-02 2.50094384e-01 -1.89239711e-01
2.63733089e-01 9.34372768e-02 1.11808956e+00 1.11216676e+00
-7.36161709e-01 6.45353734e-01 2.62973756e-01 -1.40288129e-01
-4.51183587e-01 -1.08748055e+00 -1.56283259e-01 1.37540579e+00
4.34077024e-01 -1.76441789e-01 -7.11112320e-01 -1.04286647e+00
-2.74751848e-03 9.76568937e-01 -5.87300539e-01 -3.15940768e-01
-3.57494175e-01 -1.05396271e+00 3.64717662e-01 6.16994798e-01
3.42620224e-01 -1.00607288e+00 -4.09537196e-01 2.99710065e-01
5.53360544e-02 -9.35057580e-01 -3.02529335e-01 6.18660569e-01
-6.85031593e-01 -1.10929644e+00 -6.55663192e-01 -8.08735251e-01
6.04201913e-01 5.74980438e-01 9.14558172e-01 -4.66517583e-02
-5.91035187e-01 9.46540385e-02 -5.32882452e-01 -9.25407290e-01
-1.65194441e-02 1.17731124e-01 -1.37853339e-01 1.28464743e-01
6.38914347e-01 -1.72305346e-01 -5.81729054e-01 2.77369678e-01
-8.10722411e-01 4.39924141e-03 4.00239885e-01 9.63187635e-01
6.32807374e-01 -1.81668073e-01 8.52243662e-01 -1.34449434e+00
9.52304080e-02 -6.78508162e-01 -4.67464715e-01 3.13604087e-01
-6.66878402e-01 -3.09087336e-01 4.79318738e-01 -5.86728156e-01
-1.23512864e+00 1.80057481e-01 -1.02648772e-02 -3.48662078e-01
-4.58354861e-01 3.53047810e-02 -1.36649787e-01 7.03339055e-02
7.12344944e-01 5.58478087e-02 -4.23134536e-01 -7.02909529e-01
2.69930840e-01 8.71781528e-01 5.38895607e-01 -3.26591551e-01
6.80528164e-01 6.22406840e-01 -4.86874998e-01 -7.29834259e-01
-1.31845963e+00 -8.29361439e-01 -6.44861996e-01 -2.60132372e-01
6.13381386e-01 -8.59438300e-01 -1.54872522e-01 8.51626277e-01
-8.66959691e-01 -2.71464437e-01 -5.44298470e-01 1.60057351e-01
-1.61739707e-01 1.81954294e-01 -6.32492483e-01 -9.57626402e-01
-3.28113377e-01 -1.08101547e+00 1.04360604e+00 6.24973416e-01
1.74316671e-02 -4.67587411e-01 -2.15103492e-01 3.07755589e-01
1.74417719e-01 9.27727595e-02 6.19393826e-01 -8.63258064e-01
-6.80059791e-01 -4.14225787e-01 -2.30603546e-01 4.06851232e-01
-3.32612358e-02 1.05478287e-01 -1.39366531e+00 -3.92781049e-01
-8.62930843e-04 -4.67901409e-01 1.31236851e+00 2.64947623e-01
1.12515628e+00 -1.84736978e-02 -3.93191367e-01 6.54333711e-01
1.39921713e+00 4.32098091e-01 3.07414591e-01 1.52651578e-01
6.64156973e-01 5.30046403e-01 9.81015265e-01 3.36364686e-01
1.95371851e-01 4.20125633e-01 3.69792491e-01 5.89559600e-02
-1.61161989e-01 8.44011083e-02 2.77817100e-01 6.22660220e-01
2.97365904e-01 -4.49280471e-01 -9.05721068e-01 7.55529344e-01
-2.02364826e+00 -7.86808193e-01 -1.48515344e-01 1.98394847e+00
9.26149905e-01 3.07580650e-01 1.62494928e-02 2.85569131e-01
8.91315639e-01 2.25083396e-01 -8.27071309e-01 -3.58154438e-02
2.28565261e-02 2.15467796e-01 3.77074659e-01 2.98847824e-01
-1.23255789e+00 1.08080399e+00 5.53391790e+00 1.08251750e+00
-9.84417677e-01 2.49980941e-01 6.84968054e-01 -4.23210204e-01
5.37104383e-02 -1.46629423e-01 -1.36835682e+00 5.07926166e-01
6.42259121e-01 -1.42082393e-01 1.56153724e-01 9.09865499e-01
-1.56869784e-01 -3.43604684e-01 -1.04917586e+00 8.66095960e-01
8.98280293e-02 -1.12847292e+00 -9.85997021e-02 -4.32730824e-01
8.39674711e-01 2.52641320e-01 -2.96444260e-02 6.51086986e-01
6.35218620e-02 -6.92942739e-01 7.74357140e-01 3.61691207e-01
7.02834070e-01 -4.96030986e-01 6.75809681e-01 6.47692680e-01
-1.03070199e+00 -3.56028736e-01 -3.78505856e-01 4.74706925e-02
-8.11365694e-02 6.11662924e-01 -6.49838567e-01 2.79584318e-01
8.00877035e-01 8.14921081e-01 -7.14640379e-01 1.30264199e+00
-3.24827641e-01 8.08761597e-01 -1.66910961e-01 -5.73671348e-02
2.47488722e-01 1.83369994e-01 6.52417183e-01 1.29809976e+00
6.74673393e-02 2.27996141e-01 2.86937416e-01 7.60601521e-01
-1.11586519e-01 -9.27730203e-02 -2.62377501e-01 -2.10183058e-02
6.06299162e-01 1.22536337e+00 -9.46397960e-01 -6.55686915e-01
-4.94504243e-01 8.71306539e-01 3.18372309e-01 4.43898827e-01
-8.94926190e-01 -4.97512758e-01 5.21213830e-01 2.83737574e-02
5.23936212e-01 4.10160929e-01 -4.51762259e-01 -1.33390689e+00
1.00352310e-01 -7.60954082e-01 4.93724138e-01 -1.74441725e-01
-1.51179171e+00 4.04864550e-01 -5.82318977e-02 -1.17384803e+00
6.17436357e-02 -5.59836984e-01 -7.07606792e-01 5.20461798e-01
-1.80058801e+00 -9.17451560e-01 -4.08887446e-01 3.53356630e-01
9.52476263e-01 1.72553986e-01 4.18100566e-01 4.81963098e-01
-1.06776893e+00 6.69640303e-01 2.70859271e-01 1.38210123e-02
9.01503742e-01 -1.37849069e+00 3.63036156e-01 1.05659616e+00
-1.39769260e-02 3.80009383e-01 6.53963566e-01 -7.19122887e-01
-1.00509453e+00 -1.29125607e+00 8.25441957e-01 -3.80938828e-01
4.66719866e-01 -5.58912337e-01 -1.25349402e+00 6.16762936e-01
-2.76159137e-01 4.54196155e-01 5.47373354e-01 -1.22336075e-01
-2.26336256e-01 -2.50210613e-01 -1.12801099e+00 3.63620371e-01
1.18470275e+00 -1.84509069e-01 -5.76848209e-01 3.98842931e-01
7.53426373e-01 -2.82695502e-01 -3.31681639e-01 7.96917558e-01
2.73556650e-01 -8.05859327e-01 7.73886919e-01 -6.12686753e-01
1.30297497e-01 -3.34187448e-01 -5.96644171e-02 -9.29644525e-01
-2.19078243e-01 -3.85631263e-01 -3.96747231e-01 1.47348309e+00
3.64462495e-01 -4.97955650e-01 5.73139369e-01 5.32746255e-01
-1.47824362e-01 -8.42326462e-01 -8.62456262e-01 -1.04563892e+00
-1.47703886e-01 -4.80364442e-01 2.80686110e-01 1.01446855e+00
-3.61345202e-01 3.64788085e-01 -3.15163642e-01 6.12137839e-02
8.00719440e-01 9.24633890e-02 7.89239287e-01 -1.15828502e+00
-1.67823195e-01 -2.92458802e-01 -7.56408423e-02 -8.85943055e-01
-1.16061896e-01 -7.88994133e-01 6.52306080e-01 -1.30645800e+00
4.96226907e-01 -3.83947551e-01 -6.49515510e-01 5.85043550e-01
-7.67064631e-01 3.74020666e-01 3.84658903e-01 2.00180158e-01
-9.72393751e-01 4.98044908e-01 1.04295802e+00 -2.35754371e-01
-2.97112226e-01 2.52338380e-01 -7.52142608e-01 9.08868194e-01
7.39644170e-01 -7.63532579e-01 -3.63663733e-01 -1.67215779e-01
-2.47058630e-01 -4.57935452e-01 2.42914781e-01 -9.65310991e-01
1.30612314e-01 -2.14555100e-01 2.17984572e-01 -6.02836311e-01
1.34220973e-01 -5.15869975e-01 -2.00227439e-01 4.71498221e-01
-2.27701396e-01 -6.72658205e-01 1.28614441e-01 8.69373679e-01
-9.46300104e-02 -5.12854457e-01 1.20790696e+00 -1.83764353e-01
-1.11035156e+00 3.51603150e-01 -1.91103131e-01 2.94438213e-01
1.26851010e+00 -1.93638161e-01 -5.96215427e-01 1.27710506e-01
-6.87758923e-01 5.48800051e-01 3.17719251e-01 5.34003913e-01
4.69152302e-01 -9.97173905e-01 -4.94216591e-01 1.81897551e-01
3.87692899e-01 4.02779043e-01 3.34195167e-01 7.81504154e-01
-1.18431136e-01 5.65711148e-02 1.90786451e-01 -7.19525754e-01
-1.25163400e+00 6.15835249e-01 2.18084410e-01 -6.21736124e-02
-6.56454921e-01 1.30651724e+00 3.82056475e-01 -4.73082095e-01
6.75871491e-01 -2.24313214e-01 -8.58736709e-02 3.60568732e-01
5.89689076e-01 5.84175587e-01 -6.13785349e-02 -3.97508174e-01
-4.72406656e-01 2.76297092e-01 -4.80806530e-01 2.77427137e-01
1.33485651e+00 -1.17942117e-01 1.84617475e-01 8.95232260e-01
9.48280632e-01 -3.80013168e-01 -1.47853172e+00 -4.36042488e-01
1.84457079e-01 -4.96696323e-01 -8.76324773e-02 -1.03258073e+00
-8.18591595e-01 1.00623596e+00 7.28086174e-01 -1.16539463e-01
1.01640904e+00 1.29173085e-01 8.64268959e-01 1.91416532e-01
1.88306436e-01 -1.35696197e+00 6.59616739e-02 4.59752619e-01
4.06728685e-01 -1.54333878e+00 -6.08443022e-02 -6.65895343e-01
-7.22826719e-01 7.27454722e-01 9.69696164e-01 -9.97895002e-03
4.52021301e-01 2.33520716e-01 5.75042889e-02 -1.78768411e-01
-7.70960748e-01 -3.91009957e-01 3.97830695e-01 4.57328945e-01
5.94413467e-03 2.92986631e-02 -3.94343525e-01 8.98454666e-01
4.48176026e-01 -4.10693465e-03 3.80682915e-01 1.29153502e+00
-7.88441956e-01 -7.62407839e-01 -2.19907537e-01 7.10474491e-01
-4.44874734e-01 -5.82888126e-02 -3.83805484e-01 8.04652750e-01
4.74592358e-01 8.33800197e-01 -4.69201431e-02 -2.35252500e-01
3.56429845e-01 3.48066688e-01 1.54251665e-01 -1.07599521e+00
-4.81730431e-01 2.18428835e-01 -1.75693691e-01 -5.25218248e-01
-3.71083498e-01 -7.27750301e-01 -1.27561593e+00 8.66350532e-02
-8.26890349e-01 -1.71521947e-01 3.08459342e-01 8.62560451e-01
1.45754069e-01 6.27451479e-01 5.64924538e-01 -7.74327040e-01
-7.68417954e-01 -9.31619465e-01 -6.72066331e-01 4.05212611e-01
3.65151435e-01 -6.92201614e-01 -5.21722734e-01 -1.17126070e-01] | [9.508142471313477, 1.7347148656845093] |
eb667cdf-4899-45b4-b94e-c80203bbdc1f | sangeet-a-xml-based-open-dataset-for-research | 2306.04148 | null | https://arxiv.org/abs/2306.04148v1 | https://arxiv.org/pdf/2306.04148v1.pdf | SANGEET: A XML based Open Dataset for Research in Hindustani Sangeet | It is very important to access a rich music dataset that is useful in a wide variety of applications. Currently, available datasets are mostly focused on storing vocal or instrumental recording data and ignoring the requirement of its visual representation and retrieval. This paper attempts to build an XML-based public dataset, called SANGEET, that stores comprehensive information of Hindustani Sangeet (North Indian Classical Music) compositions written by famous musicologist Pt. Vishnu Narayan Bhatkhande. SANGEET preserves all the required information of any given composition including metadata, structural, notational, rhythmic, and melodic information in a standardized way for easy and efficient storage and extraction of musical information. The dataset is intended to provide the ground truth information for music information research tasks, thereby supporting several data-driven analysis from a machine learning perspective. We present the usefulness of the dataset by demonstrating its application on music information retrieval using XQuery, visualization through Omenad rendering system. Finally, we propose approaches to transform the dataset for performing statistical and machine learning tasks for a better understanding of Hindustani Sangeet. The dataset can be found at https://github.com/cmisra/Sangeet. | ['Swarup Chattopadhyay', 'Chandan Misra'] | 2023-06-07 | null | null | null | null | ['music-information-retrieval', 'information-retrieval'] | ['music', 'natural-language-processing'] | [ 1.85924754e-01 -5.03934443e-01 -2.89116912e-02 6.08819984e-02
-9.67638612e-01 -1.04879415e+00 4.03277695e-01 4.07274336e-01
5.42874001e-02 4.17146981e-01 4.68783826e-01 3.87046598e-02
-7.03283012e-01 -7.66039848e-01 -1.41694814e-01 -6.74930215e-01
-9.53376666e-02 4.70204532e-01 4.30848263e-02 -4.35249120e-01
7.21539021e-01 5.32808602e-01 -2.00154400e+00 6.58492088e-01
4.38442200e-01 9.10119832e-01 5.75977564e-01 7.18273640e-01
-2.77248442e-01 6.33965850e-01 -9.34731543e-01 -2.90931493e-01
2.07800001e-01 -6.20564878e-01 -8.55699778e-01 -4.89643663e-01
5.37052929e-01 3.04701447e-01 6.44456819e-02 8.97380710e-01
7.17589915e-01 7.18356669e-02 4.10157651e-01 -1.23925042e+00
-3.60315859e-01 1.21046948e+00 -2.51892745e-01 -5.11480123e-02
5.30247867e-01 -2.20030263e-01 1.20108521e+00 -4.18994308e-01
1.04610729e+00 7.51592159e-01 5.02153277e-01 1.35695234e-01
-9.39607203e-01 -6.60293758e-01 -7.19623744e-01 5.39494872e-01
-1.42127693e+00 -2.75355995e-01 1.10277581e+00 -5.73822081e-01
4.64368433e-01 1.08464825e+00 1.04910195e+00 5.54399014e-01
-3.13119113e-01 5.47747850e-01 9.84788895e-01 -6.11157000e-01
4.64063995e-02 -1.51189622e-02 2.06817407e-03 3.70838553e-01
-2.25988910e-01 -4.06517327e-01 -1.14007282e+00 -2.00016573e-01
7.43871808e-01 -1.42799720e-01 -3.54468107e-01 -2.75127321e-01
-1.42548239e+00 4.63595182e-01 6.74618930e-02 7.39063203e-01
-3.47752810e-01 -7.18525657e-03 8.43349397e-01 5.75299978e-01
-2.80049890e-02 4.57379848e-01 -4.12935436e-01 -7.44352281e-01
-1.13550019e+00 4.39631015e-01 6.86748326e-01 8.68445277e-01
4.34062928e-01 3.03023793e-02 9.20343325e-02 1.01941848e+00
1.64586425e-01 3.51026356e-01 7.35128224e-01 -9.84093785e-01
2.70394713e-01 7.05606401e-01 -3.81033778e-01 -1.03043532e+00
-1.26306877e-01 -3.28672737e-01 -6.24827266e-01 3.36857855e-01
2.99955964e-01 6.36887014e-01 -3.06815088e-01 1.37630653e+00
3.03863138e-01 -2.80441165e-01 -1.59157559e-01 7.71755695e-01
1.23953426e+00 5.96853316e-01 -1.54825717e-01 -2.17231870e-01
1.49276030e+00 -3.88688028e-01 -7.31291473e-01 6.71709359e-01
4.82253462e-01 -1.45358968e+00 1.55778205e+00 9.87670541e-01
-1.23962629e+00 -4.35510099e-01 -9.88233924e-01 -4.10388052e-01
-3.32021356e-01 2.74929911e-01 4.43036407e-01 5.27350664e-01
-5.16493678e-01 8.36726010e-01 -5.77311099e-01 -4.09477621e-01
3.10577333e-01 1.62806839e-01 -3.52121651e-01 5.28938472e-01
-6.45621061e-01 3.76328290e-01 6.05005741e-01 -2.14654282e-01
-5.71191847e-01 -9.25884426e-01 -3.67589802e-01 -1.14674792e-01
3.66091311e-01 -3.15989107e-01 1.20761788e+00 -7.94454873e-01
-1.49080169e+00 1.00903726e+00 2.64352083e-01 -2.19692335e-01
4.23983544e-01 -6.46873042e-02 -4.96814311e-01 3.61732766e-02
-1.40165210e-01 -1.39144123e-01 3.79718661e-01 -1.02197409e+00
-3.97058338e-01 -7.16754913e-01 -2.10968435e-01 1.06028304e-01
-3.52984041e-01 3.90310317e-01 -6.39817536e-01 -1.23582184e+00
1.52566969e-01 -7.52051473e-01 4.17011172e-01 -2.54794449e-01
-5.55925250e-01 -8.70282669e-03 9.21351910e-01 -1.05993974e+00
1.74119353e+00 -2.30378628e+00 2.89144099e-01 3.96846950e-01
-2.03723367e-02 9.29203928e-02 4.04778086e-02 1.08724284e+00
-7.32894167e-02 2.26510949e-02 -3.36587757e-01 2.20140532e-01
2.09975213e-01 7.06007034e-02 -3.82716060e-01 1.49044737e-01
-8.07573855e-01 6.29669070e-01 -4.58728194e-01 -7.20299780e-01
1.96594045e-01 5.58410883e-01 -4.42862272e-01 -4.66234721e-02
-2.32841134e-01 7.36352503e-01 -1.79925382e-01 6.97373688e-01
2.86339343e-01 2.12097272e-01 2.71695852e-01 -4.18523550e-01
-5.91460049e-01 5.58854580e-01 -1.55015457e+00 2.35979748e+00
-3.12102467e-01 7.20337152e-01 7.84699433e-03 -6.25068903e-01
1.03589749e+00 3.73342276e-01 8.27941716e-01 -6.88799262e-01
7.32335374e-02 3.03169340e-01 -7.16255754e-02 -5.55548072e-01
8.51419926e-01 -7.88186342e-02 -1.57954663e-01 7.37560332e-01
-8.12112540e-02 -2.07430094e-01 4.68066305e-01 2.01258793e-01
6.55505955e-01 5.23604274e-01 5.59245050e-01 -1.43293679e-01
4.63492602e-01 2.08475947e-01 2.84751922e-01 1.94606915e-01
4.52127993e-01 6.27259195e-01 1.83276713e-01 -1.94369391e-01
-1.20269763e+00 -9.41904128e-01 -1.80088744e-01 1.35522509e+00
-3.89892012e-01 -1.13644350e+00 -5.47997355e-01 2.31063664e-01
-2.59984881e-01 4.47706521e-01 -3.58370334e-01 2.62100726e-01
-5.55644274e-01 -1.46855533e-01 7.01244175e-01 -2.13164780e-02
3.64929259e-01 -1.43631470e+00 -9.06708360e-01 1.29106632e-02
-3.18598151e-01 -2.54309386e-01 -3.71619344e-01 -1.54054657e-01
-9.30753648e-01 -1.26150179e+00 -5.28041959e-01 -6.85379744e-01
-2.51451612e-01 -1.50627866e-01 1.20867205e+00 -2.41693226e-03
-7.95658052e-01 2.83385187e-01 -5.28086305e-01 -6.48665607e-01
-4.40255135e-01 3.00115556e-01 -2.49047130e-01 -2.68336058e-01
-5.66303544e-02 -9.87999678e-01 -5.17160833e-01 7.78916702e-02
-1.09137034e+00 1.63153052e-01 1.48715809e-01 1.49804011e-01
8.06716383e-01 -1.14110380e-01 3.36858422e-01 -9.77682412e-01
6.10247254e-01 -1.55500248e-01 -3.56619835e-01 2.88954079e-01
-3.08857769e-01 -1.51698619e-01 5.17006576e-01 -1.81510240e-01
-7.61659324e-01 -1.76554725e-01 -1.99208334e-01 -1.82879686e-01
2.71198992e-02 6.34741902e-01 -4.06631023e-01 3.28171015e-01
6.11266077e-01 3.14483166e-01 -3.38110805e-01 -1.27686834e+00
4.64166671e-01 1.02776802e+00 9.06970918e-01 -6.52104437e-01
4.24294859e-01 2.73290277e-01 -3.67163531e-02 -9.91460621e-01
-3.21762651e-01 -5.77140927e-01 -8.55103910e-01 -4.47245032e-01
5.19402146e-01 -3.80401045e-01 -8.96412969e-01 1.36235908e-01
-7.26197600e-01 8.69425852e-03 -8.87627363e-01 3.26282710e-01
-8.35462928e-01 3.39356422e-01 -3.07644129e-01 -7.63599575e-01
-6.89575255e-01 -4.52148646e-01 7.82713294e-01 -1.63552701e-01
-5.93218565e-01 -5.96546352e-01 5.95014215e-01 4.93187845e-01
9.32634622e-02 5.22862434e-01 1.32608533e+00 -5.71992934e-01
-6.42515898e-01 5.36580123e-02 2.87368715e-01 -6.21397980e-03
1.82425186e-01 2.89918602e-01 -9.69442010e-01 -1.42913451e-02
-2.23368436e-01 -1.76952347e-01 3.29423964e-01 6.63968027e-02
1.23627663e+00 -3.78058493e-01 2.91689038e-01 6.06573701e-01
1.53649700e+00 3.62872303e-01 6.25267982e-01 5.58529854e-01
6.60590410e-01 4.19948459e-01 5.47512710e-01 8.66242766e-01
1.31094038e-01 1.11716676e+00 3.11297923e-01 4.36153978e-01
-4.65440273e-01 -4.16050255e-01 6.43813610e-02 1.44050705e+00
-6.90590620e-01 1.54947057e-01 -7.95655429e-01 3.89108956e-01
-1.79508317e+00 -1.22861421e+00 -3.09424281e-01 2.56966257e+00
9.46318030e-01 -5.85170984e-01 5.94951451e-01 9.88480151e-01
4.26277459e-01 3.77424918e-02 -1.56937107e-01 -4.04759049e-01
-2.66250014e-01 4.92456049e-01 -1.36929289e-01 3.24396566e-02
-7.31443524e-01 6.65633023e-01 5.38135147e+00 9.76612270e-01
-1.12139499e+00 2.94996828e-01 -4.08042073e-01 -4.69759732e-01
-3.99699211e-01 6.57209009e-02 -8.23731124e-02 3.41949522e-01
9.69431698e-01 -4.93637383e-01 7.00447202e-01 5.02313793e-01
3.34663540e-01 1.11215850e-02 -6.56238139e-01 1.45275521e+00
1.07938468e-01 -1.63860762e+00 2.71665037e-01 2.50158701e-02
3.03166240e-01 -1.81176558e-01 -5.13801724e-02 -7.60986134e-02
-3.33011448e-01 -7.70290315e-01 1.15633619e+00 7.53940105e-01
8.97060633e-01 -8.98122191e-01 2.82310694e-01 1.58234999e-01
-1.28600514e+00 1.68574482e-01 -3.74317728e-02 9.15150996e-03
1.22605013e-02 1.79036930e-01 -4.89500821e-01 9.14547801e-01
8.71081114e-01 7.70774782e-01 -6.18953526e-01 1.23075366e+00
2.10051775e-01 5.87349594e-01 -2.69670367e-01 4.12177891e-02
-4.65209663e-01 -3.82000178e-01 7.69486725e-01 1.07313585e+00
7.28951633e-01 -1.52752116e-01 -1.21983536e-01 5.38210750e-01
1.20542288e-01 1.07116628e+00 -4.88938868e-01 -3.24400365e-01
4.29859787e-01 1.17539322e+00 -7.54299402e-01 -1.50157228e-01
1.19752996e-01 7.05891550e-01 -5.41207828e-02 -1.62945032e-01
-5.36992252e-01 -5.24920642e-01 3.39421570e-01 4.31823045e-01
1.21676408e-01 -3.95506501e-01 -3.29086810e-01 -1.03610814e+00
4.72487025e-02 -1.18753088e+00 8.25655818e-01 -1.00875545e+00
-8.37484896e-01 7.35412717e-01 2.11336706e-02 -1.64357805e+00
-2.60952979e-01 -3.71113837e-01 -2.51264006e-01 6.23424590e-01
-5.83618760e-01 -1.39715838e+00 -3.02586555e-01 8.72227788e-01
4.18695003e-01 -4.48124558e-01 1.25762355e+00 6.58346772e-01
-1.70380220e-01 1.16263822e-01 3.37098867e-01 2.62664966e-02
7.79853404e-01 -1.23014057e+00 -2.30613992e-01 3.35491478e-01
1.07160044e+00 5.62711000e-01 7.82993674e-01 -5.77270567e-01
-1.55933559e+00 -4.64245260e-01 8.24682176e-01 -3.73769104e-01
6.39955103e-01 -1.55129135e-01 -7.01536715e-01 5.29906034e-01
3.20640504e-01 -7.30713248e-01 1.35430491e+00 2.41864353e-01
-3.46212715e-01 -2.59623230e-01 -7.30493367e-01 4.89719480e-01
1.01598251e+00 -7.25329041e-01 -5.90655506e-01 2.26045847e-01
1.05344392e-02 -2.46575147e-01 -1.25015283e+00 1.19214449e-02
9.12362158e-01 -1.04529226e+00 7.88262486e-01 -4.56895500e-01
3.36557984e-01 -5.08140743e-01 -4.44375098e-01 -9.10094976e-01
1.22811154e-01 -9.19876635e-01 3.23182829e-02 1.44082594e+00
1.48712978e-01 4.48266119e-02 6.66270971e-01 -2.44270544e-02
-1.25502780e-01 -1.32476315e-01 -8.86296630e-01 -5.52726686e-01
-3.54675531e-01 -9.20381188e-01 7.34639525e-01 1.08697248e+00
1.26789913e-01 2.48498887e-01 -5.46985984e-01 -3.60550910e-01
4.94366258e-01 7.65972793e-01 1.04329693e+00 -1.54831731e+00
-3.55943680e-01 -5.65820217e-01 -6.90717220e-01 -2.23541617e-01
-5.54219484e-01 -1.38613391e+00 -6.48595512e-01 -1.62363100e+00
2.55463779e-01 -4.95090425e-01 -2.92621702e-01 5.33549905e-01
9.01515186e-01 9.56980646e-01 7.26135850e-01 7.84117877e-01
-3.31850022e-01 1.58398792e-01 1.26678526e+00 1.20633081e-01
-4.18064266e-01 -6.78416491e-02 -5.37254810e-01 6.36844695e-01
9.07508910e-01 -5.52587330e-01 -3.45858097e-01 -1.54590130e-01
5.65909624e-01 -1.01716734e-01 3.49555850e-01 -1.04803717e+00
2.20963791e-01 -1.26923338e-01 5.06472029e-02 -8.97002161e-01
4.11731064e-01 -7.92694330e-01 1.02343106e+00 2.74491757e-01
-4.19345945e-01 1.42827049e-01 1.53920293e-01 2.37198230e-02
-4.76332694e-01 -3.27832520e-01 4.40521359e-01 -1.04958445e-01
-7.12267399e-01 2.14577839e-02 6.96029291e-02 -1.04876205e-01
6.33712411e-01 -3.09327722e-01 -1.30554289e-01 -2.70344824e-01
-1.06374466e+00 -6.26721740e-01 4.88436222e-01 4.54068959e-01
3.60484987e-01 -1.46677387e+00 -6.75500691e-01 2.82969058e-01
3.59912187e-01 -6.67400658e-01 5.33025444e-01 6.47483528e-01
-9.83198047e-01 2.25838840e-01 -8.78521442e-01 -3.24770033e-01
-1.97836363e+00 3.66648555e-01 -3.18251193e-01 1.00626424e-01
-8.74255598e-01 2.45584771e-01 -4.20949221e-01 -3.24178964e-01
3.00212830e-01 -2.68402427e-01 -3.98747385e-01 3.91976982e-01
5.46843767e-01 5.88530362e-01 1.99298292e-01 -8.25096726e-01
-1.50770515e-01 7.92948127e-01 5.04035532e-01 -4.56254691e-01
1.50778973e+00 1.97630897e-02 -5.30579329e-01 1.17189133e+00
9.47091639e-01 7.46511102e-01 -3.35225403e-01 7.98878074e-02
1.82877690e-01 -7.48020411e-01 -1.76043645e-01 -1.09332371e+00
-9.26846087e-01 8.68899882e-01 7.42224455e-01 4.16898042e-01
1.38106370e+00 1.28919691e-01 5.11094332e-01 3.23522925e-01
3.34187627e-01 -9.89039242e-01 -2.99392641e-01 1.51684508e-01
1.36829197e+00 -5.10853231e-01 1.93162128e-01 -1.44318908e-01
-6.61342740e-01 1.26822650e+00 -1.90317720e-01 1.99395344e-01
6.04079425e-01 3.19035292e-01 3.56753051e-01 -3.54063749e-01
-4.56405282e-01 -1.30690634e-01 5.89786172e-01 4.18681473e-01
8.57832253e-01 1.76771522e-01 -6.68978810e-01 8.60866785e-01
-1.20891905e+00 1.18041851e-01 1.63900465e-01 1.01260448e+00
-4.29716974e-01 -1.59997141e+00 -4.35805976e-01 2.46405900e-01
-7.33653188e-01 -1.51563287e-01 -6.13930821e-01 9.20465827e-01
3.45770717e-01 4.00490761e-01 -1.99198127e-02 -1.89217344e-01
3.70511681e-01 1.61158413e-01 7.70090342e-01 -4.30375993e-01
-9.94337916e-01 6.65461659e-01 6.85356557e-02 -2.91413993e-01
-6.64981484e-01 -6.86184764e-01 -1.19723117e+00 -4.00475800e-01
2.67269582e-01 4.77367878e-01 1.08979475e+00 2.54521400e-01
1.88172266e-01 4.88283157e-01 1.66804790e-01 -5.20878494e-01
9.41481516e-02 -1.00559902e+00 -1.01358902e+00 6.72468781e-01
-1.51757315e-01 -1.60625964e-01 1.66742831e-01 4.62986767e-01] | [15.98134708404541, 5.205749988555908] |
96af90a1-4d0e-4a29-8611-9de84b4c2f77 | cross-modal-contrastive-learning-for-speech-1 | 2205.02444 | null | https://arxiv.org/abs/2205.02444v1 | https://arxiv.org/pdf/2205.02444v1.pdf | Cross-modal Contrastive Learning for Speech Translation | How can we learn unified representations for spoken utterances and their written text? Learning similar representations for semantically similar speech and text is important for speech translation. To this end, we propose ConST, a cross-modal contrastive learning method for end-to-end speech-to-text translation. We evaluate ConST and a variety of previous baselines on a popular benchmark MuST-C. Experiments show that the proposed ConST consistently outperforms the previous methods on, and achieves an average BLEU of 29.4. The analysis further verifies that ConST indeed closes the representation gap of different modalities -- its learned representation improves the accuracy of cross-modal speech-text retrieval from 4% to 88%. Code and models are available at https://github.com/ReneeYe/ConST. | ['Lei LI', 'Mingxuan Wang', 'Rong Ye'] | 2022-05-05 | null | https://aclanthology.org/2022.naacl-main.376 | https://aclanthology.org/2022.naacl-main.376.pdf | naacl-2022-7 | ['speech-to-text-translation'] | ['natural-language-processing'] | [ 8.12372193e-02 -1.27935885e-02 -4.44007069e-01 -5.72464287e-01
-1.94510043e+00 -7.66040683e-01 9.60004151e-01 -1.03829004e-01
-3.21749568e-01 6.08664811e-01 9.14380312e-01 -3.61276805e-01
4.82603431e-01 -9.11867917e-02 -5.78295708e-01 -3.93788844e-01
4.79900986e-01 7.78066933e-01 -1.01759024e-01 -3.98283780e-01
-1.99436218e-01 6.22466281e-02 -1.12018299e+00 7.77210355e-01
6.36703372e-01 7.35034645e-01 1.10662684e-01 7.71352291e-01
-2.31204942e-01 7.50642836e-01 -5.86915374e-01 -5.07743359e-01
-1.11438610e-01 -6.11927092e-01 -1.16286695e+00 -2.64004260e-01
7.56454527e-01 -2.71278650e-01 -6.45679712e-01 7.90155590e-01
1.00150585e+00 4.54824895e-01 9.37427878e-01 -1.08931291e+00
-1.29288316e+00 8.03825915e-01 -2.41486743e-01 2.63108343e-01
6.19686902e-01 -2.35925838e-01 1.23158443e+00 -1.39737940e+00
4.82336432e-01 1.50175500e+00 2.80701160e-01 1.12266028e+00
-1.05647504e+00 -5.51183701e-01 -1.63898826e-01 1.01067357e-01
-1.54664481e+00 -1.28158903e+00 3.25080842e-01 3.32824886e-02
1.16270924e+00 6.92784548e-01 -2.23654971e-01 1.50197208e+00
-2.26599604e-01 1.15660882e+00 9.19745862e-01 -5.24068475e-01
-2.11931586e-01 7.95494989e-02 -1.23766266e-01 4.96814489e-01
-3.56288254e-01 -4.85469624e-02 -7.78849840e-01 -9.16832592e-03
2.77889729e-01 -2.86227942e-01 -5.39153278e-01 1.58231288e-01
-1.35099137e+00 7.32117534e-01 2.59485751e-01 4.33358848e-01
2.63735894e-02 2.43543625e-01 8.04018557e-01 5.96377254e-01
6.28851175e-01 2.71651655e-01 -4.68087941e-01 -4.23178315e-01
-7.65259087e-01 1.51065290e-02 6.34409785e-01 1.18551660e+00
2.53609836e-01 1.37077585e-01 -4.62831557e-01 1.35295975e+00
4.00467336e-01 1.02622283e+00 8.08468461e-01 -7.88377464e-01
8.44516218e-01 7.42693543e-02 -1.37245253e-01 -4.01577085e-01
-1.24803251e-02 -1.58067390e-01 -7.44093180e-01 -5.98105371e-01
7.67566189e-02 -1.33549273e-01 -7.89914250e-01 1.89960003e+00
-9.75807682e-02 -4.90352064e-02 6.29289567e-01 9.17943716e-01
1.54606700e+00 1.07208741e+00 8.37688074e-02 -1.53923675e-01
1.25111496e+00 -1.36197054e+00 -1.09502983e+00 -4.28014576e-01
6.58939481e-01 -1.21224463e+00 1.36495066e+00 -4.03833896e-01
-1.33799624e+00 -4.30828661e-01 -5.82373023e-01 -4.83062118e-01
-3.19296241e-01 5.02402663e-01 1.54019773e-01 4.35427248e-01
-1.33334422e+00 1.09971114e-01 -6.29919767e-01 -5.98190546e-01
2.81718224e-02 -4.96517569e-02 -3.24995220e-01 -1.80886939e-01
-1.37932384e+00 1.04572892e+00 3.15469503e-02 -3.37495863e-01
-8.90860260e-01 -3.90993178e-01 -9.18254316e-01 5.34095429e-02
-1.44766821e-02 -7.66061544e-01 1.81031895e+00 -1.00081706e+00
-1.71324253e+00 1.02417195e+00 -6.23781860e-01 -2.62096792e-01
1.69322103e-01 -3.08387667e-01 -6.68420732e-01 2.28192911e-01
-3.94883938e-02 8.40840816e-01 7.80419886e-01 -1.11734211e+00
-1.06593214e-01 -9.05453116e-02 9.79783479e-03 5.74860990e-01
-2.14096874e-01 4.24245477e-01 -7.01981127e-01 -7.17316687e-01
9.88468006e-02 -9.42331970e-01 2.97495514e-01 -4.76696305e-02
-1.47091106e-01 -4.90605772e-01 5.01244366e-01 -9.41280484e-01
1.04433644e+00 -2.11100221e+00 4.11060750e-01 -4.66587782e-01
-2.03064203e-01 1.44531280e-01 -5.10164738e-01 7.76359856e-01
-9.21655968e-02 5.84597401e-02 -1.38319418e-01 -8.09198856e-01
1.80834666e-01 2.09203348e-01 -5.65516293e-01 2.95453757e-01
7.36203939e-02 1.16196358e+00 -7.85937071e-01 -3.85812253e-01
1.13412112e-01 6.41630828e-01 -1.55119985e-01 4.57213879e-01
-4.51124907e-02 3.88090760e-01 -1.65220290e-01 6.66719139e-01
3.72097731e-01 -2.04706639e-01 1.84535176e-01 -6.27268776e-02
3.16602379e-01 1.01022768e+00 -3.34112316e-01 2.05868316e+00
-6.45833492e-01 9.63836014e-01 -1.25592232e-01 -7.53184676e-01
8.19249213e-01 8.03735971e-01 1.18160188e-01 -9.91224289e-01
2.89140165e-01 2.99867511e-01 -5.12546897e-01 -2.77694762e-01
7.31516778e-01 -5.73148504e-02 -2.41225988e-01 5.92716694e-01
3.01594973e-01 -4.87688988e-01 -1.16976209e-01 3.51670772e-01
7.06778705e-01 -1.46110877e-01 1.35677740e-01 -1.48974314e-01
5.63783705e-01 -3.02379608e-01 1.43934503e-01 3.87087584e-01
-2.12281168e-01 8.54067564e-01 1.43036637e-02 6.78133518e-02
-8.96549642e-01 -1.28205609e+00 1.18342079e-01 1.48869717e+00
-3.58578078e-02 -4.73236084e-01 -8.16609859e-01 -7.40942419e-01
-2.44925559e-01 9.71096396e-01 -3.94827724e-01 -3.63019615e-01
-3.76679391e-01 -5.33257350e-02 9.10754561e-01 5.57340741e-01
1.69307306e-01 -7.23317206e-01 1.75924316e-01 -2.09044039e-01
-9.39981699e-01 -1.36281872e+00 -9.76280332e-01 -1.33389682e-01
-6.51492596e-01 -4.35939729e-01 -1.12820101e+00 -1.06084001e+00
5.04822433e-01 6.58215404e-01 1.37937546e+00 -2.04361156e-01
2.14876145e-01 6.15973711e-01 -7.22264886e-01 -1.68210506e-01
-9.28015232e-01 1.30858317e-01 2.49512643e-01 -2.89277643e-01
3.19398195e-01 -1.96620390e-01 -3.45390201e-01 3.07942957e-01
-7.30690479e-01 8.98074359e-02 3.65324944e-01 7.90291965e-01
5.07149577e-01 -9.42042291e-01 6.31542623e-01 -1.35263726e-01
8.76747429e-01 -4.83941644e-01 -1.56113237e-01 5.37059665e-01
-2.59744704e-01 1.83304980e-01 3.28527629e-01 -4.86964762e-01
-8.16596866e-01 -3.26380312e-01 -3.83893132e-01 -6.22896731e-01
-5.17379381e-02 4.74223435e-01 4.07412387e-02 4.06273365e-01
6.55235291e-01 5.29149473e-01 5.26398607e-02 -4.83267993e-01
8.44286799e-01 1.20103991e+00 5.23803771e-01 -6.72401309e-01
4.65997487e-01 6.94672987e-02 -6.61325157e-01 -6.77507281e-01
-7.71271884e-01 -4.87824887e-01 -3.25522542e-01 -8.56510773e-02
8.24611545e-01 -1.38348484e+00 -2.43387997e-01 6.39157593e-02
-1.38512146e+00 -4.43179190e-01 -1.42540723e-01 6.53137922e-01
-8.59969914e-01 4.03292060e-01 -6.37553155e-01 -5.75665534e-01
-6.97141767e-01 -1.07873011e+00 1.51109517e+00 -1.46757767e-01
-4.34991241e-01 -1.01268566e+00 2.55744874e-01 7.49577701e-01
5.81916451e-01 -5.15457451e-01 5.50235271e-01 -8.54032278e-01
-8.89389813e-02 9.02318060e-02 -2.48038724e-01 4.03711498e-01
3.88833970e-01 -2.20656037e-01 -1.18449557e+00 -6.32403553e-01
-3.80072415e-01 -7.85033226e-01 8.58926535e-01 1.59889027e-01
8.62903237e-01 -4.81174976e-01 2.74841134e-02 2.65092909e-01
9.03481662e-01 -9.50029641e-02 5.15440702e-01 -8.81265923e-02
4.12425280e-01 5.14105737e-01 5.30562401e-01 2.02923194e-01
5.09775579e-01 1.00398874e+00 5.19742854e-02 4.74125780e-02
-7.07325220e-01 -2.32514367e-01 9.38274145e-01 1.59189975e+00
2.53218085e-01 -6.13833785e-01 -9.04615998e-01 7.00045586e-01
-1.78091133e+00 -9.87916946e-01 2.95496643e-01 1.99353480e+00
1.14083147e+00 -4.15623903e-01 2.80633569e-02 -3.89225036e-01
6.21614695e-01 3.86495084e-01 -2.30398387e-01 -5.79686761e-01
-2.01632231e-01 1.86203435e-01 1.74985871e-01 7.16986001e-01
-9.38461125e-01 1.27732837e+00 6.78714466e+00 9.71138477e-01
-1.16991735e+00 6.19749129e-01 4.09914076e-01 -1.94047555e-01
-6.16926491e-01 -3.46566647e-01 -6.76639140e-01 1.12875506e-01
1.52983582e+00 -4.99419510e-01 5.95555842e-01 6.78924024e-01
1.81276083e-01 5.14491916e-01 -1.22705948e+00 1.32533097e+00
5.49271405e-01 -1.33005941e+00 3.93148929e-01 -4.68847901e-01
7.55723715e-01 6.50986016e-01 2.33697534e-01 5.40656745e-01
3.00305486e-01 -1.00035870e+00 7.65346408e-01 3.23760331e-01
1.22548950e+00 -5.33241630e-01 6.00229263e-01 8.06966275e-02
-1.14882684e+00 3.72559965e-01 -2.81642377e-01 4.06321913e-01
1.77607313e-01 -1.09923914e-01 -1.03561199e+00 6.74422920e-01
3.95420879e-01 6.62671149e-01 -3.82438987e-01 3.98118347e-01
-1.95752174e-01 5.47433257e-01 -6.72971606e-02 -9.07041505e-02
3.24776590e-01 2.18161002e-01 4.57048774e-01 1.54955137e+00
4.32798684e-01 -2.37856224e-01 -4.00667787e-02 7.39877045e-01
-6.00126386e-01 4.05372232e-01 -7.80896664e-01 -4.73090649e-01
6.97801590e-01 7.64108539e-01 5.09856641e-02 -5.45853674e-01
-6.30527675e-01 1.45224428e+00 3.93168509e-01 5.54868639e-01
-8.32384109e-01 -1.49921015e-01 9.33173299e-01 -4.13348734e-01
1.43539757e-01 -3.24629307e-01 9.15023088e-02 -1.35559344e+00
9.94767696e-02 -1.18583393e+00 2.84215748e-01 -1.03043771e+00
-1.34865689e+00 8.37066770e-01 -1.72788173e-01 -1.41011655e+00
-6.25234485e-01 -3.00050139e-01 -2.17680991e-01 9.77073669e-01
-1.52920377e+00 -1.37515974e+00 8.15637633e-02 8.59651446e-01
1.25239563e+00 -4.73042816e-01 1.28282726e+00 3.32290024e-01
-3.09482425e-01 1.20485795e+00 5.42124510e-01 3.32318008e-01
1.12495327e+00 -7.59850025e-01 7.96586990e-01 6.90703154e-01
5.22515297e-01 6.02072716e-01 5.26263118e-01 -2.98486263e-01
-1.50087857e+00 -1.09817243e+00 1.26242685e+00 -6.11985922e-01
7.33947635e-01 -2.84302384e-01 -8.94670725e-01 7.62097716e-01
8.29028368e-01 -5.40535897e-02 7.30307519e-01 8.17310065e-03
-8.65519583e-01 4.26197797e-03 -8.87784958e-01 8.24901223e-01
8.19045484e-01 -1.36552405e+00 -7.40284562e-01 6.54746056e-01
1.22695768e+00 -4.52953488e-01 -9.07194912e-01 2.50026613e-01
4.93150681e-01 -3.87602061e-01 9.63867962e-01 -8.37679803e-01
3.25457573e-01 2.96643097e-02 -8.93694460e-01 -1.39780247e+00
4.23865430e-02 -6.76371515e-01 -1.68788344e-01 1.26035285e+00
6.97921276e-01 -5.21227598e-01 1.41172662e-01 2.34517232e-01
-4.73819882e-01 -6.29705548e-01 -1.31019890e+00 -9.84018803e-01
4.68377858e-01 -3.56365383e-01 5.84275305e-01 1.21019626e+00
1.69616744e-01 7.23786414e-01 -3.18443418e-01 1.56677917e-01
2.00584814e-01 -3.08101554e-03 6.22217894e-01 -6.54758096e-01
-1.81020677e-01 -4.93208438e-01 -7.88079724e-02 -1.46315348e+00
7.13647246e-01 -1.34729970e+00 4.74706478e-02 -1.57723033e+00
2.93482482e-01 1.33991195e-02 -1.16259634e-01 4.98237610e-01
-1.11296386e-01 1.50672406e-01 1.72671959e-01 3.92781198e-01
-7.14155018e-01 1.06928051e+00 1.17602563e+00 -5.03880799e-01
1.50457799e-01 -1.99979648e-01 -6.19548202e-01 2.14100257e-01
1.09993410e+00 -2.80502886e-01 -3.73217463e-01 -9.46822286e-01
-1.35157034e-01 2.61128038e-01 -2.38628332e-02 -6.16412342e-01
3.29786167e-02 -1.53877782e-02 -2.20083401e-01 -5.21844745e-01
7.63841987e-01 -5.07388353e-01 -3.04241568e-01 -2.48481780e-02
-9.16975141e-01 1.79974079e-01 3.42660755e-01 2.00870022e-01
-5.41526973e-01 3.62817757e-02 7.44040012e-01 7.08673000e-02
-4.89592671e-01 1.03343658e-01 -4.55739707e-01 4.01898235e-01
4.20564264e-01 4.48838174e-01 -6.74199164e-01 -9.77020741e-01
-7.09605336e-01 4.01013978e-02 2.78108388e-01 9.97894943e-01
8.77842665e-01 -1.72428083e+00 -1.13546169e+00 6.47285534e-03
5.48941433e-01 -6.40234888e-01 -2.16914155e-02 8.59497428e-01
-6.06686175e-02 7.54201889e-01 1.05238266e-01 -6.10285759e-01
-1.69652903e+00 5.50382398e-02 4.53946233e-01 1.29726276e-01
-2.48298734e-01 9.78137732e-01 -2.11037677e-02 -9.11613584e-01
3.54106724e-01 -3.19223225e-01 5.39424568e-02 -1.52837604e-01
7.68020749e-01 2.22863823e-01 1.76225901e-01 -1.19242203e+00
-4.99051094e-01 4.51261729e-01 -2.00919941e-01 -5.45666337e-01
1.01741922e+00 -6.64051592e-01 -3.38922143e-02 7.23235250e-01
1.58529592e+00 -3.15300599e-02 -5.55931926e-01 -6.51809335e-01
-2.76387602e-01 -2.95984089e-01 7.27199391e-02 -8.88155460e-01
-8.85635018e-01 9.70880568e-01 7.34081328e-01 -2.77601033e-01
8.96259069e-01 5.92544019e-01 1.00671315e+00 7.81327069e-01
2.00140730e-01 -8.98178399e-01 2.32472688e-01 9.57715750e-01
1.42405283e+00 -1.41418517e+00 -2.38623559e-01 -6.05453812e-02
-9.54739153e-01 8.63524437e-01 2.51906276e-01 3.89331758e-01
1.60786435e-01 -8.06881348e-04 5.18950999e-01 1.28153980e-01
-1.04522526e+00 -3.42233509e-01 6.25647366e-01 5.76703668e-01
1.04639935e+00 2.39601046e-01 -1.31609812e-02 2.06355199e-01
-2.23204717e-01 -4.05426413e-01 1.50713563e-01 6.78079128e-01
-3.43176365e-01 -1.12436068e+00 -1.83972687e-01 -1.43458173e-01
-4.54455346e-01 -4.80338901e-01 -8.38028312e-01 3.31886560e-01
-7.52928555e-01 1.27922904e+00 5.86312674e-02 -4.27074671e-01
3.20337325e-01 3.97834301e-01 4.26454157e-01 -6.02968693e-01
-4.29664999e-01 1.77603379e-01 5.26715994e-01 -5.72436571e-01
-4.95150715e-01 -6.11389399e-01 -1.25728500e+00 -4.18851495e-01
-2.26849914e-01 2.45152831e-01 8.17138314e-01 6.59136295e-01
5.74890137e-01 4.43581253e-01 7.36440003e-01 -6.99209154e-01
-6.73351824e-01 -1.29512644e+00 -1.13350749e-01 3.35697979e-01
6.20448053e-01 -4.04263020e-01 -5.40052891e-01 1.98864355e-03] | [14.494524002075195, 7.197785377502441] |
b89b0513-f785-4c05-b2e8-4416d4f74a86 | mem_ge-a-new-maximum-entropy-method-for-image | 2002.07921 | null | https://arxiv.org/abs/2002.07921v1 | https://arxiv.org/pdf/2002.07921v1.pdf | MEM_GE: a new maximum entropy method for image reconstruction from solar X-ray visibilities | Maximum Entropy is an image reconstruction method conceived to image a sparsely occupied field of view and therefore particularly appropriate to achieve super-resolution effects. Although widely used in image deconvolution, this method has been formulated in radio astronomy for the analysis of observations in the spatial frequency domain, and an Interactive Data Language (IDL) code has been implemented for image reconstruction from solar X-ray Fourier data. However, this code relies on a non-convex formulation of the constrained optimization problem addressed by the Maximum Entropy approach and this sometimes results in unreliable reconstructions characterized by unphysical shrinking effects. This paper introduces a new approach to Maximum Entropy based on the constrained minimization of a convex functional. In the case of observations recorded by the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI), the resulting code provides the same super-resolution effects of the previous algorithm, while working properly also when that code produces unphysical reconstructions. Results are also provided of testing the algorithm with synthetic data simulating observations of the Spectrometer/Telescope for Imaging X-rays (STIX) in Solar Orbiter. The new code is available in the {\em{HESSI}} folder of the Solar SoftWare (SSW)tree. | ['Federico Benvenuto', 'Michele Piana', 'Brian R Dennis', 'Richard Schwartz', 'Paolo Massa', 'Anna Maria Massone', 'A Kim Tolbert'] | 2020-02-18 | null | null | null | null | ['image-deconvolution'] | ['computer-vision'] | [ 3.57159525e-01 2.64270365e-01 3.09273839e-01 -3.22648972e-01
-4.03903753e-01 -2.96768427e-01 6.91935062e-01 -5.03485739e-01
-4.34190631e-01 8.56814981e-01 -8.51479098e-02 -2.44921908e-01
-4.07696009e-01 -6.47769272e-01 -3.62705261e-01 -9.58122611e-01
1.74567953e-01 5.47454357e-01 -9.01980028e-02 -2.03987807e-01
2.97117084e-01 6.62191808e-01 -1.77290595e+00 7.44770421e-03
8.74230266e-01 8.11247528e-01 7.25590527e-01 5.56692481e-01
1.29227161e-01 6.94124281e-01 -2.48410061e-01 2.05649376e-01
3.74630153e-01 -9.32876408e-01 -6.55880868e-01 4.39519882e-01
2.68961281e-01 -6.40201718e-02 -6.40352890e-02 1.31964779e+00
3.26392651e-01 1.58125579e-01 7.82708943e-01 -7.05539823e-01
-3.46013397e-01 8.99784714e-02 -3.91410351e-01 4.49584424e-01
3.79639119e-01 -1.95919737e-01 5.45399487e-01 -7.16101527e-01
6.45421386e-01 6.81180775e-01 5.38682044e-01 3.16662341e-02
-1.35441375e+00 2.92226695e-03 -8.97833049e-01 2.84314305e-01
-1.24913597e+00 -4.61237669e-01 4.77041125e-01 -4.73327458e-01
9.67659652e-01 7.23491788e-01 5.75755298e-01 6.65888250e-01
-5.31557277e-02 9.84741002e-02 1.59761631e+00 -5.12009084e-01
2.78372943e-01 4.54561621e-01 7.65792951e-02 4.17074114e-01
2.86010116e-01 7.06880033e-01 -4.62630540e-01 -1.24057442e-01
7.58961320e-01 -2.62177080e-01 -7.50220180e-01 -2.98205819e-02
-1.21946871e+00 1.02336955e+00 4.24052387e-01 6.43895328e-01
-6.83997512e-01 -4.77770180e-01 -1.47743300e-02 5.01422286e-01
5.22860527e-01 5.79907358e-01 -1.75781429e-01 1.09015234e-01
-1.25273323e+00 2.25287184e-01 9.08854723e-01 3.32412392e-01
8.17346632e-01 3.15640360e-01 4.38914180e-01 4.82675225e-01
4.18450147e-01 7.84718633e-01 5.44122100e-01 -1.02836919e+00
-2.02601507e-01 2.66532063e-01 1.95368335e-01 -6.27719760e-01
-5.46977401e-01 -4.59282011e-01 -8.39899480e-01 9.34649587e-01
1.68104932e-01 5.77404797e-02 -7.41577327e-01 1.12686944e+00
3.51922333e-01 -2.57680118e-01 4.56922948e-01 1.31384826e+00
7.45135844e-01 7.58391619e-01 -5.41647971e-01 -8.67645979e-01
1.16751683e+00 -4.24632400e-01 -8.97728801e-01 -8.10999647e-02
3.52711886e-01 -8.93470049e-01 6.11153841e-01 5.66280425e-01
-1.10406101e+00 -4.83868718e-01 -1.19877148e+00 1.71674043e-01
-2.10516557e-01 5.48467925e-03 2.44677544e-01 5.90358794e-01
-1.04136527e+00 7.49331594e-01 -6.00794017e-01 -4.26514179e-01
-1.50483996e-01 1.47303818e-02 -2.82375991e-01 5.31070173e-01
-7.88293898e-01 1.24129498e+00 6.55562997e-01 -3.10628433e-02
-3.16090286e-01 -5.20739734e-01 -6.85834408e-01 1.36535868e-01
8.58929306e-02 -5.33116400e-01 9.07840014e-01 -1.34106994e+00
-1.55511153e+00 1.05000460e+00 9.06861797e-02 -5.99427521e-01
5.34047842e-01 3.01031858e-01 -4.07301605e-01 5.02248764e-01
-4.16452587e-02 1.90331742e-01 8.90066206e-01 -1.25246334e+00
-1.08885430e-01 -5.47603011e-01 -4.19353515e-01 2.25294650e-01
2.58720130e-01 -1.55908819e-02 2.42368907e-01 -4.05592740e-01
4.49841022e-01 -8.84070754e-01 -2.69793905e-02 -1.52724385e-01
-8.79136845e-02 2.96299458e-01 7.83763945e-01 -9.73789871e-01
7.57725894e-01 -2.12407327e+00 4.39847887e-01 2.45953277e-01
-2.31347859e-01 1.79254174e-01 5.76832414e-01 4.12182510e-01
-5.29019594e-01 -4.02008265e-01 -9.02731597e-01 -1.75575912e-01
-2.22494990e-01 2.71613002e-01 -2.38745466e-01 8.19427788e-01
-5.21563053e-01 3.73506159e-01 -3.72177571e-01 -4.14926767e-01
5.48543453e-01 5.58163404e-01 -2.43622229e-01 3.16000096e-02
-1.96642131e-01 9.02707994e-01 -2.99068689e-01 3.54196668e-01
8.95767391e-01 -3.82334292e-01 -4.78527043e-03 -2.30074227e-01
-8.86770546e-01 -1.76241457e-01 -1.27855647e+00 1.58591723e+00
-1.98121682e-01 5.50736725e-01 3.82711530e-01 -1.13383925e+00
8.98945868e-01 3.90518129e-01 6.72617555e-01 -9.58318353e-01
1.47933349e-01 2.50493824e-01 -2.27539554e-01 -9.45955694e-01
5.38882136e-01 -5.98577619e-01 6.05671108e-01 3.81243676e-01
-1.20473631e-01 -4.28292722e-01 1.19245484e-01 7.29733938e-03
7.76082039e-01 3.69127959e-01 5.99912405e-01 -7.31411934e-01
6.05493009e-01 4.12015140e-01 -6.16111327e-03 4.35999632e-01
2.67541409e-01 8.97408485e-01 6.47551119e-02 -4.19586301e-01
-1.53324401e+00 -7.90311098e-01 -1.05377316e+00 2.62668401e-01
4.60083857e-02 -7.29515031e-02 -5.70393384e-01 6.43118471e-02
-2.90539235e-01 9.42764103e-01 -5.20140231e-01 2.30731577e-01
2.70991735e-02 -1.17212999e+00 4.61740084e-02 -3.26879561e-01
7.75115490e-01 -1.29067194e+00 -1.12116003e+00 -4.86792587e-02
-2.77444601e-01 -9.16007578e-01 2.83314437e-01 9.41278487e-02
-8.63867164e-01 -1.12408626e+00 -6.91538692e-01 -1.10338584e-01
4.43021506e-01 1.14654459e-01 1.06259847e+00 6.14154013e-03
-5.78334093e-01 4.40380067e-01 -5.72174430e-01 -1.52742177e-01
-6.24289215e-01 -4.21059668e-01 -6.55156597e-02 2.30590459e-02
1.15018874e-01 -5.74410617e-01 -3.74985039e-01 1.42769396e-01
-1.21736598e+00 6.29897043e-02 3.88950258e-01 7.35239923e-01
4.97929484e-01 9.38167647e-02 3.70517612e-01 -7.09904373e-01
1.56085610e-01 -3.84004533e-01 -1.17992544e+00 -1.61157653e-01
-8.02746356e-01 1.61537498e-01 4.59318906e-01 2.17737675e-01
-1.39627838e+00 2.35734910e-01 -4.27373111e-01 -3.70835274e-01
-1.87769502e-01 5.29996693e-01 1.64250940e-01 -5.09845376e-01
1.09732616e+00 5.48615217e-01 3.48817289e-01 -5.65692127e-01
-5.61962016e-02 7.74089098e-01 6.70351326e-01 1.38872832e-01
6.46882117e-01 7.47090459e-01 2.04677314e-01 -1.45760918e+00
-8.21812809e-01 -5.15149593e-01 -4.29841876e-01 -3.61498654e-01
1.24444604e+00 -7.82248259e-01 -6.13919616e-01 2.39331543e-01
-7.42543817e-01 -2.93839071e-02 -4.54354912e-01 9.20470119e-01
-9.01650012e-01 8.00853610e-01 1.32711576e-02 -9.70651329e-01
-2.09592134e-01 -9.70560908e-01 8.02084863e-01 3.53894055e-01
1.77738115e-01 -9.70809162e-01 4.45253819e-01 5.19798636e-01
3.03932905e-01 4.70837295e-01 4.67203259e-01 -4.26808465e-03
-6.22871757e-01 1.62275210e-01 -1.53531432e-01 6.19317532e-01
-3.31599116e-01 -3.03802043e-01 -1.08340299e+00 -3.81206870e-01
9.76646602e-01 -1.22478113e-01 8.88789117e-01 5.70838034e-01
7.28480220e-01 -6.19379953e-02 6.17365679e-03 1.03128684e+00
2.02642703e+00 -1.36196101e-02 9.79458511e-01 5.69304049e-01
2.31695864e-02 4.29547340e-01 5.78094125e-01 7.60438144e-01
-2.10415080e-01 7.29775429e-01 8.06857944e-01 -1.09147675e-01
2.01534167e-01 7.12894976e-01 2.42196605e-01 2.77854681e-01
-5.20830035e-01 9.42571685e-02 -4.68975663e-01 3.56118917e-01
-1.58210909e+00 -1.47262514e+00 -8.11776757e-01 2.63436246e+00
5.24527252e-01 -3.14128488e-01 -1.47207126e-01 2.44399786e-01
4.20298249e-01 2.43066251e-01 -2.74423033e-01 -3.24771911e-01
-3.11181217e-01 1.94481075e-01 6.45617664e-01 9.71642733e-01
-7.48373330e-01 1.06063724e-01 6.32738352e+00 2.51684666e-01
-1.04167688e+00 2.67952919e-01 -7.27696419e-02 -2.69406550e-02
-2.95859098e-01 3.11850607e-01 -3.65326107e-01 7.75951147e-01
9.11027312e-01 -1.84637770e-01 6.93553627e-01 6.67619705e-01
3.90040606e-01 -7.79383123e-01 -3.76564801e-01 1.20388007e+00
2.73761034e-01 -1.29456890e+00 -5.49754024e-01 4.57437366e-01
5.27146578e-01 3.14977616e-01 -2.89879829e-01 -2.39401370e-01
-4.79984879e-01 -8.18826020e-01 4.86291230e-01 8.91244113e-01
7.44917154e-01 -7.17963338e-01 4.49603289e-01 7.10021734e-01
-5.37496448e-01 1.08633647e-02 -6.52517259e-01 -2.79493891e-02
5.18317878e-01 9.86007869e-01 -6.41655445e-01 8.55614662e-01
9.01535392e-01 3.73107612e-01 -2.97909319e-01 8.84402096e-01
1.83720887e-01 2.71233261e-01 -6.72481418e-01 4.39697891e-01
1.02924518e-01 -1.09944773e+00 1.03287256e+00 7.04968035e-01
6.82857811e-01 4.34015125e-01 -2.88395137e-01 1.02163208e+00
5.65962195e-01 1.05441242e-01 -7.38534987e-01 1.35095999e-01
-2.58352280e-01 1.33462000e+00 -5.54931939e-01 -3.71651918e-01
-5.15280187e-01 1.04566860e+00 -1.56472102e-01 2.10969344e-01
-7.52756596e-01 1.18618369e-01 1.89678431e-01 4.16440666e-01
3.74802262e-01 -1.89262122e-01 -3.11877638e-01 -1.12220550e+00
-1.15660377e-01 -5.80756068e-01 3.16796422e-01 -1.50020361e+00
-7.26752043e-01 7.97247291e-01 3.73152375e-01 -1.17087007e+00
-3.61280322e-01 -5.85280299e-01 -4.56700116e-01 1.30873406e+00
-1.41759491e+00 -7.64895976e-01 -5.61073363e-01 5.33682287e-01
2.38370925e-01 -2.01140970e-01 9.49962497e-01 2.49709740e-01
-2.40252428e-02 -4.30846065e-01 6.28324628e-01 -7.52193928e-01
2.92971015e-01 -1.41054904e+00 -3.88714939e-01 8.11921656e-01
-7.37884119e-02 -1.04296729e-01 1.43928266e+00 -7.01148391e-01
-1.19612157e+00 -5.76148331e-01 8.74600112e-01 -6.71056136e-02
3.74451816e-01 3.75924796e-01 -1.10485148e+00 7.98317730e-01
6.39924109e-01 -2.63099074e-01 4.52308327e-01 -5.04102349e-01
3.30095947e-01 3.00961703e-01 -1.42966926e+00 -1.53098404e-01
5.02154469e-01 -3.39170933e-01 -8.49853396e-01 8.59898984e-01
-3.07694469e-02 -3.18666220e-01 -1.07330263e+00 4.26360905e-01
5.45423962e-02 -1.88203585e+00 1.02908111e+00 1.90397799e-01
1.81903392e-01 -4.74955648e-01 -8.89481008e-02 -1.10685456e+00
-1.63081840e-01 -5.97876906e-01 2.41428956e-01 5.89542389e-01
1.94563046e-01 -8.87548923e-01 4.28004593e-01 -1.60698652e-01
-2.49959722e-01 -2.70355716e-02 -1.13511586e+00 -6.24908805e-01
-4.99175251e-01 5.09646833e-02 -6.90406337e-02 1.03745449e+00
-1.95046559e-01 1.96246743e-01 -3.91722590e-01 5.75573027e-01
1.23847353e+00 4.22679633e-01 4.15474206e-01 -1.35260308e+00
-7.15143740e-01 -2.87379995e-02 -2.40000769e-01 -3.88069689e-01
-8.23534802e-02 -1.03801847e+00 -5.52644879e-02 -1.28453350e+00
3.74028645e-02 -4.15352136e-02 4.26561952e-01 -7.83394501e-02
3.01083595e-01 3.95368576e-01 1.75959870e-01 6.13113940e-01
2.82109499e-01 4.49068338e-01 1.07018495e+00 3.98025453e-01
3.57614011e-02 1.33491859e-01 -1.14998035e-01 7.02751935e-01
6.41825318e-01 -4.76416528e-01 -3.38392675e-01 -9.14642811e-02
5.06088853e-01 3.64584088e-01 6.98592901e-01 -1.22081387e+00
1.55208722e-01 1.59699664e-01 3.48416507e-01 -8.96599531e-01
6.30677819e-01 -1.09249330e+00 1.01626050e+00 4.71567720e-01
2.08837718e-01 -1.93916544e-01 -1.62965462e-01 1.03694111e-01
-3.17873359e-01 -9.34405267e-01 1.56526816e+00 -6.83831215e-01
-6.12339556e-01 -1.47706971e-01 -2.93280840e-01 -4.48389202e-01
1.00860405e+00 -2.84659684e-01 -1.42094016e-01 -3.10341954e-01
-1.03068626e+00 -3.10368150e-01 1.05804586e+00 -2.84054071e-01
3.26746523e-01 -1.00128651e+00 -8.98533821e-01 5.45443118e-01
-4.74584624e-02 -1.30588844e-01 5.11229217e-01 1.08390212e+00
-1.07936490e+00 2.35517561e-01 -4.11532670e-01 -7.59885073e-01
-1.18213034e+00 6.83954656e-01 7.37683475e-01 3.75290662e-02
-9.88145888e-01 3.69388103e-01 7.30368644e-02 -2.41490752e-01
-4.03743386e-01 9.07156169e-02 -3.60837191e-01 5.47741652e-02
6.32732213e-01 3.61941755e-01 2.21242070e-01 -9.30396080e-01
-9.00200009e-03 5.72321177e-01 5.70392191e-01 -2.30457008e-01
1.45847213e+00 -4.00302708e-01 -3.20159256e-01 2.52532482e-01
1.08802652e+00 -6.63596913e-02 -1.07182467e+00 4.29265276e-02
-1.64268300e-01 -3.67664367e-01 6.29306078e-01 -7.58150816e-01
-8.35338831e-01 4.23084944e-01 8.34126174e-01 8.46552551e-01
1.31063712e+00 -2.11744271e-02 9.78111848e-02 3.42980236e-01
6.18629754e-02 -1.02354777e+00 -5.75006783e-01 2.18206927e-01
1.21348488e+00 -1.38464701e+00 4.52889889e-01 -2.05431089e-01
-5.87457180e-01 1.23355758e+00 -6.94247112e-02 -1.66638821e-01
7.48080730e-01 4.12686378e-01 9.30553079e-02 -5.35040498e-01
-2.22011432e-01 -4.39909995e-01 -1.26959220e-01 3.42619300e-01
1.21489726e-01 -1.36152819e-01 -7.15708613e-01 -2.55958915e-01
-2.93776929e-01 8.89282078e-02 7.63862908e-01 8.44065428e-01
-8.25246513e-01 -7.01920986e-01 -1.06358302e+00 2.73164064e-01
-2.90362358e-01 -1.05543636e-01 -1.51088327e-01 7.88555264e-01
-2.78075114e-02 5.58401763e-01 4.07917574e-02 3.64219517e-01
8.88191611e-02 1.14315361e-01 6.77468717e-01 -1.51230246e-01
-3.23287785e-01 1.66857511e-01 9.27682295e-02 -4.01625931e-01
-9.39394951e-01 -9.34992135e-01 -1.24899602e+00 -2.16673225e-01
-2.63455480e-01 4.14057732e-01 1.25094795e+00 9.16444778e-01
-1.33106992e-01 4.34184134e-01 7.41362512e-01 -1.16859567e+00
-5.33126533e-01 -1.16226554e+00 -1.17229927e+00 1.92337781e-01
3.69334400e-01 -4.90796626e-01 -9.40873384e-01 3.33962739e-01] | [11.526118278503418, -2.595055103302002] |
e17d3b7e-8e42-4cd4-8cf7-fe630c527105 | exploiting-timegraphs-in-temporal-relation | null | null | https://aclanthology.org/W14-3702 | https://aclanthology.org/W14-3702.pdf | Exploiting Timegraphs in Temporal Relation Classification | null | ['Natsuda Laokulrat', 'Makoto Miwa', 'Yoshimasa Tsuruoka'] | 2014-10-01 | null | null | null | ws-2014-10 | ['temporal-relation-classification'] | ['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] | [-7.481617450714111, 3.5874524116516113] |
e63baa1e-8446-429e-85f4-ff855c49f699 | fine-grained-causality-extraction-from | 2107.09980 | null | https://arxiv.org/abs/2107.09980v2 | https://arxiv.org/pdf/2107.09980v2.pdf | Fine-Grained Causality Extraction From Natural Language Requirements Using Recursive Neural Tensor Networks | [Context:] Causal relations (e.g., If A, then B) are prevalent in functional requirements. For various applications of AI4RE, e.g., the automatic derivation of suitable test cases from requirements, automatically extracting such causal statements are a basic necessity. [Problem:] We lack an approach that is able to extract causal relations from natural language requirements in fine-grained form. Specifically, existing approaches do not consider the combinatorics between causes and effects. They also do not allow to split causes and effects into more granular text fragments (e.g., variable and condition), making the extracted relations unsuitable for automatic test case derivation. [Objective & Contributions:] We address this research gap and make the following contributions: First, we present the Causality Treebank, which is the first corpus of fully labeled binary parse trees representing the composition of 1,571 causal requirements. Second, we propose a fine-grained causality extractor based on Recursive Neural Tensor Networks. Our approach is capable of recovering the composition of causal statements written in natural language and achieves a F1 score of 74 % in the evaluation on the Causality Treebank. Third, we disclose our open data sets as well as our code to foster the discourse on the automatic extraction of causality in the RE community. | ['Daniel Mendez', 'Andreas Vogelsang', 'Henning Femmer', 'Julian Frattini', 'Tobias Springer', 'Jannik Fischbach'] | 2021-07-21 | null | null | null | null | ['tensor-networks'] | ['methodology'] | [ 2.52947420e-01 3.46777171e-01 -4.34245169e-01 -5.43052435e-01
-2.31240064e-01 -6.76753461e-01 3.46517146e-01 2.64700413e-01
1.40493542e-01 8.34038675e-01 6.72007322e-01 -8.42813194e-01
-5.31470656e-01 -9.26804125e-01 -5.22508919e-01 -7.03083053e-02
-2.55799979e-01 3.91186744e-01 2.94113457e-01 -3.91664177e-01
3.06401491e-01 3.28297853e-01 -1.34109712e+00 4.88617063e-01
8.65414917e-01 8.00705016e-01 -1.12322330e-01 5.68607509e-01
-2.56389558e-01 1.38033938e+00 -5.78327239e-01 -5.35996020e-01
-2.87183762e-01 -6.10983789e-01 -1.36957359e+00 -2.68845648e-01
9.56707867e-04 -3.20817322e-01 4.92551178e-02 8.86548638e-01
4.38283160e-02 -3.08674753e-01 6.13811553e-01 -1.50816071e+00
-5.69357157e-01 1.58388424e+00 -3.68647993e-01 1.62458047e-01
7.88177490e-01 -8.03793594e-02 1.66912591e+00 -5.68020105e-01
6.39319837e-01 1.16615641e+00 2.73430347e-01 4.92494911e-01
-1.22818029e+00 -5.98305643e-01 1.53894737e-01 5.16418219e-01
-1.03763294e+00 -3.72827470e-01 9.40235853e-01 -8.57473910e-01
1.45654225e+00 4.47194308e-01 3.55178326e-01 1.12216449e+00
6.38392791e-02 7.14013815e-01 8.72711778e-01 -6.49609506e-01
9.70336497e-02 -3.48458111e-01 5.91899931e-01 6.20880663e-01
3.39460135e-01 4.62046936e-02 -5.85842371e-01 -3.88013244e-01
6.28443062e-01 -7.00424075e-01 -1.09843962e-01 9.90630942e-04
-1.44629526e+00 8.40216100e-01 -2.99754888e-02 6.52780473e-01
-5.22918224e-01 4.34124589e-01 3.83647859e-01 2.82402188e-01
3.26688103e-02 6.11094177e-01 -7.89268970e-01 -2.30223626e-01
-4.57229316e-01 3.75903845e-01 1.08409142e+00 9.98054922e-01
4.63170856e-01 7.28553087e-02 -5.05419038e-02 4.58512425e-01
4.92380917e-01 2.82852471e-01 3.06911618e-02 -8.53469312e-01
4.63919461e-01 1.00838935e+00 1.32712394e-01 -1.17144978e+00
-4.17748094e-01 5.46220839e-02 -4.42574501e-01 -1.81228653e-01
4.32455271e-01 -4.76354569e-01 -3.47393751e-01 1.82107019e+00
2.11172774e-01 -1.66908756e-01 -3.59747596e-02 6.78362548e-01
8.44070971e-01 2.65548021e-01 1.07238479e-01 -4.87758368e-01
1.37120843e+00 -2.55420953e-01 -9.31640983e-01 -8.33526924e-02
6.82123005e-01 -7.00779974e-01 8.27298462e-01 5.42334497e-01
-1.08650088e+00 1.14778638e-01 -1.02477121e+00 3.72161299e-01
5.40173166e-02 -1.90913588e-01 1.10671556e+00 4.96770769e-01
-6.67550147e-01 4.50089574e-01 -6.90949678e-01 -3.63711417e-02
-7.99906626e-02 2.47918323e-01 -3.37474406e-01 7.19501302e-02
-1.49675274e+00 8.34220171e-01 6.00357831e-01 1.29876107e-01
-5.89470983e-01 -6.80964649e-01 -8.64300907e-01 -1.04888774e-01
6.96964264e-01 -3.01867753e-01 1.35021853e+00 -7.30333567e-01
-1.01952112e+00 3.24617207e-01 -1.52297050e-01 -1.68698490e-01
-6.36276677e-02 -2.71925926e-01 -7.42212534e-01 -1.25560641e-01
3.04762691e-01 6.49004281e-02 3.00088376e-01 -8.30926001e-01
-8.01841319e-01 -7.74555430e-02 6.82418764e-01 -5.15964687e-01
-3.22501749e-01 7.60088623e-01 1.15687652e-02 -5.67866027e-01
-9.87733826e-02 -6.73712790e-01 -1.91335827e-01 -8.27525198e-01
-7.00946867e-01 -7.89439797e-01 2.55812466e-01 -5.81844151e-01
1.80826020e+00 -1.79725409e+00 3.33797753e-01 8.99012163e-02
2.49601021e-01 -2.42972031e-01 -4.43537086e-02 5.84994078e-01
-7.42724836e-01 6.52791262e-01 -3.39316905e-01 7.28506863e-01
3.91468436e-01 2.10590124e-01 -4.56246495e-01 2.82918274e-01
7.44872451e-01 7.48005211e-01 -1.06814539e+00 -6.42118812e-01
-8.99933726e-02 1.73843224e-02 -8.08912992e-01 2.16578409e-01
-7.15326130e-01 2.47569874e-01 -6.92698359e-01 5.62391698e-01
1.02710754e-01 -7.48004988e-02 5.09157181e-01 -2.25613743e-01
-4.12859887e-01 1.11777782e+00 -1.31787729e+00 1.31601560e+00
-5.04872978e-01 3.69706064e-01 -4.22143221e-01 -8.08425486e-01
6.75929248e-01 8.34339738e-01 6.91826820e-01 -5.16720414e-01
1.29271716e-01 2.14944780e-01 4.26313132e-01 -8.38061571e-01
1.81127131e-01 -3.00331473e-01 -5.76744974e-01 8.55946898e-01
-1.22848384e-01 -2.29122400e-01 8.39889824e-01 2.07353696e-01
1.55906248e+00 3.63828659e-01 7.93712318e-01 -2.05992296e-01
5.59986770e-01 4.10590500e-01 9.25539672e-01 4.71753702e-02
1.93355322e-01 1.19917646e-01 1.16775274e+00 -6.08729780e-01
-5.95846415e-01 -6.93368435e-01 -8.35542902e-02 1.02717125e+00
-4.41044807e-01 -8.50433528e-01 -2.28741199e-01 -1.03687990e+00
-1.92374319e-01 1.32554746e+00 -5.47430098e-01 1.97246104e-01
-1.00567305e+00 -7.58220613e-01 8.01728487e-01 6.10757589e-01
-9.35309529e-02 -1.28028011e+00 -8.34489942e-01 5.42746603e-01
-6.83311701e-01 -1.20773053e+00 4.68714163e-02 2.85928279e-01
-5.27832389e-01 -1.55584121e+00 3.88799250e-01 -2.07721710e-01
3.24870229e-01 -4.59650248e-01 1.50257015e+00 1.06054358e-01
-2.72820555e-02 -5.77269122e-02 -7.82725036e-01 -3.84502918e-01
-6.25075638e-01 -2.21213132e-01 -1.48889795e-01 -3.62003326e-01
6.75597727e-01 -7.24146426e-01 1.43935785e-01 4.23928946e-01
-8.22320342e-01 -8.90347809e-02 4.29611295e-01 5.40491164e-01
-2.95506278e-03 3.17457557e-01 4.98653889e-01 -7.84924805e-01
8.60232055e-01 -4.82962787e-01 -6.36107028e-01 1.93993151e-01
-6.69505239e-01 3.81565601e-01 6.08913064e-01 -1.78211913e-01
-1.09557164e+00 -1.36944979e-01 -2.67384410e-01 3.82634133e-01
-3.52920771e-01 1.07269216e+00 -1.41234100e-01 5.79583406e-01
8.97890627e-01 -5.00522137e-01 -8.13743830e-01 -1.52829260e-01
3.80367517e-01 4.66355860e-01 3.05800140e-01 -1.12822914e+00
7.65486836e-01 -9.72728282e-02 3.40674771e-03 -3.00169915e-01
-6.91755772e-01 -3.64128768e-01 -6.50496602e-01 -1.44191146e-01
8.76302242e-01 -4.07955527e-01 -7.57160008e-01 -2.67005503e-01
-1.65838540e+00 -3.81134957e-01 -2.20665425e-01 5.97539425e-01
-4.43746716e-01 2.40467727e-01 -6.53686523e-01 -7.54732788e-01
-2.16062944e-02 -9.97876227e-01 6.35376036e-01 -3.05137783e-01
-1.00568008e+00 -7.52379537e-01 1.82275787e-01 2.04743475e-01
3.50241736e-02 4.62350637e-01 1.48798048e+00 -4.66731369e-01
-2.38723576e-01 -7.24138469e-02 -1.45480335e-01 8.57586265e-02
2.90429264e-01 5.10029197e-01 -4.70351249e-01 3.95560265e-01
-1.55390784e-01 -3.55022490e-01 1.13823622e-01 -3.38312313e-02
4.88198340e-01 -5.69923460e-01 -1.37879997e-01 -2.36570910e-01
1.08997774e+00 3.71980250e-01 5.72716117e-01 1.93411827e-01
7.54838049e-01 1.08577299e+00 7.14757681e-01 5.38410902e-01
5.17393649e-01 5.30351520e-01 5.37344992e-01 3.76402199e-01
-1.60153762e-01 -3.11562810e-02 3.83978784e-01 1.02457500e+00
-4.94860321e-01 9.21338499e-02 -1.41346323e+00 8.94135535e-01
-1.80549312e+00 -8.52439165e-01 -9.59438682e-01 1.56539750e+00
1.24718940e+00 1.68323278e-01 -7.99628347e-02 5.18391073e-01
5.60139537e-01 -2.12425455e-01 2.90824678e-02 -4.91727561e-01
2.58643568e-01 4.02114034e-01 3.05393398e-01 5.49377203e-01
-8.05828393e-01 8.04465890e-01 5.93131828e+00 1.81759477e-01
-7.30283797e-01 1.06845729e-01 -1.55775368e-01 1.59428328e-01
-8.07452202e-01 4.52536017e-01 -5.49602509e-01 1.37895150e-02
1.16557288e+00 -3.80999118e-01 3.86450469e-01 5.51111460e-01
3.05501461e-01 1.64479777e-01 -1.44383037e+00 4.20024425e-01
-2.33657986e-01 -1.20012581e+00 -1.54859871e-01 -1.78595796e-01
6.24468207e-01 -1.43545076e-01 -6.89724088e-01 9.91340950e-02
1.01587892e+00 -1.03611505e+00 1.08824515e+00 1.73812255e-01
6.06930733e-01 -7.23896623e-01 8.37638795e-01 2.24568054e-01
-1.06758296e+00 -2.34453499e-01 6.00618422e-02 -4.71786350e-01
2.65300006e-01 1.14134347e+00 -1.13800335e+00 9.98960793e-01
6.29080236e-01 5.33420324e-01 -2.94763058e-01 5.76711655e-01
-1.01693273e+00 9.15555656e-01 -5.88021837e-02 -3.00419569e-01
-7.01045915e-02 1.05276376e-01 5.03321350e-01 1.29247057e+00
4.39536497e-02 2.88381308e-01 -5.87981269e-02 1.08280444e+00
2.02607155e-01 9.75235105e-02 -4.89087492e-01 -4.92111504e-01
1.63313553e-01 9.59946752e-01 -4.92357582e-01 -1.05974548e-01
-4.05132920e-01 2.14946762e-01 6.68915659e-02 3.63377124e-01
-9.50952828e-01 -3.92360300e-01 4.52593833e-01 8.95322263e-02
-1.61018129e-02 -2.72428304e-01 -5.42986512e-01 -1.02156985e+00
2.96727240e-01 -1.15184617e+00 5.21261096e-01 -5.87332010e-01
-1.20172060e+00 6.49861336e-01 4.44376469e-01 -8.37420166e-01
-7.92779684e-01 -3.90358120e-01 -3.69898587e-01 8.31767380e-01
-1.21372163e+00 -9.53793645e-01 1.29587337e-01 5.31732321e-01
3.13338578e-01 2.61125714e-01 9.20489192e-01 4.67598945e-01
-5.65502942e-01 6.37984052e-02 -1.27435791e+00 3.11646134e-01
7.34080911e-01 -1.24610209e+00 3.25865149e-01 1.36517537e+00
-8.12050179e-02 1.11742485e+00 9.64212358e-01 -9.16226149e-01
-1.52846694e+00 -8.52028906e-01 1.64383662e+00 -4.52533275e-01
1.33800995e+00 -1.63621679e-01 -5.36748350e-01 9.29174781e-01
3.37572932e-01 -2.75200307e-01 6.74728215e-01 5.45523107e-01
-6.34218156e-01 -3.89822479e-03 -7.34297872e-01 7.33227789e-01
1.05148268e+00 -4.44951981e-01 -1.13369036e+00 2.04762563e-01
1.02058244e+00 1.97992641e-02 -1.11252141e+00 6.34231389e-01
4.26199406e-01 -9.45562720e-01 5.20156264e-01 -7.98173189e-01
9.59192514e-01 -6.45050526e-01 -3.39375883e-01 -1.02247834e+00
-3.76662940e-01 -6.63714767e-01 -1.29509524e-01 1.48680508e+00
9.74958301e-01 -5.75434029e-01 2.78521538e-01 8.62924635e-01
-1.81913972e-01 -2.95448989e-01 -8.21430326e-01 -3.76993477e-01
-1.40855029e-01 -1.29435229e+00 7.83006489e-01 1.37055242e+00
6.57917619e-01 6.80771708e-01 -6.37562051e-02 1.33627430e-01
4.36278462e-01 3.64879727e-01 4.38344806e-01 -1.50922179e+00
-4.26159084e-01 -5.12903988e-01 -9.62937623e-02 -3.54013979e-01
5.04360914e-01 -7.71689892e-01 2.02569798e-01 -1.71065688e+00
-7.17413649e-02 -5.88551521e-01 4.80300337e-02 1.02777863e+00
6.88198507e-02 -2.06493542e-01 -2.52942413e-01 -2.19235551e-02
-8.43359753e-02 3.05694211e-02 1.13187420e+00 -2.91238338e-01
-1.69197291e-01 -1.77224264e-01 -1.01550221e+00 8.18060517e-01
7.02531457e-01 -7.51577854e-01 -4.87914681e-01 -4.15373147e-01
1.07295120e+00 4.69319314e-01 2.26171955e-01 -3.68159413e-01
2.50367612e-01 -9.04503644e-01 -3.33730966e-01 -3.57391715e-01
-5.50954998e-01 -7.38606572e-01 3.40743065e-01 3.93477082e-01
-4.47943479e-01 2.11266771e-01 9.70857292e-02 -5.26396260e-02
-2.12618470e-01 -4.98075038e-01 8.12088624e-02 -2.89864279e-02
-6.74548566e-01 -2.72909403e-01 -4.25095409e-01 -3.26106809e-02
8.43528211e-01 3.24219584e-01 -4.58099633e-01 6.07452467e-02
-3.13474208e-01 -8.37639906e-03 -1.17745670e-03 5.77564418e-01
5.46705842e-01 -1.14855814e+00 -9.80261564e-01 -2.68367887e-01
2.21629649e-01 -4.90542836e-02 -2.94832110e-01 1.04149580e+00
-2.81410992e-01 5.92103899e-01 -1.30389661e-01 -1.04464099e-01
-1.07218122e+00 6.37173891e-01 -1.80784747e-01 -3.36367399e-01
-4.31282461e-01 6.07848287e-01 -1.32940756e-02 -3.03567559e-01
-2.03939989e-01 -7.69801557e-01 -5.64505219e-01 1.27296314e-01
6.02792203e-01 4.77157347e-02 1.47071004e-01 -4.58420873e-01
-8.24975610e-01 1.85095221e-01 4.51285511e-01 -3.83129418e-01
1.56919754e+00 3.22449148e-01 -1.00639725e+00 6.86141551e-01
5.60693622e-01 4.06384349e-01 -3.99125457e-01 2.20072508e-01
5.88957489e-01 -1.60585478e-01 -6.83239922e-02 -9.57462013e-01
-7.84134090e-01 3.76828253e-01 -3.74612302e-01 6.83859885e-01
1.12631440e+00 2.12020442e-01 4.52006966e-01 2.35908866e-01
4.15893555e-01 -6.84799731e-01 -1.39433309e-01 8.63891482e-01
1.22441173e+00 -6.39428437e-01 -1.60560757e-01 -7.82456756e-01
-3.53110939e-01 1.27180064e+00 3.48286062e-01 6.05476871e-02
2.03212634e-01 8.38197649e-01 -6.16367683e-02 -5.43510139e-01
-1.10270047e+00 -3.91503721e-01 2.44088709e-01 3.87797803e-01
1.09448791e+00 3.76713037e-01 -8.65241587e-01 7.18679070e-01
-3.67836654e-01 1.62178650e-01 8.74472439e-01 1.00370634e+00
4.46765013e-02 -1.36459744e+00 -6.10862136e-01 2.46557295e-01
-7.02029228e-01 -4.74838674e-01 -7.18538046e-01 6.91670656e-01
2.88150668e-01 1.58563185e+00 -4.29626584e-01 -6.17716193e-01
6.92105591e-01 1.13803573e-01 5.87327182e-01 -7.90786803e-01
-6.90054178e-01 -6.61379099e-02 9.80317175e-01 -6.21831119e-01
-7.56989002e-01 -6.88778400e-01 -1.56025410e+00 -1.69765949e-01
-3.06942523e-01 2.38311097e-01 5.33832967e-01 1.32905269e+00
-2.32554182e-01 1.08871627e+00 3.42895865e-01 -1.09033361e-01
-1.89692467e-01 -9.28614318e-01 -2.58308649e-01 2.36643538e-01
9.50697064e-02 -7.06215382e-01 -2.93386102e-01 4.24461335e-01] | [9.211508750915527, 9.006542205810547] |
dc323ced-5df7-465e-8c89-30c47ee3a1ea | neural-models-for-reasoning-over-multiple | 1804.05922 | null | http://arxiv.org/abs/1804.05922v1 | http://arxiv.org/pdf/1804.05922v1.pdf | Neural Models for Reasoning over Multiple Mentions using Coreference | Many problems in NLP require aggregating information from multiple mentions
of the same entity which may be far apart in the text. Existing Recurrent
Neural Network (RNN) layers are biased towards short-term dependencies and
hence not suited to such tasks. We present a recurrent layer which is instead
biased towards coreferent dependencies. The layer uses coreference annotations
extracted from an external system to connect entity mentions belonging to the
same cluster. Incorporating this layer into a state-of-the-art reading
comprehension model improves performance on three datasets -- Wikihop, LAMBADA
and the bAbi AI tasks -- with large gains when training data is scarce. | ['Ruslan Salakhutdinov', 'Bhuwan Dhingra', 'Qiao Jin', 'Zhilin Yang', 'William W. Cohen'] | 2018-04-16 | neural-models-for-reasoning-over-multiple-1 | https://aclanthology.org/N18-2007 | https://aclanthology.org/N18-2007.pdf | naacl-2018-6 | ['lambada'] | ['natural-language-processing'] | [ 1.15367778e-01 9.66933608e-01 -1.65767297e-01 -5.79757154e-01
-1.00006247e+00 -6.37442946e-01 5.51519275e-01 3.87494773e-01
-7.25557268e-01 9.44583893e-01 8.07719052e-01 -4.47529227e-01
-2.49127612e-01 -5.69114149e-01 -8.88367593e-01 -3.81912380e-01
-1.43491970e-02 1.25789833e+00 1.84970289e-01 -5.79338074e-01
-9.94743332e-02 7.48398132e-04 -9.01946962e-01 6.32421017e-01
9.45222080e-01 4.32846189e-01 1.84595555e-01 6.22837186e-01
-2.83233970e-01 1.01950109e+00 -6.24213755e-01 -7.05531061e-01
-7.45922998e-02 -3.92072946e-01 -1.57122087e+00 -9.87898767e-01
5.92625797e-01 1.39139131e-01 -4.61095810e-01 7.53471673e-01
4.71382201e-01 2.64515340e-01 6.66423738e-01 -6.98601186e-01
-7.65351951e-01 1.55159998e+00 -3.26488286e-01 5.15974045e-01
4.92702007e-01 -4.91114050e-01 1.63289690e+00 -5.76348662e-01
9.96556997e-01 1.13904583e+00 6.03880823e-01 6.70340180e-01
-1.25659919e+00 -4.88009453e-01 2.87004024e-01 4.72461253e-01
-9.48648036e-01 -5.44538796e-01 4.53217536e-01 1.80264905e-01
1.87436175e+00 1.36553198e-01 -1.32316336e-01 1.39222825e+00
-1.49530351e-01 9.55039561e-01 5.43177068e-01 -4.81152236e-01
-4.05354559e-01 -1.80094451e-01 7.16557205e-01 1.63304240e-01
1.50767311e-01 -1.24716908e-01 -4.69007790e-01 2.24173695e-01
1.25389487e-01 -6.19214714e-01 -5.62438190e-01 -2.41806954e-01
-1.31947482e+00 7.04216540e-01 8.40611517e-01 6.77768528e-01
-4.46566701e-01 -9.65597555e-02 4.76097971e-01 7.22652793e-01
3.51720870e-01 9.67438519e-01 -9.39580798e-01 4.50403318e-02
-7.67085552e-01 4.88452911e-02 1.42731297e+00 1.02357376e+00
3.69830489e-01 -5.88272870e-01 -1.50848478e-01 9.86762583e-01
2.30441704e-01 2.84070969e-01 3.17462206e-01 -7.75992572e-01
1.22445953e+00 4.90591288e-01 8.96327049e-02 -7.45263994e-01
-9.25928891e-01 -4.64106500e-01 -7.46959031e-01 -3.40448529e-01
5.53513467e-01 -5.12585998e-01 -8.19892466e-01 2.01587272e+00
2.65942812e-02 8.62387344e-02 6.87505126e-01 7.42240489e-01
1.49161363e+00 7.24650443e-01 2.25203335e-01 -1.16461933e-01
1.22558689e+00 -1.10252738e+00 -7.39175797e-01 -4.88439798e-01
9.46632683e-01 -4.90769535e-01 5.02692759e-01 1.27433874e-02
-1.27851462e+00 -2.09398463e-01 -8.93699348e-01 -4.01089579e-01
-7.27147698e-01 -6.48708194e-02 3.42744827e-01 -4.04717326e-02
-1.03097618e+00 5.55945039e-01 -5.67400932e-01 -5.45269132e-01
6.07234985e-02 6.21614516e-01 -5.13352036e-01 4.00696024e-02
-1.72759855e+00 1.75160575e+00 7.46691108e-01 3.15081865e-01
-3.73726696e-01 -6.94340169e-01 -9.55330372e-01 3.57567072e-01
4.28419083e-01 -6.85220718e-01 1.41316485e+00 -7.75816381e-01
-1.25480497e+00 9.56640840e-01 -2.74775594e-01 -7.97319770e-01
1.45812333e-01 -7.08608925e-01 -5.62465549e-01 3.79505903e-02
4.20606099e-02 7.61365354e-01 2.81275570e-01 -1.11224723e+00
-5.92192888e-01 -2.67629832e-01 1.03936568e-01 5.75870275e-01
2.52157301e-01 1.08930767e-01 -2.09812179e-01 -2.40379706e-01
-1.58224538e-01 -8.58224988e-01 -3.59554589e-03 -1.18476439e+00
-7.89306521e-01 -7.42034256e-01 6.16660297e-01 -8.44521344e-01
1.12944734e+00 -1.72781503e+00 7.71905422e-01 1.02884643e-01
7.19956756e-02 5.70830286e-01 -4.55405504e-01 4.95712340e-01
-6.61016285e-01 1.13533624e-01 -7.54023844e-04 -2.16704950e-01
9.01202485e-03 4.66697425e-01 -2.61480749e-01 5.85446022e-02
2.28592321e-01 1.09492409e+00 -1.12435579e+00 -2.70803392e-01
-5.31766638e-02 2.70025879e-01 -4.47292887e-02 4.56154883e-01
-4.02170718e-01 3.90970647e-01 -2.92387068e-01 9.27402228e-02
3.34235638e-01 -2.56219596e-01 2.72608429e-01 -2.36958772e-01
1.55927181e-01 1.18165946e+00 -7.04204142e-01 1.95138991e+00
-5.89328051e-01 7.55090892e-01 9.77063645e-03 -1.16228271e+00
7.43441284e-01 6.90314233e-01 -3.94277833e-02 -7.44728148e-01
2.74369150e-01 2.04750806e-01 3.61467630e-01 -5.66168964e-01
5.73562086e-01 6.89894855e-02 -2.90902972e-01 5.10801494e-01
5.89719296e-01 2.93757260e-01 4.55725431e-01 6.73757017e-01
1.27311599e+00 6.70513436e-02 2.92948097e-01 -2.36561909e-01
5.12186468e-01 -4.66120429e-02 4.30757076e-01 8.22383940e-01
2.10707754e-01 6.03397012e-01 6.46461487e-01 -5.40249050e-02
-8.42001736e-01 -9.30401146e-01 -1.23634227e-01 1.50514901e+00
2.32417677e-02 -4.53531593e-01 -4.81398433e-01 -1.02746463e+00
-2.44889796e-01 1.13043928e+00 -8.71955454e-01 -6.14897301e-03
-1.02394342e+00 -6.21864498e-01 7.10080147e-01 8.28788102e-01
3.13597351e-01 -1.31636274e+00 -4.57292318e-01 4.10825551e-01
-7.04385996e-01 -1.17651343e+00 -2.18845561e-01 7.65268207e-01
-4.35603201e-01 -1.12861192e+00 -6.14230037e-01 -9.93607819e-01
3.60416263e-01 -3.37711662e-01 1.85680783e+00 -2.29832437e-02
1.49041727e-01 1.51673943e-01 -5.92002094e-01 -3.73714507e-01
-4.32156384e-01 1.07222140e+00 -3.33252817e-01 -6.04463279e-01
9.74918008e-01 -4.64847535e-01 -1.36736734e-02 -2.15584952e-02
-3.27779114e-01 -1.80550113e-01 6.98514640e-01 1.01927006e+00
4.42984961e-02 -5.62819958e-01 8.36179733e-01 -1.52684855e+00
6.92467391e-01 -8.54295671e-01 -9.04874876e-02 7.27024257e-01
-2.21006781e-01 3.96015137e-01 6.45509660e-01 -2.49631554e-01
-1.59560120e+00 -2.95665532e-01 -2.83986509e-01 1.50904372e-01
-4.67445582e-01 9.10444736e-01 -2.43233696e-01 5.53185999e-01
6.72785938e-01 -3.41595292e-01 -4.99101669e-01 -5.99894226e-01
7.08723545e-01 5.90048313e-01 1.13322341e+00 -5.40385187e-01
2.97452658e-01 -3.03630441e-01 -3.08243096e-01 -6.73485756e-01
-1.46458447e+00 -6.80711865e-01 -1.06256425e+00 3.59941065e-01
8.60284567e-01 -8.29071820e-01 -6.17958426e-01 1.46162450e-01
-1.50341749e+00 -3.59513402e-01 -8.42485204e-02 5.28204441e-01
-2.64850199e-01 -4.43929508e-02 -8.48274648e-01 -2.49567851e-01
-6.03239477e-01 -5.71641982e-01 5.54473639e-01 4.83744800e-01
-7.16017306e-01 -1.07054985e+00 3.93533081e-01 4.23730075e-01
4.53028053e-01 -1.45134658e-01 1.10891891e+00 -1.54395199e+00
-1.84177607e-01 5.79265580e-02 -2.91258365e-01 -8.87071118e-02
-3.05278122e-01 -1.47631139e-01 -8.32155585e-01 -4.72320989e-02
-4.15821970e-01 -6.81730211e-01 1.16187060e+00 2.44419247e-01
3.64689469e-01 -2.40816846e-01 -6.87489212e-01 3.47792745e-01
1.06154871e+00 -1.61920816e-01 6.42507911e-01 4.19610173e-01
7.90347815e-01 8.75260234e-01 3.27874899e-01 -2.91061044e-01
6.64231479e-01 4.74822640e-01 3.83994102e-01 -1.32308677e-02
-1.75359190e-01 -1.47562534e-01 1.77114591e-01 1.06561041e+00
-1.34326860e-01 -5.75741410e-01 -1.14775300e+00 1.07332420e+00
-2.09646988e+00 -9.21850204e-01 -4.00052905e-01 1.60963082e+00
1.23366570e+00 4.47989665e-02 -2.71039754e-01 -5.52303374e-01
6.72997713e-01 9.82684046e-02 -5.43772936e-01 -6.39526784e-01
-2.79726356e-01 4.02633160e-01 4.42395538e-01 7.13998079e-01
-1.14835882e+00 1.16898942e+00 5.84195852e+00 1.92932442e-01
-5.47763288e-01 4.77140676e-03 2.51807958e-01 -5.59657961e-02
-1.33444473e-01 -1.21823261e-02 -1.00811756e+00 2.35621445e-02
1.34845614e+00 1.49379104e-01 7.93007091e-02 3.10131520e-01
-4.59601641e-01 -9.99081507e-02 -1.54369688e+00 4.61091876e-01
2.34635741e-01 -1.16131663e+00 -2.23716460e-02 -4.87085104e-01
4.67420667e-01 8.34151566e-01 -2.98453242e-01 6.17008030e-01
1.06473839e+00 -1.48906600e+00 1.84007332e-01 5.45237958e-01
3.75537783e-01 -8.65575254e-01 1.40390682e+00 4.03673232e-01
-8.38629901e-01 2.01397851e-01 -3.79022419e-01 2.07010165e-01
4.24119920e-01 9.13142562e-02 -5.64572573e-01 7.84604788e-01
8.80981803e-01 8.49870503e-01 -5.49143374e-01 7.25462735e-01
-9.12176371e-01 6.22884214e-01 -4.78799969e-01 -5.77576607e-02
4.29239929e-01 7.00947344e-02 6.42189562e-01 1.59656656e+00
-1.57922357e-01 3.83516818e-01 -2.59765446e-01 7.63464808e-01
-5.96255064e-01 -8.12780187e-02 -4.96374309e-01 1.95733368e-01
6.67607367e-01 1.27334690e+00 -9.88143906e-02 -2.29398742e-01
-6.29617274e-01 8.49230528e-01 1.06516838e+00 4.05964613e-01
-5.54448962e-01 -8.74428213e-01 3.68917853e-01 -5.46471179e-01
6.10041440e-01 1.64854437e-01 -6.94760541e-03 -1.24117422e+00
-2.77070016e-01 -8.34023297e-01 8.95145059e-01 -7.61842489e-01
-1.46036410e+00 8.75371575e-01 5.86176366e-02 -3.62518698e-01
-6.37101829e-01 -6.04393125e-01 -8.49405944e-01 1.09099782e+00
-1.61142540e+00 -1.28791428e+00 5.51963737e-03 5.16824126e-01
2.00795174e-01 -1.28640741e-01 1.20918834e+00 1.02451064e-01
-6.59926057e-01 6.78108633e-01 9.64873508e-02 7.15933681e-01
1.11765432e+00 -1.56466711e+00 5.10463953e-01 7.40074217e-01
4.49895978e-01 9.46626604e-01 8.56696427e-01 -3.84453446e-01
-7.02007711e-01 -8.02347600e-01 1.64100862e+00 -8.29512656e-01
6.81973457e-01 -1.68453559e-01 -1.37265408e+00 1.34138632e+00
1.08949959e+00 -3.92493337e-01 8.26165497e-01 1.04387832e+00
-6.29029453e-01 3.60903740e-01 -7.13939726e-01 3.75333965e-01
9.82225716e-01 -5.36777675e-01 -1.79120946e+00 3.95963937e-02
7.56159961e-01 -6.15450263e-01 -1.19049871e+00 4.18305486e-01
5.16995862e-02 -4.91859376e-01 8.85154426e-01 -1.10922480e+00
3.53777409e-01 1.29950317e-02 1.58370152e-01 -1.90792632e+00
-2.69427985e-01 -5.97713470e-01 -3.20090890e-01 1.40217125e+00
1.31656122e+00 -3.75616372e-01 4.63349342e-01 7.31422603e-01
-1.71962872e-01 -2.82058060e-01 -9.15827930e-01 -4.57624406e-01
5.94304204e-01 6.25470355e-02 3.57218802e-01 1.10382438e+00
9.49596107e-01 1.49324167e+00 -3.82495746e-02 2.15608045e-01
2.71311492e-01 1.06590584e-01 4.73375767e-01 -1.43832541e+00
-1.29570290e-01 -3.79791319e-01 2.26853818e-01 -1.25642884e+00
7.80937731e-01 -1.13350308e+00 2.50011951e-01 -1.69833708e+00
1.03768215e-01 -3.21263611e-01 -3.30843002e-01 8.05596352e-01
-2.54774451e-01 -2.96654433e-01 1.22270629e-01 4.43834290e-02
-9.11683917e-01 3.11042577e-01 6.60803854e-01 -1.12168215e-01
-9.34996605e-02 -2.31340230e-01 -7.66035259e-01 6.90991998e-01
7.94898748e-01 -5.62808573e-01 -1.92312077e-01 -9.41205561e-01
5.68050325e-01 1.04842745e-01 -2.14567497e-01 -5.08243620e-01
7.48282135e-01 5.21019578e-01 1.98203757e-01 -6.65060818e-01
1.72211245e-01 -6.88493550e-01 -3.42268795e-01 4.86215353e-02
-1.00008214e+00 4.64661904e-02 1.28979102e-01 4.21904147e-01
-3.33907723e-01 -5.58672667e-01 4.01054770e-01 -3.73914301e-01
-6.05194807e-01 -2.35192418e-01 -1.40093356e-01 6.89848959e-01
5.86557031e-01 4.76998687e-01 -8.98901105e-01 -4.77735430e-01
-1.03096151e+00 7.54435539e-01 -3.30510177e-02 6.96037233e-01
8.08022842e-02 -6.55590534e-01 -1.17332745e+00 -2.73958355e-01
-5.10816798e-02 3.20647627e-01 -1.04020841e-01 7.57663608e-01
-1.79660078e-02 9.25401509e-01 -1.66969612e-01 -1.82780325e-01
-1.38481772e+00 4.96970505e-01 4.49098468e-01 -7.97653317e-01
-6.22569501e-01 1.14115369e+00 -1.08714595e-01 -9.89430368e-01
4.34588403e-01 -3.04430783e-01 -8.72882247e-01 4.48144108e-01
4.52917993e-01 1.84895277e-01 2.20117897e-01 -8.42646897e-01
-4.71346915e-01 1.95960686e-01 -6.64480090e-01 1.31050110e-01
1.63554966e+00 -3.12285185e-01 -2.30824143e-01 4.73982483e-01
1.06680644e+00 -2.61938095e-01 -7.19273269e-01 -4.96418715e-01
5.08684337e-01 4.64157581e-01 -1.88250199e-01 -1.32105005e+00
-8.75918210e-01 8.76216948e-01 -3.17669690e-01 2.05262557e-01
4.99193966e-01 5.04625559e-01 7.97031760e-01 9.25195694e-01
-1.48937562e-02 -1.21215665e+00 -5.02360284e-01 1.22055852e+00
7.60947883e-01 -1.32169545e+00 -2.37682059e-01 -1.57711312e-01
-7.08584189e-01 1.05559707e+00 8.30311120e-01 -9.09323469e-02
4.12481546e-01 2.50361264e-01 1.21017009e-01 -5.06418765e-01
-1.04347587e+00 -3.81182820e-01 4.39097375e-01 5.50130308e-01
8.13475192e-01 -3.02030481e-02 -2.39645258e-01 7.26651788e-01
-2.09260553e-01 -5.00787616e-01 4.97191697e-01 5.53925455e-01
-1.40636206e-01 -9.77604210e-01 1.50917873e-01 5.09635985e-01
-6.49668753e-01 -5.14520168e-01 -7.44870007e-01 8.87001216e-01
-2.75532305e-01 1.07819271e+00 1.76153079e-01 1.19749092e-01
4.82940376e-01 4.14466292e-01 4.16043550e-01 -7.33159482e-01
-1.08138192e+00 -3.47226560e-01 7.80213654e-01 -4.50809091e-01
-8.16858649e-01 -7.57123768e-01 -1.49049485e+00 -3.70417908e-02
-7.14649081e-01 5.87358236e-01 1.88799724e-01 1.28054845e+00
3.26414794e-01 7.09070742e-01 -1.49149984e-01 -4.00440574e-01
-2.21421093e-01 -1.34964335e+00 -2.57798374e-01 6.04532778e-01
3.04760218e-01 -3.66629571e-01 -2.86395252e-01 -1.96192876e-01] | [9.49134349822998, 9.336053848266602] |
a701cd1c-cd52-4317-a2dc-3894da8f825b | gradient-hyperalignment-for-multi-subject | 1807.02612 | null | http://arxiv.org/abs/1807.02612v1 | http://arxiv.org/pdf/1807.02612v1.pdf | Gradient Hyperalignment for multi-subject fMRI data alignment | Multi-subject fMRI data analysis is an interesting and challenging problem in
human brain decoding studies. The inherent anatomical and functional
variability across subjects make it necessary to do both anatomical and
functional alignment before classification analysis. Besides, when it comes to
big data, time complexity becomes a problem that cannot be ignored. This paper
proposes Gradient Hyperalignment (Gradient-HA) as a gradient-based functional
alignment method that is suitable for multi-subject fMRI datasets with large
amounts of samples and voxels. The advantage of Gradient-HA is that it can
solve independence and high dimension problems by using Independent Component
Analysis (ICA) and Stochastic Gradient Ascent (SGA). Validation using
multi-classification tasks on big data demonstrates that Gradient-HA method has
less time complexity and better or comparable performance compared with other
state-of-the-art functional alignment methods. | ['Daoqiang Zhang', 'Tonglin Xu', 'Muhammad Yousefnezhad'] | 2018-07-07 | null | null | null | null | ['brain-decoding', 'multi-subject-fmri-data-alignment', 'brain-decoding'] | ['medical', 'medical', 'miscellaneous'] | [ 9.88131203e-03 -6.32885754e-01 2.70837873e-01 -7.13747084e-01
-4.42760229e-01 -3.14855903e-01 2.96561599e-01 -1.44369408e-01
-7.98973203e-01 9.71545756e-01 2.00419456e-01 -9.94314104e-02
-4.13058460e-01 -2.61472017e-01 -5.03009319e-01 -7.59224474e-01
-4.51532602e-01 6.36054158e-01 6.32014573e-02 8.82703736e-02
3.20195317e-01 4.72207636e-01 -1.12254131e+00 1.35887504e-01
1.26563537e+00 7.16056526e-01 4.08830434e-01 9.24723744e-02
5.51020280e-02 2.48584703e-01 -1.88635811e-01 -2.13632658e-01
4.48829204e-01 -7.13244557e-01 -6.57931268e-01 -2.80817628e-01
3.41914535e-01 3.97523679e-02 7.32947811e-02 1.14506948e+00
9.47465241e-01 3.75229478e-01 6.43732846e-01 -1.10582805e+00
-4.47638690e-01 5.02631664e-01 -8.01460981e-01 1.02610826e+00
1.45162363e-02 -9.41310916e-03 4.90104735e-01 -1.04154801e+00
4.51506048e-01 1.05845249e+00 4.47020262e-01 3.60656589e-01
-1.39258814e+00 -9.20921803e-01 5.67295291e-02 5.89304686e-01
-1.32219458e+00 -4.24164504e-01 6.24775648e-01 -7.39231408e-01
9.17993069e-01 3.51511985e-01 8.49695027e-01 1.06333613e+00
6.15681708e-01 3.97783160e-01 1.59234440e+00 -2.36172397e-02
5.39076924e-01 -4.14121836e-01 4.00549054e-01 4.64667648e-01
2.93859363e-01 -8.48637447e-02 -4.29068804e-01 -1.36853844e-01
6.82014525e-01 -1.82588965e-01 -2.42481232e-01 -2.54701853e-01
-1.64724052e+00 7.39466786e-01 4.90254939e-01 7.28515327e-01
-7.73342013e-01 -1.73170701e-01 4.50316221e-01 1.21196143e-01
4.70986009e-01 5.82751274e-01 -1.17934123e-01 -5.45260608e-02
-1.48692918e+00 1.85645029e-01 4.01512325e-01 2.78014183e-01
3.09193373e-01 4.41903055e-01 -2.26686314e-01 9.04339969e-01
2.77481794e-01 4.62355763e-01 1.18781519e+00 -6.03066146e-01
2.42201284e-01 3.71000439e-01 -4.78810698e-01 -1.03699696e+00
-8.77732873e-01 -6.32048666e-01 -1.18102729e+00 7.99304694e-02
4.34997112e-01 -8.86595100e-02 -7.39513874e-01 1.90979290e+00
1.97659656e-01 1.69071183e-01 -4.36261296e-01 1.21630001e+00
5.38999915e-01 1.05160378e-01 1.23459779e-01 -6.18734479e-01
1.33314788e+00 -8.78974438e-01 -8.22711885e-01 -6.77359283e-01
3.32204372e-01 -4.45813477e-01 9.84497964e-01 3.94379437e-01
-8.79272819e-01 -4.40850258e-01 -9.77648377e-01 1.15542017e-01
-2.45714813e-01 -3.07962857e-02 1.07656825e+00 6.47491276e-01
-9.96918976e-01 3.23013455e-01 -9.40046072e-01 -1.58439338e-01
7.02353895e-01 6.32757127e-01 -8.75336945e-01 -1.29287884e-01
-9.42407608e-01 1.08703840e+00 4.32179034e-01 3.46420079e-01
-9.28081930e-01 -7.13733792e-01 -6.71810925e-01 -8.39581341e-02
2.97594685e-02 -5.74345052e-01 5.25177598e-01 -1.01473653e+00
-1.35234070e+00 6.54870689e-01 -4.43446934e-01 -1.81189999e-01
4.13121849e-01 -2.12199450e-01 -4.02685136e-01 1.59660295e-01
2.13514760e-01 5.82139790e-01 8.43895972e-01 -5.84484875e-01
2.59296417e-01 -1.12070894e+00 -6.11558557e-01 4.29921240e-01
-3.12753469e-01 2.00738147e-01 7.35740811e-02 -5.11494637e-01
6.06087744e-01 -8.61953199e-01 -3.11175048e-01 -3.09037596e-01
-2.22055227e-01 2.22534657e-01 3.58842134e-01 -1.06890678e+00
1.01803243e+00 -2.00040793e+00 6.94313109e-01 2.95365870e-01
4.83896732e-01 7.17452765e-02 -5.71466610e-02 -1.43376380e-01
-5.67270815e-01 -5.63567840e-02 -5.70195615e-01 1.24653175e-01
-3.44431043e-01 -2.01776952e-01 4.87647593e-01 9.68936503e-01
-1.43537119e-01 1.04013443e+00 -7.77492106e-01 -5.41461825e-01
1.19855478e-01 3.19238782e-01 -6.82607591e-01 5.55253103e-02
4.57873613e-01 1.01753986e+00 -2.12761134e-01 4.61886227e-01
6.70961082e-01 -1.42760873e-01 3.40797663e-01 -5.16414702e-01
-2.15443820e-01 -2.05238655e-01 -1.03915167e+00 2.03725338e+00
-1.37512907e-01 5.66217244e-01 8.68360475e-02 -1.42893350e+00
6.94700003e-01 2.92733192e-01 7.25450575e-01 -8.45073223e-01
4.18767244e-01 4.79741663e-01 1.00014591e+00 -5.21951795e-01
-1.41919822e-01 -3.47052991e-01 1.83509976e-01 4.34109867e-01
4.30328637e-01 1.68801233e-01 2.72599697e-01 8.77458379e-02
1.06962562e+00 -1.77836090e-01 4.82773989e-01 -1.00193536e+00
5.35468221e-01 -4.31744576e-01 6.35725677e-01 3.41517717e-01
-4.00767595e-01 3.39235276e-01 3.39238882e-01 -3.35198700e-01
-9.33196127e-01 -9.70044553e-01 -2.73364127e-01 6.00892484e-01
-3.56606811e-01 -1.35501012e-01 -8.61535311e-01 -4.13724780e-01
-5.11455536e-02 3.88742954e-01 -7.82217562e-01 -1.43121809e-01
-5.77079892e-01 -1.62563157e+00 1.52781248e-01 3.59622061e-01
8.29828799e-01 -7.88945436e-01 -5.90226114e-01 3.34421843e-01
-1.90440238e-01 -9.68679070e-01 -3.11730921e-01 2.33706176e-01
-1.15931094e+00 -8.87579322e-01 -8.20097625e-01 -4.75107431e-01
7.75904953e-01 1.93761081e-01 6.92976773e-01 -3.45422924e-01
-5.50615072e-01 -1.47798494e-01 -1.86517350e-02 1.06764464e-02
2.02008322e-01 -9.90709290e-02 3.17516834e-01 3.92961204e-01
1.95098072e-01 -1.02907193e+00 -7.87868023e-01 2.57284552e-01
-6.58964753e-01 6.52603731e-02 7.60718107e-01 9.22186434e-01
5.81774592e-01 -1.52238607e-01 7.31357992e-01 -5.92586219e-01
1.01321614e+00 -5.59349537e-01 -4.77331758e-01 1.95978612e-01
-5.91638982e-01 7.27474242e-02 6.04467809e-01 -5.57558835e-01
-8.98250103e-01 1.04018599e-02 -9.81048346e-02 -2.25605950e-01
-2.51597696e-04 5.63084066e-01 -3.00553888e-01 -3.75895202e-01
7.73066640e-01 2.51084119e-01 2.00127184e-01 -6.22542202e-01
1.28916398e-01 4.06664580e-01 2.73356766e-01 -2.77510017e-01
1.38916329e-01 3.13617289e-01 2.34072357e-01 -7.71251798e-01
-3.99742216e-01 -3.01751852e-01 -1.23876905e+00 -3.89443964e-01
1.02589071e+00 -7.17597902e-01 -6.52328670e-01 5.55999398e-01
-6.81346476e-01 -2.07801983e-01 2.16942534e-01 1.03650105e+00
-2.06219926e-01 5.42992175e-01 -3.86552036e-01 -3.81530583e-01
-5.71608603e-01 -1.54324269e+00 4.74967897e-01 1.62183687e-01
-1.65619522e-01 -8.99150372e-01 -2.40511959e-03 4.58983332e-01
7.02119350e-01 2.48339325e-01 8.11893404e-01 -7.25826859e-01
3.75180095e-02 -2.00497568e-01 -7.26861283e-02 2.12864012e-01
-4.98010404e-02 -5.67061007e-01 -6.92065001e-01 -3.49691719e-01
5.61512053e-01 -1.30688384e-01 6.11311495e-01 7.31178641e-01
1.25460529e+00 -1.55955702e-01 -2.29498833e-01 5.50753593e-01
1.29225826e+00 1.55525386e-01 7.77354538e-01 9.37230140e-02
8.20297599e-01 5.02396345e-01 2.73441762e-01 1.37967274e-01
1.81454659e-01 6.46761835e-01 4.87534562e-03 -7.01827928e-02
-3.95802818e-02 2.97654331e-01 1.29575014e-01 1.10893798e+00
-2.18388304e-01 2.52418756e-01 -1.03823888e+00 3.05498600e-01
-1.79239523e+00 -1.18369281e+00 -5.17853260e-01 2.35549164e+00
6.77204609e-01 -1.56314209e-01 2.29055494e-01 1.60485879e-01
6.28991425e-01 -1.08979888e-01 -6.92005813e-01 -1.61806375e-01
-3.05198491e-01 2.97544450e-01 3.47007334e-01 2.79202908e-01
-8.61606359e-01 7.42188156e-01 6.83033657e+00 6.87433302e-01
-1.22200298e+00 8.83046389e-01 6.92758143e-01 -2.89030641e-01
1.23957261e-01 -8.62413868e-02 -4.27361429e-01 6.48050010e-01
1.05501533e+00 -1.65935919e-01 7.50976801e-01 5.63641012e-01
3.16847384e-01 -6.04703903e-01 -7.84187198e-01 1.26799059e+00
2.95153081e-01 -8.36972117e-01 -2.92279273e-01 1.39327675e-01
5.26817143e-01 2.72028893e-01 -2.36707479e-01 -9.51080993e-02
-2.72061497e-01 -1.13015819e+00 4.18680787e-01 4.78162706e-01
5.84905565e-01 -4.49730784e-01 6.26916587e-01 2.73423553e-01
-8.82292271e-01 -1.54584169e-01 -5.09529650e-01 -2.38386065e-01
4.14285541e-01 1.16547275e+00 -2.04938963e-01 2.96366155e-01
6.07807279e-01 5.81195712e-01 -7.42040992e-01 1.39662457e+00
-1.89438224e-01 4.76803422e-01 -2.53367394e-01 1.50886521e-01
3.35218534e-02 -5.84468663e-01 4.64784950e-01 8.90055776e-01
3.54863852e-01 1.63527459e-01 1.95976689e-01 1.00075591e+00
2.66866237e-01 5.47198236e-01 -4.31716204e-01 -1.23959586e-01
2.04563171e-01 1.57353270e+00 -1.15317082e+00 -3.32163006e-01
-3.06874156e-01 1.20362437e+00 5.57533383e-01 6.88664196e-03
-7.62469471e-01 -7.82760829e-02 2.35414147e-01 -1.03943847e-01
-2.68452376e-01 -5.71554184e-01 -4.57647920e-01 -1.39382970e+00
9.40941796e-02 -1.04340374e+00 3.25052649e-01 -4.95701164e-01
-1.15193808e+00 8.92942190e-01 1.73192844e-01 -8.26312065e-01
9.31756124e-02 -4.79349375e-01 -3.73459548e-01 9.81102645e-01
-1.09005475e+00 -7.69891560e-01 -2.63411641e-01 8.63897860e-01
3.88027519e-01 -2.36695841e-01 9.22647357e-01 7.49438047e-01
-8.87806535e-01 2.37378061e-01 6.22799657e-02 -1.29839152e-01
5.02313673e-01 -8.22355092e-01 -9.18282568e-02 1.14358687e+00
8.32707435e-02 7.42415905e-01 3.89683455e-01 -8.00268352e-01
-1.40445209e+00 -4.83186126e-01 7.00611711e-01 -1.40230432e-02
4.95892107e-01 -6.03236020e-01 -9.60932314e-01 4.82882768e-01
1.67638749e-01 2.65512675e-01 1.08475745e+00 -1.58756841e-02
5.43519743e-02 -1.68290719e-01 -1.26292491e+00 3.46433491e-01
1.11655617e+00 -2.34465718e-01 -6.15445018e-01 6.48765266e-01
-6.94088563e-02 -1.22064166e-01 -1.06587088e+00 4.25319076e-01
5.94731688e-01 -1.03167415e+00 9.51050878e-01 -5.30273914e-01
-1.80846155e-02 -1.71370536e-01 4.75369580e-02 -1.65417862e+00
-8.02806914e-01 -2.29027197e-01 1.50538445e-01 8.02363575e-01
3.13799381e-01 -7.62388706e-01 2.52323627e-01 8.08188915e-01
-1.89161167e-01 -4.78253722e-01 -1.31209254e+00 -8.88710320e-01
1.07127585e-01 -1.46823227e-01 2.79960304e-01 1.19954479e+00
3.15672219e-01 5.14918447e-01 -3.79056185e-01 -1.68947831e-01
8.00849557e-01 6.55829385e-02 1.31018624e-01 -1.29267704e+00
-8.19230899e-02 -4.94719565e-01 -5.06648958e-01 -1.40744239e-01
3.40578884e-01 -1.32122707e+00 -1.66034967e-01 -1.57923615e+00
3.20500195e-01 -1.51298761e-01 -4.01314855e-01 6.46336913e-01
-2.88154840e-01 3.63366336e-01 -1.68763269e-02 3.35344464e-01
-2.54745871e-01 6.00355446e-01 1.20444119e+00 1.88715145e-01
-1.76701516e-01 -3.97882789e-01 -5.09606659e-01 3.68726790e-01
1.03546751e+00 -5.79544067e-01 -3.80083293e-01 -3.70348811e-01
-9.42822993e-02 -3.25831100e-02 2.05290914e-01 -1.44464171e+00
4.81292931e-03 1.98477451e-02 8.08767080e-01 -9.21533536e-03
1.87049061e-01 -5.80149055e-01 4.18686718e-01 7.03866780e-01
-5.98752573e-02 6.08979583e-01 2.66502231e-01 6.05909899e-02
-9.73941609e-02 -1.13358878e-01 9.10078049e-01 -4.46748525e-01
-6.03374958e-01 3.02593797e-01 -4.59621191e-01 5.34925349e-02
1.05597329e+00 -6.20556436e-02 -5.06702550e-02 1.09054506e-01
-8.32611442e-01 -6.46317005e-02 -4.09440659e-02 3.57550532e-01
7.80920565e-01 -1.31099212e+00 -7.45801508e-01 4.50640142e-01
-4.33267802e-01 -5.82817376e-01 6.48070097e-01 1.67321813e+00
-3.04761112e-01 4.09627944e-01 -8.95251989e-01 -4.64194447e-01
-1.15682220e+00 5.29462218e-01 2.32878789e-01 -2.49565691e-01
-6.66089237e-01 8.30586255e-01 8.26828778e-02 -2.56177455e-01
-3.07062924e-01 3.05093199e-01 -3.08063567e-01 3.04232389e-01
5.66677868e-01 3.95171851e-01 2.74592996e-01 -7.96704292e-01
-7.22356677e-01 3.25243145e-01 5.34533709e-02 -3.48603010e-01
1.57934999e+00 5.58405034e-02 -6.60413384e-01 4.63816464e-01
1.20148408e+00 -2.00212881e-01 -7.37601876e-01 1.51009142e-01
-1.66963544e-02 -5.67951620e-01 5.46175897e-01 -9.21350539e-01
-1.36832488e+00 1.14350128e+00 9.49615836e-01 -4.73019451e-01
1.04269445e+00 -4.07112569e-01 3.48220378e-01 1.51809812e-01
4.74631816e-01 -1.04789841e+00 -6.22608475e-02 1.97332412e-01
1.15829110e+00 -1.20881236e+00 1.89679965e-01 -1.39062420e-01
-7.02082276e-01 8.87309372e-01 8.32092822e-01 -2.74165004e-01
8.43325377e-01 1.29987597e-01 -2.12483302e-01 -4.83805627e-01
-1.22799098e-01 6.40438870e-03 5.98029435e-01 3.77712667e-01
6.10488176e-01 1.38418227e-01 -1.22554862e+00 6.54935539e-01
-2.15127915e-01 4.64403555e-02 5.99392056e-02 7.20309913e-01
-1.60190061e-01 -1.05366611e+00 -3.62285107e-01 9.72986579e-01
-5.38105667e-01 -3.39042813e-01 -1.72427967e-01 5.80836296e-01
-8.96704495e-02 6.20519280e-01 -2.52369523e-01 -1.80937484e-01
-1.07611015e-01 4.51226890e-01 8.75827372e-01 -4.25809205e-01
-7.20011115e-01 2.63238251e-01 -1.65970117e-01 -8.09431612e-01
-4.72183555e-01 -8.22610319e-01 -1.38956201e+00 5.79756126e-02
-4.23326194e-01 2.44363904e-01 1.04142058e+00 1.03013992e+00
6.97657049e-01 5.54803967e-01 3.32742542e-01 -8.93784106e-01
-2.38998532e-01 -1.33737957e+00 -7.90803254e-01 4.04679149e-01
-2.82076567e-01 -1.10796010e+00 -2.17962533e-01 -3.33276331e-01] | [12.622687339782715, 3.378903388977051] |
b4462c19-58a3-40d0-834a-475835f2f5a9 | f2net-learning-to-focus-on-the-foreground-for | 2012.02534 | null | https://arxiv.org/abs/2012.02534v1 | https://arxiv.org/pdf/2012.02534v1.pdf | F2Net: Learning to Focus on the Foreground for Unsupervised Video Object Segmentation | Although deep learning based methods have achieved great progress in unsupervised video object segmentation, difficult scenarios (e.g., visual similarity, occlusions, and appearance changing) are still not well-handled. To alleviate these issues, we propose a novel Focus on Foreground Network (F2Net), which delves into the intra-inter frame details for the foreground objects and thus effectively improve the segmentation performance. Specifically, our proposed network consists of three main parts: Siamese Encoder Module, Center Guiding Appearance Diffusion Module, and Dynamic Information Fusion Module. Firstly, we take a siamese encoder to extract the feature representations of paired frames (reference frame and current frame). Then, a Center Guiding Appearance Diffusion Module is designed to capture the inter-frame feature (dense correspondences between reference frame and current frame), intra-frame feature (dense correspondences in current frame), and original semantic feature of current frame. Specifically, we establish a Center Prediction Branch to predict the center location of the foreground object in current frame and leverage the center point information as spatial guidance prior to enhance the inter-frame and intra-frame feature extraction, and thus the feature representation considerably focus on the foreground objects. Finally, we propose a Dynamic Information Fusion Module to automatically select relatively important features through three aforementioned different level features. Extensive experiments on DAVIS2016, Youtube-object, and FBMS datasets show that our proposed F2Net achieves the state-of-the-art performance with significant improvement. | ['Pan Zhou', 'Changhu Wang', 'Dongdong Yu', 'Daizong Liu'] | 2020-12-04 | null | null | null | null | ['unsupervised-video-object-segmentation'] | ['computer-vision'] | [ 7.51589388e-02 -3.84175509e-01 -3.00198913e-01 -3.79527718e-01
-4.20493215e-01 -2.34272093e-01 3.39019626e-01 -3.07803243e-01
-3.18895847e-01 4.11357582e-01 9.76408124e-02 2.95612782e-01
-1.76858194e-02 -6.44264877e-01 -6.08969569e-01 -8.50368559e-01
2.13192686e-01 5.13419770e-02 9.31133628e-01 9.16952044e-02
2.09376290e-01 5.10642827e-01 -1.33236325e+00 1.31879792e-01
9.23345923e-01 1.32713246e+00 4.02662426e-01 3.34327787e-01
-3.71991307e-01 8.72485220e-01 -4.78145927e-01 -5.65662375e-03
2.52988130e-01 -4.17693019e-01 -8.00225794e-01 6.49842262e-01
3.25459272e-01 -7.15604961e-01 -6.16614401e-01 1.15451717e+00
3.07998568e-01 3.93023521e-01 3.40481222e-01 -1.24781179e+00
-3.54982644e-01 3.77101272e-01 -9.74958301e-01 7.09831238e-01
2.08737388e-01 2.60127068e-01 5.95153987e-01 -9.43911552e-01
7.85453796e-01 1.33045757e+00 2.65672475e-01 1.89502701e-01
-7.68830776e-01 -6.25059366e-01 6.04981065e-01 4.73612994e-01
-1.31213748e+00 -4.25459892e-01 1.09402573e+00 -3.43907923e-01
3.28968495e-01 -7.58541524e-02 9.24270391e-01 6.52198970e-01
3.04000638e-02 1.50181091e+00 5.69418013e-01 4.37000468e-02
-4.90198992e-02 -2.45338410e-01 1.12572014e-01 8.04885209e-01
-8.35549161e-02 -1.33832665e-02 -3.08255732e-01 2.67103553e-01
1.06672931e+00 4.55866247e-01 -5.95995963e-01 -5.35806239e-01
-1.39019656e+00 5.78431606e-01 5.84058642e-01 4.45981920e-01
-4.29847449e-01 8.89553130e-02 3.01929921e-01 -1.46626368e-01
2.99400806e-01 -2.71611124e-01 -5.61024904e-01 -4.46476974e-02
-1.08771753e+00 1.39433742e-01 4.31855798e-01 1.08044040e+00
9.69873309e-01 1.02094799e-01 -3.25370520e-01 7.81738520e-01
6.04014277e-01 5.02913415e-01 4.36795592e-01 -1.25911129e+00
4.56094414e-01 7.17866004e-01 -6.50475621e-02 -1.42049479e+00
-2.98988700e-01 -3.81408751e-01 -7.50443995e-01 -2.78803349e-01
3.17593455e-01 -1.17110215e-01 -9.95920718e-01 1.50118434e+00
6.86014116e-01 7.18157709e-01 -1.08633213e-01 1.23615932e+00
1.04925942e+00 7.99424410e-01 -2.04019397e-02 -4.77965802e-01
1.17538595e+00 -1.39972329e+00 -7.87616849e-01 -1.41849384e-01
4.04254228e-01 -8.44201684e-01 5.19659102e-01 -2.13543549e-02
-1.30977869e+00 -8.87734771e-01 -8.59034896e-01 -1.33822858e-01
-3.24815847e-02 2.91980743e-01 6.45901263e-01 1.17524333e-01
-6.56564057e-01 3.20871681e-01 -9.45261955e-01 -1.06389016e-01
7.82180011e-01 3.59013766e-01 -2.80233115e-01 -3.19550961e-01
-1.00971019e+00 3.04694772e-01 6.74208999e-01 2.88114041e-01
-9.09153581e-01 -7.46635795e-01 -9.50065434e-01 4.04630136e-03
7.24439263e-01 -6.81826591e-01 9.24171448e-01 -1.32957995e+00
-1.38477743e+00 4.85562563e-01 -2.06258029e-01 -1.64736509e-01
4.56331909e-01 -1.60400942e-01 -3.60003740e-01 4.89953309e-01
2.62972385e-01 9.72530246e-01 9.51713502e-01 -1.20833600e+00
-1.17980337e+00 -2.54112929e-01 4.24150527e-02 3.78428012e-01
-1.55444279e-01 1.70544580e-01 -1.03789175e+00 -7.80741870e-01
4.54562604e-01 -6.47321165e-01 -6.31442219e-02 1.73124641e-01
-2.12675244e-01 -2.82080859e-01 1.43222833e+00 -5.73557377e-01
1.20098138e+00 -2.27136898e+00 2.01747000e-01 -5.44926822e-02
3.35566908e-01 5.32991946e-01 -1.44852400e-01 -3.54112804e-01
-6.57780468e-02 -2.07747236e-01 -2.46919960e-01 -1.20997034e-01
-3.50632370e-01 2.84747869e-01 -1.06345741e-02 5.23317814e-01
3.82146746e-01 1.00901401e+00 -1.04058278e+00 -9.28389609e-01
6.66564941e-01 4.94247347e-01 -5.95403314e-01 2.31979877e-01
-6.13618270e-02 4.80889738e-01 -8.96955371e-01 7.97399700e-01
9.35665846e-01 -3.53523433e-01 -3.13217789e-01 -5.98989069e-01
-1.56099617e-01 -1.69613749e-01 -1.51137602e+00 1.96297133e+00
6.58353195e-02 4.02730703e-01 2.15821132e-01 -9.64892268e-01
6.94163501e-01 -7.46587366e-02 8.85762870e-01 -4.51547652e-01
4.94181514e-01 5.15152141e-02 8.46515745e-02 -6.17474437e-01
2.02881470e-01 3.78770769e-01 4.65000898e-01 2.10314676e-01
2.36813918e-01 1.95602491e-01 3.93378049e-01 2.44727343e-01
7.39947796e-01 4.92497981e-01 8.24165344e-02 -2.38302231e-01
1.00777042e+00 -2.80815065e-01 1.06996584e+00 1.93513036e-01
-6.57497108e-01 8.81986797e-01 2.47516513e-01 -6.04757190e-01
-6.42349958e-01 -8.72404754e-01 -1.42353959e-03 8.45747352e-01
1.04490197e+00 -2.54092067e-01 -8.23493600e-01 -9.34091866e-01
-1.37679547e-01 1.66219324e-01 -5.65036893e-01 -2.60330051e-01
-9.04615819e-01 -5.49459100e-01 -2.86608357e-02 7.33696640e-01
1.12404251e+00 -1.02706230e+00 -4.58114415e-01 2.81786412e-01
-4.08441365e-01 -1.26945996e+00 -9.40594494e-01 -1.75477952e-01
-8.36716056e-01 -1.13257527e+00 -8.86027455e-01 -9.49306309e-01
6.11596167e-01 6.86586797e-01 7.46588528e-01 2.92671442e-01
-1.13845967e-01 1.26028135e-01 -2.87544608e-01 1.90093741e-01
2.54505217e-01 -7.69702569e-02 -2.82532066e-01 3.76241535e-01
2.98526198e-01 -4.63584304e-01 -1.03098714e+00 5.75533807e-01
-1.01528072e+00 1.79756403e-01 6.23488426e-01 7.00683713e-01
7.78785050e-01 3.44909690e-02 2.90474206e-01 -6.14848137e-01
-9.30564255e-02 -5.06912529e-01 -3.78606141e-01 2.32026353e-01
-1.14142254e-01 -4.10413295e-01 3.40347648e-01 -3.82104725e-01
-1.07362950e+00 2.37563834e-01 -1.27495289e-01 -8.28494489e-01
-2.74752885e-01 1.34003446e-01 -5.76658130e-01 -1.46295711e-01
-1.10931277e-01 5.65452397e-01 -1.79927990e-01 -3.39295566e-01
2.85414785e-01 4.39195633e-01 7.03637660e-01 -5.85621536e-01
7.99665749e-01 6.38561308e-01 -2.92283684e-01 -4.58524346e-01
-1.03624403e+00 -5.91927648e-01 -9.04605210e-01 -2.72010654e-01
1.16557240e+00 -9.72459555e-01 -6.29709184e-01 6.48178875e-01
-1.19920433e+00 -6.75472394e-02 -2.76162267e-01 4.80212480e-01
-5.43632448e-01 6.35218799e-01 -6.28844500e-01 -3.16444874e-01
-2.58839875e-01 -1.54952848e+00 1.26289809e+00 8.04442644e-01
2.13647172e-01 -9.92612064e-01 -4.53561276e-01 4.44556952e-01
1.01863608e-01 1.89402282e-01 4.44086671e-01 -4.01965916e-01
-1.04462850e+00 1.70314834e-01 -7.04793692e-01 3.04994315e-01
3.63463521e-01 2.34514251e-01 -5.85960865e-01 -1.68542966e-01
9.42303017e-02 7.11037964e-02 9.28071797e-01 6.60650790e-01
1.35944045e+00 4.86154743e-02 -5.68931818e-01 1.15393782e+00
1.05610287e+00 4.19036865e-01 5.16803861e-01 2.73501217e-01
1.12970972e+00 3.29228193e-01 9.76742923e-01 3.31894428e-01
5.20377338e-01 6.03362083e-01 2.79213786e-01 -2.41409913e-01
-3.89967203e-01 -1.16663858e-01 1.99290723e-01 9.00275767e-01
3.16830426e-02 -6.47559166e-02 -4.44699883e-01 5.81823409e-01
-2.02314329e+00 -9.42333400e-01 1.14713877e-01 1.77129734e+00
6.37567818e-01 1.06132068e-01 1.22057967e-01 -1.65761739e-01
9.75086749e-01 3.20391953e-01 -8.71424198e-01 5.24574578e-01
-3.02194655e-01 -3.67614478e-01 2.30725423e-01 2.60118186e-01
-1.42757177e+00 1.10449100e+00 5.01245546e+00 1.13266230e+00
-1.09656179e+00 5.28410822e-02 9.53062534e-01 -1.46293128e-02
-4.52136099e-02 8.95408764e-02 -8.47605824e-01 8.73326242e-01
8.16244110e-02 -3.43081653e-02 1.95985913e-01 8.66134286e-01
2.20308527e-01 -1.49705008e-01 -8.10053587e-01 1.23617458e+00
2.15872556e-01 -1.43885446e+00 1.52708828e-01 -1.42341182e-01
9.99293447e-01 -1.42973602e-01 -1.69946119e-01 1.35827631e-01
-1.48489788e-01 -5.20796239e-01 7.32648313e-01 5.73101282e-01
4.28444743e-01 -9.59030390e-01 7.59944439e-01 1.38401717e-01
-1.59436691e+00 -9.74767953e-02 -4.08197433e-01 3.17372680e-01
3.47166449e-01 6.17535114e-01 -4.51607220e-02 7.11488068e-01
8.95079851e-01 1.37339270e+00 -4.74889278e-01 1.14282787e+00
-5.34151867e-03 2.90041417e-01 -3.99213195e-01 3.50691080e-01
5.24153590e-01 -3.63772988e-01 4.47546095e-01 1.02920210e+00
7.97857568e-02 3.54344100e-01 7.20810592e-01 7.60276616e-01
8.92380066e-03 4.11065035e-02 -9.95719582e-02 2.48044997e-01
4.54349250e-01 1.51022530e+00 -1.09023178e+00 -6.44120216e-01
-6.12221897e-01 1.06863070e+00 5.73181175e-02 6.19035304e-01
-1.00775027e+00 -4.23566937e-01 7.29969144e-01 5.79216890e-03
6.79462492e-01 -1.05982758e-01 1.17236666e-01 -1.47154188e+00
3.29933800e-02 -7.65908539e-01 3.27351809e-01 -8.20007443e-01
-1.02699876e+00 3.82289320e-01 -6.90291300e-02 -1.21433973e+00
7.32485577e-02 -3.55452865e-01 -7.72873104e-01 5.05692065e-01
-1.69529438e+00 -1.10647571e+00 -5.83031893e-01 8.53095293e-01
9.32624459e-01 -1.81691758e-02 -1.12826623e-01 6.61179602e-01
-9.65951502e-01 3.52239430e-01 4.89504598e-02 4.77381229e-01
5.48073351e-01 -8.89927089e-01 6.39802366e-02 1.00384176e+00
-7.03390315e-02 6.05394125e-01 -2.10157894e-02 -7.56078243e-01
-1.24498832e+00 -1.35946679e+00 3.39501888e-01 -4.22950536e-02
5.17762899e-01 -1.81658771e-02 -9.10572112e-01 5.12087584e-01
-8.96550193e-02 6.25110924e-01 2.24383622e-01 -5.24968743e-01
2.16938421e-01 -1.49856329e-01 -9.08750415e-01 5.19844532e-01
1.25020862e+00 -2.42235303e-01 -4.57090527e-01 3.36549819e-01
9.52256322e-01 -6.81499243e-01 -8.57349634e-01 6.31854117e-01
3.47176075e-01 -9.49272156e-01 1.08688593e+00 -3.00575733e-01
4.79865640e-01 -8.26809406e-01 -2.87230201e-02 -8.08228314e-01
-3.55150104e-01 -6.76477253e-01 -4.02516693e-01 1.46233153e+00
-3.18050206e-01 -3.05460989e-01 9.29492891e-01 4.31864947e-01
-3.03557515e-01 -1.20252001e+00 -6.68028951e-01 -3.91199052e-01
-2.96546817e-01 -2.39287511e-01 6.30276918e-01 8.37017477e-01
-6.04640543e-01 2.44238004e-01 -1.90640390e-01 8.74058437e-03
5.17662466e-01 4.90289748e-01 7.97035575e-01 -9.69920039e-01
-9.64816939e-03 -5.41163087e-01 -6.19289696e-01 -1.77307129e+00
2.33140945e-01 -6.32210195e-01 2.66857184e-02 -1.38956475e+00
4.52228993e-01 -4.19978172e-01 -4.38101321e-01 2.92446464e-01
-6.58432722e-01 1.88182428e-01 3.46833169e-01 3.82397383e-01
-1.09978056e+00 8.43821645e-01 1.78792322e+00 -1.80072278e-01
-1.44143596e-01 -2.58950070e-02 -4.08921570e-01 1.06981313e+00
3.42796147e-01 -2.27671430e-01 -3.75619203e-01 -5.35688400e-01
-6.07901812e-01 7.32662901e-02 3.64514172e-01 -1.05968082e+00
3.43026578e-01 -3.02896261e-01 8.72617781e-01 -9.09900188e-01
2.43717179e-01 -9.06516254e-01 -1.81385562e-01 3.23628306e-01
-1.05640113e-01 -1.71628855e-02 -1.29805639e-01 6.64121091e-01
-4.95380849e-01 -1.28442273e-01 8.68779778e-01 -5.37495986e-02
-1.16990316e+00 9.79593098e-01 -5.12027927e-02 2.16002136e-01
1.29691720e+00 -4.81620818e-01 -1.48810759e-01 -7.23612383e-02
-6.07183158e-01 5.65078497e-01 4.82501775e-01 5.53175390e-01
7.72582173e-01 -1.45105016e+00 -4.77846801e-01 4.21894014e-01
-1.97697669e-01 4.87545699e-01 5.30903399e-01 1.17469430e+00
-5.58604956e-01 1.32644907e-01 -1.46036044e-01 -9.28199232e-01
-1.03581381e+00 5.16364753e-01 4.03253943e-01 6.14484362e-02
-7.12914646e-01 9.84569848e-01 7.90355444e-01 -6.66411873e-03
2.08265722e-01 -3.68033201e-01 -2.56747782e-01 8.02650228e-02
6.60311520e-01 3.25451523e-01 -3.46810222e-01 -1.04084611e+00
-4.01217043e-01 9.49642301e-01 -3.04990381e-01 3.65988106e-01
1.04432058e+00 -5.39664328e-01 -1.47985384e-01 1.72330528e-01
1.49473786e+00 -2.38368943e-01 -1.82442713e+00 -5.56294739e-01
-2.56249756e-01 -8.15069079e-01 1.77325770e-01 -3.18296432e-01
-1.86124754e+00 9.39233243e-01 6.05448663e-01 -2.07768694e-01
1.36503601e+00 -4.24374118e-02 1.36246681e+00 1.44914150e-01
3.17046754e-02 -1.07782066e+00 2.56831169e-01 4.63382512e-01
3.62190515e-01 -1.30859697e+00 4.54673059e-02 -6.21009707e-01
-5.59856713e-01 1.12980866e+00 9.61831331e-01 -2.50455379e-01
7.95915425e-01 1.43634558e-01 1.17195234e-01 -6.79327548e-02
-4.42951351e-01 -3.15647572e-01 5.14950097e-01 5.62080681e-01
2.03849480e-01 -4.44075137e-01 -1.73918918e-01 4.82182980e-01
3.86548400e-01 -7.03684390e-02 1.88533336e-01 7.53443956e-01
-5.43386877e-01 -7.51250923e-01 -2.29811952e-01 3.68812382e-01
-4.40593362e-01 5.27671725e-02 -6.93246871e-02 7.28407741e-01
5.35896003e-01 7.98466086e-01 2.40044668e-01 -2.38332033e-01
4.18770351e-02 -3.47202748e-01 2.95211047e-01 -3.85234416e-01
-4.00596023e-01 6.55839920e-01 -4.71520722e-01 -8.79559219e-01
-7.93765247e-01 -7.96899676e-01 -1.66957390e+00 -1.17883086e-01
-5.36840975e-01 3.92666683e-02 1.87959015e-01 1.18460810e+00
2.95479268e-01 7.62660146e-01 6.93339288e-01 -1.17928720e+00
3.70766222e-02 -6.30571544e-01 -4.66693699e-01 5.79307079e-01
2.38434255e-01 -8.94474745e-01 -1.38045281e-01 2.87286729e-01] | [9.276195526123047, -0.24351035058498383] |
b6c72516-38f9-45cc-8509-31211a502ea0 | unitopatho-a-labeled-histopathological | 2101.09991 | null | https://arxiv.org/abs/2101.09991v2 | https://arxiv.org/pdf/2101.09991v2.pdf | UniToPatho, a labeled histopathological dataset for colorectal polyps classification and adenoma dysplasia grading | Histopathological characterization of colorectal polyps allows to tailor patients' management and follow up with the ultimate aim of avoiding or promptly detecting an invasive carcinoma. Colorectal polyps characterization relies on the histological analysis of tissue samples to determine the polyps malignancy and dysplasia grade. Deep neural networks achieve outstanding accuracy in medical patterns recognition, however they require large sets of annotated training images. We introduce UniToPatho, an annotated dataset of 9536 hematoxylin and eosin (H&E) stained patches extracted from 292 whole-slide images, meant for training deep neural networks for colorectal polyps classification and adenomas grading. We present our dataset and provide insights on how to tackle the problem of automatic colorectal polyps characterization. | ['Marco Grangetto', 'Paola Cassoni', 'Luca Bertero', 'Attilio Fiandrotti', 'Enzo Tartaglione', 'Daniele Perlo', 'Carlo Alberto Barbano'] | 2021-01-25 | null | null | null | null | ['histopathological-image-classification'] | ['medical'] | [ 3.54411185e-01 3.49813014e-01 -3.51891220e-01 -2.15717450e-01
-8.29604208e-01 -7.57027328e-01 9.08890292e-02 1.00151122e+00
-6.91348076e-01 3.23125809e-01 1.21728323e-01 -7.37855792e-01
6.03955947e-02 -1.13666224e+00 -4.39557910e-01 -8.93446386e-01
-4.95469570e-01 5.16184807e-01 2.69697666e-01 2.42953375e-01
-4.74543720e-02 5.25066793e-01 -7.38278985e-01 7.46076882e-01
6.10895932e-01 8.18051338e-01 3.70184369e-02 1.31521928e+00
3.15703303e-01 6.13563836e-01 -2.06346154e-01 -3.26115131e-01
1.58207133e-01 -6.88598752e-02 -6.99330807e-01 -2.14764744e-01
2.23199815e-01 -3.21753800e-01 -2.62272358e-01 1.19457185e+00
4.82316822e-01 -4.78581458e-01 9.04003143e-01 -3.36704887e-02
-6.36957228e-01 5.97456336e-01 -2.90397853e-01 5.82618356e-01
-8.84460881e-02 2.02246115e-01 9.09108162e-01 -5.34682333e-01
5.42917490e-01 1.17161259e-01 1.18025255e+00 4.96504158e-01
-9.05900359e-01 8.11979771e-02 -6.93321466e-01 -2.91102212e-02
-1.15900266e+00 -1.07073881e-01 -3.81462835e-02 -7.33409226e-01
7.13341713e-01 5.20942807e-01 1.33448672e+00 4.58132267e-01
6.17395878e-01 7.81293273e-01 6.59729123e-01 -3.70354563e-01
1.71476230e-01 1.16880007e-01 -1.76150631e-02 1.28428423e+00
8.73683333e-01 1.38129383e-01 5.06825030e-01 -2.21409872e-01
7.94926643e-01 1.98471665e-01 -1.81903958e-01 -7.18418509e-02
-1.30920792e+00 7.34825253e-01 1.15991437e+00 5.23447633e-01
-3.02156389e-01 3.34244370e-02 8.07801485e-01 7.84633979e-02
-1.46687269e-01 8.75778735e-01 -1.74700499e-01 6.51306093e-01
-8.14286768e-01 -3.40497494e-01 7.13314772e-01 4.05305713e-01
4.74518925e-01 -4.02444243e-01 -1.56847522e-01 6.89715564e-01
1.53595999e-01 3.13872844e-01 1.19401240e+00 -4.53786105e-01
-1.99590921e-01 1.09167194e+00 -1.93176717e-02 -9.74766731e-01
-7.64015853e-01 -3.77515256e-01 -1.37771034e+00 -9.09695961e-03
4.87329870e-01 -8.14624876e-02 -9.35882032e-01 8.40691686e-01
9.52160731e-02 -4.23423618e-01 4.22331877e-02 6.59772933e-01
1.12052715e+00 2.53224224e-01 6.92644790e-02 4.48182702e-01
1.81690538e+00 -9.60848212e-01 9.03513934e-03 -2.92959005e-01
9.38042104e-01 -4.72193152e-01 7.75758445e-01 3.25543195e-01
-9.37619865e-01 -1.47535652e-01 -1.25494063e+00 -3.43870610e-01
-5.95218658e-01 9.04513896e-01 8.11772883e-01 7.05804765e-01
-1.07555544e+00 2.25149393e-01 -1.17834449e+00 -2.65728712e-01
7.16809452e-01 7.23747492e-01 -6.69176519e-01 2.08662987e-01
-8.16301346e-01 6.26920104e-01 5.16641736e-01 3.32094252e-01
-8.16054940e-01 -7.34307110e-01 -9.88464057e-01 3.06259453e-01
-3.33003551e-01 -1.01231182e+00 1.33537555e+00 -9.30026472e-01
-1.21717048e+00 1.64011252e+00 2.38792256e-01 -8.42045128e-01
2.26759031e-01 6.55612648e-01 -8.32694322e-02 3.29324722e-01
-5.18075377e-02 5.85917234e-01 2.57913142e-01 -5.16035318e-01
-1.13014388e+00 -1.59604982e-01 -5.35554700e-02 -1.99275002e-01
-3.01078200e-01 -5.06474495e-01 -3.15316945e-01 -4.93652135e-01
-2.64338583e-01 -9.28382874e-01 -9.57536757e-01 3.79383743e-01
-5.76452851e-01 2.57388592e-01 -1.50286540e-01 -8.13185573e-01
1.01174915e+00 -2.28159213e+00 -3.61856371e-01 4.74250913e-01
7.88188219e-01 6.01603329e-01 1.56051638e-02 -2.89290231e-02
1.24379024e-01 3.76118809e-01 2.79662572e-02 1.82941034e-01
-4.76893410e-02 -1.26727611e-01 4.14758205e-01 9.36149180e-01
3.59690785e-01 1.16977012e+00 -1.01833296e+00 -7.10783958e-01
3.43386710e-01 3.06673765e-01 -4.95963007e-01 -1.95100792e-02
-4.08154130e-02 4.58443984e-02 -4.27117229e-01 8.48994792e-01
3.03677410e-01 -8.35060000e-01 2.58247524e-01 -1.54123068e-01
6.57578334e-02 1.47736564e-01 -4.12750900e-01 1.10903394e+00
-5.31909227e-01 9.39105868e-01 1.19811729e-01 -7.25454867e-01
5.48948705e-01 4.25550789e-01 2.70528883e-01 -4.01352376e-01
4.20848966e-01 6.12337947e-01 3.37978512e-01 -7.31872499e-01
2.03446984e-01 -3.42560597e-02 -1.21513560e-01 7.16137215e-02
-3.65502924e-01 4.47577052e-03 5.79267025e-01 -3.22035730e-01
1.41001940e+00 -9.49054182e-01 1.25588632e+00 -5.12360573e-01
7.11473525e-01 5.48806369e-01 2.29341939e-01 5.02969086e-01
-5.36944568e-01 3.91981602e-01 3.97765189e-01 -1.24734831e+00
-1.07353938e+00 -9.51544821e-01 -4.45079923e-01 4.80516762e-01
-8.19816291e-02 2.69625664e-01 -4.30013150e-01 -6.16421103e-01
-6.42737327e-03 -2.49363333e-01 -1.29762948e+00 1.81552589e-01
-8.16315174e-01 -1.13231409e+00 6.33775175e-01 5.49751937e-01
2.59220511e-01 -1.04749405e+00 -8.74323905e-01 1.35114387e-01
1.26348808e-01 -3.55515778e-01 -3.92222434e-01 3.57479036e-01
-6.39986098e-01 -1.66452050e+00 -1.02574372e+00 -1.60041499e+00
1.40057433e+00 -3.62352817e-04 1.16419804e+00 5.32188356e-01
-9.89763558e-01 -1.21921211e-01 -5.05863801e-02 -3.35154563e-01
-9.82133567e-01 3.86424899e-01 -4.44517285e-01 -3.21052849e-01
4.82373983e-01 -3.61526042e-01 -1.25288022e+00 -8.57775360e-02
-8.15364182e-01 -4.78988402e-02 1.33899307e+00 1.09520459e+00
1.00694978e+00 -9.92755517e-02 8.26496035e-02 -1.10632682e+00
4.98991787e-01 -3.92472774e-01 -6.20676816e-01 3.55271310e-01
-1.89252451e-01 -2.99118251e-01 9.86358404e-01 -2.82222390e-01
-5.23583233e-01 1.85729370e-01 -4.79922861e-01 3.74085456e-01
-2.15150595e-01 8.01991105e-01 7.49775589e-01 -3.31204504e-01
1.10104156e+00 9.01919305e-02 -2.15440646e-01 1.29412785e-01
-1.41039610e-01 5.34025967e-01 9.13335502e-01 2.65244007e-01
1.86290503e-01 6.73891544e-01 1.00557789e-01 -7.00067341e-01
-5.32315254e-01 -9.29076850e-01 -3.35877210e-01 2.58421600e-01
7.60719061e-01 -7.94700623e-01 -7.11680233e-01 -1.94659662e-02
-6.42629921e-01 -5.29338717e-01 -4.89660859e-01 4.24962759e-01
-2.32650533e-01 2.31317624e-01 -1.27587676e+00 3.29476260e-02
-7.96319664e-01 -9.19837534e-01 9.75458622e-01 4.27495629e-01
-3.79680693e-01 -1.43708563e+00 6.22738123e-01 -1.77550972e-01
4.91653115e-01 5.80666125e-01 1.10338783e+00 -7.64286697e-01
-2.61379272e-01 -1.04615307e+00 -3.34447116e-01 -2.96886042e-02
2.68111110e-01 4.05577838e-01 -7.95604169e-01 -2.16889784e-01
-3.12237263e-01 4.94037429e-03 1.05337226e+00 8.18496943e-01
1.08001387e+00 -6.06745780e-01 -7.96319902e-01 1.09318328e+00
1.63268566e+00 1.48968518e-01 4.29083973e-01 4.55156237e-01
2.09262908e-01 4.15267318e-01 2.05861386e-02 2.32648447e-01
1.69209316e-01 -2.01608196e-01 5.59281290e-01 -5.95209420e-01
-2.03523949e-01 -2.22943723e-02 -2.53288269e-01 6.60675466e-01
1.12317793e-01 -2.64587253e-01 -1.14942086e+00 1.17963314e+00
-1.09791529e+00 -7.00230062e-01 -1.43659189e-01 1.78146017e+00
9.39567685e-01 -1.69696346e-01 -6.45221397e-02 1.78707346e-01
6.38080180e-01 -4.19105053e-01 -2.85098910e-01 -1.95797175e-01
-5.34524322e-02 8.68602544e-02 7.32011616e-01 4.76385236e-01
-1.47154093e+00 1.12150423e-01 7.05666399e+00 5.97609520e-01
-1.57811260e+00 -2.94825971e-01 1.17409694e+00 2.49207795e-01
-2.90930066e-02 -6.34552956e-01 -4.33497906e-01 1.70403287e-01
4.52374309e-01 -3.58932436e-01 -8.22482631e-02 7.60088563e-01
-1.90040916e-01 -1.70123473e-01 -1.22462273e+00 3.98886055e-01
-4.20953900e-01 -1.83376765e+00 -1.59882978e-01 1.58840820e-01
1.06160867e+00 5.07206857e-01 2.44956121e-01 3.00761908e-02
4.61096197e-01 -9.46131468e-01 -1.84680447e-01 4.31625783e-01
1.11021948e+00 -1.91479728e-01 1.31205547e+00 6.88107610e-02
-1.08686292e+00 -1.22662596e-01 -4.43782002e-01 3.50361049e-01
-4.04928356e-01 7.38511384e-01 -1.86751688e+00 -4.34001237e-02
5.06302655e-01 6.43952131e-01 -8.44196200e-01 1.54629576e+00
-4.82477508e-02 4.57837671e-01 -3.94664615e-01 -4.45100516e-01
4.25190508e-01 2.69900560e-01 1.88831136e-01 1.70549238e+00
5.56725085e-01 -2.98027217e-01 -6.89161569e-02 5.63327968e-01
6.93792105e-02 4.02844131e-01 -1.18435398e-01 -2.19199285e-01
1.57072827e-01 1.70422053e+00 -1.27126491e+00 -4.98271286e-01
-2.03247949e-01 3.81186008e-01 1.93893403e-01 -8.86658281e-02
-3.06382269e-01 -5.07247090e-01 -9.66489390e-02 2.62417823e-01
2.05272943e-01 4.56141859e-01 -4.05009240e-01 -8.70885253e-01
-3.08942527e-01 -5.85384667e-01 6.72524810e-01 -1.06600985e-01
-1.30831647e+00 6.61893189e-01 -7.75304258e-01 -1.29974759e+00
-2.69781470e-01 -1.20734191e+00 -1.00185549e+00 4.48602349e-01
-1.62551689e+00 -1.12688839e+00 -5.21481693e-01 -1.04081715e-02
1.05597347e-01 -5.40378056e-02 1.24935102e+00 3.06703061e-01
-3.87634188e-01 5.40410280e-01 2.97690123e-01 5.55088282e-01
6.38152480e-01 -1.84431016e+00 3.59527349e-01 3.88301462e-01
-3.78956705e-01 4.48238641e-01 3.92655462e-01 -4.21779335e-01
-1.04122281e+00 -1.35082424e+00 9.42215502e-01 -4.57156822e-02
8.41917098e-01 3.18370551e-01 -3.66708606e-01 7.17145085e-01
3.33121605e-02 4.10327911e-01 1.33097422e+00 -4.84860897e-01
6.04481883e-02 1.07992187e-01 -1.28262222e+00 7.77099431e-01
5.30886054e-02 -2.43483186e-01 -9.63478610e-02 5.83122671e-01
1.16476782e-01 -4.88721400e-01 -1.12052071e+00 4.79179710e-01
7.06362605e-01 -9.29924428e-01 9.92035151e-01 -9.93808806e-02
6.23118579e-01 -1.53610453e-01 2.95349389e-01 -1.06007612e+00
-7.25297749e-01 -1.67827815e-01 4.88386929e-01 -4.19435166e-02
7.41684854e-01 -4.57868725e-01 1.23766851e+00 1.48718253e-01
-2.42558882e-01 -1.07704365e+00 -6.31787121e-01 1.00508966e-01
1.32199243e-01 5.40816069e-01 4.29331720e-01 7.75404811e-01
3.19637895e-01 -3.91441107e-01 5.39790571e-01 2.21012980e-01
5.84152639e-02 3.51088457e-02 4.66721117e-01 -9.13502812e-01
-4.51368183e-01 -1.10915828e+00 -6.32049859e-01 -5.98186791e-01
-5.92374504e-01 -1.03641307e+00 1.55205503e-01 -1.65943694e+00
2.00067699e-01 -4.22172606e-01 -2.53309190e-01 3.76528263e-01
-3.41439456e-01 8.08766305e-01 -7.29065895e-01 2.01316997e-01
-5.35757124e-01 -3.47532332e-01 1.33194792e+00 -5.74054599e-01
-7.86073580e-02 4.70026284e-01 -6.85892701e-01 9.08463180e-01
7.98108697e-01 -2.92211533e-01 2.12819636e-01 -3.57301831e-01
6.98000729e-01 3.93447056e-02 3.88648987e-01 -1.03763366e+00
3.61261815e-01 -4.08712476e-02 6.65346980e-01 -4.82481480e-01
-1.68680444e-01 -6.97241008e-01 2.13533845e-02 1.41355431e+00
-6.85437322e-01 -2.09888175e-01 2.25572452e-01 4.16250348e-01
-4.73229766e-01 -5.05304635e-01 8.60864699e-01 -5.71612120e-01
-3.29919517e-01 4.04340178e-01 -1.03734303e+00 -4.29020464e-01
1.08703494e+00 -2.24177271e-01 -5.59674382e-01 1.73134953e-01
-1.02729249e+00 1.12261295e-01 6.40448451e-01 -4.94204044e-01
3.70524257e-01 -1.15886080e+00 -9.74208057e-01 2.95120507e-01
3.45516890e-01 3.31612557e-01 5.19117713e-01 1.06220686e+00
-1.68649650e+00 6.28226221e-01 -1.90822527e-01 -5.80225945e-01
-1.24453068e+00 5.75900018e-01 9.24201071e-01 -1.04357040e+00
-5.15731812e-01 1.20361304e+00 3.38075161e-01 -2.35567585e-01
-1.69760566e-02 -9.03797209e-01 -4.36210692e-01 -3.06722224e-01
7.12069809e-01 1.11914411e-01 2.12895468e-01 4.00617309e-02
-2.44172029e-02 1.15142003e-01 -2.77398258e-01 5.61558723e-01
1.08804846e+00 3.73531491e-01 -3.86440307e-01 -5.76314935e-03
1.09760094e+00 2.94360965e-01 -6.30680621e-01 -7.36845136e-02
-4.24724519e-02 -1.00560412e-01 -1.64190724e-01 -8.21583331e-01
-1.00540686e+00 7.04373300e-01 5.66894650e-01 6.19550526e-01
1.04328108e+00 -2.76963204e-01 7.66720593e-01 6.91894412e-01
-1.28428265e-01 -5.39503157e-01 -2.90272266e-01 1.78811569e-02
5.45451939e-01 -1.43459034e+00 6.60611391e-02 -3.64162982e-01
-3.98656018e-02 1.47064185e+00 2.32156530e-01 -7.64183879e-01
7.19068944e-01 5.91091990e-01 3.66982371e-01 -3.09777349e-01
-6.51040077e-01 2.09684297e-01 5.14359713e-01 4.71651345e-01
6.78397477e-01 5.35289288e-01 1.51422163e-02 8.33512425e-01
-3.75934631e-01 5.94428256e-02 3.94570649e-01 7.52043128e-01
-6.83668971e-01 -6.43800557e-01 -7.86814839e-02 1.00581622e+00
-8.68496597e-01 -2.01283544e-01 -4.87832487e-01 8.12545836e-01
1.62212908e-01 1.67511657e-01 1.49467111e-01 7.96921700e-02
-1.21941127e-01 -6.96333289e-01 3.92006248e-01 -6.26506031e-01
-1.19136739e+00 1.18587047e-01 2.12135628e-01 2.48095736e-01
-2.07305714e-01 -4.67315942e-01 -9.11201179e-01 -8.30158964e-02
-2.15329066e-01 9.97686088e-02 3.48184228e-01 5.69031060e-01
1.37628168e-02 6.43267393e-01 3.91995281e-01 -6.89430237e-01
-2.89565265e-01 -8.66737068e-01 -5.79545915e-01 8.00645649e-02
8.58134031e-01 2.69337028e-01 -4.30297256e-01 3.12462687e-01] | [15.072669982910156, -2.9896798133850098] |
999027c3-95d2-4620-9552-a8878ace0627 | cris-clip-driven-referring-image-segmentation | 2111.15174 | null | https://arxiv.org/abs/2111.15174v2 | https://arxiv.org/pdf/2111.15174v2.pdf | CRIS: CLIP-Driven Referring Image Segmentation | Referring image segmentation aims to segment a referent via a natural linguistic expression.Due to the distinct data properties between text and image, it is challenging for a network to well align text and pixel-level features. Existing approaches use pretrained models to facilitate learning, yet separately transfer the language/vision knowledge from pretrained models, ignoring the multi-modal corresponding information. Inspired by the recent advance in Contrastive Language-Image Pretraining (CLIP), in this paper, we propose an end-to-end CLIP-Driven Referring Image Segmentation framework (CRIS). To transfer the multi-modal knowledge effectively, CRIS resorts to vision-language decoding and contrastive learning for achieving the text-to-pixel alignment. More specifically, we design a vision-language decoder to propagate fine-grained semantic information from textual representations to each pixel-level activation, which promotes consistency between the two modalities. In addition, we present text-to-pixel contrastive learning to explicitly enforce the text feature similar to the related pixel-level features and dissimilar to the irrelevances. The experimental results on three benchmark datasets demonstrate that our proposed framework significantly outperforms the state-of-the-art performance without any post-processing. The code will be released. | ['Tongliang Liu', 'Mingming Gong', 'Yandong Guo', 'Xunqiang Tao', 'Qiang Li', 'Yu Lu', 'Zhaoqing Wang'] | 2021-11-30 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Wang_CRIS_CLIP-Driven_Referring_Image_Segmentation_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Wang_CRIS_CLIP-Driven_Referring_Image_Segmentation_CVPR_2022_paper.pdf | cvpr-2022-1 | ['generalized-referring-expression-segmentation', 'referring-expression-segmentation'] | ['computer-vision', 'computer-vision'] | [ 7.03186572e-01 6.45191073e-02 -3.81705076e-01 -6.22380197e-01
-1.13566053e+00 -3.54178011e-01 5.59870899e-01 -1.18708834e-01
-4.70809817e-01 3.74568939e-01 1.65958479e-01 -1.42889842e-01
3.22406411e-01 -5.15906096e-01 -1.06579542e+00 -6.31713033e-01
7.28144646e-01 1.34155631e-01 3.18321407e-01 -3.88746299e-02
3.70974869e-01 4.84769307e-02 -1.27087665e+00 6.56007290e-01
1.00401950e+00 1.14872158e+00 5.46883345e-01 2.40578309e-01
-6.22132659e-01 1.02244496e+00 -4.88251634e-02 -4.87951428e-01
1.14364713e-01 -7.45989740e-01 -9.13109541e-01 4.69560862e-01
8.67271781e-01 -1.74395159e-01 -1.48590580e-01 1.38584900e+00
1.73087001e-01 -5.63541055e-02 5.64938247e-01 -1.06497526e+00
-1.22294617e+00 4.92008805e-01 -1.02323210e+00 1.47081643e-01
2.76265085e-01 1.80677444e-01 1.07447183e+00 -8.85796905e-01
5.10711253e-01 1.34571874e+00 3.78754139e-01 4.48427826e-01
-1.02992964e+00 -3.21101010e-01 6.26811922e-01 1.31438851e-01
-1.37911069e+00 -3.73513490e-01 1.17020369e+00 -4.12251353e-01
6.13414347e-01 1.59114283e-02 4.21422809e-01 9.02591407e-01
-3.54806744e-02 1.19909406e+00 1.28265476e+00 -6.11357093e-01
-9.52041894e-02 9.84513164e-02 2.68204114e-03 9.50434744e-01
-1.45552307e-01 -9.99464840e-02 -4.46286768e-01 4.13991183e-01
9.10733104e-01 1.13681063e-01 -2.15025797e-01 -2.54133165e-01
-1.21157527e+00 5.75801671e-01 6.85694635e-01 4.48836863e-01
-4.97097671e-01 3.56769353e-01 4.07359034e-01 -1.40907720e-01
2.89697498e-01 -1.45810291e-01 -3.03383350e-01 2.90294170e-01
-1.17035306e+00 -2.51300991e-01 3.17404717e-01 1.14763188e+00
9.62479055e-01 -8.49084742e-03 -5.40123820e-01 8.07735264e-01
5.12108624e-01 4.74751920e-01 4.11628485e-01 -1.03173316e+00
6.91722095e-01 6.10819757e-01 -2.71606028e-01 -9.31680560e-01
-2.85869408e-02 -3.63556743e-01 -9.16094780e-01 -2.21095771e-01
3.41069043e-01 1.37572482e-01 -1.04594970e+00 1.70513213e+00
1.34349301e-01 1.51338220e-01 7.04159439e-02 1.15758657e+00
8.87958169e-01 7.40941703e-01 4.32154626e-01 -1.82570651e-01
1.48327196e+00 -1.36667192e+00 -7.20138669e-01 -6.94795310e-01
3.50292027e-01 -8.92466307e-01 1.48554039e+00 -4.10355926e-02
-1.34361160e+00 -7.49239624e-01 -8.03020656e-01 -6.23611331e-01
-3.11437845e-01 1.66920483e-01 3.63875240e-01 3.49139333e-01
-1.00010824e+00 -6.24040850e-02 -6.22527242e-01 -4.53940809e-01
8.18178296e-01 1.00321785e-01 -6.64073378e-02 -2.13406414e-01
-1.08488905e+00 6.88246548e-01 6.36107326e-01 1.96452022e-01
-6.88820481e-01 -4.49749142e-01 -9.09614742e-01 -1.16862431e-01
4.89094079e-01 -9.96685088e-01 1.07948077e+00 -1.68017685e+00
-1.43106425e+00 1.40313172e+00 -4.56650317e-01 -1.68195277e-01
6.23266697e-01 -6.89835176e-02 -5.51848747e-02 4.08661187e-01
4.22569573e-01 1.21787250e+00 9.28974688e-01 -1.65317297e+00
-8.13288629e-01 -3.10915500e-01 2.04015896e-01 4.33727205e-01
-2.42189601e-01 1.45961657e-01 -1.22857428e+00 -6.72073424e-01
2.87794143e-01 -6.00564778e-01 1.10001519e-01 2.39150092e-01
-6.61697686e-01 -2.39050016e-01 7.28980005e-01 -6.16702139e-01
8.41610968e-01 -2.04696131e+00 1.36817530e-01 -1.73425853e-01
-2.29356289e-02 1.36297308e-02 -3.52226734e-01 3.36298198e-02
-2.96843313e-02 -3.68680954e-02 -4.96306449e-01 -3.80086929e-01
3.21218707e-02 2.58613288e-01 -2.88683951e-01 4.73292530e-01
3.75674576e-01 1.26008034e+00 -8.20254683e-01 -1.10073829e+00
3.68429065e-01 4.31119591e-01 -4.46858168e-01 1.44009516e-01
-4.64893013e-01 5.99744618e-01 -7.28122950e-01 8.64432693e-01
6.64060533e-01 -3.10614645e-01 -2.37999961e-01 -6.90783739e-01
-9.88268554e-02 -8.98487587e-03 -6.71273410e-01 2.02905631e+00
-5.06816983e-01 5.30214489e-01 1.42371804e-01 -1.36525214e+00
8.44192743e-01 -5.37425540e-02 3.21802288e-01 -1.17999542e+00
3.15306634e-01 1.10899203e-01 -2.91599363e-01 -7.11353004e-01
2.82464415e-01 -2.13179603e-01 -1.98314905e-01 1.79114357e-01
-5.09957187e-02 -2.48846799e-01 1.41212434e-01 6.72650412e-02
2.07717076e-01 3.62604588e-01 3.15837450e-02 -1.39944956e-01
9.69661891e-01 8.87733549e-02 5.25850534e-01 5.63905358e-01
-3.80817473e-01 7.41909564e-01 4.23614651e-01 -4.38996963e-02
-7.62662709e-01 -1.06236100e+00 -1.14181796e-02 1.33899009e+00
5.68243980e-01 1.35139763e-01 -1.01442194e+00 -5.94530404e-01
-3.66592348e-01 7.98085690e-01 -6.00848913e-01 -4.37760800e-02
-4.78623658e-01 -3.98963869e-01 5.04708171e-01 6.24038637e-01
1.06016850e+00 -1.06530952e+00 -3.34598064e-01 -4.65392694e-02
-4.08690810e-01 -1.48425138e+00 -1.05468309e+00 1.04438327e-01
-6.73387110e-01 -6.96480751e-01 -8.64644349e-01 -1.56459367e+00
8.66405666e-01 3.75493258e-01 1.00716829e+00 -7.49781588e-03
5.64068463e-03 5.58828592e-01 -2.01921195e-01 -7.59586841e-02
-1.48262993e-01 -4.21455838e-02 -5.95340908e-01 1.78478032e-01
3.35287422e-01 -2.79144645e-01 -6.10853553e-01 1.46483436e-01
-1.10758960e+00 5.70174336e-01 7.48294652e-01 8.85863662e-01
1.17512071e+00 -1.07905567e-01 3.85955900e-01 -8.33501101e-01
3.47864389e-01 -1.34470791e-01 -4.02209699e-01 5.91910124e-01
-2.57330060e-01 2.10522115e-02 5.83357930e-01 -2.44317055e-01
-1.19426119e+00 3.59781474e-01 -4.96677170e-03 -6.11019373e-01
-3.21564376e-01 5.04616022e-01 -5.35870492e-01 8.64540711e-02
1.37247816e-01 7.08570063e-01 -3.14238191e-01 -4.08999920e-02
9.40131724e-01 7.13045180e-01 9.91133690e-01 -9.11444843e-01
5.22692382e-01 6.94052160e-01 -3.31105858e-01 -6.49163902e-01
-1.36249018e+00 -4.17906463e-01 -9.37073827e-01 -1.68925524e-01
1.42120516e+00 -1.02599955e+00 -4.01594400e-01 4.40489143e-01
-1.35549784e+00 -4.46190745e-01 -1.72770381e-01 2.12901399e-01
-9.12892938e-01 5.21311820e-01 -4.66807097e-01 -4.31045771e-01
-3.99050832e-01 -1.29934132e+00 1.31721985e+00 4.36188042e-01
3.26108754e-01 -1.09438515e+00 -4.55866426e-01 7.86501825e-01
1.46308169e-01 9.47989076e-02 8.31542253e-01 -2.46743098e-01
-6.81097627e-01 1.11201808e-01 -9.82547045e-01 3.73869449e-01
9.88686681e-02 -3.89973982e-03 -9.80372250e-01 2.12441734e-03
5.50921017e-04 -4.62208182e-01 1.04290938e+00 5.43306649e-01
1.36535907e+00 -1.03001133e-01 -1.71588168e-01 7.23484337e-01
1.49642825e+00 1.41941300e-02 6.74397767e-01 4.18719649e-01
1.14333379e+00 7.33705163e-01 6.88362002e-01 -2.70932280e-02
8.77926648e-01 4.83318180e-01 3.72280389e-01 -5.15325785e-01
-4.12576079e-01 -4.82653916e-01 3.27977955e-01 7.51831532e-01
3.56895000e-01 -8.03545564e-02 -7.80189037e-01 5.62866926e-01
-1.81148970e+00 -9.55683649e-01 -2.34992132e-01 1.74593651e+00
1.11563861e+00 1.81724355e-01 -1.27501264e-01 -3.41000944e-01
8.69929969e-01 2.65271842e-01 -7.60502756e-01 -3.31341505e-01
-2.94155598e-01 -2.02129453e-01 5.63091874e-01 4.96916622e-01
-1.25672376e+00 1.49132669e+00 5.15957832e+00 9.96162057e-01
-1.32032597e+00 2.65198439e-01 9.27349150e-01 2.34233126e-01
-4.11510617e-01 -1.75310788e-03 -6.33894086e-01 3.68972600e-01
4.39635724e-01 2.02696696e-02 3.52028102e-01 5.25515974e-01
3.98383617e-01 -2.18689144e-01 -1.15993297e+00 1.27732360e+00
4.62166131e-01 -1.26393890e+00 3.00236106e-01 -4.01287258e-01
1.00849915e+00 -6.04064651e-02 3.11022997e-01 2.18553394e-01
-1.14050515e-01 -7.54729211e-01 1.10626280e+00 6.42377257e-01
8.51635039e-01 -5.59922218e-01 4.78760749e-01 2.29065910e-01
-1.38309383e+00 1.63595706e-01 -2.38752857e-01 2.89485306e-01
2.29900748e-01 3.29278857e-01 -2.73351669e-01 4.97199595e-01
6.49374247e-01 1.04019046e+00 -6.44146919e-01 6.11038566e-01
-3.13872308e-01 3.26684535e-01 -1.88253950e-02 1.73662543e-01
6.58375442e-01 -2.82591432e-01 8.85085240e-02 1.35306752e+00
-9.11352262e-02 -6.40884042e-02 4.18473125e-01 1.48573649e+00
-3.52889895e-01 1.09339990e-01 -2.25091845e-01 -5.55592105e-02
3.48429233e-01 1.14596510e+00 -8.11614633e-01 -4.73555088e-01
-8.48414302e-01 1.54763770e+00 2.49697343e-01 5.64013541e-01
-9.91952777e-01 -2.70167708e-01 9.50828474e-03 -9.35107023e-02
4.44351971e-01 -7.62340128e-02 -6.03182614e-01 -9.81604159e-01
1.24258704e-01 -6.38407886e-01 3.02525848e-01 -1.17487490e+00
-1.47550142e+00 4.42833722e-01 -1.44309387e-01 -1.17739332e+00
1.87858090e-01 -4.58599895e-01 -6.45391881e-01 1.08889556e+00
-2.14142561e+00 -1.77418721e+00 -3.05346847e-01 9.73462105e-01
7.99044549e-01 1.76704481e-01 3.14151645e-01 2.90855497e-01
-7.09809244e-01 6.40693426e-01 -1.46375209e-01 4.48831052e-01
7.20341027e-01 -1.06463814e+00 -2.92688105e-02 9.43550348e-01
5.35757877e-02 5.61905861e-01 3.43300313e-01 -4.78041619e-01
-1.42306471e+00 -1.47445679e+00 6.65207565e-01 -1.62774637e-01
7.40915120e-01 -9.35840085e-02 -9.23753917e-01 7.39464760e-01
5.98158717e-01 2.25081906e-01 2.87498713e-01 -3.80620629e-01
-4.09642875e-01 -1.13031551e-01 -9.44922686e-01 6.01095438e-01
8.65650058e-01 -9.33521092e-01 -9.26513672e-01 3.27884912e-01
7.94271171e-01 -5.22868216e-01 -5.91586411e-01 1.16945826e-01
2.14103639e-01 -8.14455509e-01 9.59085524e-01 -1.94448441e-01
6.70306504e-01 -5.26786983e-01 -3.78738821e-01 -6.49150014e-01
-2.92007755e-02 -2.03331754e-01 4.20321703e-01 1.67535925e+00
2.04067990e-01 -2.62950689e-01 5.91005981e-01 5.96435666e-01
-3.20209503e-01 -5.87068319e-01 -8.28181267e-01 -5.09752452e-01
3.86073917e-01 -4.60175902e-01 2.12431028e-02 1.04307568e+00
-2.16389805e-01 5.16421616e-01 -2.25676686e-01 2.74723053e-01
6.27422988e-01 5.01366973e-01 4.46726590e-01 -7.47705042e-01
-6.79689124e-02 -7.44667590e-01 -1.12839580e-01 -1.54912376e+00
5.29330313e-01 -1.18622136e+00 3.44451964e-01 -1.87387717e+00
4.43584532e-01 -2.28434429e-01 -3.03956896e-01 5.51266015e-01
-3.27933460e-01 3.39373708e-01 2.10449338e-01 1.76886156e-01
-1.04819930e+00 6.64254427e-01 1.81660521e+00 -5.38821042e-01
-1.29541636e-01 -5.05721092e-01 -8.69190812e-01 8.45426500e-01
5.97813845e-01 -2.98862219e-01 -4.19244379e-01 -8.51726592e-01
-1.14096142e-01 1.33205894e-02 6.37381911e-01 -6.67664051e-01
4.33246255e-01 -3.08765590e-01 2.53739089e-01 -9.05890942e-01
1.34197325e-01 -8.75298142e-01 -6.01348400e-01 1.74886674e-01
-6.31453037e-01 -3.15671831e-01 2.51373678e-01 6.38706028e-01
-4.98500168e-01 -2.14896381e-01 9.16653216e-01 -1.26681179e-01
-1.01681507e+00 2.83644557e-01 -8.78695995e-02 4.24545050e-01
9.59460795e-01 -3.75723183e-01 -3.17338437e-01 -1.40186682e-01
-4.61114973e-01 4.24510688e-01 5.54965913e-01 3.93528581e-01
7.29105353e-01 -1.16853690e+00 -6.27232015e-01 5.21709137e-02
3.06649745e-01 9.62404907e-02 4.73421872e-01 1.19783211e+00
-3.82676303e-01 4.65853363e-01 -2.64024436e-01 -1.00848699e+00
-9.61050391e-01 6.79772377e-01 6.44269824e-01 1.86376184e-01
-5.43583333e-01 9.05434251e-01 6.03633940e-01 -2.98733681e-01
3.38327825e-01 -5.05102336e-01 5.23822680e-02 -5.92226088e-02
2.65222400e-01 -2.03353778e-01 -3.03542674e-01 -9.68350351e-01
-3.02342862e-01 1.19089305e+00 -2.14991421e-01 -7.24874511e-02
8.34395826e-01 -6.78573430e-01 -2.54319727e-01 4.97020692e-01
1.34243679e+00 -3.01565617e-01 -1.43065059e+00 -6.40963495e-01
-5.00112996e-02 -2.97943890e-01 4.13162291e-01 -6.86694801e-01
-1.49435914e+00 1.24250066e+00 6.17678225e-01 -1.33853734e-01
1.39992857e+00 2.37641931e-01 9.12936509e-01 1.58890039e-01
7.44038727e-03 -1.14509892e+00 2.91003346e-01 4.71111774e-01
7.23136961e-01 -1.43238294e+00 -2.30628386e-01 -4.76292104e-01
-9.77383435e-01 1.04876542e+00 7.61786819e-01 -6.96207955e-02
2.49760285e-01 3.77573655e-04 2.32170194e-01 -4.96031530e-02
-3.63865912e-01 -5.90310216e-01 4.17174071e-01 6.74497545e-01
5.71674228e-01 -2.65483439e-01 -1.41459689e-01 5.97615004e-01
3.01384389e-01 1.08645633e-01 2.22173601e-01 6.19964004e-01
-4.89785403e-01 -8.52987349e-01 -3.86855692e-01 2.19229653e-01
-2.85010815e-01 -3.83204341e-01 -3.52858722e-01 6.18737221e-01
2.76661098e-01 9.44673479e-01 2.34299079e-01 1.03114024e-01
3.72454405e-01 -6.59065694e-02 5.21449864e-01 -4.57062662e-01
-3.51823866e-01 5.15393257e-01 -3.05612922e-01 -3.90007913e-01
-9.45301473e-01 -6.45526946e-01 -1.86494291e+00 8.57208371e-02
-3.15373950e-02 -2.36573592e-01 3.85498494e-01 1.09840298e+00
1.89689875e-01 8.11025739e-01 4.37974006e-01 -8.31880510e-01
-4.84858863e-02 -5.92880487e-01 -4.17308778e-01 4.96833801e-01
2.33662978e-01 -4.23176140e-01 -4.73047458e-02 6.54754817e-01] | [10.274991989135742, 1.1622203588485718] |
e5232cde-ef16-42db-832d-65f9680e4ac0 | adaptive-and-dynamically-constrained-process | 1909.07921 | null | https://arxiv.org/abs/1909.07921v4 | https://arxiv.org/pdf/1909.07921v4.pdf | Adaptive and Dynamically Constrained Process Noise Estimation for Orbit Determination | This paper introduces two new algorithms to accurately estimate the process noise covariance of a discrete-time Kalman filter online for robust orbit determination in the presence of dynamics model uncertainties. Common orbit determination process noise techniques, such as state noise compensation and dynamic model compensation, require offline tuning and a priori knowledge of the dynamical environment. Alternatively, the process noise covariance can be estimated through adaptive filtering. However, many adaptive filtering techniques are not applicable to onboard orbit determination due to computational cost or the assumption of a linear time-invariant system. Furthermore, existing adaptive filtering techniques do not constrain the process noise covariance according to the underlying continuous-time dynamical model, and there has been limited work on adaptive filtering with colored process noise. To overcome these limitations, a novel approach is developed which optimally fuses state noise compensation and dynamic model compensation with covariance matching adaptive filtering. This yields two adaptive and dynamically constrained process noise covariance estimation techniques. Unlike many adaptive filtering approaches, the new techniques accurately extrapolate over measurement outages and do not rely on ad hoc methods to ensure the process noise covariance is positive semi-definite. The benefits of the proposed algorithms are demonstrated through two case studies: an illustrative linear system and the autonomous navigation of two spacecraft orbiting an asteroid. | ["Simone D'Amico", 'Nathan Stacey'] | 2019-09-17 | null | null | null | null | ['noise-estimation'] | ['medical'] | [ 7.21248612e-02 -1.81597665e-01 1.90115109e-01 2.21683159e-01
-1.17086388e-01 -7.48633504e-01 5.84549904e-01 -2.24021330e-01
-3.02065074e-01 8.31352353e-01 -2.12844491e-01 -5.12944579e-01
-8.52189124e-01 -4.86341923e-01 -2.30271041e-01 -8.98309052e-01
-1.08937211e-01 4.43575382e-01 -1.15095926e-02 -1.28467754e-01
7.34285265e-02 6.90935373e-01 -1.39616787e+00 -9.04761016e-01
9.98708487e-01 9.71162736e-01 1.98204499e-02 6.95489109e-01
2.65896082e-01 6.12887815e-02 -5.49146533e-01 3.65966022e-01
7.97510862e-01 -3.54569405e-01 1.14014558e-02 2.07854703e-01
-1.56170920e-01 -9.68129262e-02 -5.23622096e-01 1.43779922e+00
4.61059004e-01 5.48955739e-01 3.81261349e-01 -1.12381518e+00
1.67924181e-01 -6.99300843e-04 5.01866303e-02 1.93319842e-01
2.41429172e-02 3.69448215e-01 2.28995442e-01 -6.42012954e-01
4.19175476e-01 1.19763923e+00 8.56589675e-01 2.48466894e-01
-1.34101200e+00 -7.62475014e-01 1.21352293e-01 -1.31568477e-01
-1.46602499e+00 -5.64427137e-01 3.64269078e-01 -6.10220313e-01
4.53288615e-01 3.03434968e-01 9.06677008e-01 3.49926025e-01
5.20278990e-01 1.32490441e-01 1.26674736e+00 -1.14596598e-01
3.06650639e-01 -9.19592530e-02 1.98621914e-01 3.03078771e-01
8.00793529e-01 6.59137428e-01 -1.60771936e-01 -4.40976799e-01
7.36536860e-01 -1.74640119e-01 -5.21703899e-01 -5.45378327e-01
-1.18374491e+00 4.15840924e-01 -9.51212421e-02 6.85957968e-02
-5.90633035e-01 8.57051685e-02 2.96701014e-01 6.97199523e-01
1.69292167e-01 5.64930499e-01 -3.77699941e-01 -1.96476355e-01
-7.34602630e-01 4.31551307e-01 1.10071802e+00 9.89439130e-01
6.55321479e-01 8.30451608e-01 2.92248596e-02 -2.85878521e-03
8.23560774e-01 1.26468754e+00 2.68887937e-01 -8.77980173e-01
7.24136606e-02 2.02116892e-01 8.51975024e-01 -8.44725668e-01
-6.25979364e-01 -7.45197594e-01 -6.97098255e-01 5.43102741e-01
5.32565951e-01 -7.63683379e-01 -9.80504274e-01 1.34818792e+00
5.49680531e-01 1.73723042e-01 3.76694918e-01 8.19900870e-01
-1.75028563e-01 5.76019883e-01 -3.88528198e-01 -8.12794745e-01
1.06043422e+00 -3.25508475e-01 -1.67584705e+00 -5.39403319e-01
1.80783436e-01 -9.68210638e-01 2.71111041e-01 5.23186505e-01
-8.08601201e-01 -4.16950196e-01 -1.40127230e+00 7.50268400e-01
-2.25598067e-02 2.00120777e-01 2.41418511e-01 9.70994771e-01
-7.90373147e-01 6.27160728e-01 -1.08914006e+00 -2.32905790e-01
-5.93468547e-01 4.83943850e-01 -2.97805592e-02 5.11043668e-01
-1.38600254e+00 1.13690436e+00 6.70077384e-01 7.92168140e-01
-5.41513979e-01 -4.99665022e-01 -6.13191366e-01 -1.87014580e-01
5.99785626e-01 -9.13551569e-01 1.21738493e+00 -7.86407232e-01
-2.01095748e+00 -2.59455800e-01 -3.39280903e-01 -7.15609908e-01
6.30491257e-01 -5.24181128e-01 -7.86225736e-01 -1.44767612e-01
-3.28792632e-01 -5.43638051e-01 1.39748406e+00 -8.40425372e-01
-6.09431028e-01 6.45657023e-03 -3.90028328e-01 3.59709948e-01
1.49922103e-01 -3.27016115e-01 -2.35055834e-01 -4.15246189e-01
8.81041229e-01 -1.37040913e+00 -6.08213365e-01 -6.27177507e-02
4.37763846e-03 5.21311820e-01 1.07850075e+00 -3.81700039e-01
1.31772244e+00 -2.17380238e+00 3.28211710e-02 7.97085524e-01
-2.74746507e-01 2.06099451e-01 3.54036540e-01 5.48987210e-01
4.35352921e-02 -3.71544331e-01 -1.45236244e-02 1.80761199e-02
1.62124932e-02 1.34066701e-01 -5.66493094e-01 9.40376699e-01
5.72474487e-02 8.90113637e-02 -8.62759709e-01 3.10601383e-01
3.94954503e-01 1.76926911e-01 -1.29350170e-01 -2.29103252e-01
8.62488523e-02 6.79290414e-01 -4.42861348e-01 3.96099389e-01
7.81509340e-01 1.43346548e-01 2.81171620e-01 -3.60931307e-01
-5.07574677e-01 6.56441748e-02 -2.07848525e+00 9.22606945e-01
-2.77265668e-01 5.68241179e-01 7.68159449e-01 -4.46787834e-01
9.33369577e-01 3.57295513e-01 4.64944541e-01 -1.12540022e-01
3.70266587e-01 4.35722023e-01 3.71331751e-01 -1.35015041e-01
7.10268259e-01 -5.64563811e-01 1.23548999e-01 1.27439946e-01
-1.66467860e-01 -5.26022792e-01 1.46107838e-01 -1.20835267e-01
8.64867389e-01 -2.00079456e-02 5.48254251e-01 -5.89242160e-01
7.54525602e-01 5.25395647e-02 1.18144715e+00 6.59730196e-01
-2.69740582e-01 -5.39581105e-02 3.30299847e-02 2.37465203e-02
-8.72555017e-01 -6.37976050e-01 -1.63468927e-01 1.37080147e-03
3.58844727e-01 -3.37745339e-01 -1.30379334e-01 1.78419496e-03
2.62734443e-01 4.88847315e-01 -3.24447572e-01 -2.32414484e-01
-2.69268543e-01 -9.07041490e-01 1.63274288e-01 1.02035090e-01
4.73951817e-01 1.49628878e-01 -2.79382050e-01 6.81412160e-01
3.18206906e-01 -8.33671808e-01 -2.22904220e-01 1.23572044e-01
-1.10269785e+00 -1.08754444e+00 -7.12585151e-02 -1.60833120e-01
7.69390404e-01 1.49564788e-01 1.43121228e-01 3.11597232e-02
2.37397835e-01 6.99416876e-01 1.13111891e-01 -6.64968073e-01
-3.84136140e-01 -5.28535426e-01 1.02256346e+00 3.90451729e-01
-8.30077901e-02 -1.51649028e-01 -2.92197913e-01 7.65517771e-01
-5.70888400e-01 -2.98212469e-01 5.53857088e-01 1.08603764e+00
6.21568680e-01 7.40454912e-01 3.58267874e-01 -5.94980359e-01
6.78496957e-01 -1.91942841e-01 -1.41776025e+00 -1.21576361e-01
-9.03441250e-01 3.47539224e-03 4.61638242e-01 -5.51576793e-01
-1.36924469e+00 4.00407791e-01 2.41947219e-01 -4.95709032e-01
3.98794562e-01 9.82443094e-01 -1.82513833e-01 -4.36260313e-01
4.53434408e-01 7.39367008e-02 6.15070462e-01 -1.63619503e-01
1.47406518e-01 4.00631458e-01 5.77821314e-01 -1.07915767e-01
1.71944392e+00 6.93752706e-01 5.28400004e-01 -8.56064022e-01
-3.96121085e-01 -7.48027384e-01 -4.88727927e-01 -3.38640481e-01
4.45374042e-01 -1.22235370e+00 -7.36139357e-01 5.95494747e-01
-6.02135301e-01 -9.55569819e-02 -2.60009140e-01 1.29195797e+00
-6.16725326e-01 4.32360381e-01 -3.57336819e-01 -1.47945750e+00
-2.24656284e-01 -8.17373872e-01 4.05649275e-01 3.60225022e-01
-1.74943790e-01 -1.09254813e+00 -4.15693671e-02 -4.41616327e-01
6.62390530e-01 1.88490510e-01 6.77712411e-02 -1.37208849e-01
-9.82717752e-01 -7.61126637e-01 3.51149648e-01 2.12028220e-01
1.08976275e-01 -2.18320172e-03 -5.57494342e-01 -9.13717508e-01
6.51241243e-01 5.65872967e-01 2.25559652e-01 4.29333419e-01
-6.31361604e-02 -1.42586678e-01 -5.07374883e-01 6.09278500e-01
1.49185169e+00 4.67157453e-01 4.65395331e-01 6.75882161e-01
3.29602957e-01 2.68608481e-01 1.21977925e+00 4.67343777e-01
-5.02718687e-01 4.17718530e-01 2.72729397e-01 5.36148846e-01
4.19979125e-01 5.45500964e-02 5.55020988e-01 8.27990055e-01
-1.12718433e-01 -6.24498725e-02 -7.77534187e-01 3.32236528e-01
-1.91831076e+00 -9.72367108e-01 -4.89846408e-01 2.52158737e+00
2.19253138e-01 2.80646503e-01 -5.83698571e-01 1.94353580e-01
7.02042818e-01 -1.49246931e-01 -6.60613418e-01 3.90966050e-03
-1.13944314e-01 -4.47800726e-01 1.32828283e+00 7.20867097e-01
-1.25874865e+00 3.19564909e-01 6.31733036e+00 4.30156618e-01
-9.78015602e-01 -1.55942276e-01 -5.35494983e-01 7.92576149e-02
-3.75968739e-02 6.00808203e-01 -1.21647942e+00 3.69272381e-01
1.22636092e+00 -8.27815354e-01 3.08235317e-01 8.30652475e-01
7.74095118e-01 -2.60399818e-01 -4.94297773e-01 9.37778115e-01
-3.20232362e-01 -9.74536717e-01 -5.64267337e-01 2.36534327e-01
7.92833924e-01 -2.38031119e-01 -3.70499156e-02 4.02594954e-01
1.02828324e-01 -1.61524683e-01 6.17323399e-01 1.14071178e+00
1.50428191e-01 -6.15601301e-01 8.91252518e-01 3.11927199e-01
-1.27628708e+00 -6.00336909e-01 -2.35250622e-01 -4.19174135e-01
8.64220679e-01 8.83053422e-01 -6.69829309e-01 8.27211559e-01
4.01771069e-01 6.27828896e-01 -6.91947937e-02 1.66643775e+00
-3.00428391e-01 5.35571992e-01 -6.88712478e-01 7.15723559e-02
6.68283254e-02 -6.03839815e-01 1.34870267e+00 3.93843323e-01
8.21146965e-01 3.15453887e-01 3.01129252e-01 3.71152461e-01
7.79898882e-01 -1.55215681e-01 -4.63958412e-01 -1.48712873e-01
5.68339705e-01 1.05907261e+00 -5.55108905e-01 -4.28396314e-01
-3.61686260e-01 2.83555746e-01 -8.44260812e-01 5.16086698e-01
-4.84132558e-01 -5.77217042e-01 9.81878221e-01 -2.81165261e-02
1.31572038e-01 -8.99890184e-01 -1.86661750e-01 -8.61169040e-01
-2.91204602e-01 -8.27382326e-01 6.20294064e-02 -5.19062102e-01
-9.77672696e-01 3.98269475e-01 5.71865700e-02 -2.09611678e+00
-5.88168979e-01 -4.94450927e-01 -4.63384241e-01 1.18734014e+00
-1.00742424e+00 -5.28468251e-01 -1.17833748e-01 1.04780763e-01
2.45157108e-01 -3.36554885e-01 5.07373512e-01 1.32873043e-01
-5.23724914e-01 -2.60081649e-01 9.65231240e-01 -4.66767073e-01
8.00611973e-01 -1.06428063e+00 1.93720624e-01 1.50907862e+00
-8.08417261e-01 1.02285826e+00 1.60098028e+00 -1.20658278e+00
-1.86585402e+00 -1.06587708e+00 4.94310498e-01 -1.62384629e-01
1.29608500e+00 -8.17044526e-02 -1.02852941e+00 6.83371902e-01
-8.27148929e-02 -3.62091362e-02 1.09848499e-01 -1.18700743e-01
6.89332783e-01 -1.10788070e-01 -9.05551136e-01 5.42758465e-01
5.57339072e-01 -3.92616928e-01 -5.71847141e-01 1.96723878e-01
3.73139828e-01 -7.53738463e-01 -1.06437111e+00 6.83136106e-01
3.89507979e-01 -1.42297119e-01 6.01747632e-01 -1.88414782e-01
-1.22100759e+00 -1.39742136e+00 1.90127119e-01 -1.45075023e+00
-4.12979752e-01 -1.34713554e+00 -3.11376601e-01 8.56421173e-01
1.75134093e-01 -1.11581457e+00 5.35070479e-01 6.77139342e-01
-3.39467078e-01 1.81189612e-01 -9.15753186e-01 -1.41953194e+00
-6.07403636e-01 -3.64186853e-01 2.57323354e-01 7.35636652e-01
-2.57455885e-01 1.05981074e-01 -4.72320735e-01 1.14545047e+00
7.80369043e-01 -8.97423774e-02 1.04302824e+00 -1.42446446e+00
-2.54862040e-01 1.29256770e-01 -3.82180184e-01 -9.89030480e-01
-4.77474146e-02 -1.74928028e-02 4.33245480e-01 -1.32519114e+00
-8.33985567e-01 -3.93478215e-01 3.18269134e-02 -2.26439118e-01
-1.51869565e-01 -1.11809023e-01 3.62095684e-02 3.49632651e-01
7.44065642e-02 5.66177905e-01 8.10776770e-01 1.67331100e-02
-4.90146309e-01 6.93689525e-01 1.27452195e-01 8.65279615e-01
6.83841527e-01 -3.01684946e-01 -7.50626624e-01 -5.01761325e-02
3.95519108e-01 2.17223719e-01 9.52982903e-02 -1.57568502e+00
5.45840800e-01 -6.86551780e-02 1.83438063e-01 -9.08932567e-01
3.55565280e-01 -1.19527841e+00 1.07809508e+00 8.19248021e-01
5.65588057e-01 1.18307985e-01 4.74583834e-01 1.32169366e+00
-4.28006470e-01 -3.19490850e-01 7.31351554e-01 2.34119341e-01
-1.03344953e+00 2.03432009e-01 -1.09111571e+00 -6.24904096e-01
9.49399114e-01 -4.21094596e-01 -1.68462425e-01 -5.97084641e-01
-1.25744474e+00 4.31791604e-01 4.32539225e-01 1.66199744e-01
2.93081492e-01 -1.07429647e+00 -6.70641586e-02 5.23014545e-01
-2.71273345e-01 -3.00585419e-01 5.40761471e-01 1.32543755e+00
-3.55355293e-01 5.12027800e-01 -6.91167191e-02 -4.72730368e-01
-1.14800882e+00 5.78557611e-01 6.58413827e-01 1.96752787e-01
-4.38000470e-01 2.85450757e-01 -3.94340366e-01 -1.73325926e-01
5.78379817e-02 -3.89173627e-01 7.75473416e-02 1.94949299e-01
6.70737505e-01 3.43897015e-01 1.43691838e-01 -5.13164878e-01
-9.48344991e-02 5.93749642e-01 4.18543965e-01 -2.78884232e-01
5.78639627e-01 -8.62583756e-01 -6.38749599e-02 6.34886444e-01
3.89682114e-01 1.30997136e-01 -1.42698288e+00 -2.03209490e-01
2.71304101e-01 -5.19885302e-01 3.02702516e-01 -3.03932339e-01
-5.41082919e-01 2.24482194e-01 8.73538017e-01 2.04951406e-01
1.08074629e+00 -1.00905740e+00 1.34451523e-01 7.91355550e-01
5.28149486e-01 -1.45395076e+00 -5.63079119e-01 1.04138041e+00
6.58527553e-01 -7.18214512e-01 3.78382653e-01 -6.64837122e-01
-2.79880434e-01 1.06253839e+00 3.58433217e-01 -3.03323835e-01
9.80560780e-01 3.40484560e-01 1.95072919e-01 2.74369270e-01
-7.69348621e-01 -4.66578633e-01 2.74025232e-01 3.85992557e-01
-1.45207688e-01 -2.60447443e-01 -7.21652329e-01 6.41010702e-02
4.65083048e-02 -6.33854941e-02 8.72454166e-01 1.12314904e+00
-7.35431254e-01 -8.79441917e-01 -1.14475119e+00 4.34609532e-01
-1.89848229e-01 2.37275019e-01 3.26587409e-01 1.00754225e+00
-2.78795540e-01 8.91678751e-01 1.12345986e-01 -1.45539060e-01
7.58022964e-01 1.36600077e-01 9.59085673e-02 -4.17930186e-01
-2.45164409e-01 5.38544297e-01 4.21523035e-01 -3.58447433e-01
-3.58888179e-01 -1.04796505e+00 -1.07981634e+00 -9.64848623e-02
-8.88241827e-01 5.93197703e-01 7.92621732e-01 8.81538928e-01
1.21195212e-01 5.69352448e-01 4.56631660e-01 -7.10001528e-01
-9.99775708e-01 -9.74794090e-01 -8.47140670e-01 -1.95825696e-01
2.98557997e-01 -1.08413315e+00 -8.24507654e-01 -2.37912625e-01] | [5.541032314300537, 2.5704991817474365] |
d11c8ef1-5cb5-4aa4-bda4-025ee59a04bd | refinevis-video-instance-segmentation-with | 2306.04774 | null | https://arxiv.org/abs/2306.04774v1 | https://arxiv.org/pdf/2306.04774v1.pdf | RefineVIS: Video Instance Segmentation with Temporal Attention Refinement | We introduce a novel framework called RefineVIS for Video Instance Segmentation (VIS) that achieves good object association between frames and accurate segmentation masks by iteratively refining the representations using sequence context. RefineVIS learns two separate representations on top of an off-the-shelf frame-level image instance segmentation model: an association representation responsible for associating objects across frames and a segmentation representation that produces accurate segmentation masks. Contrastive learning is utilized to learn temporally stable association representations. A Temporal Attention Refinement (TAR) module learns discriminative segmentation representations by exploiting temporal relationships and a novel temporal contrastive denoising technique. Our method supports both online and offline inference. It achieves state-of-the-art video instance segmentation accuracy on YouTube-VIS 2019 (64.4 AP), Youtube-VIS 2021 (61.4 AP), and OVIS (46.1 AP) datasets. The visualization shows that the TAR module can generate more accurate instance segmentation masks, particularly for challenging cases such as highly occluded objects. | ['Zicheng Liu', 'Quanzeng You', 'Peng Chu', 'Jiang Wang', 'Andre Abrantes'] | 2023-06-07 | null | null | null | null | ['video-instance-segmentation'] | ['computer-vision'] | [ 2.19182804e-01 2.34277407e-03 -3.29559624e-01 -3.81484181e-01
-1.01867175e+00 -5.53803980e-01 4.83412862e-01 -1.25851691e-01
-4.85769063e-01 5.87671876e-01 -2.16214154e-02 4.39488217e-02
1.55418903e-01 -6.17246747e-01 -1.10079026e+00 -4.90522146e-01
-2.87235916e-01 3.04371119e-01 6.66169107e-01 7.85293877e-02
1.85968608e-01 2.92511076e-01 -1.77067375e+00 7.98643410e-01
7.49166787e-01 1.07579434e+00 3.57171893e-01 8.96523595e-01
-3.60912323e-01 8.51744831e-01 -6.41646147e-01 -1.89335853e-01
4.23016995e-01 -3.38582873e-01 -1.10638130e+00 4.62213755e-01
7.64226794e-01 -4.58235949e-01 -3.05925399e-01 7.25925028e-01
-1.03340298e-02 5.19977629e-01 5.50670564e-01 -1.17559552e+00
-4.17318821e-01 5.54550409e-01 -8.71066689e-01 7.45685637e-01
3.49883556e-01 5.63976347e-01 8.38405788e-01 -6.81484401e-01
1.15173244e+00 1.26272392e+00 4.20926392e-01 4.92164522e-01
-1.29875243e+00 -5.14670670e-01 7.94696093e-01 5.39985359e-01
-1.15063381e+00 -2.44061917e-01 5.00156760e-01 -5.97972095e-01
1.01916325e+00 3.01074386e-01 1.04513109e+00 1.24694431e+00
-4.15449627e-02 1.36768758e+00 7.45147884e-01 1.59879893e-01
2.25318357e-01 -4.94350761e-01 2.39269480e-01 6.37796700e-01
-2.68803358e-01 1.18668690e-01 -5.63522279e-01 3.41870755e-01
1.12055564e+00 5.65977618e-02 -2.12385446e-01 -3.66669819e-02
-1.32663643e+00 3.65267903e-01 4.17577356e-01 3.18412393e-01
-7.41908431e-01 5.81640005e-01 4.66067255e-01 2.40912866e-02
5.71129322e-01 2.74314404e-01 -5.72906256e-01 -3.05831462e-01
-1.61866486e+00 3.27850938e-01 4.20067728e-01 1.10846388e+00
8.45379293e-01 3.60075921e-01 -6.05143309e-01 5.45585632e-01
1.45878315e-01 5.85724525e-02 2.15683714e-01 -1.52039266e+00
2.80979753e-01 4.21204358e-01 1.52847230e-01 -6.15667224e-01
-3.02907806e-02 -3.38245988e-01 -3.64233255e-01 1.45656914e-01
4.24053848e-01 5.07842191e-02 -1.40978038e+00 1.41408968e+00
4.63914782e-01 1.29777431e+00 -1.44394577e-01 1.04161346e+00
1.16737974e+00 9.70146775e-01 5.65863907e-01 -3.97565305e-01
1.30720413e+00 -1.32459557e+00 -8.62462938e-01 -9.67325643e-02
1.86455131e-01 -5.13080776e-01 9.22455609e-01 4.44289625e-01
-1.67066097e+00 -9.41129088e-01 -8.34949136e-01 -3.22792053e-01
-2.18196869e-01 -9.00618285e-02 4.54453886e-01 4.69481247e-03
-1.03757596e+00 6.94968224e-01 -1.16878879e+00 1.66497659e-02
8.34636867e-01 3.12093049e-01 -2.15091571e-01 -9.89674032e-02
-8.20314944e-01 2.92460173e-01 4.33003038e-01 -1.23540009e-03
-1.39782643e+00 -1.15466201e+00 -1.04698622e+00 -6.62835836e-02
5.47120810e-01 -6.00391150e-01 1.25548279e+00 -1.66049862e+00
-1.38857460e+00 8.89752984e-01 -3.40901732e-01 -7.86792099e-01
7.79232621e-01 -6.36109054e-01 -2.19878018e-01 7.49208987e-01
1.68763071e-01 1.24324095e+00 1.31659210e+00 -1.38795316e+00
-8.68483603e-01 -3.32882516e-02 1.58291087e-01 1.41457766e-01
3.34298223e-01 -1.93830319e-02 -1.24897420e+00 -1.00300217e+00
-5.65966479e-02 -5.78684390e-01 -3.51428598e-01 7.68606514e-02
-2.97558129e-01 -2.49261230e-01 1.34208322e+00 -1.02777934e+00
1.24210608e+00 -2.07811928e+00 4.17747587e-01 -2.62970645e-02
6.12264946e-02 3.28250051e-01 -3.49846125e-01 -1.95319355e-01
-3.48241776e-01 1.43644065e-01 -3.03016961e-01 -6.09121919e-01
-4.25385088e-01 4.93205577e-01 -7.42029846e-02 3.95659417e-01
4.05969590e-01 1.14112759e+00 -1.08260810e+00 -6.58240139e-01
6.12340391e-01 6.51823997e-01 -6.63360775e-01 3.93544465e-01
-7.99135566e-01 6.38774276e-01 -2.60821283e-01 7.22511649e-01
5.21326840e-01 -2.58681923e-01 -2.65403446e-02 -5.02036333e-01
-2.09064916e-01 -1.37402430e-01 -1.24322999e+00 2.23705912e+00
-1.15962908e-01 7.46134937e-01 5.20493239e-02 -9.50476289e-01
5.57456315e-01 3.71119142e-01 9.55785036e-01 -6.37667835e-01
1.19935930e-01 -2.88219690e-01 -5.21527827e-01 -8.45898449e-01
6.52952909e-01 4.28809881e-01 2.78242707e-01 2.75457740e-01
3.00492078e-01 -9.80763584e-02 6.93814278e-01 4.75403368e-01
9.03815687e-01 9.05544043e-01 -9.79615673e-02 -2.49492913e-01
4.91032094e-01 1.08207844e-01 6.60021365e-01 5.28425455e-01
-2.29609609e-01 1.03780580e+00 4.36448187e-01 -7.17153728e-01
-7.68724203e-01 -1.11108792e+00 5.06218337e-02 1.13393807e+00
5.34517586e-01 -5.56290329e-01 -8.21696818e-01 -8.42822134e-01
-2.71118969e-01 6.61837876e-01 -8.21735024e-01 2.10673839e-01
-8.86974514e-01 -1.68932542e-01 1.03103027e-01 7.86358833e-01
6.59756064e-01 -1.34620833e+00 -6.87985718e-01 2.97238648e-01
-5.02914429e-01 -1.24228597e+00 -6.65845454e-01 -1.72301888e-01
-9.61498320e-01 -1.35208857e+00 -6.46310210e-01 -6.76862657e-01
6.44785583e-01 2.15297356e-01 1.49760759e+00 2.60221928e-01
-4.98927951e-01 7.07857966e-01 -4.38722998e-01 -1.37072569e-02
-1.92002103e-01 -1.64745376e-01 -3.54208022e-01 1.74927250e-01
8.16972032e-02 -2.39295766e-01 -8.54799330e-01 3.56759161e-01
-1.20160294e+00 3.83932590e-01 1.47477061e-01 4.99754965e-01
1.32201755e+00 -3.40693891e-01 2.98566103e-01 -8.33033919e-01
-2.75610220e-02 -6.40287399e-01 -6.81852579e-01 1.71786010e-01
3.27756628e-02 -1.22679472e-01 2.32651323e-01 -4.98046786e-01
-1.09278774e+00 3.27520609e-01 -2.79765744e-02 -1.05933642e+00
-3.19189310e-01 1.35318041e-01 2.65101999e-01 4.34583157e-01
4.76601303e-01 1.27274260e-01 -4.95020151e-02 -4.16298360e-01
4.82314914e-01 4.52526100e-02 1.05426431e+00 -7.06842840e-01
5.46387076e-01 5.46494484e-01 -4.82820630e-01 -7.20671952e-01
-9.44050431e-01 -5.47665417e-01 -8.51422548e-01 -6.44248903e-01
1.46816659e+00 -1.01782942e+00 -4.89177793e-01 2.18781322e-01
-1.06465149e+00 -9.03581798e-01 -6.55987501e-01 1.77832216e-01
-9.41358030e-01 2.33377531e-01 -6.73528969e-01 -4.78390485e-01
-2.68231571e-01 -1.30636299e+00 1.35127425e+00 4.47329074e-01
-3.82909834e-01 -7.58411050e-01 -3.06565553e-01 4.74478006e-01
-9.60342139e-02 5.71187139e-01 2.89329320e-01 -2.26641998e-01
-1.05850744e+00 2.76629537e-01 -1.87627912e-01 3.39702785e-01
-8.72288086e-03 6.18891656e-01 -7.91370034e-01 -1.77449286e-01
-4.50790316e-01 4.50797640e-02 1.10565817e+00 9.96830642e-01
1.85320652e+00 -3.10037374e-01 -2.22211912e-01 8.15198064e-01
1.30665851e+00 3.51704746e-01 9.72730875e-01 2.68522292e-01
7.98377752e-01 4.16709155e-01 9.49041605e-01 4.35882896e-01
2.01775819e-01 6.64224625e-01 6.29701495e-01 -2.09213138e-01
-3.80884379e-01 1.83176175e-02 3.22224826e-01 3.18982214e-01
-4.76463646e-01 -9.48559791e-02 -6.89711392e-01 8.29445720e-01
-2.19938421e+00 -1.37582970e+00 -1.40209824e-01 1.77184272e+00
8.14955354e-01 1.40579266e-03 3.77609432e-01 3.89818251e-02
6.90754533e-01 3.24379206e-01 -6.43247247e-01 -2.02749908e-01
6.64973781e-02 2.86631137e-01 7.81791732e-02 5.23802221e-01
-1.28303301e+00 1.20213330e+00 5.76385117e+00 7.67978549e-01
-8.83653641e-01 1.19374499e-01 1.16383946e+00 -3.47213060e-01
-1.60542741e-01 -3.04336846e-01 -4.52104658e-01 5.77098012e-01
5.99027932e-01 5.31803370e-02 2.83723086e-01 7.80264378e-01
5.78569591e-01 -2.56354570e-01 -1.11757958e+00 1.20282316e+00
-7.28443190e-02 -2.08297682e+00 1.31037056e-01 -4.19937044e-01
1.03879189e+00 -6.37839362e-02 2.31577689e-03 3.39208782e-01
1.52368248e-01 -1.03470957e+00 1.05261075e+00 7.67077029e-01
7.41673768e-01 -7.98421800e-01 4.35071260e-01 -1.10657722e-01
-1.47676706e+00 1.24903582e-01 1.79231375e-01 3.08941841e-01
4.72109050e-01 3.05115134e-01 -4.02680486e-01 4.50873882e-01
1.17997587e+00 1.35254037e+00 -4.64018643e-01 1.04395211e+00
-2.52103686e-01 5.57487190e-01 -1.83830410e-01 7.15831161e-01
5.49828887e-01 -2.81620264e-01 4.58131909e-01 1.42877352e+00
3.48882005e-02 4.37743396e-01 4.21427071e-01 8.98549497e-01
-1.12429157e-01 -3.01341891e-01 -2.00676933e-01 8.16189870e-02
2.45901912e-01 1.27445567e+00 -1.06635106e+00 -9.02585626e-01
-2.42769480e-01 1.12199163e+00 -7.91981816e-03 7.90083170e-01
-1.26270986e+00 4.50371027e-01 1.05625677e+00 1.09274462e-01
7.52100348e-01 -4.22126770e-01 -2.98430443e-01 -1.04138482e+00
-3.45014244e-01 -8.04586411e-01 7.07575977e-01 -9.50609565e-01
-9.25854981e-01 6.28134847e-01 2.92891085e-01 -1.14698350e+00
-2.54908919e-01 -1.02293655e-01 -6.32057965e-01 3.40132117e-01
-1.36940181e+00 -9.36631560e-01 -5.81869781e-01 8.61671984e-01
1.58289099e+00 1.62830994e-01 2.97369540e-01 2.42684215e-01
-8.21263850e-01 1.39082327e-01 -4.48821366e-01 1.09650046e-01
3.40451539e-01 -1.20703030e+00 3.76240879e-01 1.00237346e+00
4.66002464e-01 1.51151150e-01 5.94321489e-01 -7.44985998e-01
-1.15075731e+00 -1.57866764e+00 1.84052721e-01 -3.72081578e-01
2.31230125e-01 4.62173037e-02 -1.02949488e+00 8.13769519e-01
3.87679905e-01 4.56047595e-01 3.11017215e-01 -4.63593066e-01
-1.78158715e-01 5.23772091e-02 -1.14813244e+00 7.96234190e-01
1.28224254e+00 -1.93147644e-01 -4.45935905e-01 3.31349164e-01
8.83085728e-01 -8.66794229e-01 -8.99203777e-01 2.31486365e-01
3.82336766e-01 -9.72429454e-01 1.28287232e+00 -6.10454261e-01
7.27867186e-01 -6.12153947e-01 1.32823288e-01 -7.21665204e-01
-1.55581802e-01 -8.61115038e-01 -7.02097952e-01 1.05604184e+00
7.72548988e-02 1.37501776e-01 8.81896079e-01 6.78547621e-01
-2.60182738e-01 -9.81051624e-01 -7.97608316e-01 -5.64644694e-01
-4.81582046e-01 -8.42017174e-01 4.55871671e-01 7.48248994e-01
-5.81567228e-01 -2.49229565e-01 -7.64356330e-02 2.41224915e-01
6.52302206e-01 2.10491106e-01 7.45803058e-01 -7.18127429e-01
-2.25313947e-01 -3.96869421e-01 -4.29728210e-01 -1.21562469e+00
3.49695951e-01 -5.44051468e-01 -3.90739739e-02 -1.59596968e+00
-1.32560849e-01 -2.12356225e-01 -3.65874797e-01 6.19680882e-01
-3.40779513e-01 6.43693984e-01 3.94393533e-01 1.06933139e-01
-1.08161891e+00 3.42399985e-01 1.47272909e+00 -4.76917595e-01
-4.36411113e-01 -3.87137420e-02 -3.91171068e-01 6.91219926e-01
6.21083081e-01 -2.95619220e-01 -5.19955695e-01 -5.11573434e-01
-2.91785508e-01 8.32844526e-02 4.66167271e-01 -9.65462267e-01
-3.77645306e-02 -2.70681709e-01 5.31389296e-01 -8.44644129e-01
4.48781610e-01 -6.58166409e-01 4.40433085e-01 4.18031305e-01
-3.21055651e-01 7.66670704e-02 4.86275256e-01 7.36554563e-01
-3.26054096e-01 1.17330975e-03 9.23725724e-01 -3.28334838e-01
-1.56014729e+00 6.83524251e-01 -3.12419981e-01 2.12127462e-01
1.43040180e+00 -6.00514770e-01 -9.05108899e-02 -2.71469802e-01
-1.16765976e+00 4.41374600e-01 1.58676729e-01 6.93780899e-01
9.15961087e-01 -1.11991608e+00 -7.04636455e-01 2.62556612e-01
-3.01450253e-01 4.59453315e-01 5.58978200e-01 8.45896780e-01
-7.19871581e-01 -2.53039151e-01 -2.24734709e-01 -1.16951346e+00
-1.40244675e+00 3.95690829e-01 2.04893351e-01 -5.13417870e-02
-1.04155433e+00 1.13417661e+00 3.75331998e-01 4.55105364e-01
3.52515638e-01 -6.47632420e-01 -2.95635611e-01 2.98115164e-01
7.40945876e-01 4.79389846e-01 -2.16927633e-01 -7.47559428e-01
-1.53324142e-01 6.06113017e-01 -2.22799927e-01 -5.84364310e-02
1.42396498e+00 -2.45042682e-01 2.67849058e-01 2.24592537e-01
1.01506984e+00 -6.31076217e-01 -2.19179845e+00 1.50841558e-02
-1.97416857e-01 -6.60777032e-01 -1.01719148e-01 -5.43839872e-01
-1.54742610e+00 5.17887890e-01 5.01152933e-01 -1.70163102e-02
1.19337702e+00 -2.55525345e-03 9.67460275e-01 -2.35848308e-01
2.81598628e-01 -1.20967567e+00 5.15004694e-01 3.59016120e-01
9.70343113e-01 -1.11623645e+00 6.37613423e-03 -4.58726287e-01
-7.88103282e-01 1.03591394e+00 8.15793455e-01 -4.34441656e-01
3.89283448e-01 2.51862884e-01 -4.96120267e-02 -1.66126892e-01
-8.35221946e-01 -3.09532434e-01 5.40408313e-01 6.49906218e-01
2.48693407e-01 -2.81634539e-01 -1.70585290e-01 5.49243987e-01
3.17160666e-01 -2.50374563e-02 4.24638659e-01 7.29754090e-01
-8.39458108e-02 -7.84430146e-01 -2.30390146e-01 4.62257296e-01
-4.88232046e-01 1.08198449e-01 1.58401474e-01 8.14059854e-01
2.71442682e-01 7.62789488e-01 5.65251410e-01 -6.51121810e-02
1.08606763e-01 -3.67683992e-02 3.77942264e-01 -5.58609128e-01
-9.58492041e-01 2.80029595e-01 1.90693568e-02 -1.38407254e+00
-9.23500717e-01 -6.85065091e-01 -1.61287904e+00 -6.69891834e-02
6.93415105e-02 -1.19386837e-01 3.75677973e-01 9.45431471e-01
4.59524900e-01 1.12668192e+00 1.52514055e-01 -1.57757568e+00
4.93078828e-01 -5.28410375e-01 -2.70311952e-01 6.95583224e-01
3.05094540e-01 -5.97432017e-01 1.65163558e-02 8.49768400e-01] | [9.130172729492188, -0.078250452876091] |
19f83960-972a-4c1b-bb2f-6bc2b959aeff | development-of-the-multilingual-semantic | null | null | https://aclanthology.info/papers/N15-1137/n15-1137 | https://www.aclweb.org/anthology/N15-1137 | Development of the Multilingual Semantic Annotation System | null | ["Angela D'Egidio", 'Scott Piao', 'Carmen Dayrell', 'Paul Rayson', 'Francesca Bianchi'] | 2015-05-01 | null | null | null | hlt-2015-5 | ['multilingual-nlp'] | ['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] | [-1.5391634702682495, 15.869173049926758] |
62abb48e-e2b8-419a-9eac-5acaab839207 | single-image-calibration-using-knowledge | 2212.02379 | null | https://arxiv.org/abs/2212.02379v1 | https://arxiv.org/pdf/2212.02379v1.pdf | Single image calibration using knowledge distillation approaches | Although recent deep learning-based calibration methods can predict extrinsic and intrinsic camera parameters from a single image, their generalization remains limited by the number and distribution of training data samples. The huge computational and space requirement prevents convolutional neural networks (CNNs) from being implemented in resource-constrained environments. This challenge motivated us to learn a CNN gradually, by training new data while maintaining performance on previously learned data. Our approach builds upon a CNN architecture to automatically estimate camera parameters (focal length, pitch, and roll) using different incremental learning strategies to preserve knowledge when updating the network for new data distributions. Precisely, we adapt four common incremental learning, namely: LwF , iCaRL, LU CIR, and BiC by modifying their loss functions to our regression problem. We evaluate on two datasets containing 299008 indoor and outdoor images. Experiment results were significant and indicated which method was better for the camera calibration estimation. | ['Antoine Letienne', 'Mohamed Abbas Hedjazi', 'Oussama Hadjerci', 'Khadidja Ould Amer'] | 2022-12-05 | null | null | null | null | ['camera-calibration'] | ['computer-vision'] | [-3.10627408e-02 -2.48163179e-01 -2.89664090e-01 -6.87582672e-01
-5.25785863e-01 -7.94360518e-01 2.75063246e-01 -3.77046108e-01
-6.66286111e-01 8.12697351e-01 -2.93787301e-01 -2.30420515e-01
-2.52128951e-02 -4.24102366e-01 -9.55934823e-01 -7.21814036e-01
3.43489408e-01 1.05688624e-01 1.60035521e-01 6.50350526e-02
3.62802893e-01 7.32348800e-01 -1.09937036e+00 -2.63746798e-01
7.18807638e-01 9.97716069e-01 2.76113719e-01 9.00817811e-01
3.10327023e-01 9.63119268e-01 -3.91943604e-01 -6.17768943e-01
6.27236843e-01 -1.96782246e-01 -3.78094733e-01 1.18651196e-01
6.71523571e-01 -6.24766648e-01 -6.89959466e-01 9.19000089e-01
4.44345683e-01 1.37125134e-01 2.87825674e-01 -1.13567662e+00
-5.60783029e-01 3.52422982e-01 -5.48034608e-01 3.00421882e-02
-1.83970034e-02 6.54734373e-02 5.60784757e-01 -7.50920355e-01
2.85534918e-01 6.06806338e-01 1.38673532e+00 3.86459857e-01
-8.93029809e-01 -7.95055926e-01 1.71080247e-01 1.92260608e-01
-1.52861226e+00 -4.53788847e-01 8.80479574e-01 -1.80638164e-01
6.24614656e-01 -3.07480335e-01 5.92762828e-01 1.24183130e+00
5.26740327e-02 3.88785034e-01 8.36001515e-01 -4.81139809e-01
1.53067708e-01 3.48748624e-01 -2.80389965e-01 5.74968278e-01
3.37584317e-01 2.35185385e-01 -3.84105414e-01 2.45343477e-01
1.11426651e+00 -1.72231078e-01 -1.73508957e-01 -6.61935449e-01
-1.00633788e+00 6.74305499e-01 4.34630543e-01 -4.61065434e-02
2.11059272e-01 4.52242702e-01 1.85031295e-01 2.74504900e-01
8.00978243e-02 4.30412680e-01 -9.06372249e-01 -1.69964403e-01
-7.65758932e-01 -2.37526894e-02 5.95394194e-01 1.54097939e+00
1.27311981e+00 1.44863918e-01 4.17922467e-01 8.03224862e-01
1.24397106e-01 5.57287157e-01 5.09341359e-01 -1.19034147e+00
5.42760909e-01 2.38689870e-01 6.41160086e-02 -1.14828992e+00
-5.30421257e-01 -6.12181067e-01 -9.16286647e-01 -4.90288772e-02
4.48718011e-01 -5.36878109e-01 -7.09819674e-01 1.44841516e+00
8.19410384e-03 3.11668724e-01 -1.19160295e-01 8.20063472e-01
4.39240783e-01 4.14995313e-01 -3.87087584e-01 6.03664517e-02
4.88047063e-01 -1.14979303e+00 -3.65463883e-01 -4.67730880e-01
5.82174182e-01 -9.32644665e-01 9.00518954e-01 7.36957908e-01
-7.62330115e-01 -9.52137589e-01 -1.41312253e+00 -2.13905588e-01
-1.74869537e-01 7.73263037e-01 7.31635034e-01 8.74903262e-01
-1.08444870e+00 5.25832593e-01 -7.21509814e-01 -1.40056297e-01
2.79164672e-01 6.66949570e-01 -2.67441243e-01 2.77826097e-02
-8.14125597e-01 5.50018370e-01 5.69321036e-01 5.33552885e-01
-8.44904184e-01 -5.44065356e-01 -6.92739487e-01 -3.69748697e-02
3.79812956e-01 -4.80095297e-01 1.30824292e+00 -1.36718762e+00
-2.10401344e+00 4.77506518e-01 3.40892673e-01 -5.32843173e-01
5.02064109e-01 -4.54668075e-01 -3.09415877e-01 -4.51342948e-02
-3.31656992e-01 8.24697196e-01 1.12756348e+00 -1.38102770e+00
-6.11231089e-01 -4.28850725e-02 2.67832905e-01 5.82139008e-02
-4.62811440e-01 -4.90466237e-01 -7.93443561e-01 -4.59816158e-01
3.42257380e-01 -1.10205686e+00 -2.01779604e-01 5.29663712e-02
-1.98130295e-01 5.27742028e-01 9.60315108e-01 -5.07936835e-01
9.75144625e-01 -2.09422565e+00 -3.53572890e-02 1.87737450e-01
-2.45723240e-02 2.68131107e-01 -7.79784620e-02 -1.01661436e-01
-1.31859824e-01 -1.81714594e-01 7.60450633e-03 -4.56422418e-01
-4.79644179e-01 1.68109447e-01 -1.50499329e-01 7.07007706e-01
-2.87711788e-02 7.37833321e-01 -6.47521019e-01 -3.72283250e-01
6.45624936e-01 5.65657318e-01 -7.65767574e-01 2.71366775e-01
-8.73727128e-02 5.27571857e-01 -4.78720926e-02 6.43463969e-01
9.96708691e-01 -1.23223662e-01 2.99418986e-01 -5.49279451e-01
-1.13012739e-01 -3.80339891e-01 -1.19943619e+00 1.82409143e+00
-8.48439634e-01 1.19048750e+00 -3.54503170e-02 -9.63039577e-01
1.34174860e+00 -1.25295684e-01 4.87767637e-01 -5.80500901e-01
4.99932855e-01 4.35663797e-02 -4.14000630e-01 -5.08607328e-01
7.44906962e-01 2.70909399e-01 -5.63638359e-02 -6.40227571e-02
3.43250901e-01 -2.89673656e-01 -2.10517883e-01 -7.87482709e-02
4.79975373e-01 3.91004473e-01 5.77045083e-02 -7.75412023e-02
5.90552509e-01 -1.33057833e-01 5.57963729e-01 6.70255721e-01
-8.19593817e-02 8.74258280e-01 3.78536046e-01 -8.25688362e-01
-1.41578603e+00 -7.88942993e-01 -5.40032163e-02 7.74582446e-01
2.22151905e-01 -2.44751662e-01 -7.36996770e-01 -4.64477956e-01
-2.58905590e-01 2.47422382e-01 -5.34325182e-01 -2.90804982e-01
-9.16677833e-01 -6.56336188e-01 5.33656955e-01 6.48927748e-01
8.52159381e-01 -2.98642069e-01 -5.45353591e-01 8.94680321e-02
8.69112834e-02 -1.53399229e+00 -6.19422674e-01 4.56522912e-01
-9.59013343e-01 -1.22122192e+00 -4.78755265e-01 -7.42569745e-01
7.52527714e-01 2.46282876e-01 8.79679441e-01 -8.57040063e-02
-2.11908132e-01 5.06524146e-01 -2.76270479e-01 -3.98955882e-01
-1.21209972e-01 5.24504602e-01 1.23316839e-01 5.69581874e-02
1.44764721e-01 -4.31969970e-01 -5.62380195e-01 4.86820132e-01
-6.70050204e-01 1.18424268e-02 5.97739339e-01 8.25997591e-01
5.33061087e-01 -8.25985968e-02 1.35224074e-01 -8.15803766e-01
2.58045226e-01 -8.26628059e-02 -1.27448380e+00 1.96433678e-01
-7.05755353e-01 1.36820981e-02 8.39683592e-01 -5.52692890e-01
-1.21707606e+00 5.46000183e-01 6.07668571e-02 -6.38053894e-01
3.23246643e-02 2.51067340e-01 -1.57445744e-01 -5.60666502e-01
6.40146494e-01 -1.08031951e-01 -2.12474227e-01 -1.48441851e-01
2.32208341e-01 5.29367685e-01 9.86180604e-01 -5.60775220e-01
8.47153723e-01 4.24010158e-01 4.46168371e-02 -8.81502986e-01
-9.93672252e-01 -3.02753419e-01 -1.20529926e+00 -3.32501084e-01
7.11435974e-01 -1.29984677e+00 -6.74879968e-01 8.85723114e-01
-1.01187098e+00 -4.76788491e-01 -3.19851302e-02 8.10298681e-01
-4.67068493e-01 4.46714938e-01 -2.11992159e-01 -4.14199799e-01
1.37930065e-01 -1.20359373e+00 7.31030762e-01 5.00770032e-01
3.46989423e-01 -1.13437855e+00 1.85918868e-01 1.40904784e-01
5.66225231e-01 1.89038530e-01 4.03308600e-01 1.10574476e-01
-8.85016441e-01 -3.76794606e-01 -3.49450678e-01 7.45472610e-01
7.77560100e-02 4.27952081e-01 -1.20470560e+00 -5.36920190e-01
-4.93819034e-03 -3.25269997e-01 6.03094816e-01 6.31996155e-01
1.77501798e+00 4.74552298e-03 -1.20943338e-01 1.62495673e+00
1.74952531e+00 1.86270773e-01 7.53652453e-01 7.08192289e-01
8.11824739e-01 4.42402624e-02 3.84374499e-01 4.51399565e-01
3.26399684e-01 5.99464357e-01 6.74347699e-01 1.36823030e-02
5.59518188e-02 -3.20677280e-01 2.68830061e-01 9.15028214e-01
-2.74192035e-01 -1.02577381e-01 -7.85197973e-01 2.45841786e-01
-1.63227892e+00 -5.04131138e-01 5.65846972e-02 2.24915457e+00
5.07456362e-01 2.64350116e-01 -2.38144711e-01 -2.32310146e-01
5.12684345e-01 6.13050675e-03 -7.58180082e-01 -1.40998140e-01
-2.85501957e-01 8.91442001e-02 1.29563153e+00 3.93741101e-01
-1.23186219e+00 1.00054789e+00 6.95016766e+00 2.86329329e-01
-1.49582243e+00 -1.63938269e-01 8.21233094e-01 1.84541382e-02
4.17567678e-02 1.13109961e-01 -1.09915864e+00 2.08775833e-01
1.03036380e+00 2.03991398e-01 5.91990769e-01 1.08274090e+00
-1.30573913e-01 -1.30338505e-01 -9.99551833e-01 1.54633129e+00
2.56515652e-01 -1.34745586e+00 -2.65625417e-01 -2.39219204e-01
1.01459837e+00 1.92856938e-01 2.20431387e-01 3.69971514e-01
2.00090498e-01 -9.02746916e-01 6.64509475e-01 7.03109920e-01
9.15031254e-01 -8.31571639e-01 9.60801423e-01 9.58680585e-02
-1.08439636e+00 -4.39221203e-01 -7.65387475e-01 1.27503991e-01
-1.60988912e-01 3.16977650e-01 -8.52184832e-01 3.39019299e-01
8.76709998e-01 9.20062065e-01 -1.04808807e+00 1.21943676e+00
-1.43073693e-01 4.38188940e-01 -2.96997428e-01 1.87977534e-02
2.22210526e-01 -3.41247916e-01 1.58592069e-03 9.89193559e-01
5.68744838e-01 -1.38180643e-01 -1.42741904e-01 5.83351791e-01
-3.25485855e-01 -2.91707784e-01 -3.20796549e-01 3.20544124e-01
4.81002986e-01 1.23288894e+00 -5.36608219e-01 1.77239075e-01
-4.79945719e-01 8.88998270e-01 3.03716838e-01 4.31539446e-01
-1.23861396e+00 -3.58577639e-01 3.34376752e-01 -5.93137331e-02
4.44661766e-01 -7.58785248e-01 -2.55872756e-01 -1.26138866e+00
1.17161078e-02 -5.74534833e-01 -2.25778017e-03 -9.67473328e-01
-8.39770854e-01 5.92205286e-01 2.11175829e-01 -1.44140291e+00
-1.75181940e-01 -1.05138600e+00 -4.03773993e-01 4.56160098e-01
-1.59276390e+00 -1.14259660e+00 -9.31103289e-01 6.78826928e-01
5.19322395e-01 -5.44059098e-01 4.34036165e-01 5.23386657e-01
-6.11231506e-01 9.19516504e-01 4.00740802e-01 3.13112557e-01
9.82617676e-01 -1.09086108e+00 2.58881807e-01 8.06585431e-01
2.33237911e-02 3.81085515e-01 4.81312186e-01 -6.06276393e-02
-1.49433255e+00 -1.00011575e+00 2.39712358e-01 -4.45237637e-01
5.01127601e-01 -4.93574232e-01 -5.40980756e-01 8.43056381e-01
1.11333914e-01 1.78160682e-01 3.12647551e-01 -9.67440754e-03
-3.26934725e-01 -6.76925957e-01 -8.39039445e-01 3.52857053e-01
7.33466864e-01 -4.83292729e-01 1.49573162e-01 8.51713791e-02
7.65801132e-01 -8.01247835e-01 -6.03837609e-01 2.99713135e-01
7.16209590e-01 -1.20583916e+00 1.04398036e+00 -1.21660151e-01
1.69529736e-01 -3.15752268e-01 -1.79300725e-01 -1.15280163e+00
-1.94125012e-01 -5.20160437e-01 -9.77702253e-03 9.57231581e-01
3.99827152e-01 -4.47277665e-01 1.19071889e+00 5.76011360e-01
-2.65309989e-01 -5.33492208e-01 -6.60951793e-01 -5.48675776e-01
9.80911925e-02 -5.11877298e-01 6.58312798e-01 7.83226013e-01
-8.86431217e-01 2.21171647e-01 -7.15428472e-01 3.26178193e-01
5.87546527e-01 -3.26799989e-01 1.28512990e+00 -1.00597620e+00
-3.21064115e-01 -1.39299437e-01 -3.88561785e-01 -1.28447068e+00
1.51254848e-01 -2.43398160e-01 -9.06913728e-02 -8.16352189e-01
-1.29652908e-02 -6.13805711e-01 -4.71520051e-02 1.05209745e-01
2.95151025e-01 1.72270119e-01 -1.01794945e-02 7.89792836e-02
-4.50073361e-01 3.92107308e-01 1.03677738e+00 -2.06686616e-01
-3.27172548e-01 2.33536825e-01 -4.72617120e-01 8.39479089e-01
9.57847655e-01 -1.84556112e-01 -6.13851488e-01 -9.65533435e-01
6.88232780e-01 -2.81254023e-01 3.42956722e-01 -1.53785133e+00
6.59084380e-01 -6.69341385e-02 8.60769153e-01 -7.67604470e-01
2.07739785e-01 -1.08739090e+00 2.53785104e-01 -5.08736223e-02
-1.86755896e-01 2.27988854e-01 4.00654674e-01 4.17836517e-01
-2.16114148e-01 -4.68639284e-01 9.60389018e-01 -4.96551767e-02
-9.35542822e-01 4.80711609e-01 -4.76873703e-02 -1.72681540e-01
8.44738424e-01 -4.78270084e-01 -5.44385053e-02 -4.93945152e-01
-5.30576169e-01 -1.77445650e-01 6.00380480e-01 3.76371592e-01
6.57652617e-01 -1.22380435e+00 -2.90077895e-01 4.48114008e-01
1.19119287e-01 4.93070632e-01 1.32623106e-01 4.44850415e-01
-1.13206065e+00 3.71896535e-01 -4.02181715e-01 -7.86592424e-01
-9.27493989e-01 4.94826794e-01 7.85469890e-01 1.56759277e-01
-2.98332691e-01 9.08608854e-01 -2.90606916e-01 -7.72623777e-01
4.75674152e-01 -6.11627340e-01 1.14047118e-01 -3.80885750e-01
1.39387548e-01 2.68785834e-01 1.31181300e-01 -4.43844438e-01
1.19028371e-02 9.86419022e-01 -3.51305306e-02 2.34041721e-01
1.33240521e+00 -4.77729797e-01 4.53360714e-02 2.09339559e-01
1.56854403e+00 -1.22115709e-01 -1.89958119e+00 -1.65068716e-01
-2.08025426e-01 -5.70716083e-01 2.73644418e-01 -3.91005695e-01
-1.47442496e+00 7.75588632e-01 9.18508768e-01 -5.06853640e-01
1.16803062e+00 -4.29135799e-01 5.97926497e-01 9.78408277e-01
4.82691318e-01 -1.53952301e+00 3.38772029e-01 6.65213585e-01
4.73825634e-01 -1.50718582e+00 2.53683627e-01 -1.08704992e-01
-3.47766042e-01 1.74349368e+00 9.66223061e-01 -1.55527413e-01
8.43374491e-01 4.00286436e-01 1.04065351e-01 9.18894559e-02
-3.29180300e-01 5.17117441e-01 5.11569455e-02 5.81777930e-01
2.93982565e-01 -2.14017481e-01 5.50575435e-01 2.50481278e-01
-3.68234098e-01 -1.62158441e-02 8.25982749e-01 6.50122821e-01
-1.45916179e-01 -1.00758958e+00 -2.98716784e-01 1.75103262e-01
4.46241349e-03 1.94000289e-01 -1.41374722e-01 1.10797942e+00
4.07955348e-01 4.53754157e-01 1.33081330e-02 -5.50846159e-01
2.80639619e-01 -6.08824253e-01 6.22069359e-01 -1.81205332e-01
-2.23283425e-01 -1.45643964e-01 -3.69179100e-01 -4.27504718e-01
-7.11525321e-01 -6.53183579e-01 -6.64661229e-01 -2.86381245e-01
-5.90396047e-01 -2.86630660e-01 9.17877555e-01 6.59347653e-01
-3.00396066e-02 3.77738357e-01 9.87436831e-01 -9.03488100e-01
-5.00235677e-01 -7.99824297e-01 -3.62400085e-01 -1.09724127e-01
5.60828805e-01 -5.68392456e-01 -4.39588428e-01 4.55081433e-01] | [8.248605728149414, -2.2430670261383057] |
c0d1ab66-6f1a-4c23-bfab-50d059d4800d | idll-inverse-depth-line-based-visual | 2304.11748 | null | https://arxiv.org/abs/2304.11748v1 | https://arxiv.org/pdf/2304.11748v1.pdf | IDLL: Inverse Depth Line based Visual Localization in Challenging Environments | Precise and real-time localization of unmanned aerial vehicles (UAVs) or robots in GNSS denied indoor environments are critically important for various logistics and surveillance applications. Vision-based simultaneously locating and mapping (VSLAM) are key solutions but suffer location drifts in texture-less, man-made indoor environments. Line features are rich in man-made environments which can be exploited to improve the localization robustness, but existing point-line based VSLAM methods still lack accuracy and efficiency for the representation of lines introducing unnecessary degrees of freedoms. In this paper, we propose Inverse Depth Line Localization(IDLL), which models each extracted line feature using two inverse depth variables exploiting the fact that the projected pixel coordinates on the image plane are rather accurate, which partially restrict the lines. This freedom-reduced representation of lines enables easier line determination and faster convergence of bundle adjustment in each step, therefore achieves more accurate and more efficient frame-to-frame registration and frame-to-map registration using both point and line visual features. We redesign the whole front-end and back-end modules of VSLAM using this line model. IDLL is extensively evaluated in multiple perceptually-challenging datasets. The results show it is more accurate, robust, and needs lower computational overhead than the current state-of-the-art of feature-based VSLAM methods. | ['Deying Li', 'Xuewei Bai', 'Shuo Wang', 'Yongcai Wang', 'Yu Shao', 'Wanting Li'] | 2023-04-23 | null | null | null | null | ['visual-localization'] | ['computer-vision'] | [-1.51258126e-01 -7.19287574e-01 2.23357260e-01 -3.03892165e-01
-1.47123575e-01 -8.67435277e-01 5.46766818e-01 -4.96128649e-02
-6.24353647e-01 7.05084562e-01 -5.73912919e-01 -2.87185222e-01
-2.75557041e-01 -7.80222893e-01 -5.49257278e-01 -5.48483312e-01
7.54662380e-02 2.75871664e-01 3.96797001e-01 -3.57858241e-01
1.36917621e-01 9.74842072e-01 -1.35038757e+00 -4.87823486e-01
7.40872383e-01 9.69506621e-01 3.52233708e-01 5.87336421e-01
1.29727647e-01 1.98988363e-01 -3.92661184e-01 7.45439678e-02
7.34930694e-01 -1.06474981e-01 -2.39541680e-01 4.17452574e-01
6.45435333e-01 -4.46081191e-01 -4.22404081e-01 1.02924764e+00
4.94749308e-01 4.35409956e-02 2.56307542e-01 -1.17962623e+00
-9.31491703e-02 -3.53648335e-01 -1.04434073e+00 -2.36522943e-01
7.69706249e-01 8.35815296e-02 3.60752076e-01 -8.78505111e-01
5.92623949e-01 9.14265573e-01 1.30096900e+00 -1.05183966e-01
-8.75265718e-01 -3.46877217e-01 -1.16050847e-01 4.28392291e-02
-1.84023726e+00 -2.82390147e-01 7.04916716e-01 -4.46989954e-01
5.42297781e-01 3.16992164e-01 8.08392763e-01 4.57028925e-01
3.98530185e-01 1.52767531e-03 9.06483948e-01 -6.12603962e-01
-4.81549799e-02 -1.02667280e-01 -3.60866249e-01 1.14188540e+00
7.29277492e-01 1.01293318e-01 -4.51144040e-01 -1.18683025e-01
1.42277157e+00 4.77949768e-01 -8.38902891e-01 -1.14446044e+00
-1.70503986e+00 6.45022988e-01 6.65428936e-01 1.02549456e-01
-3.99492651e-01 -1.01083599e-01 -1.58285439e-01 2.74998575e-01
7.84009099e-02 4.03751284e-01 -4.22354281e-01 -4.68756258e-02
-9.50373769e-01 6.83229864e-02 5.80948472e-01 1.45597255e+00
1.03430724e+00 -7.49564990e-02 3.80418658e-01 4.18714285e-01
5.50420642e-01 1.18011439e+00 4.49127078e-01 -7.83036530e-01
2.41127044e-01 6.81453884e-01 5.08016229e-01 -1.41081858e+00
-9.41056252e-01 -5.57647049e-01 -8.85864139e-01 3.20130736e-01
2.84214318e-01 -3.19878638e-01 -6.09311998e-01 1.03648663e+00
4.61939484e-01 -1.20226219e-01 8.42164159e-02 1.00433147e+00
6.84397519e-01 4.83318657e-01 -7.25037932e-01 -1.76783815e-01
1.29764330e+00 -8.76001179e-01 -6.26035810e-01 -4.11490530e-01
7.70464897e-01 -1.10619116e+00 6.88660085e-01 1.21452585e-01
-4.15858060e-01 -5.78868270e-01 -1.19457996e+00 3.39021944e-02
-1.85829833e-01 7.39495397e-01 6.59411371e-01 4.19929743e-01
-1.01798737e+00 2.16977313e-01 -7.91881740e-01 -7.02681661e-01
1.59244984e-02 4.27754909e-01 -8.43863189e-01 -2.74535805e-01
-6.75273240e-01 7.11225212e-01 1.69391483e-01 4.14021730e-01
-2.11426213e-01 -3.62814873e-01 -1.12007189e+00 -4.68581051e-01
4.46246088e-01 -9.19793367e-01 7.95840681e-01 -4.13464963e-01
-1.48650348e+00 6.75484478e-01 -3.06117773e-01 -4.61448461e-01
6.53216183e-01 -2.82072186e-01 -1.47749051e-01 1.99530631e-01
2.42181644e-01 5.40649354e-01 8.03840876e-01 -1.24395525e+00
-8.10367525e-01 -4.77129966e-01 -3.03096846e-02 5.15573978e-01
2.05746800e-01 -5.64728200e-01 -4.60054696e-01 -2.79803038e-01
9.53652143e-01 -1.03119469e+00 -2.14313447e-01 5.14731765e-01
-1.96975961e-01 2.85916865e-01 1.14940786e+00 -3.45970064e-01
7.83219397e-01 -2.10410929e+00 -2.10585341e-01 1.07762538e-01
1.03124883e-02 2.12731242e-01 1.33194119e-01 3.99923027e-01
3.40639591e-01 -4.88176435e-01 -1.30023658e-01 -2.43015975e-01
-4.14915651e-01 1.71867371e-01 -6.62086308e-02 1.07097673e+00
-3.69240820e-01 5.23433626e-01 -8.46934438e-01 -4.05550987e-01
8.76727819e-01 5.50716460e-01 -1.45209745e-01 4.49418426e-02
3.67876530e-01 6.08151197e-01 -5.44145525e-01 8.87246966e-01
1.08644795e+00 3.22420388e-01 -3.53669554e-01 -5.75832546e-01
-7.81580925e-01 -4.94319558e-01 -1.33112240e+00 2.03182435e+00
-4.07556027e-01 8.77464235e-01 5.08652851e-02 -2.83505350e-01
1.36472929e+00 -8.82904157e-02 5.36398947e-01 -3.00010085e-01
1.73985943e-01 1.33857027e-01 -3.30415308e-01 -2.10526615e-01
5.51716685e-01 4.33091849e-01 9.52079371e-02 -3.32884848e-01
-1.85750231e-01 -7.81969190e-01 -9.07947645e-02 -2.72642612e-01
8.36163163e-01 5.41822970e-01 7.61429787e-01 -1.43517435e-01
8.15291107e-01 3.49345058e-01 7.20214605e-01 5.43935776e-01
-7.27172345e-02 6.76454186e-01 -1.93186834e-01 -6.24173939e-01
-8.71217430e-01 -7.22121775e-01 -2.74436831e-01 3.23524363e-02
7.60698020e-01 -4.93635207e-01 -5.26807070e-01 -2.36069277e-01
1.11170456e-01 1.49148479e-01 -4.29460891e-02 2.08471268e-01
-2.78206557e-01 -3.73799235e-01 2.22022176e-01 9.78985354e-02
1.08278692e+00 -2.39371166e-01 -1.25455523e+00 1.32945672e-01
-8.24282020e-02 -1.43248177e+00 -1.73238412e-01 -1.37724474e-01
-7.62806654e-01 -1.28924429e+00 -7.17707634e-01 -6.60900891e-01
9.72807527e-01 1.06131113e+00 5.22060752e-01 -2.20680669e-01
-2.71645337e-01 6.25963151e-01 -4.71288055e-01 -4.14281905e-01
4.49903667e-01 -2.56650299e-01 4.40036625e-01 -2.76955310e-02
1.41447470e-01 -1.41533762e-01 -5.46106219e-01 7.81139970e-01
-3.93903315e-01 1.90843776e-01 4.71151352e-01 8.35660517e-01
8.70922506e-01 3.59619372e-02 -3.81666273e-01 -2.49929696e-01
1.91560358e-01 1.23487547e-01 -1.37074256e+00 1.40899077e-01
-2.68153220e-01 -3.90241951e-01 5.12221158e-01 7.92903826e-02
-4.90502268e-01 6.73240840e-01 1.71805710e-01 -5.14144301e-01
-3.20161313e-01 3.14481705e-01 -9.67416633e-03 -9.57778752e-01
5.07791758e-01 1.79277509e-01 6.15319274e-02 -1.63931459e-01
2.12031066e-01 6.18849099e-01 6.45738304e-01 4.44149785e-03
1.17283022e+00 6.98172092e-01 5.94656110e-01 -1.30411577e+00
-5.05605638e-01 -8.52632940e-01 -1.25508940e+00 -2.68779784e-01
5.82662582e-01 -1.07814586e+00 -6.77903116e-01 6.11276984e-01
-1.32092392e+00 6.66357875e-02 -2.30364263e-01 7.90505707e-01
-6.21692717e-01 7.71218717e-01 9.20305327e-02 -6.88281536e-01
-1.42901182e-01 -1.37041342e+00 1.10698783e+00 5.25901198e-01
2.36075029e-01 -9.90026295e-01 1.07594579e-03 -3.76697183e-02
1.89701527e-01 6.68688357e-01 6.42410964e-02 1.70689121e-01
-7.77637780e-01 -5.62689185e-01 -1.69147894e-01 -2.89246812e-02
3.20044935e-01 8.27701986e-02 -7.21318126e-01 -6.34987414e-01
7.37609863e-02 2.09441110e-01 4.17397141e-01 6.33071303e-01
4.57764447e-01 2.12424491e-02 -5.76447725e-01 1.27194667e+00
1.73873842e+00 2.69303203e-01 3.11270952e-01 8.07062149e-01
8.56180668e-01 3.45131010e-01 1.26711786e+00 6.55815542e-01
5.61099529e-01 9.59527493e-01 7.05297887e-01 -5.49388111e-01
1.39524817e-01 -9.32578817e-02 3.20700765e-01 4.38052803e-01
-2.74156958e-01 -1.16257735e-01 -8.71432543e-01 1.78941920e-01
-1.97283554e+00 -6.75004482e-01 -5.76772451e-01 2.48297429e+00
1.97555214e-01 -2.54018903e-01 -3.47677201e-01 6.78823069e-02
4.71702665e-01 9.93868187e-02 -2.52427965e-01 1.35337457e-01
-3.67549330e-01 -4.19973195e-01 1.32080960e+00 5.88207126e-01
-1.41031849e+00 1.08363605e+00 5.07154942e+00 4.12445247e-01
-1.16567922e+00 -2.43619964e-01 -2.74403751e-01 5.81821978e-01
2.90777147e-01 2.94721544e-01 -8.90279591e-01 4.04254533e-02
2.53255337e-01 9.24118012e-02 1.81144789e-01 9.77123797e-01
1.97167635e-01 -5.63348174e-01 -6.88167751e-01 1.68712962e+00
1.92253441e-01 -1.14963841e+00 -2.27736875e-01 1.01093486e-01
7.71044493e-01 1.80895030e-01 -2.22668290e-01 -4.42202717e-01
-1.85919687e-01 -4.21343565e-01 6.93201363e-01 6.74972355e-01
9.39037740e-01 -6.78341627e-01 1.18616509e+00 3.80749047e-01
-1.49780095e+00 5.46336696e-02 -9.44377661e-01 -6.10272363e-02
2.01493561e-01 5.75472951e-01 -1.01112592e+00 1.07810521e+00
5.87593794e-01 8.56793761e-01 -7.59800911e-01 1.50268531e+00
-2.06156000e-01 -1.22292787e-01 -5.53686559e-01 2.22075701e-01
4.75474298e-01 -5.02785921e-01 6.44039154e-01 8.70084643e-01
7.54173160e-01 -6.77439496e-02 5.10799348e-01 5.15380204e-01
5.70953310e-01 -5.80312461e-02 -1.05247736e+00 4.96841639e-01
5.11469245e-01 1.44526672e+00 -7.40628004e-01 2.02296093e-01
-5.22603631e-01 1.13803995e+00 -3.40761721e-01 1.45991340e-01
-6.76365018e-01 -7.67849863e-01 5.53716362e-01 1.90283149e-01
1.74896717e-01 -1.18395889e+00 9.92579758e-02 -1.18249428e+00
-6.34188503e-02 -4.07376140e-01 -2.60324746e-01 -9.21126544e-01
-6.35206640e-01 8.72614384e-01 -2.21654296e-01 -1.99136937e+00
-2.78751999e-01 -7.02270508e-01 -1.51486069e-01 7.88305640e-01
-1.74168074e+00 -1.52915263e+00 -1.16067564e+00 8.45007420e-01
5.59838474e-01 -2.13598445e-01 9.41891074e-01 -2.03830466e-01
-1.81876227e-01 1.26396015e-01 4.08071101e-01 8.19672048e-02
8.75487089e-01 -9.12915051e-01 2.79615790e-01 1.23386896e+00
2.39101514e-01 5.28873563e-01 5.60900331e-01 -6.06653631e-01
-1.82318830e+00 -1.02202308e+00 4.79483753e-01 -3.53911042e-01
2.56345421e-01 -1.60274461e-01 -2.44479150e-01 5.50639749e-01
-1.74090639e-01 2.30394796e-01 3.09783816e-01 -5.25317550e-01
2.97724783e-01 -3.16407651e-01 -1.23077583e+00 3.42766225e-01
9.35614586e-01 -2.07220301e-01 -2.51019686e-01 3.97946000e-01
5.00647843e-01 -8.24720263e-01 -6.43560529e-01 4.54425156e-01
5.45979440e-01 -1.29108465e+00 1.11795008e+00 5.62950194e-01
-6.14339590e-01 -1.03857064e+00 -2.30702013e-01 -1.33071291e+00
-4.25083995e-01 -6.44193769e-01 5.20226836e-01 1.14136076e+00
-1.49200693e-01 -8.57922733e-01 5.42559147e-01 5.73074073e-02
-2.28221133e-01 -4.74372119e-01 -9.29636180e-01 -7.21845925e-01
-9.75940406e-01 -5.19790687e-02 5.33730388e-01 8.59502077e-01
-3.69626015e-01 3.42863388e-02 -3.10727388e-01 7.11281598e-01
8.49743903e-01 8.87954831e-02 1.23972237e+00 -1.49194837e+00
2.48395935e-01 4.97567691e-02 -1.10881615e+00 -1.32292306e+00
-2.37076193e-01 -2.58909345e-01 -1.87815093e-02 -1.80764449e+00
-6.90028667e-01 -6.93123996e-01 5.09762883e-01 3.18705320e-01
4.62236285e-01 3.28881323e-01 1.75011069e-01 4.75277752e-01
-4.06836361e-01 5.47131360e-01 9.74294662e-01 1.74598202e-01
-2.66145855e-01 2.81766176e-01 1.19363099e-01 1.17960942e+00
5.23567677e-01 2.93725617e-02 -3.29469472e-01 -7.02351511e-01
-1.04554452e-01 1.09685056e-01 5.25796473e-01 -1.63509786e+00
7.45778441e-01 -1.22121915e-01 7.72252142e-01 -9.93979275e-01
6.37908101e-01 -1.35328245e+00 1.91518545e-01 5.01981497e-01
5.87094486e-01 4.40272719e-01 2.08581522e-01 4.82210577e-01
-2.91141331e-01 -2.66633630e-01 6.85661018e-01 -1.58173516e-01
-1.18958056e+00 4.56517607e-01 1.18073549e-04 -6.50464773e-01
1.43501246e+00 -7.87750900e-01 -1.68536842e-01 -3.92369896e-01
-2.24441022e-01 5.62795587e-02 1.19220662e+00 1.25471830e-01
1.10873139e+00 -1.14596784e+00 -4.82991844e-01 8.77747357e-01
3.54208559e-01 6.07712507e-01 1.09776691e-01 1.14312112e+00
-1.51615918e+00 7.37897038e-01 -4.05281663e-01 -1.03608167e+00
-1.35361242e+00 2.21347913e-01 3.61383289e-01 3.71937692e-01
-6.41066730e-01 7.14708865e-01 8.69444013e-02 -4.63802576e-01
-7.81770423e-02 -4.67616916e-01 -7.35778213e-02 -2.74086982e-01
3.15372050e-01 4.61366683e-01 -5.90316430e-02 -1.07384539e+00
-6.17281139e-01 1.55700350e+00 3.39439034e-01 3.44958603e-02
1.11742640e+00 -5.79637885e-01 -1.61556229e-01 1.36134863e-01
8.03451121e-01 3.35082710e-01 -1.33347237e+00 -9.60126594e-02
-3.22089404e-01 -8.96997929e-01 2.85114139e-01 -2.49855295e-01
-7.06408560e-01 7.65649319e-01 9.17113483e-01 -5.65762147e-02
9.88104463e-01 -5.89841962e-01 4.16774839e-01 6.50614262e-01
1.19209540e+00 -6.79180861e-01 -6.16925359e-01 5.91081321e-01
7.88389206e-01 -1.33297980e+00 4.19262320e-01 -7.13850796e-01
-4.16710854e-01 1.64691246e+00 3.97003382e-01 -2.40437631e-02
3.45100999e-01 1.82020783e-01 1.73297092e-01 6.51719272e-02
2.18182400e-01 -2.09936276e-01 2.36429006e-01 1.09311509e+00
1.14778683e-01 -5.03626317e-02 -1.56213185e-02 -6.66959137e-02
-3.47555131e-01 -2.16039583e-01 5.59019327e-01 1.08410645e+00
-7.18622684e-01 -8.52259994e-01 -8.04284871e-01 -1.35761127e-01
2.76848108e-01 4.36792113e-02 -1.53556541e-01 1.02823019e+00
2.79897153e-01 9.71561611e-01 7.10657761e-02 -4.57488894e-01
5.24743438e-01 -5.62118948e-01 5.72730601e-01 -2.65336394e-01
1.11106243e-02 1.31449804e-01 -1.70181990e-01 -8.19966137e-01
-5.45148969e-01 -5.38429201e-01 -1.14886153e+00 -1.81944057e-01
-7.28685021e-01 7.85244778e-02 1.05334878e+00 5.35409093e-01
5.02665043e-01 1.75149441e-01 7.92341053e-01 -1.16419256e+00
-2.39738762e-01 -5.96794546e-01 -6.96633518e-01 -2.03215361e-01
5.17682314e-01 -9.23882663e-01 -1.65510803e-01 -8.70193541e-03] | [7.4730939865112305, -2.0389554500579834] |
05b4d60b-fff9-400c-b012-f23aa55a1a31 | relational-graph-learning-for-grounded-video | 2112.00967 | null | https://arxiv.org/abs/2112.00967v1 | https://arxiv.org/pdf/2112.00967v1.pdf | Relational Graph Learning for Grounded Video Description Generation | Grounded video description (GVD) encourages captioning models to attend to appropriate video regions (e.g., objects) dynamically and generate a description. Such a setting can help explain the decisions of captioning models and prevents the model from hallucinating object words in its description. However, such design mainly focuses on object word generation and thus may ignore fine-grained information and suffer from missing visual concepts. Moreover, relational words (e.g., "jump left or right") are usual spatio-temporal inference results, i.e., these words cannot be grounded on certain spatial regions. To tackle the above limitations, we design a novel relational graph learning framework for GVD, in which a language-refined scene graph representation is designed to explore fine-grained visual concepts. Furthermore, the refined graph can be regarded as relational inductive knowledge to assist captioning models in selecting the relevant information it needs to generate correct words. We validate the effectiveness of our model through automatic metrics and human evaluation, and the results indicate that our approach can generate more fine-grained and accurate description, and it solves the problem of object hallucination to some extent. | ['William Yang Wang', 'Yueting Zhuang', 'Jun Xiao', 'Haocheng Shi', 'Haizhou Shi', 'Siliang Tang', 'Xin Eric Wang', 'Wenqiao Zhang'] | 2021-12-02 | null | null | null | null | ['video-description'] | ['computer-vision'] | [-6.09920584e-02 3.13976198e-01 -4.72689718e-01 -3.26881677e-01
-4.38916683e-01 -3.73220205e-01 6.56022429e-01 1.40052542e-01
9.97443572e-02 5.89124382e-01 8.40520442e-01 -1.06882557e-01
4.31236327e-02 -9.90871727e-01 -8.35207224e-01 -4.66386944e-01
2.30369523e-01 3.97582442e-01 2.51521051e-01 -2.98687667e-01
2.58409411e-01 3.72309625e-01 -1.51225913e+00 5.94036579e-01
7.62398481e-01 6.79399550e-01 6.54326737e-01 1.51228875e-01
-4.56021607e-01 1.24214494e+00 -6.57978237e-01 -3.21415514e-01
-4.30549353e-01 -5.73569894e-01 -8.48255575e-01 4.87234592e-01
2.80388921e-01 -5.66453636e-01 -6.84039474e-01 1.05936241e+00
1.41109839e-01 3.11488569e-01 6.18423283e-01 -1.47883761e+00
-1.25406253e+00 8.22745681e-01 -3.42619210e-01 9.15124789e-02
7.14995384e-01 3.21699739e-01 1.00388348e+00 -1.02547312e+00
7.58800805e-01 1.57625818e+00 8.01435709e-02 7.06943572e-01
-1.04195380e+00 -6.02348268e-01 7.03240573e-01 4.21491355e-01
-1.73594952e+00 -3.65902215e-01 1.02899396e+00 -4.71031934e-01
6.65257514e-01 2.98865825e-01 8.94729972e-01 1.45785201e+00
-9.35162008e-02 9.07822073e-01 6.99622273e-01 1.10821594e-02
2.54645050e-01 1.29210442e-01 -2.88201213e-01 7.48468995e-01
3.31386030e-01 -6.16391227e-02 -6.12898469e-01 1.09517351e-01
1.21554220e+00 9.53224078e-02 -4.57701385e-01 -5.52739978e-01
-1.64944625e+00 8.79855514e-01 6.63458943e-01 2.11310923e-01
-4.72863108e-01 2.38069251e-01 2.76789874e-01 -4.04627204e-01
2.81234413e-01 4.76936549e-01 9.23346058e-02 1.26504809e-01
-6.91219330e-01 4.24667060e-01 3.49677265e-01 1.59438431e+00
8.38339329e-01 4.39596595e-03 -8.43785644e-01 6.20623469e-01
5.08974969e-01 5.19625545e-01 1.51446074e-01 -1.00961459e+00
6.35526597e-01 7.05587208e-01 1.97032452e-01 -1.47314811e+00
-1.50871113e-01 -1.80384547e-01 -9.31345701e-01 -3.36024702e-01
-6.22031093e-02 3.76106381e-01 -8.90455604e-01 1.83125758e+00
6.40532300e-02 3.24771464e-01 3.48673798e-02 1.50991833e+00
1.32427609e+00 1.01060879e+00 5.10182977e-01 -2.67149270e-01
1.44672239e+00 -9.31003869e-01 -1.01566386e+00 -4.06323671e-01
4.86478329e-01 -3.50981474e-01 1.30993450e+00 -3.28742228e-02
-1.06221151e+00 -6.88196421e-01 -7.48113930e-01 -1.64683849e-01
-1.96385369e-01 -1.81747600e-01 5.39056897e-01 -1.40068278e-01
-9.75492477e-01 1.36858732e-01 -3.81451726e-01 -3.77485454e-01
4.88524824e-01 -2.18032479e-01 -3.13755572e-01 -4.75973129e-01
-1.44105911e+00 8.05292428e-01 7.19204068e-01 1.69616252e-01
-1.21321523e+00 -3.30741048e-01 -1.15313387e+00 4.71979082e-02
6.86796784e-01 -9.17743504e-01 1.06923258e+00 -6.29496634e-01
-8.27667117e-01 7.81899035e-01 -3.43966156e-01 -1.46864817e-01
3.15424711e-01 1.79459378e-01 -4.71636683e-01 4.42283064e-01
4.45219606e-01 1.09975684e+00 6.60631776e-01 -1.75680780e+00
-3.57204348e-01 -8.38539228e-02 5.84867597e-01 4.64263916e-01
-2.02274561e-01 -2.23810852e-01 -7.87084103e-01 -8.32298756e-01
2.86084741e-01 -6.32249951e-01 -2.15825543e-01 1.84890285e-01
-6.02010369e-01 -3.37463826e-01 7.19065249e-01 -4.90994215e-01
1.30402899e+00 -2.09373426e+00 2.25359499e-01 -1.42433763e-01
4.82917637e-01 -3.85028608e-02 -2.00745150e-01 5.62059224e-01
4.44297716e-02 5.16312361e-01 -3.70066799e-02 -9.72853750e-02
-8.14061835e-02 4.55207914e-01 -4.44754064e-01 1.40684947e-01
2.92103142e-01 1.11105573e+00 -1.42186534e+00 -1.01864231e+00
4.19062972e-01 5.75320303e-01 -4.49696898e-01 4.26834166e-01
-6.00532293e-01 4.76957768e-01 -1.03343296e+00 5.72830737e-01
5.20935476e-01 -5.90692341e-01 -1.40758723e-01 -6.91564381e-01
5.41114397e-02 2.03542948e-01 -7.81120420e-01 1.84772503e+00
-4.29565072e-01 4.84418452e-01 -5.45768857e-01 -6.24444783e-01
1.09681559e+00 3.40348005e-01 1.90640271e-01 -8.05041850e-01
-1.36797994e-01 -3.05781662e-01 -4.38586026e-01 -9.32084024e-01
4.88193601e-01 -4.67659384e-02 -1.43087983e-01 2.68492937e-01
-3.18674117e-01 -1.86094314e-01 -5.90003617e-02 6.37166560e-01
6.84547246e-01 3.89578909e-01 3.77579898e-01 -4.87156250e-02
4.59189385e-01 2.45297685e-01 3.97359520e-01 7.47633636e-01
-1.77829519e-01 8.81497443e-01 4.94138926e-01 -4.36476141e-01
-1.01761627e+00 -9.96689260e-01 3.41518581e-01 9.36948895e-01
8.93240690e-01 -6.75170004e-01 -5.47950208e-01 -5.20711958e-01
-3.30732971e-01 1.18856132e+00 -6.98775828e-01 -5.30980051e-01
-4.56111372e-01 -1.59632698e-01 2.62812704e-01 6.33476198e-01
5.67454517e-01 -1.56470823e+00 -3.71334314e-01 1.23968929e-01
-8.57051551e-01 -1.24961829e+00 -5.70231676e-01 -5.21160662e-01
-6.21082246e-01 -8.98308635e-01 -6.17818832e-01 -9.74448860e-01
9.29417431e-01 7.10545242e-01 1.42080927e+00 2.31068328e-01
2.27167174e-01 3.25046420e-01 -7.82551169e-01 -7.55889788e-02
-3.23580682e-01 -3.04378599e-01 -2.10498258e-01 -3.79577614e-02
3.36000085e-01 -2.27510333e-01 -6.60778105e-01 3.92087340e-01
-1.01330781e+00 8.90578926e-01 4.80343878e-01 5.24926960e-01
6.98237598e-01 -5.00339121e-02 4.74309385e-01 -6.97556257e-01
6.65336549e-01 -7.25988805e-01 -1.69004157e-01 3.88349146e-01
-3.71653348e-01 2.20854059e-01 6.15502000e-01 -5.74726582e-01
-9.93997872e-01 -2.18192071e-01 1.57546401e-01 -9.34238970e-01
-1.97707593e-01 6.07945263e-01 -4.50045139e-01 3.91036481e-01
5.27028859e-01 4.79023963e-01 -2.66726404e-01 -1.76380798e-01
5.56882381e-01 4.26418513e-01 3.61902773e-01 -6.67475104e-01
7.33902752e-01 4.44907784e-01 -1.43265933e-01 -4.35058624e-01
-1.05805695e+00 -2.85783917e-01 -4.25837636e-01 -4.53337550e-01
1.03762186e+00 -1.09559536e+00 -5.31463683e-01 -8.77280310e-02
-1.58724940e+00 -4.02294286e-02 -1.92174703e-01 3.48040223e-01
-7.77459204e-01 2.31958404e-01 -3.51918370e-01 -6.47846520e-01
1.94350496e-01 -1.08472967e+00 1.35770321e+00 2.17635661e-01
-2.49909326e-01 -9.46314573e-01 -3.59565526e-01 3.61629784e-01
7.15290755e-02 3.71189952e-01 1.03370357e+00 -3.68855029e-01
-9.32297528e-01 1.45922184e-01 -5.43433011e-01 -1.43920302e-01
1.46860592e-02 -1.40065178e-01 -6.15460157e-01 -8.11770931e-03
-1.42588928e-01 -3.22648942e-01 5.34986675e-01 9.82991606e-02
1.56234300e+00 -7.59110987e-01 -5.10383666e-01 3.22002202e-01
1.30552578e+00 8.66167173e-02 8.49420369e-01 1.94672942e-01
1.23702645e+00 6.77138507e-01 9.47503626e-01 4.80023563e-01
7.62900233e-01 7.51306236e-01 7.46961594e-01 -1.47625417e-01
-3.10489148e-01 -9.79427516e-01 6.93321228e-02 4.56601053e-01
-7.02456385e-02 -6.73143208e-01 -8.82149756e-01 6.93996549e-01
-2.12689161e+00 -1.07653522e+00 -8.10421556e-02 1.80247176e+00
6.85647905e-01 -1.54572040e-01 -1.43858612e-01 -2.77355880e-01
1.06965709e+00 4.93729711e-01 -5.19030690e-01 -2.04279236e-02
-1.71056733e-01 -5.01586258e-01 4.95595746e-02 3.89026523e-01
-5.36173522e-01 1.30895865e+00 5.51415110e+00 7.32319534e-01
-7.85563350e-01 -3.97437206e-03 5.09261727e-01 1.26606926e-01
-9.22400236e-01 1.10025391e-01 -5.52295148e-01 4.33666527e-01
3.41311753e-01 -2.81829685e-01 4.29538876e-01 7.55203962e-01
5.51669598e-01 2.69338697e-01 -1.20864725e+00 1.27113199e+00
3.56315762e-01 -1.49682307e+00 1.02324498e+00 -6.15025014e-02
5.18040478e-01 -6.28726065e-01 -1.91665158e-01 3.56124759e-01
1.18129268e-01 -1.22780418e+00 1.03980267e+00 7.46073604e-01
7.96361208e-01 -6.70529008e-01 5.42706370e-01 4.41950530e-01
-1.35151911e+00 2.19687283e-01 -5.83596587e-01 6.99214712e-02
3.40512276e-01 2.35625491e-01 -7.92502820e-01 4.43480819e-01
4.77306455e-01 8.77963126e-01 -6.08181536e-01 8.11093748e-01
-5.30543387e-01 3.39478761e-01 3.29708457e-01 -3.11033309e-01
4.22337353e-01 -6.91688284e-02 5.72015762e-01 9.93629038e-01
4.70904082e-01 5.58747053e-01 2.94652730e-01 1.37858164e+00
8.30890611e-02 9.09500569e-02 -8.74208868e-01 -9.44929421e-02
6.72951162e-01 9.72805798e-01 -6.45357251e-01 -4.36040342e-01
-3.85527790e-01 8.95984471e-01 4.48623031e-01 7.86843896e-01
-1.08579087e+00 1.61414966e-02 4.72805560e-01 6.25360072e-01
2.16890499e-01 -8.51342082e-02 7.13955760e-02 -1.26400697e+00
-4.01936546e-02 -6.81105494e-01 1.98737174e-01 -1.66086757e+00
-1.15249753e+00 6.99952960e-01 4.03884411e-01 -1.44582415e+00
-2.85287917e-01 -2.72283286e-01 -4.05857235e-01 6.60104215e-01
-1.22890913e+00 -1.29921472e+00 -7.68657863e-01 7.87999988e-01
6.12956285e-01 1.98520422e-01 5.74916720e-01 2.01737266e-02
-2.04312459e-01 1.90144733e-01 -7.91245520e-01 1.58538163e-01
4.02730972e-01 -8.48230720e-01 2.64624417e-01 7.41252124e-01
2.61520505e-01 6.26205266e-01 9.13174510e-01 -8.24623168e-01
-1.29864001e+00 -1.42785966e+00 1.00535452e+00 -3.84510636e-01
4.27986741e-01 -4.16026592e-01 -1.14850044e+00 7.39399672e-01
-8.30321833e-02 1.13082372e-01 3.80287498e-01 -7.07533583e-02
-4.14144725e-01 1.36787996e-01 -8.84587348e-01 1.05261612e+00
1.61442244e+00 -6.15383208e-01 -7.66446114e-01 5.41688085e-01
1.15965772e+00 -3.86848807e-01 -4.64111775e-01 2.43642211e-01
7.05340281e-02 -9.36428845e-01 1.02584791e+00 -6.80333078e-01
7.10110843e-01 -6.69015169e-01 -1.22209199e-01 -1.23035514e+00
-7.09905922e-01 -2.51452684e-01 -3.73157531e-01 1.35086453e+00
2.67614335e-01 -8.36911723e-02 5.01510084e-01 4.86425400e-01
-2.99213439e-01 -6.15620255e-01 -4.88105804e-01 -5.29732287e-01
-4.73779023e-01 -4.81151968e-01 9.71229672e-01 1.02288532e+00
-4.21353057e-02 5.08755982e-01 -5.66510975e-01 2.77761549e-01
4.45182323e-01 2.51978815e-01 4.82967228e-01 -9.63864565e-01
1.58073649e-01 -2.65069366e-01 -5.52567363e-01 -1.31123388e+00
3.35365385e-01 -7.80246258e-01 1.91301063e-01 -2.12408257e+00
3.95061135e-01 -3.99151474e-01 -1.68919310e-01 6.54184759e-01
-3.30096930e-01 1.76902100e-01 2.21264035e-01 4.27192748e-01
-9.53211546e-01 7.34810174e-01 1.97800910e+00 -2.61720866e-01
3.17727472e-03 -3.44138712e-01 -1.08473516e+00 5.08642554e-01
5.03123760e-01 -3.15454185e-01 -8.74091506e-01 -6.33458853e-01
4.21712726e-01 4.97263938e-01 8.22649479e-01 -7.45345533e-01
1.92955151e-01 -6.13177240e-01 3.84386271e-01 -5.18188357e-01
4.00757790e-01 -8.09382915e-01 2.69783884e-01 1.32029327e-02
-5.37106693e-01 3.65439840e-02 -5.05182035e-02 6.57881141e-01
-4.09547567e-01 1.73187666e-02 3.57393295e-01 -4.11682129e-01
-1.09669721e+00 7.30639458e-01 -2.06456259e-01 1.03026070e-01
1.14822507e+00 -5.23459375e-01 -3.20908368e-01 -7.85326302e-01
-6.70707881e-01 4.89129752e-01 6.80256426e-01 9.26886201e-01
1.16661811e+00 -1.84464633e+00 -8.03733587e-01 -4.50667031e-02
8.33268821e-01 3.88913244e-01 3.99723202e-01 2.10505724e-01
-4.59120005e-01 3.48083496e-01 -1.50570318e-01 -6.95086122e-01
-7.70530045e-01 1.05826628e+00 1.82124823e-01 2.95955777e-01
-8.09873998e-01 6.53851449e-01 7.67046154e-01 1.10487424e-01
1.20914638e-01 -3.66587013e-01 -6.13459826e-01 -5.42712994e-02
5.90844929e-01 -1.31058484e-01 -4.98480827e-01 -9.90775824e-01
-3.68966460e-01 4.74574566e-01 1.27097815e-02 -3.62386480e-02
8.91072392e-01 -4.34051007e-01 -3.83863114e-02 5.71366847e-01
1.03293145e+00 -1.99474737e-01 -1.41294372e+00 -2.38211557e-01
-4.26064759e-01 -6.44987464e-01 2.14542914e-03 -5.68173409e-01
-8.90685141e-01 8.30588758e-01 -4.00093980e-02 1.69957682e-01
9.36691999e-01 5.12847364e-01 5.49278080e-01 2.04154775e-01
5.11679947e-01 -6.21239960e-01 4.27188158e-01 2.97419399e-01
1.22580028e+00 -1.08426738e+00 -2.37079151e-03 -6.47311151e-01
-1.08829463e+00 8.77140164e-01 1.07156157e+00 1.43806711e-01
1.15746506e-01 -3.55151623e-01 -8.21611285e-02 -5.14390647e-01
-7.25261867e-01 -2.72018909e-01 3.77100319e-01 7.82259643e-01
2.78147101e-01 7.49641955e-02 -3.41363773e-02 5.63586771e-01
-4.42474894e-02 -1.69886634e-01 4.83509570e-01 5.08243740e-01
-3.96689355e-01 -4.31143016e-01 -2.64326721e-01 1.28635406e-01
2.61996061e-01 -2.75600225e-01 -3.35870236e-01 8.09843719e-01
1.99302241e-01 9.71348107e-01 1.69768408e-01 -3.24085623e-01
3.45991075e-01 -3.67647231e-01 2.75781602e-01 -8.94546211e-01
2.65176278e-02 -1.68853074e-01 -8.22362006e-02 -6.83071554e-01
-5.61861157e-01 -2.50323772e-01 -1.50448370e+00 -3.16118449e-01
-7.62642324e-02 2.68975019e-01 2.15384707e-01 1.06479347e+00
2.96528131e-01 6.70596004e-01 4.81343806e-01 -6.32274449e-01
-1.41573139e-02 -7.06465483e-01 -4.33727622e-01 7.20615327e-01
2.54167736e-01 -7.60478377e-01 -2.44778037e-01 1.99404821e-01] | [10.636953353881836, 1.1099368333816528] |
0e90c9a1-79a6-48fb-ba9a-b3bad837c749 | temporal-relational-ranking-for-stock | 1809.09441 | null | http://arxiv.org/abs/1809.09441v1 | http://arxiv.org/pdf/1809.09441v1.pdf | Temporal Relational Ranking for Stock Prediction | Stock prediction aims to predict the future trends of a stock in order to
help investors to make good investment decisions. Traditional solutions for
stock prediction are based on time-series models. With the recent success of
deep neural networks in modeling sequential data, deep learning has become a
promising choice for stock prediction.
However, most existing deep learning solutions are not optimized towards the
target of investment, i.e., selecting the best stock with the highest expected
revenue. Specifically, they typically formulate stock prediction as a
classification (to predict stock trend) or a regression problem (to predict
stock price). More importantly, they largely treat the stocks as independent of
each other. The valuable signal in the rich relations between stocks (or
companies), such as two stocks are in the same sector and two companies have a
supplier-customer relation, is not considered.
In this work, we contribute a new deep learning solution, named Relational
Stock Ranking (RSR), for stock prediction. Our RSR method advances existing
solutions in two major aspects: 1) tailoring the deep learning models for stock
ranking, and 2) capturing the stock relations in a time-sensitive manner. The
key novelty of our work is the proposal of a new component in neural network
modeling, named Temporal Graph Convolution, which jointly models the temporal
evolution and relation network of stocks. To validate our method, we perform
back-testing on the historical data of two stock markets, NYSE and NASDAQ.
Extensive experiments demonstrate the superiority of our RSR method. It
outperforms state-of-the-art stock prediction solutions achieving an average
return ratio of 98% and 71% on NYSE and NASDAQ, respectively. | ['Tat-Seng Chua', 'Yiqun Liu', 'Cheng Luo', 'Xiang Wang', 'Fuli Feng', 'Xiangnan He'] | 2018-09-25 | null | null | null | null | ['stock-market-prediction', 'stock-prediction'] | ['time-series', 'time-series'] | [-7.96606600e-01 -4.65519547e-01 -5.80464840e-01 -2.10277393e-01
1.96502674e-02 -4.32705730e-01 5.73773026e-01 -6.61398098e-03
-1.66203193e-02 5.58300555e-01 1.70062542e-01 -5.19666255e-01
-4.05407250e-01 -1.30935347e+00 -6.18979275e-01 -5.30528247e-01
-4.54285741e-01 4.16476637e-01 2.96537668e-01 -6.73437536e-01
4.77255106e-01 6.87076390e-01 -1.10219216e+00 -2.91607231e-02
4.39103067e-01 1.62739515e+00 -1.15935400e-01 6.61746934e-02
-6.19209945e-01 1.67813110e+00 -6.14703357e-01 -7.74263620e-01
6.64102316e-01 -2.11405054e-01 -3.19462985e-01 -1.83319226e-01
-6.54180944e-02 -6.22873724e-01 -6.89334154e-01 8.58562052e-01
1.93542570e-01 -2.11290076e-01 4.13256407e-01 -1.36050320e+00
-7.58700609e-01 9.92773116e-01 -6.64100409e-01 7.18022346e-01
-5.65194786e-01 -9.42361727e-02 1.51230991e+00 -5.90835273e-01
3.56828690e-01 6.42547786e-01 6.55480146e-01 4.72104326e-02
-6.99390411e-01 -1.04064167e+00 5.01122713e-01 1.31404102e-01
-9.55979466e-01 3.64699811e-02 1.07806146e+00 -5.45749784e-01
9.01488006e-01 -3.75985317e-02 1.13759208e+00 5.82385004e-01
5.50630033e-01 9.92147386e-01 8.24976802e-01 2.81722218e-01
2.09376544e-01 2.99378056e-02 2.07580909e-01 2.88024902e-01
4.86193687e-01 3.07034701e-01 -6.52798235e-01 9.67039540e-02
9.04633105e-01 5.92647433e-01 4.67397235e-02 -1.49308309e-01
-1.00799131e+00 9.67271447e-01 7.47705221e-01 6.03805542e-01
-8.56841028e-01 3.27068835e-01 2.64322817e-01 7.06205666e-01
8.51926744e-01 3.72568935e-01 -6.70343280e-01 -1.75467471e-03
-1.06715119e+00 4.22444463e-01 8.64446938e-01 7.02717781e-01
3.31219167e-01 7.74153948e-01 4.46878374e-02 2.72266507e-01
3.67803514e-01 2.32299492e-01 8.42653632e-01 -2.53834814e-01
6.04837477e-01 7.88280845e-01 1.34840235e-01 -1.14140439e+00
-6.76747084e-01 -1.15336323e+00 -8.23454559e-01 9.31768939e-02
9.03367177e-02 -2.56194651e-01 -6.53609097e-01 1.18162286e+00
-1.95784807e-01 7.11793959e-01 2.84140587e-01 7.03995109e-01
6.77136004e-01 1.01279306e+00 -3.59667003e-01 -2.95347691e-01
8.11678529e-01 -8.58070254e-01 -6.39549017e-01 -1.56322554e-01
4.49160248e-01 -3.51370513e-01 1.80012092e-01 2.80727535e-01
-9.69923019e-01 -3.00825477e-01 -1.12099349e+00 6.19573712e-01
-5.68887413e-01 -1.55780360e-01 8.70674551e-01 1.73714817e-01
-1.08937073e+00 1.00963187e+00 -5.76941729e-01 4.03038532e-01
2.71968991e-01 5.68393052e-01 2.69691408e-01 4.56277698e-01
-1.60218298e+00 8.42323720e-01 4.33699846e-01 2.61405885e-01
-5.71309805e-01 -9.12323296e-01 -4.07209486e-01 4.14395481e-01
5.53041995e-01 -2.45395735e-01 1.27383232e+00 -1.03838146e+00
-1.35011959e+00 3.46107155e-01 5.49058378e-01 -1.10527265e+00
5.39300263e-01 -7.06985667e-02 -7.96826065e-01 -4.56793875e-01
-9.89345238e-02 -8.49363208e-02 6.67751253e-01 -8.64826441e-01
-1.01772583e+00 -2.71832108e-01 1.40962198e-01 -1.28204882e-01
-5.93298018e-01 -1.07834503e-01 -2.22837508e-01 -9.88393903e-01
5.45344763e-02 -6.64838195e-01 -1.55943736e-01 -8.30715358e-01
-2.65537024e-01 -3.42251807e-01 6.15356088e-01 -6.26116037e-01
1.49494767e+00 -1.85724008e+00 -1.63301587e-01 3.31185549e-01
3.46620262e-01 3.43427986e-01 4.31127660e-02 6.79732203e-01
-4.09528345e-01 8.92328471e-02 2.17375994e-01 -1.74155131e-01
1.45480052e-01 -8.92268792e-02 -8.62590909e-01 3.19305211e-01
2.66414672e-01 1.41832495e+00 -6.70842230e-01 7.08070397e-02
-1.25420570e-01 2.47579161e-02 1.65596023e-01 1.22166164e-01
-3.49752069e-01 -2.35434309e-01 -4.11238194e-01 7.62169659e-01
5.63141525e-01 -4.98960048e-01 4.99259718e-02 7.08904415e-02
-4.74596292e-01 4.94317144e-01 -1.16285872e+00 7.03608930e-01
-1.26452282e-01 5.13563633e-01 -5.45141935e-01 -9.77673948e-01
1.32762027e+00 1.66571006e-01 9.22063529e-01 -9.91351068e-01
9.45297182e-02 5.82113862e-01 1.35020956e-01 1.39964744e-01
6.41911507e-01 -3.89621437e-01 1.71925388e-02 6.73345268e-01
-1.83578044e-01 1.96826637e-01 3.43039632e-01 8.90591145e-02
9.19451833e-01 -2.27064580e-01 6.05542511e-02 -7.50001967e-02
8.32620859e-02 -1.95256472e-01 6.86222970e-01 2.85114437e-01
5.50386719e-02 2.17037514e-01 9.91544843e-01 -8.27677906e-01
-7.40668774e-01 -6.85979903e-01 2.48926461e-01 6.55384421e-01
-4.67962362e-02 1.53486263e-02 1.27299815e-01 -7.07486212e-01
5.81371069e-01 7.58391678e-01 -6.96637750e-01 -1.20395809e-01
-4.83975738e-01 -9.18188035e-01 1.05824322e-02 7.64775634e-01
5.26094496e-01 -1.19977343e+00 -2.85755277e-01 5.57508945e-01
5.41577697e-01 -6.29351556e-01 -3.11833620e-01 4.15626287e-01
-9.68629360e-01 -1.05066681e+00 -1.08401859e+00 -4.48754519e-01
3.34744342e-02 2.68306494e-01 1.33494270e+00 1.51947394e-01
5.02394974e-01 -3.49203870e-02 -2.41469711e-01 -8.86402726e-01
-2.42061000e-02 2.23826081e-01 2.62679234e-02 3.60470384e-01
3.89567256e-01 -5.50615907e-01 -5.78811228e-01 1.72910631e-01
-8.62863421e-01 -3.66177827e-01 8.32816780e-01 6.77148223e-01
4.05801862e-01 8.50031316e-01 1.03318191e+00 -9.35230613e-01
8.57491910e-01 -8.91156614e-01 -1.35091925e+00 2.37950742e-01
-1.30621958e+00 5.42248599e-02 4.06492233e-01 -2.78752834e-01
-5.89705586e-01 -2.42631003e-01 1.73044145e-01 -6.87977135e-01
7.36701906e-01 1.29859471e+00 4.76623863e-01 2.10303038e-01
-1.78679302e-01 6.27230108e-01 -5.72610050e-02 -3.46326560e-01
-1.20048858e-01 1.75673380e-01 2.13450998e-01 2.32522562e-01
1.15234864e+00 1.92394882e-01 2.65514523e-01 -1.12845913e-01
-7.83401608e-01 -4.82837349e-01 -5.26300371e-01 -3.18145424e-01
4.38008159e-01 -1.04413843e+00 -7.38630772e-01 9.07119393e-01
-8.12886059e-01 -2.61600822e-01 -2.09411845e-01 4.47480410e-01
-1.34781286e-01 -2.13928714e-01 -8.14832330e-01 -1.04012883e+00
-4.79500532e-01 -7.06276357e-01 5.83296120e-01 1.96124911e-01
2.37469733e-01 -1.22545481e+00 1.71717152e-01 -4.97832112e-02
6.65575564e-01 2.24254400e-01 8.12116086e-01 -1.09881318e+00
-1.05254042e+00 -6.48825407e-01 -2.97970504e-01 4.61692125e-01
2.48213321e-01 3.75603773e-02 -4.68422741e-01 -1.58644542e-01
-2.80774161e-02 -2.63375398e-02 9.84280646e-01 5.93567550e-01
5.33604443e-01 -2.65261173e-01 8.33933726e-02 4.31107908e-01
1.58654392e+00 8.60056758e-01 6.75830781e-01 7.46328115e-01
5.33972919e-01 6.81312919e-01 6.98468745e-01 6.39659345e-01
3.45811486e-01 3.82881939e-01 6.68441117e-01 1.31722063e-01
5.17717779e-01 -2.25932896e-01 4.66987550e-01 1.09311676e+00
-2.56358176e-01 -3.87390405e-01 -9.19720352e-01 4.33921903e-01
-2.03607750e+00 -1.08607984e+00 -6.07827725e-03 1.82215226e+00
3.47647846e-01 7.93700993e-01 5.38308859e-01 1.25750825e-01
2.99720973e-01 5.82270861e-01 -8.10333252e-01 2.91919291e-01
-1.76423326e-01 4.44142260e-02 9.82095420e-01 -2.13841442e-03
-1.05386043e+00 8.29148710e-01 5.59429836e+00 4.56238598e-01
-1.44881690e+00 -3.32904845e-01 8.57431293e-01 -3.17284226e-01
-4.30377573e-01 -3.69945914e-01 -1.07771635e+00 7.56807029e-01
1.05613625e+00 -6.15611315e-01 1.90182090e-01 9.63357031e-01
6.27496168e-02 4.54576582e-01 -8.06475580e-01 7.78990448e-01
-2.04522714e-01 -1.98583806e+00 -2.87712459e-02 2.65860707e-01
6.89186573e-01 1.48130298e-01 4.51322913e-01 6.18486464e-01
4.12603617e-01 -7.73604512e-01 1.03608346e+00 9.41375375e-01
6.38974365e-03 -1.21883011e+00 1.22867203e+00 2.31337190e-01
-1.45698524e+00 -4.47993070e-01 -3.00634474e-01 -1.34154305e-01
7.55123943e-02 5.87739110e-01 -5.02095401e-01 7.72380829e-01
6.99953854e-01 1.25294662e+00 -3.10540646e-01 9.03254390e-01
1.29201775e-02 5.01232445e-01 1.27689227e-01 -4.64950919e-01
5.94513059e-01 -5.15249610e-01 1.61930904e-01 4.89359558e-01
5.89900911e-01 5.56904711e-02 -1.29355475e-01 8.19822609e-01
-2.93914258e-01 2.13975850e-02 -6.95813596e-01 -6.80047929e-01
2.14427069e-01 8.93608868e-01 -6.74241662e-01 -1.72062308e-01
-6.42458141e-01 3.44804585e-01 9.06624016e-04 2.53948987e-01
-6.45747602e-01 -3.88948798e-01 6.94059432e-01 3.81277144e-01
6.11440420e-01 -1.94552720e-01 -2.77800202e-01 -1.03190255e+00
6.14415444e-02 -6.72488034e-01 2.95743942e-01 -6.34023368e-01
-1.44023681e+00 5.95293760e-01 -3.32357883e-01 -1.54004300e+00
-4.87806410e-01 -8.14151049e-01 -8.58231246e-01 9.41512048e-01
-2.07426810e+00 -8.87506545e-01 1.17428638e-01 3.52346569e-01
2.82621622e-01 -9.99140620e-01 7.23959208e-02 3.88165623e-01
-6.75556719e-01 1.56543329e-01 3.01516980e-01 5.06136298e-01
-6.32567182e-02 -1.45583677e+00 9.41451550e-01 8.80604863e-01
2.65218556e-01 3.67512643e-01 2.60338664e-01 -1.01763344e+00
-1.33819520e+00 -1.20431554e+00 9.56085742e-01 -4.54465263e-02
1.50397551e+00 -6.96463138e-02 -9.49637353e-01 8.62087965e-01
9.05195475e-02 -2.05235526e-01 4.51618433e-01 4.44711521e-02
-1.07949093e-01 -6.71801507e-01 -5.86728930e-01 4.58057553e-01
5.79687953e-01 -1.71993092e-01 -2.97186852e-01 2.16053575e-01
1.02154088e+00 -1.14110731e-01 -9.39631164e-01 3.20729434e-01
3.61317575e-01 -1.24218917e+00 7.88528323e-01 -7.74051845e-01
4.36914742e-01 3.78297679e-02 8.24479908e-02 -1.46063221e+00
-6.21641099e-01 -5.22429585e-01 -6.10048354e-01 1.23626530e+00
4.75157022e-01 -9.43999529e-01 1.13016009e+00 5.66423655e-01
-8.06418210e-02 -1.12898719e+00 -7.34935105e-01 -1.05419266e+00
-4.30110954e-02 -2.21828848e-01 1.30034900e+00 1.05780828e+00
-4.57070082e-01 1.80870891e-01 -6.27642035e-01 5.95567450e-02
2.17302769e-01 6.36099577e-01 5.26587486e-01 -1.54510748e+00
-2.71469533e-01 -1.04653394e+00 -4.64846104e-01 -7.41734207e-01
2.60375470e-01 -6.80678129e-01 -5.10599494e-01 -1.62203360e+00
-2.88290679e-01 -4.49713081e-01 -1.12089384e+00 3.81161124e-01
1.58887908e-01 -3.32690507e-01 3.03870618e-01 5.17140210e-01
-2.41471753e-01 5.65111816e-01 9.97223258e-01 -4.34370816e-01
-2.56819606e-01 8.33886325e-01 -7.78946817e-01 3.88493061e-01
9.72763777e-01 -2.43136212e-01 -4.25064892e-01 -1.18974470e-01
7.68922806e-01 3.21313828e-01 -1.77033767e-02 -7.38612831e-01
1.49739176e-01 -3.07605565e-01 2.66501725e-01 -1.16053295e+00
1.69364527e-01 -8.34902585e-01 3.41834784e-01 8.27155411e-01
-5.67356087e-02 7.35028684e-01 3.99780944e-02 5.93362689e-01
-7.21794605e-01 -1.95613116e-01 3.53474379e-01 -1.69229731e-01
-1.01427972e+00 8.36706281e-01 -1.20327309e-01 -3.50740075e-01
1.07423604e+00 2.07848117e-01 -3.88034225e-01 -7.33197153e-01
-2.90747166e-01 4.56869453e-01 -1.23548657e-01 5.19190311e-01
7.73046255e-01 -1.60816050e+00 -7.49516845e-01 -7.17343837e-02
-8.76049101e-02 -2.35090271e-01 -8.05936381e-02 6.20458007e-01
-5.59702933e-01 8.10289383e-01 -5.34181073e-02 8.86403844e-02
-6.48049593e-01 7.70900190e-01 5.83096862e-01 -8.14993143e-01
-3.30111861e-01 9.38287854e-01 -1.26216069e-01 1.01697303e-01
3.44422251e-01 -5.43588758e-01 -5.69163203e-01 7.64590383e-01
4.37166572e-01 3.98097843e-01 2.74836034e-01 -4.20815557e-01
-2.10108176e-01 3.45715046e-01 -1.71284914e-01 1.75571397e-01
2.05559754e+00 3.46958488e-01 -3.00593525e-01 9.10974085e-01
1.02506483e+00 -1.63567960e-01 -1.13246322e+00 -4.35994536e-01
6.08201683e-01 -1.68109700e-01 3.57065201e-01 -5.72738647e-01
-1.98919225e+00 4.94368464e-01 4.07142699e-01 9.41366494e-01
1.02160025e+00 -2.55301744e-01 1.10144222e+00 3.43669027e-01
3.54006559e-01 -9.88198817e-01 2.43972465e-01 5.66723049e-01
8.81021619e-01 -1.21769035e+00 3.48513126e-02 1.45094320e-01
-6.45202875e-01 1.29852057e+00 4.61753309e-01 -3.12827528e-01
1.21871257e+00 2.13834092e-01 2.22107589e-01 -5.71129143e-01
-9.74453151e-01 -1.93053946e-01 5.31899869e-01 -8.91529769e-02
3.19731802e-01 -6.18600100e-02 8.49755779e-02 7.66644716e-01
-3.00726295e-01 4.09911610e-02 5.19936204e-01 9.46809113e-01
-3.38568211e-01 -9.23592031e-01 5.95080033e-02 9.60381508e-01
-5.85228384e-01 -2.79051304e-01 -4.84668672e-01 9.66495693e-01
-3.54507715e-01 5.11060476e-01 4.53709334e-01 -7.91997254e-01
5.40606141e-01 -3.10641881e-02 -2.61995852e-01 -5.06470203e-01
-7.23631263e-01 9.15282667e-02 -1.51948556e-01 -2.94066489e-01
-2.65011609e-01 -7.55360186e-01 -1.15299022e+00 -6.22850597e-01
-2.57264316e-01 5.33006303e-02 4.89882827e-01 7.72413731e-01
1.63546145e-01 9.04053807e-01 1.18030512e+00 -5.90770125e-01
-8.34562480e-01 -7.62492716e-01 -1.46807659e+00 8.41426384e-03
3.90906334e-01 -6.77497804e-01 -2.49687761e-01 -5.02649903e-01] | [4.352682113647461, 4.315592288970947] |
538f223b-32f4-4f42-917b-7926fe1e0f0d | a-vector-quantized-approach-for-text-to | 2302.04215 | null | https://arxiv.org/abs/2302.04215v1 | https://arxiv.org/pdf/2302.04215v1.pdf | A Vector Quantized Approach for Text to Speech Synthesis on Real-World Spontaneous Speech | Recent Text-to-Speech (TTS) systems trained on reading or acted corpora have achieved near human-level naturalness. The diversity of human speech, however, often goes beyond the coverage of these corpora. We believe the ability to handle such diversity is crucial for AI systems to achieve human-level communication. Our work explores the use of more abundant real-world data for building speech synthesizers. We train TTS systems using real-world speech from YouTube and podcasts. We observe the mismatch between training and inference alignments in mel-spectrogram based autoregressive models, leading to unintelligible synthesis, and demonstrate that learned discrete codes within multiple code groups effectively resolves this issue. We introduce our MQTTS system whose architecture is designed for multiple code generation and monotonic alignment, along with the use of a clean silence prompt to improve synthesis quality. We conduct ablation analyses to identify the efficacy of our methods. We show that MQTTS outperforms existing TTS systems in several objective and subjective measures. | ['Alexander Rudnicky', 'Shinji Watanabe', 'Li-Wei Chen'] | 2023-02-08 | null | null | null | null | ['text-to-speech-synthesis', 'speech-synthesis'] | ['speech', 'speech'] | [ 3.28623772e-01 2.81226963e-01 1.56149240e-02 -5.48866808e-01
-1.38359058e+00 -7.11082220e-01 6.95278823e-01 -3.21771652e-01
-2.72079688e-02 6.55769944e-01 6.36712551e-01 -7.61567116e-01
3.66416305e-01 -7.51031190e-02 -5.25859833e-01 -3.37494493e-01
1.69340502e-02 3.32700104e-01 2.65300721e-01 -5.44677615e-01
-8.61507608e-04 8.74858126e-02 -1.33971429e+00 4.34609562e-01
1.00075448e+00 5.84799826e-01 4.43864793e-01 1.27074468e+00
-4.50794622e-02 1.12902200e+00 -1.23384821e+00 -3.52824003e-01
2.28160456e-01 -8.73453438e-01 -6.20020509e-01 3.44982520e-02
4.93777066e-01 -4.80514795e-01 -4.10666466e-01 6.73791528e-01
6.19216979e-01 1.97000116e-01 5.15386879e-01 -1.13355875e+00
-5.41194677e-01 9.42238688e-01 3.06262933e-02 2.13128120e-01
4.83046263e-01 3.93383056e-01 1.21948421e+00 -5.59840500e-01
3.82330507e-01 1.37841499e+00 4.44925040e-01 8.09233129e-01
-1.26701164e+00 -6.86933815e-01 -1.83114916e-01 -2.89734542e-01
-1.18563175e+00 -1.37887371e+00 3.64267260e-01 -2.78982520e-01
1.32082558e+00 5.18988132e-01 5.17519951e-01 1.52514887e+00
5.50868437e-02 7.70046949e-01 8.27079356e-01 -6.40434802e-01
2.43688717e-01 -7.72951916e-02 -4.02875543e-01 5.31162381e-01
-4.30225790e-01 1.14981458e-01 -9.83605683e-01 -5.54680228e-02
6.25843823e-01 -7.53422558e-01 -3.47401768e-01 3.60489041e-01
-1.35361326e+00 6.64712250e-01 -9.28190500e-02 3.16609055e-01
-8.17072988e-02 2.18857259e-01 3.64173889e-01 5.92294633e-01
4.43429738e-01 7.20643222e-01 -4.55454499e-01 -9.06350315e-01
-1.30351865e+00 1.61790758e-01 1.12499630e+00 1.30166554e+00
1.08778812e-01 8.89985740e-01 -2.65843645e-02 1.20171034e+00
2.64780253e-01 7.92066693e-01 7.38770127e-01 -1.27032053e+00
6.56062067e-01 -2.20743313e-01 7.31592393e-03 -7.43606508e-01
-1.77018233e-02 -3.13032538e-01 -3.45671505e-01 9.58223417e-02
3.88019979e-01 -4.51021880e-01 -9.67293739e-01 1.73734498e+00
-1.32066593e-01 1.08653285e-01 2.25006253e-01 5.36453068e-01
5.57335436e-01 9.45111394e-01 -2.04129189e-01 -2.57260472e-01
8.45485330e-01 -1.13573802e+00 -9.54633415e-01 -3.00422341e-01
6.39322817e-01 -1.12869895e+00 1.36912346e+00 3.46907675e-01
-1.28963137e+00 -4.66401726e-01 -1.13779783e+00 -3.18211317e-02
1.91912740e-01 6.63414001e-02 2.15689763e-01 9.23099637e-01
-1.20297956e+00 4.94840652e-01 -9.96974766e-01 -1.91558808e-01
-1.86075449e-01 2.79224724e-01 -5.55300862e-02 4.78556007e-01
-1.13859737e+00 8.33103240e-01 -5.40244319e-02 -3.28526556e-01
-7.61262596e-01 -4.16408956e-01 -8.88250530e-01 -4.87852171e-02
1.25257894e-01 -3.09115946e-01 2.06630707e+00 -1.11710083e+00
-2.04352331e+00 3.96508455e-01 -3.60732108e-01 -7.69935250e-01
2.76278883e-01 -1.32397816e-01 -8.69148135e-01 1.96431503e-01
-8.92628953e-02 6.35695815e-01 1.04308176e+00 -9.45857227e-01
-5.79621375e-01 3.65607500e-01 -2.58544594e-01 3.44309211e-01
-4.58131671e-01 1.62346721e-01 -1.54773965e-01 -1.02359831e+00
-1.07673757e-01 -1.05978286e+00 4.60850149e-02 -3.96061122e-01
-4.43267703e-01 7.63637945e-02 6.85879767e-01 -9.34031665e-01
1.51632369e+00 -2.23720717e+00 4.88810576e-02 1.08593911e-01
-6.38623838e-04 2.49488279e-01 -2.57492542e-01 6.91434920e-01
2.09709689e-01 1.69908062e-01 -9.39253867e-02 -6.26171649e-01
4.71517518e-02 3.03215027e-01 -6.82993889e-01 8.75722021e-02
7.94685706e-02 6.72353208e-01 -7.44096339e-01 -4.08093840e-01
1.30430043e-01 2.67000467e-01 -7.36708283e-01 3.54090899e-01
-3.85213286e-01 5.70672691e-01 -1.58849940e-01 4.43182707e-01
-1.78682163e-01 2.13910453e-02 1.16229072e-01 2.07269147e-01
-2.54912466e-01 1.12066090e+00 -7.27002501e-01 1.67866802e+00
-7.60148406e-01 1.08792984e+00 1.04246892e-01 -3.75002891e-01
9.14199531e-01 6.43766999e-01 2.32572645e-01 -7.41275847e-01
-8.67385864e-02 5.77503979e-01 4.17902648e-01 -4.32581037e-01
7.12990165e-01 -1.19386725e-01 9.64486226e-02 3.86663049e-01
8.58757123e-02 -9.83122885e-01 1.53019816e-01 2.28721842e-01
1.25253296e+00 -2.47357681e-01 1.04265399e-01 -2.33733281e-01
6.53392822e-02 -7.38943592e-02 2.19711810e-01 8.40999663e-01
-1.60248689e-02 9.18025792e-01 1.63266093e-01 -2.88330056e-02
-1.34343708e+00 -1.10303640e+00 1.11811213e-01 1.15804386e+00
-4.07715440e-01 -7.69757330e-01 -1.02132726e+00 -1.86128184e-01
-4.49668616e-01 1.19566536e+00 -1.37628699e-02 7.79189914e-02
-5.92140198e-01 -1.72779858e-01 1.03487682e+00 2.29685664e-01
1.36037499e-01 -9.72325981e-01 -2.58458048e-01 5.32262146e-01
-4.22910362e-01 -1.31950748e+00 -9.04143572e-01 1.88686594e-01
-5.05272269e-01 -4.92487848e-01 -5.76274097e-01 -6.28493428e-01
8.23494643e-02 2.17496663e-01 1.15434623e+00 3.59940827e-02
1.65966004e-01 1.92167416e-01 -5.99032342e-01 -4.78082687e-01
-1.37220633e+00 3.69416535e-01 2.73964494e-01 -2.89871275e-01
1.98916788e-03 -7.73586988e-01 -2.09658578e-01 3.35256100e-01
-7.29743600e-01 3.56963933e-01 3.84656876e-01 8.52093458e-01
-3.80939916e-02 -2.77520984e-01 6.32281303e-01 -5.08220732e-01
1.01182079e+00 -2.77853072e-01 -3.75520885e-01 8.35332945e-02
-3.10146034e-01 1.72238484e-01 7.48463154e-01 -5.57866096e-01
-9.56930935e-01 -1.51356071e-01 -6.21178448e-01 -1.65280357e-01
3.06573068e-03 4.19106185e-01 6.86039254e-02 1.12560511e-01
9.29791152e-01 2.90897548e-01 -3.32085229e-02 -2.13163242e-01
3.16986889e-01 1.33674562e+00 6.98449373e-01 -6.50707722e-01
6.13629103e-01 -1.49564236e-01 -6.47569954e-01 -1.31021714e+00
-5.71266472e-01 -1.25390112e-01 -1.86874911e-01 -6.97689876e-02
5.88403106e-01 -1.05776501e+00 -4.25635874e-01 2.54910082e-01
-1.24913692e+00 -7.88006544e-01 -2.74787843e-01 6.29176915e-01
-8.43454599e-01 1.11381769e-01 -8.32398057e-01 -9.15924311e-01
-2.21598312e-01 -1.25427604e+00 1.20538795e+00 -1.06282629e-01
-8.47912550e-01 -7.71540821e-01 2.59512126e-01 4.83468980e-01
4.97276008e-01 -2.40650803e-01 6.64462924e-01 -8.23490024e-01
-3.15495431e-01 1.83248162e-01 4.23301488e-01 5.47077477e-01
4.49220240e-01 3.52968544e-01 -1.03968573e+00 -1.52692884e-01
3.58135402e-02 -4.38989848e-01 2.74602830e-01 2.41514564e-01
6.65322304e-01 -5.38745522e-01 2.53301263e-01 4.68256861e-01
5.58296740e-01 6.65643811e-01 5.42526782e-01 -7.69019872e-02
5.58072567e-01 4.30366367e-01 1.80354431e-01 4.40674126e-01
2.48717248e-01 8.37082446e-01 -1.11459270e-01 -2.26923004e-02
-3.85077089e-01 -3.97864252e-01 7.33302891e-01 1.92520511e+00
2.04677254e-01 -6.54952884e-01 -9.82301414e-01 5.51487803e-01
-1.49887264e+00 -1.07990015e+00 8.07090104e-02 2.02914166e+00
1.26565862e+00 4.74157095e-01 1.37298062e-01 7.93493241e-02
5.42914391e-01 2.58472383e-01 -3.22120011e-01 -8.32149327e-01
-1.46474615e-01 3.60670984e-01 3.83385152e-01 7.88471699e-01
-7.34918773e-01 1.10612655e+00 7.58712339e+00 1.06199694e+00
-1.11306715e+00 -6.68904185e-02 5.47941506e-01 -2.62888044e-01
-4.26736355e-01 -1.76552072e-01 -5.85562706e-01 5.95374465e-01
1.78883922e+00 -2.73387730e-01 9.55926538e-01 4.85090613e-01
5.90073168e-01 2.45618954e-01 -1.15098250e+00 8.74561787e-01
2.68952586e-02 -1.30629778e+00 -2.79902760e-02 -3.22114602e-02
9.00837183e-01 3.14975262e-01 2.35869661e-01 2.95101613e-01
7.98349500e-01 -1.19777882e+00 1.14559162e+00 2.81781293e-02
9.13463235e-01 -5.59612334e-01 2.16393154e-02 4.14089829e-01
-1.08627391e+00 1.19681701e-01 1.18533738e-01 -7.74982348e-02
2.83522338e-01 2.10971206e-01 -1.34794569e+00 2.25994736e-01
1.03679918e-01 3.91268015e-01 -2.50742346e-01 7.04528391e-01
-1.52351126e-01 1.16677320e+00 -5.17335773e-01 -1.66532457e-01
3.40663135e-01 1.84110433e-01 6.44100428e-01 1.30522633e+00
5.90963781e-01 1.20254867e-01 1.02920949e-01 4.73217279e-01
-1.45265654e-01 6.16596825e-02 -6.46524310e-01 -5.14219701e-01
8.40697229e-01 6.10038519e-01 -4.01499450e-01 -2.73713529e-01
-5.42467594e-01 9.38765287e-01 -3.85170840e-02 4.19791132e-01
-7.08247721e-01 -4.66465861e-01 7.95276463e-01 1.23652937e-02
4.19523045e-02 -8.11580300e-01 -2.79864252e-01 -1.25137746e+00
-1.51850700e-01 -1.67380202e+00 -2.32902050e-01 -8.72423172e-01
-9.18151975e-01 1.02771258e+00 -1.40244350e-01 -1.17761481e+00
-1.13784611e+00 -2.36064285e-01 -3.81997854e-01 8.27291191e-01
-8.65976572e-01 -8.23417604e-01 1.67977765e-01 3.72577965e-01
1.21061027e+00 -5.77634275e-01 9.60396945e-01 2.39557475e-01
-1.80534393e-01 7.66109407e-01 2.40461215e-01 1.03019282e-01
8.21170628e-01 -1.18878496e+00 1.23321259e+00 8.44531178e-01
6.73752069e-01 7.25784659e-01 1.23002243e+00 -5.41262925e-01
-1.14238763e+00 -7.15834379e-01 8.97387624e-01 -4.97105122e-01
8.23177576e-01 -5.84220052e-01 -6.86549962e-01 6.98167503e-01
6.04627132e-01 -4.17438835e-01 6.03748381e-01 -1.74692646e-03
-3.77653927e-01 4.70083021e-02 -6.61606133e-01 1.08797359e+00
9.39971268e-01 -9.46696222e-01 -6.31414175e-01 1.38286591e-01
1.22076416e+00 -6.24437273e-01 -5.83886981e-01 7.49387220e-02
6.17573261e-01 -9.30976450e-01 5.88637590e-01 -4.08034533e-01
4.45226878e-01 -9.85697880e-02 -4.24705505e-01 -1.75207686e+00
2.27082431e-01 -1.40607345e+00 2.42030825e-02 1.25097156e+00
7.09757507e-01 -3.27888489e-01 5.05079985e-01 6.87662423e-01
-5.97078860e-01 -2.99173236e-01 -9.19903755e-01 -9.99203444e-01
1.10584781e-01 -6.15529895e-01 5.45049071e-01 7.37726629e-01
1.85701042e-01 4.93948072e-01 -6.17078900e-01 -6.04249258e-03
1.90579310e-01 -3.03920597e-01 8.72755468e-01 -5.69164336e-01
-7.02389598e-01 -3.03720504e-01 -1.17487751e-01 -1.42655468e+00
1.68760896e-01 -4.69331533e-01 4.62498039e-01 -1.06374276e+00
-5.54483950e-01 -2.39966840e-01 3.49335223e-01 3.01623791e-01
1.75775051e-01 -6.31671958e-03 3.45741481e-01 1.25979513e-01
-2.63248265e-01 7.50507236e-01 1.06174636e+00 -1.68252736e-01
-2.87599534e-01 7.69809037e-02 -4.39228207e-01 6.03986919e-01
8.88620794e-01 -4.32587594e-01 -8.27156723e-01 -6.52905822e-01
-9.12359208e-02 3.34367663e-01 -1.30591631e-01 -1.25116813e+00
4.43978161e-01 -2.13463947e-01 -1.32969946e-01 -1.73689336e-01
6.38732791e-01 -4.70963120e-01 1.71119362e-01 1.47268400e-01
-7.72117078e-01 1.22001007e-01 2.47997999e-01 2.15552613e-01
-3.26801091e-01 -1.33050740e-01 6.79411352e-01 -1.73383225e-02
-3.06181073e-01 -1.90152451e-01 -7.94840157e-01 4.61999387e-01
4.11999106e-01 -1.02931924e-01 -2.36711487e-01 -1.01112771e+00
-4.48123962e-01 -8.28004479e-02 3.53916645e-01 5.69292367e-01
5.12756288e-01 -1.17526662e+00 -7.99321949e-01 3.49158347e-01
-3.47036533e-02 -4.31083828e-01 -1.69923946e-01 2.89579511e-01
-6.52085960e-01 4.07753736e-01 4.02598344e-02 -4.62751925e-01
-1.36332214e+00 -9.16126743e-02 2.43828654e-01 1.52812451e-01
-4.62117910e-01 7.61955202e-01 -3.75898816e-02 -3.33399385e-01
4.35724020e-01 -5.55197239e-01 3.50766599e-01 -3.74576449e-01
5.22124529e-01 1.06024697e-01 1.67263389e-01 -5.94008863e-01
-9.71711352e-02 -1.48700178e-02 4.19984432e-03 -1.02625477e+00
1.07637954e+00 -4.40773964e-01 4.27071571e-01 8.05413127e-01
1.12532389e+00 4.70200628e-01 -1.34851670e+00 5.55899441e-02
5.32936640e-02 -3.86330217e-01 -2.12305978e-01 -6.71777308e-01
-5.68789124e-01 8.11900616e-01 2.37920973e-02 5.60507238e-01
8.16448569e-01 -3.53231996e-01 1.22063744e+00 5.63173175e-01
3.70038241e-01 -1.17402601e+00 1.60017788e-01 7.69901335e-01
8.68475556e-01 -1.03780639e+00 -4.18138087e-01 -1.77505881e-01
-9.29722548e-01 1.09798563e+00 3.07495862e-01 2.57564187e-01
3.16673309e-01 8.35535288e-01 4.51986283e-01 3.51817429e-01
-1.28151870e+00 5.40968366e-02 1.81388706e-01 5.77895105e-01
7.11071432e-01 4.28174026e-02 1.49242714e-01 1.76494420e-01
-1.01119387e+00 -3.99098516e-01 6.27251089e-01 7.24797964e-01
-6.07736230e-01 -1.18778729e+00 -3.34535658e-01 2.29689419e-01
-5.95982134e-01 -6.14511311e-01 -5.28021514e-01 3.76396835e-01
-4.49577928e-01 1.56389916e+00 -1.27722416e-02 -6.36064529e-01
2.11407378e-01 1.68102801e-01 2.58815706e-01 -7.07368076e-01
-6.55922234e-01 3.94555181e-01 5.95410466e-01 -3.85245919e-01
-2.54896820e-01 -5.87221444e-01 -1.17317450e+00 -5.69477797e-01
-2.08686858e-01 3.30810905e-01 7.73658395e-01 9.78393793e-01
1.94373861e-01 6.43912017e-01 7.54826903e-01 -7.08620489e-01
-6.87391400e-01 -1.01738715e+00 -3.41966569e-01 8.06964934e-02
7.77837217e-01 -3.55984345e-02 -4.22074795e-01 5.30249417e-01] | [14.764078140258789, 6.759161949157715] |
05af2f13-f22d-438c-9924-522ac42603fd | cross-lingual-question-answering-using-common | null | null | https://aclanthology.org/W16-1403 | https://aclanthology.org/W16-1403.pdf | Cross-Lingual Question Answering Using Common Semantic Space | null | ['Amir Pouran Ben Veyseh'] | 2016-06-01 | null | null | null | ws-2016-6 | ['cross-lingual-question-answering'] | ['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] | [-7.496878147125244, 3.5826504230499268] |
9bd7ef10-e2a6-4fbb-a671-98268b1e70c4 | learning-stylometric-representations-for | 1606.01219 | null | http://arxiv.org/abs/1606.01219v1 | http://arxiv.org/pdf/1606.01219v1.pdf | Learning Stylometric Representations for Authorship Analysis | Authorship analysis (AA) is the study of unveiling the hidden properties of
authors from a body of exponentially exploding textual data. It extracts an
author's identity and sociolinguistic characteristics based on the reflected
writing styles in the text. It is an essential process for various areas, such
as cybercrime investigation, psycholinguistics, political socialization, etc.
However, most of the previous techniques critically depend on the manual
feature engineering process. Consequently, the choice of feature set has been
shown to be scenario- or dataset-dependent. In this paper, to mimic the human
sentence composition process using a neural network approach, we propose to
incorporate different categories of linguistic features into distributed
representation of words in order to learn simultaneously the writing style
representations based on unlabeled texts for authorship analysis. In
particular, the proposed models allow topical, lexical, syntactical, and
character-level feature vectors of each document to be extracted as
stylometrics. We evaluate the performance of our approach on the problems of
authorship characterization and authorship verification with the Twitter,
novel, and essay datasets. The experiments suggest that our proposed text
representation outperforms the bag-of-lexical-n-grams, Latent Dirichlet
Allocation, Latent Semantic Analysis, PVDM, PVDBOW, and word2vec
representations. | ['Steven H. H. Ding', 'William K. Cheung', 'Benjamin C. M. Fung', 'Farkhund Iqbal'] | 2016-06-03 | null | null | null | null | ['authorship-verification'] | ['natural-language-processing'] | [-6.40272424e-02 -2.71399617e-01 -1.50531560e-01 -2.74626911e-01
-2.39721742e-02 -7.21946359e-01 9.87969637e-01 4.61552531e-01
-4.43421245e-01 3.39622557e-01 6.27271891e-01 -2.31113821e-01
-9.97601748e-02 -4.94315743e-01 1.01562604e-01 -5.18701732e-01
4.51136708e-01 4.14576799e-01 -3.44166547e-01 -1.17408335e-01
1.22042358e+00 6.20453715e-01 -1.61522949e+00 -6.27225116e-02
9.96689439e-01 7.66185105e-01 -2.84692615e-01 6.77076161e-01
-6.60535097e-01 6.95243180e-01 -7.07529843e-01 -4.13818240e-01
-6.92637488e-02 -4.80890065e-01 -5.30592203e-01 3.12007636e-01
2.80927271e-01 -5.60825430e-02 -1.90671042e-01 1.11901414e+00
3.58844310e-01 -3.22192647e-02 1.35474515e+00 -1.03854012e+00
-1.06385100e+00 6.01882815e-01 -6.66144907e-01 3.02818239e-01
3.39651734e-01 -1.68850213e-01 9.73907828e-01 -9.29263711e-01
6.12214267e-01 1.32924867e+00 4.25852537e-01 1.66153356e-01
-8.11508894e-01 -8.35366607e-01 -1.98774651e-01 2.26147860e-01
-1.14139080e+00 -3.80178839e-01 1.18158388e+00 -9.72660661e-01
3.30973744e-01 4.69694519e-03 4.58608210e-01 1.46415877e+00
2.19135776e-01 5.94980597e-01 1.35856926e+00 -9.15894628e-01
2.85447184e-02 5.42948902e-01 7.75282383e-01 8.02195251e-01
4.03831720e-01 -3.46419811e-01 -6.51843011e-01 -5.92199504e-01
3.81177157e-01 -1.00303173e-01 2.31689382e-02 -4.72156852e-02
-1.05122519e+00 1.12316167e+00 -4.86568570e-01 6.32267416e-01
-1.81710094e-01 -2.97055274e-01 7.75117159e-01 2.92690068e-01
6.19883120e-01 7.36256421e-01 -5.26946224e-02 -2.17402622e-01
-1.12529278e+00 8.19614902e-03 9.53916669e-01 6.29468858e-01
4.63949978e-01 1.18898757e-01 -2.47110918e-01 7.73934782e-01
3.83124143e-01 4.55747753e-01 1.16289949e+00 -5.85994959e-01
2.58422941e-01 8.49390447e-01 1.10875145e-01 -1.84854090e+00
-2.47586995e-01 -1.74158737e-01 -8.37552369e-01 -2.87726402e-01
3.33619684e-01 -1.12215951e-01 -4.44264054e-01 1.22761500e+00
-7.83060268e-02 -3.12411040e-01 -6.48727939e-02 5.26707292e-01
8.66292357e-01 6.51123345e-01 1.00293718e-01 -3.88318807e-01
1.35198247e+00 -5.51394820e-01 -1.09170318e+00 2.49584228e-01
4.30093765e-01 -1.07533610e+00 9.48031127e-01 5.13685703e-01
-5.49671054e-01 -6.18329108e-01 -1.00999808e+00 2.21895762e-02
-6.28201425e-01 5.60049891e-01 4.50723261e-01 9.69702363e-01
-3.97528887e-01 7.17237294e-01 -2.15068251e-01 -5.29783368e-01
2.27523208e-01 -1.64649487e-02 -3.61911774e-01 5.49091399e-01
-1.02602029e+00 6.35780215e-01 3.06359261e-01 4.94350605e-02
-1.85694218e-01 -1.44179299e-01 -6.49418533e-01 1.14950366e-01
-1.05082862e-01 -1.73901767e-01 6.00903809e-01 -1.11393106e+00
-1.87986803e+00 9.18463111e-01 -1.79532796e-01 5.68157062e-02
4.30366397e-01 6.84423521e-02 -5.27446806e-01 -3.51804942e-02
7.83144161e-02 -8.54402259e-02 1.08848310e+00 -8.45244288e-01
-2.12705135e-01 -6.09380424e-01 -6.18717253e-01 -3.66918072e-02
-1.11609399e+00 4.23625827e-01 7.21560493e-02 -7.97158241e-01
1.63327679e-01 -6.41312599e-01 2.36942559e-01 -4.05526668e-01
-5.67602098e-01 -6.38629556e-01 6.79757297e-01 -1.09226072e+00
1.35902596e+00 -2.19783425e+00 3.10702711e-01 4.07249838e-01
4.52645928e-01 2.67920401e-02 2.07539693e-01 4.90535855e-01
3.45607370e-01 2.66566038e-01 8.71718898e-02 -5.07598341e-01
1.23414584e-01 -2.38777936e-01 -3.58289510e-01 7.58384645e-01
-3.45544577e-01 5.18770039e-01 -6.89248443e-01 -8.84846747e-01
-1.77273974e-01 4.42356169e-02 -2.19181016e-01 3.15009028e-01
1.44776613e-01 1.97486237e-01 -4.48685974e-01 5.99114656e-01
3.42512697e-01 -3.30650769e-02 4.65196401e-01 4.02628966e-02
-3.78141969e-01 2.35349908e-02 -9.91106093e-01 1.23060703e+00
-2.16100603e-01 9.64046240e-01 -4.98077989e-01 -7.61731684e-01
1.39926684e+00 2.17174530e-01 1.15737751e-01 -2.96114236e-01
7.39786744e-01 2.98858404e-01 -4.17575501e-02 -6.53739572e-01
7.23569572e-01 1.92197450e-02 -4.79051381e-01 8.33932638e-01
1.08333573e-01 5.49501240e-01 8.86655822e-02 2.17238009e-01
5.50457060e-01 -2.14902803e-01 4.79675591e-01 -3.88178170e-01
7.23236203e-01 -2.88630158e-01 4.33986783e-01 5.79916000e-01
-1.21849343e-01 3.51384729e-01 9.24322724e-01 -3.67634416e-01
-1.19659626e+00 -6.22874916e-01 -2.55063862e-01 9.85662937e-01
-1.13127649e-01 -1.81300774e-01 -7.41724670e-01 -6.94814384e-01
2.02380791e-01 7.37554491e-01 -7.17713177e-01 -5.90726361e-02
-4.01781291e-01 -7.03687906e-01 9.48987067e-01 1.88661546e-01
2.68268645e-01 -1.14572680e+00 -4.28796649e-01 1.89896654e-02
1.32490680e-01 -8.09441805e-01 -5.92425883e-01 -1.36254147e-01
-6.52458668e-01 -1.07971895e+00 -5.95410526e-01 -6.25742733e-01
6.10637128e-01 -1.34584323e-01 5.69822013e-01 -1.65249720e-01
-2.26104289e-01 2.30616525e-01 -6.11339152e-01 -3.39127570e-01
-5.72844505e-01 3.79939586e-01 3.99358362e-01 5.63929796e-01
7.69448757e-01 -4.77401614e-01 -7.28631169e-02 -6.05461486e-02
-5.80884755e-01 -2.66604483e-01 3.87498617e-01 9.49444830e-01
-4.15824577e-02 -1.67970523e-01 5.23951173e-01 -1.09752178e+00
1.39208317e+00 -6.44556344e-01 -2.05739364e-01 3.79410088e-01
-9.86722946e-01 2.10595176e-01 6.98949993e-01 -7.69901156e-01
-8.83277297e-01 -4.27249551e-01 4.67662066e-01 -4.11780804e-01
-2.79976666e-01 5.29883265e-01 -2.39264190e-01 2.11255714e-01
5.17901540e-01 6.09987676e-01 2.45393276e-01 -7.70736456e-01
1.88471347e-01 1.38648868e+00 2.39321738e-01 -4.56238389e-01
6.87033772e-01 1.34030715e-01 -2.55655289e-01 -1.02494907e+00
-6.55090451e-01 -7.26049483e-01 -1.09297931e+00 -1.50896654e-01
7.74694443e-01 -3.39567930e-01 -8.39477420e-01 5.21330357e-01
-1.40738618e+00 6.29449308e-01 -8.94054770e-03 3.73858750e-01
-1.59858063e-01 9.86688495e-01 -4.70532477e-01 -9.14658248e-01
-4.95989293e-01 -8.51747692e-01 6.94834352e-01 1.68746009e-01
-7.51485288e-01 -1.11386228e+00 2.96015531e-01 3.22092474e-01
1.01977438e-01 2.16579199e-01 1.24214649e+00 -1.18158829e+00
1.90234587e-01 -4.53239739e-01 -2.27213055e-01 4.25776780e-01
2.76432008e-01 2.87494391e-01 -9.05770838e-01 1.34339472e-02
-5.01136743e-02 -1.94817230e-01 6.88258350e-01 -1.60126850e-01
1.23402452e+00 -6.15427911e-01 5.07227369e-02 4.97312039e-01
1.05765367e+00 -1.29053826e-02 2.58038372e-01 2.87797481e-01
8.65444243e-01 7.61585891e-01 5.75343370e-02 8.89112473e-01
1.04899935e-01 2.30806097e-01 -9.96640846e-02 4.95114028e-01
3.21011901e-01 -2.22754538e-01 3.46895307e-01 1.35943222e+00
-1.34117946e-01 -3.31707537e-01 -1.03615224e+00 3.18768203e-01
-1.52309215e+00 -1.04925776e+00 -3.18757266e-01 1.83822858e+00
7.22221255e-01 1.74050555e-02 1.44432023e-01 4.90619302e-01
8.29984009e-01 3.05379540e-01 -2.51470417e-01 -1.03222311e+00
-2.39328638e-01 -1.36150092e-01 4.02444720e-01 2.82407969e-01
-9.23318028e-01 8.23549747e-01 5.53351021e+00 7.73864269e-01
-1.01038551e+00 1.28476217e-01 3.36625099e-01 4.03853267e-01
-4.13079023e-01 -1.97101533e-01 -1.03990996e+00 8.39044631e-01
8.64115715e-01 -3.05663466e-01 2.13025913e-01 6.86122477e-01
1.83267057e-01 -1.06354989e-02 -8.84498239e-01 1.00011146e+00
8.33236992e-01 -1.10504270e+00 3.65264267e-01 2.78699845e-01
4.57580954e-01 -6.88490510e-01 2.07265943e-01 1.62039742e-01
-1.24796033e-01 -1.09938097e+00 6.36396766e-01 7.53753662e-01
6.85869992e-01 -6.56678677e-01 8.96742702e-01 5.62320173e-01
-3.67758006e-01 -1.61381558e-01 -3.65012586e-01 -2.07417667e-01
-3.24925423e-01 4.32148278e-01 -8.13838840e-01 2.62676924e-01
1.43449455e-01 8.05355251e-01 -8.34348083e-01 6.03333294e-01
3.98680614e-03 7.89924800e-01 1.21586345e-01 -6.82891846e-01
7.09694028e-02 -3.04178149e-01 4.76493776e-01 1.29382765e+00
2.85588741e-01 -1.36615306e-01 -8.56652334e-02 8.91306579e-01
-2.76802063e-01 8.01449955e-01 -6.00902021e-01 -6.15383446e-01
5.12524188e-01 1.24288821e+00 -8.63993466e-01 -2.10159749e-01
-1.25603586e-01 1.06756318e+00 3.54197443e-01 9.01690274e-02
-4.17571664e-01 -5.20115554e-01 2.26098925e-01 2.51144141e-01
-7.85952061e-02 -4.62123394e-01 -7.99346209e-01 -1.31372094e+00
-3.18688780e-01 -8.50343049e-01 2.34739512e-01 -3.18900406e-01
-1.78195739e+00 4.85233128e-01 -2.41811827e-01 -1.12206137e+00
-1.54160773e-02 -7.08866656e-01 -7.44476080e-01 1.07605600e+00
-1.09303379e+00 -9.98122215e-01 -3.35027844e-01 2.12529868e-01
6.57947958e-01 -9.78842616e-01 8.35489333e-01 6.32665977e-02
-7.68845737e-01 7.66468942e-01 5.63113391e-01 4.85609651e-01
9.18118298e-01 -1.19269681e+00 -1.44272177e-02 3.85099828e-01
7.96134099e-02 7.54968584e-01 6.97750688e-01 -9.08211231e-01
-1.17387664e+00 -5.20944595e-01 1.42468774e+00 -5.67202628e-01
9.10272896e-01 -3.37387800e-01 -7.22066045e-01 4.13842946e-01
3.20097506e-01 -4.69190598e-01 1.24026000e+00 4.56065685e-01
-5.20507693e-01 2.88892001e-01 -8.23815227e-01 4.96117234e-01
5.10467410e-01 -6.08288169e-01 -9.70068932e-01 2.29778394e-01
4.50022593e-02 3.14304709e-01 -7.70726860e-01 -2.83713698e-01
8.58174741e-01 -5.83213210e-01 4.42030787e-01 -8.76167655e-01
7.05201268e-01 1.78128436e-01 -1.92232132e-02 -1.05516779e+00
-6.55379832e-01 -3.38302016e-01 -7.11511597e-02 1.62128854e+00
3.88987958e-02 -5.56226730e-01 5.81122994e-01 3.02279472e-01
3.10027063e-01 -4.77340758e-01 -6.51970744e-01 -4.95469958e-01
2.78904855e-01 9.74049941e-02 4.34382498e-01 1.39318919e+00
2.06160590e-01 5.50920546e-01 -5.54123819e-01 -1.76024541e-01
7.05795169e-01 1.26263335e-01 7.58879244e-01 -1.88452530e+00
-9.80136264e-03 -7.46154547e-01 -5.17051280e-01 -5.54282427e-01
6.60339594e-01 -8.89734328e-01 -4.01088119e-01 -1.06478226e+00
3.53961468e-01 -2.67095029e-01 -1.25506759e-01 -1.21639639e-01
-8.89817476e-02 -1.29602969e-01 7.90323541e-02 6.68536484e-01
-3.11410576e-01 6.53647423e-01 1.09708107e+00 -1.80790469e-01
-1.93380386e-01 -1.71177641e-01 -6.15612566e-01 8.50707889e-01
9.25165236e-01 -3.99316937e-01 -6.43416047e-02 7.54640251e-02
2.99872130e-01 -1.75819293e-01 9.05942693e-02 -5.23625612e-01
1.90424830e-01 -2.42368311e-01 5.67210376e-01 -4.29776877e-01
3.15670930e-02 -4.84469593e-01 -4.64960843e-01 1.83479115e-01
-5.47158718e-01 1.91291019e-01 -3.45847607e-01 6.27655864e-01
-1.96590215e-01 -6.08615220e-01 4.48572606e-01 -2.12044612e-01
-3.12031746e-01 -2.24190447e-02 -8.02845001e-01 8.26267805e-03
7.60894418e-01 -5.19442916e-01 -2.63493001e-01 -9.45314243e-02
-2.57384241e-01 -1.80585310e-01 3.43752533e-01 4.58922058e-01
5.16820908e-01 -1.10239732e+00 -7.71921337e-01 2.75067151e-01
1.06878050e-01 -7.88307905e-01 7.13198110e-02 7.02187657e-01
-5.55739403e-01 4.06329751e-01 -4.91067797e-01 -7.74160326e-02
-1.30483115e+00 5.64915001e-01 -7.86967948e-02 -1.49845645e-01
-3.82980287e-01 4.45818454e-01 -1.82674527e-01 -3.02426994e-01
2.59102821e-01 2.27652624e-01 -1.00360823e+00 8.93471718e-01
1.53234199e-01 6.78220332e-01 -3.81540984e-01 -9.22409475e-01
-4.25925776e-02 5.38673162e-01 -1.41794592e-01 -1.10955462e-01
1.09674799e+00 -9.31308120e-02 -3.72824222e-01 1.00818980e+00
1.18834960e+00 4.88912284e-01 -3.33666801e-01 -2.74618268e-01
4.55611080e-01 -4.98774141e-01 6.18939176e-02 -3.75299692e-01
-4.20102239e-01 1.13387358e+00 3.56709450e-01 3.16884607e-01
4.30232286e-01 -2.93699116e-01 6.01102829e-01 2.70858467e-01
1.00980103e-01 -1.54524016e+00 2.08674133e-01 5.10537863e-01
8.78044307e-01 -1.15183294e+00 2.10775018e-01 -6.54406473e-02
-6.89996123e-01 1.61086452e+00 4.63819623e-01 -3.19235697e-02
9.87601161e-01 -5.11823557e-02 -2.71022804e-02 -1.90646455e-01
-2.59225875e-01 3.21974397e-01 5.63277841e-01 5.95655106e-02
5.43468595e-01 1.72540575e-01 -7.32770920e-01 1.05799699e+00
-3.79306197e-01 -3.16323042e-01 8.28778565e-01 6.12598777e-01
-4.84595537e-01 -1.06916428e+00 -4.63845193e-01 5.50116420e-01
-4.48183209e-01 2.26683393e-02 -8.46025527e-01 4.10173863e-01
2.09193915e-01 6.30324483e-01 1.37830243e-01 -4.07611489e-01
-3.12806927e-02 5.02157867e-01 2.68895358e-01 -5.73127389e-01
-6.91402912e-01 -2.47838914e-01 1.38352826e-01 3.18803847e-01
-2.05965996e-01 -1.06847036e+00 -7.22434402e-01 -5.95867336e-01
-2.87257910e-01 3.73036981e-01 8.18306684e-01 1.00937831e+00
1.56298041e-01 2.18316093e-01 8.66749763e-01 -5.90110660e-01
-8.40616643e-01 -1.42373681e+00 -8.42319608e-01 4.72112924e-01
-1.32205442e-01 -7.29882598e-01 -7.24763155e-01 2.94748582e-02] | [9.598326683044434, 10.553476333618164] |
68797e1b-4d9d-4d78-b639-9118de5c5490 | the-value-improvement-path-towards-better | 2006.02243 | null | https://arxiv.org/abs/2006.02243v2 | https://arxiv.org/pdf/2006.02243v2.pdf | The Value-Improvement Path: Towards Better Representations for Reinforcement Learning | In value-based reinforcement learning (RL), unlike in supervised learning, the agent faces not a single, stationary, approximation problem, but a sequence of value prediction problems. Each time the policy improves, the nature of the problem changes, shifting both the distribution of states and their values. In this paper we take a novel perspective, arguing that the value prediction problems faced by an RL agent should not be addressed in isolation, but rather as a single, holistic, prediction problem. An RL algorithm generates a sequence of policies that, at least approximately, improve towards the optimal policy. We explicitly characterize the associated sequence of value functions and call it the value-improvement path. Our main idea is to approximate the value-improvement path holistically, rather than to solely track the value function of the current policy. Specifically, we discuss the impact that this holistic view of RL has on representation learning. We demonstrate that a representation that spans the past value-improvement path will also provide an accurate value approximation for future policy improvements. We use this insight to better understand existing approaches to auxiliary tasks and to propose new ones. To test our hypothesis empirically, we augmented a standard deep RL agent with an auxiliary task of learning the value-improvement path. In a study of Atari 2600 games, the augmented agent achieved approximately double the mean and median performance of the baseline agent. | ['Robert Dadashi', 'André Barreto', 'Mark Rowland', 'David Silver', 'Will Dabney', 'Marc G. Bellemare', 'John Quan'] | 2020-06-03 | null | null | null | null | ['value-prediction'] | ['computer-code'] | [ 1.77259538e-02 3.67586792e-01 -7.11417854e-01 -7.76197091e-02
-8.90429378e-01 -8.16945314e-01 7.01270938e-01 2.17370927e-01
-7.32510090e-01 1.33419454e+00 6.01076186e-01 -4.28524941e-01
-3.24697524e-01 -6.07261598e-01 -8.09074759e-01 -8.54298532e-01
-1.87408194e-01 7.00248182e-01 -1.21960267e-01 -5.08621573e-01
2.85733521e-01 4.32243973e-01 -1.40635872e+00 1.11630380e-01
6.08426273e-01 9.18014228e-01 7.63930455e-02 7.68622994e-01
1.26741990e-01 1.32381165e+00 -8.60800862e-01 -1.41868249e-01
4.09931540e-01 -5.76081216e-01 -1.06258631e+00 -6.20795898e-02
-5.87367378e-02 -7.48273849e-01 -3.93459022e-01 1.05484629e+00
2.59728372e-01 7.48937964e-01 5.81950605e-01 -1.42162418e+00
-4.20882940e-01 9.68784213e-01 -4.45682019e-01 2.19603479e-01
1.73838630e-01 5.07070303e-01 1.09377038e+00 6.81932569e-02
7.14766681e-01 1.21107769e+00 4.54161137e-01 7.66590178e-01
-1.46263647e+00 -2.34869018e-01 6.71905577e-01 1.61366433e-01
-6.00685358e-01 -2.79810369e-01 6.42617106e-01 -2.58928627e-01
1.09704530e+00 -2.36533508e-01 1.05257809e+00 1.05067205e+00
2.85571605e-01 9.56650138e-01 1.18952394e+00 -1.69458464e-01
6.31825209e-01 -5.06416410e-02 -2.51827031e-01 5.18957138e-01
1.06590882e-01 6.62315726e-01 -1.65964261e-01 -1.62382141e-01
6.79971516e-01 -9.94573832e-02 -1.75159782e-01 -8.07595730e-01
-1.01194596e+00 8.86848271e-01 3.34714025e-01 2.38750741e-01
-7.80648589e-01 7.86328733e-01 4.54698205e-01 7.20630348e-01
1.42973334e-01 8.63070428e-01 -7.47795463e-01 -6.48379385e-01
-6.28669798e-01 7.21859872e-01 7.54556715e-01 4.04371440e-01
7.15827048e-01 4.31228638e-01 -3.83674175e-01 2.26765618e-01
2.43216697e-02 2.87525892e-01 6.21234179e-01 -1.73467731e+00
2.76523471e-01 4.57502790e-02 7.56555498e-01 -4.65045124e-01
-4.25185591e-01 -6.80736840e-01 -2.34356314e-01 5.77225268e-01
7.37689316e-01 -6.57082975e-01 -6.70685589e-01 2.28865933e+00
8.06270093e-02 3.29779312e-02 3.30859244e-01 5.22662103e-01
-3.29792649e-01 7.65908241e-01 6.98531941e-02 -5.90371549e-01
6.46506846e-01 -8.43837440e-01 -6.30366385e-01 -2.03483045e-01
8.22687745e-01 -1.71511486e-01 1.04656219e+00 4.86506671e-01
-1.36571217e+00 -3.40962499e-01 -8.82191539e-01 3.62286478e-01
4.63496223e-02 -4.49158698e-01 5.60125887e-01 8.57913569e-02
-1.32223415e+00 1.16176856e+00 -7.95032740e-01 -9.13814008e-02
3.76207769e-01 3.03829670e-01 1.01255804e-01 3.02155256e-01
-1.15361810e+00 1.32005966e+00 5.87800920e-01 -4.78870600e-01
-1.36721945e+00 -6.60076082e-01 -6.26417398e-01 2.74398804e-01
7.91569591e-01 -5.20125806e-01 2.09037781e+00 -1.26743650e+00
-1.68798983e+00 2.10488111e-01 -2.63568684e-02 -9.46250975e-01
5.16747475e-01 -2.22840399e-01 5.60219288e-02 -1.22228861e-01
-1.50691986e-01 5.18896103e-01 8.48602414e-01 -1.47650456e+00
-8.95412445e-01 -1.54899016e-01 5.49888313e-01 5.95982790e-01
6.15702644e-02 -5.52080035e-01 2.65909851e-01 -5.54093242e-01
-5.52197993e-01 -9.28346872e-01 -6.34673893e-01 -3.68531764e-01
2.57278621e-01 -3.19969743e-01 1.97000682e-01 -3.87406200e-01
1.27175283e+00 -1.79485130e+00 3.68293315e-01 1.55020550e-01
1.34377986e-01 1.81715444e-01 -3.60272944e-01 5.26964724e-01
-1.50849432e-01 -5.16087376e-03 -1.67023003e-01 -1.47797659e-01
1.22622758e-01 6.04307592e-01 -6.52223885e-01 4.84371662e-01
-1.01717368e-01 1.09635842e+00 -1.30880034e+00 3.35887470e-03
1.49385154e-01 1.46023966e-02 -8.21971178e-01 1.64684191e-01
-8.64216149e-01 5.28768301e-01 -5.17129183e-01 5.76749630e-02
1.23334967e-01 4.32745889e-02 4.66476262e-01 2.66381770e-01
-1.68175697e-01 3.23470443e-01 -9.91362512e-01 1.57225847e+00
-4.35941637e-01 3.50333750e-01 -1.70688599e-01 -1.39299810e+00
8.04625034e-01 1.50477812e-01 9.39970613e-01 -9.17003572e-01
1.80773303e-01 -1.34579055e-02 2.40434408e-01 -1.55536458e-01
6.02930307e-01 -3.87991488e-01 -1.26087368e-01 9.11878943e-01
-1.29174784e-01 -3.03110071e-02 2.83801585e-01 9.08436552e-02
1.01237226e+00 5.94864666e-01 5.81877887e-01 -1.92496553e-01
2.13438928e-01 4.08884108e-01 5.65790772e-01 1.11648333e+00
-5.23672640e-01 -1.41134337e-01 1.09555662e+00 -5.19215047e-01
-1.17839992e+00 -1.10081291e+00 3.76800328e-01 1.34597218e+00
-3.65638025e-02 -2.10548908e-01 -4.89649087e-01 -8.43657136e-01
2.89606750e-01 1.08824146e+00 -7.16485322e-01 -4.92953122e-01
-7.96507537e-01 -4.67610449e-01 2.66859055e-01 5.75038373e-01
3.07952315e-01 -1.35462999e+00 -1.04588223e+00 5.00898421e-01
-2.24293582e-02 -5.98712087e-01 -4.24634337e-01 3.62849981e-01
-9.73574877e-01 -9.46083128e-01 -4.86017942e-01 -3.86275411e-01
2.39861369e-01 -1.52687356e-01 1.31877816e+00 -7.99119845e-02
4.68336105e-01 8.72198939e-01 -2.64150470e-01 -3.58553648e-01
-6.33684397e-01 -3.03311758e-02 1.15680762e-01 -3.38750720e-01
2.33564898e-02 -5.27394891e-01 -4.94907916e-01 -3.03730786e-01
-5.43694913e-01 -2.32795134e-01 6.03439569e-01 9.97739673e-01
6.15709960e-01 1.39857382e-01 7.95851409e-01 -6.59590423e-01
1.13057888e+00 -6.04623258e-01 -7.49791503e-01 3.34527969e-01
-9.48505282e-01 7.15790570e-01 8.90193939e-01 -5.34858763e-01
-1.24004328e+00 5.23407869e-02 -1.85207017e-02 -2.85144836e-01
-3.20846401e-02 4.88793969e-01 2.31542930e-01 3.98215622e-01
6.31088495e-01 5.01670420e-01 4.63395327e-01 -3.34125996e-01
7.30812669e-01 1.01165414e-01 4.64732438e-01 -1.02423871e+00
5.82518339e-01 1.16009168e-01 1.04420267e-01 -3.60218585e-01
-8.21338773e-01 -7.69812614e-03 -2.69882530e-01 -3.36177975e-01
4.33129191e-01 -6.55970335e-01 -1.07442629e+00 1.49285078e-01
-8.92199278e-01 -1.06978965e+00 -1.14235520e+00 3.12065452e-01
-1.27070868e+00 3.93949784e-02 -3.51680964e-01 -9.18892741e-01
3.54115367e-02 -1.10022366e+00 4.37985718e-01 2.93778241e-01
-6.96394816e-02 -1.14237142e+00 5.36630273e-01 -1.29210338e-01
4.82569218e-01 2.59133786e-01 1.07746470e+00 -5.89097738e-01
-2.33631447e-01 3.58360022e-01 2.45876178e-01 3.17010403e-01
-5.40177152e-03 -4.59439278e-01 -5.00439703e-01 -6.61929667e-01
1.20032899e-01 -5.90014994e-01 8.15199256e-01 6.28656149e-01
1.17038536e+00 -7.78754830e-01 8.12645480e-02 4.67445642e-01
1.43586004e+00 6.83715403e-01 4.87248778e-01 6.48643076e-01
1.74320385e-01 4.15738732e-01 8.81203949e-01 8.83926272e-01
3.39888990e-01 5.35286248e-01 7.25495517e-01 4.38576460e-01
1.51805818e-01 -5.64992726e-01 6.45930946e-01 1.21055776e-02
-2.34879285e-01 -2.92286351e-02 -6.72158778e-01 3.79694790e-01
-2.13649416e+00 -1.43839562e+00 6.85877085e-01 2.32106996e+00
1.06124783e+00 2.14565963e-01 6.84433877e-01 -1.54115155e-01
3.58539104e-01 3.22570980e-01 -1.29545808e+00 -8.43187034e-01
4.38035846e-01 2.80523449e-01 5.74078619e-01 6.93576872e-01
-7.03138292e-01 1.06884015e+00 6.94573641e+00 5.73629200e-01
-1.03007925e+00 -6.60559088e-02 7.14719951e-01 -5.01561403e-01
-3.98285359e-01 -1.05207816e-01 -4.77598697e-01 2.56455302e-01
1.06918049e+00 -6.75993562e-01 1.17489314e+00 1.14108706e+00
2.57860363e-01 -1.30675256e-01 -1.11204779e+00 6.54073477e-01
-4.19603944e-01 -1.37901914e+00 -3.84030910e-03 3.53287190e-01
8.90870333e-01 8.40692744e-02 2.24269643e-01 8.84654999e-01
1.12028301e+00 -1.13333380e+00 9.51029778e-01 5.54631352e-01
5.19195676e-01 -1.15311360e+00 6.01387203e-01 5.49213648e-01
-1.01234221e+00 -6.34424388e-01 -2.85019577e-01 -5.30492961e-01
-7.64108151e-02 -2.83289373e-01 -7.89423943e-01 1.39337212e-01
2.54818499e-01 7.33855188e-01 -2.16835558e-01 6.65949762e-01
-2.60773122e-01 3.67292583e-01 7.88389295e-02 -1.95161000e-01
6.95550084e-01 -1.73066631e-01 3.69888812e-01 6.04918540e-01
1.68637633e-01 2.51256935e-02 3.71450186e-01 6.70560598e-01
-1.21543584e-02 -2.20795363e-01 -7.66279936e-01 -3.16448003e-01
5.38224578e-01 7.32155263e-01 -4.04820919e-01 -2.70011693e-01
-1.93768233e-01 4.47259545e-01 7.74319828e-01 5.71196616e-01
-8.65015864e-01 3.54155712e-02 1.18754518e+00 -1.46987692e-01
4.18048710e-01 -2.91570127e-01 -2.13278569e-02 -1.01075816e+00
-3.88448119e-01 -1.29160595e+00 3.53527755e-01 -3.58683556e-01
-8.49513650e-01 2.28618160e-01 1.67556271e-01 -1.07672334e+00
-8.96954417e-01 -3.78948003e-01 -4.36383605e-01 4.95229393e-01
-1.56744099e+00 -4.45990473e-01 3.24256390e-01 6.52204692e-01
5.95404327e-01 -2.61636376e-01 6.81175053e-01 -4.49244618e-01
-4.06116515e-01 3.90615135e-01 6.13060832e-01 -1.57960325e-01
2.72500426e-01 -1.44542766e+00 1.65684566e-01 6.80972517e-01
-4.30545323e-02 4.22262311e-01 9.45958257e-01 -4.39828783e-01
-1.44672334e+00 -8.90221894e-01 1.59401283e-01 -3.52726012e-01
9.23032045e-01 3.57540488e-01 -7.13805795e-01 9.15023565e-01
1.97192118e-01 -2.79174328e-01 1.76867992e-01 8.22787210e-02
4.50337725e-03 -1.83289990e-01 -1.13667834e+00 7.19035268e-01
9.08339560e-01 -3.18079919e-01 -8.70604038e-01 -8.91095474e-02
7.61913240e-01 -3.56369644e-01 -6.79636359e-01 7.97266439e-02
5.84985256e-01 -1.03139591e+00 1.00876725e+00 -1.27915263e+00
4.23806131e-01 5.10023497e-02 -1.94984108e-01 -1.94720078e+00
-4.32852626e-01 -7.80754864e-01 -7.46789753e-01 7.18655944e-01
7.32203498e-02 -7.20418096e-01 9.54827309e-01 4.96375859e-01
-8.44223425e-02 -1.22323656e+00 -8.84980023e-01 -9.10880744e-01
6.35640919e-01 -3.71626258e-01 8.29808891e-01 6.09110057e-01
1.54704988e-01 7.32584745e-02 -4.97465819e-01 -2.34770358e-01
6.82562053e-01 2.49967426e-01 5.94896078e-01 -8.66009235e-01
-7.23653316e-01 -7.55968750e-01 1.49794385e-01 -1.18502498e+00
3.69287163e-01 -7.77205646e-01 1.57567218e-01 -1.59489167e+00
-4.42571081e-02 -3.56511474e-01 -6.75434649e-01 5.18756986e-01
1.35307208e-01 -5.11271596e-01 5.79989970e-01 1.50263771e-01
-7.37612844e-01 7.55817235e-01 1.51664996e+00 -4.33472684e-03
-5.76897085e-01 8.66530761e-02 -1.03675377e+00 7.09160864e-01
1.25453341e+00 -5.67150116e-01 -8.94971550e-01 -2.62673676e-01
4.08350587e-01 4.86486018e-01 7.31753930e-02 -9.05317962e-01
-2.35405937e-02 -8.49936306e-01 3.26032549e-01 -1.76731899e-01
2.18463227e-01 -5.98418295e-01 -1.75731972e-01 1.02287042e+00
-8.37659717e-01 3.45202416e-01 8.54527876e-02 6.97111785e-01
1.66167319e-01 -3.99377525e-01 8.55410695e-01 -4.52607930e-01
-9.47247803e-01 3.17352206e-01 -5.97766578e-01 4.39182490e-01
1.23108017e+00 1.21781841e-01 -3.60316902e-01 -7.48486042e-01
-7.85601616e-01 4.70634401e-01 4.98921812e-01 1.61985815e-01
5.13992548e-01 -1.18939269e+00 -4.36404645e-01 -1.62343774e-02
-2.84984648e-01 -2.67949283e-01 -1.91452447e-02 4.61374521e-01
-6.29438162e-02 2.55769104e-01 -4.92384523e-01 1.09008583e-03
-6.40842617e-01 6.07808352e-01 7.14906633e-01 -8.08377504e-01
-5.01534700e-01 4.54269916e-01 -2.02556998e-01 -1.91129357e-01
4.91509706e-01 -3.22333604e-01 -4.86742556e-01 3.83329511e-01
4.98442262e-01 3.42729121e-01 -5.12095094e-01 -2.57966369e-01
1.37314200e-01 2.76894420e-01 -1.58783734e-01 -4.78606611e-01
1.54133666e+00 -1.30084962e-01 2.23257810e-01 4.60550845e-01
8.41402650e-01 -5.37485361e-01 -1.85889852e+00 -3.25624883e-01
1.34040505e-01 -4.41553056e-01 9.73159671e-02 -1.00259459e+00
-9.32295024e-01 4.93746400e-01 3.94923180e-01 2.63949931e-01
9.42431152e-01 -1.01315975e-01 6.04616821e-01 7.55579233e-01
6.75282598e-01 -1.29126287e+00 3.93645674e-01 8.43904138e-01
9.10325885e-01 -1.07135832e+00 -7.52421618e-02 8.66683483e-01
-1.05615783e+00 1.03615415e+00 6.64245903e-01 -2.92066932e-01
2.61768848e-01 9.82858464e-02 -2.32759416e-01 1.26510963e-01
-1.31056845e+00 -5.17598093e-01 -1.83170646e-01 8.91671836e-01
1.34841308e-01 1.08490266e-01 -9.89497174e-03 3.87470037e-01
-3.59181523e-01 1.71520069e-01 7.08486617e-01 1.04857147e+00
-8.20347905e-01 -1.27030611e+00 -1.89911753e-01 5.90752900e-01
-2.59313077e-01 2.80063957e-01 -2.13368349e-02 7.81419873e-01
-1.31865084e-01 7.28765726e-01 1.98114693e-01 -3.74248147e-01
2.82622963e-01 1.70352757e-01 7.17343748e-01 -5.14337122e-01
-4.60900277e-01 -2.91481376e-01 3.76698673e-02 -8.87330413e-01
-1.04844488e-01 -8.47251832e-01 -1.50810397e+00 -6.11722946e-01
5.35415828e-01 2.70180762e-01 4.32491720e-01 1.06466258e+00
5.31903878e-02 7.66665041e-01 7.81053424e-01 -7.51311600e-01
-1.50717115e+00 -5.73352456e-01 -5.29009700e-01 3.76713157e-01
7.55143404e-01 -7.86617398e-01 -2.57360309e-01 -4.18783993e-01] | [4.0744781494140625, 1.947529673576355] |
90aca550-d590-433b-bcc3-0a1d5d5008af | shadow-detection-with-conditional-generative | null | null | http://openaccess.thecvf.com/content_iccv_2017/html/Nguyen_Shadow_Detection_With_ICCV_2017_paper.html | http://openaccess.thecvf.com/content_ICCV_2017/papers/Nguyen_Shadow_Detection_With_ICCV_2017_paper.pdf | Shadow Detection With Conditional Generative Adversarial Networks | We introduce scGAN, a novel extension of conditional Generative Adversarial Networks (GAN) tailored for the challenging problem of shadow detection in images. Previous methods for shadow detection focus on learning the local appearance of shadow regions, while using limited local context reasoning in the form of pairwise potentials in a Conditional Random Field. In contrast, the proposed adversarial approach is able to model higher level relationships and global scene characteristics. We train a shadow detector that corresponds to the generator of a conditional GAN, and augment its shadow accuracy by combining the typical GAN loss with a data loss term. Due to the unbalanced distribution of the shadow labels, we use weighted cross entropy. With the standard GAN architecture, properly setting the weight for the cross entropy would require training multiple GANs, a computationally expensive grid procedure. In scGAN, we introduce an additional sensitivity parameter w to the generator. The proposed approach effectively parameterizes the loss of the trained detector. The resulting shadow detector is a single network that can generate shadow maps corresponding to different sensitivity levels, obviating the need for multiple models and a costly training procedure. We evaluate our method on the large-scale SBU and UCF shadow datasets, and observe up to 17% error reduction with respect to the previous state-of-the-art method.
| ['Minh Hoai', 'Tomas F. Yago Vicente', 'Vu Nguyen', 'Maozheng Zhao', 'Dimitris Samaras'] | 2017-10-01 | null | null | null | iccv-2017-10 | ['shadow-detection'] | ['computer-vision'] | [ 7.30658591e-01 5.73062062e-01 4.08549458e-01 -2.05928177e-01
-9.41030204e-01 -5.39179265e-01 7.74191558e-01 -2.37194344e-01
-3.03742170e-01 8.17987502e-01 -1.63827971e-01 -2.97797740e-01
5.02000034e-01 -1.03582299e+00 -9.49350536e-01 -1.12927878e+00
2.10501149e-01 6.03074014e-01 4.43936557e-01 -3.95601243e-02
-1.22270256e-01 5.04687130e-01 -1.13887370e+00 6.03258908e-02
1.00936365e+00 1.16894865e+00 3.58936280e-01 7.99987495e-01
1.41103223e-01 8.62606704e-01 -1.07515252e+00 -5.01103938e-01
3.53237987e-01 -8.31623375e-01 -2.25289956e-01 5.66434860e-02
6.15872562e-01 -3.36413532e-01 -7.86595643e-02 7.12448895e-01
6.18410349e-01 3.93308029e-02 8.27974916e-01 -1.23119295e+00
-1.39695451e-01 2.42099196e-01 -3.41030717e-01 -2.83012837e-01
1.21527016e-01 1.48190022e-01 8.59553754e-01 -5.70496559e-01
5.43259978e-01 1.05242801e+00 7.68177032e-01 5.02689540e-01
-1.44005227e+00 -5.80023587e-01 -2.52014883e-02 -6.87988624e-02
-1.20736504e+00 -1.25376701e-01 9.18832004e-01 -2.73260921e-01
3.25731665e-01 3.46752346e-01 6.38974190e-01 1.44342315e+00
1.57657266e-01 6.43294096e-01 1.53516483e+00 -4.74824071e-01
4.28305596e-01 3.08448076e-01 -6.19443655e-01 8.47775102e-01
8.41746628e-02 1.51041046e-01 -3.40793759e-01 -1.94254503e-01
7.18409240e-01 -9.70902443e-02 -3.97077054e-01 -5.75765073e-01
-7.22175896e-01 8.88070107e-01 8.70674908e-01 -6.49645030e-02
-2.56059647e-01 5.34904897e-01 -1.25559717e-01 -1.47781283e-01
4.72019285e-01 3.83555979e-01 -3.59507650e-02 3.96287441e-01
-1.08185577e+00 7.28707388e-02 8.15712392e-01 6.82851374e-01
9.49129581e-01 1.27285153e-01 -5.68592131e-01 3.74006897e-01
2.11150140e-01 9.92695391e-01 -1.15310274e-01 -7.60723650e-01
3.04692090e-01 3.99507552e-01 6.42245710e-02 -7.42086112e-01
-8.03175643e-02 -5.97731531e-01 -7.87522137e-01 6.58546209e-01
4.81964231e-01 -2.87501425e-01 -1.23828387e+00 1.86227989e+00
3.76842499e-01 3.60651761e-01 7.33082965e-02 5.67459702e-01
3.39242101e-01 5.38856566e-01 -1.14180110e-01 8.46382380e-02
1.16508043e+00 -8.95770907e-01 -4.35343683e-01 -4.21210110e-01
6.67284057e-02 -7.64707923e-01 1.01946735e+00 1.52242109e-01
-9.21891630e-01 -2.88139343e-01 -1.20535481e+00 1.33417830e-01
-3.40524763e-01 3.33393365e-01 4.55690384e-01 7.58248508e-01
-9.46103215e-01 5.03361762e-01 -6.85438514e-01 -5.56880459e-02
5.32697678e-01 2.79575258e-01 7.26284981e-02 -4.64146398e-02
-9.79308248e-01 9.19090331e-01 2.05379784e-01 1.09601960e-01
-1.25520563e+00 -6.62112474e-01 -7.71130085e-01 1.63555294e-01
5.59237242e-01 -6.00703776e-01 7.43380666e-01 -1.09231424e+00
-1.56973410e+00 5.51155686e-01 -6.34925663e-02 -6.39740944e-01
9.22800481e-01 -1.17432654e-01 2.00082511e-02 7.05725178e-02
-8.43468457e-02 4.93850470e-01 1.22066081e+00 -1.76365042e+00
-1.26125246e-01 2.50651012e-03 2.98786104e-01 2.00971946e-01
5.67439804e-03 -4.03066069e-01 -3.13157946e-01 -8.24731171e-01
-3.00860614e-01 -1.17459416e+00 -2.64848471e-01 1.98791735e-02
-5.01883030e-01 3.85338128e-01 1.03599012e+00 -7.92482972e-01
7.97910571e-01 -2.09882307e+00 1.10098034e-01 4.87242103e-01
5.89111447e-02 1.79763049e-01 1.41008556e-01 2.23066270e-01
3.37394148e-01 -1.47879019e-01 -9.64163125e-01 -7.87927389e-01
-2.22627129e-02 2.80625284e-01 -3.71372312e-01 3.84030908e-01
1.48511261e-01 9.20695424e-01 -8.09360087e-01 -5.14465570e-01
2.96830118e-01 8.35067749e-01 -3.46407682e-01 5.45003355e-01
-4.36233670e-01 5.77135921e-01 -3.10372591e-01 5.02738059e-01
8.07567656e-01 -1.72883064e-01 3.35675895e-01 -2.19226331e-02
4.13131922e-01 7.51367956e-02 -1.02170420e+00 1.47446394e+00
-9.90411639e-01 7.15527654e-01 1.52813867e-01 -6.24190092e-01
8.79092634e-01 1.30928606e-01 3.23754549e-02 -3.92915547e-01
9.76624042e-02 1.78532094e-01 -2.55223781e-01 1.42919317e-01
2.02532485e-01 -1.83751330e-01 -1.45199716e-01 3.00190359e-01
8.51980746e-02 -4.01135236e-01 -4.66833800e-01 2.24597096e-01
1.27973175e+00 3.85743111e-01 1.51716769e-01 -1.20135158e-01
4.88950491e-01 -4.60191667e-01 3.03925872e-01 7.66450346e-01
3.50329608e-01 9.55023348e-01 5.91586590e-01 -1.60320029e-02
-9.31636274e-01 -1.17608476e+00 6.02190988e-03 7.40298629e-01
6.49678782e-02 -1.57328874e-01 -1.06908512e+00 -1.09963667e+00
-1.16639845e-01 8.49498987e-01 -9.31708872e-01 -5.05029485e-02
-6.98719263e-01 -8.09899151e-01 5.90106726e-01 4.43082303e-01
6.64006948e-01 -1.20684588e+00 -8.94743800e-01 -1.18093282e-01
-1.99890524e-01 -1.16232872e+00 -2.68779725e-01 4.87239391e-01
-5.11054516e-01 -9.88560975e-01 -7.33177245e-01 -2.21696302e-01
9.47529852e-01 -1.52211905e-01 1.30478871e+00 9.37858894e-02
-4.12625849e-01 1.72231227e-01 -2.44777903e-01 -4.92234498e-01
-6.33817315e-01 7.38548860e-03 -4.93624210e-01 3.14927161e-01
-4.23276603e-01 -6.69704139e-01 -8.37531149e-01 1.68766901e-01
-9.52085376e-01 5.61143607e-02 8.91133308e-01 1.09973776e+00
4.67636824e-01 -2.06421077e-01 1.01878680e-01 -1.18817639e+00
4.38033938e-02 -2.90813595e-01 -9.30732071e-01 2.95763612e-01
-7.79982567e-01 3.53710264e-01 8.06907594e-01 -2.75098860e-01
-1.48375750e+00 1.84413940e-01 -9.12273955e-03 -3.90625447e-01
-8.10877830e-02 -1.95212662e-01 -4.97743011e-01 -4.89973098e-01
6.97442591e-01 1.74803451e-01 -3.35549206e-01 -8.63337964e-02
4.07258391e-01 5.90519346e-02 4.22936738e-01 -4.85402435e-01
1.22360885e+00 6.93070412e-01 5.14308274e-01 -5.03679991e-01
-8.34888339e-01 -2.48943958e-02 -4.17326272e-01 -2.27957204e-01
9.05640244e-01 -6.61408603e-01 -3.86106700e-01 4.05718297e-01
-1.07231557e+00 -8.05064142e-01 -6.06490016e-01 -4.63471711e-02
-6.28112435e-01 1.80985272e-01 -1.10568911e-01 -8.89251590e-01
-2.17914164e-01 -9.78585184e-01 1.39270806e+00 -1.44657018e-02
2.90619373e-01 -1.10556626e+00 1.68175362e-02 2.62146115e-01
5.63480914e-01 9.39416409e-01 7.15981305e-01 -1.91136822e-01
-1.01994133e+00 -1.99713409e-01 -1.73562840e-01 7.19885528e-01
-1.53342605e-01 -3.92217189e-01 -1.48575974e+00 -2.29268298e-01
1.31133571e-01 -3.13346803e-01 1.17628551e+00 2.37609267e-01
1.25819004e+00 -4.09516424e-01 -2.35931724e-01 7.70287812e-01
1.75904584e+00 6.42168969e-02 8.87785375e-01 1.56315211e-02
1.02081776e+00 4.08609003e-01 5.01267195e-01 2.82713473e-01
7.19215348e-02 7.82885015e-01 7.11123943e-01 -5.56211293e-01
-5.24393976e-01 -4.04763877e-01 4.52792436e-01 -5.86001314e-02
-2.72388905e-01 -6.71329677e-01 -6.17837429e-01 2.63604522e-01
-1.63855886e+00 -7.76234329e-01 1.87848836e-01 2.35696220e+00
6.00004494e-01 2.35553190e-01 -3.20558250e-01 3.97336446e-02
3.05247366e-01 5.45600593e-01 -4.47643310e-01 -1.28771782e-01
-2.49147162e-01 5.45609117e-01 9.57199991e-01 8.20865273e-01
-9.73767400e-01 9.15022194e-01 5.70816040e+00 1.04680133e+00
-1.04837584e+00 2.69377291e-01 6.33210897e-01 4.28085774e-02
-5.76150179e-01 3.58915865e-01 -4.94685411e-01 4.88423407e-01
6.65594757e-01 5.11174858e-01 5.53747177e-01 7.58370340e-01
-2.40493864e-01 -4.50894922e-01 -8.64401042e-01 5.84736288e-01
3.96270931e-01 -1.03491902e+00 -2.06098393e-01 2.37397686e-01
8.87923777e-01 -9.18534771e-02 1.58199340e-01 -1.07937949e-02
3.72063875e-01 -1.07332408e+00 7.38230586e-01 6.26597583e-01
9.05116498e-01 -6.58510923e-01 8.65861297e-01 1.32062867e-01
-1.09505260e+00 5.17559797e-03 -9.83770713e-02 4.09731477e-01
1.53459296e-01 7.05193222e-01 -1.15538812e+00 5.21952569e-01
4.74791408e-01 -3.75795662e-02 -6.58517599e-01 4.44501460e-01
-6.61985517e-01 6.92457914e-01 -3.64560336e-01 2.57710367e-01
1.86986905e-02 -1.82485238e-01 5.09346068e-01 1.19439268e+00
3.69181514e-01 -3.00570935e-01 2.09666658e-02 8.95806015e-01
-2.61917681e-01 -3.10614705e-01 -5.85475683e-01 3.98694247e-01
2.14422598e-01 1.37825251e+00 -8.49462450e-01 -3.52166384e-01
-3.42461504e-02 1.44744098e+00 2.24564746e-01 3.80243629e-01
-1.15054011e+00 -1.50667712e-01 1.91894561e-01 2.73316294e-01
5.01218379e-01 5.97007088e-02 -3.56892228e-01 -9.60191846e-01
2.73621857e-01 -6.76932037e-01 -4.02299408e-03 -6.91639841e-01
-9.19754207e-01 7.05127895e-01 -1.17451018e-02 -1.06280291e+00
-3.80757213e-01 -3.73949647e-01 -9.11316276e-01 9.10620868e-01
-1.55927038e+00 -1.53611255e+00 -7.90041029e-01 6.00890398e-01
1.61122903e-01 8.91570523e-02 9.09763336e-01 3.67712192e-02
-1.57741278e-01 6.55363083e-01 2.04894561e-02 -9.82088819e-02
8.00721645e-01 -1.62686276e+00 4.32902366e-01 9.95210409e-01
-3.03078908e-02 1.30026117e-01 8.79134715e-01 -4.90826935e-01
-8.40401530e-01 -1.07199621e+00 5.74765623e-01 -5.55993497e-01
3.53726625e-01 -9.62175429e-01 -5.85713208e-01 4.96284306e-01
3.02473128e-01 8.77616182e-02 2.83985049e-01 -3.12263727e-01
-3.12826753e-01 -3.19687992e-01 -1.47050536e+00 3.72254103e-01
9.36698258e-01 -5.43334961e-01 5.32419831e-02 4.06231672e-01
5.59451163e-01 -4.36973691e-01 -4.96943265e-01 4.77499127e-01
4.33773547e-01 -1.30127370e+00 9.78824615e-01 2.26946965e-01
3.63340378e-01 -4.03495669e-01 -1.63786113e-01 -1.29467177e+00
9.42536071e-02 -5.76030731e-01 -2.63213396e-01 1.24574912e+00
3.60089630e-01 -6.25723064e-01 8.92529666e-01 2.45167807e-01
-1.16618298e-01 -9.38579500e-01 -8.53073239e-01 -5.98136485e-01
-3.41281742e-01 -1.84861094e-01 5.07005751e-01 3.38292509e-01
-8.97315204e-01 1.76293850e-01 -6.45995319e-01 3.56091052e-01
8.97285163e-01 2.00507328e-01 9.86439288e-01 -1.03136730e+00
-8.81932199e-01 -1.38762996e-01 -2.23861486e-01 -7.72221863e-01
1.94244102e-01 -5.73771536e-01 4.77282912e-01 -1.33908629e+00
1.04262702e-01 -7.45666027e-01 -8.76035467e-02 3.96619439e-01
-2.59230465e-01 4.74179596e-01 3.76579285e-01 -5.32385334e-02
-1.70735702e-01 6.61722124e-01 1.06678307e+00 -1.03520676e-01
9.48080141e-03 1.82858661e-01 -2.81227499e-01 7.55347669e-01
6.87702239e-01 -5.48550487e-01 -4.38318819e-01 -1.20651066e-01
6.53952956e-02 -3.51550020e-02 9.39585865e-01 -1.15307009e+00
-1.01945782e-02 -2.67680679e-02 6.34497881e-01 -3.08344662e-01
8.90445769e-01 -1.00324285e+00 2.64324099e-01 3.73057872e-01
2.35763751e-02 -3.77691120e-01 -1.22451978e-02 6.70459032e-01
2.28899289e-02 -3.93017381e-02 1.01918936e+00 -1.68786943e-01
-1.98763564e-01 5.62701710e-02 -3.89337316e-02 -1.07322410e-01
1.03448868e+00 9.32308286e-02 -2.21127391e-01 -4.67256457e-01
-4.45100665e-01 -1.43935412e-01 7.77660370e-01 -9.08124745e-02
5.49585223e-01 -1.09018326e+00 -4.35410112e-01 2.35095531e-01
-1.09814063e-01 3.23010385e-01 1.62477180e-01 6.35699749e-01
-4.45356935e-01 -4.99125682e-02 -1.23398274e-01 -3.98088962e-01
-1.07749164e+00 3.50535095e-01 5.76283455e-01 -6.13128066e-01
-5.54196298e-01 8.80205929e-01 6.97292268e-01 -1.84927389e-01
2.03477010e-01 -8.41946825e-02 2.09900498e-01 -4.30046581e-02
-5.11494353e-02 3.40984672e-01 6.19511446e-03 -4.27417219e-01
-3.23706955e-01 4.47295398e-01 4.78352636e-01 -3.21204305e-01
9.49291348e-01 1.36374712e-01 2.62397658e-02 2.51873016e-01
9.64271605e-01 4.17678356e-01 -1.71557927e+00 6.55904785e-02
-4.72409070e-01 -3.75435323e-01 1.99974980e-02 -9.88455892e-01
-1.12657154e+00 7.72935867e-01 8.78548861e-01 8.84224381e-03
1.45072174e+00 9.57596526e-02 7.23396420e-01 1.07583642e-01
4.44305211e-01 -8.42058778e-01 9.37409252e-02 1.64016455e-01
8.20728898e-01 -1.27658701e+00 3.61112915e-02 -4.68307048e-01
-6.63411558e-01 7.61755168e-01 4.38705117e-01 -3.84145230e-01
4.05921012e-01 4.09070671e-01 3.85980047e-02 -8.39987993e-02
-2.09847718e-01 -1.28296047e-01 3.50828052e-01 5.09362519e-01
-2.50813346e-02 2.98960030e-01 -6.69528618e-02 4.81286726e-04
-1.34370819e-01 -4.25196439e-01 3.71517181e-01 6.93519175e-01
-1.09930336e-01 -1.24895597e+00 -4.32507277e-01 2.28634298e-01
-3.21413428e-01 -2.55399436e-01 -5.37171125e-01 6.27895653e-01
4.59591836e-01 7.49126434e-01 -1.74713731e-01 -3.43892664e-01
6.28219396e-02 -4.68171276e-02 5.09521782e-01 -3.67560089e-01
-4.90361631e-01 3.94980758e-02 -1.05124258e-01 -7.05579638e-01
-3.83157343e-01 -5.81987679e-01 -1.03864837e+00 3.65243070e-02
-4.26477760e-01 -3.83972265e-02 8.23515952e-01 8.82918119e-01
6.64049163e-02 7.22428739e-01 8.41454566e-01 -1.03941059e+00
-3.33468914e-01 -9.61512268e-01 -3.48656237e-01 2.21915185e-01
5.73102593e-01 -7.96375751e-01 -5.89953244e-01 -1.44633770e-01] | [10.846348762512207, -4.102905750274658] |
3f446d2b-95df-4101-8bb0-3826538e8459 | omnixai-a-library-for-explainable-ai | 2206.01612 | null | https://arxiv.org/abs/2206.01612v8 | https://arxiv.org/pdf/2206.01612v8.pdf | OmniXAI: A Library for Explainable AI | We introduce OmniXAI (short for Omni eXplainable AI), an open-source Python library of eXplainable AI (XAI), which offers omni-way explainable AI capabilities and various interpretable machine learning techniques to address the pain points of understanding and interpreting the decisions made by machine learning (ML) in practice. OmniXAI aims to be a one-stop comprehensive library that makes explainable AI easy for data scientists, ML researchers and practitioners who need explanation for various types of data, models and explanation methods at different stages of ML process (data exploration, feature engineering, model development, evaluation, and decision-making, etc). In particular, our library includes a rich family of explanation methods integrated in a unified interface, which supports multiple data types (tabular data, images, texts, time-series), multiple types of ML models (traditional ML in Scikit-learn and deep learning models in PyTorch/TensorFlow), and a range of diverse explanation methods including "model-specific" and "model-agnostic" ones (such as feature-attribution explanation, counterfactual explanation, gradient-based explanation, etc). For practitioners, the library provides an easy-to-use unified interface to generate the explanations for their applications by only writing a few lines of codes, and also a GUI dashboard for visualization of different explanations for more insights about decisions. In this technical report, we present OmniXAI's design principles, system architectures, and major functionalities, and also demonstrate several example use cases across different types of data, tasks, and models. | ['Tanmay Laud', 'Steven C. H. Hoi', 'Silvio Savarese', 'Hung Le', 'Wenzhuo Yang'] | 2022-06-01 | null | null | null | null | ['counterfactual-explanation'] | ['miscellaneous'] | [-3.34820479e-01 5.54780543e-01 -3.20772231e-01 -4.83132511e-01
9.94871035e-02 -4.72616225e-01 6.20091319e-01 -8.07462856e-02
4.14571494e-01 7.15742767e-01 3.84016365e-01 -8.68302941e-01
-5.10619283e-01 -3.92118126e-01 -5.34788609e-01 -3.90094191e-01
-1.28366753e-01 6.45015419e-01 -6.16325021e-01 -8.67215097e-02
5.57379425e-01 3.75645816e-01 -1.69110072e+00 4.32492703e-01
1.11803603e+00 6.37721777e-01 -8.13301057e-02 6.96020007e-01
-4.77193534e-01 9.27107990e-01 -3.73800576e-01 -4.52070653e-01
1.06249265e-02 -4.03213888e-01 -8.54625523e-01 -2.57140636e-01
-3.54314409e-02 -1.71804264e-01 -1.05134502e-01 5.94344556e-01
1.33410096e-01 -2.84237921e-01 7.63862371e-01 -2.11427903e+00
-1.37437570e+00 9.65867341e-01 2.74482486e-03 3.71975005e-02
4.92616564e-01 8.01083982e-01 6.20218337e-01 -9.26882505e-01
4.07280594e-01 1.49836278e+00 5.64315021e-01 7.40125239e-01
-1.04941976e+00 -6.80960774e-01 1.64387926e-01 5.46623290e-01
-6.84740365e-01 -4.76266891e-02 5.47341466e-01 -6.04148030e-01
9.47752118e-01 9.63404477e-01 9.12510514e-01 1.39662421e+00
6.80582225e-01 1.02322626e+00 1.05939746e+00 -1.66269571e-01
4.44469661e-01 3.06429476e-01 7.14815855e-01 8.67181897e-01
3.62565964e-01 3.58072579e-01 -6.85615420e-01 -2.34862626e-01
8.11317742e-01 5.18789232e-01 -1.21092819e-01 -1.53154433e-01
-1.57310200e+00 7.83925116e-01 5.80519438e-01 1.51544034e-01
-5.39121926e-01 3.36367786e-01 3.16326439e-01 5.39460003e-01
2.53614455e-01 8.52748752e-01 -9.41969454e-01 -6.50217384e-02
-9.85961556e-02 6.05776608e-01 9.59930778e-01 9.36870635e-01
9.12216961e-01 4.77540195e-01 -3.27098578e-01 1.59277886e-01
5.73678851e-01 4.30854112e-01 9.46775317e-01 -9.59661007e-01
6.59805685e-02 1.14900875e+00 7.11212754e-02 -8.58994424e-01
-6.87080026e-01 -5.11182845e-01 -9.94491637e-01 4.22621131e-01
9.99135245e-03 -2.03805253e-01 -9.66514945e-01 1.35444045e+00
3.51298690e-01 -6.28665388e-02 1.52627647e-01 1.23530996e+00
1.45838368e+00 4.48134214e-01 3.66507262e-01 1.61520958e-01
1.43403757e+00 -1.28373361e+00 -9.24967468e-01 -4.53864574e-01
6.08741522e-01 -3.96203637e-01 1.64679277e+00 3.90476108e-01
-8.04753959e-01 -5.63042104e-01 -8.48511100e-01 -3.00032914e-01
-8.65344048e-01 -6.23053871e-03 1.52388620e+00 1.73456401e-01
-7.01860070e-01 8.07805121e-01 -7.11466372e-01 -3.41133475e-01
2.31960356e-01 4.17560697e-01 -4.15992379e-01 8.26424137e-02
-1.01105654e+00 9.65070248e-01 1.69104591e-01 -6.38993159e-02
-8.83803427e-01 -1.02264512e+00 -8.93961668e-01 4.98604029e-01
2.15350091e-01 -1.30467522e+00 1.11795747e+00 -1.13475776e+00
-1.36212611e+00 4.37460244e-01 -1.17544025e-01 -3.04372817e-01
4.98477727e-01 -4.02337044e-01 -2.72804409e-01 -6.61821723e-01
2.28150357e-02 5.97839534e-01 5.73290467e-01 -9.16849613e-01
-3.78275453e-03 -4.62618858e-01 2.05751002e-01 1.33470923e-01
1.34283423e-01 -1.11766860e-01 -2.54314542e-02 -3.93055528e-01
-5.93355037e-02 -8.69097292e-01 -4.95435953e-01 3.51565212e-01
-9.32826698e-01 -2.79712349e-01 8.79073858e-01 -5.55174828e-01
1.06928718e+00 -2.01537156e+00 6.81899637e-02 -1.19302422e-02
5.55544913e-01 -1.89455040e-02 1.10682333e-02 3.36118519e-01
-4.83426630e-01 5.47402382e-01 -2.95571595e-01 -3.08573395e-01
3.99902999e-01 4.06825036e-01 -6.16345406e-02 1.51389409e-02
4.31155525e-02 1.20511365e+00 -9.20662045e-01 -2.95141041e-01
6.36474252e-01 3.94953549e-01 -5.18400729e-01 4.67426181e-01
-3.42254996e-01 7.03703463e-01 -7.34345853e-01 7.61521161e-01
3.53445083e-01 -4.71965402e-01 -1.08646221e-01 4.56634164e-02
-3.08344752e-01 2.66632140e-01 -1.18014514e+00 1.55108583e+00
-4.02866483e-01 8.59394610e-01 -5.02798855e-01 -5.66213608e-01
9.07640159e-01 4.79203820e-01 5.39408214e-02 -1.28420725e-01
1.30457103e-01 1.92344218e-01 1.06319882e-01 -9.13244843e-01
1.03169074e-02 3.84130448e-01 1.44042358e-01 8.28464627e-01
-1.33912414e-01 3.58481407e-02 -2.51452271e-02 -8.01349059e-02
1.05656397e+00 1.89794511e-01 8.11041057e-01 -4.33291286e-01
3.06683183e-01 3.01725328e-01 3.47464830e-01 8.45542669e-01
2.58584410e-01 4.61478710e-01 4.62944061e-01 -1.47969890e+00
-8.45912755e-01 -8.64376724e-01 -5.57317548e-02 9.26924706e-01
-8.08369964e-02 -3.42179358e-01 -5.97384393e-01 -5.70012748e-01
2.89073586e-01 1.33692884e+00 -9.54980969e-01 -2.78127074e-01
3.67772169e-02 -6.48400187e-01 -6.00102730e-02 4.00333226e-01
6.84029698e-01 -1.72896135e+00 -7.58010745e-01 -8.22011977e-02
-1.24896970e-02 -3.19913149e-01 -8.94988850e-02 3.55376810e-01
-1.07014024e+00 -1.13590789e+00 4.58261631e-02 -7.59141222e-02
8.17128241e-01 2.71324366e-01 1.45633411e+00 5.79747617e-01
-4.16836888e-01 2.21379131e-01 -1.21895418e-01 -8.52384865e-01
-2.50251412e-01 -1.24375373e-01 -2.93756127e-02 -3.66561532e-01
3.92971963e-01 -6.27464950e-01 -7.18574166e-01 2.61007220e-01
-9.39898372e-01 7.62106001e-01 4.71761346e-01 7.14301348e-01
3.85234088e-01 -5.70839345e-01 1.88247532e-01 -9.69074607e-01
9.38827455e-01 -8.95753086e-01 -2.72209555e-01 3.51180553e-01
-1.17046106e+00 5.52631557e-01 7.70421982e-01 -5.59597075e-01
-8.94743502e-01 -2.86775440e-01 1.27669305e-01 -2.67546117e-01
-4.52139676e-01 7.30808735e-01 -1.97404489e-01 2.32936218e-01
8.48320901e-01 2.98311580e-02 5.77114224e-02 -7.73976088e-01
6.35079265e-01 6.42320693e-01 4.29370880e-01 -2.73640215e-01
6.20037138e-01 1.12245843e-01 -2.09617153e-01 -2.91198981e-03
-6.60499632e-01 3.33648443e-01 -4.88166839e-01 -1.36881322e-01
7.14122236e-01 -5.25179744e-01 -1.08290327e+00 -1.74369201e-01
-1.22629583e+00 -3.35577160e-01 -4.10151392e-01 5.30041635e-01
-4.87519622e-01 -3.23606342e-01 -2.94245601e-01 -5.29480815e-01
-7.96375334e-01 -1.34135497e+00 7.23672509e-01 5.59427559e-01
-8.89803708e-01 -1.22718644e+00 2.19069310e-02 1.51916578e-01
7.37065077e-01 4.48335379e-01 1.16647756e+00 -1.06750762e+00
-7.08724082e-01 -2.58235037e-01 -6.91300705e-02 -1.41570419e-01
-4.36421856e-02 4.40535516e-01 -9.55628276e-01 4.95926052e-01
-2.83601314e-01 2.04957416e-03 2.48516351e-01 4.77553070e-01
1.77477229e+00 -9.15774405e-01 -5.07380307e-01 7.77228951e-01
1.01483881e+00 7.94169903e-02 5.66424608e-01 4.45650160e-01
7.18053579e-01 3.65367830e-01 3.42089087e-01 5.70353031e-01
5.83381355e-01 4.08704102e-01 8.53114128e-01 -5.36689997e-01
1.60284206e-01 -1.54854551e-01 2.35234231e-01 4.50430632e-01
-2.46645331e-01 -8.64185691e-02 -8.09135795e-01 2.03495428e-01
-2.38637280e+00 -9.55241919e-01 -7.24291146e-01 2.05462861e+00
3.10016543e-01 -2.61770248e-01 -2.05280736e-01 -8.03375524e-03
1.70899153e-01 -4.21192974e-01 -1.06849885e+00 -1.12133956e+00
6.10580668e-02 -2.43483126e-01 -4.00745794e-02 5.29951096e-01
-6.57927513e-01 7.13737667e-01 7.29246569e+00 2.24867702e-01
-1.16440797e+00 -7.16640800e-02 6.45758510e-01 1.64408758e-01
-9.08700824e-01 2.67357469e-01 -1.07591309e-01 4.11323994e-01
1.00042558e+00 -6.48340762e-01 7.84644425e-01 1.40315151e+00
4.38298225e-01 2.65037268e-01 -1.64555860e+00 8.37974787e-01
-4.64779258e-01 -1.78496587e+00 2.18402877e-01 -1.18425690e-01
3.85179371e-01 -7.14378580e-02 7.61966556e-02 4.90106791e-01
4.31429535e-01 -1.19346523e+00 6.85986996e-01 7.11640060e-01
4.50882226e-01 -3.00646931e-01 7.03983247e-01 2.31711611e-01
-6.83249891e-01 -3.18015575e-01 -4.06932414e-01 -6.86998904e-01
-3.95951867e-01 5.27486622e-01 -7.55594254e-01 6.22616589e-01
8.09652805e-01 8.72873127e-01 -7.11690903e-01 1.03854346e+00
-6.22773349e-01 5.05216062e-01 -3.63037623e-02 -2.77873784e-01
-1.94182415e-02 -8.46572369e-02 4.75614786e-01 1.08206451e+00
3.55644912e-01 5.96006140e-02 -2.64419407e-01 1.52341783e+00
4.86804187e-01 -5.90037070e-02 -8.10161173e-01 1.38587430e-01
5.10403872e-01 1.50072503e+00 -3.20011526e-01 -5.15008509e-01
-1.91244096e-01 7.74120450e-01 1.47733822e-01 4.01187241e-01
-9.48376834e-01 -2.13228446e-02 9.25802052e-01 -1.11123227e-01
-5.69240868e-01 1.41524941e-01 -9.79558945e-01 -1.40006590e+00
-3.88609588e-01 -1.29237437e+00 4.01578754e-01 -1.61074257e+00
-1.18146086e+00 6.38529897e-01 1.94287509e-01 -1.04539764e+00
-6.21591389e-01 -7.48819053e-01 -1.22587061e+00 7.58677006e-01
-9.91253376e-01 -9.80812728e-01 -7.25851059e-01 5.58845639e-01
7.13442206e-01 -2.68748730e-01 1.17373669e+00 -1.21044874e-01
-8.36412430e-01 2.22774446e-01 -6.53687585e-03 -3.48395109e-01
1.71691865e-01 -1.44107103e+00 6.95610464e-01 2.72863775e-01
1.45629063e-01 1.06490970e+00 1.19233060e+00 -3.27968121e-01
-1.64927840e+00 -6.72719479e-01 9.12205577e-01 -6.57455981e-01
3.38821143e-01 -1.83888882e-01 -9.93018210e-01 1.15281582e+00
3.10899854e-01 -3.36009860e-01 7.79956818e-01 5.05113721e-01
5.30594334e-05 1.44371269e-02 -1.13687062e+00 1.13671088e+00
9.94632840e-01 -3.52103673e-02 -5.10328174e-01 7.30253279e-01
8.52225661e-01 -2.84609795e-01 -6.32462502e-01 1.30624399e-01
6.87614083e-01 -1.12886071e+00 8.77605200e-01 -1.08747995e+00
7.52392292e-01 -3.76812488e-01 2.96527088e-01 -1.39440262e+00
-5.38386881e-01 -7.88321197e-01 -3.59589040e-01 9.34234083e-01
8.08833599e-01 -8.17630589e-01 3.36663753e-01 1.41428542e+00
-4.21429634e-01 -9.04417634e-01 -4.88830745e-01 -2.58755684e-01
-4.37573224e-01 -6.02640152e-01 1.35138535e+00 9.92264986e-01
3.99696082e-01 2.77583838e-01 -3.17102253e-01 9.87968296e-02
4.65392381e-01 3.28858405e-01 1.24880421e+00 -1.41138303e+00
-2.15094030e-01 -3.70915830e-01 -1.99320897e-01 -6.76806629e-01
-2.81425584e-02 -8.60365093e-01 -4.85098869e-01 -1.78521860e+00
3.90759587e-01 -2.94198275e-01 -6.62995577e-02 1.04801118e+00
-5.46605527e-01 -4.67973679e-01 2.22130924e-01 5.49712479e-01
-2.99533874e-01 5.54982960e-01 1.24671221e+00 -9.19921100e-02
-2.40613729e-01 -7.43291676e-02 -1.11447358e+00 8.78328919e-01
6.57797277e-01 -7.18196273e-01 -4.46304828e-01 -6.85732782e-01
1.42906442e-01 9.09863412e-02 7.72316456e-01 -8.80679250e-01
-1.65751472e-01 -5.09884179e-01 6.07455075e-01 -2.45280579e-01
-1.18997522e-01 -8.31809342e-01 8.25840592e-01 7.88235605e-01
-6.12438440e-01 5.30547559e-01 3.48545611e-01 1.18355207e-01
1.95358366e-01 -9.94994864e-02 1.23314606e-02 -3.04765016e-01
-6.11566722e-01 3.53001833e-01 -4.21879768e-01 -5.54159522e-01
8.16624045e-01 -1.97569609e-01 -5.96772313e-01 -6.25816524e-01
-7.92343915e-01 4.48962569e-01 2.20776439e-01 7.17292786e-01
5.70111871e-01 -1.27560484e+00 -6.13605082e-01 3.36921662e-01
1.32827014e-01 -2.11779833e-01 1.34150907e-01 5.72656870e-01
-4.52849865e-01 4.87478077e-01 -4.81751472e-01 -5.01304865e-01
-8.89918625e-01 7.75136471e-01 3.91397804e-01 -2.05128059e-01
-7.59252250e-01 4.74574715e-01 4.40188169e-01 -1.05096960e+00
-7.63772847e-03 -6.24764621e-01 -1.11970171e-01 -7.95620501e-01
6.14402175e-01 3.15252602e-01 -3.50135863e-01 3.71845961e-01
-4.00930703e-01 1.84538215e-01 2.88529813e-01 2.58940309e-01
1.40484667e+00 -7.77930841e-02 3.09742242e-02 8.53123188e-01
5.20097911e-01 -6.53603792e-01 -1.19001281e+00 2.64506191e-01
-9.87560526e-02 -2.52574325e-01 -3.23695481e-01 -1.44251120e+00
-8.61599684e-01 9.82364953e-01 4.86733407e-01 3.65869612e-01
7.36798942e-01 -5.62842190e-02 1.11258529e-01 4.20250803e-01
1.53345987e-01 -3.18503618e-01 -2.39385858e-01 5.76967709e-02
1.51587760e+00 -1.41708279e+00 2.66182989e-01 -4.53718267e-02
-9.83153939e-01 1.29935360e+00 7.31850326e-01 2.12570846e-01
4.85296607e-01 -1.02842471e-03 3.23325515e-01 -5.68214059e-01
-1.23612428e+00 1.94019079e-01 6.11538053e-01 6.60341680e-01
7.34691679e-01 2.72963971e-01 -2.77497500e-01 1.03788495e+00
-5.54336607e-01 2.14864656e-01 3.08256388e-01 3.98319662e-01
-1.34740800e-01 -7.30633020e-01 -4.79892850e-01 5.16202450e-01
-9.42538772e-03 -1.77562907e-01 -5.89980543e-01 1.22097409e+00
1.68287456e-01 8.87567997e-01 -7.36231823e-03 -3.58578146e-01
1.33072093e-01 1.55933037e-01 -1.86710045e-01 -2.40695313e-01
-9.38313723e-01 -6.59053445e-01 4.92275087e-03 -7.63993263e-01
1.35295808e-01 -2.97125548e-01 -1.40010655e+00 -8.72298241e-01
-4.14857790e-02 1.99008390e-01 1.11828971e+00 1.27766764e+00
8.03086758e-01 6.70690358e-01 3.04188907e-01 -8.95442486e-01
-6.60210103e-02 -1.10464025e+00 -1.72122866e-01 5.09639263e-01
1.87900677e-01 -6.78666532e-01 -6.84145749e-01 2.76963180e-03] | [8.8135986328125, 5.878915309906006] |
f414c712-384f-4937-9d10-74ce9755d663 | real-time-tone-mapping-a-state-of-the-art | 2003.03074 | null | https://arxiv.org/abs/2003.03074v1 | https://arxiv.org/pdf/2003.03074v1.pdf | Real-time Tone Mapping: A State of the Art Report | The rising demand for high quality display has ensued active research in high dynamic range (HDR) imaging, which has the potential to replace the standard dynamic range imaging. This is due to HDR's features like accurate reproducibility of a scene with its entire spectrum of visible lighting and color depth. But this capability comes with expensive capture, display, storage and distribution resource requirements. Also, display of HDR images/video content on an ordinary display device with limited dynamic range requires some form of adaptation. Many adaptation algorithms, widely known as tone mapping operators, have been studied and proposed in the last few decades. In this state of the art report, we present a comprehensive survey of 50+ tone mapping algorithms that have been implemented on hardware for acceleration and real-time performance. These algorithms have been adapted or redesigned to make them hardware-friendly. All real-time application poses strict timing constraints which requires time exact processing of the algorithm. This design challenge require novel solution, and in this report we focus on these issues. In this we survey will discuss those tonemap algorithms which have been implemented on GPU [1-10], FPGA [11-41], and ASIC [42-53] in terms of their hardware specifications and performance. Output image quality is an important metric for tonemap algorithms. From our literature survey we found that, various objective quality metrics have been used to demonstrate the functionality of adapting the algorithm on hardware platform. We have compiled and studied all the metrics used in this survey [54-67]. Finally, in this report we demonstrate the link between hardware cost and image quality thereby illustrating the underlying trade-off which will be useful for the research community. | ['Tetsuya Asai', 'Masato Motomura', 'Shinya Takamaeda', 'Masayuki Ikebe', 'Prasoon Ambalathankandy', 'Yafei Ou'] | 2020-03-06 | null | null | null | null | ['tone-mapping'] | ['computer-vision'] | [ 6.66175306e-01 -6.56121790e-01 2.62341917e-01 -3.93306851e-01
-2.83391684e-01 -4.42949265e-01 1.18279710e-01 -4.51198429e-01
-2.45647073e-01 4.87629265e-01 -9.88572985e-02 -2.29745567e-01
-5.22896126e-02 -7.69270599e-01 -3.93033892e-01 -5.63966334e-01
-1.46442726e-01 -1.25973761e-01 5.12636721e-01 -4.62555826e-01
3.75875652e-01 5.61207473e-01 -1.78087008e+00 1.99701145e-01
6.73133612e-01 1.08483624e+00 5.15335143e-01 1.16589034e+00
1.64234161e-01 4.92537826e-01 -6.66631341e-01 -2.72205949e-01
6.32707775e-01 -6.57984257e-01 -2.44470432e-01 -3.84773239e-02
4.71389979e-01 -5.45089304e-01 -3.84608984e-01 9.60737467e-01
9.80218828e-01 -1.09687962e-01 3.47117603e-01 -8.56559157e-01
-5.19607604e-01 2.08399631e-02 -9.14827466e-01 5.31281769e-01
6.44295871e-01 1.16022490e-01 8.48290846e-02 -6.66095495e-01
3.91409904e-01 7.74326801e-01 6.44631386e-01 4.34078366e-01
-8.64429295e-01 -7.92729259e-01 -6.48458540e-01 2.65590876e-01
-1.44714677e+00 -1.58085719e-01 7.78647721e-01 -1.94292977e-01
1.13759112e+00 7.81840026e-01 7.94777036e-01 4.43282604e-01
6.76754713e-01 3.26582836e-03 1.66488659e+00 -6.69825673e-01
1.27043696e-02 2.82870322e-01 -3.84681299e-02 6.48862302e-01
2.41298959e-01 2.98984021e-01 -6.28195524e-01 7.82985240e-02
1.10740495e+00 -2.75316834e-01 -5.06790996e-01 -6.24041632e-02
-9.86173689e-01 4.44397688e-01 2.64133930e-01 1.44535273e-01
1.92731584e-03 1.81707844e-01 4.04404879e-01 6.06697738e-01
-7.63000622e-02 2.73190320e-01 2.08638236e-02 -3.89436990e-01
-8.72521043e-01 -3.50022018e-01 4.36784923e-01 9.63363290e-01
5.45049012e-01 3.41631830e-01 1.78112447e-01 9.73789692e-01
1.15085512e-01 6.78932428e-01 5.83124042e-01 -7.49914527e-01
1.69414639e-01 2.16227189e-01 -1.62506491e-01 -1.03020453e+00
-3.42531115e-01 -1.66031703e-01 -1.05069840e+00 8.33898723e-01
-9.29342806e-02 -3.72729823e-02 -9.81678188e-01 1.08807111e+00
3.37930948e-01 5.32282889e-02 -3.14909220e-02 1.14000106e+00
8.50091159e-01 1.10405898e+00 -3.05141300e-01 -3.77697855e-01
1.42902458e+00 -5.58562875e-01 -9.06166494e-01 2.47771725e-01
-1.15642726e-01 -1.43205690e+00 1.43161416e+00 7.71926284e-01
-1.24428236e+00 -8.58481050e-01 -1.69629192e+00 -2.80243814e-01
-2.34448195e-01 -5.35965944e-03 6.69285893e-01 1.35589504e+00
-1.25335228e+00 2.80737728e-01 -5.91698289e-01 -4.58391458e-01
-1.32914320e-01 6.04560614e-01 4.62861098e-02 -1.90276075e-02
-9.65184689e-01 8.16161394e-01 8.59818608e-02 -6.97579160e-02
-2.88412362e-01 -8.64062905e-01 -2.31043637e-01 -1.39812157e-01
-8.70400891e-02 -6.06444597e-01 9.70419288e-01 -1.00553739e+00
-2.16758227e+00 9.44002986e-01 1.90689892e-01 -1.71444789e-01
5.17623842e-01 -9.03506279e-02 -9.21976030e-01 2.96712309e-01
-4.59854841e-01 2.94652581e-01 6.70963943e-01 -7.99555004e-01
-7.59766877e-01 -1.35173574e-01 -6.96362630e-02 4.77309465e-01
-1.33774519e-01 3.55979830e-01 -6.15635931e-01 -7.09049821e-01
2.34037876e-01 -9.86891985e-01 5.42145707e-02 1.14767127e-01
1.15453623e-01 5.12984693e-01 1.23267472e+00 -3.76783192e-01
1.59759629e+00 -2.19279623e+00 -5.02752662e-01 2.98195809e-01
-1.10105969e-01 2.12933362e-01 5.28612018e-01 3.45906228e-01
-7.19566494e-02 -3.20945442e-01 -2.29293611e-02 4.54195708e-01
-3.19529831e-01 -1.77146927e-01 -5.00327826e-01 7.49535382e-01
-3.43147993e-01 4.11158502e-01 -1.13051943e-01 -3.28217655e-01
7.59594679e-01 7.04811871e-01 -4.20294642e-01 1.19642913e-01
3.97437930e-01 2.97780633e-01 -3.02700307e-02 7.12531030e-01
9.95435238e-01 -2.65111197e-02 -6.78829700e-02 -6.49799645e-01
-5.85139692e-01 -2.26505011e-01 -1.36522579e+00 1.33165634e+00
-5.87857783e-01 1.08850086e+00 -1.18658476e-01 -3.72724086e-01
1.11935163e+00 2.35245600e-01 3.75831723e-01 -1.26608360e+00
1.52820185e-01 3.62395316e-01 1.06918000e-01 -5.39517939e-01
9.19228017e-01 -3.83609273e-02 -7.93496519e-02 3.45463455e-01
-5.08824825e-01 -2.90330082e-01 7.56516606e-02 1.42053878e-02
7.27455616e-01 1.17283136e-01 5.57756662e-01 -5.35647094e-01
6.37249172e-01 2.04212129e-01 7.90511593e-02 5.44820726e-01
-2.20539913e-01 7.68524170e-01 -1.05960935e-01 -5.30113339e-01
-1.39129901e+00 -1.01973045e+00 -4.50779349e-01 9.50931668e-01
6.17496490e-01 -9.45145935e-02 -4.76524711e-01 4.01488364e-01
-6.26271307e-01 8.09511915e-02 -2.74862409e-01 8.89390558e-02
-1.02751434e+00 -9.95746553e-01 4.72190589e-01 1.64806873e-01
9.65323925e-01 -6.24501526e-01 -1.63407385e+00 1.44884542e-01
4.08802360e-01 -9.57131088e-01 -3.04506838e-01 1.43823862e-01
-9.04731512e-01 -7.70748556e-01 -7.34761477e-01 -8.48997891e-01
4.77831483e-01 4.39567685e-01 1.13849401e+00 3.90388328e-03
-9.33678448e-01 5.52534699e-01 -3.52479666e-01 -2.10140646e-01
-1.04522891e-01 -3.14988732e-01 -1.12872399e-01 -2.79559493e-01
6.72320575e-02 -5.58860660e-01 -1.00536489e+00 5.31546950e-01
-9.02181625e-01 3.47981870e-01 6.93136632e-01 5.77986479e-01
9.39245045e-01 1.02457538e-01 1.19111829e-01 -9.44929779e-01
4.59840924e-01 2.73583625e-02 -9.40016031e-01 2.71932006e-01
-5.77457726e-01 -1.73245937e-01 6.34354353e-01 -3.97524387e-01
-1.24552643e+00 1.08511403e-01 -8.48242864e-02 -3.33998650e-02
2.56362796e-01 -4.34267409e-02 -3.51802595e-02 -6.24812007e-01
9.09354031e-01 2.71945894e-01 -1.54637501e-01 -5.21150716e-02
9.69479233e-02 9.59710538e-01 9.16731536e-01 -2.66525030e-01
7.20069528e-01 6.48200035e-01 3.62985104e-01 -9.99101520e-01
-1.67146936e-01 -2.73127347e-01 -9.20859650e-02 -6.07955873e-01
7.75584817e-01 -8.35921884e-01 -6.82856143e-01 5.08925319e-01
-4.86872882e-01 -1.93257272e-01 6.90443367e-02 5.49105525e-01
-6.02837622e-01 1.73587859e-01 -5.63287497e-01 -4.88490373e-01
-7.01955259e-01 -1.24055529e+00 7.07996905e-01 8.05603623e-01
1.58001054e-02 -6.89972997e-01 1.10447221e-01 -2.13403448e-01
9.11686599e-01 4.59623903e-01 6.49171412e-01 6.77967012e-01
-8.32980871e-01 1.07800871e-01 -3.77011031e-01 -1.81988850e-01
1.83161408e-01 2.62750715e-01 -9.50955927e-01 -3.92959893e-01
2.67167687e-01 6.35961816e-02 3.69807690e-01 5.62119842e-01
1.19048452e+00 8.19595307e-02 -2.13322014e-01 1.21028149e+00
2.25246596e+00 7.08983898e-01 1.02230763e+00 4.79757965e-01
2.63305098e-01 9.42398012e-02 9.55967605e-01 4.00810212e-01
-2.21330166e-01 1.00075495e+00 3.39061059e-02 -3.37860793e-01
-3.82943839e-01 2.07160324e-01 3.04222733e-01 8.74841690e-01
-2.25911677e-01 -1.81721702e-01 -7.15864062e-01 -3.05859178e-01
-1.11792684e+00 -9.01155055e-01 -4.40161914e-01 2.28561425e+00
7.87390172e-01 8.25779587e-02 3.24907899e-02 2.60734111e-01
8.08726370e-01 -9.93191451e-02 -5.13422668e-01 -9.92979288e-01
-1.64356902e-01 5.22306323e-01 9.09375310e-01 4.01158214e-01
-7.09482133e-01 4.76974308e-01 6.78508472e+00 7.38345563e-01
-1.80338609e+00 4.22078744e-02 6.78081930e-01 -2.92664409e-01
4.34455695e-03 -2.19528198e-01 -7.31008589e-01 5.75538576e-01
9.37999904e-01 -3.95846367e-01 4.86484557e-01 7.24484324e-01
1.57251552e-01 -5.37203133e-01 -7.14347780e-01 1.59504426e+00
2.90611565e-01 -1.07825434e+00 -3.08940858e-01 -1.23006171e-02
8.16951215e-01 -2.42774233e-01 5.73342562e-01 -1.82533115e-01
-2.94376701e-01 -8.90661418e-01 4.72905874e-01 3.27194571e-01
1.55520439e+00 -9.07608628e-01 3.88744682e-01 -3.37875247e-01
-1.27676022e+00 1.26027435e-01 -6.01809919e-01 1.39841391e-02
2.84253985e-01 5.21697283e-01 -5.36593497e-01 2.68343151e-01
1.05690825e+00 2.14582961e-02 -6.12879992e-01 1.17058551e+00
4.00451481e-01 3.49911302e-01 -5.46928763e-01 -5.20166643e-02
-1.06827393e-01 -3.71163726e-01 2.61345893e-01 1.29260492e+00
7.00961292e-01 5.28372347e-01 -1.85529888e-01 3.25594097e-01
3.98190826e-01 1.23112693e-01 -5.79472482e-01 5.20263970e-01
3.60984892e-01 1.25335288e+00 -9.77297306e-01 -3.89327914e-01
-6.32213235e-01 1.10452282e+00 -4.52440679e-01 -9.03625935e-02
-1.26326966e+00 -8.92991424e-01 3.90542269e-01 2.96970785e-01
-1.21648923e-01 -3.46166581e-01 -6.04906559e-01 -6.43627167e-01
-5.90219758e-02 -8.47291291e-01 4.26490158e-01 -9.72293794e-01
-6.75964892e-01 8.47292364e-01 -5.55222370e-02 -1.68303764e+00
-6.72422051e-02 -5.95085919e-01 -2.24878699e-01 9.42006409e-01
-1.43687260e+00 -6.31463289e-01 -8.53632092e-01 8.18736315e-01
6.21923745e-01 -1.24522112e-01 4.81031626e-01 6.51306808e-01
-1.64285555e-01 7.35892713e-01 1.18108608e-01 -5.18567443e-01
8.20536315e-01 -9.18089926e-01 9.77371410e-02 1.03577745e+00
-3.42998624e-01 5.78166604e-01 8.75348806e-01 -5.43335617e-01
-1.80834246e+00 -7.17776120e-01 4.71406356e-02 3.23943384e-02
1.21590823e-01 -2.75696665e-01 -7.42636323e-01 1.78486615e-01
3.96968991e-01 5.46804406e-02 7.09857941e-01 -7.11386144e-01
4.79122922e-02 -5.54953754e-01 -1.34202504e+00 6.52844608e-01
9.78863657e-01 -3.80976975e-01 -1.60277858e-02 -5.78523837e-02
4.46469903e-01 -1.06559479e+00 -8.46758723e-01 3.05710107e-01
7.82041788e-01 -1.58991063e+00 8.95323813e-01 6.46223545e-01
1.01476334e-01 -7.14020610e-01 -3.82775098e-01 -6.82427764e-01
-2.09881276e-01 -8.49420190e-01 2.30034992e-01 9.20048952e-01
1.44565687e-01 -6.16942286e-01 6.01587772e-01 4.53571349e-01
-3.85080159e-01 -5.51628590e-01 -7.84619987e-01 -5.35872936e-01
-5.27645886e-01 -1.68977126e-01 4.42375302e-01 5.90160668e-01
-1.37311518e-01 9.39266384e-02 -6.63032591e-01 2.80317873e-01
5.98286211e-01 2.45316580e-01 6.05823278e-01 -4.89992112e-01
-4.41928983e-01 -1.60179853e-01 -7.26483643e-01 -8.73713970e-01
-9.11301017e-01 -2.92823404e-01 -2.56682098e-01 -1.15376043e+00
9.79771614e-02 -5.95127463e-01 2.22697973e-01 -1.92515790e-01
1.17763110e-01 9.20700669e-01 9.56944898e-02 2.22779438e-01
-1.48911670e-01 -5.17422706e-02 1.17924953e+00 3.47441643e-01
-3.62625480e-01 -3.22893828e-01 -3.23061258e-01 3.90685588e-01
1.04322410e+00 -1.48540765e-01 -6.77542806e-01 -4.67266679e-01
4.56190497e-01 1.30393595e-01 -2.43992836e-04 -1.59664118e+00
4.37175632e-01 -3.70302796e-02 7.74836242e-01 -6.13812804e-01
4.41317469e-01 -1.16960311e+00 9.37111318e-01 7.52807438e-01
9.00467187e-02 7.54327893e-01 1.01003438e-01 1.20974801e-01
-2.35296279e-01 -5.38989790e-02 1.48744321e+00 -5.13662444e-03
-1.20784295e+00 7.47578740e-02 -2.43917674e-01 -2.71012872e-01
1.43667865e+00 -9.84670460e-01 -4.67393130e-01 -1.95255414e-01
-3.17532808e-01 -4.50755656e-01 7.23101318e-01 1.60089061e-01
5.94665945e-01 -1.25070560e+00 -3.49541247e-01 5.32825410e-01
-1.51751041e-01 -5.16731381e-01 6.57980502e-01 6.30874038e-01
-1.58818328e+00 3.37986231e-01 -9.32242930e-01 -7.38425255e-01
-1.61220527e+00 5.09406269e-01 3.46496135e-01 1.94005549e-01
-8.34833682e-01 4.14222986e-01 4.38883975e-02 4.79884535e-01
-7.71249756e-02 -2.08497897e-01 -2.06536576e-02 -4.32195753e-01
7.13839591e-01 5.91136336e-01 2.85493135e-01 -3.17857385e-01
-2.51224816e-01 1.22675681e+00 1.14481091e-01 -3.13899934e-01
8.69496644e-01 -5.96886277e-01 6.68247268e-02 1.81823522e-01
1.12145317e+00 1.73959538e-01 -1.04302645e+00 3.25403303e-01
-4.69544917e-01 -9.15813446e-01 1.25731245e-01 -8.42644155e-01
-1.03878427e+00 7.29035079e-01 1.48666942e+00 1.45279482e-01
1.97808731e+00 -5.35079241e-01 5.90022683e-01 -1.38742030e-01
6.55633926e-01 -1.23058569e+00 2.16750875e-02 1.46498263e-01
6.16998911e-01 -8.85391593e-01 2.80126035e-01 -6.84862614e-01
-5.32014787e-01 1.31132174e+00 7.29923964e-01 -2.44829968e-01
5.32945037e-01 8.81575763e-01 2.65420914e-01 9.61493328e-03
-5.62025607e-01 1.38200745e-01 -7.59840310e-02 8.34094942e-01
6.05432272e-01 -2.76469458e-02 -6.34114921e-01 -1.52873755e-01
-6.65538967e-01 1.66992396e-01 8.71983945e-01 9.84868348e-01
-5.64953446e-01 -9.21170056e-01 -6.37894392e-01 3.86070460e-01
-7.90924072e-01 6.22527190e-02 1.39581650e-01 8.11702251e-01
-2.46530045e-02 6.33244395e-01 8.95484686e-02 -3.96030903e-01
3.41915548e-01 -7.20052958e-01 7.32730508e-01 -7.07282498e-02
-6.60870790e-01 1.18199222e-01 -6.74675927e-02 -5.83717942e-01
-3.28342855e-01 -2.11874425e-01 -1.10241377e+00 -6.32438719e-01
-2.58391304e-03 -1.29966199e-01 1.07410276e+00 5.09633906e-02
2.76349366e-01 6.16428673e-01 6.40124381e-01 -4.58766341e-01
-1.77609339e-01 -5.46295881e-01 -6.40411437e-01 -8.05613026e-03
1.81191072e-01 -2.84128726e-01 -5.65933883e-02 2.82002062e-01] | [10.818747520446777, -2.401249647140503] |
6cc81608-2c70-4a2a-8a28-e52960210758 | kurdish-handwritten-character-recognition | 2210.13734 | null | https://arxiv.org/abs/2210.13734v1 | https://arxiv.org/pdf/2210.13734v1.pdf | Kurdish Handwritten Character Recognition using Deep Learning Techniques | Handwriting recognition is one of the active and challenging areas of research in the field of image processing and pattern recognition. It has many applications that include: a reading aid for visual impairment, automated reading and processing for bank checks, making any handwritten document searchable, and converting them into structural text form, etc. Moreover, high accuracy rates have been recorded by handwriting recognition systems for English, Chinese Arabic, Persian, and many other languages. Yet there is no such system available for offline Kurdish handwriting recognition. In this paper, an attempt is made to design and develop a model that can recognize handwritten characters for Kurdish alphabets using deep learning techniques. Kurdish (Sorani) contains 34 characters and mainly employs an Arabic\Persian based script with modified alphabets. In this work, a Deep Convolutional Neural Network model is employed that has shown exemplary performance in handwriting recognition systems. Then, a comprehensive dataset was created for handwritten Kurdish characters, which contains more than 40 thousand images. The created dataset has been used for training the Deep Convolutional Neural Network model for classification and recognition tasks. In the proposed system, the experimental results show an acceptable recognition level. The testing results reported a 96% accuracy rate, and training accuracy reported a 97% accuracy rate. From the experimental results, it is clear that the proposed deep learning model is performing well and is comparable to the similar model of other languages' handwriting recognition systems. | ['Amit Chhabra', 'S. Vimal', 'Seyedali Mirjalili', 'Nebojsa Bacanin', 'Abeer Alsadoon', 'Polla Fattah', 'Tarik A. Rashid', 'Rebin M. Ahmed'] | 2022-10-18 | null | null | null | null | ['handwriting-recognition'] | ['computer-vision'] | [-2.94485148e-02 -4.31782931e-01 -5.25887311e-03 -2.91950613e-01
8.16374049e-02 -4.29702044e-01 5.83446026e-01 -2.63993919e-01
-4.86267388e-01 6.78823471e-01 -6.15232438e-02 -5.63096702e-01
-1.20249219e-01 -8.68285120e-01 -2.00948775e-01 -7.58462191e-01
3.91938120e-01 5.71487606e-01 2.40328442e-02 -2.66113192e-01
8.09255123e-01 1.15934432e+00 -1.16517878e+00 3.45975071e-01
7.01909184e-01 6.16360724e-01 3.28235954e-01 9.25083160e-01
-4.90410998e-02 1.23193038e+00 -1.05215764e+00 -3.63750845e-01
1.36822253e-01 -2.95106173e-01 -7.91234672e-01 5.62210262e-01
2.86411375e-01 -8.70377839e-01 -8.61338079e-01 7.43933678e-01
7.27239788e-01 3.57222967e-02 8.12059641e-01 -4.29691583e-01
-1.36907530e+00 3.96314174e-01 -5.96579671e-01 1.79214299e-01
-8.01572353e-02 -1.46278307e-01 2.22064435e-01 -9.19577181e-01
4.03795958e-01 9.99579847e-01 5.99966049e-01 3.67007375e-01
-2.16463715e-01 -6.24427497e-01 -5.02043545e-01 2.64906555e-01
-1.31454861e+00 -1.75725203e-02 4.54395533e-01 -4.80519652e-01
1.26138175e+00 9.41873714e-02 4.24603075e-01 6.16488099e-01
4.04701263e-01 9.66234684e-01 1.34572399e+00 -9.99176025e-01
-2.25231245e-01 2.24047720e-01 6.74466372e-01 7.84186780e-01
3.08979094e-01 -2.84233898e-01 -2.87526935e-01 4.70664054e-01
1.08241093e+00 1.36355860e-02 -9.61338077e-03 7.09683001e-01
-8.52068305e-01 6.62882268e-01 2.08717123e-01 8.69172990e-01
-2.43141323e-01 -3.59093428e-01 4.29127902e-01 3.94247353e-01
-5.94979599e-02 1.74083203e-01 -7.80778900e-02 -2.86118180e-01
-8.98706317e-01 5.15369177e-02 6.34573758e-01 1.00638652e+00
2.40792319e-01 5.64616442e-01 -2.25670218e-01 1.33752036e+00
2.58327037e-01 7.26924539e-01 9.30120051e-01 -1.16476648e-01
5.84113777e-01 7.76655138e-01 -5.57354540e-02 -1.24451840e+00
-2.86829799e-01 1.39658917e-02 -1.05711198e+00 3.40008914e-01
4.21688706e-01 -1.44513473e-01 -1.49283862e+00 5.45520663e-01
-4.81138051e-01 -6.11461878e-01 3.42608511e-01 1.04721642e+00
9.79286730e-01 1.03309619e+00 -3.25293183e-01 1.25529364e-01
1.19597590e+00 -1.00721335e+00 -8.95258963e-01 7.89871290e-02
-1.97096039e-02 -1.12572396e+00 9.32457864e-01 8.44439328e-01
-8.08582067e-01 -7.17799425e-01 -1.23014593e+00 -1.75617531e-01
-5.02768397e-01 1.12223840e+00 6.34618163e-01 1.01711202e+00
-7.37866461e-01 3.14995825e-01 -6.85007393e-01 -6.61381960e-01
4.51794297e-01 4.79032576e-01 -4.28787172e-01 -1.12036861e-01
-7.16354191e-01 1.06872272e+00 8.00638914e-01 4.96383965e-01
-7.09616363e-01 4.23336864e-01 -4.20375019e-01 -9.79298428e-02
-4.29033160e-01 5.20622432e-01 9.86557007e-01 -1.15644515e+00
-1.52137089e+00 8.96655381e-01 1.70616403e-01 -1.79057211e-01
5.38351297e-01 8.71149376e-02 -8.52865517e-01 1.46038309e-01
-2.72885203e-01 2.29134597e-02 7.53535032e-01 -6.97690487e-01
-5.22835910e-01 -5.75517893e-01 -3.38025421e-01 1.93009555e-01
-6.04285538e-01 5.49021065e-01 -6.18209302e-01 -9.65125382e-01
9.02267024e-02 -5.75512052e-01 4.05093998e-01 -3.68287444e-01
-2.87381798e-01 -2.61209339e-01 1.13029063e+00 -1.41127491e+00
1.05838656e+00 -2.07559848e+00 -3.98476988e-01 4.89956439e-01
-3.14479381e-01 1.10491538e+00 -3.77380513e-02 5.76297045e-01
1.92395434e-01 -2.12343261e-01 -2.11750448e-01 3.59268785e-01
-2.85753071e-01 3.38747561e-01 -2.28961796e-01 3.42160672e-01
3.20736051e-01 5.96360147e-01 -2.84384221e-01 -2.21721292e-01
2.11965516e-01 5.54768860e-01 3.18912625e-01 2.02308565e-01
2.87416905e-01 -6.92277849e-02 -2.59987891e-01 1.19390392e+00
7.84651458e-01 6.83842078e-02 1.12481594e-01 3.30227911e-02
-2.84119219e-01 -3.62068415e-01 -1.05205607e+00 6.70474172e-01
-2.60683179e-01 1.18319535e+00 -5.04211962e-01 -9.25763309e-01
1.67829132e+00 2.11578414e-01 -2.81365037e-01 -8.08563054e-01
4.29191440e-01 6.63631082e-01 2.24186331e-01 -9.05968785e-01
7.58093655e-01 2.51664996e-01 5.62545061e-01 4.62052733e-01
-6.70205802e-02 1.18096828e-01 3.99225563e-01 -3.28573644e-01
6.08771026e-01 -1.11560822e-01 2.83895284e-02 -4.12619449e-02
8.76129687e-01 2.01988026e-01 6.39594570e-02 6.99626923e-01
3.80105264e-02 7.03625262e-01 1.26703769e-01 -8.86758089e-01
-1.31281412e+00 -5.05679369e-01 -3.57890278e-01 5.13884962e-01
-2.58926332e-01 3.37362617e-01 -7.05987811e-01 -2.30413601e-01
-1.31020799e-01 2.42711678e-01 -3.21728885e-01 2.91531980e-01
-7.69690573e-01 -7.84225285e-01 1.06572056e+00 8.71465504e-01
1.48476112e+00 -1.64682257e+00 -4.79764491e-01 2.35569850e-01
4.45454657e-01 -7.93680191e-01 -1.08161174e-01 -6.02159686e-02
-7.41330981e-01 -1.20376325e+00 -1.46216249e+00 -1.57274020e+00
9.34831858e-01 3.76021825e-02 2.96292365e-01 2.02733636e-01
-5.76613843e-01 -2.80961730e-02 -5.92323124e-01 -6.16564989e-01
-5.78071177e-01 -6.38049049e-03 -2.36218125e-02 4.95100841e-02
7.38884747e-01 1.52804628e-01 -1.14009425e-01 1.18204862e-01
-9.12189484e-01 -1.39248401e-01 1.07748532e+00 1.10108733e+00
1.21551029e-01 3.10754240e-01 6.03436232e-01 -7.39900708e-01
1.14039648e+00 1.01893932e-01 -7.22027183e-01 5.60599625e-01
-5.54065704e-01 -3.81348461e-01 8.50404084e-01 -3.08042854e-01
-1.13344026e+00 -1.96703389e-01 -1.99206948e-01 1.03117026e-01
-2.63467252e-01 6.69307709e-01 9.88540277e-02 -3.88700396e-01
5.24781108e-01 9.15436625e-01 2.08019257e-01 -5.25080919e-01
-4.25118417e-01 1.92764175e+00 5.72567403e-01 -2.77495772e-01
1.97342485e-01 -5.73152490e-02 -3.37766111e-01 -1.44511831e+00
-1.48415184e-02 -3.23116332e-01 -7.39546478e-01 -3.35257769e-01
8.16731572e-01 -5.97462118e-01 -7.13730574e-01 1.80114913e+00
-1.24708438e+00 -1.98557198e-01 3.77975106e-01 5.17680109e-01
-1.53846651e-01 5.70387661e-01 -9.77088749e-01 -9.61186945e-01
-5.59204996e-01 -1.13337004e+00 5.58736384e-01 6.39524221e-01
1.23708211e-01 -8.71576309e-01 -2.21731305e-01 3.93923521e-01
4.41576868e-01 -1.23739637e-01 9.98370171e-01 -7.63046145e-01
-2.97687382e-01 -6.41457915e-01 -6.98126972e-01 9.48512971e-01
5.32436728e-01 2.91568279e-01 -8.29593301e-01 -2.34565958e-01
-3.40734124e-01 -4.60345775e-01 8.15367401e-01 -4.83636418e-03
1.17105758e+00 -3.59230518e-01 2.31574327e-01 2.82753795e-01
1.45054018e+00 1.12640643e+00 1.09225392e+00 6.35966659e-01
6.08322561e-01 1.90573618e-01 3.51679802e-01 4.12494361e-01
-7.17209503e-02 9.43725407e-02 -2.78623194e-01 -6.12999275e-02
-1.84600040e-01 7.86544010e-02 3.38144541e-01 1.03279531e+00
-6.00624621e-01 -2.57877797e-01 -1.31317401e+00 5.11387646e-01
-1.43443608e+00 -9.09571409e-01 -3.01566452e-01 1.81568491e+00
8.66208076e-01 -3.03153276e-01 -6.13917597e-02 6.28557384e-01
7.24016964e-01 -1.68913871e-01 -3.14458966e-01 -9.71005440e-01
-3.94305170e-01 5.54181576e-01 5.86481392e-01 3.89408588e-01
-1.28373730e+00 1.19709396e+00 6.29251480e+00 5.85966349e-01
-1.62407506e+00 -5.22529542e-01 6.22050583e-01 7.66594231e-01
4.72771347e-01 -6.24371111e-01 -9.29002166e-01 5.07517695e-01
5.24020612e-01 2.20588639e-01 3.78812492e-01 4.60663050e-01
1.92757219e-01 -1.59153447e-01 -4.60771292e-01 1.05939066e+00
5.26034355e-01 -1.38702118e+00 6.19373202e-01 -9.56342667e-02
7.97349632e-01 -2.59946883e-01 -1.41979419e-02 5.29858284e-02
-2.68922392e-02 -1.47579420e+00 4.17505980e-01 6.67353570e-01
8.28220427e-01 -9.31368113e-01 1.25582337e+00 3.40153188e-01
-4.68197048e-01 -2.26648271e-01 -6.59760833e-01 -2.07278877e-01
-4.82082278e-01 2.86699273e-02 -1.15014446e+00 2.61750668e-01
6.14400685e-01 6.79553151e-01 -6.82913840e-01 1.09432173e+00
-3.59431319e-02 8.20702314e-01 4.79751565e-02 -6.21534467e-01
5.95148802e-01 -4.06348199e-01 -1.76734477e-02 1.52687538e+00
4.98840958e-01 2.83836782e-01 -2.38140032e-01 4.69402552e-01
3.18279006e-02 4.20235217e-01 -5.98251402e-01 -6.57703280e-01
1.54573575e-01 7.30132699e-01 -8.85962188e-01 -4.11915123e-01
-3.67636591e-01 1.00975549e+00 -4.67759967e-02 3.67454261e-01
-4.48207170e-01 -1.21280301e+00 -1.57053798e-01 -1.70169443e-01
2.19409272e-01 -4.78150129e-01 -4.94527727e-01 -8.51025522e-01
9.84787941e-02 -1.22180319e+00 9.18166041e-02 -8.43341470e-01
-9.70889091e-01 7.79309273e-01 -4.31405663e-01 -9.31450665e-01
-1.73321757e-02 -1.48075712e+00 -6.15963578e-01 1.20813918e+00
-1.20082188e+00 -1.45133030e+00 -4.32361573e-01 5.85214972e-01
9.20062602e-01 -1.39471257e+00 8.40108931e-01 2.72878379e-01
-7.54971623e-01 7.81827569e-01 6.41305089e-01 9.77507293e-01
5.35018623e-01 -1.04278886e+00 1.54891750e-02 1.08258748e+00
-8.41409415e-02 6.35655880e-01 7.10602105e-02 -6.52246952e-01
-1.37853801e+00 -9.41874027e-01 9.48907852e-01 9.77063626e-02
2.55471289e-01 2.40723059e-01 -8.14403415e-01 6.40161693e-01
4.60920811e-01 -4.49165255e-01 4.34130669e-01 -5.17576694e-01
-9.38106179e-02 -1.12065770e-01 -1.16571712e+00 3.97677362e-01
2.01763660e-02 -2.91523427e-01 -4.70821291e-01 3.59034002e-01
-3.37610424e-01 -3.80923390e-01 -6.78903580e-01 -5.12298308e-02
7.92224824e-01 -7.35871911e-01 3.68770927e-01 -5.96939623e-01
6.26662672e-01 -3.14859182e-01 -9.70381796e-02 -8.05007041e-01
-3.39456350e-01 -1.58722177e-01 1.08362235e-01 1.27659237e+00
3.57581437e-01 -5.77387333e-01 7.24336863e-01 3.96957666e-01
-5.19008823e-02 -5.78252196e-01 -3.80957961e-01 -7.91841805e-01
2.20905095e-01 6.84469566e-02 3.41240495e-01 1.00157034e+00
-4.39169765e-01 -4.63637412e-02 -4.03654993e-01 7.01645911e-02
1.64159939e-01 -1.20485410e-01 5.40933371e-01 -7.89634764e-01
-5.23545817e-02 -4.48359430e-01 -8.18990409e-01 -7.66170025e-01
-1.41433433e-01 -7.18698680e-01 -2.57496417e-01 -1.84162116e+00
8.83488059e-02 -4.19868201e-01 8.97808373e-02 8.76169324e-01
2.95788258e-01 4.24847156e-01 1.83362946e-01 3.26740146e-01
3.00700128e-01 5.12106866e-02 1.29892075e+00 -6.95916831e-01
-1.31996438e-01 1.99357957e-01 -3.12746435e-01 4.97364312e-01
1.04596150e+00 2.92011738e-01 -1.83849797e-01 -9.08452570e-01
-2.89363384e-01 -3.34249675e-01 -8.28698054e-02 -9.73295689e-01
2.47683480e-01 -7.82410130e-02 9.54534650e-01 -6.97145224e-01
-2.21863445e-02 -6.88377142e-01 -1.95079729e-01 5.28546333e-01
-2.00695217e-01 4.67964225e-02 2.71556586e-01 -1.49167359e-01
-5.05644202e-01 -6.95872843e-01 8.80571485e-01 -1.43669173e-01
-1.17969429e+00 -3.93996909e-02 -8.88972759e-01 -6.98606253e-01
1.04968941e+00 -7.81314731e-01 -3.44806522e-01 -1.49127901e-01
-4.46706027e-01 -2.48440742e-01 1.81533787e-02 5.22640586e-01
1.13084269e+00 -1.21978343e+00 -9.31273103e-01 5.38069963e-01
-7.04500750e-02 -2.98403889e-01 -9.03717130e-02 4.15251911e-01
-1.87353623e+00 5.96788049e-01 -9.09334719e-01 -1.91129193e-01
-1.51368856e+00 -4.52158041e-02 3.50516558e-01 1.99527040e-01
-5.93307257e-01 5.64936161e-01 -7.23225832e-01 -3.85189533e-01
4.88970041e-01 -1.37812585e-01 -6.64174855e-01 -1.17295235e-01
8.62523615e-01 5.76442420e-01 3.99561226e-01 -8.06072652e-01
-2.33576477e-01 6.94475651e-01 -4.79705095e-01 9.96811241e-02
1.32332194e+00 5.10309577e-01 -5.15060961e-01 2.00839743e-01
8.33475411e-01 -6.26265332e-02 -6.50491416e-01 1.93986744e-01
-3.22754756e-02 -5.34519196e-01 -1.26144633e-01 -1.32762671e+00
-1.10368013e+00 1.14178705e+00 9.12847400e-01 -5.50311878e-02
1.19033813e+00 -7.96633542e-01 6.32338881e-01 1.13973892e+00
3.58703405e-01 -1.55586708e+00 -1.72196686e-01 1.05292058e+00
1.14862859e+00 -1.18551672e+00 -1.32620052e-01 1.87405646e-01
-7.00094700e-01 2.02569985e+00 6.91772699e-01 -2.90332049e-01
4.60300416e-01 3.16390783e-01 4.84411448e-01 5.72065189e-02
3.44051309e-02 2.16857955e-01 9.69119594e-02 7.40176737e-01
8.58122706e-01 1.08356774e-01 -6.94836497e-01 5.77698052e-01
-8.57539177e-02 2.96017557e-01 9.48717058e-01 1.22059953e+00
-6.16513968e-01 -1.07883227e+00 -6.76628649e-01 5.91430962e-01
-6.08055353e-01 9.20151696e-02 -7.83941507e-01 8.64741802e-01
-1.06166005e-01 7.74114311e-01 1.03014186e-01 -2.50891685e-01
1.84467137e-01 -5.50827421e-02 5.61566234e-01 -2.27261692e-01
-5.19815803e-01 -2.87329733e-01 -7.00612068e-02 3.92758936e-01
-3.31025660e-01 -1.88718051e-01 -1.21634209e+00 -4.33231086e-01
-2.24344522e-01 -5.22033200e-02 7.71790087e-01 8.77470195e-01
-1.45251215e-01 2.27889299e-01 5.64364791e-01 -4.36051756e-01
-5.26839316e-01 -1.23416603e+00 -8.64917338e-01 7.06646740e-02
1.16989799e-01 -1.32579491e-01 4.01068568e-01 3.30067605e-01] | [11.843999862670898, 2.64717173576355] |
190c8dea-fc86-4f67-9e4a-8df816b1e321 | fast-shadow-detection-from-a-single-image | 1709.09283 | null | http://arxiv.org/abs/1709.09283v2 | http://arxiv.org/pdf/1709.09283v2.pdf | Fast Shadow Detection from a Single Image Using a Patched Convolutional Neural Network | In recent years, various shadow detection methods from a single image have
been proposed and used in vision systems; however, most of them are not
appropriate for the robotic applications due to the expensive time complexity.
This paper introduces a fast shadow detection method using a deep learning
framework, with a time cost that is appropriate for robotic applications. In
our solution, we first obtain a shadow prior map with the help of multi-class
support vector machine using statistical features. Then, we use a semantic-
aware patch-level Convolutional Neural Network that efficiently trains on
shadow examples by combining the original image and the shadow prior map.
Experiments on benchmark datasets demonstrate the proposed method significantly
decreases the time complexity of shadow detection, by one or two orders of
magnitude compared with state-of-the-art methods, without losing accuracy. | ['Sepideh Hosseinzadeh', 'Moein Shakeri', 'Hong Zhang'] | 2017-09-26 | null | null | null | null | ['shadow-detection'] | ['computer-vision'] | [ 5.49564898e-01 -3.95303816e-02 3.47525150e-01 -4.30093825e-01
-2.78052628e-01 -1.01071015e-01 4.54664528e-01 -1.36589691e-01
-5.25196970e-01 7.60876834e-01 -4.30923402e-01 -3.96251440e-01
3.55329573e-01 -6.20224774e-01 -8.03393066e-01 -8.97663653e-01
4.08131719e-01 2.87788987e-01 1.16446984e+00 8.49583596e-02
2.18269885e-01 3.47569227e-01 -1.60044801e+00 -5.70651516e-02
1.18148732e+00 1.24969757e+00 9.11283910e-01 2.36542970e-01
-4.61414084e-02 4.99320716e-01 -6.22771919e-01 6.35634549e-03
3.11927319e-01 -1.90458849e-01 7.11191967e-02 1.35157675e-01
4.10066158e-01 -7.73836970e-01 -4.90549952e-01 7.67717898e-01
3.65336955e-01 -8.00202489e-02 4.12818164e-01 -1.19357276e+00
-2.57750511e-01 -1.07486278e-01 -5.99224627e-01 -1.45422667e-01
-7.10359514e-02 -1.35027379e-01 6.21304810e-01 -1.08759427e+00
2.09551916e-01 1.04485655e+00 8.07720244e-01 -3.10660172e-02
-7.82978237e-01 -6.73782706e-01 9.51609164e-02 5.85597456e-01
-1.20845687e+00 -5.49495891e-02 9.92391527e-01 -2.39364784e-02
5.09652019e-01 -9.85736698e-02 6.75937057e-01 9.02219415e-01
3.17960262e-01 1.15142131e+00 1.56894946e+00 -2.41178676e-01
2.13650405e-01 1.65502414e-01 -5.53844729e-03 1.18133152e+00
4.76054400e-01 4.03053723e-02 -4.57466394e-01 -8.00267532e-02
6.26480937e-01 2.08788320e-01 -4.61778611e-01 -7.06007481e-01
-9.02801991e-01 7.32408285e-01 6.17935121e-01 4.65564765e-02
-2.79568672e-01 2.75291502e-01 7.07391351e-02 -2.45050356e-01
4.06458437e-01 -1.39658868e-01 -3.46468985e-01 3.13566089e-01
-9.75179613e-01 1.06042594e-01 7.77486920e-01 8.00795555e-01
9.85628128e-01 1.14283562e-02 -5.02884947e-02 5.34976184e-01
2.91983902e-01 9.96078849e-01 1.32029429e-01 -6.08230174e-01
-3.92026268e-02 3.88908893e-01 4.58309323e-01 -1.01706326e+00
-4.97842580e-01 -4.53058451e-01 -5.78468680e-01 6.16840601e-01
3.92787665e-01 -1.01457387e-02 -1.04349875e+00 1.21708548e+00
4.14114863e-01 5.74543417e-01 2.40553588e-01 9.43266988e-01
7.37072408e-01 6.53642893e-01 -5.13299465e-01 -1.64060712e-01
1.30153966e+00 -1.27055204e+00 -6.70588434e-01 -6.71462357e-01
2.02040561e-02 -1.02476394e+00 1.13373148e+00 5.07585704e-01
-4.04989511e-01 -4.04970944e-01 -1.53200984e+00 1.25552371e-01
-3.19692492e-01 6.36386335e-01 8.03877354e-01 6.61488593e-01
-6.42999172e-01 4.48519111e-01 -7.40416527e-01 -2.81938195e-01
6.01505756e-01 1.80991650e-01 3.95758860e-02 -2.09837854e-01
-7.83133388e-01 9.11888897e-01 2.67511278e-01 3.24940175e-01
-1.19388604e+00 -4.43089038e-01 -7.44313955e-01 1.40575498e-01
6.97630584e-01 -2.32625887e-01 1.15476608e+00 -9.35663462e-01
-1.75723040e+00 2.18262076e-01 -2.49595031e-01 -5.15898228e-01
6.95410073e-01 -7.11738110e-01 4.56420705e-02 8.86735246e-02
-4.60605025e-02 2.39697769e-01 1.05505264e+00 -1.58919048e+00
-6.69508219e-01 -1.48228958e-01 2.29211599e-01 2.07583696e-01
-3.60007018e-01 -3.41686159e-01 -6.34319007e-01 -3.59925985e-01
1.34941667e-01 -1.17748380e+00 -1.06533036e-01 4.71839964e-01
-4.11099464e-01 -2.19449941e-02 1.56087470e+00 -5.92189848e-01
5.42314589e-01 -2.02259159e+00 -2.51943856e-01 1.71377994e-02
-4.12363233e-03 5.17355621e-01 2.04545200e-01 1.99447215e-01
7.66011894e-01 -7.82027245e-01 -7.43902862e-01 -5.30792117e-01
-1.07803099e-01 3.23105365e-01 -4.77930009e-01 6.64360821e-01
-1.64309554e-02 6.64828598e-01 -7.82488227e-01 -6.40062213e-01
5.47476470e-01 5.62940121e-01 8.08449835e-03 4.52263623e-01
-3.28100055e-01 1.40689850e-01 -4.52685744e-01 7.26873517e-01
1.06342411e+00 -1.70611635e-01 1.14350073e-01 -2.44798541e-01
-2.30374977e-01 -4.31689993e-02 -9.08346653e-01 1.58634853e+00
-6.19698822e-01 9.86501217e-01 6.35825992e-02 -8.57086480e-01
9.25127923e-01 -1.63202927e-01 2.02295557e-01 -5.02910078e-01
1.69823334e-01 5.08837223e-01 -1.57983035e-01 -2.69543886e-01
3.72745097e-01 1.42811835e-02 7.95322210e-02 2.13869795e-01
-3.88663381e-01 -1.65187255e-01 -3.43835801e-01 -1.56775583e-02
1.10665607e+00 3.28114182e-01 4.90100458e-02 -1.62438259e-01
5.04970014e-01 2.09426582e-02 6.43502891e-01 6.59117937e-01
-4.70949747e-02 5.56580663e-01 1.09403752e-01 -3.14663142e-01
-7.16991246e-01 -9.76366699e-01 -1.85276002e-01 8.14956963e-01
8.10244799e-01 6.11715801e-02 -7.18500137e-01 -7.09734619e-01
2.55400091e-01 4.67398882e-01 -5.33245027e-01 1.24638736e-01
-6.06533468e-01 -5.82960546e-01 3.65945905e-01 5.65805614e-01
1.05728054e+00 -9.98820901e-01 -1.47004044e+00 9.91229042e-02
-1.57811772e-02 -1.57284236e+00 -1.44004419e-01 2.74925202e-01
-6.52597547e-01 -1.03285944e+00 -9.65106070e-01 -7.29595363e-01
6.00675523e-01 1.04563379e+00 5.65574288e-01 2.83827603e-01
-6.08046591e-01 3.69280786e-03 -4.12069827e-01 -8.06950510e-01
8.34293813e-02 -2.96028227e-01 -8.27937275e-02 2.75157213e-01
-1.31959915e-01 -5.01027584e-01 -8.76432657e-01 1.65995985e-01
-8.05155277e-01 3.78405780e-01 1.34286642e+00 9.29777741e-01
3.05239260e-01 -4.62651029e-02 1.74245551e-01 -8.91219735e-01
8.32389947e-03 -9.46535394e-02 -1.09931195e+00 2.80930519e-01
-8.13240588e-01 4.64663506e-02 6.67002499e-01 -1.26993492e-01
-1.42683983e+00 4.59870130e-01 4.06173229e-01 -5.20865560e-01
3.29793245e-03 1.71702191e-01 -1.64297476e-01 -5.76198339e-01
3.77758086e-01 4.97909874e-01 -6.46764711e-02 -3.55027378e-01
2.70207018e-01 6.83439791e-01 6.68800175e-01 -3.03589284e-01
1.03320670e+00 1.01741052e+00 1.86497852e-01 -8.56978834e-01
-1.02563441e+00 -4.82455820e-01 -6.47948682e-01 -3.49916965e-01
6.14403784e-01 -7.55374312e-01 -3.97160292e-01 8.91767323e-01
-1.19934773e+00 -6.09034061e-01 4.09468949e-01 2.56965309e-01
-4.00199950e-01 8.37527335e-01 -1.34337157e-01 -8.95953715e-01
-3.85646850e-01 -1.11791146e+00 1.26347995e+00 5.26634514e-01
6.31740034e-01 -5.75161994e-01 -2.67262369e-01 2.58351892e-01
3.52252007e-01 3.90141368e-01 7.38629639e-01 -1.54200301e-01
-1.15855348e+00 -8.31774175e-02 -7.76975930e-01 3.12634796e-01
1.84629023e-01 -4.47029740e-01 -1.25296092e+00 -1.50629640e-01
-1.87322125e-01 -3.22104961e-01 1.07102001e+00 2.78747946e-01
1.24255562e+00 4.06243317e-02 -6.36819541e-01 5.46305835e-01
1.67598319e+00 1.20307393e-01 4.23859209e-01 2.54567802e-01
7.24367142e-01 2.66522646e-01 1.11406279e+00 5.62321782e-01
2.54445195e-01 5.93996048e-01 6.57514453e-01 -2.87347168e-01
-2.61127353e-01 1.24237258e-02 2.86895245e-01 2.14039460e-01
2.48312280e-02 -1.62476406e-01 -7.83728898e-01 5.54648995e-01
-2.19228625e+00 -4.24103737e-01 -1.18391752e-01 2.09506559e+00
3.64449114e-01 3.81175667e-01 -3.36414397e-01 8.98165454e-04
3.73067170e-01 3.53047490e-01 -8.41887116e-01 1.34676859e-01
-8.67879465e-02 1.67537242e-01 8.72474015e-01 4.48413014e-01
-1.06473064e+00 1.31925523e+00 5.27776575e+00 7.52263606e-01
-1.13135791e+00 1.44282699e-01 1.15354329e-01 3.92269313e-01
-1.81148946e-01 1.93627492e-01 -5.37252903e-01 3.15601438e-01
4.26809579e-01 2.36144096e-01 3.76149774e-01 1.10249507e+00
-8.61067791e-03 -7.04567134e-01 -7.59489536e-01 1.06350255e+00
4.18103158e-01 -1.15180540e+00 -5.07061183e-01 -7.94451609e-02
6.53249681e-01 1.81464791e-01 -1.30559221e-01 1.89136773e-01
5.33943512e-02 -6.23223484e-01 8.49507809e-01 3.83215547e-01
4.95450020e-01 -5.40506184e-01 9.90363955e-01 3.75724703e-01
-1.28199923e+00 -1.19949199e-01 -6.16426468e-01 -8.85401219e-02
6.70260265e-02 8.51998985e-01 -1.07952952e+00 4.64510411e-01
9.53585565e-01 3.37053001e-01 -4.22936320e-01 1.11128557e+00
-4.51647907e-01 3.94925863e-01 -5.64936042e-01 -4.21895623e-01
3.24905545e-01 -1.13383859e-01 2.47562170e-01 1.23295093e+00
3.86569768e-01 1.52226508e-01 4.96986210e-01 5.56124330e-01
1.44359306e-01 -7.34210089e-02 -5.52913070e-01 2.56677419e-01
4.58071470e-01 1.47257245e+00 -1.12609863e+00 -4.54088330e-01
-5.23220956e-01 1.48187864e+00 2.17622787e-01 2.06418216e-01
-1.12044668e+00 -7.58480847e-01 2.69539148e-01 -1.21272936e-01
4.76926774e-01 -4.64074314e-01 -4.60785657e-01 -7.30889618e-01
2.52347142e-01 -3.18808854e-01 -3.35099876e-01 -9.17225003e-01
-6.47513211e-01 3.73892039e-01 -1.34319678e-01 -1.11485815e+00
1.26287714e-01 -8.64091873e-01 -6.65327668e-01 5.80095708e-01
-2.01572990e+00 -1.28596938e+00 -1.10953999e+00 5.14153302e-01
7.02273309e-01 -9.84471738e-02 6.37033165e-01 6.15398549e-02
-3.22353125e-01 3.04962814e-01 3.28943759e-01 -8.12910721e-02
8.27215612e-01 -1.25315607e+00 3.04017067e-01 8.01046371e-01
1.18480511e-02 8.43224749e-02 8.52531731e-01 -5.79914451e-01
-1.37189996e+00 -1.00990844e+00 4.97650892e-01 3.79819125e-02
4.87751156e-01 -4.43551093e-01 -8.65074635e-01 1.96933180e-01
2.19523087e-01 1.93397105e-01 1.41938686e-01 -1.97946519e-01
-2.89136529e-01 -4.77344275e-01 -1.02134907e+00 5.45663834e-01
9.76264119e-01 -3.71444613e-01 -4.26849097e-01 5.76446056e-01
6.66573584e-01 -5.62960804e-01 -1.53985456e-01 7.06891954e-01
8.01682651e-01 -1.23567581e+00 7.82201767e-01 4.27044421e-01
3.51054631e-02 -6.13532305e-01 -2.25878298e-01 -1.08182323e+00
-3.09380777e-02 -3.24673295e-01 -2.38366261e-01 8.01398098e-01
2.95821037e-02 -6.98039472e-01 9.72271740e-01 5.90378158e-02
-5.23076355e-01 -1.00506485e+00 -8.51360738e-01 -7.79225826e-01
-6.17551625e-01 -1.91940561e-01 2.95984000e-01 2.23406702e-01
-6.15550458e-01 1.15940399e-01 -5.76306701e-01 6.29304528e-01
8.61766577e-01 8.64850342e-01 1.05090725e+00 -1.29658353e+00
-3.79517347e-01 -9.42197144e-02 -2.84852803e-01 -1.17674768e+00
2.34614074e-01 -2.50264287e-01 8.94163370e-01 -1.91755378e+00
4.10206884e-01 -5.97891450e-01 -3.67521763e-01 4.96769845e-01
-2.35697493e-01 3.92118186e-01 1.87309012e-01 2.53369421e-01
-5.57759106e-01 8.34962606e-01 8.66719186e-01 -1.31773904e-01
4.45339531e-02 1.77986417e-02 -9.87462923e-02 1.02090251e+00
6.63416028e-01 -4.13994491e-01 -5.31702876e-01 -3.84843111e-01
-4.53517199e-01 -1.60193101e-01 6.70086801e-01 -1.50033414e+00
1.86855957e-01 -2.33024836e-01 5.28683901e-01 -9.96689618e-01
9.97937918e-01 -9.17724252e-01 -3.55602652e-01 7.66566157e-01
3.82125169e-01 -4.66913730e-01 1.58016741e-01 9.85918403e-01
1.57125682e-01 -8.58530998e-02 9.21834767e-01 5.68902940e-02
-1.03010166e+00 1.44465044e-01 -2.75656790e-01 -5.71100116e-01
1.15939903e+00 -5.05643189e-02 -2.64986038e-01 -2.43497878e-01
1.37296245e-01 -2.08731387e-02 4.53663617e-01 2.60765195e-01
1.00607431e+00 -9.62973595e-01 -2.60540515e-01 3.76529917e-02
2.52516687e-01 5.06629013e-02 5.84044643e-02 7.56897748e-01
-6.86997712e-01 4.84782904e-01 -2.54176080e-01 -8.14117074e-01
-1.35003817e+00 3.46566379e-01 1.76322013e-01 7.73859248e-02
-9.57169533e-01 7.70967841e-01 5.51245809e-01 -1.84692174e-01
4.71443504e-01 -4.39373523e-01 2.14719579e-01 -4.90840763e-01
2.60834932e-01 1.24797419e-01 -1.65150780e-02 -3.58414501e-01
-4.84896243e-01 6.85158908e-01 2.50190079e-01 -9.14630443e-02
1.15291429e+00 1.21708930e-01 1.22719966e-01 4.83753026e-01
8.92829120e-01 -1.55816376e-01 -1.82384562e+00 -4.28195059e-01
-1.93657596e-02 -7.07559109e-01 2.17817992e-01 -6.51052356e-01
-9.05851007e-01 1.00876176e+00 8.15033793e-01 -1.68768719e-01
1.12595999e+00 -1.30011424e-01 1.19313741e+00 8.44914019e-01
7.49161422e-01 -1.18564272e+00 2.91569084e-01 3.20829988e-01
7.70265937e-01 -1.53939521e+00 2.62463450e-01 -6.74898505e-01
-3.80905867e-01 1.16093409e+00 5.99506497e-01 -4.36242431e-01
7.58000910e-01 3.25839907e-01 1.24103360e-01 9.95381922e-02
-1.75909519e-01 -3.25526476e-01 7.83839300e-02 5.77005386e-01
-1.59241900e-01 9.44061130e-02 -4.30179656e-01 3.45984936e-01
2.18019888e-01 -1.29501447e-02 4.99191761e-01 9.27261889e-01
-1.06598854e+00 -1.00079024e+00 -4.23705488e-01 2.35363707e-01
3.31174303e-03 -4.60660942e-02 -4.53130931e-01 7.96188653e-01
2.64632553e-01 9.92499173e-01 -2.58577228e-01 -3.76235902e-01
1.81582328e-02 -2.03616202e-01 5.18296480e-01 -4.27948833e-01
2.52394170e-01 3.95193733e-02 2.58016847e-02 -5.89310706e-01
-4.16877925e-01 -5.34523547e-01 -1.51471543e+00 5.99753112e-02
-3.62430006e-01 -1.49552412e-02 9.94026840e-01 1.20090103e+00
1.38703078e-01 4.91758972e-01 6.13453925e-01 -1.23612523e+00
-3.53044540e-01 -9.76488650e-01 -4.37767506e-01 -1.48300290e-01
4.30296063e-01 -1.24554789e+00 -1.75092652e-01 -2.83468962e-01] | [10.850624084472656, -4.114041805267334] |
1cdf0d26-636e-4db4-8ecc-f6298ca7542b | ascertaining-price-formation-in | 2003.00803 | null | https://arxiv.org/abs/2003.00803v1 | https://arxiv.org/pdf/2003.00803v1.pdf | Ascertaining price formation in cryptocurrency markets with DeepLearning | The cryptocurrency market is amongst the fastest-growing of all the financial markets in the world. Unlike traditional markets, such as equities, foreign exchange and commodities, cryptocurrency market is considered to have larger volatility and illiquidity. This paper is inspired by the recent success of using deep learning for stock market prediction. In this work, we analyze and present the characteristics of the cryptocurrency market in a high-frequency setting. In particular, we applied a deep learning approach to predict the direction of the mid-price changes on the upcoming tick. We monitored live tick-level data from $8$ cryptocurrency pairs and applied both statistical and machine learning techniques to provide a live prediction. We reveal that promising results are possible for cryptocurrencies, and in particular, we achieve a consistent $78\%$ accuracy on the prediction of the mid-price movement on live exchange rate of Bitcoins vs US dollars. | ['Fan Wu', 'Michail Basios', 'Waichung Chung', 'Leslie Kanthan', 'Carmine Ventre', 'Lingbo Li', 'Fan Fang'] | 2020-02-09 | null | null | null | null | ['stock-market-prediction'] | ['time-series'] | [-1.07155716e+00 -2.52355903e-01 2.81060878e-02 -2.10658014e-01
-6.36281908e-01 -7.38689780e-01 8.94933403e-01 1.97336450e-02
-4.70577955e-01 8.36808920e-01 1.48902640e-01 -6.89960420e-01
-1.08960345e-01 -1.07349861e+00 -5.44441938e-01 -6.35741234e-01
-6.36489689e-01 6.29221618e-01 -2.07815275e-01 -4.84080523e-01
5.61103761e-01 3.14173371e-01 -9.64593470e-01 3.03190559e-01
1.23759858e-01 1.71794116e+00 -4.95026618e-01 2.08734900e-01
-6.19032122e-02 1.13082421e+00 -4.66655016e-01 -7.38322914e-01
9.53896284e-01 -1.66481405e-01 -3.50553513e-01 -5.86828887e-01
1.18107283e-02 -6.12446249e-01 -4.22785461e-01 9.48147595e-01
2.63206869e-01 -3.82158220e-01 3.57329220e-01 -8.69512200e-01
-2.46155769e-01 1.34964430e+00 -6.61779523e-01 8.40719938e-01
-1.56338930e-01 1.33566543e-01 1.73490870e+00 -7.80673742e-01
3.49986285e-01 6.96873486e-01 6.97146297e-01 -1.05340227e-01
-1.03039420e+00 -1.02744401e+00 -8.57879892e-02 1.77782416e-01
-7.93889344e-01 -5.03870808e-02 8.62050116e-01 -6.09197199e-01
1.07147765e+00 4.42838632e-02 1.01037157e+00 8.02847385e-01
8.69031906e-01 7.50081301e-01 1.34149718e+00 -1.44332051e-01
1.79095030e-01 -6.17066212e-02 1.64193928e-01 -1.72086060e-01
4.21429902e-01 7.18831420e-01 -6.53808415e-01 -3.54568243e-01
6.51137471e-01 9.14027914e-02 1.31278411e-01 5.23122475e-02
-1.30952871e+00 1.23518121e+00 3.69415194e-01 5.28696358e-01
-6.33010566e-01 3.00958365e-01 6.44713879e-01 9.61032987e-01
8.36733580e-01 4.22611713e-01 -1.09112644e+00 -4.89908189e-01
-1.05958462e+00 9.73389566e-01 1.28033984e+00 1.52180478e-01
1.51973352e-01 4.08165246e-01 2.76095152e-01 1.42989993e-01
2.28228629e-01 4.61567551e-01 9.18512464e-01 -7.05662549e-01
5.56532145e-01 5.40148243e-02 1.66949674e-01 -6.99045181e-01
-4.29984093e-01 -5.20722508e-01 -5.32546997e-01 4.90724027e-01
7.63558865e-01 -5.83367407e-01 -1.66198477e-01 1.16156602e+00
-3.23594570e-01 7.08895251e-02 -5.66109456e-03 6.32651150e-01
2.25372463e-02 5.68186402e-01 -4.73752499e-01 -3.61701995e-01
1.18908238e+00 -4.03899372e-01 -5.89551210e-01 2.27200985e-01
3.77460122e-01 -7.25391805e-01 3.06212962e-01 6.20827019e-01
-1.28710198e+00 -4.09795977e-02 -8.67961347e-01 5.86519778e-01
-4.00853097e-01 -7.73931146e-01 8.05736899e-01 5.55173695e-01
-8.31104159e-01 1.07429540e+00 -9.16977108e-01 5.29145420e-01
3.74259893e-03 4.07501310e-01 6.76840022e-02 9.27149951e-01
-1.47680032e+00 9.40951526e-01 4.57013339e-01 1.67527385e-02
-6.51643455e-01 -9.68429267e-01 -5.30135930e-01 2.80447006e-01
-1.40671194e-01 1.05309471e-01 1.40448332e+00 -8.58703017e-01
-1.40389168e+00 6.42456770e-01 6.87716544e-01 -1.55274296e+00
8.14884841e-01 -1.23741083e-01 -5.67436337e-01 -3.50804120e-01
-2.61694372e-01 -1.34051563e-02 4.86803323e-01 -3.48092020e-01
-9.75996315e-01 -3.46023589e-01 -2.12841526e-01 -3.26022387e-01
3.31386477e-02 3.36715132e-02 6.66379333e-01 -1.18008268e+00
1.40643209e-01 -7.02044129e-01 -1.29382193e-01 -7.52020597e-01
7.83735290e-02 -2.22888544e-01 4.02780950e-01 -9.17677820e-01
1.16371918e+00 -1.92129302e+00 -4.76447970e-01 4.56269205e-01
1.72047272e-01 -2.35427722e-01 3.62809688e-01 5.55608809e-01
-5.47843814e-01 6.06558472e-02 1.64179593e-01 1.11598760e-01
3.74052644e-01 -2.61128932e-01 -9.44976211e-01 6.37987137e-01
1.08470239e-01 1.09087968e+00 -5.78158081e-01 2.64741302e-01
-7.70676136e-02 5.18557578e-02 -4.77574170e-01 3.98178957e-02
-4.15281564e-01 2.40650792e-02 -2.05022648e-01 7.15644121e-01
7.29706407e-01 -4.60015796e-02 1.37025654e-01 1.05909616e-01
-4.95319992e-01 8.68611395e-01 -8.11988473e-01 8.56491268e-01
-1.33339167e-01 7.85489976e-01 -2.62255520e-01 -1.02056098e+00
1.03552377e+00 3.98389667e-01 8.74877155e-01 -1.31228590e+00
3.24073493e-01 8.70633066e-01 4.52707499e-01 1.95316941e-01
6.26138806e-01 -6.49040639e-01 -1.47973269e-01 1.08575714e+00
-3.39994848e-01 -1.69223741e-01 3.15654606e-01 -2.54079968e-01
6.19362354e-01 5.60049899e-03 2.16046333e-01 -8.39242041e-01
1.02304086e-01 -2.14968815e-01 6.31695092e-01 3.43104750e-01
-2.44383022e-01 3.67617577e-01 8.86541545e-01 -1.29362571e+00
-1.16199899e+00 -1.04317486e+00 -5.04077673e-01 6.04739189e-01
-5.17413795e-01 4.75093760e-02 -4.96942371e-01 -4.04770166e-01
5.72969973e-01 6.76170468e-01 -4.88892674e-01 3.61661017e-01
-6.71094298e-01 -1.27264035e+00 8.07620510e-02 3.66222471e-01
4.42435950e-01 -1.28284109e+00 -7.72835433e-01 5.49718797e-01
2.28664547e-01 -6.12028956e-01 -3.93883646e-01 5.36617815e-01
-9.31909680e-01 -1.03027987e+00 -7.37920344e-01 -3.80286723e-01
-3.03437382e-01 -4.30435270e-01 1.45894742e+00 -3.97956818e-01
9.46179852e-02 -2.41393045e-01 -7.90037587e-02 -7.46823251e-01
-1.94484159e-01 2.50499807e-02 2.06769556e-01 1.17934167e-01
8.29175353e-01 -5.74601769e-01 -7.74763882e-01 -1.52853295e-01
-7.16827214e-01 -4.83948141e-01 2.21605673e-01 7.58592308e-01
2.73512006e-01 1.31657258e-01 7.73940265e-01 -6.12977326e-01
7.27333188e-01 -7.27481961e-01 -1.36462796e+00 -2.26087272e-01
-1.19107437e+00 1.24790326e-01 3.28962982e-01 -2.08492637e-01
-5.94216883e-01 -5.32287002e-01 9.31920558e-02 -1.31576881e-01
6.56954587e-01 9.60643351e-01 6.86421752e-01 1.95307791e-01
1.34105831e-01 2.16482013e-01 -1.10973112e-01 -7.58015096e-01
-7.23510981e-02 3.85552883e-01 3.23038906e-01 -4.05919433e-01
7.43666589e-01 4.21993315e-01 -3.52631509e-01 -2.74105221e-01
-1.53671741e-01 -7.34166279e-02 -2.87054718e-01 9.32342559e-02
4.71086353e-01 -1.08174515e+00 -1.13336277e+00 1.01145017e+00
-1.01027560e+00 -2.28724882e-01 -4.78877187e-01 7.86374629e-01
-5.34268141e-01 -1.65629666e-02 -1.26791918e+00 -1.08065975e+00
-5.62601209e-01 -8.93993437e-01 3.27445269e-01 -3.63520198e-02
-2.24821270e-01 -1.23470151e+00 7.84434080e-01 1.68726463e-02
6.96410239e-01 5.54057300e-01 7.98903286e-01 -1.24217880e+00
-6.73125982e-01 -3.27462167e-01 -1.64491430e-01 3.82186592e-01
1.05305314e-01 -9.85594168e-02 -8.10323417e-01 -5.39023757e-01
6.43641174e-01 -1.70417741e-01 8.70016575e-01 6.77822471e-01
3.66111785e-01 -4.47664887e-01 4.91152048e-01 5.59142828e-01
1.46926653e+00 6.33330286e-01 3.99774700e-01 1.02449369e+00
-1.58064201e-01 5.17226398e-01 4.15762603e-01 9.46724594e-01
3.72880876e-01 1.94654822e-01 5.61252475e-01 1.60289213e-01
6.40163422e-01 -6.44639060e-02 6.57989025e-01 8.60647440e-01
-1.95661500e-01 3.77317458e-01 -9.85131860e-01 6.27796769e-01
-1.40846324e+00 -1.03812480e+00 1.45953253e-01 2.19605541e+00
7.65018702e-01 5.53301096e-01 3.61323804e-01 1.46083966e-01
2.36570075e-01 5.36794126e-01 -5.54651320e-01 -3.50353450e-01
-4.14181203e-01 7.03585148e-01 9.58636940e-01 2.91976333e-01
-1.07775831e+00 5.62690616e-01 6.87447739e+00 3.47314984e-01
-1.47941554e+00 -1.93676934e-01 1.28438830e+00 -5.08955479e-01
-4.77641433e-01 -7.21646100e-02 -5.62608957e-01 8.38139892e-01
1.25818884e+00 -7.58661807e-01 4.95157659e-01 7.81575084e-01
2.48458400e-01 2.21928000e-01 -1.11315858e+00 9.27986979e-01
-4.54316556e-01 -1.81038189e+00 -1.99160784e-01 5.19995987e-01
7.60857642e-01 7.24511862e-01 4.67100084e-01 2.40583256e-01
3.67602855e-01 -8.47751200e-01 1.05766571e+00 5.76086998e-01
5.46546429e-02 -9.41568673e-01 1.15696800e+00 -8.85656178e-02
-9.12849069e-01 -2.88132846e-01 -2.44670868e-01 -4.32410240e-01
3.74926999e-02 6.66471541e-01 -4.14979637e-01 3.48528594e-01
1.19560099e+00 7.50111341e-01 -7.55085871e-02 6.75144494e-01
4.13435042e-01 4.97668177e-01 -3.69944096e-01 5.34451082e-02
5.51465631e-01 -7.40273774e-01 1.52605474e-01 6.59493923e-01
4.38644767e-01 -1.26498654e-01 -2.49882266e-01 9.28513348e-01
-1.50641754e-01 2.84057166e-02 -4.96018827e-01 -3.09196621e-01
1.19876504e-01 7.25511611e-01 -3.54710191e-01 -2.41802841e-01
-7.69442320e-01 2.91811168e-01 -2.41132617e-01 2.34504446e-01
-6.50242329e-01 -1.93791687e-01 8.23244810e-01 2.76435554e-01
4.91052151e-01 -2.07281664e-01 -6.11926675e-01 -1.52344429e+00
1.26066506e-01 -1.12821829e+00 5.39738774e-01 -2.29590297e-01
-1.48479795e+00 5.61971545e-01 -4.05838698e-01 -1.28140783e+00
-8.94717753e-01 -8.06568921e-01 -9.09467101e-01 1.04683733e+00
-1.90615153e+00 -5.27108252e-01 7.24856257e-01 5.17109215e-01
3.49896044e-01 -8.23268414e-01 3.97634685e-01 4.28719819e-02
-1.67021900e-01 3.12430292e-01 7.28843331e-01 5.46514511e-01
1.68467790e-01 -1.41477239e+00 1.08231151e+00 7.57571638e-01
5.27797282e-01 4.60978210e-01 8.08898032e-01 -6.63338840e-01
-1.12639594e+00 -3.57503682e-01 1.01009750e+00 -3.75888914e-01
1.68013275e+00 -1.34576410e-01 -6.39108539e-01 9.68379974e-01
6.08933628e-01 -3.03710163e-01 5.44051349e-01 -1.27260789e-01
-4.51287419e-01 -4.45531875e-01 -9.66037631e-01 3.19409788e-01
-1.39127448e-01 -5.72615683e-01 -9.03702617e-01 3.72923128e-02
6.85607314e-01 -1.13846704e-01 -1.07717073e+00 3.13010991e-01
9.55089331e-01 -1.36813641e+00 6.71765685e-01 -5.86067796e-01
2.75848150e-01 3.66003752e-01 -3.18993181e-01 -1.04115033e+00
-7.66874030e-02 -1.00740445e+00 -3.35420221e-01 7.11870790e-01
6.05891228e-01 -1.05311036e+00 1.10286343e+00 8.56951058e-01
5.30665934e-01 -5.04031777e-01 -1.31192088e+00 -7.79504359e-01
8.78478467e-01 -3.04574847e-01 1.00850332e+00 1.08073258e+00
3.37194175e-01 -3.99822921e-01 -3.77745420e-01 -3.63874763e-01
6.99013233e-01 6.09296739e-01 2.61586189e-01 -1.26460361e+00
-4.56283271e-01 -8.84853780e-01 -2.03687459e-01 -6.79735780e-01
2.84198355e-02 -7.75399268e-01 -6.49897993e-01 -4.20471281e-01
-1.02365956e-01 -3.29472989e-01 -1.05617702e+00 -4.89615537e-02
4.55252409e-01 1.16570152e-01 3.45735759e-01 4.14162070e-01
9.84244421e-02 3.70801657e-01 6.57190979e-01 -3.82598430e-01
-7.32300058e-02 3.32613468e-01 -2.90753752e-01 5.21786809e-01
9.87214983e-01 -1.46684080e-01 2.71423489e-01 -2.86822498e-01
5.96741557e-01 3.58157486e-01 -3.61983366e-02 -5.98875403e-01
-1.44462660e-01 -1.62143692e-01 3.04346234e-01 -9.25210416e-01
-1.10810325e-01 -6.41079783e-01 1.28658712e-01 1.02375865e+00
-1.73099920e-01 7.28212416e-01 1.55116022e-01 1.73793793e-01
-6.68406665e-01 9.53861028e-02 6.11192346e-01 -4.79763359e-01
-3.97873908e-01 5.95986724e-01 -3.90479654e-01 4.24784981e-02
8.71346533e-01 1.91218510e-01 -7.11677596e-02 -4.48004514e-01
-5.16765535e-01 1.35214463e-01 1.37761891e-01 5.36292017e-01
3.43893766e-01 -1.27281368e+00 -1.07192016e+00 3.32690895e-01
-3.49188536e-01 -4.83006537e-01 -2.85693347e-01 7.62405932e-01
-1.04658127e+00 8.08418334e-01 -2.99396008e-01 -9.45806950e-02
-3.90912801e-01 4.50376511e-01 7.93394327e-01 -6.91782892e-01
-4.52137172e-01 5.82064509e-01 6.96200831e-03 -6.56073838e-02
1.20944820e-01 -7.28099346e-01 -5.94927818e-02 6.53394520e-01
7.11471498e-01 2.95480967e-01 3.00689250e-01 -3.21631014e-01
-7.02587739e-02 1.97301090e-01 -2.94804454e-01 -3.70655924e-01
1.87358880e+00 3.35176259e-01 -5.32155156e-01 7.19963372e-01
1.19028616e+00 -5.17926514e-02 -1.38105702e+00 -8.89465883e-02
1.02859473e+00 -2.83531815e-01 3.05846464e-02 -5.95577538e-01
-1.65428245e+00 7.18001068e-01 5.97798645e-01 7.09792852e-01
6.36175275e-01 -2.93026388e-01 1.17319000e+00 3.38912725e-01
6.59231305e-01 -1.21209538e+00 -4.14126754e-01 5.84416628e-01
5.83188057e-01 -1.23009682e+00 -1.38345003e-01 8.13404262e-01
-2.76987940e-01 1.25196266e+00 -1.28104568e-01 -6.98213041e-01
1.29972315e+00 5.06289601e-01 3.94831598e-01 -2.29804859e-01
-1.10099626e+00 4.62781221e-01 -2.06731275e-01 9.16923024e-03
7.38679826e-01 3.03423941e-01 -5.51008105e-01 8.11414003e-01
-7.29643762e-01 -1.33094728e-01 6.71107769e-01 5.89554310e-01
-2.56408364e-01 -1.14730287e+00 -4.10052925e-01 5.99261999e-01
-1.31173658e+00 -5.90566039e-01 -1.66779861e-01 9.22281921e-01
-3.54550391e-01 4.80304778e-01 7.50513434e-01 -1.67851478e-01
8.87491256e-02 1.24789923e-01 -5.30670956e-02 1.46241738e-02
-1.19723630e+00 3.93270671e-01 -2.06886888e-01 -5.85896730e-01
-1.34775653e-01 -1.11811888e+00 -9.55333173e-01 -8.59995961e-01
-2.25818884e-02 4.63651955e-01 5.91829240e-01 7.64807522e-01
-2.13609576e-01 -5.91237433e-02 1.19292724e+00 -6.98639214e-01
-1.11294937e+00 -9.18353081e-01 -1.33554494e+00 4.18845952e-01
7.44560897e-01 -1.54855505e-01 -8.16301346e-01 -2.22680941e-01] | [4.542211055755615, 4.1563286781311035] |
89ea73d7-4c26-45c1-8ca7-6567e989a17e | designing-a-prospective-covid-19-therapeutic | 2012.01736 | null | https://arxiv.org/abs/2012.01736v1 | https://arxiv.org/pdf/2012.01736v1.pdf | Designing a Prospective COVID-19 Therapeutic with Reinforcement Learning | The SARS-CoV-2 pandemic has created a global race for a cure. One approach focuses on designing a novel variant of the human angiotensin-converting enzyme 2 (ACE2) that binds more tightly to the SARS-CoV-2 spike protein and diverts it from human cells. Here we formulate a novel protein design framework as a reinforcement learning problem. We generate new designs efficiently through the combination of a fast, biologically-grounded reward function and sequential action-space formulation. The use of Policy Gradients reduces the compute budget needed to reach consistent, high-quality designs by at least an order of magnitude compared to standard methods. Complexes designed by this method have been validated by molecular dynamics simulations, confirming their increased stability. This suggests that combining leading protein design methods with modern deep reinforcement learning is a viable path for discovering a Covid-19 cure and may accelerate design of peptide-based therapeutics for other diseases. | ['Karim Beguir', 'Uğur Şahin', 'Amine Kerkeni', 'Alexandre Laterre', 'Slim Said', 'Joe Phillips', 'Thomas Pierrot', 'Nicolás López Carranza', 'Marcin J. Skwark'] | 2020-12-03 | null | null | null | null | ['protein-design'] | ['medical'] | [ 2.72505552e-01 -4.87990491e-02 -3.66048492e-03 -1.17540091e-01
-7.73028255e-01 -8.09289038e-01 2.21042216e-01 4.85132754e-01
-5.07360160e-01 1.39693105e+00 1.10374823e-01 -6.45744026e-01
2.02678949e-01 -4.59384739e-01 -9.65031087e-01 -6.77515209e-01
-3.95176321e-01 7.91534781e-01 -3.22776616e-01 -4.56824899e-01
2.38737285e-01 7.01512873e-01 -7.01195896e-01 1.58185944e-01
1.21044588e+00 3.68221045e-01 2.74562865e-01 5.57910740e-01
1.23652898e-01 3.27202119e-02 -3.66572410e-01 2.07794994e-01
3.35368067e-01 -7.78611660e-01 -5.58023870e-01 -5.02781570e-01
-1.87754229e-01 -2.25453109e-01 4.31146890e-01 4.18581247e-01
7.72095203e-01 -3.24648879e-02 7.61550486e-01 -4.58669960e-01
-4.88070846e-01 -2.00340971e-01 -4.56992745e-01 5.93169555e-02
5.14010727e-01 4.57642436e-01 9.10498023e-01 -7.21293032e-01
1.07761896e+00 9.73685443e-01 8.92987549e-01 8.24599087e-01
-1.81929052e+00 -3.81795675e-01 -2.57842809e-01 -1.45825788e-01
-1.15475297e+00 -1.54080555e-01 1.97780222e-01 -7.57860184e-01
1.61409807e+00 9.76279303e-02 1.11655533e+00 8.80846977e-01
8.28064144e-01 4.92209107e-01 1.18088973e+00 2.92973742e-02
5.91004074e-01 -3.54689479e-01 -3.62798989e-01 6.10235572e-01
3.03976446e-01 3.59994262e-01 -1.08595071e-02 -9.88277376e-01
5.36602020e-01 3.29946637e-01 -2.03964993e-01 -6.52561724e-01
-8.08488667e-01 1.15238917e+00 2.45940894e-01 1.31785236e-02
-9.23995376e-01 1.30626187e-01 2.79448390e-01 -1.34568391e-02
1.35801816e-02 1.31089413e+00 -7.58749306e-01 5.41384071e-02
-7.70528018e-01 8.72975349e-01 6.52480066e-01 1.26458287e-01
4.58553344e-01 9.45577025e-02 -7.20495209e-02 2.30746403e-01
2.89871752e-01 7.59413183e-01 -1.77703470e-01 -4.94301409e-01
-2.49316037e-01 2.96691060e-01 6.45105302e-01 -4.24514323e-01
-8.36848080e-01 -2.80811369e-01 -3.14756006e-01 2.59222537e-01
4.11904007e-01 -7.58086979e-01 -7.14735389e-01 1.86720216e+00
5.62913060e-01 1.77337006e-02 2.59577483e-01 9.05005515e-01
1.37122236e-02 9.02521849e-01 4.37807977e-01 -6.37240589e-01
1.51382649e+00 -3.67318630e-01 -3.79736423e-01 1.64350420e-01
4.80577260e-01 -4.13226634e-01 4.34074312e-01 2.92283595e-01
-1.04102850e+00 3.06540970e-02 -1.16458488e+00 5.64935207e-01
-2.30632409e-01 -2.92888880e-01 5.81448019e-01 7.28368044e-01
-9.68018234e-01 7.81019688e-01 -8.72348666e-01 -3.64715904e-01
4.45795089e-01 6.76474392e-01 -1.04817897e-01 1.46142840e-01
-1.22452676e+00 1.09023273e+00 2.26223081e-01 -1.10239573e-01
-1.02142501e+00 -9.83868241e-01 -3.77246767e-01 -9.51518789e-02
-3.81736876e-03 -1.20298970e+00 9.13889110e-01 -5.92640340e-01
-1.66373014e+00 5.78917325e-01 -6.16451986e-02 -7.45149434e-01
1.47202164e-01 -2.32781842e-01 -8.56534988e-02 5.08286357e-02
-1.60152674e-01 7.66359925e-01 6.95219278e-01 -8.67821276e-01
-3.32428932e-01 -4.99855906e-01 -2.55268008e-01 1.63997054e-01
6.50019288e-01 3.12355757e-01 7.21839011e-01 -5.55912673e-01
-6.89051747e-01 -1.10179448e+00 -8.80584240e-01 -4.75373089e-01
6.21951111e-02 -1.68277189e-01 2.50952035e-01 -5.56552231e-01
6.43543363e-01 -1.32616687e+00 3.18464130e-01 4.32917565e-01
3.51983935e-01 7.27180302e-01 -1.80696577e-01 1.02387214e+00
6.09484799e-02 1.23185016e-01 -1.02330841e-01 7.03906000e-01
-5.31741738e-01 -3.07978272e-01 -4.32095587e-01 5.02783000e-01
5.20659685e-01 1.24942160e+00 -1.09536064e+00 2.24082574e-01
-3.89846057e-01 5.46734929e-01 -9.40824330e-01 3.06432784e-01
-9.11355257e-01 1.00009763e+00 -8.25502634e-01 4.69281286e-01
7.36301720e-01 -3.41187716e-01 1.00109434e+00 2.40720883e-01
-1.96901664e-01 1.77177742e-01 -3.08713317e-01 1.47370875e+00
6.47604838e-02 -4.41027917e-02 -1.79399028e-01 -7.17202365e-01
8.95639479e-01 4.04540718e-01 8.74571264e-01 -7.97858298e-01
1.78695843e-01 3.95436704e-01 1.70443431e-01 -1.71340302e-01
-9.20196921e-02 -7.69987106e-01 1.21582672e-01 2.60755539e-01
-4.45685983e-01 1.39757186e-01 1.03619032e-01 -2.25691572e-01
1.21579885e+00 3.94318670e-01 2.04041719e-01 -7.65174806e-01
3.39326590e-01 5.59376299e-01 9.01524246e-01 6.90705478e-01
-1.67448029e-01 2.11332552e-03 4.99251485e-01 -7.95815110e-01
-1.52585685e+00 -9.57332730e-01 2.50150431e-02 8.16476524e-01
-2.53225923e-01 -2.58526593e-01 -9.88701880e-01 -2.68985063e-01
2.00331181e-01 5.09639919e-01 -3.37962210e-01 -1.45187363e-01
-8.64037275e-01 -7.62910426e-01 7.04493821e-01 1.04980305e-01
-1.74653485e-01 -9.44398940e-01 -9.58082199e-01 8.59847367e-01
3.10000598e-01 -5.60292482e-01 -3.75738114e-01 3.12687576e-01
-8.91418636e-01 -1.02896476e+00 -1.04726815e+00 -8.17577302e-01
3.30288380e-01 -2.12748632e-01 6.52888596e-01 4.91796657e-02
-4.62393612e-01 -1.58982277e-01 -7.63377249e-02 -5.65455854e-01
-4.96297956e-01 -2.75691837e-01 4.85328048e-01 -5.01196563e-01
4.65613425e-01 -3.95120531e-01 -1.00029516e+00 -7.01232553e-02
-4.16872054e-01 1.97634101e-02 5.95419705e-01 1.07290411e+00
6.87933624e-01 -6.72808647e-01 1.17242432e+00 -7.64816523e-01
9.58553433e-01 -5.26972115e-01 -9.77934718e-01 1.39703840e-01
-8.34591568e-01 4.55805093e-01 9.37327564e-01 -2.01882064e-01
-8.76708388e-01 3.71801138e-01 -4.13288027e-01 1.45854130e-01
2.65322804e-01 2.66475290e-01 1.57050312e-01 -3.17967422e-02
7.19394028e-01 3.82132381e-01 5.38687110e-01 -2.56254554e-01
4.25730556e-01 5.46021238e-02 -6.26751631e-02 -7.58911550e-01
3.94187897e-01 2.10154846e-01 5.35795465e-02 -8.45896304e-01
-1.10479414e-01 -2.48715952e-01 2.88349420e-01 3.39898318e-01
1.29368901e+00 -8.77788961e-01 -1.36830330e+00 1.85350657e-01
-1.22047138e+00 -4.53882784e-01 4.90106121e-02 5.29427469e-01
-8.39744031e-01 4.76945657e-03 -6.56669021e-01 -4.38445538e-01
-7.35620856e-01 -1.43712962e+00 8.91646624e-01 9.82575938e-02
-4.19437468e-01 -4.51493353e-01 1.00972009e+00 2.36073092e-01
4.66048986e-01 6.33541405e-01 1.21580005e+00 -7.11869538e-01
-6.91339672e-01 3.82390946e-01 -1.05050681e-02 -8.71445909e-02
-1.93981960e-01 -2.51404494e-01 -2.72168249e-01 -5.46495497e-01
-2.95533091e-01 -4.07893896e-01 7.45609581e-01 7.10843801e-01
3.32528651e-01 -5.32174051e-01 -5.21790147e-01 4.75068331e-01
1.53843868e+00 9.17732060e-01 6.89910293e-01 1.24069393e-01
3.49823982e-01 1.75787479e-01 5.01436830e-01 5.62622249e-01
-3.05599421e-02 6.16038740e-01 -1.68736558e-02 -2.05238551e-01
6.17238939e-01 -8.28120708e-02 2.46282667e-01 -4.65386994e-02
-3.60464960e-01 -1.44901454e-01 -1.04247665e+00 1.47329450e-01
-1.62209642e+00 -1.20542395e+00 1.42160982e-01 2.09652090e+00
1.16344869e+00 -2.70552427e-01 4.18043464e-01 -7.89041698e-01
3.71558577e-01 -3.48095387e-01 -8.19945276e-01 -9.15876210e-01
-1.35422900e-01 6.37750149e-01 6.19829535e-01 5.26616991e-01
-6.82227731e-01 9.12365615e-01 7.56200457e+00 4.07611579e-01
-1.16324449e+00 -2.89270848e-01 6.75701261e-01 1.38626382e-01
-4.79864001e-01 3.34718376e-01 -6.64185822e-01 2.60914564e-01
1.26686943e+00 -8.56713504e-02 5.44752121e-01 4.53915685e-01
7.99202979e-01 2.40370035e-01 -7.48578131e-01 5.96569419e-01
-5.51716566e-01 -2.21730590e+00 -1.94247931e-01 4.18264389e-01
9.23417211e-01 8.13437998e-02 -2.69675791e-01 -3.22308764e-02
4.20960873e-01 -1.12866116e+00 2.40835592e-01 6.09513819e-01
6.63407028e-01 -1.05366898e+00 4.24511254e-01 1.11416191e-01
-7.72450268e-01 1.05006685e-02 -9.13607553e-02 7.23833069e-02
3.22022110e-01 2.95395613e-01 -1.30592239e+00 -1.55185327e-01
1.77121833e-01 1.44952998e-01 9.93148163e-02 8.77375603e-01
1.55636370e-01 3.48910511e-01 -2.06125662e-01 -4.98477966e-01
3.59020740e-01 -4.86016572e-01 6.00152194e-01 1.14257777e+00
-1.56294823e-01 5.41922331e-01 4.58258927e-01 1.08210433e+00
6.35104701e-02 3.28428596e-01 -5.24637997e-01 -5.82329154e-01
1.78345591e-01 7.63193965e-01 -6.28166914e-01 -2.86268502e-01
-2.40880959e-02 9.24032509e-01 2.38267019e-01 3.99689436e-01
-7.25317717e-01 -2.22898498e-01 1.13837683e+00 1.83155075e-01
5.63611925e-01 -1.51650816e-01 1.93569973e-01 -6.15082085e-01
-5.75273633e-01 -1.43699551e+00 1.22284956e-01 -4.85982150e-01
-1.00863266e+00 3.46320659e-01 -3.67633730e-01 -6.37695134e-01
-3.43376935e-01 -4.97687548e-01 -3.45283628e-01 9.24423158e-01
-1.05516517e+00 -8.32669735e-01 6.55139148e-01 -3.83108831e-03
2.56923944e-01 -1.31797403e-01 1.04348767e+00 1.75056607e-02
-2.60037005e-01 2.13797987e-01 6.54794574e-01 -6.34299517e-01
4.92191404e-01 -1.12019384e+00 5.96938014e-01 3.32620293e-01
-6.70363367e-01 1.04472077e+00 1.21671665e+00 -1.16011274e+00
-1.75905466e+00 -7.48068511e-01 7.24399984e-01 -3.29086214e-01
4.70967054e-01 -4.08644736e-01 -7.15569079e-01 2.92664081e-01
3.27857703e-01 -6.53526127e-01 8.03744078e-01 -2.10832745e-01
-2.66957253e-01 2.97543257e-01 -1.34348822e+00 6.72672033e-01
6.22399271e-01 -2.97212720e-01 -4.78401631e-01 5.27394295e-01
9.69023347e-01 -2.45079622e-01 -8.47754061e-01 4.55482602e-01
6.85235202e-01 -6.28930509e-01 1.02164686e+00 -1.30181229e+00
3.64739969e-02 -5.01915157e-01 1.37453496e-01 -1.36595297e+00
-6.65927351e-01 -1.27421451e+00 8.68789777e-02 1.79858178e-01
6.09528720e-01 -5.54344535e-01 7.53277421e-01 1.67120531e-01
1.23575240e-01 -1.21007097e+00 -6.44230306e-01 -7.02250898e-01
3.91162306e-01 4.50459987e-01 4.88729686e-01 5.75194716e-01
3.29520911e-01 5.76873839e-01 -6.02048039e-01 -2.24516705e-01
2.92642027e-01 1.72722414e-01 4.92999882e-01 -9.97740030e-01
-5.01574039e-01 -4.54104125e-01 -1.60672545e-01 -7.78553486e-01
-1.92589357e-01 -7.13628650e-01 3.69636789e-02 -1.49765491e+00
2.48689264e-01 -4.39706087e-01 -1.63283408e-01 2.55095363e-01
-2.78399270e-02 -2.53837138e-01 -3.02163977e-02 -7.08499104e-02
-2.59086043e-01 4.74330753e-01 1.09471416e+00 3.54915112e-02
-6.17802858e-01 -1.83596760e-01 -8.10825348e-01 2.26233914e-01
9.65547621e-01 -6.62734449e-01 -2.83580333e-01 4.28122908e-01
6.33165061e-01 2.43607953e-01 3.79404910e-02 -4.21736568e-01
-3.91292095e-01 -6.60026550e-01 5.27123332e-01 -5.14725745e-01
1.20909408e-01 -4.80392039e-01 5.06013393e-01 1.22317266e+00
-3.85630541e-02 2.79031456e-01 4.73697960e-01 7.26435125e-01
6.32873178e-01 3.56423736e-01 9.54808354e-01 -1.85570508e-01
-1.89514980e-01 8.92489329e-02 -9.74073172e-01 3.42132866e-01
1.07837176e+00 -3.29959244e-01 -8.33560675e-02 2.86608897e-02
-6.70344532e-01 8.35526362e-02 5.65146387e-01 2.63202965e-01
6.10799730e-01 -8.14748228e-01 -9.43147063e-01 1.95694685e-01
3.40939872e-02 -8.36938500e-01 1.03021309e-01 7.54983425e-01
-1.17394423e+00 8.59138310e-01 -5.34914434e-01 -5.75864553e-01
-1.01457787e+00 9.31183219e-01 6.82507157e-01 -3.53073865e-01
-5.17872155e-01 5.35794258e-01 2.62381285e-01 -2.57570803e-01
-4.46987748e-01 7.39698391e-03 -1.38730360e-02 -2.49326184e-01
6.65808618e-01 1.84816480e-01 -1.29189387e-01 -2.11002797e-01
-7.34270573e-01 2.84819990e-01 -9.38517824e-02 3.61576796e-01
1.57918227e+00 2.93878824e-01 -3.21177803e-02 -3.45068753e-01
9.62683678e-01 -3.79695632e-02 -1.26256967e+00 1.20699041e-01
1.02098227e-01 -9.43765044e-02 -4.58854973e-01 -9.61952984e-01
-3.37054878e-01 5.43257110e-02 9.81977463e-01 -4.42450225e-01
6.96832299e-01 -1.42893285e-01 1.12589061e+00 5.21378219e-01
4.81516987e-01 -9.79055703e-01 6.07810616e-02 3.49374980e-01
7.87814558e-01 -8.58003855e-01 -6.01820350e-02 3.63777786e-01
-7.01572239e-01 7.49608159e-01 2.25581601e-01 -3.63657385e-01
3.34674984e-01 1.94922164e-01 -4.99036647e-02 -4.95824486e-01
-9.38788176e-01 5.08084744e-02 1.26092602e-02 7.61335194e-01
4.76373494e-01 2.65185893e-01 -1.08013213e+00 1.64972290e-01
4.24559236e-01 1.25651255e-01 2.94583112e-01 1.15323722e+00
-7.31888056e-01 -1.65425432e+00 -2.04575881e-01 1.33695811e-01
-4.75836664e-01 -2.35204577e-01 -4.37810868e-01 2.93666273e-01
1.35421846e-02 6.81500912e-01 -2.67234266e-01 1.28740296e-01
2.35730469e-01 2.25943565e-01 6.97431266e-01 -3.62107754e-01
-7.70059943e-01 4.54204917e-01 -1.12411752e-02 -4.82198119e-01
-2.16948148e-02 -5.97237766e-01 -1.90222931e+00 -4.08431977e-01
-1.05040409e-01 3.96571130e-01 6.41504109e-01 8.71838808e-01
1.05951488e+00 3.01910043e-01 5.44373870e-01 -6.45586610e-01
-7.11929977e-01 -1.73832297e-01 -3.77343178e-01 -1.97349628e-03
3.17422152e-01 -3.77665460e-01 4.85678196e-01 -2.04797573e-02] | [4.794200897216797, 5.592466354370117] |
287e0482-e948-4f25-a713-083ca17ffc34 | confidence-regularized-self-training | 1908.09822 | null | https://arxiv.org/abs/1908.09822v3 | https://arxiv.org/pdf/1908.09822v3.pdf | Confidence Regularized Self-Training | Recent advances in domain adaptation show that deep self-training presents a powerful means for unsupervised domain adaptation. These methods often involve an iterative process of predicting on target domain and then taking the confident predictions as pseudo-labels for retraining. However, since pseudo-labels can be noisy, self-training can put overconfident label belief on wrong classes, leading to deviated solutions with propagated errors. To address the problem, we propose a confidence regularized self-training (CRST) framework, formulated as regularized self-training. Our method treats pseudo-labels as continuous latent variables jointly optimized via alternating optimization. We propose two types of confidence regularization: label regularization (LR) and model regularization (MR). CRST-LR generates soft pseudo-labels while CRST-MR encourages the smoothness on network output. Extensive experiments on image classification and semantic segmentation show that CRSTs outperform their non-regularized counterpart with state-of-the-art performance. The code and models of this work are available at https://github.com/yzou2/CRST. | ['B. V. K. Vijaya Kumar', 'Yang Zou', 'Zhiding Yu', 'Xiaofeng Liu', 'Jinsong Wang'] | 2019-08-26 | confidence-regularized-self-training-1 | http://openaccess.thecvf.com/content_ICCV_2019/html/Zou_Confidence_Regularized_Self-Training_ICCV_2019_paper.html | http://openaccess.thecvf.com/content_ICCV_2019/papers/Zou_Confidence_Regularized_Self-Training_ICCV_2019_paper.pdf | iccv-2019-10 | ['synthetic-to-real-translation'] | ['computer-vision'] | [ 3.87176901e-01 4.81411844e-01 -5.86271822e-01 -8.63546133e-01
-9.42779481e-01 -4.81000423e-01 4.36073750e-01 -1.96689352e-01
-4.09818858e-01 9.94370818e-01 -1.04242027e-01 -1.67034239e-01
1.28119931e-01 -5.08261740e-01 -9.15832818e-01 -7.05623388e-01
4.81618196e-01 6.04852378e-01 2.19922766e-01 3.27454329e-01
2.01627403e-01 -7.40825608e-02 -1.15802705e+00 2.26798162e-01
1.33867586e+00 1.08701813e+00 6.59388229e-02 1.87291846e-01
-2.41132572e-01 8.15182745e-01 -4.70649958e-01 -3.98609996e-01
1.73286095e-01 -4.13050979e-01 -9.35156524e-01 4.57156003e-01
2.24445224e-01 -8.54911879e-02 1.26260847e-01 1.36736381e+00
9.53275263e-02 2.19587311e-01 8.71184826e-01 -1.27283764e+00
-9.70703185e-01 5.65443575e-01 -6.53444052e-01 -2.19219834e-01
-7.36651421e-02 -4.99830022e-03 7.63792992e-01 -9.40426826e-01
5.75496733e-01 1.10967445e+00 6.52541935e-01 9.05139029e-01
-1.50940347e+00 -9.55956161e-01 4.76206005e-01 -2.15710010e-02
-1.33438969e+00 -3.60346556e-01 7.97592223e-01 -5.25866389e-01
3.70564997e-01 -7.60349408e-02 2.13144273e-01 1.36861837e+00
-3.24870110e-01 1.07431602e+00 1.39938891e+00 -4.07465845e-01
5.55356264e-01 6.22174025e-01 1.94682822e-01 5.18672466e-01
1.04146615e-01 1.55909568e-01 -2.87004679e-01 -1.85355023e-01
9.37836528e-01 -1.14117071e-01 -3.44774388e-02 -5.56622148e-01
-1.10667956e+00 8.86661291e-01 3.62571001e-01 7.31209964e-02
-2.43355021e-01 -6.00201301e-02 2.59553313e-01 2.12368026e-01
1.03804314e+00 1.33495018e-01 -7.15299189e-01 2.54305243e-01
-8.98073792e-01 -1.87816992e-02 5.42139232e-01 1.14395583e+00
9.07039642e-01 6.18579499e-02 -1.79200172e-01 1.29635286e+00
3.72798413e-01 3.11827779e-01 6.91268146e-01 -1.25994766e+00
1.87500805e-01 3.90146226e-01 2.72471339e-01 -4.79202688e-01
-1.55569687e-01 -6.52621567e-01 -8.08204472e-01 1.87768519e-01
4.99791771e-01 -2.90376544e-01 -1.23976111e+00 1.91675353e+00
4.21265602e-01 5.76854646e-01 1.33305207e-01 9.26879764e-01
4.90236372e-01 4.98729497e-01 4.95627403e-01 -2.11485386e-01
8.43697250e-01 -1.27853858e+00 -6.88250721e-01 -5.34656286e-01
5.94604075e-01 -5.91065705e-01 1.18635547e+00 3.94819170e-01
-8.43751192e-01 -6.20586097e-01 -7.86106884e-01 1.79333612e-01
-1.23773232e-01 3.21550936e-01 4.78641540e-01 3.88837457e-01
-9.09527004e-01 7.50764847e-01 -9.17411208e-01 -1.11054458e-01
7.57499099e-01 2.55581677e-01 -2.48212680e-01 3.38745490e-02
-1.00136888e+00 6.57502472e-01 5.51039159e-01 -2.56455839e-01
-7.68149018e-01 -5.32636940e-01 -8.29383969e-01 -3.13879132e-01
6.27775609e-01 -4.60234433e-01 1.56070673e+00 -1.42949998e+00
-1.91040218e+00 1.28320122e+00 -3.03147942e-01 -4.11845952e-01
6.44424319e-01 -4.06261891e-01 -4.24216360e-01 -5.11991605e-02
5.71355581e-01 8.99697185e-01 1.13434744e+00 -1.48333931e+00
-5.87093472e-01 -1.12937160e-01 -3.73452216e-01 1.65048733e-01
-1.95601255e-01 -2.12822780e-01 -4.19132382e-01 -8.22935581e-01
4.91967380e-01 -1.08600056e+00 -3.40149462e-01 -1.08628072e-01
-6.04647398e-01 -5.12070060e-01 4.97763067e-01 -4.05653149e-01
9.65941370e-01 -2.28924918e+00 -1.05678268e-01 1.96575150e-02
4.35113423e-02 3.00557435e-01 -8.36857408e-02 -3.15311819e-01
-2.54576057e-01 1.73924297e-01 -7.36531734e-01 -6.28124952e-01
5.70596345e-02 4.13555533e-01 -3.95980805e-01 3.58430654e-01
5.35874724e-01 7.91634858e-01 -9.75155771e-01 -6.40963435e-01
8.10653418e-02 3.28962207e-01 -5.10396421e-01 3.81839484e-01
-5.66434026e-01 9.88434374e-01 -7.18702316e-01 5.20615578e-01
8.28850389e-01 -8.56696963e-01 6.91562444e-02 1.23554930e-01
1.09761000e-01 3.02048385e-01 -1.02797163e+00 1.59495437e+00
-1.77812040e-01 2.11425364e-01 2.26167236e-02 -1.31809020e+00
9.56911564e-01 2.37063527e-01 3.54600608e-01 -4.69202846e-01
5.29213026e-02 3.24040592e-01 -7.31459141e-01 -3.08173537e-01
1.76652342e-01 -2.99082339e-01 6.46844273e-03 4.23343003e-01
2.55131364e-01 -6.96391463e-02 -1.73945669e-02 8.31975713e-02
6.04552567e-01 6.86687768e-01 6.42349124e-02 -1.47067234e-01
4.52305049e-01 1.35931611e-01 1.16033757e+00 6.18808508e-01
-3.58437926e-01 6.03411198e-01 3.61786366e-01 -1.56533107e-01
-1.00327408e+00 -1.18005943e+00 -3.95056844e-01 1.19381070e+00
1.77778170e-01 4.32025939e-02 -8.64567757e-01 -1.16026378e+00
-1.59826241e-02 9.08135951e-01 -4.45868462e-01 -1.39441401e-01
-3.94486815e-01 -6.93161786e-01 2.30064735e-01 5.44235170e-01
6.64073169e-01 -1.02269232e+00 -1.01718046e-02 2.28783652e-01
-2.65226901e-01 -1.21970654e+00 -4.30244297e-01 3.76027614e-01
-1.17159665e+00 -7.75076568e-01 -9.46447015e-01 -1.02404237e+00
1.10853124e+00 5.00033982e-03 1.10416400e+00 -2.00374007e-01
2.90812105e-01 1.53888211e-01 -3.80395174e-01 -2.14995682e-01
-5.00781596e-01 -5.29130250e-02 1.74646556e-01 1.29895017e-01
4.86255378e-01 -5.31612575e-01 -5.13747275e-01 5.53213835e-01
-8.39037180e-01 1.05237946e-01 5.11160791e-01 1.00803494e+00
1.02791643e+00 -1.25357226e-01 8.28837872e-01 -1.53792310e+00
4.29616779e-01 -6.73555672e-01 -6.87320709e-01 2.18762860e-01
-1.03769231e+00 1.38047695e-01 5.42608619e-01 -7.35723495e-01
-1.43690705e+00 1.60776362e-01 -2.95141991e-02 -5.41222155e-01
-5.96246421e-01 2.74332583e-01 1.63357690e-01 8.88448730e-02
7.15373337e-01 9.93357822e-02 -1.89965189e-01 -6.25855982e-01
4.00020301e-01 7.04343259e-01 5.94779134e-01 -6.74538076e-01
7.10203350e-01 5.00330150e-01 -5.31532526e-01 -3.39983791e-01
-1.50401902e+00 -4.13846880e-01 -6.62256300e-01 2.21863184e-02
8.61775160e-01 -1.12974012e+00 -9.33247060e-02 5.76935589e-01
-8.32013607e-01 -7.36651480e-01 -3.99891317e-01 4.71102864e-01
-6.72695637e-01 4.67248201e-01 -5.62043250e-01 -7.65839934e-01
-9.80704725e-02 -9.38679218e-01 9.01340365e-01 4.33367491e-01
-1.65032119e-01 -1.13534164e+00 -8.67344067e-02 4.95711833e-01
2.79294401e-01 6.16850927e-02 5.88523805e-01 -8.22164357e-01
-3.65159631e-01 -1.13089338e-01 -2.82555729e-01 7.62594581e-01
6.62789196e-02 -3.22093159e-01 -1.09689331e+00 -2.24126294e-01
1.84000537e-01 -7.20047355e-01 8.78969431e-01 6.58972383e-01
1.31380701e+00 -2.07204998e-01 -3.79683822e-01 7.82160163e-01
1.24634933e+00 -4.59285676e-02 3.23689818e-01 4.87854660e-01
6.23614609e-01 4.42046970e-01 1.02353311e+00 4.09904242e-01
3.94355118e-01 3.95889968e-01 1.50205970e-01 -2.92477310e-01
1.24440081e-01 -3.96932006e-01 2.30133370e-01 6.36988342e-01
1.80884555e-01 2.64064241e-02 -8.08435023e-01 4.85507905e-01
-2.08570790e+00 -5.89068234e-01 -4.61712293e-02 2.20018005e+00
1.18359244e+00 4.28514689e-01 1.51144359e-02 -2.15660423e-01
1.03650188e+00 -2.02254310e-01 -1.00910878e+00 9.01805994e-04
-1.83430344e-01 1.19415686e-01 5.74613571e-01 4.72265810e-01
-1.37743926e+00 1.39336944e+00 5.96667385e+00 8.98098111e-01
-1.02808928e+00 4.27132428e-01 9.73694146e-01 1.92109630e-01
-1.48222640e-01 1.61099732e-01 -7.67847717e-01 5.45971990e-01
7.45634198e-01 8.90816972e-02 2.30573207e-01 1.28203499e+00
1.25066452e-02 -1.81945145e-01 -8.79085004e-01 8.16291988e-01
-2.15329096e-01 -9.80751932e-01 -3.38593006e-01 -1.96114242e-01
1.18934250e+00 1.82215676e-01 1.97234020e-01 5.37570655e-01
7.77611911e-01 -6.01214230e-01 7.85311460e-01 3.93925518e-01
9.86952066e-01 -4.65405822e-01 4.52527314e-01 4.64347601e-01
-6.58538878e-01 -2.94056702e-02 -5.01710653e-01 1.78758293e-01
1.62920758e-01 8.39057386e-01 -6.46093130e-01 1.86388850e-01
7.44662821e-01 9.77734745e-01 -3.50363821e-01 7.47920454e-01
-7.44293511e-01 9.19256151e-01 -2.60631293e-01 3.63885373e-01
1.59818590e-01 -4.59757030e-01 4.56896216e-01 1.08531487e+00
-4.31336127e-02 4.75389101e-02 3.44709367e-01 1.23254037e+00
6.42225752e-03 -1.38374731e-01 -1.60314381e-01 4.13735807e-02
5.05691350e-01 9.54487205e-01 -8.46176207e-01 -6.21555746e-01
-2.12930933e-01 1.28820634e+00 5.69302022e-01 6.87089562e-01
-7.85506606e-01 4.66994522e-03 4.89810973e-01 -1.13789991e-01
2.75206268e-01 -1.10855922e-01 -5.91888130e-01 -1.39688718e+00
-1.57651920e-02 -5.48597395e-01 4.22174782e-01 -6.72148287e-01
-1.65868807e+00 5.76959431e-01 -5.72871119e-02 -1.34736371e+00
-1.75494358e-01 -5.32040834e-01 -4.15946305e-01 7.34978914e-01
-1.86364961e+00 -9.00717497e-01 -2.15018675e-01 5.24402559e-01
6.22077048e-01 -1.18691944e-01 8.78583670e-01 2.41504595e-01
-7.03514397e-01 7.20746994e-01 3.73339176e-01 1.38941646e-01
1.05227768e+00 -1.31361449e+00 2.83870727e-01 5.93042731e-01
-4.85844985e-02 4.67403322e-01 5.51527917e-01 -8.14279258e-01
-3.82308334e-01 -1.27371252e+00 6.93230450e-01 -2.96730429e-01
5.55077493e-01 -1.41800404e-01 -1.18909967e+00 8.93746972e-01
-1.40510693e-01 1.26276091e-01 8.54462981e-01 2.49850806e-02
-4.51835155e-01 1.92695662e-01 -1.22309804e+00 3.34856868e-01
9.25508261e-01 -4.13351655e-01 -5.47022104e-01 6.21042848e-01
8.91094506e-01 -3.56481522e-01 -7.69592583e-01 5.54419219e-01
5.60956448e-02 -9.12645400e-01 8.20076346e-01 -4.37561452e-01
3.36081386e-01 -2.54591316e-01 9.79074240e-02 -1.29826319e+00
-3.97357136e-01 -4.33853298e-01 8.49626865e-03 1.26521420e+00
5.36333978e-01 -7.09858179e-01 1.12527454e+00 8.36697996e-01
-1.50511399e-01 -5.75846553e-01 -7.81271100e-01 -1.00222421e+00
2.95013160e-01 -5.17613173e-01 2.42248669e-01 1.33447826e+00
-7.60311186e-02 2.49491051e-01 -2.72591829e-01 1.88306600e-01
1.05658305e+00 -2.55351886e-02 4.35011059e-01 -1.33157158e+00
-2.34727770e-01 -2.72199512e-01 1.01004444e-01 -1.28746295e+00
4.98914778e-01 -8.39707315e-01 2.14299098e-01 -1.35538363e+00
5.66823594e-02 -7.79005766e-01 -4.44982111e-01 7.62582004e-01
-2.90763348e-01 3.18398684e-01 -2.11210176e-01 5.58696270e-01
-8.64706933e-01 6.33123815e-01 1.18680584e+00 1.23561285e-02
-3.61894280e-01 3.09105992e-01 -6.86390460e-01 9.85806465e-01
1.04621828e+00 -8.61357749e-01 -3.82827520e-01 -4.02339518e-01
2.01540962e-02 -1.50298625e-01 2.42540196e-01 -9.05444801e-01
1.26449481e-01 -2.89347619e-01 3.94580066e-01 -3.55118036e-01
1.98338524e-01 -6.68940544e-01 -1.27101690e-01 1.54543370e-01
-6.03311837e-01 -5.80268562e-01 1.03956185e-01 6.83831036e-01
-2.74234802e-01 -3.66502672e-01 1.10202587e+00 -2.10551649e-01
-6.83970630e-01 1.61677241e-01 -2.64191747e-01 2.34302372e-01
8.61406446e-01 -2.88393438e-01 -1.07417665e-01 -3.36834550e-01
-1.15901053e+00 5.19456327e-01 4.59476501e-01 3.06188703e-01
5.21589518e-01 -1.32378662e+00 -6.45078123e-01 1.96205258e-01
1.08911112e-01 4.15210128e-01 1.51423290e-01 6.26236022e-01
4.46702763e-02 2.96752244e-01 5.81202470e-03 -8.15987051e-01
-7.97792733e-01 4.53276217e-01 1.85234144e-01 -2.49903321e-01
-4.77673799e-01 1.28006256e+00 5.19886553e-01 -9.06130791e-01
3.20831299e-01 -8.67248327e-02 -9.23373029e-02 -2.80697763e-01
1.88368291e-01 1.37330756e-01 -3.35279197e-01 -4.40458089e-01
-9.68940482e-02 6.25495791e-01 -3.26837182e-01 -1.34917364e-01
1.30683577e+00 -3.23895097e-01 9.36894268e-02 5.89752734e-01
1.00370657e+00 -3.60689700e-01 -1.82316554e+00 -7.22093821e-01
3.64446312e-01 -2.89160281e-01 6.16571307e-03 -1.05334508e+00
-8.51174831e-01 6.40236318e-01 5.33215165e-01 -1.12707019e-01
1.11058533e+00 9.91876423e-02 6.09019279e-01 2.20932022e-01
2.99334735e-01 -1.50613642e+00 2.19001159e-01 5.56115687e-01
5.72674572e-01 -1.67675567e+00 -2.02248886e-01 -6.46353483e-01
-1.16581857e+00 6.69380307e-01 9.00182843e-01 -2.57363260e-01
7.74502337e-01 -1.29274189e-01 4.03194129e-01 2.20502064e-01
-5.01457632e-01 -1.25293240e-01 3.06621939e-01 6.03280902e-01
3.77769411e-01 1.24185003e-01 -2.32014865e-01 8.18377256e-01
3.26243117e-02 2.60193646e-01 1.91443026e-01 9.49552655e-01
-4.12166655e-01 -1.21080029e+00 -4.17820483e-01 3.80960315e-01
-5.16012967e-01 -2.21272698e-03 -1.98367059e-01 3.03565383e-01
1.96258932e-01 9.49707329e-01 -4.67317365e-02 -2.82330424e-01
1.18691161e-01 3.55306387e-01 1.66960374e-01 -9.03327048e-01
-1.41902298e-01 4.90334004e-01 -2.17142954e-01 -4.38229054e-01
-5.04589260e-01 -5.95400393e-01 -1.53136683e+00 2.12938786e-01
-5.57412684e-01 3.22575241e-01 6.74800336e-01 9.35390174e-01
3.66745949e-01 3.86118174e-01 5.77096641e-01 -6.94222748e-01
-8.13987553e-01 -1.06852686e+00 -5.54538250e-01 7.02791452e-01
3.19945127e-01 -8.06301355e-01 -5.12025118e-01 3.16670179e-01] | [9.611597061157227, 1.413488745689392] |
1fecd337-31e2-49a1-bc1a-9a4f9a30e56b | sequential-3d-human-pose-and-shape-estimation | null | null | http://openaccess.thecvf.com/content_CVPR_2020/html/Wang_Sequential_3D_Human_Pose_and_Shape_Estimation_From_Point_Clouds_CVPR_2020_paper.html | http://openaccess.thecvf.com/content_CVPR_2020/papers/Wang_Sequential_3D_Human_Pose_and_Shape_Estimation_From_Point_Clouds_CVPR_2020_paper.pdf | Sequential 3D Human Pose and Shape Estimation From Point Clouds | This work addresses the problem of 3D human pose and shape estimation from a sequence of point clouds. Existing sequential 3D human shape estimation methods mainly focus on the template model fitting from a sequence of depth images or the parametric model regression from a sequence of RGB images. In this paper, we propose a novel sequential 3D human pose and shape estimation framework from a sequence of point clouds. Specifically, the proposed framework can regress 3D coordinates of mesh vertices at different resolutions from the latent features of point clouds. Based on the estimated 3D coordinates and features at the low resolution, we develop a spatial-temporal mesh attention convolution (MAC) to predict the 3D coordinates of mesh vertices at the high resolution. By assigning specific attentional weights to different neighboring points in the spatial and temporal domains, our spatial-temporal MAC can capture structured spatial and temporal features of point clouds. We further generalize our framework to the real data of human bodies with a weakly supervised fine-tuning method. The experimental results on SURREAL, Human3.6M, DFAUST and the real detailed data demonstrate that the proposed approach can accurately recover the 3D body model sequence from a sequence of point clouds.
| [' Jian Yang', ' Lei Liu', ' Guofeng Zhang', ' Jin Xie', 'Kangkan Wang'] | 2020-06-01 | null | null | null | cvpr-2020-6 | ['3d-human-pose-and-shape-estimation'] | ['computer-vision'] | [-5.45784123e-02 -3.11718255e-01 3.00438583e-01 -4.16320115e-01
-5.33700466e-01 -1.00596100e-01 2.77827412e-01 -2.16076940e-01
-5.66909432e-01 2.39932001e-01 -1.17733501e-01 5.22546768e-01
-6.98706717e-05 -6.61300957e-01 -1.00864303e+00 -3.61694515e-01
-1.41668320e-01 1.11202312e+00 3.48356158e-01 -9.55112576e-02
1.68979436e-01 7.89185703e-01 -1.80427134e+00 7.28141749e-03
5.57280064e-01 1.21890581e+00 2.39556670e-01 6.40734732e-01
-1.39646769e-01 1.33199066e-01 -3.24904054e-01 -7.17778131e-02
4.14463967e-01 2.26157278e-01 -5.10692298e-01 4.09308106e-01
5.35077214e-01 -4.31914002e-01 -1.43397987e-01 8.18585932e-01
5.49811006e-01 1.26156345e-01 6.04342461e-01 -1.18627429e+00
-3.04631442e-01 -3.83171499e-01 -7.97231376e-01 -7.99613446e-02
5.39772213e-01 1.81620821e-01 3.39569122e-01 -1.24620271e+00
5.21630049e-01 1.67826498e+00 9.43598807e-01 5.15128314e-01
-1.07754350e+00 -8.05519819e-01 1.68700695e-01 1.67391151e-01
-1.71690392e+00 -6.71230033e-02 9.60010409e-01 -6.95389569e-01
8.40449393e-01 4.69643660e-02 1.20946813e+00 8.50110114e-01
2.90019661e-02 5.22061110e-01 9.35009778e-01 -1.70883611e-01
-1.23733394e-02 -3.46974164e-01 -2.83029765e-01 8.81089568e-01
-9.79126096e-02 2.33759150e-01 -6.57659709e-01 -2.74292797e-01
1.53054810e+00 4.25983816e-01 -1.25204697e-01 -7.62888372e-01
-1.44537508e+00 5.39993823e-01 6.47616446e-01 -1.58526361e-01
-6.10691667e-01 3.08023006e-01 1.38650179e-01 -1.83190167e-01
6.85034335e-01 -1.62625849e-01 -6.94820225e-01 -5.90267312e-03
-7.76237547e-01 3.86908740e-01 4.68831688e-01 1.30629086e+00
7.77683139e-01 -2.49370068e-01 6.77701756e-02 4.78464663e-01
5.61120510e-01 1.01890826e+00 2.44934708e-01 -9.95619416e-01
5.25532186e-01 6.66102707e-01 4.72269386e-01 -1.16547525e+00
-5.56287348e-01 7.17336833e-02 -7.66356885e-01 3.25100690e-01
2.97720522e-01 1.54143855e-01 -1.08779836e+00 1.35055673e+00
9.11225140e-01 3.89577717e-01 -5.29865086e-01 1.47769773e+00
7.16347873e-01 4.57096338e-01 -8.14599618e-02 1.63178250e-01
1.29497087e+00 -5.73852897e-01 -5.57174623e-01 -2.16212377e-01
3.87937874e-02 -3.92634600e-01 9.33092237e-01 1.73012540e-01
-1.24221289e+00 -9.41165805e-01 -8.60611975e-01 -3.90481323e-01
-1.64243430e-01 1.53871194e-01 2.21793488e-01 5.21490872e-02
-6.83779180e-01 6.57046914e-01 -1.30593669e+00 -2.51231790e-01
3.02764893e-01 6.41810954e-01 -5.00196278e-01 -6.17198972e-03
-8.62153471e-01 6.59364462e-01 2.65843924e-02 5.67156553e-01
-5.18692195e-01 -5.69244623e-01 -1.04195011e+00 -2.82434165e-01
2.49877483e-01 -1.14394844e+00 9.73630011e-01 -5.89912057e-01
-1.43040526e+00 1.21411359e+00 -2.25943387e-01 -1.84979066e-01
6.71882570e-01 -7.39009976e-01 1.31946594e-01 1.93182230e-01
1.27771407e-01 7.38627791e-01 1.09243608e+00 -1.15858173e+00
-4.99974042e-01 -1.02964306e+00 -3.99116158e-01 4.47670639e-01
2.07453221e-01 -2.90247619e-01 -7.81471431e-01 -4.54960167e-01
6.34215117e-01 -1.09325171e+00 -2.90630817e-01 4.79819030e-01
-1.51536450e-01 -1.92974433e-01 3.52762371e-01 -6.49201095e-01
5.21732271e-01 -2.10491729e+00 6.49322629e-01 3.34356368e-01
1.66470379e-01 -2.41325587e-01 2.23823339e-01 -2.13429958e-01
1.39277846e-01 -2.43978485e-01 -1.93397909e-01 -5.68325162e-01
3.26647121e-03 2.77601749e-01 -2.79578846e-02 8.38260591e-01
1.49478257e-01 9.85257685e-01 -8.35302055e-01 -8.03814530e-01
5.44246554e-01 8.57226431e-01 -5.20305455e-01 4.44596916e-01
-2.17978731e-01 7.29819059e-01 -6.39190793e-01 6.51270330e-01
7.68310666e-01 -4.26956743e-01 -3.67143512e-01 -3.78075749e-01
-5.11394590e-02 -1.57445535e-01 -1.12875664e+00 2.40888047e+00
-1.48850083e-01 9.89810079e-02 4.58076261e-02 -5.89355290e-01
9.94279027e-01 3.21030170e-01 8.52919757e-01 -5.14182448e-01
1.80306897e-01 8.16254690e-02 -4.33789313e-01 -4.94076043e-01
3.37674499e-01 -2.01640636e-01 -2.56652713e-01 4.60337065e-02
-1.48784801e-01 -3.37612629e-01 -6.82065070e-01 -3.61999780e-01
6.46432877e-01 5.71405888e-01 1.31545678e-01 1.79156899e-01
4.65373725e-01 9.32293609e-02 5.02319932e-01 3.38311821e-01
-1.54255524e-01 8.71919274e-01 -2.78497577e-01 -7.28211462e-01
-1.37919033e+00 -1.30372477e+00 5.71693107e-03 6.62929773e-01
3.42409343e-01 -1.73842952e-01 -5.57960033e-01 -3.26119632e-01
4.45002198e-01 2.68898788e-03 -8.39953125e-01 8.14730451e-02
-9.77608919e-01 -6.37206808e-02 7.98681527e-02 7.35351264e-01
5.07783532e-01 -1.03956282e+00 -1.13634145e+00 8.82011876e-02
-2.34264657e-01 -1.26502490e+00 -6.48650229e-01 -1.31354719e-01
-1.10553467e+00 -9.53168213e-01 -9.63412702e-01 -6.41939878e-01
7.27828801e-01 -3.91325280e-02 1.00792885e+00 -5.82614616e-02
-3.62440675e-01 7.59547651e-01 -1.13156229e-01 -3.63201618e-01
3.28831851e-01 -2.29276061e-01 4.32698995e-01 9.61672217e-02
4.54317540e-01 -7.41401434e-01 -6.68116629e-01 4.15695459e-01
-3.17378759e-01 2.52945215e-01 4.01141137e-01 5.13646901e-01
1.23044455e+00 -2.77293146e-01 -2.57840425e-01 -3.65557760e-01
2.94956528e-02 -2.70808369e-01 -6.99530602e-01 -1.38297096e-01
-7.79124051e-02 6.01301268e-02 -3.50182094e-02 -8.31020653e-01
-7.14489639e-01 6.42256558e-01 3.73070985e-02 -1.34638178e+00
-3.37207377e-01 8.60400125e-02 -1.53545663e-01 1.02131099e-01
4.55807269e-01 2.18869731e-01 2.05262482e-01 -7.50725627e-01
2.41894454e-01 3.08835596e-01 7.88623273e-01 -6.86970949e-01
8.87445748e-01 8.70473385e-01 2.25999877e-01 -5.91221213e-01
-6.30637586e-01 -6.64543509e-01 -1.43493032e+00 -3.70934129e-01
1.21896923e+00 -1.29690421e+00 -1.10906374e+00 4.16401774e-01
-1.41546893e+00 -2.20778167e-01 -2.37781554e-01 5.02066910e-01
-9.99818444e-01 3.74550045e-01 -4.81009781e-01 -9.54719424e-01
-3.54426384e-01 -1.04326296e+00 1.89654040e+00 -9.39866528e-02
-3.55042070e-01 -5.42227924e-01 6.23891950e-02 2.46722817e-01
-3.66595894e-01 7.44432867e-01 4.30566162e-01 4.57406975e-02
-6.31623209e-01 -1.97166607e-01 -8.85185506e-03 -6.35107011e-02
4.02708165e-02 -4.70330656e-01 -6.41479850e-01 -1.96040437e-01
2.12651551e-01 -1.75704628e-01 2.69672364e-01 7.03355789e-01
1.13113070e+00 1.49007263e-02 -4.78437245e-01 1.05235994e+00
1.15573168e+00 -2.26517081e-01 2.03003109e-01 2.99965799e-01
1.08680451e+00 6.84346974e-01 9.84164715e-01 7.26147056e-01
5.22539198e-01 9.19437051e-01 5.66261172e-01 4.42187712e-02
1.76193371e-01 -5.34758210e-01 1.46700516e-01 8.78717005e-01
-5.44117510e-01 6.09891057e-01 -1.05712402e+00 4.67808247e-01
-1.93871176e+00 -5.56718469e-01 -1.45765364e-01 2.28068185e+00
6.11682177e-01 2.83662695e-02 3.17265660e-01 3.52866901e-03
8.07490230e-01 -2.58147001e-01 -8.45789313e-01 1.85060933e-01
2.27099374e-01 1.62593648e-01 4.74393636e-01 3.13065231e-01
-9.36002135e-01 7.31371582e-01 5.48197889e+00 3.53744060e-01
-7.22658455e-01 4.99571115e-02 -8.71926174e-02 -6.86163545e-01
1.10425539e-01 -4.21466947e-01 -7.88409293e-01 5.65090775e-01
5.80399513e-01 1.92917675e-01 2.88432837e-01 7.70596147e-01
9.24296230e-02 2.81182587e-01 -1.37237501e+00 1.68514454e+00
4.30187657e-02 -9.48654771e-01 -5.95808364e-02 2.99958497e-01
3.69989097e-01 1.12589799e-01 6.18198179e-02 5.98109374e-03
1.08744442e-01 -1.07595646e+00 1.09532607e+00 1.12604046e+00
8.83331954e-01 -8.02550673e-01 3.66773158e-01 8.08222473e-01
-1.32728183e+00 7.39137754e-02 -4.70902473e-01 -1.25654444e-01
2.40264133e-01 2.30181262e-01 -7.06884921e-01 3.72031450e-01
1.21756566e+00 8.32904637e-01 -3.77177447e-01 8.89539182e-01
-3.22348811e-02 -4.86659668e-02 -5.72974682e-01 1.33074701e-01
-1.66476279e-01 -2.04742908e-01 5.07264316e-01 7.05346346e-01
4.24584299e-01 6.09611750e-01 3.08648556e-01 1.03060794e+00
3.38198811e-01 3.60141275e-04 -4.37171489e-01 4.41854537e-01
4.06180650e-01 8.56050611e-01 -5.25169671e-01 -1.95126280e-01
-2.11152986e-01 1.17760491e+00 4.38758761e-01 1.92818329e-01
-8.38606894e-01 8.19062293e-02 6.53333008e-01 4.67115253e-01
5.03488123e-01 -6.17884815e-01 -4.50076967e-01 -1.10716581e+00
2.41696820e-01 -5.54510057e-01 2.25407705e-01 -1.20199275e+00
-1.27690780e+00 5.33272147e-01 1.66562408e-01 -1.46374142e+00
-2.09614187e-01 -4.76008564e-01 2.00233255e-02 1.04297173e+00
-1.12818098e+00 -1.08106434e+00 -6.55523181e-01 1.00716722e+00
5.35882831e-01 2.04067200e-01 6.11146808e-01 -1.31174540e-02
1.45843690e-02 1.45927161e-01 -3.61076027e-01 3.01291525e-01
5.17458975e-01 -1.10868621e+00 8.42257142e-01 1.52442440e-01
3.98694277e-02 3.85194272e-01 5.61865330e-01 -9.71475899e-01
-1.39989841e+00 -7.93930054e-01 5.24876356e-01 -9.64337528e-01
-2.87143234e-02 -6.33852780e-01 -9.23992932e-01 8.29787910e-01
-6.14139199e-01 3.65877956e-01 2.04165891e-01 -2.90499516e-02
-1.69923097e-01 1.16047248e-01 -1.25288081e+00 3.05233151e-01
1.53911483e+00 -3.97289425e-01 -7.71476150e-01 2.64628172e-01
7.71542072e-01 -1.00692677e+00 -1.04000652e+00 6.43588781e-01
8.24811935e-01 -6.59249425e-01 1.45199120e+00 -4.72181499e-01
2.49733299e-01 -4.10252810e-01 -2.76764035e-01 -9.56531584e-01
-3.70303243e-01 -1.87188983e-01 -3.79126757e-01 6.48903728e-01
-3.80058169e-01 -9.10949036e-02 1.09104931e+00 8.11826289e-01
1.28802955e-01 -6.48750901e-01 -1.28252554e+00 -5.53953290e-01
-6.05105460e-02 -6.27652824e-01 8.34599614e-01 4.73483294e-01
-5.22810102e-01 3.01034264e-02 -4.24388885e-01 5.99788487e-01
1.08534181e+00 3.41415316e-01 1.19913971e+00 -1.59253132e+00
-8.17502737e-02 2.64767092e-02 -7.74996221e-01 -1.47003126e+00
2.46266484e-01 -5.65888226e-01 2.37299427e-01 -1.14926338e+00
-9.32946503e-02 -2.48114333e-01 6.65487498e-02 1.17829986e-01
-8.38460177e-02 2.40289003e-01 2.27731109e-01 3.96628618e-01
-4.67381358e-01 6.93362057e-01 1.39351678e+00 -4.20017578e-02
-8.11288506e-02 9.48427841e-02 2.29906723e-01 9.18014765e-01
2.81592131e-01 -4.70862120e-01 -1.78218678e-01 -5.80782890e-01
6.92811683e-02 3.23204309e-01 9.52074587e-01 -1.06894624e+00
2.87329078e-01 -9.96562243e-02 1.03223848e+00 -1.37221575e+00
9.53747690e-01 -1.20721734e+00 4.24538970e-01 3.74661475e-01
2.25151051e-02 2.45745227e-01 1.77679554e-01 7.53561974e-01
1.95647135e-01 4.48316187e-01 5.97733259e-01 -5.49356639e-01
-6.63743556e-01 9.13116455e-01 3.31250638e-01 -9.53499451e-02
9.03210580e-01 -6.26144290e-01 6.86164796e-01 -1.93179548e-02
-1.19380736e+00 2.96596944e-01 7.92722523e-01 5.14685810e-01
1.28658724e+00 -1.72119832e+00 -6.15035474e-01 6.35021985e-01
1.07637450e-01 7.43482709e-01 3.78203124e-01 7.10175633e-01
-3.52580637e-01 2.48810738e-01 -3.75279337e-01 -1.35102713e+00
-1.39281607e+00 6.75347328e-01 5.61680019e-01 9.49321091e-02
-1.08945155e+00 8.35527360e-01 3.33698034e-01 -6.98635459e-01
2.60216385e-01 -6.05277181e-01 5.79716973e-02 -3.78044993e-01
2.58817971e-01 3.53708357e-01 -2.52714366e-01 -1.20212686e+00
-4.35718358e-01 1.31699085e+00 4.05535132e-01 -2.02031717e-01
1.42716956e+00 -2.93771118e-01 5.58998026e-02 8.70518029e-01
1.23858237e+00 -2.93966234e-01 -1.70971632e+00 -4.64661479e-01
-2.19373465e-01 -8.40500534e-01 -2.42296651e-01 -2.90116966e-01
-9.98847187e-01 1.04388285e+00 8.08871448e-01 -3.99532765e-01
8.88550103e-01 1.58791706e-01 8.88248742e-01 -2.20573246e-02
7.68723249e-01 -8.66068900e-01 5.27636781e-02 3.79744321e-01
1.13264811e+00 -1.14569867e+00 8.80926922e-02 -4.53140914e-01
-3.26672643e-01 9.63625610e-01 6.68471932e-01 -4.64341223e-01
7.41552591e-01 1.18060470e-01 -2.24306464e-01 -5.59878767e-01
-4.05424684e-01 -1.96277484e-01 5.63994706e-01 7.88254678e-01
-2.71369424e-02 2.03415472e-02 2.63356388e-01 6.23902857e-01
-3.74352902e-01 2.02842936e-01 -1.64158687e-01 7.56551027e-01
-3.19850773e-01 -6.21838748e-01 -9.03631568e-01 1.22272611e-01
-2.20354050e-02 2.97708899e-01 -1.49818867e-01 8.01801085e-01
3.96416664e-01 3.30955744e-01 5.07253051e-01 -4.05196875e-01
8.36043417e-01 1.48748606e-01 9.82442677e-01 -5.95527530e-01
-3.12845379e-01 2.64459342e-01 -5.01237154e-01 -8.13236594e-01
-6.78290486e-01 -8.89115810e-01 -1.49266601e+00 -1.40925854e-01
2.59876847e-02 -1.21989854e-01 6.41577601e-01 9.11775470e-01
2.98894107e-01 2.73361266e-01 2.35224605e-01 -1.61725676e+00
-3.23601604e-01 -9.38286901e-01 -7.43797958e-01 8.10323894e-01
5.22124648e-01 -1.18263125e+00 -7.73712024e-02 1.88575730e-01] | [7.019306182861328, -1.094046950340271] |
38761ff7-16de-44ba-87a2-52bc22af9882 | unified-feature-and-instance-based-domain | null | null | https://aclanthology.org/2020.emnlp-main.572 | https://aclanthology.org/2020.emnlp-main.572.pdf | Unified Feature and Instance Based Domain Adaptation for Aspect-Based Sentiment Analysis | The supervised models for aspect-based sentiment analysis (ABSA) rely heavily on labeled data. However, fine-grained labeled data are scarce for the ABSA task. To alleviate the dependence on labeled data, prior works mainly focused on feature-based adaptation, which used the domain-shared knowledge to construct auxiliary tasks or domain adversarial learning to bridge the gap between domains, while ignored the attribute of instance-based adaptation. To resolve this limitation, we propose an end-to-end framework to jointly perform feature and instance based adaptation for the ABSA task in this paper. Based on BERT, we learn domain-invariant feature representations by using part-of-speech features and syntactic dependency relations to construct auxiliary tasks, and jointly perform word-level instance weighting in the framework of sequence labeling. Experiment results on four benchmarks show that the proposed method can achieve significant improvements in comparison with the state-of-the-arts in both tasks of cross-domain End2End ABSA and cross-domain aspect extraction. | ['Rui Xia', 'Jianfei Yu', 'Chenggong Gong'] | null | null | null | null | emnlp-2020-11 | ['aspect-extraction'] | ['natural-language-processing'] | [ 3.73273462e-01 -2.24612392e-02 -1.99579760e-01 -8.58300090e-01
-1.10983014e+00 -6.12580955e-01 6.23213530e-01 -4.48728167e-02
-4.16665524e-01 6.16179109e-01 2.85311460e-01 -9.29332674e-02
1.97786793e-01 -7.36420214e-01 -5.81445515e-01 -6.25695825e-01
4.94032115e-01 4.27343935e-01 3.26625668e-02 -4.13559109e-01
-5.75667061e-02 -2.91178469e-03 -1.26646256e+00 5.83829343e-01
1.00423753e+00 1.06957996e+00 -1.20069347e-01 9.63786766e-02
-6.48001492e-01 4.78098124e-01 -6.07731998e-01 -8.44849825e-01
1.62562177e-01 -5.34527242e-01 -6.15689337e-01 3.35698873e-01
-1.55096678e-02 1.15829729e-01 -6.46159053e-02 9.01977420e-01
6.67483330e-01 8.03387910e-02 6.60054922e-01 -1.35645008e+00
-6.72776282e-01 4.51486856e-01 -6.62028193e-01 -2.51049153e-03
1.36328846e-01 9.87464190e-02 1.28038490e+00 -1.13194633e+00
5.36740005e-01 1.10970199e+00 6.55597150e-01 6.44291401e-01
-8.80999565e-01 -7.20185220e-01 8.42513442e-01 -1.15696830e-03
-1.00012863e+00 -4.36818391e-01 1.14750123e+00 -2.36463770e-01
1.17248917e+00 -9.82375666e-02 4.76129115e-01 1.40345013e+00
-1.90211713e-01 1.18357313e+00 1.12355995e+00 -5.55820525e-01
2.44515374e-01 1.32627115e-01 3.56916636e-01 6.92297876e-01
1.96453348e-01 -3.85656022e-02 -6.70931458e-01 -3.13374281e-01
2.65321553e-01 5.80479950e-03 8.30131769e-02 -4.75202352e-01
-1.15931749e+00 9.28448677e-01 3.07272896e-02 -2.72514746e-02
-3.03010792e-01 -5.77106535e-01 9.31742787e-01 4.28887695e-01
9.09254491e-01 3.63180190e-01 -1.29302323e+00 -6.62449524e-02
-5.41742623e-01 1.70543373e-01 8.66251886e-01 1.17511129e+00
7.85688281e-01 2.12978661e-01 -3.29328299e-01 1.25391042e+00
2.30980143e-01 5.76942384e-01 7.52723932e-01 -1.88562796e-01
7.25398004e-01 9.67170298e-01 -1.85890123e-01 -4.99756664e-01
-3.46714437e-01 -4.46961790e-01 -5.82183123e-01 7.24521130e-02
3.54021758e-01 -3.36281598e-01 -1.16652405e+00 1.87097371e+00
5.71122050e-01 8.38064328e-02 3.64921629e-01 7.81471789e-01
8.38099957e-01 4.94459212e-01 3.19783837e-01 -1.11469693e-01
1.67531180e+00 -1.33768559e+00 -7.59289801e-01 -5.63963592e-01
7.68934369e-01 -8.51702631e-01 1.44739735e+00 1.41737908e-01
-6.88721359e-01 -5.00180006e-01 -1.12899303e+00 4.19754051e-02
-6.43450856e-01 1.06527992e-01 5.91337025e-01 6.42890096e-01
-3.31891268e-01 2.84377113e-02 -8.51048887e-01 -1.48848847e-01
6.04406357e-01 6.13426007e-02 -3.84637564e-01 -2.46956080e-01
-1.29871523e+00 5.18572688e-01 3.06786537e-01 -1.84834138e-01
-7.92559087e-01 -9.37509775e-01 -1.24986494e+00 -1.06064558e-01
4.76448774e-01 -9.69088197e-01 1.36257100e+00 -1.31949699e+00
-1.67087817e+00 1.15858412e+00 -1.10067986e-01 -2.74035633e-01
1.13824315e-01 -5.04578054e-01 -6.34078860e-01 -3.06523085e-01
2.72564739e-01 2.45491192e-01 1.00664008e+00 -1.01703501e+00
-7.77510166e-01 -7.05779016e-01 3.54225129e-01 3.08061004e-01
-5.72849810e-01 8.25968906e-02 -4.76814568e-01 -1.00304914e+00
-2.21173108e-01 -9.37216759e-01 -3.54229987e-01 -3.14765126e-01
-3.32683235e-01 -3.24202925e-01 9.06849265e-01 -6.26640022e-01
1.05678570e+00 -2.26542926e+00 1.14747986e-01 -1.43364385e-01
-1.86137304e-01 4.79454637e-01 -4.75168258e-01 2.65393257e-01
-1.27166376e-01 -2.48642668e-01 -6.03077292e-01 -4.64580059e-01
2.62619983e-02 9.52469930e-02 -3.06346506e-01 1.91541806e-01
4.79599744e-01 7.61722505e-01 -8.62618208e-01 -4.30202603e-01
-7.98113123e-02 3.97426575e-01 -5.75204372e-01 6.07071996e-01
-5.08537948e-01 4.23538864e-01 -8.76673996e-01 9.09823239e-01
7.09493697e-01 -6.29191697e-02 1.76217169e-01 -2.66396284e-01
3.47896993e-01 6.26570344e-01 -9.39162493e-01 2.15743589e+00
-1.03421640e+00 -5.33826133e-05 1.27120480e-01 -1.20130324e+00
1.03750134e+00 3.17101479e-01 4.46601093e-01 -6.75855458e-01
1.05415456e-01 1.17612429e-01 -1.32397622e-01 -3.42909425e-01
1.16849221e-01 -4.12590772e-01 -6.21120811e-01 3.71190220e-01
4.56054628e-01 -4.82594855e-02 5.42424023e-02 1.41361337e-02
1.08250654e+00 5.74649692e-01 5.81240118e-01 -9.39264819e-02
8.31866980e-01 2.02125549e-01 9.63049054e-01 2.31816188e-01
-1.90660298e-01 7.10648775e-01 3.01167399e-01 -4.78101075e-01
-9.21670556e-01 -8.09155285e-01 -1.56515297e-02 1.55277085e+00
1.33843105e-02 -4.26542759e-01 -8.15524101e-01 -1.48099315e+00
-1.17093042e-01 7.92920768e-01 -5.78341305e-01 -3.42941314e-01
-4.82137799e-01 -1.01610613e+00 3.66777956e-01 6.70908093e-01
5.25168955e-01 -1.27288342e+00 5.62056750e-02 3.49498928e-01
-1.63463071e-01 -1.42623425e+00 -6.80237710e-01 3.07677150e-01
-7.40277588e-01 -1.07587612e+00 -7.10155725e-01 -9.04285789e-01
5.48889518e-01 1.49780080e-01 1.51502562e+00 -4.87882584e-01
-9.36050415e-02 3.77371371e-01 -6.62651420e-01 -5.72242796e-01
-2.10031912e-01 4.59810942e-01 -9.29053500e-02 2.36379564e-01
8.52778912e-01 -7.66072035e-01 -4.61069793e-01 3.97786528e-01
-8.55382264e-01 -1.43080711e-01 7.92569995e-01 1.09442484e+00
8.69173825e-01 -3.19362998e-01 9.65362430e-01 -1.65032005e+00
6.86519444e-01 -3.99081022e-01 -5.07235527e-01 2.62452245e-01
-5.79558492e-01 1.78620622e-01 8.39699030e-01 -4.25498933e-01
-1.49463367e+00 2.32935548e-01 -2.70467848e-01 -2.09671021e-01
-2.44840980e-01 5.84807336e-01 -9.16147053e-01 3.23818803e-01
5.19935966e-01 3.90845001e-01 -1.32691905e-01 -5.66425741e-01
5.75421810e-01 6.91844404e-01 1.54818162e-01 -7.61964083e-01
6.98261738e-01 3.66875440e-01 -4.12344933e-01 -4.62546647e-01
-1.49116802e+00 -5.99614859e-01 -6.68044925e-01 3.42094481e-01
8.26925039e-01 -1.39755023e+00 3.42472754e-02 5.07272363e-01
-9.84879673e-01 -2.67132577e-02 -5.38051724e-01 3.36430997e-01
-6.48734927e-01 2.24268660e-01 -3.70720387e-01 -4.29853112e-01
-7.35623837e-01 -1.05060899e+00 1.10929453e+00 1.66640535e-01
-1.09972544e-01 -9.92094040e-01 3.85624051e-01 5.92218876e-01
2.62104243e-01 4.68038879e-02 1.10725045e+00 -1.28500068e+00
-1.66579276e-01 -4.59936410e-01 -6.73351064e-03 6.99735165e-01
3.26681793e-01 -5.11911988e-01 -1.13467050e+00 -1.82963610e-01
6.85434490e-02 -4.88746941e-01 7.12954342e-01 9.42680687e-02
9.88604784e-01 -1.11173742e-01 -1.56989917e-01 7.27823734e-01
1.22780991e+00 -4.87763397e-02 4.45869118e-01 4.90396142e-01
7.66823471e-01 5.06348193e-01 1.19963300e+00 5.84553897e-01
4.71148014e-01 5.88758290e-01 1.61063030e-01 -3.74448150e-02
-3.57961982e-01 -3.59678775e-01 4.40657467e-01 1.09322846e+00
1.40051737e-01 -1.59519643e-01 -6.19088054e-01 9.34307754e-01
-1.81686604e+00 -4.21754777e-01 2.98809260e-01 1.88478470e+00
1.10158670e+00 3.52039039e-01 2.98352867e-01 -9.95952114e-02
5.95416844e-01 4.38744485e-01 -7.19003201e-01 -4.14616913e-01
-8.20540115e-02 2.50503421e-01 1.72948003e-01 5.14275357e-02
-1.40900958e+00 1.12476122e+00 5.05453157e+00 1.11627436e+00
-8.31982076e-01 4.62596923e-01 3.89560640e-01 2.01810920e-03
-4.44333196e-01 -8.71385932e-02 -9.27233338e-01 4.21087503e-01
8.67566824e-01 -6.63403794e-02 6.44690171e-02 1.31189287e+00
-2.76171297e-01 6.64937377e-01 -1.00052047e+00 7.08323479e-01
1.60884261e-01 -7.14562953e-01 2.22188771e-01 -1.98814079e-01
8.53107095e-01 -5.22503518e-02 1.08904310e-01 9.70579147e-01
4.96545911e-01 -5.45273542e-01 2.31148735e-01 6.28257468e-02
7.64681995e-01 -9.09061909e-01 9.09650266e-01 1.55771270e-01
-1.20723188e+00 2.31018856e-01 -3.50983173e-01 1.65924523e-02
1.61058903e-01 8.09753835e-01 -7.08667994e-01 6.99850440e-01
4.65437800e-01 7.14832067e-01 -2.39972308e-01 5.34467041e-01
-4.59566146e-01 9.18287992e-01 -1.28217727e-01 1.92541331e-02
3.16916704e-01 -1.11377604e-01 4.49144840e-01 1.30362654e+00
3.12028434e-02 -1.24774732e-01 2.89942265e-01 3.52134407e-01
-3.34481329e-01 4.26870376e-01 -6.03941023e-01 -2.34212562e-01
2.71574646e-01 1.37723577e+00 -3.22048515e-01 -2.77101010e-01
-1.10613120e+00 1.13482130e+00 4.38942462e-01 3.07820410e-01
-8.51546228e-01 -5.19256175e-01 1.09718609e+00 1.08738523e-02
6.80623531e-01 7.24673346e-02 -4.48746622e-01 -1.47308528e+00
2.42315680e-01 -1.29927468e+00 6.81547344e-01 -2.02474222e-01
-1.92551172e+00 7.49108553e-01 -3.10339689e-01 -1.44600141e+00
-1.76231787e-01 -5.91605306e-01 -5.62713623e-01 8.56816411e-01
-1.82221580e+00 -1.75310063e+00 3.90781350e-02 7.46438205e-01
9.94939148e-01 -6.80223644e-01 1.10925210e+00 4.13361996e-01
-4.47226077e-01 1.05556214e+00 4.63579688e-03 3.30096275e-01
1.03314257e+00 -1.12557280e+00 6.26630127e-01 6.92207694e-01
2.01527461e-01 4.27218199e-01 3.57880980e-01 -4.45407093e-01
-1.28751814e+00 -1.30857158e+00 8.02878082e-01 -5.31698823e-01
5.19708037e-01 -5.16549766e-01 -9.78515327e-01 7.82825112e-01
1.96988285e-01 4.28288907e-01 1.11817932e+00 7.40924418e-01
-9.48067367e-01 -2.65448928e-01 -1.14717972e+00 3.88508648e-01
1.19892836e+00 -5.75810492e-01 -8.77548516e-01 3.53892535e-01
1.11143577e+00 -1.85883567e-01 -7.02281475e-01 7.27982700e-01
3.35155815e-01 -5.71044683e-01 1.02002084e+00 -1.11039412e+00
4.64024216e-01 -1.84857741e-01 -1.92330256e-01 -1.55761456e+00
-1.73515096e-01 -1.96083069e-01 7.92523753e-03 1.65078223e+00
7.30294526e-01 -5.61380625e-01 7.50286937e-01 4.16334718e-01
-3.92239839e-01 -7.96687305e-01 -8.76482546e-01 -7.65166044e-01
1.58557594e-01 -2.88960278e-01 7.68522024e-01 1.12864316e+00
-1.84620410e-01 9.46176767e-01 -3.22385252e-01 1.67547360e-01
4.43347216e-01 4.92031157e-01 8.15437317e-01 -9.35123026e-01
-4.17027324e-01 -8.24636891e-02 -3.03138196e-01 -8.32580984e-01
5.96114218e-01 -9.73629415e-01 -3.72044323e-03 -1.25066769e+00
1.98399290e-01 -3.48512858e-01 -6.65670872e-01 5.54839492e-01
-5.64575553e-01 -1.09204203e-02 -3.71312015e-02 -1.50069103e-01
-8.76835346e-01 9.47069645e-01 1.16351545e+00 -3.28714311e-01
-1.10177010e-01 2.34349489e-01 -9.08829570e-01 8.50841105e-01
7.06352353e-01 -7.35541940e-01 -5.70737004e-01 -5.23275197e-01
5.69007546e-02 -2.08927929e-01 -3.55708003e-01 -6.94130182e-01
-2.17478260e-01 -8.45298916e-02 3.89295891e-02 -4.26136017e-01
3.94894749e-01 -1.00681090e+00 -6.51970565e-01 1.09038071e-03
-3.42947483e-01 -1.93271011e-01 3.87846887e-01 8.95380795e-01
-6.02428555e-01 -2.24486932e-01 6.56017244e-01 -1.78329766e-01
-8.30040216e-01 4.50960785e-01 -2.13257223e-02 7.01155305e-01
8.86528969e-01 2.36602530e-01 1.76363438e-02 -6.81010783e-02
-6.43057525e-01 2.86211092e-02 2.79414207e-01 5.82700133e-01
3.67289603e-01 -1.36353898e+00 -8.50663722e-01 3.23316097e-01
7.68593967e-01 8.29521120e-02 3.79276127e-01 3.41753215e-01
2.77981833e-02 1.58042833e-01 -1.13535888e-01 -3.64806861e-01
-1.08993280e+00 8.05303752e-01 2.86826435e-02 -8.48436952e-01
-3.37922633e-01 9.92604256e-01 5.76792717e-01 -1.16383791e+00
2.06410252e-02 7.17874244e-02 -1.64471388e-01 1.05925202e-01
4.26829189e-01 -1.84771597e-01 4.12141442e-01 -3.74034107e-01
-5.90939641e-01 4.69441831e-01 -5.39330065e-01 1.78389512e-02
1.42895496e+00 -1.65564731e-01 3.09274852e-01 3.13826174e-01
1.19956350e+00 1.54457420e-01 -1.17785490e+00 -7.78679013e-01
-1.56217441e-01 -3.81191880e-01 -1.60628825e-01 -1.17795944e+00
-1.14714539e+00 1.01568186e+00 3.67059767e-01 -1.88385710e-01
1.30951130e+00 -1.20311081e-01 1.15316153e+00 3.82604361e-01
2.57016510e-01 -1.04265201e+00 1.24626592e-01 6.32253230e-01
6.27229095e-01 -1.42400944e+00 -1.08431920e-01 -5.56328118e-01
-1.14650273e+00 5.93226731e-01 8.79584551e-01 -2.80295700e-01
7.98599958e-01 2.78477460e-01 4.01473880e-01 8.52209106e-02
-7.49221861e-01 -3.90226185e-01 2.33397231e-01 8.83312762e-01
5.78894019e-01 -1.57928150e-02 -4.10953492e-01 1.46006095e+00
1.21660508e-01 -1.18921392e-01 -1.08893551e-01 1.06105971e+00
-9.90834832e-02 -1.51576746e+00 1.15157962e-01 2.66127914e-01
-7.46903300e-01 -4.40193772e-01 -2.53182858e-01 7.35547841e-01
-1.16234139e-01 7.38967896e-01 -3.05592597e-01 -2.10074246e-01
7.86280811e-01 5.04048109e-01 3.51456940e-01 -8.45494688e-01
-6.23282135e-01 -3.83039122e-04 4.97962058e-01 -3.97666931e-01
-5.07507384e-01 -5.71812987e-01 -9.63554740e-01 2.82468319e-01
-1.75511181e-01 7.39364177e-02 6.11203074e-01 1.03269517e+00
6.35339975e-01 7.48740017e-01 7.20291853e-01 -3.28642011e-01
-6.62681639e-01 -1.21019042e+00 -4.92352307e-01 8.24773490e-01
7.98474178e-02 -5.52660465e-01 -2.50908792e-01 1.20488890e-01] | [11.428532600402832, 6.660813808441162] |
5b6a9969-fda9-44b7-890a-95359cb19945 | decomposed-soft-prompt-guided-fusion | 2211.10681 | null | https://arxiv.org/abs/2211.10681v1 | https://arxiv.org/pdf/2211.10681v1.pdf | Decomposed Soft Prompt Guided Fusion Enhancing for Compositional Zero-Shot Learning | Compositional Zero-Shot Learning (CZSL) aims to recognize novel concepts formed by known states and objects during training. Existing methods either learn the combined state-object representation, challenging the generalization of unseen compositions, or design two classifiers to identify state and object separately from image features, ignoring the intrinsic relationship between them. To jointly eliminate the above issues and construct a more robust CZSL system, we propose a novel framework termed Decomposed Fusion with Soft Prompt (DFSP)1, by involving vision-language models (VLMs) for unseen composition recognition. Specifically, DFSP constructs a vector combination of learnable soft prompts with state and object to establish the joint representation of them. In addition, a cross-modal decomposed fusion module is designed between the language and image branches, which decomposes state and object among language features instead of image features. Notably, being fused with the decomposed features, the image features can be more expressive for learning the relationship with states and objects, respectively, to improve the response of unseen compositions in the pair space, hence narrowing the domain gap between seen and unseen sets. Experimental results on three challenging benchmarks demonstrate that our approach significantly outperforms other state-of-the-art methods by large margins. | ['Jingcai Guo', 'Song Guo', 'Ziming Liu', 'Xiaocheng Lu'] | 2022-11-19 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Lu_Decomposed_Soft_Prompt_Guided_Fusion_Enhancing_for_Compositional_Zero-Shot_Learning_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Lu_Decomposed_Soft_Prompt_Guided_Fusion_Enhancing_for_Compositional_Zero-Shot_Learning_CVPR_2023_paper.pdf | cvpr-2023-1 | ['compositional-zero-shot-learning', 'novel-concepts'] | ['computer-vision', 'reasoning'] | [ 0.39340416 -0.2747917 -0.30217898 -0.21143156 -0.72625786 -0.42651564
0.8585403 -0.09914853 -0.18289852 0.38978693 0.11810937 0.27652922
0.14710878 -0.5101245 -0.8160197 -1.1496173 0.38622752 0.22756033
0.39261192 -0.06402393 -0.11870778 -0.03272313 -1.8868563 0.5420902
0.6560556 1.4747643 0.3648602 0.2958414 -0.3888543 1.3199774
-0.21312797 -0.10213749 0.17899497 -0.5476846 -0.44102472 0.63782686
0.5573154 -0.4444897 -0.6748439 1.3005928 0.22107488 0.35152423
0.5343808 -1.4388517 -1.0568789 0.70339745 -0.27853915 0.1301912
0.4427487 0.59431344 1.0036633 -1.0983139 0.50345105 1.4086074
0.04634485 0.92824084 -1.3764656 -1.004628 0.765123 0.5982899
-1.3966974 -0.7668393 1.0456425 -0.5587819 0.68737924 0.26262814
0.5109557 1.594306 -0.39029178 1.3917202 1.0436747 -0.2332399
0.4240041 0.29675356 0.3681704 0.9176776 0.05034718 0.1338081
-0.75781643 -0.1867121 0.43071198 0.4708206 -0.49950954 -0.7884945
-1.4697675 0.4795908 0.30793473 0.23990217 -0.32245448 -0.05755591
0.38312835 0.3116115 0.14189321 0.05744703 -0.38255852 0.23468927
-0.4873083 -0.07778919 0.5325558 1.10033 1.0045443 0.2851511
-0.53381974 0.57527524 0.2113335 0.53702414 0.8500669 -0.45313567
0.36058977 0.9695198 -0.18611708 -0.48274878 0.14238517 -0.32703146
-0.9076385 -0.21330535 -0.04826995 0.1325717 -1.2481253 1.9397478
0.31764 0.88051164 0.48416248 0.9811319 1.0721382 0.9303144
0.06426996 -0.41354528 1.3057641 -1.0239881 -0.76755947 -0.4130678
0.21969138 -0.24198145 0.8953492 0.27983972 -0.86602247 -0.7981657
-1.0728209 0.13326739 -0.33985883 0.01161309 0.43993604 0.06161653
-0.6483012 0.26347604 -0.9033856 -0.10049359 0.27002555 0.28295502
-0.21697989 -0.25440305 -1.1395526 0.51278156 0.83939236 -0.05650093
-1.5154226 -0.5908357 -1.3020833 0.07010645 0.9539892 -0.63598096
1.2343825 -1.2448375 -1.4515088 0.49241295 -0.08463964 -0.15356891
0.17282209 0.17568837 -0.66794693 0.1159651 -0.02319021 0.44831973
1.3827022 -1.6742208 -0.86538756 -0.41328344 -0.13479742 0.11071348
-0.5442021 -0.12305852 -0.56528074 -0.40126762 0.2714036 -0.61248696
0.01828136 0.02233827 -0.28319687 -0.5152781 1.0506425 -0.3420496
0.9433546 -2.4890804 0.52888536 -0.14150539 0.28263015 0.2862144
-0.5022887 0.08099453 -0.18839592 -0.5075397 -0.10734898 -0.3515282
0.04590485 0.5678786 -0.73429525 0.4039756 0.56183034 1.1145995
-1.2481548 -0.4567916 0.21688275 0.26924598 -0.11309806 0.46895853
-0.3337052 0.38243023 -0.5298353 0.887043 0.37418267 -0.40034485
0.13918501 -0.36934313 0.09572575 -0.13711073 -1.3299807 1.8425077
-0.13195063 -0.09148046 0.09648079 -1.3145139 0.86083055 0.43747866
0.4038059 -0.5394521 0.13727495 -0.07615929 -0.05288108 -0.7399651
-0.01238581 -0.31209132 -0.17296566 0.13603236 0.8160388 0.24934335
0.1123565 0.2553108 0.9123073 0.02955729 0.23607188 0.03052892
0.8466847 -0.41697612 1.0014611 0.65738964 -0.5936261 0.25286096
0.13143988 -0.42216167 -0.6269364 -1.3791789 0.20644198 1.2493408
0.75813067 -0.32370952 -0.26960245 -0.8901576 0.01873103 0.6505634
-0.65992033 -0.6778817 -0.36140722 -0.41400194 0.1480256 0.66371506
0.35322687 -1.035707 -0.23162127 0.11002948 -0.12812623 -1.3381227
-0.5551025 0.48180246 -0.43843457 -0.980914 -0.56704247 -1.184764
0.7508235 0.39914274 0.64640176 -0.08426785 -0.20313947 0.6939897
-0.4660074 -0.04533778 -0.45420548 -0.50384897 0.40376875 0.8743205
0.3689191 -0.34666014 -0.30348426 0.12451544 -1.1005472 0.23859304
0.76082444 1.1763594 0.5171319 0.04728098 0.47190642 -0.39450443
0.131431 -0.5473324 -0.42248386 0.72116107 -0.49124452 0.3267126
0.66862446 -1.0739268 -1.1667516 0.36880437 0.5293963 -1.2918515
-0.27015007 0.16103253 -0.43328264 0.09694223 0.32032463 1.0321978
0.12422183 -0.4588707 0.44342285 0.55553305 0.8139689 -0.73620504
1.0366766 0.5184514 -0.37606192 -0.5830171 -0.9994853 -0.7754229
-0.4474641 -0.15453245 0.87616926 -1.3072109 -0.5318159 0.57636
-0.9078869 0.05169773 -0.5333404 0.3330817 -0.47534022 0.49102417
-0.708647 -1.0678695 -0.13693933 -1.3595735 1.2282556 0.34580988
0.14567095 -0.5810773 -0.17869498 0.40505582 -0.0070444 -0.05680874
0.9658717 -0.8696492 -0.76581645 -0.13424669 -0.22557111 0.62607974
0.18575118 -0.2341071 -1.0462353 -0.50807184 0.30409175 -0.7556946
1.1924969 -0.24839373 1.152593 -0.4153267 -0.30574262 0.30425397
1.2469434 0.3300446 0.1613145 -0.15588367 0.8521102 0.4452546
0.38149515 0.47363654 0.4567303 0.4513801 0.3816845 0.27386898
-0.2051553 -0.49658632 0.8900822 1.1856369 0.17904358 0.03515918
-0.5196375 0.44237792 -2.1382334 -0.8974614 0.5074108 1.8628733
0.8712777 -0.09452356 -0.15572146 -0.02576631 0.7827027 0.31805113
-0.96409756 0.40654194 -0.28235775 -0.01482372 0.05978302 0.03599252
-1.1152667 0.9175322 5.1014752 0.9329406 -1.1679806 0.19373599
0.12745358 -0.21128894 -0.25963682 0.14399093 -0.8603328 0.4981599
0.41455874 -0.14679109 0.67523426 0.88249224 -0.38500664 0.3612024
-1.4678286 1.1674815 0.6058456 -1.081025 0.44825193 -0.20364745
0.93572176 -0.1380863 0.15223034 0.83946836 0.32259837 -0.3454304
0.97330415 0.73221165 0.50641674 -0.1751665 0.38075173 0.6994322
-1.342108 -0.60583955 -0.22194727 0.02095257 -0.10943444 0.01960096
-0.40053225 0.66391873 0.47598296 0.97577417 -0.46946907 0.55910015
-0.14212456 0.4133421 0.07891235 0.05780926 0.2399674 -0.10487404
0.7438015 0.6831606 0.05770357 0.19198184 0.8646696 1.2196366
0.01665966 -0.1854392 -0.31789708 -0.32008722 0.13078666 1.2657899
-0.30746722 -0.8586331 -0.68807006 1.0197973 0.51054806 0.68347
-0.70429206 0.05941363 0.837255 -0.3810431 0.5668379 0.02104174
0.46825197 -1.9742641 0.13094659 -0.9691303 0.6109721 -0.58984804
-1.6708703 0.46489304 -0.07912446 -1.4296604 0.0422466 -0.6415962
-0.57356507 0.5202616 -1.5870183 -1.5182405 -0.406942 0.8898933
0.8963524 -0.42469788 0.7205963 0.1771166 -1.0237944 0.488152
-0.09877332 0.18823843 0.624962 -0.9889221 -0.12836115 0.93217856
0.33103907 0.5834421 0.36840189 -0.71869457 -2.0337293 -1.2512131
0.18550105 -0.413262 0.8507759 -0.64977384 -1.1666608 0.6526119
-0.02631 0.48111346 0.57673806 -0.21695487 -0.8437375 -0.30208915
-0.6387377 0.5417885 1.019456 -0.9291957 -0.9503107 0.15497872
1.0246103 -0.1936557 -0.60563517 0.6760011 0.35589784 -0.5616577
0.87600446 -0.96808904 0.19092382 -0.601673 -0.4443559 -1.1952819
-0.5860554 -0.34490335 -0.7822877 1.3194685 0.11017945 -0.5084246
0.4713894 0.5890466 -0.17083721 -0.678552 -0.8347961 -1.0913187
-0.40186575 -0.30766767 0.5723522 1.052344 0.07932778 0.61555827
-0.46302593 0.2803023 0.7045485 0.4069832 0.61185336 -1.0151443
-0.47176015 -0.33664653 -0.62314457 -1.0234158 0.655357 -1.0457457
0.38229984 -1.0238944 0.70910424 -0.04305978 -0.9540713 0.75809056
-0.46414584 -0.15321083 0.36268348 0.29110166 -1.1059074 1.0892063
1.4195569 -0.8167173 -0.02680495 -0.31797308 -0.68280125 0.65271705
0.13038749 -0.15086448 -0.53679293 -0.20174745 -0.39234656 -0.08515301
0.654065 -0.954614 0.5174423 -0.3853901 0.46620804 -0.49844489
0.58656085 -0.95200646 0.04370663 0.43583676 -0.45263496 -0.6784118
-0.10912574 1.0603566 -0.31465074 -0.07843306 0.6201969 -0.18596393
-1.2946737 0.7042105 0.05194945 -0.2418777 1.4188976 -0.06872113
-0.46110034 -0.06551676 -0.75376606 0.5472823 0.27352783 0.91129386
0.78118235 -1.5360566 -0.46419534 0.62634206 0.8151489 0.1142952
0.6419534 0.77990645 0.37133396 0.15543537 -0.0505692 -0.7107853
-1.0497323 1.2023857 0.14374961 0.04365645 -0.671787 0.9470059
0.9059104 -0.27340642 0.45916823 -0.3941073 0.06769224 0.18808557
0.88451725 0.04007761 -0.3910262 -0.81070817 -0.34915736 0.38309726
-0.3533369 0.1438106 1.1257689 -0.16924179 -0.03708404 0.90612125
1.2276462 -0.6392113 -1.6163728 -1.0895058 -0.10113901 -0.2361362
0.08402127 -0.62608033 -1.0438509 0.7731729 0.7192171 0.0811376
1.1327538 0.41650113 0.9183114 0.44776616 0.2902314 -0.8522341
0.5009512 0.29603514 0.7770219 -1.5383431 -0.4247245 -0.20604098
-0.7804188 0.9172708 1.032243 -0.02371893 0.43491882 0.12496029
-0.12133543 -0.1740258 -1.1490211 -0.51461154 0.6043304 0.46527314
-0.3846808 0.16950375 0.41846603 1.1883484 0.5228616 -0.07763836
0.00955856 1.0046443 -0.4668168 -0.8632022 -0.22053558 0.39534286
0.2100338 0.07312654 -0.3145543 0.21809077 0.39789206 0.6934293
0.14652218 -0.73265606 0.23362538 0.35042715 0.4376482 -0.9089771
-0.1681471 0.05440728 -0.3157565 -0.55379736 -0.35053495 -0.6485198
-1.1078887 0.4096552 -0.57683736 -0.09089693 0.24576685 1.277072
0.2468665 0.6025692 0.7893171 -0.77729267 -1.1195098 -0.7903109
-0.75474733 0.69236004 0.7155486 -0.8970561 -0.48202893 0.2598355 ] | [10.164388656616211, 2.2918319702148438] |
3c5f3fd3-50cd-4352-b2ab-03345f604b6a | pointnorm-normalization-is-all-you-need-for | 2207.06324 | null | https://arxiv.org/abs/2207.06324v4 | https://arxiv.org/pdf/2207.06324v4.pdf | PointNorm: Dual Normalization is All You Need for Point Cloud Analysis | Point cloud analysis is challenging due to the irregularity of the point cloud data structure. Existing works typically employ the ad-hoc sampling-grouping operation of PointNet++, followed by sophisticated local and/or global feature extractors for leveraging the 3D geometry of the point cloud. Unfortunately, the sampling-grouping operations do not address the point cloud's irregularity, whereas the intricate local and/or global feature extractors led to poor computational efficiency. In this paper, we introduce a novel DualNorm module after the sampling-grouping operation to effectively and efficiently address the irregularity issue. The DualNorm module consists of Point Normalization, which normalizes the grouped points to the sampled points, and Reverse Point Normalization, which normalizes the sampled points to the grouped points. The proposed framework, PointNorm, utilizes local mean and global standard deviation to benefit from both local and global features while maintaining a faithful inference speed. Experiments show that we achieved excellent accuracy and efficiency on ModelNet40 classification, ScanObjectNN classification, ShapeNetPart Part Segmentation, and S3DIS Semantic Segmentation. Code is available at https://github.com/ShenZheng2000/PointNorm-for-Point-Cloud-Analysis. | ['Gaurav Gupta', 'Changjie Lu', 'Jinqian Pan', 'Shen Zheng'] | 2022-07-13 | null | null | null | null | ['point-cloud-classification'] | ['computer-vision'] | [-2.35745877e-01 -4.13285404e-01 -9.47017514e-04 -5.00287414e-01
-6.20935977e-01 -4.80618358e-01 4.57978278e-01 3.77661824e-01
-2.35266626e-01 1.70272931e-01 -4.45218891e-01 -2.57221073e-01
-2.68582493e-01 -1.09915650e+00 -7.87980199e-01 -6.45041704e-01
6.52732179e-02 6.20223999e-01 5.15385449e-01 -2.09320728e-02
5.74734330e-01 1.14349413e+00 -1.73245776e+00 -2.50694931e-01
1.02906990e+00 1.41166091e+00 1.66412324e-01 4.01821256e-01
-5.90672910e-01 -2.26204589e-01 -3.26576054e-01 -8.84767342e-03
6.02079630e-01 7.24666566e-02 -4.69475269e-01 4.04647971e-03
3.84975404e-01 -3.35393578e-01 9.65784192e-02 1.28341472e+00
4.26025450e-01 2.15975165e-01 5.38629591e-01 -1.38394427e+00
-2.71288633e-01 -9.00310576e-02 -7.85303295e-01 3.23252603e-02
4.11414765e-02 2.38577411e-01 8.55083287e-01 -9.20761943e-01
2.82833695e-01 1.20983326e+00 8.46109509e-01 1.21425826e-03
-8.50843132e-01 -8.69520426e-01 -1.64599672e-01 4.18168083e-02
-1.68379176e+00 -1.60549492e-01 9.09777403e-01 -3.73382419e-01
6.83434963e-01 3.75623763e-01 7.80405223e-01 2.03845933e-01
4.25345674e-02 4.01230007e-01 5.70127428e-01 -6.66572899e-02
4.16218489e-01 -3.72017324e-01 1.73372105e-01 4.72003907e-01
1.94925979e-01 9.60741751e-03 -2.50738617e-02 -4.37992215e-01
1.22480404e+00 4.60981041e-01 4.26815860e-02 -4.48452622e-01
-1.12804317e+00 8.00270319e-01 7.53153980e-01 9.01027694e-02
-5.73771060e-01 1.00319847e-01 3.21336389e-01 -3.80023532e-02
5.32994390e-01 1.42053649e-01 -5.15789211e-01 -6.93492666e-02
-1.20702839e+00 4.17307913e-01 3.83518279e-01 1.29165602e+00
1.32374597e+00 -1.49838358e-01 2.75762916e-01 7.72403300e-01
4.30681407e-01 7.44846880e-01 2.42425948e-01 -1.02255690e+00
4.84296173e-01 9.89977479e-01 -1.94044694e-01 -1.39231431e+00
-4.25674170e-01 -3.25287968e-01 -9.46962416e-01 3.21117371e-01
9.58136469e-02 2.19792798e-01 -1.08258426e+00 1.09141779e+00
7.52020717e-01 1.94861144e-01 -4.07790542e-01 9.16798055e-01
8.16618860e-01 5.12998283e-01 4.97442763e-03 1.77301645e-01
1.31324363e+00 -5.57693958e-01 -3.36446613e-01 1.65626824e-01
3.52764636e-01 -9.99915063e-01 1.10705256e+00 9.41873342e-02
-1.10226357e+00 -5.05366981e-01 -1.04762101e+00 -2.40047306e-01
-2.80779630e-01 -1.08253479e-01 5.88047326e-01 1.97086409e-01
-6.59253895e-01 6.00234926e-01 -1.05379665e+00 -2.79576749e-01
7.43116200e-01 5.01965702e-01 -2.25159109e-01 8.19636434e-02
-6.70078933e-01 3.97096127e-01 4.26183462e-01 2.14710236e-01
-1.15573443e-01 -1.02416003e+00 -6.93823814e-01 6.09142818e-02
3.59200865e-01 -6.85106874e-01 1.09804630e+00 -4.54123884e-01
-1.24694061e+00 5.88465035e-01 -2.01162249e-01 -1.80047020e-01
4.35514122e-01 9.24380496e-02 -9.13492069e-02 2.79152900e-01
2.96785057e-01 5.49974382e-01 6.61026895e-01 -1.16649580e+00
-6.86709762e-01 -8.45385730e-01 -2.53897220e-01 2.21956193e-01
8.23092833e-02 -2.50606179e-01 -6.61916196e-01 -5.38727701e-01
9.32792425e-01 -6.22892499e-01 -3.15835983e-01 2.90707409e-01
-3.60730916e-01 -3.11029375e-01 1.13582134e+00 -4.20795798e-01
1.00452888e+00 -2.39908051e+00 -4.83453602e-01 7.75986910e-01
2.58493632e-01 1.63621843e-01 1.64951414e-01 1.83063090e-01
-1.70910433e-02 9.45323110e-02 -5.48717082e-01 -1.93572983e-01
-1.07191749e-01 2.91160107e-01 -7.07503110e-02 7.10653663e-01
1.21050969e-01 6.74075067e-01 -7.39972651e-01 -7.10590184e-01
6.04668915e-01 3.80657375e-01 -6.63848698e-01 -1.17429867e-01
-1.80296004e-01 2.34832093e-01 -7.67392933e-01 1.10956359e+00
1.33546114e+00 1.34697836e-02 -5.68245888e-01 -3.67749333e-01
-2.42650598e-01 6.52545616e-02 -1.50628245e+00 1.69439626e+00
-2.57856071e-01 8.80142748e-02 3.33766103e-01 -6.97436512e-01
1.30125451e+00 6.85790032e-02 8.82563770e-01 -5.57748735e-01
3.24708045e-01 5.15174210e-01 -1.64698824e-01 -2.27235273e-01
4.75772262e-01 -1.26513436e-01 1.61318555e-01 7.44480714e-02
-3.09367478e-01 -5.87142229e-01 -1.32193834e-01 -8.56181458e-02
7.41278172e-01 1.51951984e-01 3.69402736e-01 -3.26603413e-01
5.76019943e-01 3.40273410e-01 6.11231029e-01 3.48074466e-01
-7.33449981e-02 9.28942084e-01 2.44782999e-01 -3.48613143e-01
-1.07009089e+00 -1.05234182e+00 -4.47919726e-01 5.26732326e-01
4.90297616e-01 -3.73321235e-01 -6.67319357e-01 -3.86190474e-01
3.55168432e-01 5.10651827e-01 -2.28832513e-01 1.09759673e-01
-5.99346578e-01 -4.45258856e-01 3.91732037e-01 6.01565659e-01
6.82532430e-01 -8.31401348e-01 -4.55689251e-01 1.68252483e-01
1.88840419e-01 -8.67189050e-01 -4.92462188e-01 -8.04122686e-02
-1.36488783e+00 -1.01014256e+00 -2.93674439e-01 -5.68076849e-01
8.98906469e-01 3.87168378e-01 6.35859549e-01 3.60937566e-01
-8.31697583e-02 3.28559801e-02 -2.24048167e-01 -4.96129692e-01
1.80120334e-01 1.36519626e-01 -1.63459927e-01 -1.92393497e-01
4.82676923e-01 -8.65338743e-01 -6.39859736e-01 5.37611604e-01
-9.49703336e-01 -1.34870976e-01 4.85105932e-01 4.33450133e-01
1.03402781e+00 1.62877798e-01 1.59985676e-01 -5.01495779e-01
3.04809272e-01 -4.33890641e-01 -7.87111700e-01 -1.99028164e-01
-3.32370877e-01 -3.36769462e-01 4.80026901e-01 -1.89075507e-02
-6.37245893e-01 1.43128559e-01 -4.79877412e-01 -6.88757002e-01
-3.35792571e-01 3.28232348e-01 -4.44179893e-01 -3.04970801e-01
3.90335441e-01 1.87203601e-01 3.25325847e-01 -6.83049142e-01
3.28825086e-01 7.21031070e-01 6.38872802e-01 -7.46075809e-01
1.00954437e+00 8.56672466e-01 2.03048944e-01 -1.01247883e+00
-3.55851620e-01 -8.52213204e-01 -9.04861033e-01 -2.03718524e-02
8.14090073e-01 -7.10227370e-01 -7.50408769e-01 4.40077841e-01
-1.28136361e+00 2.19478935e-01 -4.76379991e-01 4.17416036e-01
-4.58984554e-01 4.78747934e-01 -2.92217761e-01 -5.08957863e-01
-4.96370167e-01 -1.20896304e+00 1.18962383e+00 2.38886431e-01
4.63433824e-02 -7.77416229e-01 -2.58386970e-01 1.13042809e-01
2.65423208e-01 4.82207477e-01 7.49123037e-01 -5.72990298e-01
-8.55825663e-01 -4.85181719e-01 -4.60240781e-01 4.00505722e-01
1.64479837e-01 3.39288920e-01 -7.09187448e-01 4.00790162e-02
1.91433474e-01 4.01257455e-01 2.70915270e-01 4.62399453e-01
1.60017478e+00 -2.35055890e-02 -2.96714127e-01 1.06749356e+00
1.50457132e+00 1.41469315e-01 6.38246119e-01 4.36552554e-01
9.31635797e-01 2.64143556e-01 7.35431850e-01 4.73331630e-01
4.13163066e-01 4.80866760e-01 6.75932944e-01 -1.51106432e-01
7.51483813e-02 -1.61080256e-01 -3.65632862e-01 9.48524952e-01
-3.38296920e-01 2.95721918e-01 -1.22282302e+00 4.41337913e-01
-1.69302952e+00 -6.93629801e-01 -5.59076786e-01 2.11363149e+00
4.64845717e-01 1.77773908e-01 -4.91801575e-02 1.10122062e-01
8.43768477e-01 4.33985852e-02 -6.27317667e-01 -2.01713920e-01
1.09880641e-01 2.92466313e-01 8.14653039e-01 4.18798655e-01
-9.61295366e-01 7.77977288e-01 4.97450352e+00 1.17855394e+00
-1.14117503e+00 -2.06365418e-02 1.28145501e-01 8.54081064e-02
-1.76749185e-01 1.61352769e-01 -7.82313526e-01 6.12275004e-01
3.99599195e-01 -1.05831377e-01 2.25725114e-01 1.11975384e+00
3.02587837e-01 -1.24414722e-02 -7.17043281e-01 1.12363434e+00
-2.54392058e-01 -1.22591555e+00 2.60394607e-02 3.20748270e-01
4.09320384e-01 3.38763118e-01 -2.67544448e-01 7.85854552e-03
-2.63054043e-01 -6.78310692e-01 8.16599071e-01 5.58296800e-01
6.23276651e-01 -9.21904206e-01 6.42490089e-01 5.44062197e-01
-1.41745436e+00 2.22653612e-01 -5.83007157e-01 8.16669166e-02
3.19078974e-02 8.67697358e-01 -8.33895564e-01 8.14537168e-01
8.04707229e-01 7.15535879e-01 -5.12540758e-01 1.23000813e+00
1.73281923e-01 3.32017154e-01 -9.00998950e-01 2.02191517e-01
2.86406249e-01 -6.41404986e-01 7.70555437e-01 9.68745887e-01
5.06485879e-01 1.24176428e-01 1.96318522e-01 9.62179840e-01
1.52984917e-01 1.98309436e-01 -3.73184294e-01 3.30857903e-01
7.38911152e-01 1.32751536e+00 -1.02986693e+00 -3.41673374e-01
-2.10531726e-01 2.53345639e-01 -6.86147809e-02 1.27130419e-01
-7.18506575e-01 -6.78417146e-01 8.79339457e-01 4.19919997e-01
2.89233565e-01 -6.84093714e-01 -1.06924951e+00 -8.11248064e-01
2.98023909e-01 -4.85748082e-01 1.60029337e-01 -5.68527520e-01
-1.26784730e+00 4.14152116e-01 2.22038701e-01 -1.54884052e+00
1.24964684e-01 -3.92288834e-01 -9.47734177e-01 9.09568310e-01
-1.42783439e+00 -1.11359942e+00 -5.71499288e-01 6.22529864e-01
4.71948087e-01 6.75361082e-02 2.85093844e-01 4.32329297e-01
-3.63748521e-01 2.70517617e-01 8.96148533e-02 -4.90224063e-02
2.50836462e-01 -1.09166896e+00 4.49134767e-01 6.18329048e-01
-4.82830405e-01 8.55711877e-01 4.49199975e-01 -7.96888471e-01
-1.28343928e+00 -1.03121924e+00 4.51267064e-01 -1.40033454e-01
5.10904193e-01 -3.69220912e-01 -1.27840698e+00 5.71160078e-01
-3.60025555e-01 3.66830558e-01 2.54676580e-01 -1.70702040e-01
6.16251975e-02 -2.23542571e-01 -1.45983458e+00 4.58448857e-01
1.03401721e+00 -2.41668910e-01 -5.66001773e-01 2.45243892e-01
5.59856474e-01 -6.11168385e-01 -1.13097847e+00 7.03197300e-01
3.97092253e-01 -8.49111974e-01 1.23191130e+00 -5.30637726e-02
1.95148915e-01 -6.78949118e-01 -2.60753840e-01 -8.97107959e-01
-2.96301961e-01 -1.67302385e-01 1.40978441e-01 1.30033183e+00
9.40173939e-02 -8.65322530e-01 8.83816838e-01 4.88227129e-01
-5.10216475e-01 -9.37789261e-01 -1.16609967e+00 -7.46138453e-01
1.92182176e-02 -7.57203162e-01 1.29705477e+00 8.91221285e-01
-5.94205976e-01 -2.59930253e-01 4.76582795e-01 4.76177007e-01
8.28720510e-01 9.62190032e-02 9.88727033e-01 -1.45956933e+00
3.41158003e-01 -6.13342166e-01 -7.25354433e-01 -1.03253317e+00
-1.65444493e-01 -8.98729980e-01 -3.10982522e-02 -1.58339202e+00
-3.25143486e-01 -8.78276527e-01 9.69087183e-02 3.30913186e-01
1.79959517e-02 2.27611229e-01 1.11106485e-01 5.73326707e-01
-2.25949734e-01 6.67033315e-01 1.43519258e+00 1.51597559e-01
-3.92279685e-01 1.63303077e-01 -4.21710134e-01 1.05943930e+00
9.61418211e-01 -4.42858726e-01 -1.82150096e-01 -5.52720547e-01
-7.40826651e-02 -2.62849480e-01 5.55263937e-01 -1.22316611e+00
5.52841067e-01 -2.97929555e-01 3.85334671e-01 -1.25031805e+00
4.50865239e-01 -1.17048621e+00 1.60521686e-01 1.87383950e-01
3.71920466e-01 1.81273237e-01 1.79355174e-01 3.24808985e-01
-2.58582324e-01 -1.82211727e-01 8.50067735e-01 -1.67627156e-01
-5.00943422e-01 6.91317856e-01 2.64252424e-01 -3.25330675e-01
1.16668808e+00 -6.10885501e-01 -5.34198247e-02 1.84659883e-01
-5.45349002e-01 3.52465540e-01 7.78164089e-01 1.58852980e-01
8.09756756e-01 -1.49069142e+00 -5.38874686e-01 5.35244882e-01
-3.46980356e-02 8.49621058e-01 3.48937333e-01 9.53536153e-01
-1.08750701e+00 7.13843256e-02 -2.02997416e-01 -1.12320983e+00
-1.10840356e+00 3.00787270e-01 2.76827604e-01 3.72059405e-01
-8.14647615e-01 5.45304060e-01 -4.16187420e-02 -7.12447643e-01
-1.14689492e-01 -7.25466549e-01 2.05055609e-01 6.18015788e-03
3.48071679e-02 7.46038139e-01 3.90993655e-01 -7.07956493e-01
-5.19763827e-01 1.13396895e+00 1.19995616e-01 1.58944875e-01
1.25317097e+00 -4.03463505e-02 -3.80337656e-01 2.25387320e-01
1.30443227e+00 -1.49762174e-02 -9.87967849e-01 -1.96092814e-01
-8.77915770e-02 -8.09578538e-01 2.19526783e-01 -1.02765404e-01
-1.08196497e+00 8.55177104e-01 3.99363816e-01 2.25210756e-01
1.03646743e+00 1.71929058e-02 1.05035830e+00 1.17463343e-01
4.46977228e-01 -9.43380237e-01 -6.51012778e-01 5.50325751e-01
8.48780513e-01 -1.09502316e+00 3.20955724e-01 -7.56146073e-01
-1.27120554e-01 1.14916217e+00 5.26867568e-01 -4.06037092e-01
9.51865435e-01 1.16293237e-01 -1.98905859e-02 -5.43063641e-01
-1.46845973e-03 -3.46469916e-02 2.71511048e-01 3.37966979e-01
-2.81522460e-02 1.08370997e-01 -1.33185163e-01 4.19843942e-01
-5.97090840e-01 -1.32201701e-01 -3.92906591e-02 1.13157904e+00
-6.39269412e-01 -7.96192527e-01 -8.77736509e-01 6.31701589e-01
-7.71189928e-02 1.53523758e-01 6.12300858e-02 9.51623976e-01
4.42439109e-01 5.63560367e-01 5.82376897e-01 -2.73133904e-01
6.38052881e-01 -2.19269335e-01 1.04842007e-01 -4.48669493e-01
-3.97453725e-01 2.97531039e-01 -5.77623606e-01 -7.38853335e-01
-2.58243203e-01 -8.16626966e-01 -1.69487202e+00 -5.43036103e-01
-3.99140447e-01 9.50320885e-02 1.24786282e+00 7.60731518e-01
4.70082551e-01 3.16425979e-01 6.94028139e-01 -1.14482677e+00
-4.89483505e-01 -8.55568707e-01 -6.94334269e-01 2.48788342e-01
2.48126879e-01 -6.98235691e-01 -5.04946709e-01 -2.03839317e-01] | [7.9549970626831055, -3.230024576187134] |
3f0b6103-0c83-4148-b7d2-d9f419e42ab5 | towards-robust-video-instance-segmentation | 2301.09416 | null | https://arxiv.org/abs/2301.09416v1 | https://arxiv.org/pdf/2301.09416v1.pdf | Towards Robust Video Instance Segmentation with Temporal-Aware Transformer | Most existing transformer based video instance segmentation methods extract per frame features independently, hence it is challenging to solve the appearance deformation problem. In this paper, we observe the temporal information is important as well and we propose TAFormer to aggregate spatio-temporal features both in transformer encoder and decoder. Specifically, in transformer encoder, we propose a novel spatio-temporal joint multi-scale deformable attention module which dynamically integrates the spatial and temporal information to obtain enriched spatio-temporal features. In transformer decoder, we introduce a temporal self-attention module to enhance the frame level box queries with the temporal relation. Moreover, TAFormer adopts an instance level contrastive loss to increase the discriminability of instance query embeddings. Therefore the tracking error caused by visually similar instances can be decreased. Experimental results show that TAFormer effectively leverages the spatial and temporal information to obtain context-aware feature representation and outperforms state-of-the-art methods. | ['Siyu Zhu', 'Zuozhuo Dai', 'Fangtao Shao', 'Zhenghao Zhang'] | 2023-01-20 | null | null | null | null | ['video-instance-segmentation'] | ['computer-vision'] | [ 1.17260031e-01 -2.79445201e-01 -2.28359312e-01 -3.76134127e-01
-7.89882362e-01 -5.32155335e-01 3.56987864e-01 -4.36243936e-02
-4.75131482e-01 4.14437294e-01 2.02269971e-01 2.37299904e-01
-2.04999402e-01 -6.78948581e-01 -8.37102592e-01 -6.29806519e-01
2.09727615e-01 -7.21173286e-02 7.70436108e-01 1.13094591e-01
-7.65702873e-02 2.25079775e-01 -1.38068175e+00 3.48286897e-01
9.66729641e-01 1.41986680e+00 4.20157015e-01 4.11167741e-01
-8.35621282e-02 6.36800349e-01 -2.43830383e-01 -3.65073532e-01
3.27832520e-01 -3.42116922e-01 -6.05444968e-01 3.52759063e-01
7.34031737e-01 -6.03676558e-01 -7.39698589e-01 8.64274979e-01
3.90981793e-01 3.49138737e-01 2.34553978e-01 -1.23451316e+00
-7.86815584e-01 2.17144281e-01 -8.76601100e-01 6.15937591e-01
1.84457496e-01 5.18888414e-01 1.10854399e+00 -9.19759214e-01
6.20538771e-01 1.22870314e+00 3.33868831e-01 1.06948562e-01
-9.23954129e-01 -6.60039186e-01 7.93474197e-01 5.13333976e-01
-1.23258722e+00 -2.59052843e-01 1.04658866e+00 -2.69147336e-01
8.09854209e-01 2.02377528e-01 8.70943367e-01 8.49004567e-01
-2.13321745e-01 1.23117530e+00 7.61049032e-01 4.17114459e-02
-7.97003731e-02 -1.83105424e-01 -1.75537229e-01 8.25932324e-01
-2.34738052e-01 1.09577753e-01 -5.68155587e-01 3.49900544e-01
1.23590183e+00 4.05608684e-01 -3.89710873e-01 -4.23486412e-01
-1.27812159e+00 4.49460775e-01 6.54395044e-01 4.42930549e-01
-4.28810745e-01 4.41991448e-01 4.77202952e-01 8.28909315e-03
5.40663421e-01 7.46689597e-03 -7.32741416e-01 -3.11716437e-01
-1.04773068e+00 2.81240046e-02 -1.01460166e-01 1.16449916e+00
7.34198630e-01 3.20001617e-02 -7.28152990e-01 7.44472742e-01
3.30939114e-01 2.55287945e-01 3.83411646e-01 -9.66176152e-01
5.64501703e-01 7.61192501e-01 3.67131196e-02 -1.01964641e+00
1.54534262e-02 -5.81809223e-01 -4.71987545e-01 -3.12783390e-01
1.81360483e-01 2.45273411e-01 -1.00179911e+00 1.63797760e+00
4.33947146e-01 8.06511819e-01 -3.54187906e-01 1.12640500e+00
7.81642497e-01 5.40345669e-01 2.66238064e-01 -2.98612207e-01
1.46461451e+00 -1.30119467e+00 -8.23094547e-01 -1.12648554e-01
3.02374363e-01 -6.42503977e-01 1.13277912e+00 -2.66632915e-01
-1.27074957e+00 -8.16964447e-01 -8.66250873e-01 -4.10222560e-01
-1.39917061e-01 2.30685920e-01 4.83834922e-01 2.54253030e-01
-6.58091724e-01 3.92038047e-01 -1.03955889e+00 -8.21317509e-02
6.89465880e-01 3.82253408e-01 -2.71943837e-01 -2.24837109e-01
-1.12968540e+00 3.01112950e-01 4.69716102e-01 7.48239756e-02
-6.99408770e-01 -8.51265550e-01 -1.04550588e+00 3.45469601e-02
5.19340813e-01 -7.21491277e-01 1.11515200e+00 -9.54619050e-01
-1.30199862e+00 5.36589563e-01 -5.70777059e-01 -2.26819962e-01
5.63857317e-01 -3.51545841e-01 -1.89327985e-01 4.51317668e-01
8.42639208e-02 6.88108385e-01 8.92423034e-01 -1.10944057e+00
-9.26110268e-01 -3.54785562e-01 3.59366626e-01 4.07028198e-01
-5.96795142e-01 -2.22935244e-01 -1.29538321e+00 -1.00766706e+00
1.61717683e-01 -5.39196789e-01 -7.79474825e-02 4.43737954e-01
-1.02227904e-01 -2.85374731e-01 1.10306180e+00 -7.30553687e-01
1.50135100e+00 -2.25795722e+00 2.55703330e-01 -7.07496330e-02
2.81425446e-01 2.95365989e-01 -2.19958082e-01 -9.09004807e-02
1.56540304e-01 1.19786084e-01 -1.76167190e-01 -4.10014480e-01
4.46845777e-03 2.57258385e-01 4.54323785e-03 2.70283908e-01
5.69111645e-01 1.27844739e+00 -1.21171904e+00 -7.64259219e-01
5.66631198e-01 7.60293305e-01 -7.27310598e-01 1.57373175e-01
-2.82960951e-01 4.98483777e-01 -8.82738948e-01 6.53292298e-01
7.35903382e-01 -3.46420258e-01 -2.75904685e-01 -7.67575979e-01
-4.16549519e-02 1.13129258e-01 -7.87384868e-01 2.03646469e+00
-4.90949959e-01 6.44049048e-01 -2.00379074e-01 -8.34695935e-01
4.74685907e-01 1.93829611e-01 6.72805011e-01 -9.96175885e-01
5.65872304e-02 6.69444399e-03 -3.73975098e-01 -6.76335633e-01
5.43698788e-01 1.81197494e-01 1.58897027e-01 8.44718963e-02
8.27812701e-02 3.79023194e-01 1.03605790e-02 1.25423595e-01
8.73064935e-01 5.07903636e-01 3.23157385e-02 1.80212379e-01
6.20116174e-01 -4.40854162e-01 9.23023224e-01 2.46011108e-01
-5.09007752e-01 8.50100279e-01 2.45037243e-01 -2.82675803e-01
-7.53610313e-01 -1.16258097e+00 4.96704429e-02 1.16911197e+00
7.16401219e-01 -3.52793843e-01 -5.50325274e-01 -9.84131038e-01
-4.87021208e-02 2.65453070e-01 -7.72227526e-01 -1.88206211e-01
-7.56466448e-01 -2.72420883e-01 1.54104382e-01 9.55854595e-01
8.17470551e-01 -8.54048789e-01 -6.84591532e-01 3.99155378e-01
-3.51672769e-01 -1.33961475e+00 -1.30029523e+00 -2.24761069e-01
-8.95972311e-01 -9.51323628e-01 -9.01228786e-01 -8.18436921e-01
5.48175216e-01 3.98085654e-01 8.73715699e-01 2.56861411e-02
-3.63594949e-01 4.94894892e-01 -5.94260037e-01 1.67380482e-01
3.74028474e-01 -3.96385975e-02 -4.53160048e-01 3.38751942e-01
2.60642767e-01 -7.39310443e-01 -1.19249856e+00 3.75598848e-01
-1.04012930e+00 1.29982516e-01 5.68449795e-01 7.82761753e-01
9.03069258e-01 -7.36017600e-02 4.39868659e-01 -3.55272889e-01
1.08307458e-01 -2.15574011e-01 -3.93454611e-01 4.21714425e-01
-1.16133362e-01 9.18308645e-02 2.53624618e-01 -6.72050536e-01
-1.03632927e+00 -1.52611323e-02 2.34709196e-02 -9.41281796e-01
6.28250539e-02 2.02046782e-01 -4.98485178e-01 -2.87819318e-02
-2.16086537e-01 5.83454013e-01 -4.29451287e-01 -5.34034550e-01
4.84980643e-01 2.63819069e-01 4.11145389e-01 -5.81316710e-01
7.60406971e-01 4.80178863e-01 -3.34615260e-01 -4.24076915e-01
-8.39380026e-01 -6.26969397e-01 -5.38071394e-01 -2.64988661e-01
1.25590193e+00 -9.64040399e-01 -6.62918031e-01 3.29448730e-01
-1.15547502e+00 -2.14315966e-01 -3.58165234e-01 3.44856679e-01
-5.87593794e-01 4.51576203e-01 -5.96726239e-01 -5.76931119e-01
-2.76938885e-01 -1.44694364e+00 1.54823554e+00 4.74734128e-01
1.82662249e-01 -8.63914847e-01 -2.86965281e-01 3.27105284e-01
1.27457678e-01 2.55048603e-01 5.98633945e-01 -2.22296059e-01
-1.17502868e+00 1.33978024e-01 -7.43995786e-01 2.23553225e-01
3.61702710e-01 -1.56685054e-01 -7.95515060e-01 -2.57734329e-01
-2.62468457e-01 4.37204503e-02 1.07478917e+00 3.26001644e-01
1.53818059e+00 -3.50972027e-01 -3.56094599e-01 8.44998717e-01
1.32899916e+00 2.81260610e-01 6.53385580e-01 2.46133953e-01
1.09061205e+00 3.64285737e-01 9.63926136e-01 3.85960609e-01
6.66574657e-01 1.04018331e+00 3.30537856e-01 -1.43003672e-01
-2.75381744e-01 -2.90910959e-01 3.03657949e-01 6.56845570e-01
-9.45195407e-02 -9.98524427e-02 -4.54930723e-01 7.56904721e-01
-2.10602427e+00 -1.00689042e+00 2.47727334e-01 2.02126908e+00
8.87229562e-01 -1.42785320e-02 8.36324096e-02 5.86380325e-02
6.11861169e-01 3.00318629e-01 -6.38574541e-01 1.76476851e-01
-7.42439777e-02 1.48029774e-01 3.05182099e-01 2.96805710e-01
-1.31924617e+00 1.00563598e+00 4.99701405e+00 1.15007699e+00
-1.06567669e+00 3.85415226e-01 6.85446024e-01 -3.89928281e-01
-4.39321488e-01 -2.30633467e-01 -3.50863069e-01 6.82709217e-01
2.26969823e-01 -8.58112201e-02 3.79202515e-02 6.16402984e-01
3.77015650e-01 1.50661290e-01 -1.07347226e+00 1.03547001e+00
-2.77463477e-02 -1.12754977e+00 1.78254247e-01 9.47714411e-03
6.35445774e-01 -3.27192903e-01 3.20272982e-01 2.54848868e-01
-2.72526354e-01 -7.18410730e-01 8.54001045e-01 6.29401147e-01
7.95643330e-01 -8.79905879e-01 4.57614958e-01 -1.12700239e-01
-1.89816749e+00 -1.05156519e-01 1.59515403e-02 4.45340037e-01
3.41573983e-01 2.85464615e-01 -2.45885178e-01 8.09788644e-01
7.80882716e-01 1.29189694e+00 -6.79823279e-01 1.27640939e+00
2.13320423e-02 3.57754141e-01 -3.22282910e-01 3.97404194e-01
3.96485537e-01 -9.09318849e-02 5.11568367e-01 1.18340969e+00
2.41765410e-01 2.39988208e-01 3.51331830e-01 7.81482160e-01
-5.54339401e-02 -4.01937328e-02 -6.07651323e-02 2.59138979e-02
5.28401494e-01 1.14113986e+00 -6.11917377e-01 -2.61015266e-01
-6.13643587e-01 1.44578183e+00 2.01976955e-01 6.48319244e-01
-1.22316825e+00 -1.46027416e-01 8.20604563e-01 1.01069041e-01
8.95471513e-01 -1.31121263e-01 -1.08535230e-01 -1.31098330e+00
4.61028308e-01 -5.85007071e-01 3.83400410e-01 -7.53573775e-01
-1.04496312e+00 5.94892085e-01 -1.92165211e-01 -1.48119724e+00
2.03490347e-01 -2.33514205e-01 -4.49075162e-01 5.97796082e-01
-1.75074267e+00 -1.62349391e+00 -4.78882402e-01 7.40241885e-01
9.72719610e-01 2.75603205e-01 2.13358864e-01 6.10432148e-01
-7.69145310e-01 1.02283990e+00 -2.48339012e-01 3.10324639e-01
6.66877568e-01 -1.07700443e+00 1.09059401e-01 8.35452497e-01
1.06068924e-01 5.89443088e-01 2.57160753e-01 -4.95033979e-01
-1.24048018e+00 -1.51184845e+00 4.20698166e-01 -3.66343409e-01
4.42829400e-01 -7.36853629e-02 -9.67449188e-01 5.66789687e-01
7.94589869e-04 5.71838796e-01 4.17314589e-01 -4.06886995e-01
-5.15388966e-01 -3.16713214e-01 -9.85075116e-01 5.17259717e-01
1.31893981e+00 -8.14206779e-01 -4.22294736e-01 -2.71576084e-02
1.12571108e+00 -4.28448021e-01 -9.84016359e-01 6.23011768e-01
6.04900718e-01 -9.15611744e-01 1.18644488e+00 -2.95421124e-01
4.38659906e-01 -6.04492426e-01 -1.51405275e-01 -7.26783037e-01
-3.00597847e-01 -4.75480765e-01 -4.17661518e-01 1.40444255e+00
2.82118842e-03 -4.00386214e-01 6.52009428e-01 4.62526917e-01
-1.55900925e-01 -1.17437112e+00 -1.21704876e+00 -7.26196766e-01
-2.16528192e-01 -4.44067121e-01 6.12395108e-01 7.99617648e-01
-3.17449689e-01 7.79051408e-02 -2.64809906e-01 1.30701572e-01
4.88864213e-01 2.74812520e-01 3.79156172e-01 -8.01888704e-01
-3.50677669e-01 -4.98107404e-01 -6.20916843e-01 -1.68283653e+00
-4.94999550e-02 -5.81250966e-01 -6.52720034e-02 -1.41083741e+00
3.50173324e-01 -4.28998619e-01 -7.65993774e-01 4.50250745e-01
-7.10554779e-01 3.59303623e-01 3.28346223e-01 1.94124691e-02
-1.05186117e+00 9.33559358e-01 1.66078162e+00 -1.85834229e-01
-1.45793378e-01 -4.43843544e-01 -3.46646458e-01 4.72190499e-01
4.36783671e-01 -1.81362048e-01 -6.04603350e-01 -5.84296465e-01
-2.21319184e-01 -2.20783390e-02 5.77088833e-01 -8.89204383e-01
1.44693166e-01 -2.37896115e-01 5.32426715e-01 -7.45511353e-01
5.38889408e-01 -9.27704513e-01 -1.64627329e-01 9.05028507e-02
-2.46863142e-01 -7.12863309e-03 3.24680299e-01 1.03539860e+00
-4.20519412e-01 3.73247743e-01 6.57122135e-01 6.75721690e-02
-9.24835384e-01 9.98857796e-01 6.15154766e-02 1.59177646e-01
1.14830446e+00 -4.54800636e-01 2.28470322e-02 -1.83400214e-01
-5.80794394e-01 4.43393022e-01 5.88264108e-01 7.59896278e-01
8.28177035e-01 -1.45977056e+00 -3.09289992e-01 8.55857953e-02
2.30720773e-01 5.31027243e-02 6.84446812e-01 1.11052370e+00
-8.96571875e-02 2.15664983e-01 -9.72362468e-04 -7.51649439e-01
-1.32515931e+00 6.16034687e-01 3.91039759e-01 -2.00556457e-01
-6.51104808e-01 1.10783803e+00 6.52102470e-01 1.73854679e-01
2.61761874e-01 -4.51320708e-01 -1.94872290e-01 7.70688355e-02
4.83761758e-01 1.20318029e-02 -4.11977082e-01 -8.04861724e-01
-4.20089483e-01 9.26279426e-01 -2.65038490e-01 -2.09688712e-02
1.20215666e+00 -3.70981157e-01 2.04461202e-01 2.43581206e-01
1.48393059e+00 -1.66820243e-01 -1.78773618e+00 -4.59307373e-01
-2.42484018e-01 -8.79851460e-01 2.06447184e-01 -5.31679451e-01
-1.59599638e+00 9.58523393e-01 7.14236677e-01 -1.44797832e-01
1.58070791e+00 1.87585037e-02 1.24596179e+00 -3.14322144e-01
2.14896008e-01 -8.10931921e-01 5.52396953e-01 2.71556824e-01
7.68654346e-01 -1.22703302e+00 -1.68734208e-01 -7.01731682e-01
-4.82509226e-01 9.40436423e-01 8.95417571e-01 4.32332531e-02
4.88326043e-01 5.13899773e-02 -2.76010782e-01 -3.12257297e-02
-6.28160238e-01 -5.03517210e-01 7.68045545e-01 5.15773535e-01
4.39808160e-01 -2.39830405e-01 -2.15899140e-01 7.72671759e-01
3.53513896e-01 -3.57870050e-02 -8.82249046e-03 8.77493262e-01
-2.46422768e-01 -1.10288489e+00 1.14485778e-01 3.11610609e-01
-3.54919761e-01 -2.05210179e-01 -3.48685794e-02 6.73308194e-01
5.25351107e-01 7.18829989e-01 1.41880721e-01 -4.25267905e-01
3.57064396e-01 -2.92493016e-01 5.95301449e-01 -3.09649974e-01
-5.41909516e-01 4.40729797e-01 -1.10275425e-01 -8.27232420e-01
-5.58469594e-01 -6.15869939e-01 -1.32136643e+00 1.60698786e-01
-3.38778824e-01 -1.78321958e-01 5.98915704e-02 9.03986275e-01
6.00671768e-01 9.88225341e-01 5.35619318e-01 -7.79745042e-01
6.61661178e-02 -7.25065231e-01 -2.32574880e-01 5.08750319e-01
5.80006182e-01 -9.28916156e-01 -1.80870742e-01 2.12320030e-01] | [9.187912940979004, -0.02568202279508114] |
95c49548-eaba-4d4d-bd30-7d13b5191e84 | bert-based-simplification-of-japanese | null | null | https://aclanthology.org/2020.inlg-1.31 | https://aclanthology.org/2020.inlg-1.31.pdf | BERT-Based Simplification of Japanese Sentence-Ending Predicates in Descriptive Text | Japanese sentence-ending predicates intricately combine content words and functional elements, such as aspect, modality, and honorifics; this can often hinder the understanding of language learners and children. Conventional lexical simplification methods, which replace difficult target words with simpler synonyms acquired from lexical resources in a word-by-word manner, are not always suitable for the simplification of such Japanese predicates. Given this situation, we propose a BERT-based simplification method, the core feature of which is the high ability to substitute the whole predicates with simple ones while maintaining their core meanings in the context by utilizing pre-trained masked language models. Experimental results showed that our proposed methods consistently outperformed the conventional thesaurus-based method by a wide margin. Furthermore, we investigated in detail the effectiveness of the average token embedding and dropout, and the remaining errors of our BERT-based methods. | ['Satoshi Sato', 'Rei Miyata', 'Taichi Kato'] | null | null | null | null | inlg-acl-2020-12 | ['lexical-simplification'] | ['natural-language-processing'] | [ 4.38642055e-02 5.41040152e-02 5.34598269e-02 -4.24394697e-01
-5.36799371e-01 -5.33085108e-01 1.64004102e-01 2.79883713e-01
-8.98676932e-01 1.04015160e+00 2.12328956e-01 -6.69391155e-02
7.07197487e-02 -8.17299068e-01 -6.56849921e-01 -7.18729258e-01
4.82456297e-01 2.38935485e-01 4.44511920e-01 -5.93446195e-01
-9.09700170e-02 1.91008866e-01 -1.55186987e+00 3.36602516e-02
1.58866739e+00 2.24454939e-01 3.67706478e-01 -2.46066198e-01
-3.70048702e-01 4.51293468e-01 -8.55728686e-01 -1.13563216e+00
-7.45277479e-02 -1.75428718e-01 -3.18020344e-01 -4.21206027e-01
3.04647654e-01 -1.89493045e-01 -2.13412195e-01 1.22161055e+00
4.40845102e-01 4.35757995e-01 4.59753543e-01 -5.45138597e-01
-1.02732575e+00 1.32616448e+00 -2.79319882e-01 1.70006230e-01
3.46440315e-01 -2.38939866e-01 1.07439756e+00 -8.37045908e-01
4.59830403e-01 1.39090657e+00 6.49343610e-01 7.80602276e-01
-1.25131953e+00 -7.66708910e-01 5.58772027e-01 3.06478053e-01
-1.72618663e+00 -1.26731500e-01 6.99091971e-01 -1.68925116e-03
1.23836112e+00 3.02793592e-01 8.69295120e-01 7.13354826e-01
8.64461735e-02 6.91206992e-01 7.54648566e-01 -7.05377936e-01
-2.64819980e-01 2.31412277e-01 2.77521789e-01 5.60737371e-01
6.26630783e-01 -3.32409889e-01 -4.23115045e-01 3.29668671e-01
3.26960295e-01 -7.21708015e-02 -3.07489693e-01 6.67717531e-02
-1.03331459e+00 7.21307695e-01 1.83032230e-01 4.04113948e-01
-5.41838072e-02 -5.46636693e-02 3.31107676e-01 3.32732201e-01
2.34579742e-01 6.37536585e-01 -6.49010479e-01 -7.45358467e-02
-6.03984594e-01 3.28766435e-01 5.51753581e-01 1.36738932e+00
7.89666772e-01 1.37063518e-01 -1.18657872e-01 9.55661952e-01
-1.46335391e-02 7.00819433e-01 6.17048264e-01 -5.51385343e-01
5.17905056e-01 5.38144290e-01 -1.22446150e-01 -4.88970250e-01
-1.48230776e-01 -1.99975923e-01 -3.64103973e-01 -5.21997154e-01
2.71069318e-01 -1.63210198e-01 -8.23222518e-01 2.24739170e+00
6.11197472e-01 -2.02113211e-01 1.88018367e-01 4.00919914e-01
1.01936460e+00 5.23833513e-01 4.98560458e-01 -3.85825366e-01
1.45768321e+00 -1.06280386e+00 -1.09124899e+00 -1.00566819e-01
9.66409922e-01 -8.99589539e-01 1.60798538e+00 1.58712342e-01
-1.22771370e+00 -3.04960847e-01 -1.05144811e+00 -4.59227294e-01
-6.24926567e-01 -6.78055212e-02 7.26498544e-01 9.67955172e-01
-5.91999412e-01 6.22916162e-01 -4.97110069e-01 -1.36421621e-01
4.81927507e-02 4.06943977e-01 -4.38274115e-01 -2.32031215e-02
-1.64340925e+00 1.26552474e+00 6.68638289e-01 2.42229756e-02
-3.69931340e-01 -9.65955555e-01 -1.10049903e+00 3.47563863e-01
4.07170117e-01 -6.02199614e-01 1.29467952e+00 -7.46542752e-01
-1.68824911e+00 7.70457506e-01 -1.28469378e-01 -1.43259525e-01
4.20249403e-02 -6.83414340e-01 -4.76235658e-01 -1.03052847e-01
1.97744712e-01 4.96597201e-01 6.54771030e-01 -6.31703734e-01
-6.55337691e-01 -4.00930978e-02 2.05768973e-01 5.10035217e-01
-4.89912301e-01 1.24080017e-01 -4.30127740e-01 -9.39548850e-01
1.53376639e-01 -7.51267612e-01 -5.28590642e-02 -4.60912257e-01
-2.84489125e-01 -7.44831502e-01 3.62215698e-01 -6.47064805e-01
1.63211429e+00 -2.44688916e+00 1.92145571e-01 -2.96294689e-02
6.28284290e-02 5.70372581e-01 -1.12493232e-01 4.91909683e-01
-9.46696699e-02 2.14531645e-01 -9.86957774e-02 -3.18248361e-01
7.08051696e-02 3.60797256e-01 -3.52534384e-01 2.86103506e-02
1.55269012e-01 8.65163863e-01 -8.98871839e-01 -6.86857700e-01
1.61034450e-01 3.94792765e-01 -7.12168753e-01 8.25446770e-02
-2.67013937e-01 1.39795557e-01 -3.14324856e-01 5.96852422e-01
5.49063623e-01 4.92147535e-01 2.56774455e-01 -8.58049169e-02
-1.58192202e-01 1.00002027e+00 -1.05680382e+00 1.40627527e+00
-5.74623704e-01 2.07441032e-01 -2.79600888e-01 -7.23973691e-01
5.69591701e-01 3.29112291e-01 -7.95813277e-02 -7.15862453e-01
4.69102859e-02 5.04403055e-01 3.62481207e-01 -3.32434535e-01
4.91304487e-01 -3.75681520e-01 -2.66457051e-01 6.92724213e-02
-8.69585294e-03 -4.90865529e-01 5.73223472e-01 5.94100729e-02
6.15890265e-01 2.86602378e-01 5.55602849e-01 -3.22571546e-01
5.96252978e-01 -1.29516259e-01 9.51415598e-01 4.76313323e-01
8.87481421e-02 2.95486808e-01 3.14691335e-01 -2.15204462e-01
-6.76239491e-01 -1.23169982e+00 -3.03575993e-01 1.15016735e+00
1.12215847e-01 -5.86542368e-01 -8.82263899e-01 -6.54061913e-01
-1.58853799e-01 1.30155516e+00 -2.83707708e-01 -2.11799607e-01
-1.08029938e+00 -5.05364299e-01 4.81659025e-01 3.66953343e-01
2.03208745e-01 -1.10714388e+00 -3.20546031e-01 4.62531865e-01
-1.16439812e-01 -1.17815506e+00 -6.04003310e-01 2.71037132e-01
-6.30258620e-01 -6.72512770e-01 -4.14841145e-01 -1.17092967e+00
7.52225637e-01 3.65904838e-01 9.43638384e-01 3.89870197e-01
2.57207781e-01 -2.30710760e-01 -4.80910510e-01 -8.43418121e-01
-1.29011959e-01 1.86821744e-01 2.46576354e-01 -4.98524487e-01
7.18768716e-01 -8.36599886e-01 -4.45335917e-02 -1.95386112e-01
-1.07405651e+00 6.75903261e-02 5.25747657e-01 8.70301187e-01
5.45180500e-01 1.62747279e-01 4.96278942e-01 -1.15642476e+00
5.09641111e-01 2.26113596e-03 -5.91281116e-01 4.34174716e-01
-4.26793873e-01 2.39153579e-01 8.25505674e-01 -8.64782751e-01
-1.19534028e+00 -2.41333395e-01 -4.63015914e-01 1.11993141e-01
7.61212409e-02 4.64189082e-01 -7.29447365e-01 -2.64086183e-02
2.44551331e-01 2.63436198e-01 -4.14046466e-01 -5.76861024e-01
5.25582135e-01 2.02539191e-01 2.64466465e-01 -8.15942347e-01
8.37222338e-01 9.15344432e-03 -2.33522668e-01 -9.49472070e-01
-1.12832117e+00 -3.09712011e-02 -5.45713484e-01 4.18325305e-01
7.62753427e-01 -1.10181284e+00 -3.55662495e-01 3.90634626e-01
-1.40133274e+00 -1.51351909e-03 -5.18861294e-01 7.99095690e-01
6.75366959e-03 6.61736727e-01 -7.09360600e-01 -2.92741805e-01
-2.79641420e-01 -1.08837283e+00 7.26509511e-01 5.26844800e-01
-4.30381209e-01 -9.01200056e-01 -1.90681919e-01 2.75784016e-01
3.64762664e-01 -2.19003111e-01 1.73141432e+00 -9.75863814e-01
-3.90089720e-01 2.57800221e-02 5.91292605e-02 5.39120853e-01
2.66448587e-01 7.75028244e-02 -6.56076372e-01 1.38126239e-01
1.25023827e-01 -6.38672560e-02 6.80752456e-01 1.43063605e-01
6.70101523e-01 -3.16501647e-01 -1.87195539e-01 5.79815090e-01
1.36299396e+00 2.35927239e-01 6.22032702e-01 2.87717730e-01
9.73305106e-01 5.84402263e-01 5.67501068e-01 8.35123099e-03
5.91103792e-01 5.50136983e-01 -1.52846992e-01 1.14667468e-01
-2.51649857e-01 -3.22218865e-01 3.33139151e-01 1.48953664e+00
2.12472975e-01 1.88178290e-02 -5.07312536e-01 8.60146105e-01
-1.52856898e+00 -5.86855352e-01 -1.20493628e-01 2.18949914e+00
1.64761150e+00 4.13175404e-01 -2.64435977e-01 8.52449834e-02
7.50771105e-01 -2.11097952e-02 9.63619426e-02 -5.97279370e-01
-3.28221768e-01 5.98156869e-01 2.08967388e-01 5.51452041e-01
-7.66237438e-01 1.74845266e+00 6.15110350e+00 1.20993114e+00
-1.03063953e+00 2.11035445e-01 -1.77167565e-01 -5.77793606e-02
-8.37042391e-01 1.65286422e-01 -1.30912673e+00 3.67130578e-01
5.03714740e-01 -2.61652201e-01 4.92201447e-01 6.46364927e-01
-5.99168204e-02 -1.15308546e-01 -1.14992249e+00 6.98751867e-01
1.14538714e-01 -7.99683750e-01 4.46810335e-01 -5.09414494e-01
7.80847132e-01 -4.09508795e-01 -1.60826162e-01 6.18273675e-01
1.08276971e-01 -7.86240160e-01 8.56479824e-01 1.62038341e-01
5.04671633e-01 -7.39470303e-01 6.28674924e-01 1.32854789e-01
-1.00354946e+00 2.86050707e-01 -4.79331464e-01 -5.44331551e-01
1.08376935e-01 6.61770761e-01 -3.96507174e-01 2.76295006e-01
5.03067911e-01 3.84667516e-01 -4.84975129e-01 9.05638516e-01
-1.09565175e+00 6.26899183e-01 -5.00695884e-01 -5.43412268e-01
-3.83853652e-02 -3.74869734e-01 7.44096220e-01 1.11492312e+00
3.10197085e-01 1.30549207e-01 -2.33343512e-01 6.64533436e-01
3.06875538e-02 4.97120470e-01 -5.19273758e-01 7.95023702e-03
9.06500041e-01 9.36806321e-01 -4.43963319e-01 -4.56982017e-01
-7.38558710e-01 6.51387215e-01 6.16674602e-01 3.56372505e-01
-8.65615308e-01 -7.14403689e-01 8.02816451e-01 -1.33052364e-01
7.54757941e-01 -2.16354847e-01 -3.83005261e-01 -1.37520170e+00
3.29219580e-01 -9.14880395e-01 1.78375378e-01 -4.27761585e-01
-1.19354677e+00 4.40466106e-01 3.16094786e-01 -9.25506651e-01
1.68840110e-01 -4.44897145e-01 -5.67878187e-01 8.40674460e-01
-1.53759766e+00 -1.31194794e+00 3.76056224e-01 5.63197315e-01
4.65506196e-01 1.65284961e-01 9.75466669e-01 4.50422645e-01
-8.81615579e-01 9.64100242e-01 -1.60461117e-03 -1.19597889e-01
6.05676055e-01 -1.13515425e+00 2.30779976e-01 1.20250773e+00
1.16728216e-01 1.16071820e+00 9.28754449e-01 -7.55299985e-01
-1.05633783e+00 -7.64317274e-01 1.69711375e+00 -2.32874691e-01
7.00426996e-01 -4.35938239e-01 -1.14781415e+00 7.11899817e-01
2.95525700e-01 -5.27419269e-01 8.00443709e-01 2.81122953e-01
-3.12611133e-01 -4.10203844e-01 -8.55198145e-01 1.25346768e+00
1.11816835e+00 -2.91391104e-01 -1.43789625e+00 2.19591320e-01
1.40279150e+00 -4.69943166e-01 -5.84672749e-01 4.91277248e-01
3.51896465e-01 -5.10658801e-01 6.54778779e-01 -4.79646027e-01
1.61439776e-01 -3.64695936e-01 -7.80956969e-02 -1.39193535e+00
-3.92734975e-01 -7.69770026e-01 1.88852653e-01 1.51132786e+00
6.40371978e-01 -5.91785014e-01 2.63052970e-01 6.82154894e-01
-5.01108944e-01 -6.70959830e-01 -1.07071388e+00 -6.72550559e-01
2.38336995e-01 -3.26806575e-01 9.40383554e-01 7.60966241e-01
-6.29261956e-02 6.34907007e-01 -8.56494084e-02 4.20414610e-03
2.00643167e-01 -6.30240282e-03 4.51686740e-01 -9.41013932e-01
-4.32678312e-02 -4.63140041e-01 -2.70305336e-01 -8.85327101e-01
3.71600240e-01 -6.20343328e-01 -6.68676198e-02 -1.24324906e+00
-1.73416704e-01 -4.49723899e-01 -3.22791576e-01 5.12260199e-01
-7.14408517e-01 1.48027996e-03 3.10104162e-01 -1.24978215e-01
-4.60953355e-01 8.19925666e-01 1.33403552e+00 1.11810572e-01
-3.10178578e-01 -1.02583796e-01 -8.72785926e-01 8.90319288e-01
4.92067695e-01 -8.50711465e-01 -4.14217860e-01 -8.25273335e-01
4.24360245e-01 -5.42281568e-01 -3.30686539e-01 -8.05896342e-01
-2.34551355e-02 -3.41815084e-01 -1.65075570e-01 -4.08708781e-01
2.79132098e-01 -9.07925010e-01 -1.09733969e-01 4.33500260e-01
-8.03375766e-02 1.30343020e-01 4.04173940e-01 -1.01417350e-02
-2.79003233e-01 -5.80123723e-01 5.62388718e-01 -7.40442574e-02
-7.82015562e-01 -2.53697813e-01 -3.08872044e-01 3.77414256e-01
9.42822695e-01 -2.68425077e-01 -1.98815241e-01 1.06612235e-01
-3.57127756e-01 9.17461216e-02 2.77692854e-01 2.65442878e-01
4.37103361e-01 -1.20453179e+00 -6.03160918e-01 4.33556080e-01
5.67064509e-02 2.13098779e-01 2.43441194e-01 7.69167840e-01
-6.93306983e-01 3.19507390e-01 -5.19648455e-02 8.19459707e-02
-1.22016561e+00 6.45111024e-01 1.47414327e-01 -1.30516812e-01
-4.36522216e-01 1.17315483e+00 5.29171944e-01 -4.65875894e-01
4.06843394e-01 -8.09983015e-01 -3.55365157e-01 5.66186160e-02
4.45556909e-01 3.24948989e-02 1.33084923e-01 -5.21709025e-01
-4.15088356e-01 4.87718105e-01 -3.05224210e-01 2.82817315e-02
1.17076111e+00 -2.87687838e-01 -3.62063974e-01 4.21771228e-01
7.67542779e-01 7.36727953e-01 -6.48204088e-01 -4.66091603e-01
-2.61376277e-02 -3.11828136e-01 -4.33986664e-01 -5.44425011e-01
-6.51923239e-01 7.10258424e-01 -1.78156957e-01 -4.17960547e-02
1.32724035e+00 -8.49119350e-02 1.08132756e+00 5.19085944e-01
2.46698081e-01 -1.18417084e+00 -4.36939508e-01 8.94210637e-01
6.06274605e-01 -8.44836473e-01 1.25112146e-01 -1.07917583e+00
-5.31046867e-01 7.27526069e-01 9.93131459e-01 -8.83815512e-02
4.41837579e-01 9.61925741e-03 1.39610916e-01 1.52216002e-01
-7.45102942e-01 -2.33407155e-01 2.37962812e-01 5.04428327e-01
5.59914052e-01 2.53885061e-01 -1.18802643e+00 1.06913149e+00
-6.63714945e-01 -5.08798361e-01 4.84633535e-01 8.76643419e-01
-2.83631235e-01 -1.40481079e+00 -1.96066037e-01 1.93462625e-01
-7.99883902e-01 -8.38634908e-01 -7.45275915e-02 1.12654257e+00
5.29852152e-01 5.98065794e-01 -1.57204673e-01 -1.17799610e-01
6.39468789e-01 1.91224232e-01 7.37739563e-01 -1.07670987e+00
-7.94952929e-01 1.00017404e-02 2.46149108e-01 -4.62107100e-02
-2.16753528e-01 -5.86539507e-01 -1.40374458e+00 -2.50195444e-01
-6.61704302e-01 1.75686538e-01 4.74123269e-01 1.34118271e+00
-1.45497933e-01 4.83597070e-01 2.99397796e-01 -2.97947556e-01
-7.02907383e-01 -1.02615499e+00 -5.65737069e-01 4.24980938e-01
1.75893530e-02 -7.64085591e-01 -3.44097823e-01 -2.21274152e-01] | [10.801602363586426, 10.212812423706055] |
c4241d1c-b62d-45eb-aed1-faad2cb1d502 | learning-physical-intuition-of-block-towers | 1603.01312 | null | http://arxiv.org/abs/1603.01312v1 | http://arxiv.org/pdf/1603.01312v1.pdf | Learning Physical Intuition of Block Towers by Example | Wooden blocks are a common toy for infants, allowing them to develop motor
skills and gain intuition about the physical behavior of the world. In this
paper, we explore the ability of deep feed-forward models to learn such
intuitive physics. Using a 3D game engine, we create small towers of wooden
blocks whose stability is randomized and render them collapsing (or remaining
upright). This data allows us to train large convolutional network models which
can accurately predict the outcome, as well as estimating the block
trajectories. The models are also able to generalize in two important ways: (i)
to new physical scenarios, e.g. towers with an additional block and (ii) to
images of real wooden blocks, where it obtains a performance comparable to
human subjects. | ['Sam Gross', 'Adam Lerer', 'Rob Fergus'] | 2016-03-03 | null | null | null | null | ['physical-intuition'] | ['reasoning'] | [-4.07453150e-01 1.11033797e-01 1.12506092e-01 6.09760219e-03
3.98978412e-01 -6.49962902e-01 5.03191888e-01 -1.84482674e-03
-1.41606694e-02 4.88285929e-01 1.09584227e-01 -4.27879125e-01
-1.25673831e-01 -1.05558932e+00 -1.17947888e+00 -4.56014603e-01
-5.35853863e-01 4.08299923e-01 6.72907829e-01 -3.35635155e-01
2.21194848e-02 6.11023068e-01 -1.73835897e+00 2.57763624e-01
7.24383652e-01 5.32362521e-01 6.62995815e-01 9.70406473e-01
4.50310379e-01 1.22940826e+00 8.24661367e-03 -1.50809199e-01
2.76862264e-01 1.23206981e-01 -8.35454345e-01 -5.20970464e-01
1.82468370e-01 -9.26231146e-01 -6.49231911e-01 6.84064507e-01
1.44181587e-02 3.93014789e-01 6.50869131e-01 -9.10223901e-01
-8.03428531e-01 7.05701113e-01 -5.54117501e-01 3.05729918e-02
5.15842617e-01 2.53084421e-01 7.52839863e-01 -4.42944616e-01
4.48977143e-01 1.10493708e+00 7.44251907e-01 9.46979880e-01
-1.19576359e+00 -5.45729518e-01 1.99528560e-01 3.15176308e-01
-8.49920213e-01 -1.22684032e-01 5.40145874e-01 -8.29462707e-01
1.16198230e+00 1.02416933e-01 1.34467196e+00 1.25510931e+00
2.51383364e-01 8.14042509e-01 9.56743181e-01 -2.52774507e-01
3.22128236e-01 -4.64906484e-01 5.20249307e-02 7.67306864e-01
3.16048920e-01 5.98025739e-01 -4.13795143e-01 3.07551891e-01
1.08600044e+00 1.06024265e-01 -8.79297480e-02 -9.27001059e-01
-1.08861160e+00 4.51195091e-01 8.35083604e-01 7.37407431e-02
-3.46314669e-01 6.94199443e-01 1.06551297e-01 -9.62631330e-02
2.89856712e-03 5.50503552e-01 -7.20640659e-01 -3.70638072e-01
-6.23748899e-01 9.20457304e-01 6.77002668e-01 1.00834644e+00
3.02785456e-01 -1.49805039e-01 2.54832417e-01 5.63665271e-01
3.41877043e-01 3.45563620e-01 2.18679860e-01 -9.87643301e-01
5.59905767e-01 1.25218242e-01 3.72121751e-01 -6.12050712e-01
-7.99542427e-01 -1.86366420e-02 -6.88561440e-01 8.36362898e-01
4.80160862e-01 -2.05799993e-02 -1.08540201e+00 2.06866503e+00
3.13865185e-01 3.36711854e-01 -2.82437891e-01 7.13141143e-01
8.94709587e-01 7.44546771e-01 -4.40081805e-02 3.71643454e-01
1.19961417e+00 -8.92940342e-01 1.78575516e-04 -1.24495149e-01
3.50511640e-01 -1.03518933e-01 1.19633722e+00 5.47905862e-01
-1.40913630e+00 -5.23348033e-01 -1.18904316e+00 -9.02508721e-02
-4.03223932e-01 -3.71407539e-01 1.06567574e+00 3.32748711e-01
-1.35420907e+00 1.24033511e+00 -1.57196343e+00 -2.96782374e-01
6.75709605e-01 4.83928621e-01 -4.64959711e-01 1.44583866e-01
-1.02368081e+00 1.16808391e+00 4.97668952e-01 -6.23554066e-02
-1.09626424e+00 -1.05294490e+00 -8.98732185e-01 3.04481059e-01
-2.03087926e-01 -9.00302768e-01 1.54446435e+00 -3.86566967e-01
-1.63358104e+00 8.73869300e-01 2.93454885e-01 -4.52373326e-01
3.88108999e-01 -5.15539944e-01 1.89910501e-01 -9.43455026e-02
-3.39196362e-02 8.65731418e-01 6.76133513e-01 -8.44427407e-01
-2.33266547e-01 -1.67723328e-01 6.73601031e-01 1.12053014e-01
-1.15871020e-01 -2.07072243e-01 1.72885597e-01 -6.66651368e-01
1.51584744e-01 -1.05451214e+00 -3.69208008e-01 1.36645705e-01
-4.37864959e-01 -4.90530171e-02 3.62290025e-01 -6.87635541e-01
9.81062949e-01 -1.79761052e+00 3.29408497e-01 -9.56209153e-02
5.71429014e-01 1.69723898e-01 -3.90145928e-02 5.32706857e-01
-4.13461566e-01 3.26373093e-02 1.24814376e-01 2.33518958e-01
-9.73587111e-02 6.59664860e-03 -2.31707860e-02 7.38318032e-03
1.77730531e-01 1.19567132e+00 -1.11659729e+00 8.38971958e-02
4.24298614e-01 1.71706274e-01 -9.28927302e-01 1.66488528e-01
-4.66697156e-01 8.52773309e-01 -4.69626367e-01 1.26608191e-02
5.61201811e-01 -1.03106990e-01 -1.72115341e-01 1.94370806e-01
-2.89866090e-01 7.42758572e-01 -6.31178916e-01 1.46175933e+00
-6.63996041e-01 8.15674424e-01 -2.79998988e-01 -8.55071545e-01
2.52311796e-01 1.05412930e-01 3.58758003e-01 -4.05634850e-01
7.17674941e-02 -3.11015904e-01 3.36901814e-01 -6.69888198e-01
6.51746765e-02 -3.47273707e-01 9.06737894e-02 6.15604937e-01
-2.04205558e-01 -6.24575436e-01 9.93549675e-02 -4.65784185e-02
1.42099440e+00 5.17477751e-01 3.24385524e-01 -4.88469213e-01
-3.81905772e-02 -2.82409668e-01 2.57667303e-02 5.80730736e-01
1.35746330e-01 6.30437315e-01 4.33050990e-01 -9.38420951e-01
-1.28080523e+00 -1.67665851e+00 9.32160392e-02 1.00698423e+00
7.76733235e-02 -1.53624788e-01 -9.42621112e-01 2.31945291e-02
1.13994136e-01 8.07310820e-01 -1.10878277e+00 -5.16853750e-01
-7.38759279e-01 -3.68854195e-01 3.51062082e-02 9.00876462e-01
3.33966613e-01 -1.34617472e+00 -1.19553220e+00 2.98473626e-01
5.20741642e-02 -8.05885375e-01 4.27907221e-02 3.42409670e-01
-8.99436414e-01 -9.83372808e-01 -4.73159701e-01 -6.83126688e-01
3.04361790e-01 4.15902697e-02 1.18619001e+00 1.20127685e-01
-2.74366587e-01 1.05716512e-01 -3.18517745e-01 -5.46680152e-01
-4.76900727e-01 3.96778211e-02 2.80025601e-01 -7.99160779e-01
-8.58637094e-02 -1.25183356e+00 -8.08427632e-01 -6.56975899e-03
-5.67508936e-01 9.34291005e-01 2.61171252e-01 6.02412403e-01
2.59343069e-02 1.25625372e-01 1.49612129e-01 -3.01047295e-01
3.18534136e-01 -4.45602864e-01 -7.13897049e-01 -5.43905869e-02
1.84746610e-03 2.97660530e-01 8.00415814e-01 -7.24993944e-01
-9.14943993e-01 -1.76441193e-01 -3.58989537e-01 -3.30587626e-01
-1.24957308e-01 -7.24875415e-03 9.05989259e-02 -3.57134431e-03
4.95949626e-01 -1.77713290e-01 -5.50559759e-01 -5.30107915e-01
6.54713690e-01 -5.06048128e-02 5.77125192e-01 -8.07389200e-01
9.76010859e-01 1.48593217e-01 7.12170079e-02 -7.54304230e-01
-7.66761720e-01 1.85341805e-01 -1.02164519e+00 -2.08193615e-01
1.24069500e+00 -6.66758835e-01 -1.22748828e+00 8.59690487e-01
-1.11612713e+00 -1.20784891e+00 -2.86679864e-01 5.73094845e-01
-1.00799513e+00 -2.97964990e-01 -9.53966200e-01 -3.78922582e-01
3.91657501e-02 -8.36065412e-01 7.19897687e-01 4.92104739e-01
-5.58799207e-01 -8.57098401e-01 3.71037543e-01 6.02737367e-02
1.22142345e-01 3.92450452e-01 1.43791270e+00 -6.82508722e-02
-6.54373825e-01 1.54333115e-01 -7.95295835e-02 -2.51437295e-02
1.39096836e-02 4.62972462e-01 -8.79817605e-01 -3.12737107e-01
-2.34519660e-01 -4.12498564e-01 5.77007055e-01 7.06264079e-01
1.47041667e+00 -1.02831729e-01 -4.85297471e-01 7.03666210e-01
9.61435318e-01 3.48219275e-01 6.79227769e-01 2.78614551e-01
7.86482334e-01 4.05484229e-01 1.58914641e-01 2.95715004e-01
5.53521514e-01 6.24270797e-01 6.32186294e-01 1.93871528e-01
-1.54213443e-01 -8.55198860e-01 2.04629321e-02 8.39089751e-01
-5.63269556e-01 -8.14945716e-03 -1.10755777e+00 5.73506951e-01
-1.56359804e+00 -1.01818764e+00 -1.14889871e-02 1.99495852e+00
6.99984729e-01 4.48674768e-01 2.23554209e-01 6.47782385e-02
4.49696839e-01 4.89975512e-02 -7.77151585e-01 -7.13100433e-01
5.17314434e-01 8.28567386e-01 1.80494517e-01 1.41419917e-01
-9.86817956e-01 1.13993239e+00 7.36706305e+00 4.52452630e-01
-9.62157011e-01 -1.19473740e-01 4.45522517e-01 -3.42308640e-01
-8.83665383e-02 -1.22983165e-01 -3.12545151e-01 4.82389212e-01
7.05575645e-01 7.02303424e-02 9.58396554e-01 6.53296232e-01
2.28105798e-01 -2.20953271e-01 -1.61881042e+00 6.84295297e-01
-7.32191205e-01 -1.20649254e+00 -2.16121137e-01 -2.37202704e-01
8.53965998e-01 2.34808072e-01 6.29457161e-02 4.13232505e-01
1.08072281e+00 -1.43999386e+00 1.10192966e+00 4.53639925e-01
7.36035943e-01 -3.26681137e-01 -6.96019083e-02 8.72179627e-01
-1.17121041e+00 -2.01223806e-01 -4.08563226e-01 -7.97023177e-01
-1.47224426e-01 6.66516572e-02 -7.48310626e-01 -8.37875381e-02
1.28872859e+00 6.51930332e-01 -3.96824390e-01 9.07653987e-01
-6.73968971e-01 5.13049841e-01 -5.71271181e-01 -6.39503479e-01
2.13508651e-01 -3.07523459e-01 3.44658792e-01 5.11808157e-01
2.60727316e-01 5.19671619e-01 -6.08110547e-01 1.15546179e+00
1.18327782e-01 -3.85241210e-01 -1.00291073e+00 1.62457317e-01
2.47255951e-01 8.73119652e-01 -8.08497131e-01 1.36621762e-02
-3.53128999e-01 6.31288648e-01 9.09783721e-01 2.63200521e-01
-8.76998305e-01 -8.61573517e-02 9.45371211e-01 4.41215724e-01
5.90786517e-01 -3.95398945e-01 -2.48866722e-01 -1.06925380e+00
2.02292800e-02 -4.36326146e-01 -4.87865031e-01 -1.48958313e+00
-8.20676088e-01 2.01930717e-01 2.94297636e-01 -8.80248904e-01
-1.92953393e-01 -8.82539809e-01 -1.04426718e+00 3.74790519e-01
-6.78374171e-01 -1.01064718e+00 -2.26094291e-01 1.98923022e-01
3.08512747e-01 2.89540082e-01 5.91237366e-01 -1.74646769e-02
-3.08773011e-01 2.99982786e-01 5.55464998e-02 1.48951367e-01
-3.09363037e-01 -1.41022146e+00 1.17643499e+00 6.33450627e-01
3.67122948e-01 6.92275643e-01 1.00308859e+00 -4.00438040e-01
-9.89879429e-01 -6.77875042e-01 1.08407475e-01 -7.25001216e-01
8.47106516e-01 -9.30271983e-01 -9.27829802e-01 8.56998324e-01
-9.30500478e-02 -7.89199173e-02 1.62139207e-01 1.88566804e-01
-2.29037657e-01 4.19001654e-02 -9.84468102e-01 1.05475628e+00
1.73744965e+00 -2.07648143e-01 -8.51031244e-01 1.86713248e-01
8.13571632e-01 -7.08169997e-01 -6.59893036e-01 4.68026519e-01
1.21295559e+00 -1.34087622e+00 1.23650181e+00 -1.20087135e+00
1.01883209e+00 1.17398508e-01 3.44152331e-01 -1.70546687e+00
-6.70234621e-01 -4.36871529e-01 -4.13321465e-01 5.90930045e-01
2.51301676e-01 -5.37394047e-01 9.73591626e-01 6.95443392e-01
1.27431914e-01 -1.17327094e+00 -7.55171537e-01 -6.10106289e-01
7.44011819e-01 -5.06147921e-01 9.64688897e-01 4.60700274e-01
3.14947128e-01 8.81449580e-02 -1.07417576e-01 1.00212738e-01
2.65522808e-01 1.94259197e-01 6.24658287e-01 -1.27743459e+00
-3.40213329e-01 -6.70769691e-01 -5.42303324e-01 -1.43052292e+00
1.09534308e-01 -6.26280785e-01 8.18779245e-02 -1.66737747e+00
2.85456151e-01 -7.08268881e-01 -5.79742454e-02 3.04720879e-01
-1.43989727e-01 -8.54044780e-02 1.07524164e-01 -3.87045354e-01
-1.85179710e-01 4.40122753e-01 1.36545694e+00 1.84119791e-01
-2.00186595e-01 5.27192414e-01 -4.81223345e-01 1.20186877e+00
8.49346399e-01 -3.05959046e-01 -6.43045902e-01 -7.44912326e-01
6.27369642e-01 2.13839680e-01 6.45567656e-01 -1.14450562e+00
-2.23820090e-01 -2.64957041e-01 6.94596469e-01 -8.10006857e-01
2.85532773e-01 -6.80065930e-01 1.84838161e-01 8.31879318e-01
-1.62821949e-01 4.44202088e-02 5.60885549e-01 3.47983360e-01
4.30619836e-01 -2.34084845e-01 8.50077152e-01 -1.91638052e-01
-4.18906778e-01 4.98869777e-01 -6.46054387e-01 1.11648344e-01
1.21250784e+00 -2.45994464e-01 -1.44843906e-01 -3.88770640e-01
-9.98943865e-01 3.13992143e-01 7.70150661e-01 6.08644843e-01
4.66527075e-01 -1.42354608e+00 -4.76891696e-01 4.97272670e-01
-2.11960956e-01 3.56653631e-01 2.40483612e-01 3.35176170e-01
-1.37035942e+00 2.72577912e-01 -4.30392414e-01 -5.21495759e-01
-9.75322545e-01 8.73546541e-01 4.20606077e-01 -2.46597946e-01
-9.80692744e-01 1.01874912e+00 8.41946185e-01 -6.30513012e-01
-9.81501639e-02 -1.14605618e+00 -2.69914269e-01 -6.21434033e-01
3.60445112e-01 3.09145004e-01 -2.28938445e-01 -1.29999802e-01
-2.48611644e-01 7.94884026e-01 2.11870953e-01 2.04655662e-01
1.72253084e+00 1.77885279e-01 2.60237213e-02 4.89577383e-01
8.45994413e-01 -1.65625259e-01 -1.58186042e+00 3.14226806e-01
-8.23305696e-02 -2.29139775e-01 -5.02847254e-01 -5.66998839e-01
-8.19123089e-01 1.31881309e+00 2.23160610e-01 3.59177858e-01
8.33822846e-01 2.35802203e-01 8.42194259e-01 4.52941120e-01
6.70867026e-01 -4.42901611e-01 1.74727380e-01 5.65216064e-01
9.93969858e-01 -8.91963899e-01 -8.16311035e-03 -2.60504931e-01
-1.08242668e-01 1.11729848e+00 8.04506361e-01 -5.70373714e-01
8.63217413e-01 5.61217427e-01 -5.49589992e-01 -2.36920938e-01
-1.12541330e+00 7.19755515e-02 2.45360106e-01 9.21887457e-01
8.30138251e-02 4.03662622e-01 3.69792670e-01 7.06233203e-01
-9.20478702e-01 -1.26939164e-02 5.43509066e-01 6.64412141e-01
-5.10348499e-01 -5.62900364e-01 -5.52286021e-02 6.08868957e-01
-2.01339796e-01 -1.75605580e-01 -4.41259407e-02 8.76535237e-01
1.88155115e-01 4.09871787e-01 9.30152237e-02 -3.66721898e-01
3.29569638e-01 -2.18145043e-01 1.25951076e+00 -8.85109961e-01
-1.84505656e-01 -7.46757448e-01 -1.07547469e-01 -5.10346770e-01
1.53069794e-01 -6.96499109e-01 -1.24335301e+00 -5.32399714e-01
-1.16765261e-01 -3.48016441e-01 2.39936143e-01 8.58782232e-01
-2.24426419e-01 6.76254988e-01 4.95694190e-01 -1.27936053e+00
-2.08123833e-01 -9.61778164e-01 -5.40155470e-01 2.26973563e-01
4.85159039e-01 -1.08052421e+00 -3.76601145e-02 4.13259231e-02] | [8.404227256774902, 0.9464970827102661] |
b67b6560-46ac-4213-b76c-fc2d8579a28e | benchmark-of-deep-learning-models-on-large | 1710.08531 | null | http://arxiv.org/abs/1710.08531v1 | http://arxiv.org/pdf/1710.08531v1.pdf | Benchmark of Deep Learning Models on Large Healthcare MIMIC Datasets | Deep learning models (aka Deep Neural Networks) have revolutionized many
fields including computer vision, natural language processing, speech
recognition, and is being increasingly used in clinical healthcare
applications. However, few works exist which have benchmarked the performance
of the deep learning models with respect to the state-of-the-art machine
learning models and prognostic scoring systems on publicly available healthcare
datasets. In this paper, we present the benchmarking results for several
clinical prediction tasks such as mortality prediction, length of stay
prediction, and ICD-9 code group prediction using Deep Learning models,
ensemble of machine learning models (Super Learner algorithm), SAPS II and SOFA
scores. We used the Medical Information Mart for Intensive Care III (MIMIC-III)
(v1.4) publicly available dataset, which includes all patients admitted to an
ICU at the Beth Israel Deaconess Medical Center from 2001 to 2012, for the
benchmarking tasks. Our results show that deep learning models consistently
outperform all the other approaches especially when the `raw' clinical time
series data is used as input features to the models. | ['Sanjay Purushotham', 'Chuizheng Meng', 'Zhengping Che', 'Yan Liu'] | 2017-10-23 | null | null | null | null | ['length-of-stay-prediction'] | ['medical'] | [-2.74918437e-01 -3.17823648e-01 -1.08363688e-01 -4.50782418e-01
-7.89955616e-01 8.27959105e-02 1.23602934e-01 6.98159575e-01
-5.72427809e-01 7.61809766e-01 5.20227373e-01 -6.85868442e-01
-6.05388105e-01 -6.24788046e-01 4.05568928e-02 -8.18103731e-01
-6.57060385e-01 1.01443315e+00 -2.78686106e-01 3.82203492e-03
-1.08730905e-01 4.97059137e-01 -1.01427186e+00 8.04075181e-01
4.28183526e-01 1.30584657e+00 -3.64098519e-01 8.86107206e-01
1.92889675e-01 1.23272562e+00 -3.12042654e-01 -6.56541884e-02
-4.98702973e-02 -3.71943265e-01 -9.23588395e-01 -7.68732309e-01
-4.00713384e-01 -2.36738995e-01 -3.91210318e-01 1.72293872e-01
1.15716600e+00 -1.79709688e-01 8.82041454e-01 -9.08563018e-01
-2.82698303e-01 4.11134809e-01 3.91640179e-02 5.86930811e-01
-1.35385916e-01 1.01428427e-01 6.43300891e-01 -7.00688779e-01
3.48179698e-01 8.66130590e-01 1.33996105e+00 5.82252920e-01
-6.94411874e-01 -5.29616773e-01 -5.15523374e-01 6.04788780e-01
-1.18314052e+00 -6.34507015e-02 2.47771934e-01 -7.42372215e-01
1.16816890e+00 4.34633829e-02 4.56428498e-01 1.39021623e+00
7.82338321e-01 5.04397511e-01 8.69877100e-01 -1.61802351e-01
1.19342953e-01 -1.02580711e-01 5.27427495e-01 7.17627168e-01
-6.83098286e-02 3.26044410e-01 -3.99189591e-01 -6.53670728e-01
3.73977035e-01 5.07219374e-01 -1.43593460e-01 -6.65465966e-02
-1.36229968e+00 9.18357611e-01 1.22358918e-01 5.36893666e-01
-9.07135129e-01 -7.51769394e-02 1.06156647e+00 5.91668904e-01
5.50009549e-01 1.68878540e-01 -1.32745028e+00 -3.33685577e-01
-5.86073935e-01 1.15176737e-02 7.31225014e-01 2.11201712e-01
-1.30211294e-01 6.71755057e-03 -3.17838699e-01 1.04999542e+00
2.60914505e-01 1.34007022e-01 1.23264611e+00 -5.76972425e-01
2.66384870e-01 6.18167698e-01 -6.18038438e-02 -6.05118096e-01
-1.04327428e+00 -6.63981020e-01 -1.42219245e+00 -3.44759449e-02
7.90319592e-02 -6.56306744e-01 -7.62026072e-01 1.30517244e+00
-1.16971798e-01 1.50816828e-01 5.98320782e-01 5.61904967e-01
1.25574005e+00 5.24694979e-01 3.67202491e-01 -3.56722504e-01
1.27819932e+00 -7.95897901e-01 -6.02550447e-01 3.06048393e-01
1.33620906e+00 -4.06635523e-01 4.94313091e-01 7.98300743e-01
-7.90013313e-01 -4.57438439e-01 -5.04627943e-01 2.28455260e-01
-4.50339228e-01 1.79646164e-01 4.41923171e-01 2.56058961e-01
-1.04580796e+00 9.38673973e-01 -9.07024026e-01 -6.24039531e-01
6.45833969e-01 5.95220566e-01 -6.28240466e-01 1.15396651e-02
-1.36510432e+00 1.22599745e+00 4.78970438e-01 -1.72127813e-01
-1.07063818e+00 -7.92141557e-01 -4.93659198e-01 1.27216473e-01
-4.62767363e-01 -1.25760019e+00 1.22828293e+00 -8.29334319e-01
-1.15250492e+00 1.03598893e+00 2.31882706e-01 -8.28231394e-01
3.91507626e-01 -3.61173987e-01 -5.14946163e-01 -1.50548413e-01
-5.02734661e-01 5.65612242e-02 2.14339823e-01 -6.74848020e-01
-5.03999949e-01 -7.02733040e-01 -5.04430413e-01 -2.01417699e-01
-1.87024325e-01 3.71000797e-01 5.75855494e-01 -6.56965554e-01
-3.19243401e-01 -7.19364524e-01 -4.50406075e-01 -5.43739140e-01
-1.47583395e-01 -3.66125733e-01 6.32931292e-01 -8.81565332e-01
1.47905743e+00 -2.17999077e+00 1.51380703e-01 -1.26539525e-02
1.66581929e-01 5.28364360e-01 1.11607283e-01 7.60305822e-01
-5.84676147e-01 9.72866639e-02 -1.42315850e-01 -4.67158586e-01
-5.80541193e-01 4.13203359e-01 3.73364948e-02 4.06051844e-01
-2.35766470e-01 8.21083724e-01 -7.46453822e-01 -4.64731842e-01
4.17649657e-01 5.21516740e-01 -3.50618660e-01 6.40753090e-01
4.42969859e-01 6.94437087e-01 -3.03451121e-01 4.60678101e-01
3.92771214e-02 -2.77030081e-01 -1.09546423e-01 1.75945461e-01
1.32545456e-01 1.23372087e-02 -2.96879679e-01 1.57582748e+00
-4.45060879e-01 5.33757627e-01 -4.88312691e-01 -1.45345759e+00
9.24942315e-01 8.80375862e-01 1.02372503e+00 -3.47570002e-01
5.65751076e-01 2.19024628e-01 2.88707048e-01 -9.47764814e-01
-6.85161352e-01 -5.93293726e-01 1.23519301e-01 9.45549086e-02
2.18445301e-01 4.18412119e-01 -4.41966891e-01 -3.00488323e-01
1.38314486e+00 -4.46959525e-01 6.21096909e-01 -2.87219405e-01
6.41621292e-01 -2.64462903e-02 6.91652358e-01 6.36229217e-01
-5.29752493e-01 7.82135725e-01 4.21478629e-01 -1.29474413e+00
-8.86205018e-01 -9.09610689e-01 -4.29456174e-01 1.01107514e+00
-7.87441015e-01 -2.74561971e-01 -5.52075326e-01 -4.87872690e-01
4.41751350e-03 5.73362291e-01 -1.01117158e+00 -5.03533006e-01
-5.61109006e-01 -1.21270573e+00 8.51745486e-01 8.45654011e-01
6.81738928e-02 -1.43157649e+00 -9.22785163e-01 5.46931863e-01
-6.14477955e-02 -5.16142905e-01 2.70879328e-01 5.65157354e-01
-1.30851793e+00 -1.36697614e+00 -7.52955496e-01 -8.49722087e-01
-4.20086607e-02 -6.57939553e-01 1.13792992e+00 2.60091405e-02
-5.24349570e-01 1.44546062e-01 -4.28933412e-01 -8.54620278e-01
-5.62361062e-01 9.87167358e-02 3.52813721e-01 2.83378772e-02
5.29215753e-01 -5.57206869e-01 -8.07947576e-01 -2.33121678e-01
-5.91509283e-01 1.22154877e-01 5.84086239e-01 1.18710887e+00
4.02227074e-01 -4.45857823e-01 1.10888362e+00 -9.23154235e-01
6.88623667e-01 -9.96024311e-01 2.16208190e-01 8.09375197e-02
-9.32981551e-01 -1.89168528e-01 8.21762085e-01 1.59952845e-02
-5.77542305e-01 -2.79064417e-01 -5.21434426e-01 -3.89540851e-01
-5.31861603e-01 8.54654431e-01 4.55269814e-01 5.12992382e-01
6.28337741e-01 1.32827967e-01 -8.20841826e-03 -8.11801434e-01
-3.16094995e-01 1.16535223e+00 3.94265562e-01 -4.92892005e-02
-9.60513055e-02 2.41275191e-01 9.66375619e-02 -3.43645722e-01
-8.54419410e-01 -7.48133123e-01 -8.59313309e-01 1.21450722e-02
1.18907344e+00 -8.78431499e-01 -7.07815409e-01 6.89856350e-01
-1.14034235e+00 -2.84316689e-01 -1.75497625e-02 7.90065229e-01
-7.60668039e-01 -7.90511370e-02 -9.65756595e-01 -5.56597352e-01
-1.06457996e+00 -9.85683203e-01 7.90307701e-01 -3.30169529e-01
-5.39749384e-01 -1.41823792e+00 6.53651834e-01 1.71456113e-01
6.35094583e-01 9.69365656e-01 1.61175644e+00 -1.24203193e+00
6.02400482e-01 -4.98191416e-01 3.58063020e-02 8.15644562e-01
2.25141153e-01 -1.00137942e-01 -8.66753399e-01 -2.64485985e-01
-3.82056041e-03 -2.57961124e-01 8.88827801e-01 5.85309982e-01
1.67956090e+00 -2.00545594e-01 -2.79417098e-01 6.84839308e-01
1.42486680e+00 7.52123237e-01 4.16313171e-01 4.70160186e-01
5.32540619e-01 2.74779379e-01 8.42682868e-02 8.95630002e-01
3.39233339e-01 1.51409328e-01 3.66161585e-01 -1.63648382e-01
3.59681845e-01 5.04142702e-01 6.83253184e-02 1.16194236e+00
-3.94282073e-01 -1.19145632e-01 -1.65621018e+00 5.59835017e-01
-2.08466578e+00 -6.95967436e-01 -3.19433659e-01 1.96652782e+00
6.96907878e-01 -2.00242355e-01 1.26055535e-02 5.08126915e-01
2.38943711e-01 -3.43111008e-01 -3.06725085e-01 -8.76847267e-01
1.64673463e-01 5.10350406e-01 1.93219945e-01 3.04696169e-02
-1.26314044e+00 4.49305952e-01 6.26199198e+00 2.28962019e-01
-1.17749214e+00 5.83108008e-01 1.08031952e+00 -2.06885383e-01
4.58441079e-01 -6.99494123e-01 -2.89086159e-02 4.00807172e-01
1.84078264e+00 -1.38553157e-01 -3.40090990e-02 8.66856933e-01
5.26195765e-01 3.35587740e-01 -1.31980073e+00 1.16668987e+00
-1.34084508e-01 -1.20827579e+00 -8.37712437e-02 -1.51818588e-01
5.40354788e-01 5.53288639e-01 -3.27310413e-02 5.65396070e-01
1.55311808e-01 -1.45001197e+00 4.92468290e-02 8.13066959e-01
7.91431129e-01 -9.56039310e-01 1.74303770e+00 4.69468415e-01
-4.01149780e-01 -5.21524012e-01 -5.47231808e-02 -1.14043541e-01
3.62920761e-02 5.28825939e-01 -7.93922365e-01 5.41021705e-01
1.10242486e+00 6.98798001e-01 -2.78135329e-01 9.75715935e-01
4.39439207e-01 1.01734865e+00 1.38324574e-01 2.66144276e-01
3.86312842e-01 3.68550152e-01 2.43042270e-03 1.35614681e+00
3.72103155e-01 4.20247704e-01 1.31072355e-02 -3.12711555e-03
-2.64120866e-02 5.13739347e-01 -6.88521802e-01 1.75659463e-01
-3.36829513e-01 8.58558714e-01 -9.06789824e-02 -5.97058475e-01
-4.10010248e-01 4.89051670e-01 7.41329193e-02 1.50723204e-01
-6.99496150e-01 -2.51866966e-01 6.40874386e-01 1.51107609e-01
-2.14072540e-01 2.08668426e-01 -6.74891293e-01 -1.01947117e+00
-6.80445671e-01 -8.98064911e-01 9.36636627e-01 -6.80912733e-01
-1.53691828e+00 8.91804099e-01 -1.83247998e-01 -1.17790592e+00
-4.74794805e-01 -8.29818428e-01 -6.73727334e-01 8.80212009e-01
-1.29941940e+00 -7.59329677e-01 -3.59960079e-01 7.75940478e-01
3.68698329e-01 -6.41624093e-01 1.50375497e+00 4.63854879e-01
-5.86734235e-01 3.41288060e-01 6.31169081e-01 3.89786392e-01
8.50065053e-01 -9.98739302e-01 -1.27313942e-01 5.10595292e-02
-5.54920137e-01 3.89340937e-01 3.85693997e-01 -1.92019254e-01
-7.55759656e-01 -1.35298955e+00 1.28175652e+00 -6.27777457e-01
4.88653839e-01 3.19682479e-01 -9.20953155e-01 6.62356853e-01
2.77351648e-01 1.43379763e-01 1.39628303e+00 -5.18835410e-02
7.83918332e-03 -2.74510860e-01 -1.19857073e+00 -3.51544358e-02
5.39630413e-01 -1.03848539e-01 -6.26714289e-01 4.60925847e-01
2.96755016e-01 -1.60082564e-01 -1.46847260e+00 1.08104587e+00
7.41254032e-01 -1.02252829e+00 8.58756125e-01 -1.62069213e+00
7.46898174e-01 3.85453939e-01 -6.60372674e-02 -1.43473113e+00
-5.22426307e-01 -1.65132955e-01 -3.14537197e-01 5.46995878e-01
2.67892212e-01 -6.09310329e-01 3.83690566e-01 3.21006596e-01
-3.11091572e-01 -1.64762640e+00 -1.08733225e+00 -4.37857300e-01
5.65804005e-01 -4.19323355e-01 4.87718225e-01 1.00506246e+00
-7.45066702e-02 1.62192300e-01 -4.38654989e-01 -1.48647189e-01
1.11805975e-01 -1.37214407e-01 2.92479992e-01 -1.65583932e+00
-1.88998878e-01 -4.70507026e-01 -7.06795394e-01 3.30768794e-01
-1.02950633e-01 -7.33936965e-01 -2.92672366e-01 -2.00366831e+00
2.42310658e-01 -4.20125633e-01 -1.26945436e+00 7.02904820e-01
-3.51217724e-02 -7.99425915e-02 -1.41765565e-01 1.98603287e-01
-3.34630221e-01 2.53191382e-01 5.78491449e-01 7.08290488e-02
-1.55507639e-01 2.74162561e-01 -4.52357858e-01 9.41160262e-01
1.22099638e+00 -6.08105600e-01 -1.01910226e-01 -4.75780636e-01
-1.78881899e-01 5.34211516e-01 1.33374244e-01 -1.18908668e+00
-1.32658929e-01 -7.49732554e-02 6.10231221e-01 -3.60427856e-01
-3.85417864e-02 -6.82838202e-01 1.74571410e-01 1.08725512e+00
-4.74492610e-01 6.07650042e-01 2.41658762e-01 1.63331404e-01
-4.44209218e-01 5.99652827e-02 1.13237691e+00 -2.60214359e-01
-3.80179793e-01 7.56543100e-01 -6.89015388e-01 -6.84536025e-02
1.14085090e+00 1.49625406e-01 -2.03324109e-02 -1.27712473e-01
-1.18079269e+00 -3.69555317e-02 -2.36134171e-01 5.56315303e-01
6.49689138e-01 -1.33681309e+00 -1.12171865e+00 1.01337284e-01
4.14374888e-01 -1.25983030e-01 4.49287415e-01 1.24531949e+00
-9.80974197e-01 6.66400850e-01 -3.84660065e-01 -3.85278821e-01
-1.41406333e+00 7.41676271e-01 6.14680648e-01 -6.39569879e-01
-7.36300886e-01 7.55414069e-01 1.44784704e-01 -2.70159602e-01
5.09564281e-01 -5.29437780e-01 -7.18851507e-01 -4.01221365e-02
5.72089016e-01 4.93087918e-01 4.08648312e-01 -4.50471312e-01
-6.69609308e-01 2.58262962e-01 5.46707548e-02 4.12169844e-01
1.76787293e+00 3.84328932e-01 -1.97553933e-01 8.34354281e-01
1.53545821e+00 -9.60461140e-01 -1.69003710e-01 -9.61366296e-02
2.09527627e-01 2.23805413e-01 1.80336133e-01 -1.08106160e+00
-1.13120925e+00 1.39231610e+00 1.20166993e+00 -4.22485918e-02
1.25326085e+00 -1.17961533e-01 1.00148904e+00 5.35545409e-01
2.69519389e-01 -6.96890950e-01 -1.49762765e-01 7.14036167e-01
8.21586847e-01 -1.36549366e+00 -6.31658137e-01 4.12248880e-01
-7.82709956e-01 1.38092041e+00 2.92400479e-01 -1.68860689e-01
1.28397739e+00 8.31081644e-02 4.51723754e-01 -8.06074217e-02
-1.25225937e+00 1.55441567e-01 6.48647770e-02 5.21479130e-01
8.12188745e-01 3.80120933e-01 -3.48059416e-01 1.23075855e+00
-1.12711797e-02 5.66881955e-01 2.77522087e-01 6.86077237e-01
-2.98560649e-01 -9.00146186e-01 -1.55710080e-03 9.46080089e-01
-1.20375550e+00 -3.81195635e-01 -3.39053720e-01 4.08076376e-01
3.90916049e-01 7.47670472e-01 -2.13723838e-01 -5.26273251e-01
4.24447715e-01 5.64576507e-01 -8.74007866e-03 -5.20249724e-01
-1.04502237e+00 -5.87539554e-01 6.14338107e-02 -3.83992583e-01
-2.66430467e-01 -4.67529297e-01 -1.25592339e+00 -1.70625448e-01
2.73541242e-01 2.15758547e-01 5.45533538e-01 9.99356747e-01
6.75209761e-01 8.95326793e-01 4.56376314e-01 -3.19429219e-01
-5.45938730e-01 -1.44234443e+00 -4.74614739e-01 4.36644316e-01
6.35649085e-01 -5.33963084e-01 -1.80649221e-01 1.75900444e-01] | [8.020005226135254, 6.237532615661621] |
d2d75bbf-0d10-49ef-b888-3e8b84267b21 | semantic-aware-generation-of-multi-view | 2305.02618 | null | https://arxiv.org/abs/2305.02618v1 | https://arxiv.org/pdf/2305.02618v1.pdf | Semantic-aware Generation of Multi-view Portrait Drawings | Neural radiance fields (NeRF) based methods have shown amazing performance in synthesizing 3D-consistent photographic images, but fail to generate multi-view portrait drawings. The key is that the basic assumption of these methods -- a surface point is consistent when rendered from different views -- doesn't hold for drawings. In a portrait drawing, the appearance of a facial point may changes when viewed from different angles. Besides, portrait drawings usually present little 3D information and suffer from insufficient training data. To combat this challenge, in this paper, we propose a Semantic-Aware GEnerator (SAGE) for synthesizing multi-view portrait drawings. Our motivation is that facial semantic labels are view-consistent and correlate with drawing techniques. We therefore propose to collaboratively synthesize multi-view semantic maps and the corresponding portrait drawings. To facilitate training, we design a semantic-aware domain translator, which generates portrait drawings based on features of photographic faces. In addition, use data augmentation via synthesis to mitigate collapsed results. We apply SAGE to synthesize multi-view portrait drawings in diverse artistic styles. Experimental results show that SAGE achieves significantly superior or highly competitive performance, compared to existing 3D-aware image synthesis methods. The codes are available at https://github.com/AiArt-HDU/SAGE. | ['Gang Xu', 'Nannan Wang', 'Chang Jiang', 'Fei Gao', 'Biao Ma'] | 2023-05-04 | null | null | null | null | ['3d-aware-image-synthesis'] | ['computer-vision'] | [ 2.98923850e-01 -2.05132559e-01 4.79635084e-03 -5.25647044e-01
-4.23488498e-01 -7.98888326e-01 5.10512233e-01 -8.73550296e-01
5.72861731e-01 3.93323988e-01 1.21462591e-01 1.28681839e-01
1.98562890e-01 -1.01201916e+00 -6.78419352e-01 -2.59759009e-01
7.03875065e-01 1.65662453e-01 -3.18013936e-01 -5.21661997e-01
2.66080528e-01 7.89647043e-01 -1.49403358e+00 5.70576966e-01
6.72636628e-01 8.25533986e-01 1.72093049e-01 5.34904242e-01
-3.35509598e-01 4.90323663e-01 -7.24537194e-01 -8.13298643e-01
4.63541597e-01 -7.44207382e-01 -3.98933083e-01 6.08437359e-01
1.09028256e+00 -4.50817317e-01 -2.21823663e-01 9.86949205e-01
4.04364139e-01 -2.12765902e-01 8.42779815e-01 -1.64836097e+00
-1.41893280e+00 9.55679417e-02 -9.66482043e-01 -3.25753987e-01
5.90465248e-01 -7.45862871e-02 7.92300284e-01 -1.05933797e+00
9.62399960e-01 1.50859344e+00 7.98220217e-01 8.23316813e-01
-1.02880991e+00 -9.19932425e-01 3.68106693e-01 -9.13915262e-02
-1.62036836e+00 -4.38172132e-01 1.44829595e+00 -1.33861721e-01
4.22613680e-01 3.97456974e-01 8.18510652e-01 1.38094592e+00
7.19278380e-02 7.65707612e-01 1.24317515e+00 -1.96947992e-01
1.25938356e-01 2.23129913e-01 -7.55190551e-01 6.42551064e-01
-9.08059105e-02 -3.57699662e-01 -5.88292062e-01 1.54481426e-01
1.28393078e+00 9.84494239e-02 -1.81067675e-01 -1.97781816e-01
-9.59675968e-01 5.78438818e-01 5.06346107e-01 2.71050334e-01
-1.49670333e-01 1.34299353e-01 5.92200384e-02 2.88876504e-01
6.81111634e-01 2.60671258e-01 -2.73448955e-02 2.34288633e-01
-8.88819098e-01 2.63897628e-01 3.82758796e-01 1.49033701e+00
6.22693658e-01 3.16939354e-01 1.72578320e-01 1.10745132e+00
1.09459877e-01 7.81519473e-01 1.14387408e-01 -1.08832598e+00
6.08645856e-01 7.48629630e-01 -1.52622357e-01 -1.36314309e+00
-1.43855810e-01 -5.91690727e-02 -1.15332007e+00 3.01572859e-01
-7.78601989e-02 6.36761077e-03 -9.06074524e-01 1.46682334e+00
2.24433154e-01 9.48334765e-03 2.58626100e-02 1.01907921e+00
9.75699961e-01 7.53205180e-01 -2.13399813e-01 2.75184214e-01
1.29940319e+00 -8.28853905e-01 -7.98615158e-01 -2.55402565e-01
8.81497338e-02 -9.51385200e-01 1.18218780e+00 3.96113962e-01
-1.16072357e+00 -7.61181295e-01 -1.00061762e+00 -3.27752441e-01
-4.61083472e-01 4.26762849e-01 5.37608266e-01 7.04634607e-01
-1.02746749e+00 3.21322739e-01 -1.52284756e-01 -3.50689471e-01
6.82914913e-01 -1.64798066e-01 -4.84556824e-01 -2.80833513e-01
-1.01444387e+00 7.53407896e-01 -1.03478841e-01 -1.72156151e-02
-4.08515990e-01 -8.77851486e-01 -7.55062461e-01 -1.04284272e-01
8.03728476e-02 -9.24908936e-01 1.07549453e+00 -1.47066963e+00
-1.54174972e+00 9.58460867e-01 -1.89241543e-01 2.94632822e-01
6.67891741e-01 2.66259223e-01 -7.62686849e-01 4.49611604e-01
1.50497451e-01 1.11986053e+00 1.04428434e+00 -1.74308562e+00
-3.83301049e-01 -3.62572044e-01 3.62361856e-02 4.06238347e-01
-3.58847111e-01 -2.87107885e-01 -4.99700308e-01 -9.78208959e-01
2.12874487e-01 -8.23786736e-01 8.97119269e-02 4.96263653e-01
-4.89805520e-01 4.06481996e-02 1.04537606e+00 -5.50403237e-01
7.31488407e-01 -2.17030978e+00 -1.08039051e-01 2.31046274e-01
4.43127472e-03 1.82400513e-02 -4.25788730e-01 5.75691283e-01
-1.88767210e-01 3.04468840e-01 -1.73835143e-01 -3.55609268e-01
-1.11097604e-01 1.07898094e-01 -4.81057823e-01 2.33142421e-01
3.87072861e-01 9.62771535e-01 -7.38688111e-01 -3.87552828e-01
4.15596217e-01 7.87337065e-01 -4.37492341e-01 -3.67923416e-02
-2.40924627e-01 4.24332321e-01 -5.57934284e-01 7.92553008e-01
1.09372115e+00 -3.57954085e-01 1.40636683e-01 -4.45811003e-01
5.32032549e-02 -1.95671394e-01 -1.13073838e+00 2.13943958e+00
-8.21596682e-01 7.42883444e-01 -2.46958688e-01 -5.12475133e-01
1.30483842e+00 2.27172345e-01 2.83979654e-01 -8.11016560e-01
7.02669695e-02 -8.66085961e-02 -5.97317576e-01 -2.94614911e-01
6.31132007e-01 -3.06593090e-01 -4.86707594e-03 4.03341353e-01
-1.53248042e-01 -7.30991840e-01 -1.16548702e-01 2.41108641e-01
5.36804974e-01 2.88001657e-01 1.08443186e-01 1.15799621e-01
3.71968269e-01 -2.16404572e-01 3.34410936e-01 9.26269665e-02
3.94153208e-01 1.29788530e+00 5.76156437e-01 -5.88163793e-01
-1.21680057e+00 -1.16761827e+00 1.97834037e-02 5.93792021e-01
3.21704686e-01 -3.69867563e-01 -8.63000035e-01 -4.86899137e-01
-6.64320365e-02 8.56176138e-01 -7.66566932e-01 1.18939400e-01
-4.01731640e-01 -2.70472258e-01 4.30264443e-01 5.65594554e-01
7.97004580e-01 -6.67096853e-01 -4.03347552e-01 -2.52028167e-01
-1.61057502e-01 -1.24187088e+00 -7.77640462e-01 -6.39907181e-01
-7.02411056e-01 -9.31192100e-01 -1.01478517e+00 -8.72206092e-01
1.18403304e+00 8.14174175e-01 1.08033848e+00 -1.83292925e-02
-2.82461584e-01 5.58963954e-01 -3.80758137e-01 -2.76528418e-01
-5.11086047e-01 -1.32637218e-01 -9.07009393e-02 1.95081234e-01
-2.07667192e-03 -8.63229752e-01 -7.46339560e-01 5.92187881e-01
-1.12756193e+00 6.28680706e-01 5.65525889e-01 5.11710703e-01
5.72550476e-01 -8.94843340e-02 5.82453549e-01 -8.96055222e-01
5.25689900e-01 -1.46446437e-01 -3.63571823e-01 4.83389676e-01
-2.98242152e-01 -2.49154747e-01 1.04259419e+00 -4.28399771e-01
-1.23858440e+00 1.22306645e-02 -6.98091909e-02 -9.82460976e-01
-2.59166926e-01 -2.32594937e-01 -4.46550488e-01 -1.45323634e-01
6.16866469e-01 3.45554620e-01 6.94415793e-02 -2.57401705e-01
7.47195661e-01 5.91516435e-01 4.34362531e-01 -3.52408558e-01
1.10321987e+00 7.21964419e-01 -4.60951552e-02 -8.79959226e-01
-6.16777778e-01 2.24948525e-02 -6.54454648e-01 -6.08267903e-01
7.16727436e-01 -1.01613879e+00 -4.50327367e-01 3.34926218e-01
-1.29733431e+00 -8.60163718e-02 -1.17530987e-01 -2.15844601e-01
-6.36550665e-01 2.98938274e-01 -2.41091207e-01 -4.69725490e-01
-2.34944090e-01 -9.56998885e-01 1.42629075e+00 4.13021713e-01
-2.89157659e-01 -9.89606917e-01 -2.74494380e-01 4.98985887e-01
2.98935652e-01 4.64203566e-01 7.80906737e-01 1.33081391e-01
-5.65596938e-01 -1.18857466e-01 -4.28629279e-01 3.83105695e-01
4.81777877e-01 2.13178769e-01 -1.12812853e+00 7.42340237e-02
-3.07522416e-01 -1.88578427e-01 3.50709170e-01 2.17360988e-01
1.37132943e+00 -3.61538470e-01 -2.01489374e-01 8.53795707e-01
1.42850327e+00 1.96755633e-01 7.90843189e-01 -1.69558693e-02
9.42811549e-01 6.30222082e-01 3.22216421e-01 4.12270069e-01
4.36607689e-01 7.29819775e-01 1.43052056e-01 -1.54867053e-01
-4.76593554e-01 -9.36863661e-01 1.75884187e-01 7.62391329e-01
-9.68775526e-02 -4.81318295e-01 -4.97907132e-01 2.39273638e-01
-1.32503438e+00 -1.09023881e+00 -1.22353151e-01 1.64811206e+00
5.95075369e-01 -3.82977009e-01 -3.24573368e-02 -3.32080722e-02
7.45895267e-01 2.87785232e-01 -6.74048960e-01 -5.50762415e-01
-2.98534095e-01 -4.78268676e-02 3.53993535e-01 1.69576302e-01
-6.34747565e-01 1.04491663e+00 5.77506590e+00 8.42155814e-01
-1.25949144e+00 -8.44647586e-02 8.85202527e-01 -1.49809405e-01
-8.47164810e-01 -2.12053224e-01 -5.39549470e-01 3.46550465e-01
2.58962721e-01 -9.50205028e-02 4.79821324e-01 9.38420892e-01
9.95723978e-02 2.08435208e-01 -9.37936604e-01 1.52033508e+00
6.44270658e-01 -1.54615891e+00 5.91336429e-01 -3.74328978e-02
1.16095793e+00 -6.89770103e-01 4.66022164e-01 -2.54050761e-01
8.88758451e-02 -9.70704615e-01 8.51262808e-01 5.07254124e-01
1.25586879e+00 -7.82480717e-01 4.91208062e-02 -9.16793942e-02
-1.14100969e+00 2.49895930e-01 -3.67938846e-01 2.98736066e-01
1.97655186e-01 3.53174448e-01 -7.09646404e-01 8.01868081e-01
4.87223834e-01 1.04412508e+00 -5.73728144e-01 5.07449806e-01
-3.46220970e-01 -2.07155943e-01 5.54849058e-02 -1.76597741e-02
1.38527513e-01 -4.84962523e-01 2.52249807e-01 7.58239746e-01
7.72191882e-01 2.21245512e-01 -3.61887306e-01 1.30923307e+00
-3.82075697e-01 -1.63352620e-02 -1.07209694e+00 6.43345490e-02
5.37056923e-01 1.45331693e+00 -7.65663207e-01 -2.02706590e-01
-4.38667744e-01 1.64212799e+00 -4.43001203e-02 4.56492037e-01
-9.23698187e-01 -2.67253548e-01 6.14251316e-01 2.32575074e-01
1.52659625e-01 -1.32635683e-01 -5.38049281e-01 -1.04410005e+00
2.34669104e-01 -7.88196564e-01 -1.23484224e-01 -1.58597898e+00
-1.51690400e+00 7.50865579e-01 -5.68489358e-02 -1.64162409e+00
-4.85152155e-02 -5.60623944e-01 -6.26909971e-01 7.22953498e-01
-1.33049250e+00 -1.64518869e+00 -5.84504604e-01 5.49766958e-01
8.61667573e-01 -1.66860282e-01 6.85418248e-01 1.85598224e-01
-3.29788983e-01 6.93038881e-01 -8.71986970e-02 1.53403237e-01
7.43158817e-01 -1.00976622e+00 9.11938727e-01 5.40412068e-01
2.65891463e-01 4.24537241e-01 3.19875032e-01 -6.24209940e-01
-1.69423103e+00 -1.31997550e+00 5.23652971e-01 -5.79990447e-01
2.72807002e-01 -4.75028247e-01 -6.83129489e-01 4.83915299e-01
3.65840375e-01 -2.36828715e-01 7.09135592e-01 -2.79218435e-01
-6.44619346e-01 -1.84024602e-01 -1.23402166e+00 9.98810232e-01
1.49520838e+00 -5.49339175e-01 -2.62643456e-01 3.66746217e-01
3.37679148e-01 -4.78417367e-01 -8.75837088e-01 -2.36727074e-02
7.93651402e-01 -1.12523377e+00 1.18859005e+00 -1.57861382e-01
7.98425853e-01 -3.07709962e-01 -1.44436195e-01 -1.60449326e+00
-1.78451940e-01 -5.62633634e-01 5.57353675e-01 1.40188932e+00
1.61247239e-01 -4.73525077e-01 6.83246613e-01 5.52862465e-01
-6.62411675e-02 -4.64592308e-01 -7.48088360e-01 -7.79869378e-01
4.76284549e-02 -2.20873162e-01 9.46031570e-01 1.22923577e+00
-4.07286257e-01 3.88564974e-01 -5.37807643e-01 6.95242062e-02
4.46930945e-01 3.42186421e-01 9.37803864e-01 -1.04546809e+00
2.46955007e-01 -5.00494838e-01 -3.51754189e-01 -7.55321085e-01
1.94290280e-01 -9.51331675e-01 -3.40575308e-01 -1.67494416e+00
-4.67923358e-02 -4.31494713e-01 4.22165841e-01 3.81763041e-01
2.01643333e-01 5.43007016e-01 3.87716591e-01 5.73034436e-02
-2.16913372e-01 6.47577107e-01 1.93277299e+00 -9.80108008e-02
-7.43699968e-02 -3.38156313e-01 -1.03852654e+00 6.15819216e-01
9.32187259e-01 -1.30840063e-01 -7.27464616e-01 -7.08664238e-01
3.38631988e-01 -3.22689936e-02 5.23774922e-01 -9.49593961e-01
-4.62058485e-02 -2.29686946e-01 9.49472904e-01 -5.78609586e-01
6.57686710e-01 -8.97110403e-01 6.16680205e-01 -4.11842428e-02
-3.21797073e-01 3.49738777e-01 1.14170931e-01 4.41950202e-01
-9.32397470e-02 9.33112726e-02 7.56381094e-01 -3.47285807e-01
-6.49840176e-01 4.10573512e-01 -1.25982212e-02 -8.07383657e-02
9.51802850e-01 -6.89465702e-01 -4.34464723e-01 -5.80752432e-01
-2.17036456e-01 -3.38632427e-02 9.35482502e-01 8.22303355e-01
1.02039063e+00 -1.76459587e+00 -6.62079155e-01 4.46100056e-01
2.30526447e-01 -1.14965625e-02 7.62961507e-01 1.73684150e-01
-6.27777219e-01 1.68659836e-01 -6.10620499e-01 -3.41494620e-01
-1.25776851e+00 3.38669419e-01 1.95830807e-01 5.97791731e-01
-8.87346387e-01 6.09123647e-01 3.94010752e-01 -4.16600347e-01
-7.47213960e-02 -3.26465935e-01 3.91824506e-02 1.00987270e-01
6.78379893e-01 1.11989439e-01 -1.07390985e-01 -6.69834316e-01
-7.25937039e-02 1.12654293e+00 9.41019058e-02 -1.42366856e-01
1.28845549e+00 -3.25767100e-02 1.66067913e-01 2.81637162e-01
1.39130306e+00 -3.39789316e-02 -1.25857377e+00 5.67770004e-02
-7.00063527e-01 -8.69607091e-01 7.50163430e-03 -7.20750928e-01
-1.45715868e+00 9.46105540e-01 4.89731252e-01 7.72287846e-02
1.36351717e+00 -2.10632697e-01 8.89488041e-01 1.02424920e-01
3.52970272e-01 -1.02000368e+00 1.64608419e-01 -2.20678393e-02
1.45459998e+00 -9.75150764e-01 1.16280220e-01 -7.34741747e-01
-1.03751004e+00 1.37149894e+00 8.08953166e-01 -1.68410480e-01
2.35668391e-01 7.59690851e-02 1.88939914e-01 -2.34966516e-01
-5.66322207e-01 1.56442910e-01 3.17491174e-01 8.45579982e-01
2.48054177e-01 -7.75029510e-02 3.04832697e-01 2.10627437e-01
-4.06257480e-01 1.19103659e-02 5.94908893e-01 6.05398059e-01
3.12336702e-02 -1.08024573e+00 -4.86077428e-01 1.69792309e-01
1.36930868e-01 1.89022776e-02 -8.66350174e-01 8.61527145e-01
4.55354266e-02 7.02167332e-01 2.71141648e-01 -4.80696350e-01
4.95301038e-01 1.28576963e-03 7.90117621e-01 -2.88898528e-01
-2.49834135e-01 -4.42986488e-02 -7.63946772e-02 -4.22724992e-01
-6.09781861e-01 -4.21073765e-01 -8.77351344e-01 -6.47099435e-01
2.00148627e-01 -1.67863175e-01 7.11004019e-01 3.58613521e-01
5.79959929e-01 5.30209005e-01 1.06503451e+00 -9.46315706e-01
8.59123319e-02 -5.90186775e-01 -6.08472645e-01 4.80725467e-01
-4.33635190e-02 -6.38317585e-01 -2.15539724e-01 3.14716011e-01] | [12.013776779174805, -0.46215716004371643] |
a07b4687-d7d3-42a7-8f9c-04a4a82dc968 | ccml-a-novel-collaborative-learning-model-for | 2012.10715 | null | https://arxiv.org/abs/2012.10715v6 | https://arxiv.org/pdf/2012.10715v6.pdf | Multi-Label Noise Robust Collaborative Learning for Remote Sensing Image Classification | The development of accurate methods for multi-label classification (MLC) of remote sensing (RS) images is one of the most important research topics in RS. The MLC methods based on convolutional neural networks (CNNs) have shown strong performance gains in RS. However, they usually require a high number of reliable training images annotated with multiple land-cover class labels. Collecting such data is time-consuming and costly. To address this problem, the publicly available thematic products, which can include noisy labels, can be used to annotate RS images with zero-labeling cost. However, multi-label noise (which can be associated with wrong and missing label annotations) can distort the learning process of the MLC methods. To address this problem, we propose a novel multi-label noise robust collaborative learning (RCML) method to alleviate the negative effects of multi-label noise during the training phase of a CNN model. RCML identifies, ranks and excludes noisy multi-labels in RS images based on three main modules: 1) the discrepancy module; 2) the group lasso module; and 3) the swap module. The discrepancy module ensures that the two networks learn diverse features, while producing the same predictions. The task of the group lasso module is to detect the potentially noisy labels assigned to multi-labeled training images, while the swap module is devoted to exchange the ranking information between two networks. Unlike the existing methods that make assumptions about noise distribution, our proposed RCML does not make any prior assumption about the type of noise in the training set. The experiments conducted on two multi-label RS image archives confirm the robustness of the proposed RCML under extreme multi-label noise rates. Our code is publicly available at: http://www.noisy-labels-in-rs.org | ['Begüm Demir', 'Mahdyar Ravanbakhsh', 'Ahmet Kerem Aksoy'] | 2020-12-19 | null | null | null | null | ['remote-sensing-image-classification'] | ['miscellaneous'] | [ 2.45713368e-01 -1.99187934e-01 2.07734033e-02 -5.77879310e-01
-9.56927121e-01 -4.41646963e-01 2.21496031e-01 4.52832319e-02
-3.94535154e-01 5.71466446e-01 -1.61122978e-01 -1.70038059e-01
-2.46802062e-01 -9.63143826e-01 -6.14846647e-01 -1.02868509e+00
1.95934922e-01 1.37119204e-01 -4.44604084e-02 2.70996373e-02
-1.33203715e-01 4.10902053e-01 -1.80072689e+00 4.46078092e-01
8.80679786e-01 1.15250564e+00 3.89504731e-01 2.16786504e-01
-3.30436409e-01 8.70715976e-01 -3.28213990e-01 -1.09927021e-01
4.44239736e-01 -2.73272902e-01 -5.54170251e-01 1.46191409e-02
3.38574052e-01 -2.81756483e-02 2.17704013e-01 1.50562418e+00
5.56967795e-01 3.01919933e-02 4.27420437e-01 -1.22747266e+00
-2.71459848e-01 6.49206817e-01 -5.10859668e-01 -2.81848669e-01
-4.87916082e-01 -2.13648617e-01 8.91086638e-01 -1.02070332e+00
3.20351183e-01 1.27719986e+00 1.09373116e+00 4.16525364e-01
-1.14208710e+00 -1.07476187e+00 1.37704954e-01 -9.72225592e-02
-1.72367787e+00 -3.51716608e-01 5.46833158e-01 -5.77718616e-01
9.26318243e-02 4.36332464e-01 3.17034096e-01 8.09951603e-01
-1.07340172e-01 4.46451068e-01 1.41142058e+00 -1.57781854e-01
2.09138975e-01 9.81520340e-02 1.25233635e-01 5.52138686e-01
1.25270724e-01 -4.09649163e-02 -9.77966413e-02 -3.73241454e-01
4.50424403e-01 2.92928874e-01 -1.77171186e-01 8.28343444e-03
-8.59100819e-01 8.77235472e-01 5.10908544e-01 2.40265727e-01
-2.63438821e-01 1.78968996e-01 4.35090959e-01 1.83822677e-01
9.29792702e-01 1.24980956e-01 -5.05960405e-01 5.94218016e-01
-9.17110384e-01 -9.70360264e-02 4.88307625e-01 7.76582062e-01
1.12419426e+00 -1.39565200e-01 -1.02542937e-01 1.30264378e+00
6.67943120e-01 7.00964987e-01 2.36518696e-01 -8.66191804e-01
4.17947531e-01 5.87434947e-01 8.54431316e-02 -1.39353311e+00
-5.95967829e-01 -7.59066880e-01 -1.36264086e+00 8.45313147e-02
-2.70058382e-02 -3.33980590e-01 -8.43687773e-01 1.66181028e+00
3.88956755e-01 4.08184499e-01 5.16415946e-02 9.75872576e-01
1.08840537e+00 6.86102629e-01 3.88410330e-01 -2.59451687e-01
9.48926985e-01 -7.87787914e-01 -8.00149202e-01 -2.27448940e-01
7.99926937e-01 -7.28132665e-01 6.35851502e-01 2.01999415e-02
-3.63247037e-01 -5.30434310e-01 -8.26936841e-01 3.00753176e-01
-4.05233771e-01 7.54666865e-01 4.71587002e-01 4.78929549e-01
-8.05762589e-01 6.13679051e-01 -5.29814780e-01 -1.03064127e-01
6.37029111e-01 1.69561192e-01 -3.32254529e-01 -2.71174788e-01
-1.18055868e+00 6.80867970e-01 3.87624502e-01 7.55584896e-01
-9.42296982e-01 -4.06395465e-01 -5.94474316e-01 -1.74856000e-02
5.93778312e-01 9.83497780e-03 8.39547098e-01 -1.41594541e+00
-9.67133343e-01 7.57377386e-01 7.48373494e-02 1.65156603e-01
5.77859163e-01 -1.10684887e-01 -5.25786221e-01 -2.37571195e-01
4.46014583e-01 4.23861504e-01 6.05539322e-01 -1.74788082e+00
-7.62041450e-01 -3.36003870e-01 -3.79966944e-01 1.06451035e-01
-1.18511871e-01 6.65940270e-02 -2.01723546e-01 -7.99591243e-01
6.81512296e-01 -1.07976735e+00 -3.66278261e-01 1.26282573e-01
-3.63391072e-01 -1.82052419e-01 7.99721241e-01 -4.11975533e-01
9.17064369e-01 -2.27141142e+00 -2.53982812e-01 4.31318671e-01
1.55656725e-01 4.61919338e-01 -3.75064522e-01 1.60452366e-01
-1.30781099e-01 5.82595885e-01 -3.30258429e-01 -4.47882473e-01
-3.77159446e-01 3.71236026e-01 -5.13860248e-02 6.71796381e-01
2.40988895e-01 4.87010926e-01 -9.42813218e-01 -4.91881430e-01
6.15293235e-02 2.75136530e-01 1.64763033e-02 1.76424637e-01
-1.83490202e-01 6.67982399e-01 -5.43791711e-01 7.39252627e-01
1.05561650e+00 -3.22671771e-01 3.45750183e-01 -4.89124417e-01
-2.97894329e-01 -1.54603481e-01 -1.62382019e+00 1.22977138e+00
-4.16228145e-01 1.58790484e-01 2.45487779e-01 -1.10492885e+00
1.17435670e+00 4.71267372e-01 5.73569000e-01 -2.91780889e-01
1.86894119e-01 5.68604410e-01 -3.66570264e-01 -6.34223998e-01
1.78164423e-01 -2.60983557e-01 7.65308142e-02 4.33933288e-01
-8.22118744e-02 2.01036602e-01 -3.54034975e-02 -1.69762552e-01
7.13357091e-01 -1.85185019e-02 -6.66675973e-04 -2.91813016e-01
5.02680302e-01 -5.27474913e-05 1.25287807e+00 9.42578495e-01
-2.23543018e-01 6.75562918e-01 1.34664550e-01 -5.84476709e-01
-8.93422544e-01 -5.24173915e-01 -2.59166807e-01 1.07929611e+00
1.92564741e-01 1.57284606e-02 -2.43636519e-01 -8.12586367e-01
4.66217054e-03 1.13916941e-01 -4.67690021e-01 1.08835727e-01
-2.58186013e-01 -1.31917965e+00 7.83014834e-01 2.16174409e-01
7.57593274e-01 -8.42644334e-01 9.57692564e-02 2.10922539e-01
-5.81635177e-01 -9.38873768e-01 1.32934833e-02 4.07574624e-01
-6.13444984e-01 -1.26880598e+00 -3.56819063e-01 -7.78652966e-01
8.54777575e-01 5.22611678e-01 8.15436244e-01 4.10384536e-01
2.60850098e-02 -1.90871537e-01 -6.85694098e-01 -3.98378342e-01
-5.23293018e-01 -3.74594480e-02 -3.85924950e-02 4.57763255e-01
3.02123725e-01 -3.66518885e-01 -2.80861318e-01 6.77995622e-01
-1.05119264e+00 2.03275055e-01 6.05881810e-01 8.78006160e-01
8.73663604e-01 3.94991875e-01 8.31564188e-01 -1.11989784e+00
1.49961486e-01 -7.59932101e-01 -7.45736361e-01 4.84112620e-01
-4.55587327e-01 -2.20055491e-01 5.64455032e-01 -3.74902666e-01
-1.04728281e+00 5.65212309e-01 -1.92468241e-01 -2.26029590e-01
-2.60757476e-01 7.48885334e-01 -2.29872361e-01 -2.34133109e-01
5.91610014e-01 -4.43757586e-02 -2.08395526e-01 -6.60967410e-01
1.43499210e-01 1.07396829e+00 1.68148041e-01 -3.62713248e-01
7.81772733e-01 4.91471440e-01 6.61843047e-02 -5.68390667e-01
-1.40278578e+00 -7.78196573e-01 -4.81873870e-01 -4.77627546e-01
7.39382923e-01 -1.40371037e+00 -4.29906249e-01 7.95213819e-01
-1.01502812e+00 -2.03603894e-01 1.27685085e-01 5.37891865e-01
-2.73304768e-02 2.05285564e-01 -4.16158080e-01 -1.01389992e+00
-3.46927762e-01 -1.23443151e+00 9.90829945e-01 2.21238539e-01
4.03980345e-01 -6.87931120e-01 -2.31380239e-01 4.49014604e-01
5.25827169e-01 3.20528686e-01 7.77673721e-01 -7.05191374e-01
-4.42388594e-01 -1.01753697e-01 -5.47465920e-01 7.62967706e-01
2.21224278e-01 -3.35637890e-02 -1.25874531e+00 -3.22505325e-01
-7.79579729e-02 -5.77442229e-01 9.76229131e-01 1.91181555e-01
1.24934876e+00 -1.46697402e-01 -1.76547632e-01 4.86412287e-01
1.80768275e+00 9.78352781e-03 4.02948827e-01 1.73975304e-01
9.46492314e-01 6.56129956e-01 7.80554116e-01 4.37490612e-01
2.56479084e-01 3.21446657e-01 7.06440866e-01 -3.51555854e-01
3.46986316e-02 4.40294445e-02 -6.57441840e-03 8.64209175e-01
-5.24250083e-02 -3.47293913e-01 -1.06002986e+00 4.25698996e-01
-2.12928414e+00 -7.57278502e-01 -6.06095195e-01 2.01542974e+00
8.38591158e-01 -4.25569504e-01 -5.36211252e-01 6.60852194e-02
1.11172259e+00 2.31646627e-01 -4.29156870e-01 3.64576191e-01
-4.45049882e-01 -1.55452013e-01 8.99984002e-01 2.96108693e-01
-1.60104239e+00 7.91981518e-01 5.30956697e+00 9.74187136e-01
-1.00244522e+00 4.28591758e-01 7.26958871e-01 1.97828740e-01
-2.09927171e-01 -6.23555966e-02 -8.90584767e-01 3.80995959e-01
5.14770865e-01 5.77982783e-01 2.22341016e-01 8.35404456e-01
4.81204867e-01 -2.63158530e-01 -4.96181875e-01 9.04877186e-01
-1.01367131e-01 -1.20008540e+00 -1.51722118e-01 -1.60091460e-01
1.15691483e+00 4.36370611e-01 -2.67898023e-01 1.14593893e-01
6.91959500e-01 -8.73767555e-01 6.70951903e-01 6.85387254e-01
7.76620746e-01 -6.37936711e-01 1.26100075e+00 5.32257497e-01
-1.30312359e+00 -3.92278999e-01 -7.17552602e-01 1.87875867e-01
-1.26835525e-01 1.27367115e+00 -2.46751890e-01 7.53028929e-01
9.42505240e-01 8.64689827e-01 -6.19852185e-01 9.94245350e-01
-3.44599098e-01 6.44968748e-01 -2.77092040e-01 2.34687015e-01
2.19522208e-01 -3.73378307e-01 2.14113593e-01 9.45099890e-01
3.92089158e-01 1.28146335e-01 5.96469164e-01 6.96492016e-01
-3.07416588e-01 2.91097194e-01 -4.56918329e-01 1.61073461e-01
6.51960194e-01 1.40841770e+00 -7.08703339e-01 -1.66603193e-01
-3.74162436e-01 5.45336604e-01 3.02708924e-01 3.35938871e-01
-6.55892253e-01 -3.21640968e-02 4.11314398e-01 -3.35648179e-01
-6.18065931e-02 -2.64145099e-02 -3.88256878e-01 -1.00940883e+00
7.14739226e-03 -7.52408564e-01 2.53672421e-01 -6.58799827e-01
-1.52597213e+00 4.20807123e-01 -3.79706323e-01 -1.34166062e+00
5.50944567e-01 -2.30345801e-01 -2.29064211e-01 9.91002858e-01
-1.89123261e+00 -1.41397166e+00 -7.00500011e-01 2.96031326e-01
3.32774937e-01 -6.79474324e-02 8.89557540e-01 8.04341018e-01
-7.63525009e-01 1.68034881e-01 6.42264307e-01 2.73270786e-01
7.75045037e-01 -8.36248279e-01 -2.71713853e-01 7.48168468e-01
-2.01487377e-01 1.90981209e-01 2.08954692e-01 -8.47557664e-01
-6.70986891e-01 -1.86300206e+00 8.95716131e-01 9.42393094e-02
4.13025200e-01 -2.08836615e-01 -9.53395069e-01 5.67720830e-01
-6.35138273e-01 5.08127511e-01 9.16619718e-01 -3.16598006e-02
-3.06512237e-01 -3.99117351e-01 -1.11929619e+00 6.89587519e-02
7.44077921e-01 -4.77002025e-01 9.08233821e-02 7.33094096e-01
5.44109464e-01 -1.21090859e-01 -9.11853492e-01 6.19321525e-01
4.09034282e-01 -7.39587963e-01 7.18681395e-01 -1.44019455e-01
3.70190918e-01 -6.95670664e-01 -3.73632401e-01 -1.24944603e+00
-3.83077085e-01 2.76347756e-01 7.18598068e-01 1.47245097e+00
4.09886211e-01 -4.77595359e-01 3.79513085e-01 3.28221172e-01
-2.32467189e-01 -3.06325465e-01 -8.34772885e-01 -6.93887413e-01
-1.68789148e-01 -5.20478547e-01 5.91398954e-01 1.20449173e+00
-8.08802843e-01 -1.01018049e-01 -6.89416051e-01 7.09533632e-01
6.21456921e-01 -1.92282861e-03 4.98152047e-01 -1.78601372e+00
2.22476304e-01 4.12681624e-02 5.57745360e-02 -4.40045118e-01
3.07328433e-01 -9.95206892e-01 5.06808698e-01 -1.51248801e+00
2.38178775e-01 -1.30048430e+00 -3.04257274e-01 1.01106620e+00
-1.71296760e-01 5.21149039e-01 8.67871195e-02 7.45296955e-01
-6.83668137e-01 6.22949958e-01 9.68336582e-01 -1.88641638e-01
-5.24346903e-03 2.36641437e-01 -4.82977897e-01 8.39074254e-01
9.91742671e-01 -1.07404101e+00 -8.32489431e-02 -4.72762555e-01
5.32840788e-01 -2.41078004e-01 4.11006957e-01 -9.37775135e-01
2.74193704e-01 -3.91782105e-01 1.42511144e-01 -5.25912702e-01
-5.65551519e-02 -1.13987327e+00 5.96840143e-01 3.32546055e-01
-3.44305933e-01 -4.07591969e-01 -7.88587481e-02 6.11999750e-01
-2.24936754e-01 -5.01699448e-01 1.08138764e+00 -4.01338845e-01
-5.88718534e-01 3.74759644e-01 -4.24732804e-01 -3.23654324e-01
8.20886374e-01 3.53715152e-01 -3.79588395e-01 -9.31062102e-02
-7.82653630e-01 4.81772542e-01 1.42371863e-01 3.69146407e-01
3.07106197e-01 -1.45780373e+00 -8.90224338e-01 1.19069993e-01
3.42818618e-01 3.76611322e-01 4.14021760e-01 7.04646289e-01
-4.08970922e-01 -4.55408581e-02 1.70704961e-01 -6.26687646e-01
-1.27856719e+00 4.17412519e-02 6.18860781e-01 -2.31886223e-01
-1.48339719e-01 6.10640049e-01 -1.72968224e-01 -1.10583973e+00
2.00854853e-01 1.18319146e-01 -4.98306870e-01 3.72174650e-01
3.76989514e-01 3.72821480e-01 2.54604578e-01 -9.35400844e-01
-2.01968014e-01 6.64576113e-01 2.94171602e-01 3.93463463e-01
1.57287180e+00 -3.51722389e-01 -5.90179980e-01 6.11618400e-01
1.12902308e+00 -4.17556673e-01 -1.01581848e+00 -4.90289450e-01
3.89858745e-02 -3.40043485e-01 3.85868043e-01 -8.76676202e-01
-1.36160553e+00 5.32544196e-01 9.71099794e-01 2.69440040e-02
1.05584669e+00 -2.80008584e-01 3.85221541e-01 4.29960102e-01
3.44555438e-01 -1.34406042e+00 -1.60008475e-01 4.44356471e-01
6.47314370e-01 -1.77232742e+00 -1.33930873e-02 -6.10864758e-01
-3.92192543e-01 9.62026656e-01 5.37371874e-01 2.49822989e-01
8.99162769e-01 -8.07040837e-03 5.82047582e-01 -3.73818696e-01
-1.97121635e-01 -3.58571440e-01 3.45947370e-02 4.02572393e-01
1.63853705e-01 2.25339621e-01 -4.55119282e-01 4.16616410e-01
4.35556084e-01 6.53335033e-03 3.66525054e-01 7.81486452e-01
-5.52549779e-01 -9.11645412e-01 -6.33000612e-01 5.78128278e-01
-5.60947120e-01 -8.38148817e-02 -1.03588286e-03 3.11614871e-01
7.90947437e-01 1.46441126e+00 -1.73200548e-01 -4.83187735e-01
1.92974851e-01 -3.90165821e-02 -5.39019763e-01 -7.87309229e-01
-6.97537422e-01 1.11265212e-01 4.62994613e-02 -4.80813503e-01
-1.16907120e+00 -4.53180671e-01 -9.67315674e-01 -1.56970039e-01
-8.26016784e-01 1.79712638e-01 8.79121721e-01 1.05570197e+00
2.09696904e-01 3.36739510e-01 9.72680032e-01 -7.62038112e-01
-4.85994458e-01 -1.11375642e+00 -8.90448451e-01 5.38655519e-01
3.38664144e-01 -6.27997696e-01 -5.87975204e-01 -4.46841903e-02] | [9.54385757446289, 3.8437845706939697] |
c80b65dd-e0c8-4e1f-80e4-69e701f508c6 | network-compression-for-machine-learnt-fluid-1 | null | null | https://openreview.net/forum?id=6Qy9aoCms0C | https://openreview.net/pdf?id=6Qy9aoCms0C | NETWORK COMPRESSION FOR MACHINE-LEARNT FLUID SIMULATIONS | Multi-scale, multi-fidelity numerical simulations form the pillar of scientific applications
related to numerically modeling fluids. However, simulating the fluid
behavior characterized by the non-linear Navier Stokes equations are often times
computational expensive. Physics informed machine learning methods is a viable
alternative and as such has seen great interest in the community [refer to
Kutz (2017); Brunton et al. (2020); Duraisamy et al. (2019) for a detailed review
on this topic]. For full physics emulators, the cost of network inference is often
trivial. However, in the current paradigm of data-driven fluid mechanics models
are built as surrogates for complex sub-processes. These models are then used in
conjunction to the Navier Stokes solvers, which makes ML model inference an
important factor in the terms of algorithmic latency. With the ever growing size
of networks, and often times overparameterization, exploring effective network
compression techniques becomes not only relevant but critical for engineering
systems design. In this study, we explore the applicability of pruning and quantization
(FP32 to int8) methods for one such application relevant to modeling fluid
turbulence. Post-compression, we demonstrate the improvement in the accuracy
of network predictions and build intuition in the process by comparing the compressed
to the original network state. | ['Anonymous'] | 2021-03-04 | null | null | null | null | ['physics-informed-machine-learning'] | ['graphs'] | [ 1.98082581e-01 -1.06002413e-01 6.69397935e-02 3.30918789e-01
-1.41957954e-01 -4.66570765e-01 7.44600594e-01 5.44657111e-01
-3.77716780e-01 1.09070551e+00 -2.29664087e-01 -6.15988851e-01
-7.60644376e-01 -9.32738304e-01 -5.72479486e-01 -6.91485405e-01
-4.46929067e-01 7.91202247e-01 2.66278684e-02 -1.71118289e-01
4.38429117e-01 9.91062522e-01 -1.55614960e+00 -2.03085601e-01
5.60810089e-01 1.10433733e+00 -1.13960057e-01 1.05748260e+00
-1.69839993e-01 5.92833638e-01 -7.23510236e-02 -1.27670228e-01
2.75182128e-01 -2.49414399e-01 -5.99966168e-01 -3.66779089e-01
2.18523875e-01 -4.09571268e-02 -4.00713414e-01 8.26949775e-01
5.56919694e-01 7.37971425e-01 6.16249740e-01 -9.93490100e-01
4.28700745e-01 5.37402332e-01 -2.19668224e-01 3.06555092e-01
-6.69648573e-02 4.66927469e-01 8.36484909e-01 -6.25212014e-01
4.15074557e-01 1.11599374e+00 8.22114408e-01 1.45429745e-01
-1.44021749e+00 -6.17819428e-01 -3.01968515e-01 1.50911018e-01
-1.32569158e+00 -4.71067488e-01 6.69942856e-01 -6.09235287e-01
6.87715590e-01 3.63539875e-01 9.92269993e-01 6.80749357e-01
2.34471038e-01 4.62263301e-02 1.22220004e+00 -2.73399204e-01
7.77947724e-01 1.47951052e-01 5.77777624e-02 5.09690702e-01
7.04834461e-01 4.31260496e-01 -6.97107434e-01 -5.13681114e-01
8.98991942e-01 -1.76353544e-01 -1.29144087e-01 -2.15361372e-01
-6.92566574e-01 8.60885739e-01 1.67949826e-01 -1.80686768e-02
-4.32122558e-01 5.10255933e-01 5.65608740e-01 5.89242540e-02
5.13336360e-01 9.25012350e-01 -4.72183496e-01 -6.86536252e-01
-1.24246538e+00 8.80932033e-01 1.37537527e+00 5.34509301e-01
7.54381955e-01 1.93508282e-01 5.21807849e-01 5.28794408e-01
3.13342422e-01 3.21675271e-01 6.21723309e-02 -1.51459670e+00
-9.28307101e-02 3.58350337e-01 7.87896737e-02 -1.04044747e+00
-3.01963896e-01 -7.01533079e-01 -1.30288684e+00 3.59005600e-01
4.74982291e-01 -3.20623845e-01 -4.35310125e-01 1.10176277e+00
5.90953231e-01 5.27012944e-01 8.68550676e-04 8.08045328e-01
3.82296950e-01 8.08494091e-01 2.07047135e-01 -3.51666719e-01
9.97455418e-01 -5.90393960e-01 -4.05898869e-01 9.65431258e-02
6.10188663e-01 -6.81629181e-01 5.98045647e-01 6.60213888e-01
-1.27821779e+00 -4.83652949e-01 -8.96164894e-01 2.16822311e-01
-4.16222602e-01 -7.89084733e-01 8.87214422e-01 4.59443718e-01
-9.58517611e-01 1.57540548e+00 -8.89668405e-01 -3.75991255e-01
4.95657414e-01 3.60976368e-01 5.71394116e-02 6.86116610e-03
-1.19882345e+00 8.39491189e-01 -1.43569680e-02 1.87798038e-01
-6.48799181e-01 -1.19100583e+00 -4.52939272e-01 1.00639552e-01
4.09653217e-01 -8.48579228e-01 9.20116782e-01 -3.19551378e-01
-1.78655171e+00 2.28597969e-01 -9.98217016e-02 -5.97611010e-01
8.59277904e-01 -1.15846574e-01 -7.41470978e-02 1.34697691e-01
-5.70698977e-01 2.73313850e-01 5.99942029e-01 -1.24915266e+00
-3.32697034e-01 1.23638593e-01 1.23541385e-01 1.20108411e-01
-3.89552899e-02 -3.40311110e-01 1.02877855e-01 -4.75637019e-01
-1.75957873e-01 -1.06225157e+00 -6.58182025e-01 4.59392458e-01
-4.39624816e-01 -3.52994017e-02 8.73043478e-01 -3.31078440e-01
9.87913191e-01 -1.72894585e+00 2.67607123e-01 7.06738472e-01
4.86007690e-01 4.17649120e-01 3.62405092e-01 6.64110601e-01
-2.14457046e-02 3.31825674e-01 -5.28727353e-01 -3.94354999e-01
-4.50550653e-02 3.60153675e-01 -3.20610791e-01 4.84639108e-01
9.22204331e-02 6.06875837e-01 -7.72580743e-01 -5.33754647e-01
5.18091142e-01 5.74508309e-01 -5.59009016e-01 6.87268749e-02
-2.19766766e-01 7.74996936e-01 -4.59638000e-01 3.37906390e-01
2.92686671e-01 -4.46732104e-01 -3.97103420e-03 -2.49648705e-01
-1.80508330e-01 2.06114352e-01 -1.34990382e+00 1.31022894e+00
-6.19396508e-01 6.84305787e-01 3.87225389e-01 -1.18797922e+00
6.11442208e-01 2.69560516e-01 5.61221838e-01 -3.02107871e-01
3.53164554e-01 2.52345145e-01 3.44332397e-01 -2.55224109e-01
5.48711061e-01 -4.27985162e-01 3.54467064e-01 3.69885236e-01
-4.56624180e-01 -7.67777264e-01 4.56366330e-01 1.55211508e-01
1.22428322e+00 1.17877543e-01 2.19229802e-01 -6.74057484e-01
4.89047617e-01 2.09247068e-01 9.78816822e-02 7.24113166e-01
-1.41988307e-01 1.31895900e-01 5.04913270e-01 -5.19573569e-01
-1.34382212e+00 -7.53866792e-01 -3.78775269e-01 4.23692137e-01
-2.23302126e-01 -3.57339174e-01 -6.54485941e-01 1.37247771e-01
4.37769711e-01 5.15262961e-01 -4.91576344e-01 -2.23061834e-02
-7.54678369e-01 -5.19959450e-01 3.93010199e-01 8.31306279e-02
2.91064322e-01 -8.71697068e-01 -6.88925266e-01 4.14034963e-01
5.25908947e-01 -1.21555173e+00 4.75649983e-01 3.03543746e-01
-1.25396073e+00 -1.12505853e+00 -3.26310903e-01 4.25366312e-02
1.66716650e-01 -3.32715631e-01 1.05141997e+00 4.29889590e-01
-5.43397367e-01 2.41112337e-01 -3.77714001e-02 -3.77003312e-01
-7.23098218e-01 1.88547850e-01 2.16601387e-01 -3.68033856e-01
-2.54013062e-01 -1.14498138e+00 -6.85576320e-01 -1.67383209e-01
-8.48832309e-01 1.53138682e-01 4.58471209e-01 2.70128250e-01
5.64091384e-01 2.15804920e-01 3.44952762e-01 -7.92443216e-01
5.38539648e-01 -5.87117553e-01 -9.09094334e-01 -4.08818364e-01
-7.60635853e-01 -6.90216050e-02 1.12467384e+00 -4.08389628e-01
-5.86670101e-01 -1.60222933e-01 2.51861569e-02 -5.37609041e-01
-1.21242300e-01 6.57246828e-01 3.31051707e-01 -5.35698831e-01
2.84636706e-01 -9.21606719e-02 2.78206557e-01 -4.09624308e-01
1.80304140e-01 1.85798347e-01 7.36341700e-02 -9.00394261e-01
9.62961078e-01 3.65798414e-01 1.16261446e+00 -1.45954633e+00
-4.52894330e-01 -1.13400981e-01 -5.15263677e-01 -4.77273583e-01
3.31565738e-01 -3.73337716e-01 -1.06418240e+00 1.69103712e-01
-1.05171967e+00 -5.45110226e-01 -6.53940678e-01 5.45511007e-01
-6.20287657e-01 1.41112521e-01 -8.03648829e-01 -1.03654873e+00
-1.38002709e-01 -1.10187948e+00 5.51028192e-01 6.45119473e-02
-6.53091908e-01 -1.26840138e+00 7.87875652e-02 3.22313458e-02
8.26663852e-01 5.43365479e-01 1.05246878e+00 -5.24430692e-01
-7.90853679e-01 -2.20951568e-02 -2.55003721e-01 1.84428155e-01
-1.58101991e-01 4.51138139e-01 -9.46048558e-01 -3.76462005e-02
-2.15036124e-02 -1.17534120e-02 5.93966544e-01 5.01372576e-01
1.37375152e+00 -1.56325221e-01 -3.58777881e-01 4.56350714e-01
1.42531049e+00 -1.96745694e-01 2.92955637e-01 -1.34549573e-01
7.69806147e-01 8.67402375e-01 1.42510265e-01 6.95302427e-01
-1.64465204e-01 2.91550994e-01 2.61615247e-01 4.64870632e-02
-1.04562044e-01 1.27592042e-01 -1.01833709e-01 1.28678393e+00
-2.14247346e-01 -3.05546433e-01 -1.03500974e+00 2.49952361e-01
-1.58033192e+00 -8.15503657e-01 -3.69378775e-01 2.02205086e+00
8.86005878e-01 3.76484305e-01 -2.18263462e-01 3.31181318e-01
4.42520261e-01 -7.65716564e-03 -7.11975932e-01 -6.07441127e-01
1.93508103e-01 6.06208742e-01 6.82466030e-01 6.89215183e-01
-8.11894059e-01 4.92408872e-01 5.53612280e+00 9.26253438e-01
-1.08236372e+00 -1.25517607e-01 8.82826149e-01 -2.91666985e-01
1.44015804e-01 1.30751804e-01 -7.93827653e-01 5.03361046e-01
1.72127080e+00 -3.09672982e-01 6.74530685e-01 4.78527218e-01
7.08379269e-01 -4.65453774e-01 -1.17822516e+00 8.11389565e-01
-4.74609911e-01 -1.72560668e+00 -5.38091436e-02 3.11260015e-01
5.97239375e-01 4.84570935e-02 -2.84263372e-01 2.24343576e-02
3.58599067e-01 -1.20026386e+00 3.83425534e-01 8.66024792e-01
5.00742078e-01 -6.41843140e-01 4.75510687e-01 2.58252949e-01
-1.26811349e+00 2.62911618e-01 -1.22121304e-01 -4.19604272e-01
4.83041316e-01 1.13467586e+00 -4.48060364e-01 3.34092647e-01
5.74254870e-01 6.97850525e-01 -7.03278482e-02 1.19433045e+00
4.17431533e-01 1.07920802e+00 -9.99130368e-01 -2.81805724e-01
1.12621091e-01 -3.19767475e-01 8.09469819e-01 9.18637216e-01
3.42819154e-01 9.25965458e-02 1.28798679e-01 9.01952386e-01
-4.93011102e-02 -1.37224957e-01 -4.91205335e-01 -2.26909682e-01
4.66148108e-01 1.32239473e+00 -9.28448737e-01 -4.29194480e-01
1.11524463e-02 1.07624210e-01 1.91160701e-02 4.06167120e-01
-5.34085333e-01 2.57391046e-04 1.00115156e+00 5.09193838e-01
1.68372259e-01 -4.63229001e-01 -4.85270739e-01 -3.86676162e-01
-3.84542614e-01 -7.55203724e-01 -1.47278696e-01 -3.74137819e-01
-1.47371197e+00 2.97820847e-02 2.81167179e-01 -9.67225969e-01
-8.61077905e-02 -7.50768185e-01 -4.18924451e-01 9.42990839e-01
-1.74916589e+00 -5.76420486e-01 -1.87222198e-01 5.07910922e-02
2.37068936e-01 2.43882954e-01 8.07875812e-01 5.32300770e-01
-4.92741585e-01 -1.09280407e-01 4.42568392e-01 -2.53727436e-01
3.30123425e-01 -9.29059684e-01 3.48520756e-01 4.10449922e-01
-4.92007315e-01 6.53432965e-01 1.12559938e+00 -9.45427716e-01
-1.77679801e+00 -1.09259248e+00 4.90508229e-01 -8.69798139e-02
9.74011838e-01 -1.54956132e-01 -1.18802691e+00 6.10583229e-03
-2.90705301e-02 3.36380392e-01 5.12062848e-01 -2.48200282e-01
3.70088935e-01 -1.59815162e-01 -1.11301100e+00 5.37991583e-01
8.90051901e-01 -3.61036837e-01 2.29864746e-01 2.82847703e-01
4.73823875e-01 -3.34401518e-01 -1.22685540e+00 3.83589834e-01
4.38159317e-01 -7.83292472e-01 8.67196679e-01 -4.19719249e-01
6.79567635e-01 -1.91527620e-01 7.45588541e-02 -1.01084328e+00
1.30774260e-01 -9.13897097e-01 -5.98997474e-01 9.66862381e-01
1.84113368e-01 -6.65454209e-01 1.03699052e+00 6.80494964e-01
1.16224431e-01 -1.09249270e+00 -1.16949427e+00 -5.37977874e-01
3.04053694e-01 -6.45925641e-01 2.89644420e-01 9.26270008e-01
-2.36150950e-01 -1.20222485e-02 9.96178463e-02 8.23289976e-02
8.73568952e-01 -1.75933391e-01 6.99639142e-01 -1.68734729e+00
-5.29936492e-01 -7.72540271e-01 -2.71100670e-01 -7.71763623e-01
-4.48073149e-02 -6.17205381e-01 -4.65832055e-02 -1.15307474e+00
-1.75466329e-01 -8.68315041e-01 -1.19360397e-02 -2.04406589e-01
1.15081847e-01 1.97033688e-01 6.60537137e-03 2.92010248e-01
-2.61741310e-01 3.71795654e-01 1.18289351e+00 2.42316082e-01
1.37440369e-01 -1.03001460e-01 -2.44881921e-02 6.21793211e-01
9.69247520e-01 -4.06612307e-01 -3.41296166e-01 2.11741269e-01
4.44750667e-01 1.34205729e-01 6.29998565e-01 -1.28619480e+00
5.79007268e-01 -1.00497834e-01 2.28946447e-01 -3.80395502e-01
6.91873312e-01 -8.91387880e-01 2.05489546e-01 7.62207389e-01
-3.46570283e-01 3.34803313e-01 5.38685203e-01 6.90888703e-01
5.56383356e-02 -1.92722678e-01 8.28432918e-01 -3.20405871e-01
-2.26932302e-01 3.53672057e-01 -5.76469898e-01 1.60690293e-01
8.61810625e-01 -2.65791565e-01 -1.05933577e-01 -3.29248488e-01
-6.22011483e-01 1.08613372e-01 3.75664353e-01 -2.06734955e-01
3.25719863e-01 -6.20398223e-01 -4.90118533e-01 7.74337351e-02
-7.13554323e-01 4.94346082e-01 2.68709630e-01 8.54967535e-01
-9.45463359e-01 1.48333922e-01 9.67267007e-02 -4.78461683e-01
-8.49519610e-01 -1.53535213e-02 4.69375908e-01 -4.38801140e-01
-3.47433418e-01 7.98931658e-01 -2.56257683e-01 -2.11725742e-01
2.81514823e-02 -2.88096577e-01 9.08057988e-02 -1.06112309e-01
1.69393063e-01 9.48824763e-01 2.03130215e-01 -2.34918579e-01
-1.50191458e-02 5.27804673e-01 2.82808512e-01 2.88119987e-02
1.48756409e+00 1.09264910e-01 -3.28872770e-01 7.44573116e-01
1.11962306e+00 -2.08820045e-01 -1.26254177e+00 -1.32008269e-01
-3.35770585e-02 -1.62572384e-01 3.29027832e-01 -4.34804678e-01
-1.09448946e+00 1.02135503e+00 3.60559881e-01 5.36834836e-01
4.91897702e-01 -3.75459611e-01 8.91813397e-01 4.45144355e-01
3.49193633e-01 -1.01079142e+00 -4.91796643e-01 5.13221204e-01
5.44936061e-01 -8.64901841e-01 6.30326629e-01 -5.49660742e-01
2.19394073e-01 1.31465840e+00 2.31188789e-01 -4.28946108e-01
1.14508760e+00 7.07129419e-01 -3.87584686e-01 -1.45868033e-01
-7.39075065e-01 4.13002610e-01 -1.88425422e-01 1.40186861e-01
2.62627512e-01 -9.82252285e-02 -3.13733220e-02 -6.08414188e-02
-3.89112473e-01 -7.34906178e-03 4.56882358e-01 9.37237918e-01
-3.14928114e-01 -1.16757643e+00 -2.27599517e-01 9.29040015e-01
-1.56093538e-01 -1.64871469e-01 4.22738455e-02 8.64589691e-01
1.76813882e-02 7.65381813e-01 1.07860595e-01 -1.13427434e-02
1.70560479e-01 7.40223974e-02 2.56515414e-01 -3.26049209e-01
-7.11818337e-01 -1.42814279e-01 2.22986236e-01 -4.64174569e-01
-4.42971021e-01 -7.63097703e-01 -1.29487550e+00 -1.06645846e+00
-1.71593979e-01 3.98290336e-01 8.65346134e-01 1.00206220e+00
4.51068729e-01 8.27665985e-01 1.63573831e-01 -1.41817880e+00
-4.61151183e-01 -8.11790347e-01 -5.29443741e-01 7.04510733e-02
3.62455577e-01 -8.29966366e-01 -6.48923814e-01 -6.19532913e-02] | [6.414342403411865, 3.444453477859497] |
9a1cc607-0c4b-41ca-8de4-6431dc59826b | comprehensive-and-delicate-an-efficient | null | null | http://openaccess.thecvf.com//content/CVPR2023/html/Zhao_Comprehensive_and_Delicate_An_Efficient_Transformer_for_Image_Restoration_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Zhao_Comprehensive_and_Delicate_An_Efficient_Transformer_for_Image_Restoration_CVPR_2023_paper.pdf | Comprehensive and Delicate: An Efficient Transformer for Image Restoration | Vision Transformers have shown promising performance in image restoration, which usually conduct window- or channel-based attention to avoid intensive computations. Although the promising performance has been achieved, they go against the biggest success factor of Transformers to a certain extent by capturing the local instead of global dependency among pixels. In this paper, we propose a novel efficient image restoration Transformer that first captures the superpixel-wise global dependency, and then transfers it into each pixel. Such a coarse-to-fine paradigm is implemented through two neural blocks, i.e., condensed attention neural block (CA) and dual adaptive neural block (DA). In brief, CA employs feature aggregation, attention computation, and feature recovery to efficiently capture the global dependency at the superpixel level. To embrace the pixel-wise global dependency, DA takes a novel dual-way structure to adaptively encapsulate the globality from superpixels into pixels. Thanks to the two neural blocks, our method achieves comparable performance while taking only 6% FLOPs compared with SwinIR. | ['Xi Peng', 'Jiancheng Lv', 'Dezhong Peng', 'Boyun Li', 'Yuanbiao Gou', 'Haiyu Zhao'] | 2023-01-01 | null | null | null | cvpr-2023-1 | ['superpixels'] | ['computer-vision'] | [ 2.79759526e-01 -3.16635430e-01 -9.06026810e-02 -2.69321710e-01
-7.85076678e-01 1.53035680e-02 4.49187070e-01 6.14077412e-02
-2.21410275e-01 4.81386542e-01 4.81339842e-01 -1.99017450e-01
4.38857339e-02 -9.20625210e-01 -7.55165458e-01 -9.49199319e-01
2.99870253e-01 -2.72522271e-01 6.22453451e-01 -1.26635239e-01
3.12985569e-01 4.21401054e-01 -1.51089263e+00 2.52553761e-01
1.28119922e+00 1.16070759e+00 4.25654858e-01 4.45969790e-01
-1.87659964e-01 1.05335760e+00 -3.58953744e-01 -1.22839101e-01
2.20267266e-01 -3.61131012e-01 -6.10999525e-01 2.34828413e-01
7.06093848e-01 -5.17765343e-01 -2.45994419e-01 1.25308216e+00
4.25820529e-01 -1.83897570e-01 2.23453671e-01 -9.58992720e-01
-9.90694046e-01 6.51273072e-01 -8.70505869e-01 3.34024131e-01
-4.94536348e-02 2.80014992e-01 1.05817342e+00 -9.13953841e-01
1.83239385e-01 1.24655688e+00 5.35336673e-01 8.43430459e-02
-1.14418054e+00 -5.93851268e-01 4.11099494e-01 4.93359506e-01
-1.28126454e+00 -3.32147419e-01 7.82822311e-01 -1.64947569e-01
9.74914670e-01 2.21340179e-01 8.13658953e-01 3.06462914e-01
2.54002988e-01 8.20298433e-01 1.18495131e+00 -2.15079963e-01
7.98176676e-02 -3.06941807e-01 3.61297339e-01 5.48979759e-01
-1.94596201e-02 -1.04963534e-01 -3.46980929e-01 3.30687404e-01
1.09301829e+00 2.46617690e-01 -6.37550116e-01 4.19676155e-02
-1.05003870e+00 5.47939181e-01 9.53719437e-01 1.80286095e-01
-6.31560683e-01 3.40788841e-01 3.93717378e-01 1.64789647e-01
3.69133025e-01 -1.10206090e-01 -4.57160354e-01 2.60799736e-01
-1.04045761e+00 -5.84662408e-02 1.48625299e-01 7.88607359e-01
9.33198333e-01 8.29719082e-02 -5.03912985e-01 8.49605739e-01
2.13488579e-01 3.13611597e-01 4.99761075e-01 -8.05735469e-01
3.27762216e-01 7.90141881e-01 6.04245812e-03 -1.11400890e+00
-1.14829212e-01 -6.50878787e-01 -1.24728870e+00 4.30959284e-01
2.71547943e-01 3.31121057e-01 -1.12304211e+00 1.55005491e+00
3.71197045e-01 3.11649561e-01 -2.73777515e-01 1.06511521e+00
7.03537226e-01 8.62714291e-01 2.35980019e-01 -2.69925982e-01
1.54424870e+00 -1.42783046e+00 -5.96935570e-01 -1.40715465e-01
-1.97500014e-03 -8.45546603e-01 1.15152597e+00 2.93010950e-01
-1.42586184e+00 -9.85012949e-01 -1.14642799e+00 -7.00233579e-01
-1.11122675e-01 2.00921744e-01 5.79644501e-01 4.27428722e-01
-1.27449954e+00 4.23279166e-01 -7.69210756e-01 -8.97916034e-02
6.23063207e-01 3.71256202e-01 -1.56493485e-01 -2.03871518e-01
-9.23498929e-01 4.84191030e-01 8.67376179e-02 3.69256496e-01
-9.15994287e-01 -7.65975654e-01 -7.64549732e-01 5.42343140e-01
1.93863332e-01 -7.55306244e-01 8.67590368e-01 -1.16910946e+00
-1.43652201e+00 4.56237435e-01 -3.47517401e-01 -4.69806224e-01
3.88997048e-01 -3.22277665e-01 -1.70348939e-02 1.09403841e-01
9.88760665e-02 6.42465711e-01 1.20470047e+00 -1.06821680e+00
-9.15998101e-01 -3.03860605e-01 1.52957857e-01 1.73975751e-01
-5.88497281e-01 5.92645817e-02 -8.39528263e-01 -9.05000985e-01
3.22529107e-01 -3.90718371e-01 -3.30886155e-01 1.68728024e-01
-3.62127066e-01 -1.10129006e-01 8.03818405e-01 -7.76594996e-01
1.34945309e+00 -2.18829966e+00 1.36750072e-01 -1.91465423e-01
3.62471223e-01 3.92585963e-01 -1.58469900e-01 4.04473245e-02
-2.06900164e-01 -5.64092472e-02 -4.31081146e-01 -4.25175339e-01
-2.92138338e-01 1.52036145e-01 -3.85426581e-01 4.45760369e-01
3.13296825e-01 8.40525508e-01 -7.90505826e-01 -5.25480866e-01
4.29772407e-01 6.89562500e-01 -7.07305789e-01 -3.15944897e-03
-1.72472745e-02 2.84750611e-01 -4.57482368e-01 8.74939561e-01
1.15042615e+00 -2.93453991e-01 4.37013619e-02 -8.42110515e-01
-4.59416002e-01 5.18408269e-02 -8.79074633e-01 1.44753599e+00
-5.26924908e-01 4.16206330e-01 4.47454065e-01 -1.06834364e+00
8.30231130e-01 4.71195206e-02 3.46946329e-01 -9.48203683e-01
1.58251300e-02 7.69441500e-02 -2.27705523e-01 -2.67238379e-01
5.94917595e-01 1.41219953e-02 2.42784053e-01 9.24360454e-02
-1.82950616e-01 2.27468938e-01 9.60420519e-02 5.47756404e-02
9.55178738e-01 1.60773963e-01 2.99445331e-01 -3.82815361e-01
9.26666975e-01 -6.77466975e-04 7.77806759e-01 4.79387641e-01
-2.84737110e-01 8.30838382e-01 2.78351933e-01 -5.98596811e-01
-8.82450581e-01 -9.49909747e-01 5.23735769e-02 1.04080713e+00
4.83401418e-01 -2.82169819e-01 -8.28776360e-01 -4.68787074e-01
-2.03265488e-01 1.16834469e-01 -5.87081313e-01 -2.80223731e-02
-7.78560519e-01 -9.58249032e-01 5.67686372e-02 7.77209401e-01
1.15290487e+00 -1.06226647e+00 -5.79705596e-01 3.12832147e-01
-1.41850144e-01 -8.53185177e-01 -8.76540780e-01 1.00579053e-01
-9.37023640e-01 -8.81315708e-01 -7.85879254e-01 -9.67804372e-01
7.52445638e-01 7.09960222e-01 1.02038467e+00 2.90324897e-01
-2.49548838e-01 -2.43652269e-01 -3.63986224e-01 1.57502279e-01
2.85946518e-01 -1.07253246e-01 -4.00447458e-01 2.95868963e-01
-2.24885754e-02 -8.65365624e-01 -1.05039823e+00 3.36128533e-01
-9.75453854e-01 2.32861280e-01 1.01251900e+00 1.02328515e+00
8.89566720e-01 9.24523994e-02 2.66169935e-01 -6.59237921e-01
4.05449301e-01 -1.38471648e-01 -5.84661961e-01 3.90302300e-01
-5.09977818e-01 -1.10251017e-01 8.74709547e-01 -1.44017711e-01
-1.15511513e+00 1.89354852e-01 -2.07465962e-01 -4.78712380e-01
6.08957559e-02 3.08108240e-01 -4.96778786e-01 -2.28065059e-01
1.56385273e-01 5.08138895e-01 -2.80381471e-01 -7.34940946e-01
3.99387926e-01 5.78748286e-01 7.97402143e-01 -4.34673756e-01
5.65963805e-01 6.80872560e-01 -2.21230924e-01 -5.86079001e-01
-6.55558407e-01 -2.70310938e-01 -4.94296730e-01 2.91046426e-02
9.75478768e-01 -1.06190228e+00 -8.25952113e-01 9.73232925e-01
-1.11231124e+00 -4.12535369e-01 -3.28494668e-01 1.43696383e-01
-1.43691033e-01 4.39744323e-01 -9.33777869e-01 -4.25436795e-01
-6.45557940e-01 -1.25800991e+00 9.99464989e-01 5.73996365e-01
4.35731590e-01 -6.27884984e-01 -3.54311466e-01 3.97349417e-01
7.39184082e-01 -6.19730204e-02 8.62587571e-01 1.88228101e-01
-1.11114717e+00 9.18331221e-02 -1.06825280e+00 4.58664596e-01
2.60468721e-01 -1.88673288e-03 -9.32239652e-01 -2.63800383e-01
3.37580293e-02 5.90404607e-02 1.31078732e+00 6.62310600e-01
1.33519232e+00 -2.04755813e-01 -2.13145137e-01 8.59029651e-01
1.61041987e+00 9.54631791e-02 9.56311226e-01 2.75390297e-01
9.49366093e-01 2.20603958e-01 4.42651689e-01 3.24333042e-01
5.78640521e-01 5.67594767e-01 6.81946993e-01 -5.69526911e-01
-5.73274732e-01 -7.25545362e-02 4.24353123e-01 8.93125117e-01
-1.69258475e-01 -1.73121125e-01 -5.01195788e-01 8.03856075e-01
-1.76423538e+00 -7.76158392e-01 -2.11554274e-01 2.13138103e+00
7.54969835e-01 2.77374852e-02 -2.73321092e-01 2.04587802e-01
7.67320096e-01 3.89856011e-01 -5.01206815e-01 -2.39137307e-01
-2.43233159e-01 2.63425946e-01 5.54309249e-01 6.53712571e-01
-1.01368821e+00 9.06645000e-01 5.76825762e+00 1.20049155e+00
-1.25301611e+00 2.13617101e-01 9.02750671e-01 1.39567524e-01
-2.64910251e-01 1.41886503e-01 -7.19956517e-01 4.85639721e-01
2.36979976e-01 2.57656455e-01 2.86996216e-01 5.69357097e-01
2.63386041e-01 -2.24162519e-01 -5.35962045e-01 8.26353133e-01
3.37102301e-02 -1.27535379e+00 3.13537657e-01 -1.00388467e-01
7.11064637e-01 -1.21635407e-01 8.03524703e-02 6.09740354e-02
1.69330940e-01 -9.29988861e-01 7.65141845e-01 3.75045329e-01
6.19535387e-01 -7.25016713e-01 6.59521759e-01 7.85186440e-02
-1.57807744e+00 -2.21830249e-01 -5.05843937e-01 -7.44225308e-02
2.27048844e-01 8.96366417e-01 -5.89375608e-02 7.03971386e-01
1.01943970e+00 1.06325305e+00 -4.68421251e-01 1.00619304e+00
-3.66339952e-01 5.35579026e-01 -1.39029369e-01 4.83281285e-01
3.36944729e-01 -3.80549103e-01 2.85737395e-01 1.11187243e+00
2.96659321e-01 1.78386092e-01 1.35011658e-01 8.19449246e-01
-5.57144023e-02 -7.52026215e-02 2.44498014e-01 2.79165387e-01
2.81513065e-01 1.44109702e+00 -7.62153447e-01 -5.60915649e-01
-6.55493200e-01 1.30403924e+00 3.22969198e-01 2.92109221e-01
-8.54218125e-01 -5.13683677e-01 7.87852287e-01 1.03040449e-01
6.81524813e-01 -1.34902030e-01 -5.54117382e-01 -1.03903842e+00
1.93540871e-01 -6.99351192e-01 2.83117533e-01 -8.78267765e-01
-1.13341975e+00 7.57654190e-01 -5.74458599e-01 -1.20597446e+00
5.77282667e-01 -4.10149723e-01 -8.11930239e-01 9.73129511e-01
-1.95788956e+00 -1.54267848e+00 -7.68441975e-01 9.04206872e-01
7.03290343e-01 4.64689761e-01 3.78140569e-01 5.19472837e-01
-9.11972463e-01 5.43960214e-01 -5.36467172e-02 9.27474275e-02
7.07389653e-01 -1.15214062e+00 3.18707019e-01 1.07735395e+00
-2.02770948e-01 5.89365005e-01 3.25754911e-01 -5.40675282e-01
-1.18982410e+00 -1.14803350e+00 7.68597305e-01 2.16279671e-01
5.25283515e-01 1.44695723e-02 -1.04901886e+00 5.94866157e-01
3.39160830e-01 3.98807585e-01 -7.91024119e-02 -1.98982179e-01
-4.06275511e-01 -7.23791242e-01 -9.70463514e-01 5.87085724e-01
8.69540989e-01 -4.16240722e-01 -2.77570963e-01 7.66138136e-02
7.98272967e-01 -3.12930971e-01 -8.05004954e-01 3.08474094e-01
4.88996089e-01 -1.26649892e+00 1.23683214e+00 1.29954293e-02
7.06980407e-01 -8.14760804e-01 -1.21040553e-01 -1.00111115e+00
-6.83501899e-01 -4.71352249e-01 1.02677375e-01 1.40619946e+00
1.68148309e-01 -6.73839390e-01 5.38661599e-01 2.57556945e-01
-4.43785906e-01 -8.33217561e-01 -8.62065196e-01 -4.02723759e-01
-1.44776657e-01 -1.54540345e-01 7.32134223e-01 7.14773834e-01
-2.95111746e-01 1.60579786e-01 -5.23783743e-01 3.64025712e-01
7.27978647e-01 4.64396268e-01 4.46563005e-01 -7.89115369e-01
-2.57527173e-01 -5.34659505e-01 -2.33982518e-01 -1.55646384e+00
-3.21867794e-01 -5.19416094e-01 1.51131198e-01 -1.59279752e+00
5.05537271e-01 -5.19890785e-01 -6.53108001e-01 6.03176415e-01
-5.11967480e-01 6.44094944e-01 2.36217916e-01 2.92358100e-01
-4.52690303e-01 5.31174481e-01 1.53903437e+00 -3.52227330e-01
-2.07917646e-01 -3.48328441e-01 -8.63724113e-01 6.31540716e-01
6.70462132e-01 -1.09485976e-01 -2.32117534e-01 -8.88919055e-01
-1.89275995e-01 -7.61988088e-02 6.06472135e-01 -1.05207407e+00
6.14899099e-01 -7.41361156e-02 3.54265898e-01 -6.07305586e-01
2.99216062e-01 -6.91418171e-01 -4.80700098e-03 4.90517646e-01
-1.45060392e-02 -7.80485719e-02 5.98239824e-02 5.56104660e-01
-5.92243910e-01 3.33559692e-01 9.67928529e-01 -8.77821222e-02
-9.35861409e-01 5.02177536e-01 -2.76810795e-01 -2.78737307e-01
7.61658370e-01 -2.66069263e-01 -3.37208301e-01 -1.08494714e-01
-5.53429663e-01 2.78141975e-01 4.85668778e-01 1.19534157e-01
6.56356454e-01 -1.09754026e+00 -7.65021384e-01 4.33419704e-01
-2.53822356e-01 1.15134224e-01 8.11740637e-01 1.10616505e+00
-6.60044849e-01 1.86665043e-01 -4.38675135e-01 -6.33387089e-01
-1.29287863e+00 6.25443757e-01 3.11546862e-01 -4.50286925e-01
-7.38003790e-01 1.16580212e+00 5.92453420e-01 1.92488894e-01
-1.49229839e-02 -6.66410148e-01 -2.29027301e-01 -1.22136632e-02
7.86586106e-01 4.15829390e-01 -1.01443186e-01 -5.82047522e-01
-2.86850661e-01 1.09576464e+00 -2.38085091e-01 3.33198935e-01
1.25341618e+00 -4.74557102e-01 -5.17619967e-01 -1.80915833e-01
9.25396442e-01 1.92427889e-01 -1.55051231e+00 -2.29071051e-01
-4.21232611e-01 -7.00423658e-01 4.84056354e-01 -7.28366077e-01
-1.73253989e+00 1.07769120e+00 5.71692824e-01 1.48035228e-01
1.86385667e+00 -2.29900792e-01 1.19511116e+00 -3.81887823e-01
1.68739155e-01 -7.67586112e-01 -1.16476022e-01 3.36947948e-01
7.70187318e-01 -1.04903936e+00 1.79883823e-01 -7.72019982e-01
-3.12944263e-01 9.65629637e-01 5.60964465e-01 -4.76327330e-01
5.71709275e-01 3.57142329e-01 -5.55067025e-02 1.15417041e-01
-4.55870777e-01 -3.91624331e-01 1.00858919e-01 4.79928136e-01
3.55138391e-01 1.50689874e-02 -3.22090119e-01 5.72339118e-01
3.28075141e-01 1.12386003e-01 2.79821724e-01 5.07174790e-01
-5.59806108e-01 -1.05857539e+00 -3.52943718e-01 1.55350864e-01
-5.29118538e-01 -5.39569557e-01 5.29809035e-02 4.77088004e-01
3.34943205e-01 9.56065297e-01 1.99091867e-01 -2.72251397e-01
2.15222731e-01 -5.81804812e-01 2.38628790e-01 -1.68406770e-01
-9.08998489e-01 2.59244174e-01 -3.54460895e-01 -8.56512189e-01
-4.93807793e-01 -3.94848704e-01 -9.83582854e-01 -3.14246953e-01
-2.06770793e-01 -9.67644304e-02 3.07994723e-01 8.28120768e-01
4.41223681e-01 8.21537733e-01 6.47047579e-01 -1.02280629e+00
-9.14211571e-02 -8.45358074e-01 -5.26057184e-01 2.40877509e-01
5.67373633e-01 -2.37498268e-01 -2.67497569e-01 1.48935080e-01] | [10.973854064941406, -1.8547015190124512] |
78548b15-cb09-4f60-af83-dd21a15a32b9 | learning-logic-programs-from-noisy-failures | 2201.03702 | null | https://arxiv.org/abs/2201.03702v2 | https://arxiv.org/pdf/2201.03702v2.pdf | Learning Logic Programs From Noisy Failures | Inductive Logic Programming (ILP) is a form of machine learning (ML) which in contrast to many other state of the art ML methods typically produces highly interpretable and reusable models. However, many ILP systems lack the ability to naturally learn from any noisy or partially misclassified training data. We introduce the relaxed learning from failures approach to ILP, a noise handling modification of the previously introduced learning from failures (LFF) approach which is incapable of handling noise. We additionally introduce the novel Noisy Popper ILP system which implements this relaxed approach and is a modification of the existing Popper system. Like Popper, Noisy Popper takes a generate-test-constrain loop to search its hypothesis space wherein failed hypotheses are used to construct hypothesis constraints. These constraints are used to prune the hypothesis space, making the hypothesis search more efficient. However, in the relaxed setting, constraints are generated in a more lax fashion as to avoid allowing noisy training data to lead to hypothesis constraints which prune optimal hypotheses. Constraints unique to the relaxed setting are generated via hypothesis comparison. Additional constraints are generated by weighing the accuracy of hypotheses against their sizes to avoid overfitting through an application of the minimum description length. We support this new setting through theoretical proofs as well as experimental results which suggest that Noisy Popper improves the noise handling capabilities of Popper but at the cost of overall runtime efficiency. | ['John Wahlig'] | 2021-12-28 | null | null | null | null | ['inductive-logic-programming'] | ['methodology'] | [ 4.32751924e-01 5.88957489e-01 -3.76542151e-01 -3.75851274e-01
-9.22304332e-01 -7.19683826e-01 4.54314739e-01 4.97223496e-01
-2.53287017e-01 1.02144408e+00 -3.37268621e-01 -5.90667963e-01
-4.53792959e-01 -1.03840685e+00 -1.06429291e+00 -4.52849537e-01
-2.01838329e-01 8.20580959e-01 3.91353726e-01 1.55234069e-01
2.22416580e-01 1.65234372e-01 -1.83500457e+00 4.78331715e-01
8.56746376e-01 9.00097549e-01 6.91682100e-02 5.86838126e-01
-1.87073573e-01 8.31629813e-01 -4.20497060e-01 -2.60050148e-01
3.64575088e-01 -2.30325177e-01 -7.97430634e-01 -1.31186277e-01
2.98251718e-01 3.40855308e-02 3.09581578e-01 1.00887477e+00
1.07555009e-01 5.92462271e-02 9.41086560e-02 -1.68219423e+00
-1.84201181e-01 1.03295588e+00 -3.82493399e-02 -3.58344801e-02
5.53986788e-01 1.31120846e-01 1.24476743e+00 -8.89207542e-01
5.21174431e-01 1.39758873e+00 7.41061509e-01 5.23694277e-01
-1.71831274e+00 -4.70721334e-01 2.59770691e-01 8.20641872e-04
-1.30247390e+00 -2.15952307e-01 5.64871848e-01 -1.15507849e-01
1.02565885e+00 6.88224375e-01 3.85209322e-01 1.01098120e+00
-6.59234151e-02 8.79046082e-01 1.17267072e+00 -9.34922576e-01
7.46929467e-01 4.30880696e-01 1.76788643e-01 7.35932469e-01
5.44869900e-01 8.98804665e-02 -4.53385204e-01 -4.95487660e-01
2.71417230e-01 -3.77725124e-01 -8.13717172e-02 -3.89506370e-01
-8.99702132e-01 8.34670007e-01 8.79785395e-04 4.25917432e-02
-1.44397970e-02 2.49192014e-01 4.32017893e-01 3.60584736e-01
1.69199109e-01 8.08006942e-01 -7.94637024e-01 3.55533734e-02
-8.76372755e-01 6.48887813e-01 1.23130941e+00 8.40395570e-01
8.15938354e-01 -2.00938985e-01 -2.04394966e-01 6.97812676e-01
4.55459565e-01 2.82269388e-01 1.59260139e-01 -1.08166802e+00
4.48885083e-01 7.30781555e-01 1.12619020e-01 -7.99254060e-01
-2.41441891e-01 -2.42590442e-01 -1.90946534e-01 3.60969603e-01
3.46710384e-01 -4.66244407e-02 -9.68497574e-01 1.81045139e+00
1.56907022e-01 1.14840999e-01 -1.73871014e-02 3.31756055e-01
2.63176769e-01 6.50055349e-01 1.18061066e-01 -5.16331017e-01
8.91967654e-01 -7.13283241e-01 -5.16983151e-01 -3.54146063e-01
8.31852496e-01 -3.43238980e-01 1.14909232e+00 9.60426331e-01
-1.19323611e+00 -2.78962255e-01 -1.21683002e+00 2.23323405e-01
-3.53350282e-01 -3.11141104e-01 8.44715416e-01 8.04547131e-01
-8.15570593e-01 5.89765251e-01 -7.86063254e-01 8.00475702e-02
4.30967510e-01 6.48247898e-01 -2.96407968e-01 -1.87715992e-01
-1.05132973e+00 8.61849010e-01 8.85646701e-01 2.08368003e-02
-6.20969057e-01 -7.76869178e-01 -1.05818236e+00 -1.27963856e-01
9.70703125e-01 -3.93766403e-01 1.30163884e+00 -9.51589525e-01
-1.19661295e+00 4.15658712e-01 -1.98009551e-01 -7.12127984e-01
5.94900787e-01 -7.09178671e-02 -2.16786280e-01 -5.71922250e-02
-2.31250320e-02 6.17098749e-01 4.59978700e-01 -1.64525640e+00
-6.99914753e-01 1.35182783e-01 1.22008406e-01 -2.33206779e-01
2.98857719e-01 -2.07101896e-01 -4.21802491e-01 -3.40955108e-01
3.06683630e-01 -9.04596746e-01 -3.79615933e-01 -1.92039251e-01
-6.69821560e-01 -3.61007601e-01 6.97950661e-01 -9.02121589e-02
1.51174617e+00 -1.89345610e+00 1.51596209e-02 9.03499067e-01
4.14087772e-02 1.74366251e-01 1.27332127e-02 3.53370577e-01
-2.14594617e-01 6.98540032e-01 -3.55872005e-01 -3.13217640e-01
2.86034912e-01 8.43622744e-01 -3.19116473e-01 4.20926213e-02
5.28540194e-01 5.52475989e-01 -1.03185034e+00 -5.26153326e-01
1.95465475e-01 -2.46579602e-01 -8.19474816e-01 -7.64139518e-02
-9.01215911e-01 -1.03713095e-01 -2.14098215e-01 8.10784161e-01
5.60599864e-01 4.56898883e-02 4.59669560e-01 1.73634291e-01
4.99754064e-02 4.19234514e-01 -1.63948452e+00 1.00225198e+00
-3.84399742e-01 1.37611583e-01 -2.25930929e-01 -1.15806341e+00
9.60034847e-01 2.14703918e-01 1.00197658e-01 -3.56567919e-01
9.41836461e-02 5.15726984e-01 2.85247574e-03 -6.22066200e-01
1.91545099e-01 -3.87623191e-01 -2.70201683e-01 2.77610302e-01
-3.59655544e-02 -2.17397138e-01 6.09101295e-01 1.28883548e-04
1.21702659e+00 1.76292360e-01 3.16780508e-01 -1.55366078e-01
5.98228455e-01 4.36055996e-02 1.10285866e+00 1.39620769e+00
3.24303918e-02 4.86583829e-01 8.67090583e-01 -4.63869423e-01
-9.86708045e-01 -1.08556926e+00 -4.21382695e-01 7.94803739e-01
-2.28798673e-01 -6.96656823e-01 -4.03700054e-01 -1.08565605e+00
2.34125137e-01 1.00038588e+00 -4.67184782e-01 5.22321910e-02
-4.88480598e-01 -8.57504904e-01 5.45222819e-01 5.89791238e-01
2.63112873e-01 -1.24436450e+00 -3.73637557e-01 3.10907304e-01
-4.83285487e-02 -8.56071293e-01 2.65203625e-01 7.94726551e-01
-6.15985215e-01 -1.26345038e+00 4.40608650e-01 -7.00520933e-01
8.36216867e-01 -5.67285955e-01 1.12172198e+00 3.90233517e-01
-2.72837043e-01 2.17256442e-01 -4.47845340e-01 -6.33456290e-01
-8.57082129e-01 -2.59912938e-01 2.52267987e-01 -5.02742231e-01
6.05364919e-01 -5.39417207e-01 1.81155458e-01 2.66576074e-02
-1.16331565e+00 -4.35916483e-02 5.13208091e-01 1.07237613e+00
8.70922685e-01 6.54359043e-01 5.27997017e-01 -1.17745292e+00
5.41244328e-01 -6.55196965e-01 -8.86154175e-01 4.23919886e-01
-8.25884759e-01 4.54344660e-01 8.36936414e-01 -6.25347912e-01
-5.93062878e-01 1.05910242e-01 -3.60745862e-02 -2.78937489e-01
-9.39528868e-02 7.43123710e-01 -3.00590992e-01 -3.50165851e-02
7.07594752e-01 -1.32442012e-01 -1.59541622e-01 -3.09332013e-01
1.82343759e-02 3.79389435e-01 4.98075634e-01 -1.02615750e+00
9.05690074e-01 5.67019312e-03 1.27928495e-01 -1.99894905e-01
-8.41657698e-01 3.55491489e-02 -2.45820940e-01 5.62603064e-02
1.49048716e-01 -3.11328769e-01 -8.33791256e-01 -1.02808222e-01
-7.20165431e-01 -4.38289464e-01 -5.98674595e-01 1.81656435e-01
-6.16971672e-01 2.98643380e-01 -2.57354230e-01 -1.30234778e+00
3.05155888e-02 -1.06689191e+00 7.08556235e-01 2.88396869e-02
-4.86491352e-01 -7.94265211e-01 -1.67672649e-01 9.25745592e-02
-3.97688076e-02 4.83009219e-01 1.31033456e+00 -1.12353516e+00
-4.66738850e-01 -3.86530191e-01 2.40218788e-01 6.33364975e-01
-3.07388693e-01 9.14894119e-02 -6.79190934e-01 -1.14238024e-01
-9.80216712e-02 -5.25123298e-01 6.25354350e-01 -2.03823191e-04
1.40498364e+00 -6.09726250e-01 -3.50986332e-01 1.96479261e-01
1.72389591e+00 2.19078675e-01 6.34611070e-01 6.21959865e-01
1.79409429e-01 4.56734568e-01 6.34099066e-01 2.68723488e-01
2.83165486e-04 3.78207445e-01 4.02745396e-01 1.59568146e-01
3.81504953e-01 -4.79608744e-01 1.88759521e-01 1.40183970e-01
3.44506085e-01 -3.29877913e-01 -1.12576866e+00 4.80162770e-01
-2.02662945e+00 -9.94759679e-01 -3.73306349e-02 2.54922724e+00
1.24872756e+00 8.20088387e-01 -9.98150259e-02 8.92616749e-01
3.87131721e-01 -3.27541202e-01 -1.74721971e-01 -8.65623057e-01
-1.32232577e-01 5.71307182e-01 4.82541561e-01 8.35024893e-01
-9.65092003e-01 6.76980495e-01 6.53403997e+00 7.06421256e-01
-5.42108595e-01 -2.39346310e-01 4.92900789e-01 -1.45738453e-01
-5.00378609e-01 3.80323142e-01 -9.00654376e-01 5.30189693e-01
8.11478376e-01 -4.30190675e-02 5.73741913e-01 1.00434554e+00
4.73273210e-02 -5.68718851e-01 -1.56639433e+00 5.81809044e-01
-9.14475024e-02 -1.28254962e+00 -5.31273372e-02 -1.71111524e-01
6.18254006e-01 -3.37083429e-01 -1.89330041e-01 5.87786853e-01
4.94198650e-01 -1.13549101e+00 7.65100300e-01 4.66237247e-01
3.09264332e-01 -9.72796977e-01 9.88221765e-01 5.21134913e-01
-7.52802074e-01 -5.47139168e-01 -1.95915952e-01 -2.60737181e-01
-1.43004343e-01 8.23301792e-01 -1.15075707e+00 3.88476372e-01
5.50334454e-01 4.66959849e-02 -5.33731282e-01 1.01438963e+00
-4.09527093e-01 7.84734547e-01 -6.90642774e-01 8.91567096e-02
1.28351167e-01 2.34261751e-01 6.67710185e-01 1.11108577e+00
-1.63855210e-01 -2.02323779e-01 6.82881355e-01 1.13140213e+00
6.02963045e-02 -2.61439711e-01 -4.22271550e-01 1.97409630e-01
9.16973770e-01 8.86227548e-01 -6.16249323e-01 -3.06444436e-01
-1.05485164e-01 3.12173009e-01 2.69118816e-01 1.87066481e-01
-7.30848491e-01 -2.27933183e-01 1.04714990e-01 6.55797347e-02
2.48280466e-01 6.90125749e-02 -4.17146355e-01 -6.33052230e-01
3.64651203e-01 -1.17566729e+00 4.41117257e-01 -4.69403744e-01
-1.33498204e+00 1.47055820e-01 3.67352843e-01 -8.66441905e-01
-3.59539568e-01 -6.01197183e-01 -5.46421647e-01 7.24456489e-01
-1.24839175e+00 -8.51551294e-01 9.06172097e-02 7.69292489e-02
4.85140115e-01 3.30895990e-01 9.36291039e-01 -9.01224241e-02
-6.71427488e-01 6.86173379e-01 -2.95167834e-01 -1.91419199e-01
4.95035559e-01 -1.58610117e+00 -1.34231761e-01 9.97562528e-01
-1.97005495e-01 7.92328000e-01 1.03460121e+00 -8.52845371e-01
-1.39691794e+00 -1.17419457e+00 1.05803871e+00 -3.95593196e-01
5.47786832e-01 -5.32084346e-01 -9.72349524e-01 8.09550941e-01
-4.85460013e-01 1.78806305e-01 6.61118746e-01 4.35334563e-01
-5.36397040e-01 -1.42659575e-01 -1.41394889e+00 5.67628026e-01
6.90823734e-01 -2.23801017e-01 -8.68825495e-01 3.40432465e-01
5.60388386e-01 -3.30162168e-01 -6.27737999e-01 6.81642592e-01
3.31567317e-01 -7.74432540e-01 6.32761002e-01 -5.08831561e-01
2.65441418e-01 -7.22576261e-01 -3.26884866e-01 -7.56718814e-01
-2.46263117e-01 -8.15318346e-01 -3.22371244e-01 1.30773115e+00
8.24049711e-01 -5.77751279e-01 6.87697649e-01 9.00407434e-01
-1.36768192e-01 -1.11993742e+00 -9.93295074e-01 -9.89313424e-01
-1.68438807e-01 -1.05478561e+00 4.50449497e-01 7.09889889e-01
4.16248024e-01 -3.21736485e-02 -1.11078545e-01 2.67086029e-01
5.16742170e-01 -1.91950455e-01 5.13985515e-01 -1.27502644e+00
-8.25244904e-01 -1.81101680e-01 -3.24114561e-01 -2.69952893e-01
1.40633896e-01 -8.82942259e-01 4.79451418e-01 -1.26134527e+00
2.10714519e-01 -9.32987094e-01 -2.79013574e-01 1.06958461e+00
-1.94064781e-01 2.09328130e-01 1.53497636e-01 -1.30014643e-01
-6.35164082e-01 -7.55157396e-02 4.85712647e-01 7.21261427e-02
-4.68195349e-01 -2.58072279e-02 -8.15111816e-01 8.47957611e-01
7.86438942e-01 -5.42599022e-01 -3.90648037e-01 2.01102868e-02
7.85767019e-01 -2.69564390e-01 3.40621561e-01 -9.86813128e-01
9.69301090e-02 -2.47639090e-01 3.50640327e-01 -5.78107297e-01
-1.14438981e-01 -7.42758632e-01 2.73293674e-01 4.67777908e-01
-4.70800936e-01 -5.84528930e-02 4.37625885e-01 4.87424254e-01
7.63235465e-02 -6.39700830e-01 5.75331032e-01 -1.74338430e-01
-4.91621733e-01 -2.01861739e-01 -3.88552040e-01 2.34577116e-02
1.12860370e+00 -2.96904773e-01 -2.15665236e-01 7.88387805e-02
-7.56228030e-01 5.65224707e-01 3.83921981e-01 2.71685362e-01
3.95896137e-01 -1.02401924e+00 -4.09056813e-01 3.65054131e-01
1.19402789e-01 4.37951267e-01 -2.73428172e-01 7.57847488e-01
-3.09495300e-01 3.88179839e-01 2.69958615e-01 -3.66464972e-01
-1.20406663e+00 7.92818666e-01 2.25126520e-01 -4.99773949e-01
-5.29033959e-01 8.80756557e-01 -3.74763101e-01 -4.43861097e-01
5.57265460e-01 -5.58224678e-01 1.94041163e-01 -3.37521255e-01
6.69722855e-01 2.23067835e-01 2.15810031e-01 2.48166382e-01
-4.52046454e-01 -3.71501446e-02 -1.03936307e-01 -8.29871446e-02
1.29016876e+00 1.48431644e-01 -2.21205175e-01 6.06861115e-01
6.67665184e-01 2.82909930e-01 -9.40234601e-01 -1.64277151e-01
4.90449518e-01 -2.79817790e-01 -9.15345773e-02 -1.25765944e+00
-2.97085255e-01 1.89389363e-01 2.15250611e-01 3.97932827e-01
1.20671034e+00 -4.83012311e-02 3.25250357e-01 6.38870478e-01
2.93344498e-01 -1.31946039e+00 -1.38232559e-01 5.02813041e-01
6.49473786e-01 -1.10854387e+00 9.40778106e-02 -5.59727967e-01
-2.35380441e-01 1.13388383e+00 7.19771743e-01 -6.96659088e-03
1.78394645e-01 7.87009120e-01 -4.45664972e-01 1.19021140e-01
-1.00212455e+00 -3.35647464e-02 -8.83341655e-02 4.88945544e-01
1.18717225e-02 -2.42208950e-02 -5.36970496e-01 9.14840043e-01
-2.51167089e-01 1.52191326e-01 4.70579088e-01 1.33638060e+00
-5.88012397e-01 -1.63614845e+00 -5.95557928e-01 5.30197978e-01
-4.29361552e-01 -1.29033357e-01 -2.79752284e-01 9.18049276e-01
7.87310421e-01 1.12135363e+00 -6.77115470e-02 -3.56672466e-01
2.22463846e-01 4.88631874e-01 5.37501991e-01 -9.50952351e-01
-5.78645349e-01 -2.62087584e-01 4.81842369e-01 -4.63622719e-01
-1.16270542e-01 -3.95348698e-01 -1.39725888e+00 -1.08232751e-01
-6.33941531e-01 2.91554511e-01 5.65918922e-01 1.24636185e+00
3.61843295e-02 3.25026155e-01 5.70182979e-01 -4.41207111e-01
-7.10609615e-01 -6.30789161e-01 -3.37002009e-01 2.64439851e-01
2.41781995e-01 -7.04904854e-01 -5.70792913e-01 -1.77513510e-01] | [8.725825309753418, 6.6622138023376465] |
fd4bcf17-9ebe-4ab3-ab62-09440e18763e | ntire-2023-challenge-on-light-field-image | 2304.10415 | null | https://arxiv.org/abs/2304.10415v1 | https://arxiv.org/pdf/2304.10415v1.pdf | NTIRE 2023 Challenge on Light Field Image Super-Resolution: Dataset, Methods and Results | In this report, we summarize the first NTIRE challenge on light field (LF) image super-resolution (SR), which aims at super-resolving LF images under the standard bicubic degradation with a magnification factor of 4. This challenge develops a new LF dataset called NTIRE-2023 for validation and test, and provides a toolbox called BasicLFSR to facilitate model development. Compared with single image SR, the major challenge of LF image SR lies in how to exploit complementary angular information from plenty of views with varying disparities. In total, 148 participants have registered the challenge, and 11 teams have successfully submitted results with PSNR scores higher than the baseline method LF-InterNet \cite{LF-InterNet}. These newly developed methods have set new state-of-the-art in LF image SR, e.g., the winning method achieves around 1 dB PSNR improvement over the existing state-of-the-art method DistgSSR \cite{DistgLF}. We report the solutions proposed by the participants, and summarize their common trends and useful tricks. We hope this challenge can stimulate future research and inspire new ideas in LF image SR. | ['Yulan Guo', 'Radu Timofte', 'Jungang Yang', 'Zhengyu Liang', 'Longguang Wang', 'Yingqian Wang'] | 2023-04-20 | null | null | null | null | ['image-super-resolution'] | ['computer-vision'] | [ 7.00669289e-01 -2.74626583e-01 -1.42169592e-03 -2.68341243e-01
-1.08408093e+00 -3.48412067e-01 2.04267800e-01 -9.38806474e-01
-7.53303692e-02 1.07241535e+00 5.65428078e-01 1.38279393e-01
4.62731067e-03 -2.78929055e-01 -7.49812782e-01 -6.35596395e-01
8.77191722e-02 -2.02965364e-01 4.08017069e-01 -3.06256592e-01
5.76219261e-01 4.78588164e-01 -1.75602496e+00 7.25846767e-01
8.53301466e-01 9.58543003e-01 6.49461806e-01 7.79163241e-01
2.21622512e-01 1.07424152e+00 -3.18968266e-01 -5.34728728e-02
4.78712589e-01 -4.04087782e-01 -9.71471667e-01 -3.61866981e-01
1.54957128e+00 -7.34374166e-01 -5.27986169e-01 1.08196700e+00
9.10278857e-01 5.90679832e-02 -1.09260511e-02 -6.29422307e-01
-1.16895354e+00 4.17369694e-01 -8.52898359e-01 9.24869061e-01
6.23738647e-01 2.73013026e-01 7.31759131e-01 -1.30101287e+00
1.11271226e+00 1.19578445e+00 5.07887006e-01 8.31673861e-01
-1.53099084e+00 -9.56044614e-01 -2.36372784e-01 6.30107045e-01
-1.22357512e+00 -9.53740656e-01 2.56146312e-01 -1.34983927e-01
8.23020160e-01 2.92963713e-01 2.91193038e-01 9.55989182e-01
2.15226114e-01 5.24330437e-01 1.97236562e+00 -1.28145158e-01
-1.43137127e-01 -2.53273606e-01 2.01213658e-02 1.53980970e-01
4.02606726e-02 4.54437733e-01 -1.16481400e+00 1.67591840e-01
1.23999345e+00 -4.78641182e-01 -8.65231633e-01 1.83424637e-01
-1.63536823e+00 2.08428383e-01 6.04471147e-01 3.58745158e-01
5.85412085e-02 1.76825449e-01 -3.72799337e-02 3.90244722e-01
7.45957077e-01 5.37816107e-01 -2.22095534e-01 1.24240316e-01
-9.83517230e-01 3.16026688e-01 -4.45011556e-02 8.16823304e-01
5.46313345e-01 5.19648530e-02 -3.25676918e-01 9.96599615e-01
-2.75020242e-01 6.88377082e-01 -4.30720448e-02 -1.72211027e+00
2.48369515e-01 -2.59485185e-01 2.99316138e-01 -6.96095943e-01
-1.86723620e-01 -7.85201073e-01 -8.62523556e-01 4.61836457e-01
4.14619558e-02 4.62235957e-01 -6.34825766e-01 1.63201237e+00
-3.90488766e-02 6.79566383e-01 -2.09846631e-01 1.34363687e+00
1.37472510e+00 3.22154015e-01 -5.23567975e-01 -5.21845698e-01
1.04320848e+00 -9.72245872e-01 -7.06378043e-01 -5.52511550e-02
-3.76450062e-01 -1.15684402e+00 1.12192750e+00 8.42930138e-01
-1.67680454e+00 -9.51423585e-01 -8.58271718e-01 -5.27195990e-01
4.80642229e-01 9.97133851e-02 5.64225554e-01 2.97067791e-01
-1.70042408e+00 7.26588666e-01 -2.58152157e-01 -5.61642013e-02
7.07014263e-01 1.17810942e-01 -3.08544785e-01 -5.57917595e-01
-1.08225787e+00 7.99019456e-01 -1.87497973e-01 -2.85774879e-02
-9.63194311e-01 -1.30803382e+00 -3.26171696e-01 -3.57004315e-01
1.68919161e-01 -1.00241649e+00 1.09026217e+00 -3.96794885e-01
-1.52621877e+00 1.20326030e+00 -5.46729863e-01 -3.85453999e-01
4.21261400e-01 -3.91678751e-01 -7.89762914e-01 5.24486721e-01
3.15895081e-01 6.56316936e-01 1.01158345e+00 -1.41834497e+00
-7.11705923e-01 -2.69275695e-01 1.07163843e-02 2.10457519e-01
2.09626302e-01 6.09473944e-01 -1.85495481e-01 -6.23968303e-01
2.06694603e-01 -5.59531093e-01 2.56861504e-02 -3.76082659e-02
-8.05518925e-02 8.88999104e-02 6.48325026e-01 -6.12454712e-01
9.66853976e-01 -1.91218913e+00 1.83634579e-01 -6.27842903e-01
7.71347642e-01 1.83552057e-01 -1.51950449e-01 -9.69711095e-02
-2.80172706e-01 -1.65272027e-01 -6.99611157e-02 -4.16977257e-01
-5.65512657e-01 -2.43528306e-01 -8.22808444e-01 6.89095676e-01
-2.05121189e-01 7.71729767e-01 -1.06208825e+00 -2.98399806e-01
2.66595930e-01 7.30272472e-01 -3.96866173e-01 2.25705262e-02
5.72551377e-02 1.25919604e+00 -2.13603675e-02 5.53105056e-01
1.29951191e+00 -6.83002472e-01 -2.98829019e-01 -7.02321351e-01
-4.07011479e-01 3.60793740e-01 -1.04327250e+00 1.93708193e+00
-4.43982750e-01 9.42283154e-01 7.74575248e-02 -2.78895527e-01
7.41557360e-01 -7.24277869e-02 4.43899184e-01 -8.67769659e-01
-4.98647630e-01 4.69934076e-01 -4.94935155e-01 -2.75129229e-01
5.38035095e-01 -6.38481826e-02 6.84025168e-01 4.01701003e-01
-5.76809533e-02 -2.60117233e-01 1.41499788e-01 3.02863985e-01
1.05917490e+00 1.38278663e-01 -1.27925530e-01 -2.79555827e-01
7.61178136e-01 -5.77501714e-01 3.57065111e-01 7.46907532e-01
-1.83484331e-01 1.36843193e+00 -1.99202493e-01 -5.61456203e-01
-9.96054053e-01 -1.59801745e+00 -6.04308844e-01 9.65583682e-01
4.85558897e-01 -3.04752290e-01 -2.83852577e-01 -3.06407779e-01
-5.36415160e-01 3.57356668e-01 -4.26365733e-01 4.87993687e-01
-7.87349403e-01 -6.32138371e-01 2.80147463e-01 1.60231709e-01
1.15928602e+00 -8.47190261e-01 -4.52331513e-01 -2.65748501e-01
-8.96037340e-01 -1.67035890e+00 -5.80101252e-01 -7.85607815e-01
-7.04942346e-01 -9.72989738e-01 -9.98811245e-01 -5.27486742e-01
2.41575196e-01 9.90450501e-01 1.51563811e+00 -2.13547312e-02
-4.78785396e-01 1.48886859e-01 -1.55416831e-01 1.34632975e-01
1.68122686e-02 -3.54270190e-01 1.67882234e-01 -5.37055023e-02
-2.35254079e-01 -8.89160931e-01 -1.40899897e+00 5.90913773e-01
-6.98381126e-01 3.61423850e-01 5.47689915e-01 5.91067255e-01
7.70144939e-01 -1.97238117e-01 4.42126989e-01 -5.42177677e-01
2.86948621e-01 3.59108783e-02 -4.52872574e-01 4.63764668e-02
-7.16012239e-01 -3.08841527e-01 4.76845205e-01 -1.14065610e-01
-1.38680434e+00 -5.71629107e-01 5.06301895e-02 -5.35845995e-01
8.77831131e-02 -2.76882797e-01 4.04815525e-01 -7.75868177e-01
1.01502597e+00 4.47598577e-01 -1.62739649e-01 -6.87177479e-01
3.36231947e-01 5.16580820e-01 1.10994220e+00 -2.96560109e-01
6.49657845e-01 1.02351868e+00 2.82871634e-01 -6.79993570e-01
-1.16659176e+00 -4.57133591e-01 -2.76596129e-01 -3.73423040e-01
7.31836140e-01 -1.23998535e+00 -7.79650450e-01 5.17457604e-01
-1.11613715e+00 -3.00455123e-01 -2.86517590e-01 3.84804815e-01
-8.65273952e-01 5.58128476e-01 -7.94183075e-01 -3.43738824e-01
-3.61980617e-01 -1.18166280e+00 1.41038477e+00 3.70864362e-01
3.76944184e-01 -4.69630986e-01 8.44294950e-02 1.00339496e+00
1.08296061e+00 2.02889815e-02 8.99150372e-02 5.49740314e-01
-1.20616448e+00 7.30584800e-01 -9.76075053e-01 5.55414319e-01
-2.78340746e-02 -4.93947357e-01 -1.24639702e+00 -7.34043300e-01
2.31036589e-01 -4.22050804e-01 1.23351145e+00 6.59337461e-01
1.32155192e+00 1.22299939e-01 -2.13282816e-02 1.20223022e+00
1.53508604e+00 -4.15801942e-01 1.14246297e+00 3.28989685e-01
5.65479755e-01 2.64632434e-01 6.54855549e-01 1.58421442e-01
3.21980804e-01 1.10731971e+00 4.27717596e-01 -1.75981238e-01
-1.06010234e+00 4.05725650e-02 4.68183696e-01 3.79096866e-01
-7.10881174e-01 2.63634384e-01 -3.87741506e-01 1.45290449e-01
-1.22339118e+00 -1.26573241e+00 -3.69696021e-01 2.17912769e+00
9.95117605e-01 -3.26234251e-01 -1.08749963e-01 -1.10287711e-01
6.78581059e-01 6.32933676e-01 -5.11020422e-01 2.33715162e-01
-7.50193059e-01 5.99424124e-01 5.66913366e-01 6.96477354e-01
-8.33633602e-01 9.04029250e-01 7.19640350e+00 1.08821201e+00
-1.28701246e+00 4.88384306e-01 7.51470745e-01 -4.26120102e-01
-1.80926621e-01 -5.91195673e-02 -1.04095924e+00 4.60740089e-01
6.25769854e-01 -1.09102160e-01 9.73893046e-01 2.59185463e-01
3.77263069e-01 -2.73404866e-01 -7.48350501e-01 1.30059993e+00
3.69798362e-01 -1.74783897e+00 -1.95302635e-01 5.45723960e-02
1.08896887e+00 5.35995543e-01 5.10238707e-01 -2.54660666e-01
9.29264054e-02 -1.43392098e+00 2.83071697e-01 7.64254630e-01
1.76577842e+00 -2.98257381e-01 3.26279134e-01 -1.91275198e-02
-1.12085569e+00 -7.98505843e-02 -6.61228120e-01 1.48640007e-01
3.76715183e-01 7.95161009e-01 -1.77505404e-01 8.89783144e-01
1.39883208e+00 1.18299305e+00 -6.10697150e-01 9.27731335e-01
-2.38196746e-01 3.53011072e-01 -2.43199673e-02 1.02307034e+00
-5.97576976e-01 -3.48533951e-02 1.09267783e+00 8.83147240e-01
3.68606240e-01 2.97466010e-01 -1.92331240e-01 1.08315480e+00
-2.59930883e-02 -3.55338454e-01 -3.11151475e-01 5.85821450e-01
2.68253446e-01 1.24364662e+00 -3.48436505e-01 -6.20726161e-02
-2.31007591e-01 1.04115629e+00 -5.33413962e-02 4.56473202e-01
-6.79389596e-01 1.74257383e-01 6.68240845e-01 4.82035220e-01
1.67239383e-01 -1.79832652e-01 -3.65116030e-01 -1.37348247e+00
7.72922337e-02 -9.71989989e-01 1.42166182e-01 -1.55756104e+00
-1.27835429e+00 8.31954360e-01 -5.64139560e-02 -1.42690742e+00
2.31273800e-01 -2.87385583e-01 -2.26407811e-01 1.08027124e+00
-2.25371337e+00 -1.16614437e+00 -7.23210394e-01 8.07229161e-01
5.75994968e-01 -2.43568309e-02 2.92832315e-01 5.15898705e-01
-1.15142472e-01 3.03644001e-01 1.11760376e-02 -4.88540322e-01
1.25692546e+00 -9.48821425e-01 4.11722332e-01 9.65674341e-01
-2.72451788e-02 6.59758568e-01 9.25526917e-01 -5.63076019e-01
-1.26008379e+00 -7.96229303e-01 7.97228336e-01 -6.80576801e-01
3.91037703e-01 -4.14141640e-02 -6.71865284e-01 5.69799721e-01
1.50718629e-01 6.93770766e-01 1.09597839e-01 -3.09876919e-01
-5.42027414e-01 -3.29096466e-01 -1.27313924e+00 3.78883779e-01
1.70932472e+00 -5.69654405e-01 -3.47489476e-01 3.26925099e-01
9.62777734e-01 -8.00660193e-01 -7.61891663e-01 7.28128850e-01
5.11160970e-01 -1.86253917e+00 1.68433714e+00 -1.18698040e-02
8.49653959e-01 -6.47913456e-01 -2.84006745e-01 -1.06107450e+00
-4.71489578e-01 -9.99098003e-01 -1.19903676e-01 7.41795838e-01
-1.35362655e-01 -6.81510568e-01 5.55468202e-01 -3.23992446e-02
-1.32371053e-01 -4.02239740e-01 -9.52020645e-01 -9.47942793e-01
-2.78275013e-01 -2.05886856e-01 3.29466879e-01 7.35229611e-01
-7.48673201e-01 2.40362003e-01 -6.24157965e-01 2.47912351e-02
1.38357008e+00 3.40342969e-01 8.26763391e-01 -9.61294591e-01
-4.59875464e-01 -1.48167789e-01 -1.16691798e-01 -1.50219119e+00
-1.77732944e-01 -8.18394125e-01 -3.40969235e-01 -1.40980685e+00
5.28578997e-01 -2.49032319e-01 -1.94294825e-01 -6.54588863e-02
-1.58247173e-01 9.56827402e-01 4.02680904e-01 5.37413776e-01
-8.41181219e-01 2.96130866e-01 1.94287884e+00 2.37962380e-01
1.11170225e-01 -1.64081216e-01 -9.51528847e-01 4.18982595e-01
2.90631294e-01 1.06469855e-01 -3.05287421e-01 -6.22623086e-01
4.52783883e-01 1.81681469e-01 8.07653844e-01 -1.02899253e+00
2.69056886e-01 -1.48488760e-01 4.46278006e-01 -6.77425385e-01
5.24225533e-01 -3.35275561e-01 4.44903791e-01 -1.98587626e-02
-3.82231504e-01 -3.88541996e-01 -1.03809461e-01 3.52194250e-01
-1.37508094e-01 4.38671678e-01 1.23854911e+00 -2.70059139e-01
-6.51232541e-01 4.00157362e-01 4.03948456e-01 2.80525178e-01
4.34312820e-01 -2.50183254e-01 -1.17074752e+00 -5.22684753e-01
-6.83730483e-01 -9.60423127e-02 6.69753253e-01 2.48977974e-01
9.74230051e-01 -1.25461662e+00 -1.33643007e+00 2.39263564e-01
-2.17787489e-01 -1.70366630e-01 8.98340046e-01 1.20219135e+00
-4.15435165e-01 3.27777803e-01 -5.09054542e-01 -7.49432385e-01
-1.51020300e+00 2.76817977e-01 3.77635926e-01 -2.66323775e-01
-9.52958822e-01 1.14954889e+00 5.55158496e-01 7.56147504e-02
-1.71914428e-01 5.56342974e-02 -3.00098449e-01 -5.63328326e-01
1.25006819e+00 8.38682234e-01 4.28289473e-02 -4.93707478e-01
-2.06529632e-01 1.16536748e+00 -1.20944433e-01 9.74927461e-05
1.64863515e+00 -7.63831496e-01 -4.45714027e-01 1.62086442e-01
9.33386028e-01 3.45423162e-01 -1.26183510e+00 -5.44939697e-01
-6.52387857e-01 -1.31998837e+00 4.18813080e-01 -9.98029530e-01
-1.24783993e+00 6.69497430e-01 8.94823551e-01 -2.58289665e-01
1.62535191e+00 1.84369192e-01 1.02910113e+00 -3.03449571e-01
8.37748230e-01 -5.89415073e-01 2.54803360e-01 3.80167305e-01
1.39460993e+00 -1.21938586e+00 2.97084033e-01 -8.33449483e-01
-3.51854265e-01 1.12482345e+00 4.10303563e-01 -2.22962931e-01
4.21458364e-01 2.48422980e-01 -1.36729673e-01 -1.70790285e-01
-7.80389905e-01 -2.08308131e-01 4.42943096e-01 7.03691602e-01
4.76187587e-01 -3.56961846e-01 -3.13151777e-01 7.54776746e-02
-3.75506401e-01 7.01833248e-01 7.22260475e-01 2.59915352e-01
-4.19351041e-01 -8.73143435e-01 -3.99903119e-01 1.48294419e-01
-7.68109798e-01 -4.68471110e-01 7.45466202e-02 1.75239488e-01
1.52549386e-01 1.10306132e+00 -2.03601927e-01 -2.90845424e-01
1.99167266e-01 -8.43725085e-01 1.13500619e+00 -4.54498798e-01
-2.56985247e-01 3.97820175e-02 6.37992918e-02 -1.29555500e+00
-7.41830885e-01 -5.17413914e-01 -7.77463257e-01 -5.34010172e-01
7.32563287e-02 -4.34749901e-01 3.34131688e-01 4.35230285e-01
5.56455314e-01 2.89666086e-01 8.16914082e-01 -1.22542679e+00
-2.41582021e-01 -7.33171880e-01 -4.55803216e-01 2.56861210e-01
7.33433664e-01 -6.85690582e-01 -7.18764544e-01 5.39218709e-02] | [10.919062614440918, -2.130563735961914] |
ba7bdbaf-9fa6-499f-b23d-9378a15f887f | calibration-of-p-values-for-calibration-and | 2202.00100 | null | https://arxiv.org/abs/2202.00100v7 | https://arxiv.org/pdf/2202.00100v7.pdf | Calibration of P-values for calibration and for deviation of a subpopulation from the full population | The author's recent research papers, "Cumulative deviation of a subpopulation from the full population" and "A graphical method of cumulative differences between two subpopulations" (both published in volume 8 of Springer's open-access "Journal of Big Data" during 2021), propose graphical methods and summary statistics, without extensively calibrating formal significance tests. The summary metrics and methods can measure the calibration of probabilistic predictions and can assess differences in responses between a subpopulation and the full population while controlling for a covariate or score via conditioning on it. These recently published papers construct significance tests based on the scalar summary statistics, but only sketch how to calibrate the attained significance levels (also known as "P-values") for the tests. The present article reviews and synthesizes work spanning many decades in order to detail how to calibrate the P-values. The present paper presents computationally efficient, easily implemented numerical methods for evaluating properly calibrated P-values, together with rigorous mathematical proofs guaranteeing their accuracy, and illustrates and validates the methods with open-source software and numerical examples. | ['Mark Tygert'] | 2022-01-31 | null | null | null | null | ['mathematical-proofs'] | ['miscellaneous'] | [ 4.23831819e-03 7.87880719e-02 -3.94313961e-01 -4.66963947e-01
-1.06357419e+00 -5.11610210e-01 5.35810411e-01 5.23267210e-01
-2.02878684e-01 1.58228195e+00 2.05962569e-01 -4.36937809e-01
-6.94330037e-01 -9.22826648e-01 -6.77789092e-01 -9.83829916e-01
-3.63735855e-01 4.67748612e-01 -1.01749949e-01 3.06214988e-01
5.29388070e-01 2.96517193e-01 -1.69631827e+00 -2.86615402e-01
1.23636198e+00 8.16509962e-01 -5.01028001e-01 6.79825187e-01
1.00966476e-01 1.13672286e-01 -9.48902607e-01 -5.43888807e-01
-1.29792392e-01 -4.02456105e-01 -1.35865808e-01 -3.10253799e-01
4.09171462e-01 1.75165813e-02 1.00934535e-01 1.05691659e+00
7.03911185e-01 -2.09918678e-01 1.12571383e+00 -1.55385268e+00
-6.62528038e-01 8.72043312e-01 -4.96720493e-01 -1.61065340e-01
3.94558460e-01 8.77171848e-03 7.71797836e-01 -5.70059657e-01
6.94179628e-03 1.34216726e+00 1.03612685e+00 -1.77777894e-02
-1.40317380e+00 -6.27853453e-01 -2.23928422e-01 -4.55571115e-01
-1.55031157e+00 -2.77713742e-02 3.14671248e-02 -7.83541977e-01
3.13408554e-01 8.35374534e-01 5.72795272e-01 9.86042559e-01
5.64778984e-01 1.70137450e-01 1.38123596e+00 -3.60410005e-01
6.23297513e-01 1.87029853e-01 5.60719132e-01 2.02518240e-01
9.51108158e-01 4.41000164e-01 -3.06582958e-01 -7.27387905e-01
7.30907500e-01 -2.17618123e-01 -6.24787807e-02 -3.47134054e-01
-1.32970440e+00 8.61806154e-01 -5.21401428e-02 1.13351122e-01
-3.84176046e-01 2.95068562e-01 1.01777919e-01 -2.28173077e-01
6.87062323e-01 -1.74118400e-01 -4.71856326e-01 -2.62906160e-02
-8.78814042e-01 5.14477968e-01 1.08985007e+00 9.02453363e-01
3.95608097e-01 -7.18663931e-02 -8.69507670e-01 5.46276510e-01
2.91079283e-01 1.13979900e+00 1.15049049e-01 -7.56467462e-01
2.15676591e-01 2.84312576e-01 5.71055472e-01 -7.57933080e-01
-5.86924136e-01 -3.88571739e-01 -1.01205885e+00 -5.27000017e-02
6.38022542e-01 -7.74302959e-01 -4.24100816e-01 1.88683856e+00
3.89761239e-01 1.48539124e-02 -1.72535926e-01 3.69826764e-01
5.85379541e-01 5.32142460e-01 4.47886616e-01 -4.85820144e-01
1.29466593e+00 -8.78610238e-02 -6.80177629e-01 3.50429624e-01
2.69333273e-01 -4.38960314e-01 9.21788931e-01 3.18551272e-01
-1.36289561e+00 -2.87985772e-01 -8.27790380e-01 5.58483064e-01
-4.41045523e-01 7.49612972e-02 7.92059541e-01 1.07024229e+00
-1.24080813e+00 5.46228945e-01 -4.63537395e-01 -4.43225056e-01
1.90726921e-01 4.30001944e-01 -1.31884053e-01 4.96032208e-01
-1.21330941e+00 6.45691335e-01 6.87618414e-03 -5.24271466e-02
-5.82781196e-01 -1.17198122e+00 -5.18381894e-01 2.34064266e-01
-1.44070879e-01 -1.12009227e+00 8.07787776e-01 -3.17077070e-01
-1.37803209e+00 6.29038453e-01 1.51501387e-01 -3.10613602e-01
8.31898153e-01 -8.20479244e-02 -3.71547014e-01 -3.40786904e-01
1.13753438e-01 3.43385428e-01 1.71394289e-01 -9.50156510e-01
-7.76114583e-01 -4.63793755e-01 -3.72879863e-01 -3.45629513e-01
-5.56133725e-02 1.82740822e-01 6.20403029e-02 -6.83697164e-01
-2.11225584e-01 -4.82099742e-01 -5.21887004e-01 -3.38597149e-01
-6.97898567e-01 -2.24021927e-01 1.83312856e-02 -3.59377265e-01
1.35282910e+00 -1.96300066e+00 -2.83408374e-01 5.55832922e-01
-5.23802750e-02 -5.64693809e-01 1.75321922e-01 6.11886919e-01
4.75299582e-02 2.85750777e-01 -3.98664474e-01 8.19904283e-02
3.15159261e-01 -2.79635284e-02 -2.86218733e-01 8.87158513e-01
6.15982823e-02 6.94269419e-01 -6.03494406e-01 -4.65167552e-01
8.33742693e-02 5.29698312e-01 -4.01410401e-01 1.41327060e-03
2.70534009e-01 3.38797778e-01 -2.33786926e-01 4.69111681e-01
9.61900413e-01 -1.94718167e-01 1.50808454e-01 6.33506179e-02
-5.59096754e-01 -2.39525735e-01 -1.28412330e+00 5.95235527e-01
-2.15554517e-02 1.92178279e-01 -1.17018387e-01 -1.02938771e+00
1.09545803e+00 1.33678705e-01 2.19769627e-01 -1.27932310e-01
2.09440500e-01 4.03186142e-01 -5.13965249e-01 -3.48363034e-02
2.60823339e-01 -4.03438210e-01 -7.00654984e-01 4.54043597e-02
-2.34730676e-01 -2.46085167e-01 2.45722815e-01 -1.34377591e-02
8.09780955e-01 -1.60505340e-01 6.19822025e-01 -8.56929302e-01
6.26340389e-01 -2.03156814e-01 5.64812779e-01 1.07775009e+00
-6.11465611e-02 2.44065687e-01 1.13217235e+00 7.92305097e-02
-7.97252715e-01 -1.59420204e+00 -7.79276133e-01 6.86464071e-01
-1.91317692e-01 1.38560114e-02 -6.99660599e-01 -5.89532405e-02
5.86386263e-01 7.84274220e-01 -1.11861575e+00 2.06396148e-01
6.09850660e-02 -1.12718022e+00 5.02691269e-01 5.33835292e-01
2.21835122e-01 -2.74865806e-01 -2.34797478e-01 -2.04380751e-01
1.44942448e-01 -3.99478674e-01 6.14571013e-02 1.67239591e-01
-1.06274235e+00 -8.47267091e-01 -1.07622361e+00 -4.55372423e-01
7.25866377e-01 -4.61361110e-01 9.70099390e-01 -3.99989605e-01
8.93350318e-02 1.82120502e-01 9.12067592e-02 -4.35967088e-01
-1.53486237e-01 -3.76764566e-01 1.00153856e-01 -3.04561436e-01
3.71842563e-01 -5.68173766e-01 -6.26028180e-01 5.37769258e-01
-5.06688833e-01 -5.26665747e-01 7.00978160e-01 8.97573471e-01
7.59569585e-01 -1.31905004e-01 9.96558011e-01 -8.21014524e-01
6.33870602e-01 -6.20103240e-01 -1.31375837e+00 2.81190574e-01
-6.72899067e-01 1.75819680e-01 2.49537051e-01 -1.94104180e-01
-7.91275442e-01 -4.37652171e-01 2.90934056e-01 2.75290102e-01
-1.89172644e-02 5.28460383e-01 -2.81634033e-01 1.86671525e-01
3.95288289e-01 -1.89155757e-01 -9.64021981e-02 -4.36568439e-01
2.44623274e-01 6.22969687e-01 8.83608639e-01 -8.23879123e-01
4.05552864e-01 4.07459944e-01 5.15465021e-01 -3.81791651e-01
-6.23981118e-01 -1.58915505e-01 -4.08584952e-01 -6.61969632e-02
5.73340654e-01 -7.85585344e-01 -1.18909407e+00 2.08373055e-01
-6.85939789e-01 -2.81237304e-01 -4.19922918e-01 8.43528450e-01
-7.56429136e-01 1.52650088e-01 -1.68802246e-01 -1.11791193e+00
-1.56949058e-01 -7.98663139e-01 1.06823385e+00 3.13803315e-01
-2.15900972e-01 -1.45540583e+00 5.66861987e-01 -1.36261493e-01
3.37275147e-01 7.69038737e-01 9.76890385e-01 -6.38200760e-01
-1.88384265e-01 -5.25634229e-01 -3.26697648e-01 3.17205824e-02
-2.52397776e-01 6.47644937e-01 -8.98030758e-01 -2.11302444e-01
-2.19064668e-01 4.13138241e-01 3.77140582e-01 1.23731184e+00
1.37130344e+00 -5.51552773e-01 -6.94644988e-01 3.46643269e-01
1.41169965e+00 -1.55495092e-01 7.43140101e-01 -9.47142243e-02
-2.05342118e-02 7.87370682e-01 5.58775127e-01 1.07985783e+00
1.58894062e-01 2.56751209e-01 5.99784590e-02 2.40686134e-01
2.84710616e-01 -2.44771302e-01 2.94229299e-01 6.10510349e-01
-4.88582812e-02 -1.43868104e-01 -6.62020087e-01 4.47948396e-01
-1.53832674e+00 -9.57855225e-01 -6.47031248e-01 2.99258614e+00
8.41334462e-01 1.50430985e-02 4.64177370e-01 -1.62649248e-02
1.26645982e+00 -3.08611006e-01 -2.85650492e-01 -3.40522915e-01
-4.90411937e-01 1.74350113e-01 1.02942860e+00 6.48286641e-01
-8.48441064e-01 1.62809521e-01 8.59996700e+00 9.16883349e-01
-7.11486697e-01 -5.13410158e-02 9.04535949e-01 9.64287296e-02
-6.50157690e-01 4.01507884e-01 -9.25695598e-01 6.28269196e-01
1.61975336e+00 -6.28589988e-01 -5.18590152e-01 6.81220949e-01
4.99093473e-01 -5.51563621e-01 -1.05772960e+00 4.81600791e-01
-4.12539542e-01 -1.13412941e+00 -9.29305926e-02 3.53426248e-01
1.03871524e+00 -3.77330214e-01 4.21052456e-01 2.51556724e-01
5.71786702e-01 -1.04312193e+00 6.26200736e-01 9.93817985e-01
8.86445343e-01 -8.17131102e-01 1.04115283e+00 -3.41545977e-02
-6.32339001e-01 6.43570572e-02 -5.72091520e-01 -2.70275950e-01
-2.45526582e-02 1.40018535e+00 -3.77503008e-01 5.70714474e-01
5.14988124e-01 2.77426124e-01 -5.55692792e-01 1.41040897e+00
-1.13290034e-01 7.55834341e-01 -4.57724422e-01 -5.62737882e-01
-2.45003700e-01 -3.41509312e-01 4.21126366e-01 1.12338853e+00
8.84210885e-01 -1.14404730e-01 -3.24858367e-01 1.02592540e+00
3.19690973e-01 2.76615411e-01 -2.62507081e-01 1.18774362e-01
7.00862348e-01 1.05544782e+00 -5.98588645e-01 -4.85897630e-01
-2.71076947e-01 3.14555271e-03 -4.02092516e-01 3.83576006e-01
-9.30534661e-01 -7.31947005e-01 4.49454516e-01 2.07142696e-01
3.52351181e-02 1.83643222e-01 -1.02286005e+00 -7.43291378e-01
-4.03011471e-01 -3.09753656e-01 6.47616506e-01 -7.56992161e-01
-1.61137593e+00 -1.10757433e-01 7.04218447e-01 -1.22507668e+00
-2.39666387e-01 -7.12350190e-01 -5.87874532e-01 1.12645304e+00
-7.10109174e-01 -6.79880619e-01 -2.51932353e-01 4.09222394e-01
-4.58977848e-01 4.27938662e-02 7.06715524e-01 7.28331730e-02
-5.61990499e-01 9.82978523e-01 7.89167285e-01 -3.77034873e-01
7.48069406e-01 -1.31584859e+00 6.63506240e-02 5.05242229e-01
-7.44230509e-01 6.13530993e-01 1.33075833e+00 -9.12600279e-01
-9.77706552e-01 -5.00808775e-01 1.02358532e+00 -5.66256642e-01
1.03900349e+00 -2.22331211e-01 -4.67658818e-01 4.55456465e-01
-2.63039153e-02 -1.76590741e-01 1.00444555e+00 3.52082402e-01
-1.10380009e-01 -2.95902371e-01 -1.45620322e+00 7.25352645e-01
6.28190279e-01 1.33945465e-01 -4.88215715e-01 3.08863878e-01
4.55818415e-01 8.98706615e-02 -1.30071628e+00 5.98123014e-01
8.28029513e-01 -1.26264417e+00 8.94396782e-01 -4.42233413e-01
9.59121585e-02 -1.73974931e-01 -2.95777410e-01 -1.13563097e+00
-2.77862903e-02 -7.70542622e-01 3.32549781e-01 1.29851973e+00
4.45318192e-01 -1.17703021e+00 4.60693777e-01 6.80247486e-01
1.36081800e-01 -6.42137229e-01 -1.11632740e+00 -8.15060675e-01
5.40433049e-01 -2.82260418e-01 8.93775463e-01 5.22370875e-01
3.84450197e-01 -2.73717254e-01 -9.71135776e-03 3.25393051e-01
1.19973373e+00 -5.47807850e-02 7.65751839e-01 -1.57899833e+00
-9.27905738e-02 -6.21463120e-01 -8.17105651e-01 -3.66233289e-01
-1.91851631e-01 -4.19140577e-01 -4.95006889e-02 -1.29643369e+00
6.55172348e-01 -3.10707301e-01 -2.79971510e-01 1.71648264e-01
-2.79148221e-01 -2.57035140e-02 -5.36076784e-01 -1.08409852e-01
-1.64847851e-01 4.08508152e-01 8.67230117e-01 2.04880044e-01
-4.08163480e-02 2.79742092e-01 -1.06200826e+00 5.75108290e-01
6.69308007e-01 -4.24197435e-01 -1.46957293e-01 5.46723485e-01
3.89047593e-01 3.82294625e-01 6.56729698e-01 -1.05300617e+00
3.53066921e-02 -3.77292126e-01 5.84569752e-01 -1.10716426e+00
-1.78438604e-01 -3.59802306e-01 5.83878577e-01 6.94514453e-01
-4.68182534e-01 2.80638725e-01 3.08123827e-01 4.95073020e-01
-1.26847699e-01 -1.08796857e-01 7.22626269e-01 4.99325663e-01
1.33315340e-01 -9.11714658e-02 -4.28951085e-01 1.06730290e-01
1.19768834e+00 -2.97749648e-04 -7.23168135e-01 -5.48472703e-01
-6.10255241e-01 3.36379796e-01 2.65058905e-01 -3.99166644e-01
3.12143147e-01 -1.41336834e+00 -1.00962794e+00 -1.58951599e-02
-5.11756912e-03 -5.63785195e-01 5.10422409e-01 1.30034697e+00
-5.04648566e-01 6.28417909e-01 -3.15831929e-01 -6.21157169e-01
-8.12136948e-01 5.79769492e-01 7.88270459e-02 -1.59280404e-01
7.41999149e-02 6.04514539e-01 3.85172993e-01 -3.63762408e-01
1.42500624e-01 -6.10515475e-01 3.78083773e-02 1.24135934e-01
5.92376828e-01 9.22448456e-01 -3.27922374e-01 -1.69899553e-01
-4.74713564e-01 6.89222813e-01 4.59192485e-01 -4.68574494e-01
1.14506352e+00 -4.15654272e-01 -3.85119498e-01 8.49968135e-01
9.76286411e-01 4.28839713e-01 -1.14461505e+00 3.42490643e-01
9.75419357e-02 -3.89223099e-01 -4.07113642e-01 -9.28304851e-01
-5.66485226e-01 6.91931188e-01 4.48045433e-01 3.61175328e-01
9.03544784e-01 2.62525622e-02 -7.74595216e-02 -3.33002731e-02
2.81117380e-01 -8.40114057e-01 -4.80853021e-01 1.30639508e-01
9.30600643e-01 -7.66597211e-01 3.65864217e-01 -3.84335369e-01
-4.07936633e-01 7.44674027e-01 1.07914237e-02 -1.91055223e-01
8.49263251e-01 4.23550010e-01 -2.41919339e-01 1.13340832e-01
-5.65353870e-01 1.83567867e-01 2.73252875e-01 7.99116135e-01
5.82171798e-01 4.33209777e-01 -9.70991552e-01 9.30370748e-01
-5.04896641e-01 4.01684791e-02 6.08270884e-01 4.40962344e-01
-5.39625049e-01 -8.48793685e-01 -7.90483475e-01 7.17341781e-01
-5.15138924e-01 -2.70064566e-02 -1.26630381e-01 1.03407240e+00
-2.91391581e-01 9.59700048e-01 3.09514940e-01 -2.84140632e-02
2.98879385e-01 -4.68572266e-02 3.28617156e-01 -3.66963334e-02
-3.68246406e-01 4.80878465e-02 8.76693353e-02 -2.05163315e-01
-3.41025144e-01 -8.95728052e-01 -8.44150543e-01 -9.81887162e-01
-2.43105486e-01 7.16980517e-01 6.13433421e-01 4.35341895e-01
4.41173390e-02 2.58830935e-01 4.71554756e-01 -6.63519502e-01
-9.71390843e-01 -9.11376238e-01 -1.04681528e+00 -1.46807462e-01
-5.01299836e-02 -8.14117551e-01 -8.99044454e-01 -4.03992444e-01] | [7.580937385559082, 4.595036029815674] |
35ce3820-8fb3-4f60-a106-98342234a547 | directed-acyclic-graph-neural-networks-1 | 2101.07965 | null | https://arxiv.org/abs/2101.07965v3 | https://arxiv.org/pdf/2101.07965v3.pdf | Directed Acyclic Graph Neural Networks | Graph-structured data ubiquitously appears in science and engineering. Graph neural networks (GNNs) are designed to exploit the relational inductive bias exhibited in graphs; they have been shown to outperform other forms of neural networks in scenarios where structure information supplements node features. The most common GNN architecture aggregates information from neighborhoods based on message passing. Its generality has made it broadly applicable. In this paper, we focus on a special, yet widely used, type of graphs -- DAGs -- and inject a stronger inductive bias -- partial ordering -- into the neural network design. We propose the \emph{directed acyclic graph neural network}, DAGNN, an architecture that processes information according to the flow defined by the partial order. DAGNN can be considered a framework that entails earlier works as special cases (e.g., models for trees and models updating node representations recurrently), but we identify several crucial components that prior architectures lack. We perform comprehensive experiments, including ablation studies, on representative DAG datasets (i.e., source code, neural architectures, and probabilistic graphical models) and demonstrate the superiority of DAGNN over simpler DAG architectures as well as general graph architectures. | ['Jie Chen', 'Veronika Thost'] | 2021-01-20 | directed-acyclic-graph-neural-networks | https://openreview.net/forum?id=JbuYF437WB6 | https://openreview.net/pdf?id=JbuYF437WB6 | iclr-2021-1 | ['graph-property-prediction'] | ['graphs'] | [ 1.18436471e-01 6.49251938e-01 -2.89945722e-01 -4.46270376e-01
2.84331024e-01 -4.39224124e-01 7.15345204e-01 3.37808430e-01
-6.41667917e-02 4.68809426e-01 3.68499398e-01 -6.87250555e-01
-5.47011614e-01 -1.25772381e+00 -8.22790563e-01 -5.12705803e-01
-6.35344684e-01 4.99331862e-01 1.67256340e-01 -2.43592367e-01
2.99593154e-02 6.28133833e-01 -1.39153004e+00 7.49898925e-02
5.21721780e-01 5.62636256e-01 9.49301496e-02 4.32276130e-01
-4.73143816e-01 1.42780364e+00 -3.25904965e-01 -4.99859929e-01
-1.20918021e-01 -2.01538265e-01 -6.93422318e-01 -5.26533008e-01
4.42766458e-01 -8.48331228e-02 -1.02980435e+00 9.98185277e-01
2.85184056e-01 6.78754896e-02 7.15496361e-01 -1.36953676e+00
-1.10764587e+00 1.51049149e+00 -3.59911025e-01 2.89691359e-01
9.32388008e-02 -1.25593081e-01 1.37865746e+00 -6.98236585e-01
8.08721185e-01 1.46840990e+00 1.10627174e+00 5.00783920e-01
-1.49158561e+00 -4.52657312e-01 6.37167215e-01 2.07371801e-01
-1.28586280e+00 -2.38207102e-01 1.05567431e+00 -3.24652970e-01
9.81875479e-01 2.21370067e-02 7.69907057e-01 1.47811878e+00
4.75321949e-01 9.02180016e-01 5.62951088e-01 -3.74293804e-01
3.49752963e-01 -4.65980679e-01 8.16149294e-01 9.50768769e-01
8.61001909e-01 2.93140322e-01 -7.89688885e-01 -2.39959434e-01
7.44184852e-01 1.07120223e-01 -1.35166466e-01 -7.25700319e-01
-7.79126108e-01 8.60108316e-01 1.00328839e+00 4.61539477e-01
-4.21011239e-01 7.94278979e-01 2.17949152e-01 2.86448061e-01
3.63903016e-01 3.00660521e-01 -6.66340962e-02 2.48886779e-01
-5.79824209e-01 1.33086428e-01 1.18198860e+00 1.06734765e+00
9.03553545e-01 4.80088890e-01 -2.63714194e-01 5.42441308e-01
6.17976367e-01 1.38496980e-01 1.89018533e-01 -4.98702943e-01
1.92929149e-01 8.07907939e-01 -7.92891562e-01 -1.44155526e+00
-7.81251848e-01 -8.12512577e-01 -1.27219033e+00 -3.14802825e-01
1.47648990e-01 -4.06194441e-02 -1.19700491e+00 2.07698536e+00
-7.16223493e-02 7.26304054e-02 -5.57621904e-02 3.81339669e-01
1.25570333e+00 3.82250309e-01 2.13413537e-01 2.70613402e-01
9.03019369e-01 -5.07770360e-01 -7.31080234e-01 -4.35478777e-01
6.87178671e-01 1.18198559e-01 7.61832118e-01 1.95151076e-01
-9.03616548e-01 -2.31304571e-01 -1.07317138e+00 -2.44425610e-02
-7.16551840e-01 -3.67253125e-01 1.11670125e+00 8.26537430e-01
-1.99454486e+00 8.59858871e-01 -8.45466673e-01 -6.35028780e-01
4.19172138e-01 4.43140894e-01 -2.21293688e-01 -7.61278719e-02
-1.31696522e+00 6.83346033e-01 6.43956959e-01 5.18628299e-01
-9.82576609e-01 -4.17388469e-01 -1.10886753e+00 2.36637637e-01
3.60438198e-01 -8.84887934e-01 9.88251626e-01 -6.44409418e-01
-1.14637053e+00 5.37945628e-01 -5.99160045e-02 -8.00872266e-01
-7.90145099e-02 2.73512360e-02 -2.70220876e-01 -2.33435884e-01
-2.53230959e-01 6.23170495e-01 5.55717885e-01 -1.24201930e+00
-1.12943925e-01 -5.23669362e-01 2.13014051e-01 -1.05073810e-01
-5.78799963e-01 -3.23628128e-01 -3.25581878e-01 -7.34133780e-01
4.13842231e-01 -9.32404339e-01 -4.32317376e-01 -4.04133558e-01
-8.55979681e-01 -5.28754413e-01 7.46186554e-01 -1.59230024e-01
1.63158703e+00 -1.96405566e+00 5.25423028e-02 7.06563771e-01
9.89596963e-01 -1.21777654e-01 -1.82984382e-01 7.68340945e-01
-3.08285713e-01 3.39295775e-01 -2.55257636e-01 -8.22196156e-02
1.41278476e-01 5.18296063e-01 -1.89506158e-01 2.73077697e-01
2.93164998e-01 1.26728380e+00 -1.02752125e+00 -3.13594431e-01
-1.72564194e-01 4.11495447e-01 -4.11891609e-01 -1.26203999e-01
-3.61127466e-01 -2.74691015e-01 -4.84197229e-01 6.24565303e-01
3.98731172e-01 -7.54574835e-01 5.89461088e-01 -1.75812736e-01
2.97487676e-01 3.42676133e-01 -1.03823256e+00 1.53363633e+00
-7.00489357e-02 7.31877029e-01 -1.69386417e-01 -1.03035378e+00
1.07877469e+00 -1.11655705e-01 1.71841770e-01 -3.48340660e-01
-5.08461101e-03 -2.14631915e-01 3.80716860e-01 -1.22247972e-01
5.34598172e-01 3.74373049e-01 -6.53263601e-03 4.94427562e-01
3.40112597e-01 2.26342946e-01 4.44851160e-01 8.51941347e-01
1.65440440e+00 2.32373569e-02 3.74953210e-01 -5.77517092e-01
-4.13057879e-02 -3.94357771e-01 2.79669821e-01 1.32321382e+00
6.34644106e-02 4.11969543e-01 9.04821932e-01 -5.97192228e-01
-4.82178360e-01 -1.23194456e+00 1.92363918e-01 1.34630477e+00
-1.16855592e-01 -9.48233485e-01 -4.57797319e-01 -6.08626902e-01
1.21481903e-01 7.04261661e-01 -1.03460586e+00 -5.84810257e-01
-6.41964078e-01 -9.24439728e-01 8.27500880e-01 9.13576126e-01
1.83858112e-01 -1.12286484e+00 -1.10696070e-01 1.85195416e-01
3.49226564e-01 -8.32998753e-01 2.65643522e-02 6.33233786e-01
-1.11628211e+00 -8.66504848e-01 -1.59722880e-01 -8.80114019e-01
5.97298861e-01 2.12605689e-02 1.75523400e+00 1.88656852e-01
9.17319432e-02 6.10836804e-01 -1.56257540e-01 -4.52405214e-01
-3.71825814e-01 5.45522630e-01 -6.59606233e-02 4.49699387e-02
2.41304815e-01 -1.16622627e+00 -2.51705855e-01 -5.48561215e-02
-1.05282807e+00 5.85376099e-02 8.74500811e-01 7.10583508e-01
2.14530975e-01 -1.20508872e-01 5.26569247e-01 -1.57395613e+00
9.09508467e-01 -7.33115971e-01 -4.21609640e-01 2.02337801e-01
-8.84233534e-01 5.92953801e-01 5.20407975e-01 -1.85054570e-01
-7.06357300e-01 -1.23453550e-01 8.47352669e-02 -2.74925679e-01
3.94087434e-02 1.23138356e+00 -1.24991089e-01 -1.11639105e-01
9.12429631e-01 1.52787983e-01 -5.18780537e-02 -2.54274547e-01
6.03853643e-01 -5.06563708e-02 4.87063736e-01 -7.78640449e-01
7.73777783e-01 3.21648389e-01 4.92227912e-01 -7.48987675e-01
-4.53353554e-01 3.15475687e-02 -3.82492900e-01 -3.28015476e-01
4.18862104e-01 -5.97846508e-01 -7.23884344e-01 3.98135334e-01
-1.29751027e+00 -4.06911135e-01 -3.03408772e-01 1.57089397e-01
-1.04846589e-01 2.42075115e-01 -8.68141949e-01 -8.39579761e-01
-1.64752886e-01 -7.38660932e-01 5.39444745e-01 7.49100149e-02
-2.69091308e-01 -1.42745864e+00 1.90338399e-02 -7.61332870e-01
6.83800042e-01 3.77410710e-01 1.32839084e+00 -1.02823138e+00
-6.81490004e-01 4.58278321e-02 -3.68527651e-01 -3.52597632e-03
-3.53786387e-02 1.85299799e-01 -1.01113379e+00 -1.78009063e-01
-3.92310113e-01 -1.44585982e-01 1.23020518e+00 5.23629844e-01
1.35665143e+00 -3.60748559e-01 -6.99063361e-01 4.31325823e-01
1.28453803e+00 -4.95954268e-02 4.67648089e-01 -5.82626686e-02
1.18250346e+00 6.55117869e-01 -5.63874543e-01 1.45773470e-01
7.74024546e-01 6.96149841e-02 9.10685539e-01 1.97353922e-02
-2.31913030e-01 -6.00688100e-01 3.14731508e-01 1.16041625e+00
-7.76794404e-02 -6.69353664e-01 -1.19654298e+00 4.35698539e-01
-2.02442503e+00 -6.87425792e-01 -3.19549024e-01 1.75364244e+00
5.98120987e-01 3.08534771e-01 -9.96615440e-02 -2.96516661e-02
8.12340379e-01 5.74600756e-01 -5.94235480e-01 -1.19150490e-01
-3.01563650e-01 3.25722694e-01 4.44796234e-01 3.49365950e-01
-9.30345714e-01 8.15478206e-01 6.98546219e+00 5.77221453e-01
-8.43284428e-01 -2.23696694e-01 5.90148091e-01 1.68785989e-01
-7.70128429e-01 6.74288347e-02 -6.65688276e-01 9.96000022e-02
1.34225881e+00 -1.70130745e-01 3.75843763e-01 8.17730367e-01
-3.48976046e-01 2.93318033e-01 -1.61327159e+00 8.22819650e-01
4.06832583e-02 -1.69050348e+00 5.71532667e-01 1.33767560e-01
4.34711605e-01 3.82809222e-01 2.45327409e-02 5.03884912e-01
1.24309921e+00 -1.18469822e+00 7.52817392e-01 6.41846061e-01
4.24440891e-01 -3.71261656e-01 5.41414917e-01 3.76132019e-02
-1.40290093e+00 -1.30623817e-01 -2.54179060e-01 -6.22473955e-02
-1.53661460e-01 8.43960345e-01 -7.56709933e-01 9.45777178e-01
7.98346221e-01 1.19054163e+00 -9.82952952e-01 7.83609331e-01
-2.93857872e-01 8.74264538e-01 -4.64960635e-01 -4.85242814e-01
3.39004040e-01 -5.96849732e-02 4.85120177e-01 1.31963766e+00
1.22772016e-01 -2.61738092e-01 1.03050657e-02 1.20535541e+00
-2.55167186e-01 -3.10400307e-01 -1.31348860e+00 -1.91632628e-01
5.59813201e-01 1.17793381e+00 -8.20109725e-01 -3.73993181e-02
-2.02911735e-01 2.44342327e-01 8.85703266e-01 6.74864531e-01
-5.50658166e-01 -4.27337915e-01 2.47574806e-01 1.61966041e-01
1.93524599e-01 -3.05560112e-01 -2.69763082e-01 -7.20603824e-01
-1.21488243e-01 -5.65644026e-01 7.95916975e-01 -6.32254720e-01
-1.50777149e+00 8.21187317e-01 1.13659404e-01 -6.02653444e-01
-5.07092535e-01 -7.80606031e-01 -5.82332611e-01 4.86865789e-01
-1.21679926e+00 -1.16970837e+00 -2.83320010e-01 6.23804331e-01
-2.06793934e-01 -2.07310938e-03 6.64327145e-01 1.83157716e-02
-5.96622646e-01 5.51568151e-01 -7.03788921e-02 3.74289185e-01
8.20839331e-02 -1.53733182e+00 8.78456175e-01 9.81761873e-01
6.16687655e-01 1.08069170e+00 5.78920245e-01 -7.45825708e-01
-1.90749240e+00 -1.30799854e+00 6.61476851e-01 -5.22544801e-01
9.07722890e-01 -7.34723747e-01 -1.08533192e+00 1.19465601e+00
2.22448140e-01 1.75294310e-01 4.68474597e-01 6.70540035e-01
-5.48954785e-01 -2.67585456e-01 -6.54355824e-01 8.69462788e-01
1.88871908e+00 -5.81618965e-01 -3.11090261e-01 1.20576672e-01
9.74053979e-01 -2.95616031e-01 -7.44273067e-01 5.41112661e-01
3.78655851e-01 -1.04002464e+00 8.47515285e-01 -8.33954394e-01
1.74137980e-01 -2.23697230e-01 -1.12810254e-01 -1.40434337e+00
-7.51656234e-01 -9.61154938e-01 -6.63971305e-01 1.13225949e+00
5.82270563e-01 -1.04190671e+00 1.00510979e+00 2.56068170e-01
-2.86330551e-01 -7.44988501e-01 -7.72315323e-01 -6.30847096e-01
-6.04799353e-02 -6.72833443e-01 6.34233534e-01 8.68114173e-01
-1.51733860e-01 6.53721392e-01 7.71338344e-02 1.67822003e-01
6.16581440e-01 -1.55048668e-01 5.92105150e-01 -1.81235456e+00
-9.47143063e-02 -7.26417899e-01 -5.61765015e-01 -1.09108579e+00
4.59927231e-01 -1.38109708e+00 -3.14022265e-02 -1.83900595e+00
-1.09568499e-02 -3.79459202e-01 -5.66430271e-01 6.85676098e-01
8.07272829e-03 -2.45144829e-01 -7.45853931e-02 -1.53200096e-03
-6.37311935e-01 5.46637356e-01 8.19557607e-01 -2.47910410e-01
-3.71636115e-02 -1.81771711e-01 -9.52574134e-01 6.14607513e-01
6.95624828e-01 -4.29785013e-01 -8.16147149e-01 -6.23042643e-01
7.61620402e-01 -2.67561913e-01 5.62616944e-01 -9.64303136e-01
6.81211412e-01 2.79771209e-01 3.09001416e-01 -6.20106101e-01
-2.11208574e-02 -6.70718610e-01 3.28059554e-01 4.45912510e-01
-5.48694730e-01 5.57952166e-01 1.95661575e-01 1.03833818e+00
-5.12881503e-02 3.27843428e-02 8.68229419e-02 -1.97096035e-01
-6.55531764e-01 5.64029336e-01 -5.43332279e-01 6.65508956e-02
4.38674450e-01 -3.25696349e-01 -5.87259591e-01 -6.16586208e-01
-8.57057333e-01 1.88490734e-01 -2.47195456e-02 5.63080609e-01
6.99039340e-01 -1.36053753e+00 -5.85325897e-01 1.98837832e-01
1.97390854e-01 1.14181332e-01 -9.25091356e-02 6.25356376e-01
-2.87569642e-01 2.72626549e-01 9.83144194e-02 -8.21619630e-01
-5.76607466e-01 5.43483138e-01 3.47535580e-01 -5.55561244e-01
-7.47281015e-01 7.25441039e-01 4.04693067e-01 -7.02516496e-01
3.85128349e-01 -7.97477007e-01 -2.71608561e-01 -1.27981573e-01
3.52232275e-03 2.16118053e-01 2.33873531e-01 -8.97152871e-02
-3.78159225e-01 1.01903314e-02 -1.23444207e-01 1.91705018e-01
1.35329366e+00 1.97768793e-01 -5.66646934e-01 7.40904152e-01
6.90708041e-01 -3.91704947e-01 -7.99053550e-01 -5.38385510e-01
5.72252572e-01 2.18874469e-01 -1.17466912e-01 -5.27083099e-01
-1.13521004e+00 7.54620433e-01 1.48765609e-01 8.98225784e-01
7.97305107e-01 2.25263298e-01 6.84906915e-02 7.92974114e-01
3.50551575e-01 -5.39672911e-01 -1.19350627e-01 8.27338159e-01
8.03541243e-01 -6.21275246e-01 -9.55436230e-02 -3.90701711e-01
7.20713437e-02 1.25807309e+00 7.81089187e-01 -1.94933012e-01
1.09625328e+00 3.99889499e-01 -4.34942871e-01 -6.35156333e-01
-9.54448521e-01 -1.09202109e-01 1.49158105e-01 8.07307005e-01
5.02871215e-01 5.33555374e-02 1.50343090e-01 5.50444782e-01
-2.93398738e-01 -1.26202047e-01 5.38173378e-01 9.18770134e-01
-1.28373027e-01 -8.17386150e-01 8.31657499e-02 8.85191023e-01
-4.72270995e-02 -4.92669642e-01 -7.25586057e-01 1.00915504e+00
-3.75929058e-01 8.05480599e-01 1.53150290e-01 -6.38844728e-01
2.40775526e-01 -8.04347768e-02 4.09915388e-01 -6.44393682e-01
-7.11863101e-01 -6.50068223e-01 3.67618054e-01 -7.15590477e-01
-3.37479889e-01 -3.20172817e-01 -1.02441239e+00 -4.41616237e-01
-1.96750119e-01 1.31829113e-01 3.51180971e-01 6.13976181e-01
6.80860817e-01 9.42831755e-01 1.65496811e-01 -7.42133975e-01
-4.01692957e-01 -1.02585649e+00 -6.21117353e-01 1.63712669e-02
2.81509578e-01 -5.89409292e-01 -2.25274280e-01 -3.81485820e-01] | [6.932179927825928, 6.308597564697266] |
f6fccff5-40fa-4b41-bbfa-5960f5d7901f | depression-recognition-using-remote | 2206.04399 | null | https://arxiv.org/abs/2206.04399v1 | https://arxiv.org/pdf/2206.04399v1.pdf | Depression Recognition using Remote Photoplethysmography from Facial Videos | Depression is a mental illness that may be harmful to an individual's health. The detection of mental health disorders in the early stages and a precise diagnosis are critical to avoid social, physiological, or psychological side effects. This work analyzes physiological signals to observe if different depressive states have a noticeable impact on the blood volume pulse (BVP) and the heart rate variability (HRV) response. Although typically, HRV features are calculated from biosignals obtained with contact-based sensors such as wearables, we propose instead a novel scheme that directly extracts them from facial videos, just based on visual information, removing the need for any contact-based device. Our solution is based on a pipeline that is able to extract complete remote photoplethysmography signals (rPPG) in a fully unsupervised manner. We use these rPPG signals to calculate over 60 statistical, geometrical, and physiological features that are further used to train several machine learning regressors to recognize different levels of depression. Experiments on two benchmark datasets indicate that this approach offers comparable results to other audiovisual modalities based on voice or facial expression, potentially complementing them. In addition, the results achieved for the proposed method show promising and solid performance that outperforms hand-engineered methods and is comparable to deep learning-based approaches. | ['Miguel Bordallo López', 'Manuel Lage Cañellas', 'Constantino Álvarez Casado'] | 2022-06-09 | null | null | null | null | ['heart-rate-variability'] | ['medical'] | [ 2.8453562e-01 7.2450370e-02 1.1478646e-01 -6.0562348e-01
-2.8306645e-01 -1.5316191e-01 3.2570732e-01 2.8766650e-01
-3.8849583e-01 6.4935911e-01 1.7956330e-01 1.2335186e-01
3.1624809e-02 -7.1428645e-01 -8.3891407e-02 -8.3296508e-01
-1.7679307e-01 -1.2724850e-01 -3.5252061e-01 -2.7005452e-01
8.0575839e-02 7.2810745e-01 -1.7034523e+00 1.5846661e-01
7.1527034e-01 1.2515630e+00 -5.8896506e-01 4.9637681e-01
1.8068162e-01 3.5876054e-01 -6.9867694e-01 -1.8303454e-02
2.3609448e-02 -5.9085482e-01 -1.8112391e-01 -1.9677448e-01
3.0075103e-01 -3.5807365e-01 2.5675496e-02 5.3887320e-01
1.0766302e+00 1.5647834e-02 3.8350824e-01 -1.0039089e+00
-6.6706657e-02 1.3122408e-01 -4.5243013e-01 1.9345157e-01
7.2303385e-01 3.4621418e-01 2.8988510e-01 -6.7535746e-01
4.3025902e-01 1.0701361e+00 9.4246876e-01 6.4003009e-01
-1.5759795e+00 -6.0575354e-01 -3.9021710e-01 2.3482066e-01
-1.2580733e+00 -8.6089957e-01 9.3008220e-01 -5.1892745e-01
8.8924342e-01 2.8971779e-01 9.9621624e-01 1.2911627e+00
2.4112338e-01 -1.6003290e-01 1.3405626e+00 -1.7749512e-01
2.6736593e-01 3.6343786e-01 -4.4821776e-02 5.5881941e-01
1.3921651e-01 -6.0145296e-02 -5.7733583e-01 -2.4200197e-01
3.6467755e-01 -1.0155924e-02 -5.7814509e-01 -1.0741541e-01
-8.2748294e-01 4.8937100e-01 4.6697002e-02 6.4098454e-01
-8.3692515e-01 -6.2548958e-02 5.9663182e-01 4.1073534e-01
5.5010748e-01 1.4210263e-01 -2.8462851e-01 -2.2095281e-01
-1.0058630e+00 -2.0985720e-01 8.6766911e-01 -2.4935836e-01
6.0160846e-01 1.6760148e-01 -4.7648591e-01 7.9016250e-01
2.0453730e-01 6.2585896e-01 5.9427631e-01 -7.5188309e-01
-1.0684861e-01 7.1744525e-01 -1.1368151e-01 -1.4936237e+00
-9.5146549e-01 -1.0766804e-01 -9.2072517e-01 1.9862279e-01
2.4295130e-01 -4.1800678e-01 -4.3025187e-01 1.5462404e+00
4.8427939e-01 2.8590220e-01 -7.3682800e-02 1.0068140e+00
1.2107106e+00 2.8398722e-01 1.2237887e-01 -7.0842743e-01
1.3323736e+00 5.5302605e-02 -8.8034952e-01 9.3913086e-02
3.4975725e-01 -2.5057128e-01 8.3580321e-01 5.6377208e-01
-8.1456625e-01 -4.5477360e-01 -9.7910583e-01 9.0136528e-02
-3.4165120e-01 2.4920717e-01 1.9996308e-01 1.2331688e+00
-1.0751121e+00 1.1304977e+00 -7.8349125e-01 -5.2567863e-01
3.6367130e-01 3.9961845e-01 -4.9306962e-01 2.8561622e-01
-1.1444082e+00 8.6035925e-01 -2.2657931e-01 5.0298679e-01
-4.2326611e-01 -8.4678823e-01 -9.1277170e-01 6.2238626e-02
-1.5386906e-01 -4.1588834e-01 3.2230857e-01 -9.9003363e-01
-2.0220244e+00 1.0153310e+00 -1.3341993e-01 -3.0237544e-01
4.2183149e-01 -3.8239285e-02 -4.9592435e-01 6.1138314e-01
-5.1340485e-01 3.0795607e-01 1.1239042e+00 -7.3018438e-01
3.5531369e-01 -6.1343807e-01 -4.0878296e-01 -6.0165942e-02
-6.9186687e-01 1.4263982e-01 1.6196412e-01 -2.7108595e-02
-1.9067425e-01 -6.8572199e-01 9.7486168e-02 9.0837359e-02
-2.7406722e-01 7.5852178e-02 6.3856882e-01 -9.0194333e-01
8.3731246e-01 -1.9415140e+00 -6.9133341e-02 3.9725739e-01
2.5205356e-01 6.6238475e-01 -6.2076554e-02 3.7150952e-01
-2.3244020e-01 1.6721675e-02 -7.6709040e-02 -2.4977186e-01
-1.4099666e-01 -1.7929883e-01 8.5176386e-02 9.3990839e-01
3.2837898e-01 7.3975086e-01 -4.8678100e-01 -2.2774130e-01
5.8374000e-01 1.1879067e+00 -2.1186252e-01 2.9633957e-01
3.9409584e-01 7.8233689e-01 2.2600908e-02 5.0669897e-01
6.5019411e-01 3.3437982e-01 3.5702795e-01 -3.9749670e-01
3.0953323e-02 1.7057616e-01 -8.4017515e-01 1.2446886e+00
-4.8777980e-01 7.8681433e-01 1.8860567e-01 -1.4361365e+00
1.5085510e+00 5.4374349e-01 8.2754707e-01 -9.0437686e-01
4.8236689e-01 -1.0157334e-02 -5.8558654e-02 -9.9273825e-01
-2.7228063e-01 -3.6162770e-01 2.3462462e-01 2.1586779e-01
2.4917940e-02 1.6049087e-01 -1.9198725e-01 -3.9336443e-01
1.0172558e+00 1.5898646e-01 3.2345212e-01 -1.9063793e-01
8.6761481e-01 -6.5318191e-01 5.5385798e-01 2.2319396e-01
-4.1283059e-01 5.0627315e-01 8.0556262e-01 -3.1385770e-01
-4.3138328e-01 -6.0865414e-01 -2.3757887e-01 5.9702790e-01
-3.0792102e-01 -2.2542834e-01 -7.5193858e-01 -6.4156927e-02
7.8363359e-02 3.5982725e-01 -7.7347165e-01 -4.2830014e-01
-2.6744020e-01 -8.7522954e-01 6.4681625e-01 3.3924425e-01
2.3484176e-01 -1.1679742e+00 -1.0787568e+00 1.2278207e-01
-1.8562093e-01 -1.1157340e+00 3.3299983e-01 -1.2671939e-01
-1.0243104e+00 -9.6596080e-01 -6.7506385e-01 -6.5116987e-02
2.4687912e-01 -2.1780095e-01 7.5892889e-01 2.1175977e-02
-7.9952705e-01 7.2406524e-01 -1.2625512e-01 -5.1483727e-01
-2.2771297e-01 -1.6549805e-01 2.0202485e-01 5.9088022e-01
5.1557380e-01 -1.0491867e+00 -9.2699015e-01 -4.9067672e-02
-4.2703646e-01 -5.7119429e-01 3.1088734e-01 2.2759332e-01
3.8476443e-01 -4.8691303e-01 8.1505978e-01 -5.8096915e-01
7.7721918e-01 -4.2521065e-01 -2.6945657e-01 -1.8605544e-01
-6.4242887e-01 -3.4118980e-01 6.6143674e-01 -3.5699233e-01
-8.3583212e-01 2.0635553e-02 -1.9307533e-01 -4.3445033e-01
-5.2164471e-01 3.7291193e-01 2.2612348e-02 -3.6876950e-02
8.5312933e-01 1.4776951e-01 3.6418712e-01 -4.3253425e-01
-1.4952590e-01 7.7139616e-01 3.8932049e-01 -1.0443421e-01
4.2044353e-01 5.5852622e-01 3.2131931e-01 -1.5470175e+00
-3.2713798e-01 -4.2105773e-01 -6.5314770e-01 -5.7640302e-01
7.4961513e-01 -8.4999037e-01 -1.2730484e+00 4.4739589e-01
-8.0688381e-01 -1.5647027e-01 -1.0918178e-01 5.6303447e-01
-3.7799114e-01 3.4484413e-01 -4.2956340e-01 -1.1095920e+00
-7.8247392e-01 -4.6160853e-01 8.3425045e-01 3.5741255e-01
-3.9040616e-01 -9.8592603e-01 4.3947095e-01 3.8382471e-01
7.0568091e-01 9.9345207e-01 5.3291750e-01 -4.8608041e-01
3.2608488e-01 -3.6409533e-01 -4.2042132e-02 5.1654387e-01
2.3514776e-01 1.5407644e-01 -1.4501786e+00 -1.9779541e-01
2.7005970e-01 -4.0350205e-01 6.2274611e-01 4.8601067e-01
8.6975157e-01 -4.4384217e-03 -2.1907408e-04 5.3122818e-01
1.3665402e+00 -4.1294798e-02 1.0448589e+00 -1.9428626e-01
3.8699815e-01 8.8294798e-01 1.4212629e-01 7.7620381e-01
1.2095033e-01 5.6107211e-01 3.8022953e-01 -3.6744273e-01
8.1493735e-02 2.5655723e-01 7.1813607e-01 4.5068294e-01
-6.2567008e-01 2.8454262e-01 -6.4485466e-01 3.2674479e-01
-1.3583386e+00 -1.0873066e+00 -4.7677821e-01 2.3662210e+00
7.1991158e-01 -3.6152202e-01 3.5051033e-01 4.0329412e-01
6.6673183e-01 3.0314315e-02 -5.3323466e-01 -8.3803606e-01
8.8860638e-02 8.4345734e-01 -9.0998365e-04 9.6961334e-02
-8.5077065e-01 2.1380943e-01 6.1929994e+00 -2.0037569e-01
-1.8060224e+00 -1.4319823e-02 4.9848136e-01 -2.9677635e-01
7.5360298e-02 -5.1569748e-01 -3.2471746e-01 3.9938956e-01
1.5224934e+00 9.2496321e-02 3.4063751e-01 4.7110605e-01
9.1554260e-01 -1.9244766e-01 -8.9664459e-01 1.2097372e+00
2.2739875e-01 -7.6711023e-01 -6.5443796e-01 -2.2743521e-02
-2.1771977e-02 -2.6814824e-01 -9.8264515e-02 8.0282092e-02
-8.5115492e-01 -1.0661042e+00 -1.1026268e-01 1.0746843e+00
9.8290706e-01 -6.5907264e-01 7.5843203e-01 6.3834186e-03
-8.2584286e-01 5.4868050e-02 -2.0447221e-01 -2.2554660e-01
-1.7219278e-01 9.8778486e-01 -7.6826322e-01 3.0333248e-01
6.4025223e-01 7.3548138e-01 -4.1020149e-01 8.1765848e-01
-2.1257022e-01 7.2086537e-01 -3.1505406e-01 -1.6520839e-02
-4.8582244e-01 -1.9152889e-01 3.9804411e-01 1.2573688e+00
3.9850110e-01 2.2455220e-01 -4.0122515e-01 9.9712312e-01
1.7069514e-01 4.1567791e-01 -6.9879735e-01 -1.2212364e-01
-8.0629289e-02 1.7041743e+00 -5.9403050e-01 -1.3911310e-01
-3.5541350e-01 8.7911630e-01 -1.2110425e-01 1.9233556e-01
-7.3939931e-01 -6.4236552e-01 8.2328856e-01 3.4551549e-01
5.0585873e-02 1.2151424e-01 -8.5672230e-02 -1.1622030e+00
3.8699768e-02 -7.7537203e-01 2.0134676e-01 -5.9330797e-01
-9.5700812e-01 4.5560613e-01 -2.9988050e-01 -9.7207344e-01
-2.3662695e-01 -3.5005438e-01 -8.3854055e-01 1.0675130e+00
-1.5881073e+00 -5.4043752e-01 -8.0062306e-01 6.8455368e-01
-2.2796452e-01 3.0589920e-01 1.1270800e+00 3.8563997e-01
-8.6165684e-01 4.2142934e-01 -3.6452293e-01 2.8517239e-02
8.7811613e-01 -8.8893700e-01 -2.6453710e-01 5.0374496e-01
-4.0408754e-01 3.4845531e-01 5.8599710e-01 -2.0045644e-01
-1.4216058e+00 -7.5785995e-01 9.5129716e-01 -1.4878772e-01
1.6639818e-01 -3.2788438e-01 -9.4900000e-01 5.7711650e-02
1.5186563e-01 3.4739292e-01 9.4563001e-01 -5.5152003e-02
-5.0720364e-02 -6.4049107e-01 -1.5277126e+00 2.9690538e-02
6.3274467e-01 -5.3061533e-01 -4.5847982e-01 3.6802053e-02
-6.6099033e-02 -5.1920872e-02 -1.2951288e+00 4.5080802e-01
9.7108436e-01 -1.5182368e+00 7.6248860e-01 -3.5995108e-01
3.2569438e-01 1.0991156e-01 3.5585922e-01 -1.2902560e+00
1.3663056e-01 -6.5365189e-01 1.6075673e-02 1.3772033e+00
-7.9687253e-02 -9.7504842e-01 5.8480269e-01 8.7010419e-01
3.6898822e-01 -5.6357324e-01 -7.1404529e-01 -4.9247470e-01
-3.3287662e-01 -2.2075012e-01 2.5102040e-01 1.0778325e+00
4.0656006e-01 2.5918406e-01 -3.7284946e-01 -2.1281976e-02
3.5527354e-01 2.7929995e-02 6.9017154e-01 -1.6396554e+00
5.6141131e-03 -1.6831337e-01 -7.1469933e-01 1.1145939e-01
9.8032892e-02 -6.4037287e-01 -1.6161630e-01 -1.3822172e+00
-8.5384220e-02 7.1291640e-02 -5.6708902e-01 6.6520882e-01
1.7217448e-01 6.6949028e-01 -1.1342482e-01 -3.9892301e-01
1.1897546e-01 5.3063494e-01 7.8439343e-01 2.2734542e-01
-8.1680799e-01 -1.5739994e-01 -4.7476009e-01 6.0839254e-01
9.1464943e-01 -3.6191341e-01 -2.0108703e-01 4.1850483e-01
4.4941373e-02 3.2061487e-01 5.4969412e-01 -1.2286975e+00
-2.6543811e-01 2.4526235e-01 7.4003398e-01 -3.5350915e-02
5.7095653e-01 -7.0533276e-01 4.6155870e-02 7.3166931e-01
-1.2048848e-01 -4.1127118e-01 2.7329302e-01 1.2389384e-01
-4.1278034e-02 7.0204929e-02 9.3503547e-01 1.1201912e-01
-1.9624230e-02 -2.4501922e-02 -5.7778108e-01 -2.0515324e-01
8.1124729e-01 -3.0033359e-01 -1.4182070e-01 -5.5422580e-01
-1.0025594e+00 -2.6310766e-01 5.9338942e-02 2.0743793e-01
5.7949322e-01 -1.0238905e+00 -7.7145642e-01 3.9162305e-01
-6.6788733e-02 -7.8103155e-01 4.3864796e-01 1.6086876e+00
-2.4393970e-01 2.8658444e-01 -6.2397730e-01 -7.2817737e-01
-1.4532982e+00 2.5700080e-01 6.9238800e-01 1.7837638e-01
-7.3948485e-01 3.3765751e-01 -4.6583468e-01 -9.0382554e-02
6.7800015e-02 -2.5138533e-01 -7.8637898e-01 6.9858742e-01
7.0129573e-01 7.2276771e-01 4.4665006e-01 -6.0942411e-01
-5.6328928e-01 5.8117557e-01 6.0203087e-01 6.4707279e-02
1.6144359e+00 -1.7798853e-01 -2.8572044e-01 6.0831594e-01
1.2014048e+00 4.4498451e-02 -8.6517167e-01 1.5260816e-01
-3.2183027e-01 -2.4912411e-01 1.3727874e-01 -6.1996579e-01
-1.3308557e+00 1.2173564e+00 1.1701580e+00 3.3250624e-01
1.4980888e+00 -4.9035963e-01 5.8377850e-01 2.4440277e-01
-3.9717536e-02 -1.0410993e+00 6.3755274e-02 -2.8451890e-01
8.6856711e-01 -8.7342489e-01 4.1941531e-02 -2.3592113e-01
-4.4305161e-01 1.5115547e+00 1.9228429e-01 -1.9960704e-01
7.6508605e-01 1.3780668e-01 2.4227403e-01 -1.8936430e-01
-5.8783805e-01 -2.3624289e-01 2.4989542e-01 8.3000016e-01
7.0648789e-01 -1.4914224e-01 -6.1952394e-01 5.0790185e-01
3.0654082e-02 3.3447221e-01 6.1917561e-01 5.3063512e-01
-2.8191704e-01 -8.1886643e-01 -3.5629398e-01 4.9417347e-01
-6.5477598e-01 1.5461241e-01 -5.3204674e-01 4.1890991e-01
1.5339726e-01 1.0973005e+00 1.8170055e-02 -2.1500967e-01
5.5891418e-01 6.1908424e-01 6.3968778e-01 -4.8467159e-01
-7.5757521e-01 -1.4713015e-03 1.3822140e-01 -1.0516565e+00
-8.4652221e-01 -7.6835644e-01 -1.0318427e+00 -1.8260032e-01
4.7805164e-02 -4.1157961e-01 9.2811286e-01 8.4234178e-01
6.4522004e-01 3.9685953e-01 8.4175849e-01 -1.0127654e+00
-4.8250116e-02 -1.0050529e+00 -6.5895450e-01 4.6524805e-01
4.8157397e-01 -5.0896466e-01 -4.0151203e-01 -4.3411888e-02] | [13.77697467803955, 2.9236643314361572] |
d84cb436-8760-4c5d-808f-628bb2084a40 | a-high-resolution-chest-ct-scan-image-dataset | 2205.03408 | null | https://arxiv.org/abs/2205.03408v1 | https://arxiv.org/pdf/2205.03408v1.pdf | A High-Resolution Chest CT-Scan Image Dataset for COVID-19 Diagnosis and Differentiation | During the COVID-19 pandemic, computed tomography (CT) is a good way to diagnose COVID-19 patients. HRCT (High-Resolution Computed Tomography) is a form of computed tomography that uses advanced methods to improve image resolution. Publicly accessible COVID-19 CT image datasets are very difficult to come by due to privacy concerns, which impedes the study and development of AI-powered COVID-19 diagnostic algorithms based on CT images. To address this problem, we have introduced HRCTv1-COVID-19, a new COVID-19 high resolution chest CT Scan image dataset that includes not only COVID-19 cases of Ground Glass Opacity (GGO), Crazy Paving, and Air Space Consolidation, but also CT images of cases with negative COVID-19. The HRCTv1-COVID-19 dataset, which includes slice-level, and patient-level labels, has the potential to aid COVID-19 research, especially for diagnosis and differentiation using artificial intelligence algorithms, machine learning and deep learning methods. This dataset is accessible through web at: http://databiox.com and includes 181,106 chest HRCT images from 395 patients with four labels: GGO, Crazy Paving, Air Space Consolidation and Negative. Keywords- Dataset, COVID-19, CT-Scan, Computed Tomography, Medical Imaging, Chest Image. | ['Hamidreza Bolhasani', 'Bentolhoda Otroshi Shahreza', 'Mahsa Vali', 'Iraj Abedi'] | 2022-05-06 | null | null | null | null | ['covid-19-detection'] | ['medical'] | [-3.25164534e-02 -4.47252005e-01 -2.61826694e-01 1.29325673e-01
-6.75571620e-01 -5.19856453e-01 1.39431849e-01 4.05620635e-01
-4.05845970e-01 7.20744014e-01 2.37473488e-01 -8.98119092e-01
-3.71281862e-01 -8.77869070e-01 -3.28018308e-01 -7.10795462e-01
-1.25596300e-01 1.49829257e+00 6.84666112e-02 3.47759813e-01
-3.23569119e-01 6.08396053e-01 -5.39612591e-01 4.42411691e-01
3.70933443e-01 9.12775636e-01 4.42002565e-01 1.14548397e+00
4.07691568e-01 7.99813211e-01 -1.35837182e-01 -1.69090524e-01
2.24319562e-01 -6.04021370e-01 -8.15394998e-01 -3.59421730e-01
3.25123966e-02 -8.89117599e-01 -4.16299328e-02 3.88933748e-01
6.76749647e-01 -3.81788194e-01 1.11790013e+00 -9.02647436e-01
-3.04797620e-01 -9.70547646e-03 -7.28222907e-01 1.10650063e+00
2.40953490e-01 4.14931834e-01 5.97597361e-01 -8.82550657e-01
9.31374073e-01 8.26107204e-01 1.03058124e+00 4.07859057e-01
-4.99127567e-01 -6.77346706e-01 -5.94646454e-01 9.71412137e-02
-1.30910361e+00 8.58780682e-01 -2.35561237e-01 -1.01979423e+00
9.73768353e-01 5.66321075e-01 1.31065011e+00 1.16259408e+00
5.91145217e-01 2.89030045e-01 1.16699755e+00 -7.87409022e-03
-3.65390722e-03 -3.16339731e-01 1.17272943e-01 9.72788692e-01
5.40420651e-01 3.97748619e-01 6.47694767e-02 -5.25345325e-01
1.02665257e+00 6.23440981e-01 -3.81747633e-01 1.47685930e-01
-1.11496568e+00 1.19522643e+00 5.25101006e-01 3.64412159e-01
-5.63252509e-01 -1.41297057e-01 5.91139793e-01 -9.25115570e-02
5.63588850e-02 3.29115868e-01 -4.41334724e-01 5.49182519e-02
-6.07771456e-01 3.01411718e-01 4.21895832e-02 4.94181812e-01
-3.36267911e-02 -1.73428968e-01 -2.18553767e-01 6.37333810e-01
3.19179207e-01 8.38713109e-01 7.83532143e-01 -7.03256607e-01
5.02709329e-01 3.30815464e-01 -2.02054560e-01 -6.12218916e-01
-9.84058499e-01 -2.47272998e-01 -1.23620415e+00 -2.64213443e-01
-1.48011521e-02 -2.43887708e-01 -1.24402404e+00 1.08754671e+00
3.32993567e-01 4.56538424e-02 -3.62115324e-01 1.15857184e+00
9.49114084e-01 5.05381346e-01 5.18377163e-02 -2.27931470e-01
1.93817794e+00 -6.79374278e-01 -4.24456924e-01 4.13183600e-01
7.18782902e-01 -3.94549400e-01 8.33369970e-01 5.53788006e-01
-8.42401087e-01 4.99727428e-02 -6.86732590e-01 3.58556122e-01
-4.23926920e-01 -3.69892955e-01 4.64474618e-01 8.63259077e-01
-7.71790743e-01 1.22952789e-01 -8.87318671e-01 -5.10092795e-01
5.99503636e-01 2.29773521e-01 -2.35257402e-01 -3.24669152e-01
-1.24128413e+00 9.60809648e-01 2.59888917e-01 -5.05032599e-01
-1.23364103e+00 -1.02530801e+00 -4.40946043e-01 -2.66466320e-01
3.95317674e-01 -1.48458970e+00 1.13623452e+00 5.83453141e-02
-5.87901711e-01 1.13019264e+00 1.46679670e-01 -3.63945752e-01
5.54155707e-01 1.04621395e-01 -4.43204820e-01 7.13186204e-01
1.83052287e-01 3.26400071e-01 7.14312434e-01 -1.12245536e+00
-6.01193368e-01 -7.72648513e-01 -5.01567185e-01 -8.58962536e-02
4.26504344e-01 3.85668069e-01 -1.37167498e-01 -7.87051380e-01
-1.48689210e-01 -1.17612553e+00 -3.51863980e-01 -3.01097244e-01
-3.58588219e-01 -2.86336374e-02 9.89240229e-01 -7.62703538e-01
8.68812680e-01 -1.78930879e+00 -6.05114281e-01 1.76130340e-01
9.00068998e-01 5.57308137e-01 4.92880642e-01 -1.18819391e-02
-1.26848355e-01 7.95961738e-01 -5.81525862e-01 2.58660197e-01
-5.14846504e-01 2.18256623e-01 1.71734944e-01 5.99705935e-01
-1.86563656e-02 1.04529321e+00 -6.82669580e-01 -8.48483503e-01
3.53295237e-01 3.36440563e-01 -5.39808989e-01 2.90905565e-01
-6.30824044e-02 1.02991164e+00 -6.21263623e-01 7.83596039e-01
7.35088289e-01 -8.53459954e-01 -2.07868770e-01 3.04461598e-01
-1.80825479e-02 -2.22005501e-01 -3.42333913e-01 9.31482494e-01
-2.22642198e-01 3.99945021e-01 4.94363457e-02 -4.34379667e-01
2.47437000e-01 7.25266039e-01 9.24945414e-01 -7.03196108e-01
4.64238614e-01 1.42069414e-01 1.10967375e-01 -7.98992693e-01
6.51254952e-02 -7.74679482e-01 1.40623555e-01 9.29569185e-01
-5.77395439e-01 -5.23548782e-01 -1.44373611e-01 1.45433217e-01
1.23859167e+00 -8.41806293e-01 4.27577078e-01 1.47452295e-01
1.34682685e-01 2.85844564e-01 4.51404929e-01 9.28435862e-01
-3.09596926e-01 1.15536094e+00 1.37394026e-01 -6.21920407e-01
-1.23127651e+00 -1.47295785e+00 -7.48975158e-01 4.81428772e-01
-2.71386147e-01 -3.85920890e-02 -5.23247778e-01 -4.73783493e-01
-1.98750854e-01 3.82637769e-01 -6.91326857e-01 3.34790587e-01
-8.34508836e-01 -1.09908736e+00 8.91188622e-01 6.31853640e-01
3.58109444e-01 -8.30210447e-01 -1.04384565e+00 1.39133289e-01
-3.86999935e-01 -1.01075161e+00 -1.74813688e-01 1.38284387e-02
-9.55506027e-01 -1.32411397e+00 -1.20309365e+00 -2.86280274e-01
3.24479431e-01 5.02582565e-02 1.20788121e+00 5.56780756e-01
-8.53091061e-01 3.98233652e-01 -4.54265237e-01 -6.99436188e-01
-3.66069376e-01 -3.23494077e-01 8.65797624e-02 -7.66576052e-01
1.51474830e-02 -3.57402921e-01 -8.13327789e-01 2.09021747e-01
-9.69317436e-01 9.28942561e-02 5.20371437e-01 8.54935348e-01
1.07073414e+00 -1.02645159e-01 1.08697288e-01 -9.89215136e-01
4.29515451e-01 -9.88545656e-01 -3.19299549e-01 -1.97861865e-02
-8.69899690e-01 -7.92085350e-01 6.05682254e-01 2.41764784e-01
-7.74657667e-01 -5.54832458e-01 -2.62990236e-01 -7.64768720e-01
-3.33547682e-01 5.50738037e-01 7.71080375e-01 5.09934366e-01
6.20536745e-01 8.84293839e-02 -1.62660852e-01 -4.57936317e-01
-1.73858166e-01 7.68250465e-01 5.78479230e-01 -1.90089330e-01
5.56146801e-01 7.83814907e-01 2.41422430e-01 -8.24916422e-01
-5.73898077e-01 -7.79478788e-01 -6.87492967e-01 -7.02857971e-02
1.72120011e+00 -8.65386426e-01 -4.88585293e-01 4.00432408e-01
-9.15065527e-01 -1.35612294e-01 -2.11543903e-01 9.87151861e-01
-3.07835758e-01 1.29705921e-01 -9.97457027e-01 -2.17639595e-01
-8.84813309e-01 -1.37927127e+00 1.01667666e+00 -9.04464349e-02
-3.21685582e-01 -9.57700908e-01 5.87401330e-01 7.24995613e-01
4.80381370e-01 6.85270786e-01 1.48987687e+00 -7.45462775e-01
-6.86491668e-01 -6.85783476e-02 -4.84044045e-01 -3.25735472e-02
-4.82858233e-02 -2.79449485e-02 -5.39536655e-01 -1.69602498e-01
1.78775445e-01 -4.67901796e-01 7.63016164e-01 7.12987363e-01
1.37973380e+00 -4.18745466e-02 -3.69855344e-01 1.04921758e+00
1.49352753e+00 7.94994831e-01 3.00151438e-01 3.11139077e-01
7.99019217e-01 -2.04611167e-01 4.07638580e-01 8.80052209e-01
2.21369103e-01 3.52100879e-01 6.46116912e-01 -4.66572255e-01
-1.87501647e-02 1.21462524e-01 -5.78066409e-01 7.90288627e-01
-6.75270319e-01 -3.70309353e-01 -1.57512164e+00 3.48842859e-01
-8.37182224e-01 -7.20490217e-01 -7.79704213e-01 1.50798321e+00
2.86973506e-01 -2.50738651e-01 2.75825948e-01 -4.08665314e-02
5.97668290e-01 -1.59423053e-01 -5.46017826e-01 -6.03957355e-01
1.28828213e-01 5.59046507e-01 3.48789036e-01 -2.07393318e-02
-7.98096657e-01 1.68000340e-01 6.77085257e+00 4.39456910e-01
-1.28639174e+00 7.20204294e-01 6.93124294e-01 1.80822723e-02
-3.06353539e-01 -5.51127374e-01 -3.84966969e-01 4.61257994e-01
8.53048801e-01 8.06870908e-02 5.21016084e-02 6.11545384e-01
3.30105931e-01 -2.88053662e-01 -5.93021750e-01 9.14030194e-01
-9.46526676e-02 -1.65741432e+00 3.42070535e-02 2.80066639e-01
5.68903983e-01 6.95291281e-01 3.53278428e-01 8.82871673e-02
2.77875572e-01 -1.33277440e+00 3.14303219e-01 4.79766987e-02
1.50276434e+00 -5.37050545e-01 1.10106421e+00 1.66212499e-01
-1.26258588e+00 1.14875317e-01 -9.38042477e-02 3.26101601e-01
2.67172217e-01 4.48754102e-01 -1.62677383e+00 6.94640577e-01
1.13036788e+00 2.42684186e-01 -5.00431597e-01 9.41763937e-01
2.33499594e-02 8.41727495e-01 -2.73318857e-01 1.59909889e-01
4.11694825e-01 -9.37730297e-02 5.16857564e-01 1.38030767e+00
4.48745847e-01 8.33512902e-01 3.95150408e-02 8.40281904e-01
4.73262742e-02 -4.37677354e-02 -8.54630291e-01 9.98432487e-02
2.40482137e-01 9.33458030e-01 -1.14367628e+00 -5.91389656e-01
-3.72650027e-01 4.84268814e-01 -3.94230247e-01 1.50045678e-02
-1.15779233e+00 3.58149976e-01 1.77847326e-01 5.23861945e-01
1.07041799e-01 1.31748855e-01 -5.31740010e-01 -7.78534770e-01
-6.76840603e-01 -9.56838667e-01 1.06706488e+00 -1.02902627e+00
-1.18623674e+00 6.89957917e-01 1.95020303e-01 -1.04658902e+00
-4.71470237e-01 -5.68379581e-01 -5.37419915e-01 5.86092412e-01
-9.95132029e-01 -8.92988861e-01 -5.08776903e-01 8.81964326e-01
3.88222724e-01 -2.05907673e-01 9.38011229e-01 2.67510682e-01
-4.60372537e-01 2.58735657e-01 1.99846461e-01 1.66171595e-01
3.55097428e-02 -1.14483964e+00 -7.07607018e-03 2.83141941e-01
-5.64110518e-01 3.69696975e-01 7.18612224e-02 -8.87973547e-01
-9.98726368e-01 -1.30952644e+00 3.36358547e-01 -5.86230695e-01
2.37840846e-01 2.61064529e-01 -6.29224837e-01 8.40841532e-01
1.21615723e-01 2.50234276e-01 1.00224423e+00 -6.94019139e-01
-2.95337364e-02 5.17873228e-01 -1.59340143e+00 5.40851392e-02
8.55719209e-01 -4.80529577e-01 -7.58144379e-01 6.34148717e-01
7.60630012e-01 -5.15203536e-01 -1.28730524e+00 6.86254561e-01
4.73520875e-01 -1.10166681e+00 1.26785147e+00 -5.76621354e-01
4.45732385e-01 1.81977227e-01 -9.02411342e-02 -8.95844698e-01
-4.86423850e-01 1.01826690e-01 2.66415894e-01 2.46201903e-01
3.43279123e-01 -7.05018580e-01 8.15030813e-01 2.07546294e-01
-2.14241862e-01 -9.00773466e-01 -9.23518896e-01 -7.57602155e-01
2.61382580e-01 -6.19948685e-01 6.98657930e-01 1.24624574e+00
-5.49427807e-01 -1.14506595e-01 6.45181164e-03 2.92529259e-03
2.92937785e-01 1.63332582e-01 3.00296366e-01 -1.12516391e+00
-4.34650183e-01 -3.82575721e-01 -5.80703095e-02 -2.92365402e-01
-6.23863995e-01 -9.08763885e-01 -4.86464918e-01 -1.83449996e+00
8.02447915e-01 -6.79337382e-01 -5.30142635e-02 1.79875508e-01
-4.45263870e-02 3.90357554e-01 2.95616657e-01 5.00738502e-01
-1.34897426e-01 -3.04024927e-02 1.78111124e+00 -1.24993026e-02
1.98940545e-01 6.69544712e-02 1.48251904e-02 9.68922853e-01
9.14426446e-01 -9.76536572e-01 -3.26187015e-01 -8.03669319e-02
1.11883655e-01 7.95537531e-01 4.35252488e-01 -8.26350629e-01
-3.64357054e-01 -4.04269606e-01 6.83139086e-01 -1.30170870e+00
4.28748816e-01 -7.82244623e-01 6.05283558e-01 1.43342102e+00
1.55082777e-01 8.32290471e-01 -4.59209271e-02 3.10115188e-01
-9.72292423e-02 -2.49629632e-01 1.02612519e+00 -8.76899362e-01
-3.47879469e-01 8.11605453e-01 -1.28281724e+00 5.87975264e-01
1.12347531e+00 -1.18885629e-01 -6.23903155e-01 -1.57973036e-01
-6.55868292e-01 -6.76312447e-02 3.62723947e-01 -1.73494726e-01
9.44735765e-01 -1.03632641e+00 -7.92893648e-01 1.76257208e-01
2.61493959e-02 5.11538625e-01 3.62393469e-01 1.44292021e+00
-1.39893973e+00 8.07731092e-01 -3.05063903e-01 -1.01090896e+00
-1.40054476e+00 8.01435113e-01 5.07354975e-01 -8.55558276e-01
-9.52154815e-01 9.73101854e-01 5.62466025e-01 -3.33527237e-01
-1.49676174e-01 -4.40297753e-01 -7.27781653e-02 -1.14664972e-01
4.41935867e-01 3.85002941e-01 2.42634118e-01 -6.48782074e-01
-5.73400438e-01 6.41048372e-01 4.05681171e-02 -2.49132570e-02
1.23410094e+00 -8.38758126e-02 -6.63714483e-02 1.36518970e-01
1.35581303e+00 -2.60088414e-01 -2.75534570e-01 1.33581340e-01
-3.07246059e-01 -6.09674454e-01 -9.15551856e-02 -9.08894122e-01
-1.01783955e+00 8.79900634e-01 1.06336367e+00 -7.69469561e-03
1.07217407e+00 2.67364115e-01 1.05147111e+00 -5.56975827e-02
2.87912875e-01 -4.90475833e-01 4.41612303e-01 5.77701986e-01
6.61102712e-01 -1.19340742e+00 1.24680631e-01 -2.60617703e-01
-6.51036203e-01 7.27871895e-01 4.56320554e-01 2.37889886e-01
6.62966013e-01 4.82793361e-01 1.41806111e-01 -9.20183957e-01
-7.06966043e-01 3.32043320e-02 -2.72582471e-01 8.30522537e-01
2.74304897e-01 6.23761654e-01 -1.91893816e-01 6.49951518e-01
-3.66917789e-01 8.54950324e-02 4.91655469e-01 9.93107855e-01
-2.85140455e-01 -7.11798728e-01 -8.95853639e-01 1.11988068e+00
-9.20320213e-01 -2.13179082e-01 -4.40381348e-01 1.24945533e+00
6.84009254e-01 4.54200417e-01 1.17672846e-01 -3.01344931e-01
-3.47599983e-02 -2.09767848e-01 4.29520518e-01 -5.38002551e-01
-7.46254742e-01 -1.45156533e-02 7.52070099e-02 -1.53727710e-01
-8.09865594e-02 -6.12514377e-01 -1.67333639e+00 -4.88544434e-01
-1.93499506e-01 2.53176526e-03 5.37470222e-01 8.81857991e-01
-8.36866423e-02 5.74409604e-01 3.86858761e-01 -2.51615554e-01
-2.26400420e-02 -7.88109601e-01 -8.54082465e-01 2.52825320e-01
6.43701255e-01 -6.43069327e-01 -1.13356024e-01 -6.38546944e-02] | [15.442106246948242, -1.8317177295684814] |
d71e7f91-149a-4d54-b54e-1af25fdb92cb | weakly-supervised-body-part-parsing-with-pose | 1907.13051 | null | https://arxiv.org/abs/1907.13051v2 | https://arxiv.org/pdf/1907.13051v2.pdf | Weakly Supervised Body Part Segmentation with Pose based Part Priors | Human body part segmentation refers to the task of predicting the semantic segmentation mask for each body part. Fully supervised body part segmentation methods achieve good performances but require an enormous amount of effort to annotate part masks for training. In contrast to high annotation costs needed for a limited number of part mask annotations, a large number of weak labels such as poses and full body masks already exist and contain relevant information. Motivated by the possibility of using existing weak labels, we propose the first weakly supervised body part segmentation framework. The core idea is first converting the sparse weak labels such as keypoints to the initial estimate of body part masks, and then iteratively refine the part mask predictions. We name the initial part masks estimated from poses the "part priors." With sufficient extra weak labels, our weakly supervised framework achieves a comparable performance (62.0% mIoU) to the fully supervised method (63.6% mIoU) on the Pascal-Person-Part dataset. Furthermore, in the extended semi-supervised setting, the proposed framework outperforms the state-of-art methods. Moreover, we extend our proposed framework to other keypoint-supervised part segmentation tasks such as face parsing. | ['Yuncheng Li', 'Jiebo Luo', 'Linjie Yang', 'Zhengyuan Yang', 'Ning Zhang'] | 2019-07-30 | null | null | null | null | ['face-parsing'] | ['computer-vision'] | [ 5.14643967e-01 9.09928977e-01 -4.79779065e-01 -6.50419295e-01
-9.12968040e-01 -3.96786034e-01 3.63037527e-01 -2.25969195e-01
-2.71556735e-01 5.40709674e-01 1.78988233e-01 3.55822265e-01
3.05666715e-01 -4.72143918e-01 -9.52314615e-01 -4.34218705e-01
2.30160162e-01 8.18995357e-01 5.17982841e-01 -9.30017829e-02
-2.87666053e-01 1.29938081e-01 -1.63320136e+00 5.74514531e-02
6.62486672e-01 1.04782212e+00 -5.65472990e-02 2.53788024e-01
-1.85610905e-01 3.31553340e-01 -3.28511178e-01 -8.32839251e-01
4.11992908e-01 -5.18573105e-01 -1.14771891e+00 6.75181448e-01
7.30448544e-01 -2.38398388e-01 1.76726043e-01 1.09570003e+00
4.36761528e-01 -6.25493228e-02 4.68777955e-01 -1.09979737e+00
2.08649963e-01 6.72602296e-01 -8.39572310e-01 -4.59209651e-01
4.60285872e-01 -2.05144778e-01 8.62218261e-01 -7.67882943e-01
6.86731696e-01 1.28213882e+00 9.00992274e-01 8.60436916e-01
-1.11686039e+00 -5.89919746e-01 4.95884091e-01 -3.22099894e-01
-1.40125847e+00 -6.51726842e-01 7.99036980e-01 -5.25069535e-01
2.72929132e-01 5.56439161e-02 6.72344089e-01 7.52799034e-01
-2.69164234e-01 1.09444916e+00 8.59141409e-01 -2.78300494e-01
2.90329922e-02 -3.61423083e-02 3.85589898e-01 1.21104252e+00
2.96863616e-01 -4.58570957e-01 -4.64293331e-01 -2.63806701e-01
7.33403087e-01 -8.85383785e-02 -3.99721041e-02 -5.46087086e-01
-1.05135334e+00 5.28949261e-01 2.32072383e-01 -1.34557812e-02
-3.16374630e-01 2.87235588e-01 2.76859701e-01 -3.90989959e-01
5.40423810e-01 -2.94717431e-01 -6.83892608e-01 1.24817170e-01
-1.21143270e+00 1.21239901e-01 8.03866506e-01 1.14065766e+00
1.10080016e+00 -3.83178622e-01 -2.35501856e-01 9.95375276e-01
6.02435291e-01 4.25874770e-01 1.36212319e-01 -1.14841950e+00
6.18141651e-01 6.88866496e-01 1.59718677e-01 -4.35521513e-01
-5.09157002e-01 -4.57875758e-01 -6.73766494e-01 -1.99376449e-01
7.44808257e-01 -1.72443852e-01 -1.31430757e+00 1.69541502e+00
1.00954771e+00 -6.10616617e-02 -3.21648389e-01 8.58780384e-01
9.11011398e-01 1.44131199e-01 4.24084961e-01 -8.09391886e-02
1.71815276e+00 -1.18781936e+00 -8.27936947e-01 -6.45095766e-01
3.98457646e-01 -6.86271369e-01 6.99290097e-01 1.74936175e-01
-1.14548492e+00 -7.90267348e-01 -6.91582859e-01 -1.19793907e-01
7.86867589e-02 6.14408314e-01 8.36779714e-01 7.76507080e-01
-8.61414552e-01 5.01098752e-01 -1.06535029e+00 -3.53938788e-01
7.74883568e-01 6.99412942e-01 -5.39801002e-01 -1.14352390e-01
-6.41825616e-01 3.59784305e-01 4.83656734e-01 2.98082769e-01
-6.88148320e-01 -2.63545334e-01 -1.32261610e+00 -1.59318015e-01
6.92699850e-01 -7.36769795e-01 1.23508012e+00 -1.07754004e+00
-1.49886715e+00 1.33853495e+00 -4.21139598e-01 -2.78570324e-01
7.10937381e-01 -6.38404310e-01 1.44349501e-01 5.16379654e-01
3.73198569e-01 1.24953711e+00 1.04192579e+00 -1.26853192e+00
-6.38924658e-01 -5.11221170e-01 -1.33350581e-01 2.18023241e-01
8.59866068e-02 5.82166277e-02 -1.25486815e+00 -5.84938109e-01
5.10148585e-01 -1.04094601e+00 -3.84733796e-01 2.09234491e-01
-6.12333715e-01 -4.21869159e-01 5.84682703e-01 -8.87565374e-01
8.48401248e-01 -2.04879665e+00 1.02959841e-01 1.35665283e-01
9.94260833e-02 7.56572038e-02 1.59515113e-01 -1.18967079e-01
5.83641939e-02 -1.18028164e-01 -7.22728014e-01 -8.74352932e-01
8.67475346e-02 3.91636521e-01 2.04512790e-01 6.51340008e-01
9.60945114e-02 1.03160059e+00 -6.66392148e-01 -1.07722139e+00
2.27980480e-01 2.64002651e-01 -6.56150579e-01 1.44405290e-01
-2.70371377e-01 6.21916592e-01 -5.24108648e-01 1.01266861e+00
7.77781248e-01 -4.26912129e-01 2.46247381e-01 -1.55869514e-01
3.18937778e-01 -2.29596999e-02 -1.28025150e+00 2.11029959e+00
1.59027815e-01 -2.23241478e-01 6.21661544e-01 -1.01951122e+00
6.17421091e-01 3.85211259e-01 7.70970345e-01 -6.01565428e-02
2.46456191e-01 2.85851322e-02 -1.77515477e-01 -3.99436593e-01
9.38879550e-02 -4.13432956e-01 -2.96993345e-01 1.95827052e-01
4.94980335e-01 -1.23978287e-01 2.69250661e-01 -4.39544693e-02
7.57273614e-01 8.95323217e-01 4.48030114e-01 -2.05588698e-01
7.39919424e-01 -8.19403604e-02 8.94593000e-01 3.60592365e-01
-4.77315217e-01 7.86115587e-01 4.59104180e-01 -2.76582539e-01
-4.07967359e-01 -8.12570453e-01 -2.74966538e-01 1.30467451e+00
2.57545799e-01 -5.77927768e-01 -1.34596264e+00 -1.22575617e+00
-7.71340951e-02 1.68639049e-02 -7.22260177e-01 2.45895162e-01
-7.56175280e-01 -6.87903523e-01 5.53122580e-01 7.50354767e-01
7.02790201e-01 -1.06026793e+00 -2.58751631e-01 4.84637730e-02
-3.56748909e-01 -1.36159515e+00 -5.76297343e-01 6.80944622e-02
-9.86211598e-01 -1.21726489e+00 -1.01814210e+00 -9.30761278e-01
1.00163305e+00 -3.33823264e-01 1.14042544e+00 2.21914798e-01
-2.35912919e-01 4.02940154e-01 -2.30621666e-01 -2.04435289e-01
-1.09851919e-01 -5.40341474e-02 -5.88440374e-02 1.79526746e-01
1.26426667e-01 -1.66170076e-01 -7.95592368e-01 5.12732625e-01
-5.70666850e-01 2.27742001e-01 8.93787742e-01 6.62923157e-01
1.13315201e+00 -1.38351932e-01 1.93433270e-01 -1.27426541e+00
-3.75910252e-01 -9.39016193e-02 -3.00992876e-01 1.93952024e-01
-1.16783336e-01 -1.82810128e-02 2.72586495e-01 -3.12316269e-01
-1.36895239e+00 8.09481323e-01 -7.03339159e-01 -1.61044389e-01
-6.51799321e-01 -9.35247466e-02 -7.31932044e-01 -4.24238183e-02
2.58727759e-01 -7.70760477e-02 -1.12880304e-01 -8.90256882e-01
4.30421978e-01 3.57040226e-01 6.95917845e-01 -8.30760241e-01
7.79415965e-01 7.04460859e-01 5.73611930e-02 -6.30502820e-01
-1.20057797e+00 -8.22316945e-01 -1.19416368e+00 -3.85272056e-02
1.24134076e+00 -1.07485676e+00 -3.69512081e-01 4.40812826e-01
-9.28812325e-01 -4.93263900e-01 -4.16267127e-01 2.46405020e-01
-6.66065454e-01 8.70639741e-01 -7.02470243e-01 -6.97157145e-01
-3.79186809e-01 -8.99523854e-01 1.84034002e+00 1.00833148e-01
-3.68166924e-01 -6.41907752e-01 -1.73170611e-01 1.00585365e+00
-4.51114923e-01 3.01662147e-01 4.08419967e-01 -5.45545042e-01
-2.07987353e-01 -4.90377098e-01 -4.35060486e-02 3.55696172e-01
5.63957207e-02 -4.87640768e-01 -1.04825592e+00 -1.46426171e-01
-1.68990090e-01 -4.28747892e-01 8.14556718e-01 6.04288876e-01
1.12022531e+00 -1.12618253e-01 -7.00917244e-01 6.30056500e-01
9.84152496e-01 -2.20122039e-01 2.74478674e-01 -3.24633896e-01
1.10328686e+00 8.95528555e-01 9.05717790e-01 3.30897689e-01
3.77249926e-01 6.97535932e-01 1.83633223e-01 -3.49339545e-01
-2.64461547e-01 -6.76247656e-01 8.81372094e-02 4.47692931e-01
-2.46920973e-01 3.22830305e-02 -7.03785062e-01 4.04114187e-01
-1.87593305e+00 -5.41687369e-01 -9.04029384e-02 2.03338480e+00
8.95286620e-01 2.87389904e-01 4.94497567e-01 8.87410045e-02
9.15300608e-01 2.82884091e-02 -3.87481153e-01 2.81622767e-01
2.84072459e-01 3.61474067e-01 4.42892283e-01 4.28123713e-01
-1.56284893e+00 1.22282922e+00 6.15069675e+00 8.22982073e-01
-4.85984057e-01 3.31519693e-01 6.00553632e-01 1.79410309e-01
1.35682538e-01 3.58022526e-02 -1.18238688e+00 3.70376050e-01
5.39346695e-01 7.78075278e-01 -1.28783703e-01 1.11703598e+00
-1.70931891e-01 -2.37711489e-01 -1.08600140e+00 9.73078609e-01
1.70041665e-01 -6.14481807e-01 -1.36122122e-01 1.67340755e-01
7.36576736e-01 -3.44928712e-01 -4.11415726e-01 3.43281358e-01
5.50691299e-02 -8.08372915e-01 9.68282163e-01 2.54100055e-01
8.45222950e-01 -4.57227528e-01 7.19556928e-01 6.20271981e-01
-1.43450344e+00 1.91519275e-01 -3.74497622e-01 -5.48203178e-02
3.18170279e-01 5.64428329e-01 -4.91601408e-01 3.96419227e-01
6.61773086e-01 6.02469504e-01 -5.59820652e-01 8.67226005e-01
-6.94823444e-01 8.18905592e-01 -5.61945796e-01 4.73657489e-01
-1.23080097e-01 -1.18546613e-01 1.23414211e-01 1.06555223e+00
-1.73564792e-01 2.77526110e-01 6.07200682e-01 6.08585000e-01
-3.00037831e-01 1.75484985e-01 -1.07600562e-01 1.77204400e-01
-6.38484508e-02 1.43226540e+00 -1.24695754e+00 -6.26832187e-01
-5.38026989e-01 1.28535926e+00 2.17733309e-01 1.95623338e-01
-7.38972723e-01 1.42634496e-01 3.77995163e-01 2.79742986e-01
4.36699808e-01 9.20554027e-02 -2.61470407e-01 -1.06755269e+00
1.45352527e-01 -6.06124699e-01 6.83614016e-01 -3.19765121e-01
-9.33571935e-01 2.90514141e-01 3.18349451e-01 -9.93924677e-01
-3.27406496e-01 -5.71244597e-01 -3.90821874e-01 5.20787954e-01
-1.17644322e+00 -1.68979037e+00 -3.12096924e-01 4.66938585e-01
7.88887978e-01 4.78315473e-01 6.92955971e-01 2.24505082e-01
-6.28458261e-01 5.32403648e-01 -7.32680023e-01 4.82234359e-01
6.30608141e-01 -1.30846190e+00 3.37716430e-01 8.75739396e-01
1.68278977e-01 6.16300285e-01 4.81906861e-01 -9.90723133e-01
-9.96145666e-01 -1.05663383e+00 7.06281304e-01 -5.67202032e-01
-6.35337457e-02 -6.06465161e-01 -5.27412891e-01 8.63295257e-01
-2.41964832e-01 4.50576961e-01 7.35962093e-01 1.58920273e-01
-2.83945333e-02 1.34612963e-01 -1.27806103e+00 3.01425755e-01
1.51915288e+00 -1.33958653e-01 -7.12132037e-01 4.86450851e-01
4.79911894e-01 -6.60840273e-01 -7.42375314e-01 8.15851927e-01
5.78893483e-01 -7.86467731e-01 1.10554135e+00 -3.20420235e-01
1.07283123e-01 -4.03944880e-01 3.02681699e-02 -4.88698453e-01
-1.68504015e-01 -6.68159902e-01 -2.85298645e-01 1.41990232e+00
1.05458401e-01 -1.68718264e-01 1.34128058e+00 7.68114805e-01
-1.97227001e-01 -7.03936636e-01 -8.94277394e-01 -4.08433497e-01
-2.21052542e-01 -4.44468766e-01 3.10926080e-01 5.50194383e-01
-2.76682347e-01 1.65645838e-01 -3.29290330e-01 1.89209312e-01
9.10226583e-01 4.33420032e-01 9.17649269e-01 -1.34658825e+00
-4.27046239e-01 -1.60732761e-01 -4.05993015e-01 -1.30836713e+00
4.08003807e-01 -7.39441812e-01 5.15104353e-01 -1.56199062e+00
4.21184093e-01 -2.17580304e-01 -7.89991617e-02 9.82093453e-01
-5.74471176e-01 7.93621957e-01 -7.76639059e-02 1.37035713e-01
-9.33763206e-01 3.74035537e-01 1.26965404e+00 -1.19627722e-01
-1.45152003e-01 4.30861324e-01 -5.80370426e-01 1.36869574e+00
4.39736336e-01 -3.92997980e-01 -1.68366283e-01 -4.84979190e-02
-2.45028332e-01 7.57497773e-02 2.67822593e-01 -9.91450965e-01
-9.59040150e-02 6.12152889e-02 6.96624100e-01 -7.88189471e-01
5.19619524e-01 -7.12383211e-01 4.32960078e-04 4.23108876e-01
6.15456142e-02 -7.36565173e-01 5.55750839e-02 5.45180142e-01
-1.25573277e-01 -3.32078189e-01 6.92358613e-01 -4.62991118e-01
-6.47556543e-01 4.70243454e-01 -3.11376508e-02 2.50745356e-01
9.33801234e-01 -3.80473793e-01 3.42658967e-01 -2.17116475e-02
-1.20232046e+00 2.54934460e-01 5.42026699e-01 1.89245000e-01
2.72691756e-01 -9.89319623e-01 -4.11957771e-01 2.40921453e-01
-3.91223244e-02 5.19959569e-01 2.91844845e-01 8.96493912e-01
-2.69059569e-01 3.62498105e-01 3.31572779e-02 -7.07726479e-01
-1.47801566e+00 4.37838018e-01 2.45597467e-01 -1.28721595e-01
-7.98870087e-01 1.10509610e+00 6.02252185e-01 -5.78674257e-01
3.24101388e-01 -3.54082316e-01 -1.51752546e-01 5.21381870e-02
1.85630411e-01 2.85248369e-01 -1.57101601e-01 -1.24934042e+00
-6.19877338e-01 1.07983458e+00 3.06762397e-01 6.20847521e-03
1.06219923e+00 -2.00149789e-02 -3.44877280e-02 -6.38908371e-02
9.36940610e-01 2.36034155e-01 -1.45333076e+00 -9.53154713e-02
5.88018401e-03 -1.28620893e-01 -4.53822970e-01 -6.97389483e-01
-1.07170844e+00 8.63260567e-01 4.65305895e-01 -4.87660676e-01
9.98704493e-01 5.98815084e-01 8.68033648e-01 1.75477505e-01
6.17458582e-01 -1.14114773e+00 -2.17473865e-01 1.54362544e-01
5.70352376e-01 -1.26276433e+00 2.84802794e-01 -1.30174196e+00
-5.80396473e-01 6.26736343e-01 7.67395377e-01 3.65478732e-02
5.91132760e-01 1.57589838e-01 1.17691733e-01 -3.58817905e-01
8.06878321e-03 -7.29948044e-01 7.49697328e-01 5.66820681e-01
5.10457516e-01 3.04187499e-02 -5.74114561e-01 1.07770860e+00
-2.81261891e-01 -6.37631565e-02 -2.88363099e-01 8.93610954e-01
-3.67790490e-01 -1.37712669e+00 -6.58255935e-01 3.80529016e-01
-8.49770308e-01 3.79822433e-01 -5.17243087e-01 8.12538803e-01
6.71243310e-01 8.10487330e-01 -1.71122670e-01 4.75813299e-02
1.73552305e-01 5.76622963e-01 5.82667947e-01 -1.02255571e+00
-4.35833305e-01 6.43589735e-01 1.85059384e-01 -8.15728545e-01
-7.42277801e-01 -6.96022332e-01 -1.56627464e+00 5.23140192e-01
-3.61565262e-01 1.43139780e-01 3.87711465e-01 1.24611449e+00
-9.24754664e-02 2.25639209e-01 -1.21990204e-01 -1.17274618e+00
-1.49958223e-01 -1.05768430e+00 -5.65376759e-01 7.70558596e-01
-3.93172540e-02 -8.24726641e-01 -3.67134511e-02 4.55828279e-01] | [8.22274398803711, -0.2438848614692688] |
6fc1a918-a212-4469-96d8-89ffa2df9f5f | machine-learning-for-advancing-low | 2307.00131 | null | https://arxiv.org/abs/2307.00131v1 | https://arxiv.org/pdf/2307.00131v1.pdf | Machine learning for advancing low-temperature plasma modeling and simulation | Machine learning has had an enormous impact in many scientific disciplines. Also in the field of low-temperature plasma modeling and simulation it has attracted significant interest within the past years. Whereas its application should be carefully assessed in general, many aspects of plasma modeling and simulation have benefited substantially from recent developments within the field of machine learning and data-driven modeling. In this survey, we approach two main objectives: (a) We review the state-of-the-art focusing on approaches to low-temperature plasma modeling and simulation. By dividing our survey into plasma physics, plasma chemistry, plasma-surface interactions, and plasma process control, we aim to extensively discuss relevant examples from literature. (b) We provide a perspective of potential advances to plasma science and technology. We specifically elaborate on advances possibly enabled by adaptation from other scientific disciplines. We argue that not only the known unknowns, but also unknown unknowns may be discovered due to an inherent propensity to spotlight hidden patterns in data. | ['Tobias Gergs', 'Luca Vialetto', 'Jan Trieschmann'] | 2023-06-30 | null | null | null | null | ['known-unknowns'] | ['miscellaneous'] | [ 3.47338915e-01 8.02266747e-02 -4.03727069e-02 -3.54454339e-01
-1.80687845e-01 -6.06821813e-02 1.03566074e+00 1.14203833e-01
2.93156393e-02 7.64994740e-01 -2.32927158e-01 -5.36011219e-01
-2.92763501e-01 -6.67142153e-01 -1.97041810e-01 -1.18835032e+00
-3.82048696e-01 1.05465508e+00 -2.35907927e-01 -4.66038674e-01
3.86771113e-01 1.11990428e+00 -1.36428261e+00 3.23132247e-01
5.61822712e-01 8.85259092e-01 -1.11453356e-02 3.70005995e-01
-2.73617566e-01 7.66865194e-01 -4.11900699e-01 1.66623399e-01
-3.48808467e-02 -4.07820970e-01 -6.63871646e-01 -8.25131610e-02
-5.68420649e-01 6.19145215e-01 -4.70057607e-01 5.09762466e-01
3.08631122e-01 2.87515461e-01 1.15578997e+00 -1.11441672e+00
-1.40163898e-01 2.15762183e-01 -7.55453289e-01 4.43794250e-01
-1.26722962e-01 5.63784242e-01 2.61501789e-01 -7.09985256e-01
3.92308354e-01 9.54773009e-01 1.49815395e-01 6.31022811e-01
-1.11151958e+00 -3.97175014e-01 -3.10599227e-02 3.49902004e-01
-1.14006543e+00 -4.78878886e-01 1.07578766e+00 -8.32305908e-01
1.46385694e+00 2.39719242e-01 4.14114773e-01 8.69485915e-01
6.81589067e-01 4.00259495e-01 1.04393756e+00 -6.71643972e-01
5.43555856e-01 5.40273190e-01 1.23066820e-01 3.51670861e-01
8.71545747e-02 1.03392112e+00 -5.44315338e-01 -3.99330109e-01
4.19695407e-01 -3.81261408e-01 -1.40225977e-01 -6.11885130e-01
-7.82128751e-01 1.03611553e+00 -2.48003174e-02 6.89674795e-01
-3.42274636e-01 -5.14722347e-01 3.99333030e-01 -3.23996171e-02
3.17245096e-01 1.06305850e+00 -7.71796346e-01 -1.55515596e-01
-7.41498590e-01 4.90603149e-01 1.02219617e+00 4.68162954e-01
6.86544061e-01 4.38832074e-01 1.04780532e-01 7.71156490e-01
1.11123964e-01 5.10695338e-01 1.90685675e-01 -4.12156999e-01
-9.42180119e-03 1.77649967e-02 3.90196025e-01 -5.60344100e-01
-7.24276602e-01 -2.77774315e-02 -1.11535060e+00 4.35179234e-01
1.75953433e-01 -6.23723984e-01 -7.41117835e-01 1.15600812e+00
1.40567496e-01 1.45995036e-01 2.64049053e-01 6.38599932e-01
6.53337836e-01 9.11386728e-01 1.86526626e-01 -7.04187393e-01
1.09419119e+00 -8.54161620e-01 -6.58728957e-01 -2.65758306e-01
4.67106402e-01 -6.05435014e-01 2.59543896e-01 4.47642922e-01
-1.19819605e+00 -4.87124443e-01 -9.52661514e-01 2.08373785e-01
-6.35697901e-01 -6.06628299e-01 1.16467071e+00 7.63015985e-01
-5.55412292e-01 8.07641983e-01 -1.17555296e+00 -4.02385920e-01
1.51451394e-01 3.10610831e-01 4.93161492e-02 5.16055405e-01
-1.17205513e+00 1.23372757e+00 1.17719792e-01 -4.03252095e-02
-7.34303415e-01 -1.23506868e+00 -6.14046693e-01 -2.18887240e-01
1.21068656e-02 -8.32355917e-01 1.42885876e+00 7.07199499e-02
-1.96704841e+00 6.89731836e-01 -9.09844875e-01 -5.48380494e-01
2.52354205e-01 2.21060202e-01 -9.00603652e-01 3.06445919e-02
-3.72500181e-01 2.32257903e-01 4.16924387e-01 -1.32885134e+00
-7.24725842e-01 -2.61330009e-01 -6.24393582e-01 -1.46933988e-01
2.40423888e-01 2.72047907e-01 2.09982783e-01 -1.32650778e-01
-8.34202543e-02 -5.51292360e-01 -8.79660666e-01 -6.02210939e-01
-1.97279915e-01 -3.25163633e-01 1.07386398e+00 1.45102367e-01
6.98992431e-01 -1.70750058e+00 2.84809738e-01 2.53117412e-01
2.79891547e-02 1.25452787e-01 4.45741206e-01 8.67259979e-01
-4.87319380e-01 2.99477987e-02 -6.12642705e-01 -2.97926843e-01
-2.91703314e-01 -9.99064296e-02 -7.21164167e-01 5.62948287e-01
3.11038673e-01 9.99506533e-01 -7.43882716e-01 2.66095221e-01
1.09602189e+00 5.77418447e-01 -4.69841063e-03 2.60069013e-01
-3.83039534e-01 1.09958720e+00 -4.81839627e-01 5.73832333e-01
6.49609745e-01 -2.11529121e-01 -2.48268563e-02 -9.18917283e-02
-5.96302152e-01 7.50260502e-02 -4.98761237e-01 1.02179742e+00
-3.71871322e-01 3.96353632e-01 2.46960506e-01 -1.43589091e+00
1.00788593e+00 4.60995048e-01 1.00152910e+00 -7.40554094e-01
3.57245743e-01 9.80223194e-02 1.94374785e-01 -5.32431006e-01
3.08067292e-01 -1.06036496e+00 2.89430976e-01 4.07693654e-01
-2.09612384e-01 -6.09712243e-01 5.81286289e-03 -1.51558548e-01
7.86700845e-01 -5.14187694e-01 4.28351104e-01 -7.54932821e-01
7.88891315e-01 2.06024170e-01 1.32871404e-01 7.45589197e-01
-1.22326210e-01 -1.39517067e-02 3.85221750e-01 -7.06983149e-01
-1.11852598e+00 -8.92969668e-01 -8.48923802e-01 3.78270566e-01
8.16549882e-02 -4.33773488e-01 -3.29992920e-01 -6.78939447e-02
2.39855424e-01 8.15361857e-01 -7.71937966e-01 -1.59503892e-01
-5.66004634e-01 -1.72820699e+00 5.93843944e-02 4.04484391e-01
-9.69460756e-02 -1.41948473e+00 -5.01624584e-01 2.26272315e-01
5.62343895e-01 -8.76010358e-01 6.27626598e-01 5.93910813e-01
-8.81784499e-01 -1.09200549e+00 -2.71363050e-01 -2.90469974e-01
2.78284997e-01 1.15794905e-01 1.04679537e+00 -2.90395617e-01
-6.91172183e-01 1.38769999e-01 -7.43607283e-02 -9.82512653e-01
-6.37699962e-01 8.93547833e-02 4.88066763e-01 -1.02451816e-01
1.01538742e+00 -7.52214551e-01 -2.41913378e-01 2.61121154e-01
-4.84674007e-01 -6.55006021e-02 3.61161590e-01 5.74097633e-01
3.83203536e-01 4.18455333e-01 7.29878008e-01 -1.11679816e+00
4.06465948e-01 -7.41600692e-01 -7.46276259e-01 -2.32823730e-01
-7.20422566e-01 -7.90869445e-02 7.30146706e-01 8.70952681e-02
-1.34495366e+00 -2.90674597e-01 -2.74042845e-01 -2.95071900e-01
-7.01507568e-01 2.54328519e-01 -1.37169793e-01 -1.97773933e-01
8.65059793e-01 8.06239322e-02 2.00558558e-01 -5.01307309e-01
2.30854124e-01 5.84532619e-01 2.18667969e-01 -8.01747739e-01
8.50583911e-01 9.12022173e-01 5.11242449e-01 -1.59285200e+00
-4.41438496e-01 -4.43867326e-01 -6.96534455e-01 5.53267151e-02
8.99937212e-01 -4.83102173e-01 -9.24255908e-01 3.53862613e-01
-9.24808145e-01 -4.88246232e-01 -4.03765023e-01 3.91106814e-01
-6.75187290e-01 2.79819965e-01 -6.42526150e-01 -1.26312399e+00
-2.33063132e-01 -1.24607551e+00 7.12374449e-01 4.50206459e-01
-4.25237894e-01 -1.44066048e+00 9.75669548e-02 3.04020271e-02
5.91425180e-01 4.18881297e-01 1.30925417e+00 -4.22249734e-01
-4.21473771e-01 8.91100466e-02 1.55655026e-01 -2.79052585e-01
3.88949126e-01 -1.84752829e-02 -1.36396182e+00 -3.70070934e-01
8.07074547e-01 -1.58197433e-02 1.04853833e+00 7.13814914e-01
1.32997823e+00 3.27196330e-01 -1.17582405e+00 6.87722147e-01
1.20276594e+00 8.40775371e-01 3.15797269e-01 2.08416268e-01
3.70139569e-01 9.62502003e-01 3.38896245e-01 4.68372166e-01
-3.18947285e-01 3.80275130e-01 -1.71296075e-02 -5.55156291e-01
3.83382380e-01 2.45427907e-01 -1.45361617e-01 5.36542296e-01
-1.22768339e-02 -1.28309339e-01 -1.06798482e+00 2.14828670e-01
-1.59976804e+00 -9.13591564e-01 -3.98088187e-01 1.93716967e+00
2.38570154e-01 2.18151733e-01 6.02388158e-02 -5.64108230e-02
4.82588202e-01 7.49537796e-02 -9.03554022e-01 -5.05993664e-01
-3.51275831e-01 4.05074924e-01 3.19309294e-01 9.76537406e-01
-8.63503397e-01 8.19170654e-01 8.44293785e+00 5.81128716e-01
-1.47051227e+00 -1.07157074e-01 9.09010351e-01 -2.70293236e-01
-8.73561800e-02 -1.86339676e-01 -8.47319603e-01 2.30322376e-01
1.21578503e+00 -5.70907295e-01 4.24373269e-01 7.09083080e-01
6.42874181e-01 -2.83665806e-01 -1.30420041e+00 9.89047050e-01
-1.08711459e-01 -1.65748310e+00 -2.99793575e-03 3.09480608e-01
7.04204977e-01 3.80551517e-01 1.10964835e-01 2.85768539e-01
-1.97343230e-01 -1.36688352e+00 4.47628275e-02 5.32248199e-01
6.11035347e-01 -8.88631582e-01 4.58365232e-01 2.48858958e-01
-8.32010269e-01 -5.00642136e-02 -6.35529160e-01 -5.14011025e-01
7.38521218e-01 1.20492733e+00 -6.99356794e-01 7.44780004e-01
5.71751356e-01 7.58096218e-01 -1.98037531e-02 8.41471434e-01
2.54684150e-01 5.76465249e-01 -3.19223404e-01 -1.21234141e-01
1.79859102e-01 -6.91200137e-01 5.85317552e-01 1.33522213e+00
-5.82349636e-02 5.20910084e-01 1.81939915e-01 1.19945836e+00
5.28301358e-01 -1.10015810e-01 -7.73641348e-01 -1.12006389e-01
2.75571253e-02 1.44636405e+00 -6.25478208e-01 -3.20554942e-01
-4.79922533e-01 1.56069517e-01 6.11997209e-02 4.53203321e-01
-6.09778225e-01 -1.20135257e-02 1.06992292e+00 3.02267522e-01
-1.44379646e-01 -2.60964513e-01 -6.52858853e-01 -8.75984013e-01
-5.72132468e-01 -3.96941990e-01 2.60102123e-01 -4.42804098e-01
-1.73821902e+00 6.97074234e-01 3.58804852e-01 -2.04461068e-01
-1.99306697e-01 -1.12858820e+00 -9.12472129e-01 1.20442677e+00
-1.54533219e+00 -3.32540303e-01 3.50334309e-02 -3.50619070e-02
4.39184368e-01 -6.81671083e-01 9.87700820e-01 -1.55883282e-01
-6.50249362e-01 -1.42159045e-01 5.14341533e-01 -6.53581202e-01
4.64985341e-01 -1.07269216e+00 7.85220444e-01 1.69764832e-01
-1.96695551e-01 5.85928798e-01 1.20549333e+00 -6.24557197e-01
-1.48903608e+00 -1.00733685e+00 4.70464557e-01 -8.43379617e-01
6.23428285e-01 -5.69351494e-01 -1.07225430e+00 4.38045830e-01
2.31283545e-01 -4.18694615e-01 8.84085953e-01 4.74745840e-01
3.06955904e-01 2.01923504e-01 -1.18633091e+00 5.56612372e-01
6.29867017e-01 -3.69485199e-01 -5.29711962e-01 7.76652157e-01
1.10687234e-01 -2.71558255e-01 -6.82089865e-01 8.61716509e-01
5.61878942e-02 -9.49364364e-01 6.51941001e-01 -8.82500768e-01
9.44525599e-02 -2.72307813e-01 2.31713548e-01 -1.15475631e+00
-5.73881507e-01 -7.99959004e-01 -2.90041506e-01 5.03750026e-01
5.76574743e-01 -8.28938067e-01 1.28824925e+00 8.84713590e-01
-2.22797930e-01 -1.12191737e+00 -8.38595152e-01 -6.32685602e-01
9.10475910e-01 -3.72989863e-01 4.90506321e-01 8.38909149e-01
8.64556253e-01 4.83984172e-01 -2.84031332e-01 1.50911897e-01
5.28096914e-01 1.89652279e-01 5.26814580e-01 -1.35627496e+00
-3.01626593e-01 -4.64138478e-01 -3.33509892e-02 -7.21417606e-01
3.45693976e-01 -8.56067955e-01 -1.11016870e-01 -1.21955252e+00
1.47351297e-02 -4.00714397e-01 -2.59813726e-01 -5.14014289e-02
2.67412692e-01 -1.56747550e-01 -3.93187970e-01 8.21565762e-02
1.58097614e-02 5.84321856e-01 9.25377369e-01 -1.61364928e-01
-3.96213949e-01 -1.42463863e-01 -7.88003743e-01 4.41813201e-01
1.31514978e+00 -1.07618226e-02 -3.66882592e-01 6.71252906e-02
-2.39843920e-01 9.41853300e-02 1.86590552e-02 -1.02821183e+00
2.75454253e-01 -5.70061922e-01 5.79188645e-01 -4.76725399e-01
5.77590764e-01 -4.24716473e-01 7.27137849e-02 2.31526375e-01
1.78704932e-01 -1.93698362e-01 6.99319005e-01 3.55738282e-01
-9.24812537e-03 -8.72831792e-02 1.25953996e+00 -4.70262021e-01
-3.42838258e-01 4.61975396e-01 -1.16910911e+00 -3.74456584e-01
1.37558770e+00 1.36455968e-01 -1.12044178e-01 -1.58349879e-03
-7.32928872e-01 1.70162648e-01 3.98111165e-01 5.35090089e-01
-2.37552170e-02 -5.34648895e-01 -5.49548984e-01 4.49407429e-01
9.13584381e-02 -1.42426863e-01 6.00022078e-01 5.95635831e-01
-1.43786669e-01 1.30743313e+00 1.18571497e-01 -7.12086916e-01
-8.14162970e-01 1.03809309e+00 9.83512998e-01 -1.32588567e-02
-6.87485754e-01 5.59273481e-01 5.58455646e-01 -3.14143330e-01
-2.65069246e-01 9.09506828e-02 -7.68003017e-02 -5.57257771e-01
8.13370585e-01 3.15370947e-01 4.27244931e-01 -5.45307457e-01
-2.08177358e-01 4.32722539e-01 -3.62602770e-01 7.08792135e-02
1.28263974e+00 2.41718933e-01 -3.04812789e-01 6.62372708e-01
8.50482404e-01 -2.46563539e-01 -8.38148236e-01 2.47352988e-01
8.22373182e-02 1.08763529e-02 1.00088455e-01 -9.37254727e-01
-7.74261773e-01 1.09840274e+00 4.14831668e-01 2.69769579e-01
6.21014953e-01 1.80422828e-01 4.41894561e-01 4.17961806e-01
5.10206342e-01 -7.22532690e-01 -3.03017080e-01 5.91210067e-01
7.65374303e-01 -1.03288984e+00 -1.78066507e-01 -7.42707431e-01
-2.72375941e-01 1.10178065e+00 6.63961172e-01 9.58293751e-02
1.27177966e+00 7.75189340e-01 2.12399125e-01 -5.61311603e-01
-1.05460095e+00 2.75145203e-01 -2.03812733e-01 9.29691672e-01
6.37683630e-01 -1.22338302e-01 -5.43654860e-05 3.20348620e-01
-3.79115522e-01 -1.23670146e-01 4.55917925e-01 7.54026830e-01
-6.22716010e-01 -1.54945862e+00 -4.07428712e-01 7.43251562e-01
-1.27622947e-01 1.55281082e-01 -2.23139122e-01 5.33966243e-01
2.81950068e-02 1.28335512e+00 1.28130049e-01 1.10548725e-02
3.08967352e-01 3.55643868e-01 4.94671732e-01 -5.29186428e-01
-2.81874716e-01 -2.54896879e-01 -5.76847196e-02 -1.59255475e-01
-1.37827531e-01 -8.81998301e-01 -1.21009529e+00 -3.13005716e-01
-2.47615531e-01 9.66699421e-01 6.38275743e-01 1.22099745e+00
3.05473119e-01 6.96525931e-01 4.33693558e-01 -1.16717720e+00
-6.73387870e-02 -9.02797163e-01 -1.00380659e+00 -3.93402129e-02
2.46355742e-01 -1.00869656e+00 -7.24451900e-01 -2.68273503e-01] | [6.37091588973999, 3.5888671875] |
23ccb018-7522-4b98-a86e-e7d1562064be | low-rank-random-tensor-for-bilinear-pooling | 1906.01004 | null | https://arxiv.org/abs/1906.01004v2 | https://arxiv.org/pdf/1906.01004v2.pdf | Frontal Low-rank Random Tensors for Fine-grained Action Segmentation | Fine-grained action segmentation in long untrimmed videos is an important task for many applications such as surveillance, robotics, and human-computer interaction. To understand subtle and precise actions within a long time period, second-order information (e.g. feature covariance) or higher is reported to be effective in the literature. However, extracting such high-order information is considerably non-trivial. In particular, the dimensionality increases exponentially with the information order, and hence gaining more representation power also increases the computational cost and the risk of overfitting. In this paper, we propose an approach to representing high-order information for temporal action segmentation via a simple yet effective bilinear form. Specifically, our contributions are: (1) From the multilinear perspective, we derive a bilinear form of low complexity, assuming that the three-way tensor has low-rank frontal slices. (2) Rather than learning the tensor entries from data, we sample the entries from different underlying distributions, and prove that the underlying distribution influences the information order. (3) We employed our bilinear form as an intermediate layer in state-of-the-art deep neural networks, enabling to represent high-order information in complex deep models effectively and efficiently. Our experimental results demonstrate that the proposed bilinear form outperforms the previous state-of-the-art methods on the challenging temporal action segmentation task. One can see our project page for data, model and code: \url{https://vlg.inf.ethz.ch/projects/BilinearTCN/}. | ['Yan Zhang', 'Qianli Ma', 'Heiko Neumann', 'Siyu Tang', 'Krikamol Muandet'] | 2019-06-03 | null | null | null | null | ['action-parsing'] | ['natural-language-processing'] | [ 1.53049484e-01 -1.74112022e-01 -2.19714075e-01 -1.29365772e-01
-5.46656251e-01 -4.75554824e-01 3.31570894e-01 -2.40357950e-01
-3.78658831e-01 5.58553040e-01 2.57131517e-01 -2.58402713e-02
-4.23384100e-01 -4.37335759e-01 -9.15608644e-01 -7.69892693e-01
-3.65396768e-01 3.24023068e-01 2.59837389e-01 -6.55154660e-02
6.95662722e-02 4.86292690e-01 -1.32636225e+00 3.85671496e-01
9.57682431e-01 1.39425516e+00 -1.44921750e-01 4.44503307e-01
2.80870825e-01 7.71945119e-01 -2.25127742e-01 -3.52581620e-01
5.83217025e-01 -2.48531774e-01 -9.77787912e-01 2.37581655e-01
5.80267131e-01 -6.18468404e-01 -5.11752903e-01 1.06745934e+00
-3.39443125e-02 2.36461803e-01 4.80855525e-01 -1.22177207e+00
-3.34213048e-01 3.37089986e-01 -6.38164282e-01 -8.87942836e-02
1.24785048e-03 3.42106134e-01 1.08004797e+00 -7.57098377e-01
6.18081450e-01 1.17826891e+00 7.02211618e-01 2.54079461e-01
-1.17324936e+00 -6.42938614e-01 2.65048623e-01 4.64683801e-01
-1.19289756e+00 -1.63038820e-01 8.84806752e-01 -8.15682054e-01
3.64990711e-01 2.74352819e-01 9.06947792e-01 1.25825763e+00
1.77168384e-01 1.03083301e+00 1.29012406e+00 1.89503968e-01
2.12753210e-02 -5.29018342e-01 3.19902360e-01 7.87178874e-01
1.54150769e-01 5.94350062e-02 -5.37429988e-01 -2.69172131e-03
9.28969026e-01 1.36813223e-01 -2.33784780e-01 -4.95221555e-01
-1.50899482e+00 8.09608877e-01 6.32521808e-01 4.17888522e-01
-5.11946201e-01 3.46331209e-01 4.33970869e-01 -8.39617774e-02
4.17814881e-01 3.39235693e-01 -5.11144996e-01 -5.63764334e-01
-6.62196934e-01 4.58867997e-01 5.90579271e-01 6.26433849e-01
8.11689913e-01 -2.14807138e-01 -1.48789331e-01 6.98023677e-01
6.21261597e-02 2.24193826e-01 1.77875683e-01 -1.33720887e+00
5.78650057e-01 4.82621163e-01 7.70823285e-02 -1.20968437e+00
-5.84336758e-01 -4.22105014e-01 -1.24179161e+00 2.05697976e-02
9.23194587e-01 -4.91858348e-02 -7.35458016e-01 1.76387310e+00
3.22349787e-01 6.95816278e-02 -3.58289242e-01 1.07867420e+00
2.65329242e-01 3.03826243e-01 -2.01255098e-01 -7.49405697e-02
1.28082085e+00 -9.22286272e-01 -7.04771340e-01 -9.54836756e-02
5.18440008e-01 -5.51306486e-01 1.01799965e+00 6.80488169e-01
-8.65421593e-01 -4.78203952e-01 -9.60217535e-01 -3.35229844e-01
-1.99262336e-01 3.74852926e-01 9.80997801e-01 6.10157400e-02
-7.78018475e-01 9.64467287e-01 -1.10858953e+00 -1.69742882e-01
6.49315298e-01 2.05599725e-01 -5.34702361e-01 3.44299199e-03
-1.27111137e+00 4.70807701e-01 3.86155576e-01 5.16385138e-01
-6.84893548e-01 -7.11663902e-01 -7.09998071e-01 -1.31105036e-01
7.34274209e-01 -5.49780130e-01 1.15367973e+00 -7.52104163e-01
-1.45944190e+00 4.42810297e-01 9.51521620e-02 -3.64015073e-01
8.26326907e-01 -6.01783633e-01 1.40873909e-01 4.00687158e-01
1.96692552e-02 5.31963587e-01 9.90687847e-01 -9.36954200e-01
-2.53323525e-01 -6.34073377e-01 6.00425959e-01 -5.22908941e-02
-3.43273997e-01 -3.08567852e-01 -3.98431689e-01 -7.95175195e-01
2.61069506e-01 -1.20974314e+00 -3.17203134e-01 3.34794313e-01
-4.15754467e-01 -6.48554191e-02 6.45399272e-01 -9.88673806e-01
1.18999565e+00 -2.10839248e+00 5.07936835e-01 -9.32697579e-03
4.85824734e-01 2.31047511e-01 1.45480614e-02 2.83728421e-01
4.89477627e-03 9.43969861e-02 -4.54600334e-01 -1.59470495e-02
2.66360492e-01 2.33373433e-01 -2.92295609e-02 5.40053129e-01
3.14224035e-01 9.25140679e-01 -9.54467356e-01 -3.53682607e-01
9.73783880e-02 5.59233725e-01 -6.49578810e-01 -1.00728564e-01
-1.03851467e-01 7.24198580e-01 -6.24198675e-01 4.67262536e-01
5.79667270e-01 -3.62200707e-01 9.64213512e-04 -8.02712917e-01
-8.25238675e-02 9.26318094e-02 -1.27334821e+00 1.66631103e+00
-1.96188152e-01 5.65339506e-01 1.41637906e-01 -1.16427326e+00
3.56853127e-01 -7.56574841e-03 9.52671230e-01 -5.21938205e-01
2.26353124e-01 1.54382840e-01 8.24400783e-02 -5.49098492e-01
3.82566720e-01 1.67180955e-01 -1.61368713e-01 1.24569170e-01
7.50563946e-03 1.85597911e-01 5.20460844e-01 7.75518566e-02
1.05532169e+00 3.51455480e-01 1.13438442e-01 -1.36487365e-01
3.91004771e-01 -2.25861356e-01 7.58784473e-01 3.21244210e-01
-2.55879730e-01 3.00092787e-01 8.62737894e-01 -5.25724292e-01
-7.49239862e-01 -8.27258527e-01 -2.03702569e-01 7.75578976e-01
-6.31560013e-02 -3.79563302e-01 -1.00662804e+00 -6.23800874e-01
1.51413307e-02 1.10179156e-01 -7.14570940e-01 -5.50119318e-02
-8.28859389e-01 -5.37867486e-01 4.85355645e-01 6.77529633e-01
7.95289338e-01 -8.13526094e-01 -5.39536417e-01 7.22550824e-02
-4.94416416e-01 -1.45085943e+00 -7.09017158e-01 -1.28729209e-01
-9.36349273e-01 -1.22140598e+00 -8.04443002e-01 -7.40405694e-02
5.70771515e-01 1.11621566e-01 7.93156326e-01 -1.81771517e-01
-2.28117555e-01 3.40933174e-01 -3.60708416e-01 2.92779785e-02
8.21049511e-02 -2.04762891e-02 1.06857359e-01 4.99090225e-01
1.53005019e-01 -7.01898873e-01 -8.05274904e-01 3.52098584e-01
-1.18694913e+00 1.51537329e-01 7.95114517e-01 9.22426283e-01
6.31935954e-01 4.91867438e-02 1.44769803e-01 -5.90188682e-01
4.57295179e-01 -1.59284353e-01 -6.42870605e-01 -6.45641759e-02
3.41031253e-02 2.04129100e-01 6.15944088e-01 -5.47696590e-01
-6.62903845e-01 2.07769498e-01 -9.28892866e-02 -6.31705046e-01
-1.79255009e-01 7.03464866e-01 -1.93711504e-01 -1.39095649e-01
2.50987589e-01 4.59729843e-02 -8.74280408e-02 -6.99564755e-01
3.40650648e-01 1.85109854e-01 3.98038864e-01 -8.21943700e-01
7.60316074e-01 6.58211887e-01 4.55383509e-01 -9.35080349e-01
-9.84202623e-01 -3.60157698e-01 -1.11975789e+00 -3.44511062e-01
1.03122640e+00 -6.71072483e-01 -9.56142545e-01 9.78422225e-01
-1.23946607e+00 -5.13090849e-01 -1.66920677e-01 6.84071839e-01
-6.09487414e-01 5.94117105e-01 -6.46317184e-01 -5.54647326e-01
8.00949708e-02 -1.29302418e+00 1.10117567e+00 -6.46710768e-02
3.72224115e-02 -6.52337730e-01 -1.15410447e-01 7.47268617e-01
6.43898696e-02 7.01806605e-01 6.42178953e-01 -2.51031518e-01
-9.71497178e-01 -5.67500368e-02 -2.69278616e-01 6.16597950e-01
-6.60179257e-02 -9.78628993e-02 -7.19018102e-01 -9.63092074e-02
-4.12267298e-02 -3.46889883e-01 8.73084605e-01 3.38137984e-01
1.41205180e+00 -4.88454252e-01 6.67318180e-02 6.67272866e-01
9.53804016e-01 -1.20357841e-01 5.54679096e-01 1.09854870e-01
1.17384601e+00 8.21595430e-01 8.11809719e-01 5.32970011e-01
4.35559124e-01 9.12249207e-01 4.13140744e-01 8.23424663e-03
1.20110117e-01 -1.42665684e-01 4.95531172e-01 8.69759917e-01
-5.07295072e-01 2.40714803e-01 -7.87331760e-01 4.92860049e-01
-2.11381197e+00 -8.97958994e-01 -3.32490027e-01 2.14172482e+00
8.67835462e-01 3.90778407e-02 3.28426093e-01 1.86590269e-01
2.83422828e-01 4.02075171e-01 -7.34551847e-01 1.82778649e-02
1.31113946e-01 5.72701097e-02 6.33213937e-01 4.07962680e-01
-1.29536712e+00 7.84724116e-01 4.52852011e+00 9.20747101e-01
-1.10230386e+00 1.42969102e-01 4.25667763e-01 -1.12935804e-01
1.11075431e-01 -1.13766998e-01 -6.39795184e-01 4.64038491e-01
6.68014586e-01 8.99548605e-02 5.56576192e-01 4.97357935e-01
3.57213616e-01 -5.05817346e-02 -1.07916260e+00 9.14075553e-01
-1.97396979e-01 -1.09280527e+00 5.75743709e-03 4.42077398e-01
5.77184379e-01 -4.88629155e-02 7.32059404e-02 1.77081570e-01
-3.86857614e-02 -7.17319608e-01 7.49715030e-01 8.16941440e-01
5.93151450e-01 -5.84872842e-01 3.70012760e-01 3.69293153e-01
-1.32050884e+00 -1.64085254e-01 -1.61114499e-01 -9.76261124e-02
2.31102213e-01 9.11977351e-01 -1.85301006e-01 6.86742008e-01
8.10678422e-01 9.84499633e-01 -4.74946916e-01 9.17519093e-01
-9.07107741e-02 5.06327391e-01 -2.94774354e-01 2.08668470e-01
5.42800784e-01 -6.16274834e-01 6.05396211e-01 8.74510527e-01
2.75566071e-01 2.01997221e-01 3.95425856e-01 7.15260088e-01
-6.35747761e-02 -6.11341931e-02 -4.49463874e-01 -3.96620423e-01
-2.22295061e-01 1.16769445e+00 -5.71425974e-01 -5.24854213e-02
-3.39724958e-01 9.23884571e-01 2.84489214e-01 4.68898505e-01
-8.80998731e-01 -2.11649492e-01 9.83616650e-01 1.26867428e-01
4.73335236e-01 -7.65577853e-01 -2.09323645e-01 -1.36999750e+00
6.81308746e-01 -1.01218593e+00 1.34538636e-01 -3.93125355e-01
-1.13097012e+00 3.87760162e-01 2.68322676e-01 -1.35344219e+00
-2.38756955e-01 -9.42830265e-01 2.14003026e-02 4.95811969e-01
-1.21946824e+00 -1.31022489e+00 -2.89589673e-01 6.76531315e-01
4.05922741e-01 3.46307576e-01 4.21454579e-01 5.79117239e-01
-6.00881159e-01 3.62374842e-01 -1.09874248e-03 4.04136389e-01
3.97148401e-01 -1.12272441e+00 1.55883953e-01 8.08452785e-01
1.93284139e-01 5.66402733e-01 3.99342000e-01 -5.01639545e-01
-1.49688911e+00 -1.01861250e+00 5.79433680e-01 -3.99536610e-01
1.20567358e+00 -5.89278936e-01 -8.82225931e-01 5.46193063e-01
-1.62043661e-01 2.11992800e-01 4.31211561e-01 1.27750430e-02
-2.87092924e-01 -3.00857008e-01 -7.96342790e-01 6.44393444e-01
1.24765277e+00 -6.07828200e-01 -1.75050467e-01 5.00439644e-01
6.69155121e-01 -4.76051271e-01 -1.17798150e+00 6.21898115e-01
8.99262965e-01 -1.02430546e+00 9.39535320e-01 -6.05439544e-01
4.95974362e-01 -3.98802310e-01 -1.69755936e-01 -1.03520870e+00
-2.55435824e-01 -8.58068228e-01 -5.04499972e-01 8.23211014e-01
9.32964757e-02 -6.90388799e-01 5.72957098e-01 5.61938345e-01
-1.08460896e-01 -1.28615642e+00 -9.77852046e-01 -8.22920442e-01
9.33130980e-02 -5.49752533e-01 2.63539225e-01 6.70243323e-01
-3.23548824e-01 2.01203488e-02 -6.95380270e-01 1.71807408e-01
6.55004740e-01 1.02843888e-01 7.38575935e-01 -1.26648080e+00
-4.53969449e-01 -4.33799386e-01 -5.47124386e-01 -1.47043025e+00
2.12869912e-01 -4.79934543e-01 -7.71773830e-02 -1.31035578e+00
6.92708045e-02 -2.15511233e-01 -2.80000448e-01 5.01749635e-01
-3.30775678e-02 3.39102358e-01 4.21226025e-01 9.98533964e-02
-5.84905565e-01 7.95210063e-01 1.64084220e+00 -2.10541114e-01
1.40550733e-01 -1.78943481e-02 -4.31275338e-01 9.66592133e-01
6.90348446e-01 -2.63801485e-01 -3.09673995e-01 -3.59356910e-01
1.66748628e-01 -3.10872681e-03 8.49812567e-01 -1.06247723e+00
2.32213903e-02 -2.67574549e-01 1.33859947e-01 -5.57725728e-01
6.59442186e-01 -9.22191501e-01 -1.20231453e-02 3.02965701e-01
-2.12848157e-01 -1.14854589e-01 6.39141276e-02 4.99542624e-01
-3.38744879e-01 2.22563110e-02 5.91135144e-01 -5.82253821e-02
-5.69672883e-01 6.92654133e-01 -1.33724287e-01 2.34641835e-01
7.87935495e-01 -1.09076843e-01 -2.00727522e-01 -3.27067256e-01
-7.51877189e-01 8.69365688e-03 3.27086389e-01 4.64316756e-01
4.05868709e-01 -1.48641479e+00 -4.92229521e-01 9.08500776e-02
-7.53413662e-02 1.06494464e-02 5.08862257e-01 1.52612400e+00
-3.67807418e-01 4.51594234e-01 -3.19450319e-01 -7.60240436e-01
-1.02999866e+00 2.70492911e-01 2.37877265e-01 -5.83987653e-01
-5.18132627e-01 5.92955351e-01 4.01836693e-01 -3.03299159e-01
2.13099375e-01 -6.96141660e-01 -1.53340220e-01 2.30986968e-01
3.90429646e-01 6.00905478e-01 -1.09861515e-01 -9.53360558e-01
-2.57491350e-01 9.57020164e-01 1.35846407e-04 3.46946232e-02
1.31800520e+00 2.24390537e-01 -2.02316046e-01 5.36079943e-01
1.48146439e+00 -2.96050817e-01 -1.64338720e+00 -1.60401389e-01
-1.11671519e-02 -6.12909198e-01 -1.99191179e-03 -3.98981541e-01
-1.30777097e+00 1.07199907e+00 4.64982003e-01 2.28549719e-01
1.25543773e+00 -1.49795726e-01 1.03853929e+00 5.30537248e-01
5.48214436e-01 -1.07869327e+00 2.40720749e-01 6.75695300e-01
1.22705019e+00 -1.09133804e+00 3.60947587e-02 -4.67686772e-01
-6.26317739e-01 1.13896286e+00 4.40346122e-01 -1.27213895e-01
7.75695026e-01 -4.50826325e-02 -9.89708006e-02 -3.15848857e-01
-3.53864998e-01 -2.31284827e-01 5.99162936e-01 3.04779887e-01
2.03999728e-01 1.36689261e-01 -4.59410697e-01 5.58018863e-01
-9.06255990e-02 1.52014717e-01 2.61564106e-01 6.46688223e-01
2.83148978e-02 -1.11622250e+00 -2.07632825e-01 4.44293618e-01
-5.94973266e-01 5.10256849e-02 -3.84743631e-01 6.79927468e-01
2.71094948e-01 7.04597652e-01 -2.28941023e-01 -4.44859862e-01
3.60228539e-01 -1.82154655e-01 5.22524595e-01 -1.04064845e-01
-1.97811037e-01 -8.18118751e-02 1.16063394e-01 -1.27369368e+00
-6.44960344e-01 -8.74615669e-01 -1.10206985e+00 -1.06269144e-01
1.18727721e-01 -2.07488582e-01 5.52469194e-01 1.08460367e+00
3.79174083e-01 4.60788459e-01 4.59974289e-01 -1.21958542e+00
-6.42958343e-01 -9.65118289e-01 -5.64723790e-01 5.06635368e-01
5.09119451e-01 -1.07653761e+00 -3.77622664e-01 1.28905743e-01] | [7.952866077423096, 0.2758459746837616] |
edf8a00d-5ed4-4114-85db-f376f4f43c9a | attention-based-point-cloud-edge-sampling | 2302.14673 | null | https://arxiv.org/abs/2302.14673v2 | https://arxiv.org/pdf/2302.14673v2.pdf | Attention-based Point Cloud Edge Sampling | Point cloud sampling is a less explored research topic for this data representation. The most commonly used sampling methods are still classical random sampling and farthest point sampling. With the development of neural networks, various methods have been proposed to sample point clouds in a task-based learning manner. However, these methods are mostly generative-based, rather than selecting points directly using mathematical statistics. Inspired by the Canny edge detection algorithm for images and with the help of the attention mechanism, this paper proposes a non-generative Attention-based Point cloud Edge Sampling method (APES), which captures salient points in the point cloud outline. Both qualitative and quantitative experimental results show the superior performance of our sampling method on common benchmark tasks. | ['Jürgen Beyerer', 'Julius Pfrommer', 'Junwei Zheng', 'Chengzhi Wu'] | 2023-02-28 | null | http://openaccess.thecvf.com//content/CVPR2023/html/Wu_Attention-Based_Point_Cloud_Edge_Sampling_CVPR_2023_paper.html | http://openaccess.thecvf.com//content/CVPR2023/papers/Wu_Attention-Based_Point_Cloud_Edge_Sampling_CVPR_2023_paper.pdf | cvpr-2023-1 | ['3d-point-cloud-classification', '3d-part-segmentation'] | ['computer-vision', 'computer-vision'] | [-8.45066980e-02 -1.98545069e-01 -9.70005020e-02 -1.13578811e-01
-7.03250051e-01 3.92881781e-02 7.68860936e-01 3.48228551e-02
-1.79322973e-01 4.71242130e-01 -9.21825692e-02 8.56919363e-02
-7.05028847e-02 -1.04697776e+00 -7.68203139e-01 -6.41011119e-01
2.72585690e-01 6.54984176e-01 2.82054722e-01 5.77623025e-02
7.83776700e-01 8.77432168e-01 -1.69997978e+00 3.03432234e-02
1.10231364e+00 9.62978840e-01 4.00312364e-01 2.80900449e-01
-6.48801625e-01 2.70160198e-01 -5.71846664e-01 -3.55039984e-01
2.57858098e-01 -2.94547737e-01 -4.56429303e-01 3.05679172e-01
4.13835019e-01 -2.97069877e-01 7.46125181e-04 1.26534343e+00
4.80212361e-01 2.78167993e-01 9.29702640e-01 -1.26879692e+00
-9.84769106e-01 3.03270996e-01 -8.29905450e-01 4.67845984e-02
2.80837536e-01 6.88571483e-02 9.52036917e-01 -1.52308142e+00
4.80096638e-01 1.48440981e+00 6.76077187e-01 3.43405515e-01
-1.00759709e+00 -6.75644159e-01 -3.70623730e-02 4.33478892e-01
-1.53885126e+00 5.78141920e-02 1.44557762e+00 -4.20729131e-01
5.88290393e-01 2.70523667e-01 9.74050999e-01 7.58289218e-01
5.44493347e-02 1.18376029e+00 1.02085388e+00 -4.18061972e-01
4.82995600e-01 1.02025114e-01 -6.00771680e-02 3.20983827e-01
4.44832951e-01 2.85525769e-01 -1.82479694e-01 -4.47182119e-01
1.12921810e+00 7.52931714e-01 -3.04641694e-01 -5.89337707e-01
-1.05297089e+00 1.08201802e+00 8.80786836e-01 1.32183313e-01
-8.62791061e-01 1.82945833e-01 2.55599827e-01 -2.37646222e-01
8.43216300e-01 1.13572821e-01 2.41247565e-01 1.93425968e-01
-1.26885104e+00 4.47724432e-01 4.64631975e-01 1.43401492e+00
9.00377452e-01 2.44745344e-01 -3.94107491e-01 5.62713087e-01
4.09269303e-01 5.49592972e-01 4.03625697e-01 -5.92790484e-01
4.28614616e-02 8.02384794e-01 3.03361356e-01 -1.26458704e+00
2.07960010e-01 -3.49172354e-01 -9.30368364e-01 6.08673513e-01
-1.45937771e-01 1.17970198e-01 -8.22579265e-01 8.43841553e-01
4.46781754e-01 3.03495973e-01 -3.82297188e-01 9.67256188e-01
7.66930461e-01 6.43787205e-01 4.12345529e-02 1.19640507e-01
1.00748694e+00 -7.96357751e-01 -4.55472380e-01 1.69367358e-01
-2.63877124e-01 -6.65630102e-01 1.18970120e+00 5.05572021e-01
-1.05559611e+00 -6.89293206e-01 -8.79038453e-01 -8.67149979e-02
-1.97421342e-01 1.56326547e-01 6.35599792e-01 3.13557208e-01
-7.90200531e-01 7.54125953e-01 -7.15443730e-01 -2.17812300e-01
1.01304257e+00 4.59460430e-02 1.56652614e-01 -1.93942770e-01
-4.96880263e-01 4.14488077e-01 2.61510313e-01 -1.01531222e-01
-6.38781846e-01 -5.34516037e-01 -5.46320081e-01 3.30577195e-01
8.21725503e-02 -8.10320079e-01 9.53955472e-01 -8.92179191e-01
-1.62305093e+00 5.95886648e-01 -3.43753487e-01 -4.65275258e-01
5.62993228e-01 -4.64100808e-01 2.56147146e-01 2.99275637e-01
1.37272179e-01 9.05713916e-01 1.15250683e+00 -1.52420020e+00
-6.44206285e-01 -2.61853755e-01 -3.80775094e-01 2.60956347e-01
-4.70937155e-02 -3.01710851e-02 -3.24557453e-01 -5.66853464e-01
2.38114133e-01 -6.64760113e-01 -3.75649631e-01 2.06179604e-01
-4.13106918e-01 -8.87925744e-01 1.09750915e+00 -2.79150605e-01
7.75533319e-01 -2.02055454e+00 -1.45019710e-01 4.29025702e-02
2.54350185e-01 2.28573978e-01 3.45915675e-01 4.33673263e-01
2.31960565e-02 9.31085497e-02 -2.12508574e-01 -4.01229978e-01
2.38715839e-02 -2.70934284e-01 -5.44495761e-01 3.67610931e-01
3.88743013e-01 9.00464475e-01 -1.02254367e+00 -5.85195422e-01
6.73892736e-01 6.58000946e-01 -5.48068166e-01 1.16383597e-01
-3.13924402e-01 2.64130533e-01 -7.11542606e-01 8.26088965e-01
9.08721268e-01 -1.27878875e-01 -6.74455881e-01 -1.48462355e-01
-6.16293550e-02 -1.55799136e-01 -9.17322397e-01 1.49396086e+00
-2.80993491e-01 6.79661632e-01 -3.22208375e-01 -6.01789176e-01
1.29124045e+00 7.45288059e-02 4.84336078e-01 3.81478705e-02
2.60342002e-01 1.73099801e-01 -2.01423034e-01 -2.90119916e-01
7.49682844e-01 -2.02063471e-01 3.75804245e-01 -6.54121935e-02
-3.03078264e-01 -5.10859549e-01 -7.53695667e-02 -8.51542354e-02
6.27555847e-01 4.05587375e-01 3.82721156e-01 -3.83844048e-01
3.14060599e-01 3.30354691e-01 4.21487123e-01 5.49952745e-01
-4.95121889e-02 1.21216905e+00 -3.83261442e-02 -6.20317101e-01
-1.12119198e+00 -8.04236114e-01 -2.59729147e-01 5.23729861e-01
3.19065958e-01 -1.64062262e-01 -6.90742671e-01 -5.22519231e-01
1.80027574e-01 8.63035500e-01 -4.86382753e-01 3.09895650e-02
-3.74834061e-01 -2.50342280e-01 1.01163965e-02 5.59656203e-01
6.22942388e-01 -1.46012449e+00 -8.22047949e-01 2.46334106e-01
5.62146828e-02 -6.46460950e-01 -3.77239734e-01 -2.40014598e-01
-1.23771095e+00 -7.71850348e-01 -1.22733617e+00 -6.21916592e-01
9.59715962e-01 6.13674760e-01 1.26985013e+00 1.50948882e-01
-7.58814961e-02 2.82624304e-01 -3.65870565e-01 -9.59855914e-01
1.32134050e-01 -1.22562144e-02 -4.03678626e-01 1.49180636e-01
9.11852837e-01 -5.23293197e-01 -7.76593149e-01 6.79304823e-02
-8.31781328e-01 -2.83391718e-02 9.81405616e-01 7.51234412e-01
1.01820803e+00 -9.87678766e-02 3.49402159e-01 -8.19727838e-01
8.29331517e-01 -4.50239658e-01 -5.97822309e-01 -2.72549927e-01
-3.46991271e-01 -9.09228846e-02 5.31181037e-01 -3.74564707e-01
-6.61700130e-01 -7.65281022e-02 -8.83360878e-02 -1.28921914e+00
-4.72841024e-01 1.86083406e-01 -5.69152161e-02 -1.64993584e-01
5.19415796e-01 6.81000292e-01 -1.12379707e-01 -4.04817820e-01
1.51612923e-01 7.03201532e-01 2.04056844e-01 -4.65490937e-01
7.71290243e-01 9.11468744e-01 -7.56965019e-03 -1.07921374e+00
-3.23034286e-01 -5.72420657e-01 -3.47091287e-01 -3.09773505e-01
8.25528145e-01 -5.86660385e-01 -5.94139755e-01 1.89734206e-01
-1.33397043e+00 2.28587285e-01 -4.73461390e-01 3.56199980e-01
-6.21599793e-01 4.91224051e-01 -2.42470741e-01 -1.16069102e+00
-8.36015046e-01 -1.16047633e+00 1.62735081e+00 3.25306594e-01
8.23920667e-02 -6.50321841e-01 -7.07415193e-02 -2.47602999e-01
4.47716504e-01 4.58697557e-01 6.01067901e-01 -4.08904940e-01
-1.11577356e+00 -5.74557483e-01 -3.53565186e-01 1.44766003e-01
-1.28834359e-02 1.51558980e-01 -8.32396507e-01 -1.54050156e-01
6.57387599e-02 1.84565470e-01 6.78575158e-01 7.37431705e-01
1.47688639e+00 -1.27310634e-01 -6.83710575e-01 6.64891124e-01
1.73430550e+00 2.72482514e-01 9.41074789e-01 2.59207338e-01
7.65913785e-01 2.23746300e-01 5.88714600e-01 4.30907100e-01
9.06184092e-02 4.15967047e-01 6.64403796e-01 -2.70678867e-02
-3.02882493e-02 -4.02020305e-01 -1.98868826e-01 5.54638684e-01
-4.03249979e-01 -4.31859642e-02 -7.12119579e-01 6.98231578e-01
-1.62866735e+00 -1.08654308e+00 -1.59633979e-01 2.17395568e+00
3.65836501e-01 2.27786332e-01 1.26678735e-01 9.73922089e-02
8.70867491e-01 -8.91624659e-04 -5.30590177e-01 -1.62836835e-01
2.73043871e-01 2.34609038e-01 4.45035934e-01 -6.01404440e-03
-9.25600886e-01 7.53175378e-01 5.87712145e+00 1.06394279e+00
-1.06856370e+00 -1.25778034e-01 5.50592899e-01 2.39223167e-01
-4.41756994e-01 -8.00164789e-02 -9.30272818e-01 8.43576670e-01
2.85456955e-01 -1.45379603e-01 1.35355279e-01 1.46349871e+00
1.51357159e-01 8.44511464e-02 -7.73131371e-01 1.30630517e+00
3.38570178e-02 -1.37575364e+00 4.06838894e-01 3.69718820e-02
7.56897211e-01 4.95699309e-02 1.00235117e-03 3.67311537e-01
2.05548838e-01 -8.17524254e-01 7.48247325e-01 8.62469494e-01
5.72714865e-01 -9.58488524e-01 6.71588421e-01 3.75097156e-01
-1.07251608e+00 2.45052233e-01 -9.70829070e-01 9.53681767e-02
6.62641041e-03 8.23040068e-01 -7.93695927e-01 5.44553459e-01
8.87385428e-01 7.72741914e-01 -3.46238136e-01 1.77605045e+00
-7.60229155e-02 6.26416326e-01 -3.70981842e-01 -5.87022781e-01
3.12919438e-01 -6.15339637e-01 9.35924530e-01 9.90563571e-01
5.32419860e-01 -1.74984619e-01 1.19467400e-01 1.52819514e+00
-4.93554361e-02 2.27691814e-01 -9.70843494e-01 2.78084129e-01
6.48576379e-01 1.27211237e+00 -9.63304400e-01 -5.29573500e-01
-2.92782605e-01 7.51216412e-01 1.76601589e-01 2.73660868e-01
-6.25507772e-01 -6.56091273e-01 3.75597358e-01 3.91697466e-01
6.11577570e-01 -3.10432881e-01 -3.76168549e-01 -7.11666346e-01
-6.19139299e-02 -4.80874538e-01 -2.71258295e-01 -1.14054334e+00
-1.40855455e+00 7.17805088e-01 3.26795816e-01 -1.87416363e+00
-4.16965932e-02 -3.06171656e-01 -1.07238650e+00 1.06723583e+00
-1.48980224e+00 -9.22288239e-01 -6.08784497e-01 2.84584731e-01
1.10007012e+00 -1.26973823e-01 3.84387404e-01 9.01026949e-02
-9.06559378e-02 -4.33266461e-02 2.19466165e-01 8.20541382e-03
9.10376981e-02 -1.36976469e+00 6.19316697e-01 7.13227272e-01
1.93373293e-01 7.44594157e-01 5.63387454e-01 -7.24323988e-01
-1.23288345e+00 -1.09889400e+00 4.00105447e-01 -2.14514375e-01
-1.52954003e-02 -2.12714702e-01 -1.01389098e+00 3.89590830e-01
4.36970741e-01 9.43358988e-02 8.14555585e-02 -3.68542433e-01
2.99917817e-01 1.73016354e-01 -1.20357704e+00 7.07700551e-01
8.52178037e-01 -3.38233300e-02 -7.73028433e-01 3.32706630e-01
5.49601018e-01 -3.67263347e-01 -4.65750635e-01 3.49799484e-01
1.31224953e-02 -1.01459348e+00 1.11510432e+00 -8.08445662e-02
4.69924390e-01 -4.11640763e-01 1.65759772e-01 -1.45230484e+00
-6.84615433e-01 -4.21089351e-01 -2.06515603e-02 1.07613468e+00
-2.74865896e-01 -5.73587894e-01 1.17940629e+00 1.41709358e-01
-3.56572181e-01 -9.51542497e-01 -6.50648534e-01 -6.63214624e-01
-8.59406441e-02 -1.77092239e-01 8.83315921e-01 6.16287291e-01
-5.74201703e-01 2.67138243e-01 7.80248865e-02 3.10894046e-02
8.28342915e-01 4.45893794e-01 8.81870270e-01 -1.63042271e+00
1.16417758e-01 -8.20801318e-01 -6.23787463e-01 -1.18072104e+00
-1.77247137e-01 -7.99225509e-01 8.73019248e-02 -1.71233368e+00
-1.67090580e-01 -3.64476591e-01 1.75658807e-01 -2.46941239e-01
-4.55545574e-01 -1.10958681e-01 9.25420299e-02 6.69830918e-01
-3.12850207e-01 1.00942767e+00 1.22775161e+00 -1.57217383e-01
-1.07940212e-01 2.00960532e-01 -4.50393260e-01 8.61644864e-01
6.29718661e-01 -5.60358703e-01 -4.31386799e-01 -3.01519185e-01
-9.82780680e-02 -2.41620183e-01 4.66785401e-01 -1.38936806e+00
2.04313055e-01 -1.04115680e-01 5.30075192e-01 -1.33453953e+00
2.69178510e-01 -1.05076563e+00 4.75370139e-02 3.71801168e-01
3.14695537e-02 1.38826773e-01 -1.64419897e-02 8.85930538e-01
-3.16237628e-01 -4.62500513e-01 7.66748369e-01 -4.05019283e-01
-6.77062035e-01 6.09623790e-01 2.18121260e-02 -2.06093669e-01
1.01762295e+00 -6.16836846e-01 2.76446432e-01 -2.22081602e-01
-3.08511794e-01 -2.00167999e-01 5.12382507e-01 1.05241969e-01
1.12720382e+00 -1.65976477e+00 -8.02410603e-01 2.64998108e-01
4.36136760e-02 6.02377057e-01 -4.00947556e-02 5.60687661e-01
-7.60715008e-01 2.81197250e-01 -1.85047597e-01 -1.08858943e+00
-7.48738289e-01 8.37301433e-01 8.99178758e-02 2.48158664e-01
-9.81500685e-01 7.08342969e-01 1.65951148e-01 -1.47525921e-01
2.39350870e-02 -6.79318368e-01 -2.51756251e-01 -3.26450616e-01
3.45050156e-01 5.34097910e-01 4.70063984e-02 -3.84171844e-01
-3.71745937e-02 6.90371156e-01 1.09888598e-01 1.82173878e-01
1.26546788e+00 2.48750150e-01 1.41847074e-01 3.87631059e-01
9.02652800e-01 -1.32480226e-02 -1.33425438e+00 -2.66346306e-01
-1.88091230e-02 -9.28312838e-01 2.28106931e-01 -8.58864188e-02
-1.06335914e+00 1.01842046e+00 4.85784084e-01 6.27508938e-01
8.19509864e-01 -2.32313201e-01 7.08397806e-01 1.86251868e-02
7.93100476e-01 -7.40374207e-01 -1.05919868e-01 1.74518198e-01
1.23400903e+00 -1.25399721e+00 9.41374227e-02 -5.11268377e-01
-5.41801810e-01 9.86067355e-01 5.43899775e-01 -9.75267053e-01
5.90693057e-01 3.75912078e-02 -4.05356914e-01 -4.22158420e-01
-4.04675514e-01 -2.03253269e-01 3.65295351e-01 7.86496937e-01
2.69813836e-01 -2.85978783e-02 -2.33662814e-01 2.66712219e-01
-9.99552533e-02 2.20244691e-01 1.50025576e-01 8.24231803e-01
-6.75342143e-01 -5.68488836e-01 -7.48010457e-01 7.24939823e-01
-1.52191520e-01 -1.34094492e-01 -2.73583382e-01 8.62687767e-01
-1.32433683e-01 4.33398455e-01 3.01867127e-01 -1.13455936e-01
3.52067769e-01 -7.58435950e-02 2.18939364e-01 -6.14246070e-01
-3.84853452e-01 1.57328486e-01 -7.85706580e-01 -1.55838400e-01
-3.93075347e-01 -7.47474194e-01 -1.13959754e+00 -1.64524421e-01
-6.99194252e-01 2.64274895e-01 6.78292155e-01 4.82364357e-01
6.95531726e-01 5.50689042e-01 5.87652564e-01 -1.69435644e+00
-5.26978016e-01 -1.23146403e+00 -4.39622372e-01 4.91387635e-01
4.35815603e-02 -9.70434129e-01 -3.39259356e-01 -3.15483570e-01] | [8.248215675354004, -3.5071539878845215] |
d700fef9-e661-48ef-8aa4-6cb0e45d38b9 | deep-image-prior-inpainting-of-ancient | 2306.14209 | null | https://arxiv.org/abs/2306.14209v1 | https://arxiv.org/pdf/2306.14209v1.pdf | Deep image prior inpainting of ancient frescoes in the Mediterranean Alpine arc | The unprecedented success of image reconstruction approaches based on deep neural networks has revolutionised both the processing and the analysis paradigms in several applied disciplines. In the field of digital humanities, the task of digital reconstruction of ancient frescoes is particularly challenging due to the scarce amount of available training data caused by ageing, wear, tear and retouching over time. To overcome these difficulties, we consider the Deep Image Prior (DIP) inpainting approach which computes appropriate reconstructions by relying on the progressive updating of an untrained convolutional neural network so as to match the reliable piece of information in the image at hand while promoting regularisation elsewhere. In comparison with state-of-the-art approaches (based on variational/PDEs and patch-based methods), DIP-based inpainting reduces artefacts and better adapts to contextual/non-local information, thus providing a valuable and effective tool for art historians. As a case study, we apply such approach to reconstruct missing image contents in a dataset of highly damaged digital images of medieval paintings located into several chapels in the Mediterranean Alpine Arc and provide a detailed description on how visible and invisible (e.g., infrared) information can be integrated for identifying and reconstructing damaged image regions. | ['Rosa Maria Dessì', 'Luca Calatroni', 'Elena Loli Piccolomini', 'Elena Morotti', 'Oceane Acquier', 'Perrine Saillard', 'Fabio Merizzi'] | 2023-06-25 | null | null | null | null | ['image-reconstruction'] | ['computer-vision'] | [ 6.47519469e-01 -1.01839140e-01 4.26516891e-01 1.12635046e-01
-5.57037234e-01 -2.86206663e-01 6.95774376e-01 2.47885644e-01
-5.10960996e-01 8.76171470e-01 1.65577978e-01 1.43430784e-01
-3.30031544e-01 -1.06234384e+00 -6.99308872e-01 -8.12010884e-01
3.04242790e-01 3.57654691e-01 8.22942108e-02 -4.30547982e-01
4.09069657e-01 9.45728898e-01 -1.83880997e+00 5.03243983e-01
6.73396707e-01 9.39755440e-01 4.43981588e-01 5.17835259e-01
-1.25143364e-01 5.16702354e-01 -5.87549269e-01 -4.48152184e-01
1.76235169e-01 -5.03108203e-01 -4.43212450e-01 1.49658009e-01
5.33538461e-01 -4.22976971e-01 -5.46432614e-01 6.87981009e-01
5.52419603e-01 2.55145967e-01 6.49219096e-01 -3.63795698e-01
-5.61092973e-01 -1.26668394e-01 -5.13177037e-01 2.11514711e-01
3.99294734e-01 4.80794199e-02 4.23575580e-01 -8.19081903e-01
9.85248148e-01 1.02458787e+00 8.77413809e-01 3.08879942e-01
-1.62944460e+00 -1.54782981e-01 -5.31790972e-01 5.46689034e-01
-9.84070003e-01 -5.48787951e-01 1.17831326e+00 -4.33096975e-01
5.42676270e-01 3.75004768e-01 7.81205654e-01 1.17989945e+00
2.63713419e-01 4.95576501e-01 1.12325263e+00 -6.36243343e-01
3.28418016e-01 -2.22779572e-01 -7.31929183e-01 2.63755977e-01
-2.61375219e-01 3.28477740e-01 -6.20159686e-01 1.95380691e-02
1.04244196e+00 2.53489107e-01 -4.90939140e-01 -1.59962066e-02
-9.80867326e-01 5.75660229e-01 4.71321404e-01 5.96999526e-01
-8.56572032e-01 8.23618770e-02 3.58957708e-01 3.09126139e-01
7.45008826e-01 2.36738384e-01 -1.37430876e-01 1.15307476e-02
-1.60723543e+00 3.11349154e-01 3.77136707e-01 5.55882603e-02
8.43518436e-01 2.72979826e-01 1.18294753e-01 1.03165078e+00
2.86639053e-02 2.07688570e-01 1.56891495e-01 -1.18054545e+00
1.52188003e-01 3.81832719e-01 4.57153209e-02 -1.19928074e+00
-1.54384226e-01 -2.30293617e-01 -1.15390301e+00 7.98106372e-01
3.31644356e-01 4.51508492e-01 -8.92279148e-01 1.16335940e+00
4.72114623e-01 -1.07798547e-01 -1.47220567e-01 8.25450659e-01
5.31049311e-01 6.63430214e-01 -6.87033534e-02 -3.84543836e-01
1.17626524e+00 -3.73269916e-01 -7.51241803e-01 -2.27889732e-01
-2.47712865e-01 -9.82736349e-01 8.80541503e-01 7.65024781e-01
-1.29165578e+00 -6.26458645e-01 -9.28949356e-01 -3.59737813e-01
-1.77993178e-01 -1.40152097e-01 -2.01131906e-02 2.43789077e-01
-1.06117964e+00 1.19935858e+00 -6.01548731e-01 -2.94610262e-01
7.33106613e-01 2.24960968e-02 -6.13416433e-01 -3.75956953e-01
-8.71591091e-01 9.60076332e-01 2.70862579e-01 5.35653949e-01
-8.74630094e-01 -9.29252684e-01 -6.12287164e-01 -1.17298737e-01
1.66735470e-01 -4.83993292e-01 6.93641424e-01 -1.36308265e+00
-1.54267144e+00 9.97135043e-01 3.04672152e-01 -3.55414867e-01
1.02674353e+00 -8.08192119e-02 -2.77662605e-01 4.17245477e-01
-1.25314891e-01 2.50390917e-01 1.10742593e+00 -1.52547097e+00
-6.98158741e-02 -4.50078011e-01 -3.39419574e-01 -1.61544442e-01
3.66649404e-02 -4.69956361e-02 -2.51907885e-01 -1.06878054e+00
1.88129991e-01 -3.20875049e-01 6.42609075e-02 6.42202020e-01
-5.37826270e-02 4.43540007e-01 9.56262946e-01 -1.64523757e+00
7.76747227e-01 -2.09117174e+00 5.58078110e-01 4.77801412e-02
5.51620647e-02 4.78468955e-01 6.52951188e-03 7.98234284e-01
-5.62486202e-02 -3.80634993e-01 -1.11462665e+00 -7.03853726e-01
-3.13237190e-01 5.25302827e-01 -1.88166648e-01 7.97587752e-01
4.95168678e-02 6.22199178e-01 -6.81859553e-01 -4.25442666e-01
7.09978998e-01 1.27655602e+00 -1.41683906e-01 2.42674366e-01
-4.14357901e-01 9.12520707e-01 4.37317975e-02 7.12913871e-01
9.09600258e-01 4.54597682e-01 -4.80936952e-02 -2.93655485e-01
-4.24216598e-01 -1.10824332e-01 -9.76871669e-01 2.09704804e+00
-6.82106376e-01 9.11237359e-01 6.29949927e-01 -1.13323498e+00
9.41818178e-01 3.81525695e-01 4.17172611e-01 -1.19990063e+00
3.26775685e-02 3.64456177e-01 -6.21175051e-01 -7.30000973e-01
5.80983162e-01 -6.89061940e-01 5.21340132e-01 3.62606943e-01
-2.16634035e-01 -2.26943538e-01 -1.87955901e-01 -1.61872536e-01
1.05563998e+00 6.35533214e-01 -3.23643014e-02 4.76623653e-03
7.20797658e-01 -1.92583352e-01 2.53493011e-01 1.99145734e-01
1.41336888e-01 1.28505671e+00 1.76183894e-01 -7.21358001e-01
-1.58096659e+00 -8.38324428e-01 -3.24405938e-01 4.37564552e-01
-3.56741846e-01 2.12639961e-02 -7.74279356e-01 -4.02031541e-02
-1.60049483e-01 5.91105580e-01 -9.73589599e-01 -1.01625286e-02
-8.30587029e-01 -4.85306531e-01 2.31667578e-01 1.43876702e-01
7.22850323e-01 -1.64012647e+00 -8.80757689e-01 6.40022874e-01
-2.84744442e-01 -8.03411961e-01 1.21612892e-01 -2.16997385e-01
-9.31306005e-01 -1.02936077e+00 -1.10742080e+00 -4.05772090e-01
5.12723029e-01 -7.39903897e-02 1.17945290e+00 3.46575767e-01
-6.02410316e-01 4.24279898e-01 -2.89892852e-01 6.73209280e-02
-8.25227439e-01 -3.85605276e-01 -5.36631703e-01 3.72809768e-01
-4.03262377e-01 -1.02824974e+00 -8.49596262e-01 -1.47889592e-02
-1.53549063e+00 -2.60373875e-02 5.14957190e-01 8.35109830e-01
5.92256725e-01 1.45166844e-01 1.60454243e-01 -6.19493604e-01
3.89405727e-01 -4.10842836e-01 -4.48929995e-01 1.58448428e-01
-2.71552294e-01 -8.65505934e-02 5.38845539e-01 -3.06325495e-01
-1.29823375e+00 -1.91110261e-02 -4.40623641e-01 -4.73564208e-01
-2.26078764e-01 3.67494404e-01 -7.28784800e-02 -1.38622269e-01
6.88728094e-01 4.01721060e-01 1.67392895e-01 -8.56511235e-01
1.40755087e-01 4.77651924e-01 9.35263932e-01 -3.86370301e-01
6.33600533e-01 8.82390261e-01 2.65917212e-01 -1.13340664e+00
-2.53398120e-01 -3.66812870e-02 -8.99929285e-01 -6.55612588e-01
8.53920639e-01 -3.95799190e-01 -4.89022881e-01 6.30603790e-01
-1.24064708e+00 -4.25706744e-01 -6.92607760e-01 -6.36915937e-02
-6.70148313e-01 7.26145983e-01 -5.45631468e-01 -9.24969912e-01
-3.77440572e-01 -9.02284563e-01 1.18418527e+00 -3.67182903e-02
1.39508191e-02 -1.02278745e+00 3.56879592e-01 4.56239313e-01
5.24347186e-01 1.05797410e+00 8.28318596e-01 4.41766530e-01
-5.37197530e-01 -2.46772721e-01 3.42216194e-02 8.01821113e-01
-2.81140245e-02 -1.15236863e-01 -1.12847531e+00 -1.01081386e-01
3.39899778e-01 -1.10739945e-02 9.73357379e-01 1.80249602e-01
7.93885708e-01 -2.06235290e-01 1.65266559e-01 4.53016311e-01
1.76584148e+00 -1.58140361e-01 1.33463240e+00 6.61259949e-01
5.03766119e-01 8.75950515e-01 1.61577612e-01 5.55941105e-01
-6.03393242e-02 8.99904311e-01 7.17717290e-01 -1.84066728e-01
-6.19351745e-01 -4.94814068e-02 1.76877379e-01 2.72989661e-01
-6.42830253e-01 5.85304201e-03 -6.47524953e-01 7.04522014e-01
-1.61015224e+00 -1.01455426e+00 -4.08899248e-01 2.46563721e+00
8.77431154e-01 -2.78322935e-01 -1.29500002e-01 6.55032158e-01
6.98339462e-01 2.21330792e-01 -2.17224643e-01 -3.33531618e-01
-5.69251537e-01 6.09044075e-01 1.60865605e-01 5.21116614e-01
-5.88412046e-01 4.35587227e-01 5.44161654e+00 9.42261100e-01
-9.49534655e-01 3.45504105e-01 4.65390921e-01 1.40384324e-02
-3.25579405e-01 -6.27722219e-02 2.28148565e-01 6.27865791e-01
9.57354903e-01 6.34969473e-01 6.65927172e-01 1.57064110e-01
4.29719508e-01 -5.85050464e-01 -7.31614649e-01 1.00495994e+00
2.42203310e-01 -1.56267369e+00 -2.05321595e-01 1.75708771e-01
6.48730814e-01 -1.72452539e-01 -1.46764487e-01 -4.24696982e-01
-4.99199182e-01 -9.32066679e-01 1.12072897e+00 1.15957618e+00
8.10581386e-01 -7.82420278e-01 4.58585918e-01 2.34913692e-01
-6.42376304e-01 -5.05040661e-02 -2.71310121e-01 1.15627460e-02
4.83047783e-01 9.91325021e-01 -1.40165538e-01 7.59953141e-01
8.01027238e-01 6.62495255e-01 -3.90724719e-01 9.63657856e-01
-3.20710868e-01 2.40962386e-01 -2.80802846e-01 8.01538885e-01
2.08392143e-02 -5.79137444e-01 6.24157429e-01 8.21933150e-01
4.66910303e-01 -1.11468975e-02 -6.13292933e-01 1.17570424e+00
3.55175324e-03 2.00975500e-03 -3.72287095e-01 3.41278285e-01
-1.61384508e-01 1.10910749e+00 -7.60644615e-01 -1.09685637e-01
-9.43399370e-02 1.42213511e+00 1.46663040e-01 2.60898560e-01
-2.37141311e-01 2.52800751e-02 1.21681221e-01 7.22745895e-01
4.74191934e-01 -3.66737038e-01 -2.96781242e-01 -7.96455979e-01
2.52804190e-01 -6.46608353e-01 1.86043859e-01 -1.00364280e+00
-1.13196254e+00 5.96595645e-01 -2.28521258e-01 -1.02161741e+00
5.51426075e-02 -3.13548952e-01 -6.39786959e-01 8.74638319e-01
-1.58593118e+00 -1.35543919e+00 -3.85725975e-01 4.68978703e-01
5.52591801e-01 1.09198786e-01 7.92669177e-01 6.44221306e-01
-3.05764496e-01 -1.31579638e-01 5.81722200e-01 -3.21159124e-01
5.18144131e-01 -8.63847852e-01 1.03586510e-01 8.61969471e-01
-7.91411698e-02 -6.05872758e-02 1.04820383e+00 -6.67430878e-01
-1.22269249e+00 -6.27125978e-01 6.75289392e-01 -1.67248592e-01
2.57799029e-01 -1.14654087e-01 -1.34452772e+00 3.29228133e-01
3.36354345e-01 1.43060973e-02 1.28829896e-01 -5.97008228e-01
-1.15514196e-01 -1.39369100e-01 -1.59919047e+00 2.73987472e-01
5.88508368e-01 -5.70380926e-01 -4.17804837e-01 2.52026677e-01
-1.13930598e-01 -1.96415931e-01 -9.95990038e-01 1.07717730e-01
5.86731553e-01 -1.59840584e+00 1.13504505e+00 2.74257421e-01
7.69590020e-01 -2.43255407e-01 -1.00341327e-01 -1.07878482e+00
2.95779686e-02 -6.99759781e-01 -6.08499497e-02 1.23107457e+00
-3.12640727e-01 -3.51808906e-01 5.77667356e-01 4.74911213e-01
-2.08276093e-01 -4.62730020e-01 -1.39434361e+00 -2.61202931e-01
3.52028087e-02 -3.72878760e-01 3.18890303e-01 7.26825356e-01
-8.02309632e-01 -2.98310786e-01 -6.68501377e-01 -2.52392381e-01
8.05616498e-01 -1.00530133e-01 5.29060900e-01 -1.37882590e+00
-3.09036672e-01 -3.68330538e-01 -4.20778751e-01 -2.73491770e-01
1.41665526e-03 -5.38808227e-01 -3.21015641e-02 -1.67018962e+00
-2.97046542e-01 -3.00498545e-01 1.37011141e-01 2.64258295e-01
2.70195097e-01 8.75719547e-01 1.28650069e-01 4.91339326e-01
1.89458475e-01 7.45559931e-01 1.21770000e+00 -9.92420539e-02
5.52268326e-02 -2.04362094e-01 -2.47297868e-01 4.16514844e-01
4.82785344e-01 -4.35361892e-01 1.19528189e-01 -5.21007359e-01
5.17860651e-01 7.42165148e-02 8.86427224e-01 -1.14265919e+00
2.87781730e-02 2.02668190e-01 6.42274499e-01 -5.44658124e-01
6.17932856e-01 -1.12803173e+00 8.26353550e-01 4.22171295e-01
5.21851704e-03 -1.50756091e-01 2.67692029e-01 6.40335321e-01
-2.97553182e-01 -3.19016695e-01 1.13988447e+00 -4.44703400e-01
-5.08868873e-01 3.81162688e-02 -1.99954912e-01 -4.52489704e-01
7.51961648e-01 -5.04610002e-01 8.01585615e-03 -3.02755445e-01
-7.61491060e-01 -5.62910795e-01 8.53707254e-01 4.00253758e-03
8.71281385e-01 -1.17671490e+00 -9.28061545e-01 3.51044863e-01
-2.56650746e-01 2.52372295e-01 9.36221123e-01 8.97235870e-01
-1.12935424e+00 -3.44294071e-01 -6.32784605e-01 -4.46197867e-01
-9.88028884e-01 5.01714826e-01 4.65912670e-01 -1.75990053e-02
-1.28412783e+00 4.01657820e-01 -2.48204887e-01 -5.46052568e-02
-4.31559756e-02 1.71916872e-01 -9.34084654e-02 2.94415116e-01
6.01753712e-01 6.02663934e-01 5.84358931e-01 -9.20469820e-01
-1.84776578e-02 7.62656927e-01 2.56310135e-01 -3.15412045e-01
1.95000553e+00 -1.07369095e-01 -4.85468686e-01 3.81288886e-01
9.27715778e-01 -1.37383237e-01 -1.66997790e+00 -1.38873041e-01
-2.76671290e-01 -7.90015101e-01 5.69612026e-01 -8.38029504e-01
-1.32999051e+00 1.02376580e+00 7.88827777e-01 2.29669243e-01
1.39501035e+00 -1.26507074e-01 8.36971998e-01 -3.50374609e-01
1.61151737e-01 -1.19295454e+00 -7.03903213e-02 -1.84990674e-01
1.60953975e+00 -8.60226750e-01 4.68015343e-01 5.42869270e-02
-2.64423728e-01 1.42642450e+00 -1.48685187e-01 -2.50976980e-01
4.70783383e-01 -3.94861922e-02 -8.67296904e-02 -2.39615470e-01
-7.71611780e-02 7.27060884e-02 1.76932201e-01 7.68443167e-01
1.12530656e-01 -4.25161332e-01 -3.38446409e-01 -4.21950705e-02
9.46184918e-02 1.60997212e-02 5.51387250e-01 1.15231240e+00
-8.48902762e-02 -1.20557332e+00 -9.25405204e-01 3.34783755e-02
-3.32393467e-01 -9.04125348e-03 -2.45283991e-01 6.72146082e-01
4.46792126e-01 5.69851458e-01 1.09891959e-01 8.26013610e-02
4.65317160e-01 -1.42437831e-01 8.42078447e-01 -1.07733570e-01
-7.88892388e-01 2.14332059e-01 -1.92806855e-01 -4.66035336e-01
-8.56346250e-01 -8.83671403e-01 -8.55202436e-01 -4.13203955e-01
4.79224995e-02 -4.65089023e-01 1.09802890e+00 1.02734828e+00
1.57824129e-01 6.59571469e-01 3.83323193e-01 -1.63516831e+00
9.24143791e-02 -1.01135230e+00 -7.94089735e-01 5.68379045e-01
6.08152270e-01 -7.14692712e-01 -3.42715740e-01 2.87919372e-01] | [11.369420051574707, -2.116116762161255] |
92fdffd6-1d14-4f51-bbf8-2b93e9dca507 | deep-learning-models-for-multilingual-hate | 2004.06465 | null | https://arxiv.org/abs/2004.06465v3 | https://arxiv.org/pdf/2004.06465v3.pdf | Deep Learning Models for Multilingual Hate Speech Detection | Hate speech detection is a challenging problem with most of the datasets available in only one language: English. In this paper, we conduct a large scale analysis of multilingual hate speech in 9 languages from 16 different sources. We observe that in low resource setting, simple models such as LASER embedding with logistic regression performs the best, while in high resource setting BERT based models perform better. In case of zero-shot classification, languages such as Italian and Portuguese achieve good results. Our proposed framework could be used as an efficient solution for low-resource languages. These models could also act as good baselines for future multilingual hate speech detection tasks. We have made our code and experimental settings public for other researchers at https://github.com/punyajoy/DE-LIMIT. | ['Sai Saketh Aluru', 'Punyajoy Saha', 'Animesh Mukherjee', 'Binny Mathew'] | 2020-04-14 | null | null | null | null | ['question-similarity'] | ['natural-language-processing'] | [-6.28050327e-01 -4.77483928e-01 -3.17243546e-01 1.37668356e-01
-8.66050005e-01 -7.34526157e-01 6.21304631e-01 1.31383557e-02
-6.56718433e-01 7.42336631e-01 3.86865437e-01 -2.77611375e-01
6.00579083e-01 -4.63814110e-01 -3.68961990e-01 -7.43728757e-01
8.35806355e-02 9.05312821e-02 4.00326401e-01 -1.37149841e-01
2.88409084e-01 3.20396155e-01 -9.55747664e-01 9.37671289e-02
8.08681846e-01 6.88325167e-02 1.27834275e-01 6.44430995e-01
4.54465114e-02 8.14095974e-01 -7.32485652e-01 -9.16307330e-01
1.86487168e-01 2.32346699e-01 -5.51000357e-01 -3.01039487e-01
3.58086675e-01 -4.33432549e-01 -6.68246746e-01 1.30766344e+00
1.01254702e+00 9.88449622e-03 5.15331268e-01 -1.26178229e+00
-1.10241675e+00 5.59035063e-01 -7.40245521e-01 5.28062880e-01
3.89071882e-01 3.46728623e-01 6.81438625e-01 -1.21607530e+00
6.38871551e-01 1.27206612e+00 4.57844406e-01 5.36510050e-01
-7.45107412e-01 -1.07719874e+00 -3.45784485e-01 3.59647453e-01
-1.53436255e+00 -6.06694996e-01 8.72501671e-01 -5.31509459e-01
1.02202320e+00 1.22432642e-01 2.28630342e-02 1.80698550e+00
-1.61385491e-01 9.35750782e-01 1.38690639e+00 -4.02963281e-01
-2.45615929e-01 7.13325620e-01 1.60180852e-01 7.38762856e-01
3.11830103e-01 -2.65788555e-01 -5.61646044e-01 -5.05373836e-01
2.72015750e-01 1.23044103e-03 -2.36413389e-01 2.06427053e-01
-9.01320934e-01 1.21120846e+00 1.53346494e-01 6.56547427e-01
-1.08193390e-01 -1.42283946e-01 7.76026368e-01 1.77915126e-01
9.93622005e-01 3.86586517e-01 -6.83630109e-02 -3.75079811e-01
-8.28007817e-01 1.05818762e-02 6.86832190e-01 7.87624598e-01
5.08852541e-01 6.78661913e-02 -7.76110291e-02 1.35912955e+00
1.47904992e-01 6.47534728e-01 4.52677011e-01 -5.46205521e-01
7.67735183e-01 2.61782203e-03 1.35791838e-01 -9.73940969e-01
-2.05678418e-01 -1.37859643e-01 -5.36314785e-01 -9.96860564e-02
4.61411923e-01 -5.50779164e-01 -6.03858709e-01 1.29839694e+00
2.41690978e-01 2.85485238e-01 -1.02394581e-01 9.16393697e-01
5.82761586e-01 9.32175040e-01 1.21888481e-01 -5.23162149e-02
1.71181488e+00 -1.34738839e+00 -9.27616775e-01 -3.55324149e-02
7.73160994e-01 -1.28248751e+00 1.21942878e+00 2.64991224e-01
-6.80696607e-01 -9.39434581e-03 -7.42336810e-01 -2.53367901e-01
-8.58121693e-01 2.04745665e-01 4.10287857e-01 9.71921682e-01
-8.20762575e-01 4.88191508e-02 -6.05651855e-01 -8.38152111e-01
4.78840917e-01 -8.81526247e-02 -5.42561173e-01 -1.75602853e-01
-1.26931047e+00 1.05212891e+00 1.87548384e-01 -7.41624683e-02
-9.96442378e-01 -3.79295796e-01 -7.02145278e-01 -3.60558450e-01
3.98024917e-01 1.00895353e-01 9.59811985e-01 -5.11614144e-01
-1.03607488e+00 1.08598864e+00 -1.70929849e-01 -3.85016024e-01
3.41439456e-01 -5.95824659e-01 -4.67628062e-01 1.02652043e-01
7.14875758e-02 3.25881153e-01 8.25093448e-01 -1.03417468e+00
-2.16115758e-01 -3.28432411e-01 1.45269260e-01 -6.03921115e-02
-9.52360749e-01 9.76168692e-01 -2.42902398e-01 -8.51763487e-01
-7.70067155e-01 -1.11843121e+00 2.70031780e-01 -1.67595372e-01
-5.55453181e-01 -3.40713233e-01 1.07513189e+00 -1.25036645e+00
1.61237907e+00 -2.20955038e+00 -2.10471824e-01 -3.99199396e-01
7.06921369e-02 8.45497727e-01 -1.29599394e-02 7.65479386e-01
2.01984495e-01 4.09498483e-01 -8.88178870e-02 -5.45787573e-01
2.38303542e-02 -1.45260394e-01 -2.18952805e-01 7.79299378e-01
2.66683847e-01 5.56112349e-01 -8.04413021e-01 -6.01315856e-01
1.38469681e-01 8.18742633e-01 -2.98708469e-01 3.27439606e-01
3.07279587e-01 1.43571556e-01 -2.40363941e-01 8.47199261e-01
6.90804660e-01 1.91262245e-01 -2.44427517e-01 1.06035784e-01
-2.95730919e-01 3.06482434e-01 -5.94295084e-01 1.52476716e+00
-6.55942142e-01 1.02911842e+00 1.19967140e-01 -4.79426652e-01
8.38993192e-01 5.78068018e-01 -1.41093537e-01 -3.93639863e-01
1.18123785e-01 1.46974623e-01 1.08356275e-01 -8.96329165e-01
4.66439933e-01 1.33467153e-01 -1.15921395e-02 3.97173256e-01
7.49906600e-02 2.67813504e-01 2.45836496e-01 1.96494445e-01
1.08857095e+00 -1.27630532e-01 2.91004449e-01 -4.02534753e-02
4.44710195e-01 -2.07815930e-01 3.35746050e-01 5.28966069e-01
-8.64529192e-01 5.94151437e-01 4.13074046e-01 -1.54386505e-01
-1.33423376e+00 -8.62042785e-01 -2.29148626e-01 1.44627786e+00
-4.66015399e-01 -4.06764179e-01 -7.22558677e-01 -7.47974336e-01
-2.66523838e-01 5.42715430e-01 -3.42465013e-01 1.87419742e-01
-5.07409871e-01 -9.94559705e-01 1.06860316e+00 1.89686462e-01
4.74668920e-01 -1.12604940e+00 -2.04654157e-01 -1.42318562e-01
-1.89390168e-01 -1.36268747e+00 -5.62855005e-01 1.40077174e-01
-1.55422240e-01 -7.93032408e-01 -1.13913572e+00 -8.24717522e-01
2.70002484e-01 3.75864029e-01 6.14883125e-01 9.58214998e-02
-2.97842681e-01 -1.14289001e-01 -6.61396801e-01 -5.52926481e-01
-2.49616265e-01 3.11545372e-01 1.49526626e-01 -4.38424712e-03
7.46233225e-01 -3.48812431e-01 -5.05552709e-01 -3.06787193e-02
-7.78797388e-01 -3.96293789e-01 1.79999620e-01 6.19445920e-01
-2.09985539e-01 -5.62545732e-02 3.15380305e-01 -9.96415973e-01
8.36356997e-01 -1.01650572e+00 -4.41525429e-01 6.57248572e-02
3.16849053e-02 -2.60558695e-01 8.61550152e-01 -4.85180706e-01
-8.94365013e-01 -3.45926821e-01 -3.17270219e-01 -6.22544646e-01
-4.53226209e-01 9.98774543e-02 1.01437069e-01 8.04873407e-02
5.88770449e-01 -2.14182809e-02 -4.77048337e-01 -7.82592595e-01
3.83260883e-02 1.29492295e+00 -5.06928787e-02 -4.19430733e-01
1.10351706e+00 3.22542012e-01 -7.65839398e-01 -1.22834742e+00
-6.07864201e-01 -8.38304281e-01 -6.57922685e-01 -9.27070156e-03
1.11043155e+00 -1.12905777e+00 -3.17887366e-01 7.39315927e-01
-1.27642190e+00 -1.39489353e-01 6.72610164e-01 4.48574662e-01
2.32438609e-01 5.91672957e-01 -1.00145853e+00 -1.36001945e+00
-4.60943103e-01 -1.12350762e+00 9.58730698e-01 1.67878062e-01
-2.38447875e-01 -1.20688260e+00 4.19511288e-01 6.14387631e-01
3.83457661e-01 1.29001081e-01 6.22568488e-01 -1.04371226e+00
-9.30628851e-02 -5.65158166e-02 -2.15080231e-01 3.27838749e-01
1.98166415e-01 3.95994872e-01 -1.32730854e+00 -6.16614163e-01
-2.99638331e-01 -6.59147739e-01 7.46730089e-01 -1.13095924e-01
7.89813340e-01 -4.55384523e-01 -8.69588703e-02 3.41334432e-01
1.56493282e+00 5.36620319e-02 3.92086536e-01 5.30760884e-01
8.40278506e-01 7.10827410e-01 5.13868272e-01 5.73913515e-01
2.49072731e-01 6.67579114e-01 -2.09977664e-02 1.17198534e-01
-3.11764449e-01 -4.17722404e-01 7.69700766e-01 1.27361631e+00
-5.37633710e-02 -3.74680698e-01 -1.35981858e+00 9.46193755e-01
-1.63004303e+00 -1.26960278e+00 -2.97486819e-02 2.06501555e+00
7.41487384e-01 -6.48007318e-02 7.10477233e-01 -1.43723503e-01
9.98599768e-01 4.90239650e-01 -4.43375185e-02 -7.62922168e-01
-1.32479474e-01 8.63360539e-02 5.65918267e-01 5.02347767e-01
-1.39767742e+00 1.30633211e+00 5.77975273e+00 1.07387888e+00
-1.13267279e+00 9.14634109e-01 4.58379269e-01 -2.74987519e-01
7.90632367e-02 -1.28666550e-01 -8.29423428e-01 9.59483027e-01
1.31876814e+00 -5.58781996e-02 5.16942501e-01 7.67664313e-01
2.72435784e-01 9.57527608e-02 -4.10954535e-01 1.03873515e+00
1.81513026e-01 -9.49439585e-01 -3.24672341e-01 1.53907076e-01
6.47653759e-01 5.36721289e-01 2.27995187e-01 3.29680592e-01
4.48421270e-01 -9.66125667e-01 3.85353237e-01 -1.00380167e-01
6.91597104e-01 -7.50102878e-01 7.41165459e-01 4.50949311e-01
-9.20198798e-01 -6.22889735e-02 -6.48387253e-01 1.69015765e-01
2.21422650e-02 3.12741280e-01 -9.57250774e-01 1.58055454e-01
7.89150238e-01 5.50059676e-01 -8.00900519e-01 9.79408145e-01
-3.70268464e-01 9.90278184e-01 -2.53653347e-01 -1.68513313e-01
3.50103557e-01 -2.76123416e-02 5.85972846e-01 1.79025400e+00
2.49279320e-01 -3.00379992e-01 3.30337659e-02 4.14477050e-01
-1.65174648e-01 5.32667637e-01 -1.33193552e+00 -3.91296804e-01
7.54695296e-01 1.32282305e+00 -4.60082263e-01 -3.65494043e-02
-9.36667442e-01 1.01756263e+00 7.35610664e-01 3.63118976e-01
-1.16005135e+00 -5.66922307e-01 7.64630258e-01 5.89199318e-03
1.96214661e-01 -4.67332184e-01 2.24241853e-01 -1.44094443e+00
-6.41450435e-02 -1.05809677e+00 3.56746525e-01 -7.44837046e-01
-1.40675771e+00 6.17134392e-01 -1.75451204e-01 -9.18973029e-01
-1.70559004e-01 -8.15981984e-01 -7.13498056e-01 7.21447229e-01
-1.50035512e+00 -1.35180235e+00 2.29759187e-01 6.70790970e-01
8.49644601e-01 -4.76419836e-01 7.80795991e-01 6.45904899e-01
-1.07787275e+00 8.41364026e-01 2.30003729e-01 7.14179277e-01
1.05827308e+00 -1.11857355e+00 2.07867622e-01 1.12265563e+00
4.34783548e-01 6.06305718e-01 8.16821337e-01 -5.80347061e-01
-1.19793868e+00 -9.16366100e-01 1.02858937e+00 -5.46867669e-01
1.28766012e+00 -7.63307750e-01 -1.06994736e+00 5.48464894e-01
9.20375228e-01 -1.99081019e-01 9.31710482e-01 2.32989699e-01
-7.75081635e-01 2.82873333e-01 -1.11612225e+00 6.50920868e-01
7.20310509e-01 -1.09291458e+00 -3.30616593e-01 6.98832750e-01
6.91976547e-01 -7.91789556e-04 -8.02838862e-01 -2.39074051e-01
4.71589178e-01 -7.75597394e-01 6.29708707e-01 -5.67274690e-01
4.47952420e-01 -6.15749471e-02 -1.31944269e-01 -1.22728741e+00
-2.62773186e-01 -7.60584354e-01 -1.74640696e-02 1.64603186e+00
4.07122642e-01 -7.72385955e-01 3.73793066e-01 3.26445162e-01
2.76318073e-01 -5.69095433e-01 -9.07029390e-01 -7.74805427e-01
5.55581391e-01 -3.49115312e-01 1.47356540e-01 1.21429801e+00
-1.49600044e-01 3.00626785e-01 -9.56793666e-01 3.38234007e-01
4.81280565e-01 -4.38687801e-01 7.77800977e-01 -5.07211089e-01
-3.64584774e-02 -7.89170936e-02 -2.39594534e-01 -2.36687690e-01
7.53693402e-01 -8.73884916e-01 -2.89262444e-01 -9.86373246e-01
6.87761009e-01 -1.71981946e-01 -3.30742687e-01 5.77357352e-01
-2.01842532e-01 5.86839259e-01 4.38736141e-01 2.99919188e-01
-4.27485287e-01 2.84799725e-01 7.44315147e-01 -1.54011384e-01
2.20560834e-01 -3.24734807e-01 -4.07372445e-01 7.33373165e-01
1.19301033e+00 -8.43031526e-01 -5.20225465e-02 -5.47103584e-01
-2.21155435e-01 -4.76585150e-01 2.61490762e-01 -7.77020514e-01
1.33119673e-01 -8.50200467e-03 -1.08538205e-02 -1.43184945e-01
5.02688646e-01 -7.08197713e-01 -3.76276016e-01 2.46440470e-01
-1.33680567e-01 3.13421100e-01 1.41233295e-01 1.28444597e-01
-2.61774927e-01 -6.25834823e-01 7.40617692e-01 -7.65183792e-02
-5.55151999e-01 2.63133407e-01 -4.46084648e-01 2.42611840e-01
9.97406483e-01 2.01677129e-01 -1.01020586e+00 -4.43399996e-01
-3.24802548e-01 -7.49532273e-03 6.04505002e-01 6.95818961e-01
3.61096978e-01 -1.26985800e+00 -9.52193379e-01 -1.78039849e-01
2.29817212e-01 -1.00241220e+00 -1.03449486e-01 9.11833227e-01
-6.53331280e-01 6.73599958e-01 -2.40851626e-01 -3.52043658e-02
-1.55257916e+00 6.68192923e-01 -5.44753671e-02 -1.98478565e-01
-2.47227535e-01 6.00853920e-01 1.19543880e-01 -4.91847396e-01
1.70387387e-01 4.77326661e-01 -1.70667350e-01 2.64773577e-01
7.33032227e-01 4.30505008e-01 -2.84788221e-01 -1.23646438e+00
-6.00991905e-01 2.72933483e-01 -2.28791878e-01 -1.59801409e-01
1.29876864e+00 -8.09077993e-02 -1.41065031e-01 6.43471897e-01
1.63007176e+00 6.12751424e-01 -7.45289743e-01 -1.17441222e-01
1.13536224e-01 -7.77496040e-01 -9.15122963e-03 -4.14553344e-01
-8.43924880e-01 1.08377635e+00 6.54336989e-01 2.62647986e-01
8.51345479e-01 1.35699168e-01 9.09196138e-01 2.96572059e-01
4.37696636e-01 -1.06576777e+00 2.32065115e-02 5.28267443e-01
8.05404484e-01 -1.62395227e+00 -2.21219704e-01 -1.92926556e-01
-8.41168523e-01 8.84684265e-01 6.93916440e-01 -2.34858856e-01
6.11529648e-01 3.25158298e-01 2.85440922e-01 2.24485230e-02
-7.92493403e-01 -4.66378987e-01 6.68593571e-02 6.95835531e-01
1.00752342e+00 1.56039208e-01 -4.10669625e-01 1.66478470e-01
-2.06470281e-01 -5.02959013e-01 6.15945220e-01 7.44162440e-01
-2.39971817e-01 -1.10166681e+00 -6.14698887e-01 1.62432030e-01
-1.12958372e+00 -3.80749017e-01 -5.72885633e-01 8.62710595e-01
9.42376480e-02 1.07411146e+00 -2.49225318e-01 -3.44763666e-01
-3.35436431e-03 3.25815856e-01 2.42917776e-01 -7.00791895e-01
-7.05623507e-01 1.63405716e-01 4.18887377e-01 -1.77053839e-01
-3.20484877e-01 -6.30572736e-01 -6.49047017e-01 -6.81779563e-01
-1.75077140e-01 7.20561668e-02 5.61780870e-01 4.57418382e-01
2.38877416e-01 -2.03969851e-01 6.71107590e-01 -4.61195499e-01
-2.55715519e-01 -1.19426465e+00 -4.71212476e-01 5.05435705e-01
4.01207954e-01 -6.34617090e-01 -5.53146660e-01 -1.41891122e-01] | [8.763197898864746, 10.53953742980957] |
50ab7288-9f54-4f58-89df-dbd40c92fe58 | memory-efficient-network-for-large-scale | 2103.03089 | null | https://arxiv.org/abs/2103.03089v2 | https://arxiv.org/pdf/2103.03089v2.pdf | Memory-Efficient Network for Large-scale Video Compressive Sensing | Video snapshot compressive imaging (SCI) captures a sequence of video frames in a single shot using a 2D detector. The underlying principle is that during one exposure time, different masks are imposed on the high-speed scene to form a compressed measurement. With the knowledge of masks, optimization algorithms or deep learning methods are employed to reconstruct the desired high-speed video frames from this snapshot measurement. Unfortunately, though these methods can achieve decent results, the long running time of optimization algorithms or huge training memory occupation of deep networks still preclude them in practical applications. In this paper, we develop a memory-efficient network for large-scale video SCI based on multi-group reversible 3D convolutional neural networks. In addition to the basic model for the grayscale SCI system, we take one step further to combine demosaicing and SCI reconstruction to directly recover color video from Bayer measurements. Extensive results on both simulation and real data captured by SCI cameras demonstrate that our proposed model outperforms previous state-of-the-art with less memory and thus can be used in large-scale problems. The code is at https://github.com/BoChenGroup/RevSCI-net. | ['Xin Yuan', 'Zhengjue Wang', 'Ruiying Lu', 'Hao Zhang', 'Guanliang Liu', 'Bo Chen', 'Ziheng Cheng'] | 2021-03-04 | null | http://openaccess.thecvf.com//content/CVPR2021/html/Cheng_Memory-Efficient_Network_for_Large-Scale_Video_Compressive_Sensing_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Cheng_Memory-Efficient_Network_for_Large-Scale_Video_Compressive_Sensing_CVPR_2021_paper.pdf | cvpr-2021-1 | ['video-compressive-sensing'] | ['computer-vision'] | [ 4.50404286e-01 -7.31882453e-01 1.30428806e-01 -7.40466192e-02
-6.79139197e-01 -4.00706351e-01 5.22843711e-02 -7.85611570e-01
-4.26764816e-01 4.57361847e-01 -1.53696258e-02 -3.23194623e-01
7.24351108e-02 -5.51767647e-01 -9.93217468e-01 -7.70616949e-01
-1.37542889e-01 -1.43388808e-01 8.94863307e-02 2.65700549e-01
2.52333373e-01 5.11212230e-01 -1.19702756e+00 2.61322469e-01
3.85002047e-01 1.22133076e+00 5.03805459e-01 8.58959317e-01
3.16476166e-01 1.29132187e+00 -3.38886172e-01 1.17527517e-02
8.08369875e-01 -6.86516166e-01 -1.75036609e-01 3.89247507e-01
5.74382067e-01 -1.16736734e+00 -1.11256337e+00 1.31932688e+00
6.01400018e-01 1.58122167e-01 2.42478713e-01 -8.43457222e-01
-6.77752137e-01 3.02763373e-01 -6.71956778e-01 3.30127716e-01
2.88811147e-01 2.57011861e-01 4.24006075e-01 -1.06526780e+00
3.00456941e-01 9.33010459e-01 5.11673868e-01 2.38260373e-01
-7.18109906e-01 -8.03879917e-01 -2.29163989e-01 6.08283103e-01
-1.50680089e+00 -7.90044725e-01 1.03812897e+00 -1.21148586e-01
7.04340637e-01 9.07640010e-02 6.69417739e-01 7.14603961e-01
2.51198441e-01 8.43453169e-01 1.14043498e+00 -3.34717244e-01
1.75221980e-01 -5.00553012e-01 -3.26603591e-01 9.30037439e-01
3.57160956e-01 2.64227062e-01 -5.88280380e-01 6.90153539e-02
1.07298076e+00 5.90978384e-01 -8.69969249e-01 -2.48100958e-03
-1.17945433e+00 4.46496278e-01 3.59213799e-01 1.13841696e-02
-3.71542990e-01 5.11636019e-01 4.71483879e-02 4.62575018e-01
2.33506382e-01 3.30044553e-02 -3.76343988e-02 1.52307793e-01
-1.23394549e+00 -6.85143843e-02 7.01320469e-01 9.02578115e-01
7.39367247e-01 5.26124775e-01 1.56436160e-01 5.47262549e-01
2.22255901e-01 7.01938510e-01 3.40262413e-01 -1.56501853e+00
3.49784017e-01 -8.14539846e-03 1.22977927e-01 -1.03745091e+00
-6.17117956e-02 -3.96510065e-01 -1.12773895e+00 4.33941334e-01
9.34550241e-02 -3.13716501e-01 -8.69615197e-01 1.29049444e+00
1.21505111e-01 8.75025511e-01 1.85262822e-02 1.39891577e+00
5.16305566e-01 1.03961313e+00 -8.02784204e-01 -4.48312730e-01
1.03531408e+00 -9.75681067e-01 -6.14682794e-01 -3.29514563e-01
9.51524153e-02 -8.60970914e-01 5.30183494e-01 6.95076287e-01
-1.37414706e+00 -5.30669153e-01 -1.51565039e+00 -1.49236724e-01
2.13370562e-01 1.62712127e-01 2.12086886e-01 3.76267314e-01
-1.10216153e+00 6.75194025e-01 -8.80017698e-01 -2.55493429e-02
5.14040411e-01 2.08595186e-01 -1.21238582e-01 -6.62994921e-01
-8.99386227e-01 4.13265646e-01 3.16853076e-01 3.88561368e-01
-1.33509851e+00 -5.35546243e-01 -5.29154062e-01 1.44064084e-01
5.86846530e-01 -6.07274234e-01 1.09330368e+00 -9.15813565e-01
-1.68038726e+00 5.38083613e-01 7.30303228e-02 -4.76453513e-01
4.55747098e-01 -3.85923326e-01 -1.61181718e-01 6.99087262e-01
-2.53145158e-01 3.30842406e-01 1.22523868e+00 -1.07609558e+00
-4.32975858e-01 -2.16394350e-01 1.85461640e-02 3.25482756e-01
-2.13797480e-01 4.15996313e-02 -8.30378771e-01 -3.74832898e-01
2.28479519e-01 -1.03727758e+00 -1.55853465e-01 3.70335102e-01
-3.24976861e-01 6.52658284e-01 9.27745819e-01 -8.82324219e-01
8.97045791e-01 -2.25470257e+00 1.04867414e-01 -2.63173938e-01
3.63612443e-01 4.62506682e-01 -1.48060411e-01 3.27053010e-01
1.96468290e-02 -4.36701119e-01 -5.07293820e-01 -3.80058020e-01
-3.93590927e-01 -1.88179873e-02 -3.04948509e-01 9.83226180e-01
-2.79823571e-01 7.08466411e-01 -6.48760736e-01 -3.74602377e-01
5.32782912e-01 4.84026521e-01 -5.53555429e-01 4.41109836e-01
-4.95964312e-04 4.64282542e-01 -2.97513962e-01 7.60054648e-01
1.05161059e+00 -4.68932569e-01 2.24780411e-01 -3.22085768e-01
-1.94754541e-01 -7.64503255e-02 -1.24722135e+00 1.96741700e+00
-1.75805286e-01 8.30648482e-01 4.81798857e-01 -1.28118849e+00
5.48067093e-01 2.47951701e-01 5.97353756e-01 -7.38898277e-01
3.27121526e-01 2.27089733e-01 -3.28949660e-01 -8.21949184e-01
4.53814447e-01 -2.90240906e-02 2.37462103e-01 4.67258900e-01
-1.37250841e-01 -5.05556865e-03 2.54571680e-02 9.61678177e-02
1.19572318e+00 9.87519100e-02 2.09854886e-01 8.95468071e-02
4.55512553e-01 -2.67684251e-01 7.58452833e-01 7.58461058e-01
-1.97199911e-01 9.84649062e-01 4.86500449e-02 -3.55597764e-01
-1.19560742e+00 -9.54233110e-01 2.33131811e-01 6.03195548e-01
4.30150658e-01 8.49454030e-02 -5.25219083e-01 -2.57551782e-02
-3.22102457e-01 2.81059384e-01 -1.01074301e-01 -2.90096235e-02
-8.36290598e-01 -7.12036252e-01 4.93607819e-01 2.76483417e-01
1.02762914e+00 -7.24476218e-01 -9.11009192e-01 1.96592554e-01
-1.54240802e-01 -1.48503268e+00 -6.07431233e-01 -1.50080740e-01
-8.92452419e-01 -1.20489883e+00 -8.19078803e-01 -8.71251464e-01
5.15421510e-01 1.13703489e+00 7.26534545e-01 1.95183888e-01
-1.97453126e-01 2.31160417e-01 -3.29837799e-01 -2.93647889e-02
-1.20641075e-01 -5.67939758e-01 1.50410654e-02 2.46602446e-01
7.88391605e-02 -8.62138212e-01 -1.00466275e+00 -5.68851419e-02
-1.35562587e+00 2.56375939e-01 7.46442735e-01 7.79277861e-01
4.03458327e-01 8.10411274e-02 2.36150511e-02 -4.31104720e-01
1.84302106e-01 -4.22703803e-01 -9.68529105e-01 1.05477400e-01
-3.89062941e-01 -2.07550585e-01 8.24644744e-01 -4.04085457e-01
-9.78209376e-01 1.55467123e-01 5.19820116e-02 -1.13712311e+00
7.18523562e-02 5.92854619e-01 5.00769354e-02 -3.84513825e-01
2.27010593e-01 7.24021137e-01 2.01397479e-01 -4.05232221e-01
2.41502635e-02 6.37173891e-01 8.21449876e-01 -2.10318163e-01
8.15078199e-01 9.27924693e-01 1.31672785e-01 -9.24111366e-01
-6.77459359e-01 -3.22914153e-01 -3.23994935e-01 -4.79886413e-01
8.59157205e-01 -1.40974474e+00 -6.93935215e-01 8.07333767e-01
-1.05765593e+00 -3.45763385e-01 1.64572269e-01 8.26983690e-01
-3.90222162e-01 8.34914625e-01 -9.73502994e-01 -6.20026290e-01
-4.69090134e-01 -1.10330153e+00 7.80861795e-01 3.02065343e-01
6.95840478e-01 -5.97101808e-01 -1.45546973e-01 3.67114574e-01
4.68214095e-01 6.55864254e-02 3.70262206e-01 1.08978003e-01
-1.38495755e+00 -3.24021190e-01 -3.43411833e-01 7.02618301e-01
-1.14608005e-01 -2.46591523e-01 -8.36106598e-01 -6.16079390e-01
6.70102715e-01 -2.42299810e-01 9.13331747e-01 5.79915881e-01
1.34814346e+00 -2.19585642e-01 1.25212178e-01 1.28426230e+00
1.88129783e+00 5.05703092e-01 7.96144605e-01 1.03528216e-01
8.14189672e-01 -7.79702216e-02 2.22017571e-01 5.67470551e-01
5.60815670e-02 4.01851267e-01 5.79894602e-01 2.81557348e-02
-3.14934045e-01 -1.26544669e-01 6.84884310e-01 1.15522385e+00
-1.87383872e-02 -5.12133420e-01 -4.15466368e-01 3.35060120e-01
-1.65318608e+00 -1.05206645e+00 2.26376094e-02 2.32071996e+00
2.93793797e-01 -2.51016200e-01 -4.58387792e-01 -8.58010259e-03
7.47695267e-01 6.96450293e-01 -7.20528901e-01 2.51926333e-01
-2.73906320e-01 -2.01612934e-02 8.24716926e-01 4.12102878e-01
-7.71025300e-01 5.38921177e-01 5.54131937e+00 6.99476659e-01
-1.36988771e+00 3.82023722e-01 5.96807241e-01 -5.99944413e-01
2.46789530e-01 8.86326656e-02 -3.39960724e-01 5.72542071e-01
7.78141141e-01 -7.35523775e-02 9.47730780e-01 4.78534549e-01
2.62120306e-01 -1.41273156e-01 -8.97248566e-01 1.44910014e+00
4.45447773e-01 -1.62093103e+00 -1.27587005e-01 7.58654065e-03
9.60906804e-01 4.35555637e-01 6.08422086e-02 -1.76719457e-01
-7.40999952e-02 -5.55636585e-01 7.33169258e-01 5.12884438e-01
1.00993013e+00 -4.49579805e-01 4.46208149e-01 3.23221236e-01
-1.05687916e+00 -1.19040281e-01 -6.70376480e-01 -3.05936188e-01
4.72465843e-01 7.24314988e-01 -1.90505028e-01 5.30307055e-01
6.91133142e-01 9.23671186e-01 -1.26536161e-01 1.20610082e+00
-1.04444295e-01 7.02623367e-01 -3.18762571e-01 2.77007401e-01
2.89386660e-01 -6.38279319e-01 6.67565584e-01 9.14421916e-01
9.30216491e-01 5.45255423e-01 2.07658455e-01 7.35143900e-01
-3.76771033e-01 -5.21616578e-01 -6.64106131e-01 5.25740348e-02
3.62527162e-01 1.14500058e+00 -5.00371397e-01 -4.91954863e-01
-7.73131251e-01 1.25523388e+00 -5.89306839e-02 6.62815332e-01
-9.00079072e-01 -1.53190523e-01 5.27814329e-01 -4.64635305e-02
5.44080317e-01 -5.34514010e-01 7.82505795e-03 -1.54106116e+00
1.06150970e-01 -8.76679778e-01 1.24736525e-01 -1.14176607e+00
-9.45642710e-01 3.20816338e-01 -2.76723832e-01 -1.70084918e+00
1.19518586e-01 -6.20312393e-01 -6.40548587e-01 5.32126725e-01
-1.80872107e+00 -7.49810636e-01 -6.04228079e-01 7.60471225e-01
7.12147951e-01 -1.80287510e-01 3.08608055e-01 5.92822134e-01
-6.40441298e-01 1.35488972e-01 7.11685181e-01 2.55762160e-01
6.00557566e-01 -5.48038602e-01 2.80688614e-01 1.45578945e+00
1.09892795e-02 2.56315291e-01 5.47545373e-01 -5.75911641e-01
-2.12208867e+00 -1.06070733e+00 1.85077026e-01 3.32941085e-01
4.27258044e-01 -7.93148577e-02 -8.94371867e-01 6.88818574e-01
5.42186081e-01 3.62993866e-01 3.34807515e-01 -8.35652709e-01
-1.79933459e-01 -2.86360413e-01 -8.03706110e-01 4.22377408e-01
1.05331564e+00 -6.59414291e-01 -2.66554862e-01 4.46104437e-01
5.91692507e-01 -6.57706559e-01 -5.04028440e-01 1.99689701e-01
5.01501262e-01 -1.27851558e+00 1.08562624e+00 2.63090674e-02
7.70818949e-01 -6.02157772e-01 -4.33157116e-01 -9.99875009e-01
-2.83806115e-01 -8.87691140e-01 -4.30031836e-01 4.94556367e-01
-1.99094102e-01 -6.49485886e-01 7.99418747e-01 2.09162295e-01
-2.98186868e-01 -6.03727162e-01 -8.79418433e-01 -7.57960677e-01
-3.12812954e-01 -4.94020671e-01 4.75126058e-01 7.85187006e-01
-4.32014197e-01 7.46065527e-02 -1.06888306e+00 7.07139432e-01
1.21634316e+00 4.17034984e-01 6.28529191e-01 -5.36938965e-01
-6.69350207e-01 -2.92636938e-02 -3.88375759e-01 -1.55961931e+00
-5.29365316e-02 -5.53737283e-01 1.31313369e-01 -1.29043162e+00
3.15473050e-01 -2.39816904e-02 -4.14699554e-01 -1.03224434e-01
-4.67280373e-02 4.98785198e-01 5.53292036e-01 4.96753812e-01
-7.08984733e-01 5.60831428e-01 1.13950455e+00 -1.04881421e-01
1.77328527e-01 -3.87090564e-01 -4.10743415e-01 6.15436018e-01
6.29013360e-01 -5.98947763e-01 -3.12677026e-01 -1.09266818e+00
-1.00337705e-02 7.87119985e-01 5.55109680e-01 -1.39844060e+00
6.88076854e-01 -7.19870180e-02 5.77221036e-01 -4.97157902e-01
6.44547641e-01 -8.17443609e-01 3.92660797e-01 5.50630391e-01
-4.70510684e-02 5.69341742e-02 -1.71066225e-01 6.46858454e-01
-2.69754082e-01 -2.35587463e-01 8.48861575e-01 -3.84660512e-01
-7.73733020e-01 6.67097151e-01 -3.01160306e-01 -1.59938559e-01
8.48766923e-01 -2.30260596e-01 -4.69999284e-01 -6.29159033e-01
-1.55745342e-01 -1.46781402e-02 7.01284647e-01 7.33966529e-02
1.04990876e+00 -1.16441572e+00 -8.18135381e-01 4.26438242e-01
-3.64823967e-01 -4.81481105e-03 6.26307249e-01 8.81999075e-01
-1.14101636e+00 1.82015553e-01 -2.59131491e-01 -5.22868514e-01
-1.11276138e+00 6.81402385e-01 4.12318617e-01 3.00391257e-01
-9.29003000e-01 8.15873802e-01 1.76463887e-01 1.06290817e-01
1.43114835e-01 -7.41618723e-02 4.34129149e-01 -5.15982747e-01
8.71556342e-01 4.34046537e-01 -2.44760200e-01 -4.40373003e-01
-2.02038009e-02 5.97531974e-01 1.15455382e-01 -1.37885705e-01
1.47493589e+00 -3.54828864e-01 -1.66191116e-01 1.96888611e-01
1.61336720e+00 -1.98855430e-01 -1.80029035e+00 -3.93512160e-01
-7.43189454e-01 -9.14623380e-01 4.50337797e-01 -1.64582402e-01
-1.57157302e+00 8.10670197e-01 7.19422281e-01 -8.68766159e-02
1.52542377e+00 -3.93620372e-01 1.18225729e+00 5.89833617e-01
3.59211087e-01 -9.25300419e-01 1.45340830e-01 2.86744118e-01
5.92759073e-01 -1.22726047e+00 2.71701366e-01 -8.40612575e-02
-2.46257469e-01 1.23596299e+00 1.92437142e-01 -5.90184510e-01
4.42418724e-01 3.72910172e-01 3.62228677e-02 -6.78199083e-02
-6.12600923e-01 1.47110492e-01 -1.87051460e-01 2.16288120e-01
-3.90269198e-02 -1.72045648e-01 8.75854567e-02 -1.54431000e-01
3.40352625e-01 4.22859877e-01 9.89743054e-01 1.04131401e+00
-4.94468123e-01 -4.75444734e-01 -6.73771918e-01 4.21435475e-01
-5.75853109e-01 -1.40727863e-01 1.58813387e-01 4.24790889e-01
-1.13032810e-01 9.28089321e-01 -6.97542801e-02 -4.57905591e-01
-1.42292693e-01 -3.21551949e-01 6.06829643e-01 -1.70208260e-01
2.81620678e-02 2.15071946e-01 -2.10937947e-01 -7.66494632e-01
-5.50682366e-01 -6.12387598e-01 -1.00903404e+00 -6.27629340e-01
-3.71713862e-02 -2.28643924e-01 5.67316651e-01 8.60897005e-01
3.81627142e-01 3.24900150e-01 9.21289980e-01 -1.25556827e+00
-5.18425941e-01 -6.97244525e-01 -7.56407559e-01 1.72931224e-01
6.43818259e-01 -1.59368053e-01 -6.03733957e-01 3.36001128e-01] | [11.031530380249023, -2.127836227416992] |
904dfa2f-c080-4438-b97b-03fff410a91b | improving-weakly-supervised-visual-grounding | 2007.01951 | null | https://arxiv.org/abs/2007.01951v2 | https://arxiv.org/pdf/2007.01951v2.pdf | Improving Weakly Supervised Visual Grounding by Contrastive Knowledge Distillation | Weakly supervised phrase grounding aims at learning region-phrase correspondences using only image-sentence pairs. A major challenge thus lies in the missing links between image regions and sentence phrases during training. To address this challenge, we leverage a generic object detector at training time, and propose a contrastive learning framework that accounts for both region-phrase and image-sentence matching. Our core innovation is the learning of a region-phrase score function, based on which an image-sentence score function is further constructed. Importantly, our region-phrase score function is learned by distilling from soft matching scores between the detected object names and candidate phrases within an image-sentence pair, while the image-sentence score function is supervised by ground-truth image-sentence pairs. The design of such score functions removes the need of object detection at test time, thereby significantly reducing the inference cost. Without bells and whistles, our approach achieves state-of-the-art results on visual phrase grounding, surpassing previous methods that require expensive object detectors at test time. | ['Liwei Wang', 'Kun Xu', 'Yin Li', 'Jing Huang', 'Zhengyuan Yang', 'Dong Yu'] | 2020-07-03 | null | http://openaccess.thecvf.com//content/CVPR2021/html/Wang_Improving_Weakly_Supervised_Visual_Grounding_by_Contrastive_Knowledge_Distillation_CVPR_2021_paper.html | http://openaccess.thecvf.com//content/CVPR2021/papers/Wang_Improving_Weakly_Supervised_Visual_Grounding_by_Contrastive_Knowledge_Distillation_CVPR_2021_paper.pdf | cvpr-2021-1 | ['phrase-grounding'] | ['natural-language-processing'] | [ 3.85293961e-01 3.04095060e-01 -1.95817545e-01 -4.54299569e-01
-1.35432637e+00 -6.42709672e-01 6.19181514e-01 4.89962131e-01
-4.54091817e-01 3.81887794e-01 -2.06439227e-01 -9.21583474e-02
3.54828745e-01 -6.44093275e-01 -1.09356630e+00 -5.24836659e-01
1.81082889e-01 3.82831544e-01 7.76315033e-01 -3.26470099e-02
2.51342982e-01 1.87566504e-01 -1.45093203e+00 5.69234729e-01
3.39666307e-01 1.06937444e+00 6.43239915e-01 6.79655850e-01
-1.27830133e-01 6.97681725e-01 -3.66766095e-01 -5.76391876e-01
3.78606647e-01 -4.79051620e-01 -5.38009048e-01 4.01644856e-01
1.13365519e+00 -3.72356504e-01 -6.15751781e-02 1.32988274e+00
2.54281282e-01 -1.34933949e-01 5.24590492e-01 -1.14254069e+00
-5.41948974e-01 3.08680296e-01 -7.59892941e-01 2.06884533e-01
4.17591780e-01 2.11786196e-01 1.55280733e+00 -1.43480921e+00
5.37583530e-01 1.07776964e+00 6.37118638e-01 2.84584343e-01
-1.44022799e+00 -4.00536835e-01 2.84583330e-01 -1.33028626e-01
-1.35413945e+00 -2.38827437e-01 7.40415752e-01 -6.11451387e-01
1.03098428e+00 2.61306390e-02 6.26708984e-01 6.16696954e-01
3.65736894e-02 9.95156467e-01 9.12508845e-01 -4.90861982e-01
1.01335287e-01 4.59905744e-01 -5.20169958e-02 1.36501586e+00
9.55470838e-03 6.74530864e-03 -7.54792035e-01 1.04328655e-02
1.02759123e+00 6.19514622e-02 -1.40371650e-01 -7.23828733e-01
-1.25669241e+00 6.80676699e-01 5.71366549e-01 6.09565713e-02
-2.67852582e-02 3.13422203e-01 9.26516727e-02 5.00132516e-02
4.27285433e-01 2.37021759e-01 -4.19323415e-01 2.87593991e-01
-1.22274971e+00 3.64750534e-01 4.56237376e-01 9.84369695e-01
1.11004925e+00 -4.58795488e-01 -4.87818360e-01 5.38891375e-01
4.25908715e-01 6.47089005e-01 -1.42786562e-01 -7.90713906e-01
6.34068727e-01 6.61850333e-01 1.46428421e-01 -1.01296592e+00
-1.01629786e-01 -6.90579414e-01 -2.32752725e-01 1.60213187e-01
5.17288804e-01 3.28228742e-01 -9.76909578e-01 1.66765881e+00
3.20274889e-01 2.68029094e-01 -2.95018017e-01 9.21585143e-01
8.14717114e-01 4.46152627e-01 4.54338863e-02 -7.09769782e-03
1.67688632e+00 -1.15920091e+00 -1.58935264e-01 -5.92822552e-01
4.30847675e-01 -7.05724716e-01 1.32304132e+00 -1.06542572e-01
-1.40054178e+00 -6.23464823e-01 -1.00896645e+00 -3.30975950e-01
-1.50922015e-01 1.47325903e-01 3.24307084e-01 2.82845318e-01
-1.08036852e+00 3.53550196e-01 -6.60343945e-01 -2.85881102e-01
4.08459663e-01 4.33957666e-01 -3.00819635e-01 1.55808982e-02
-8.37657928e-01 8.81150723e-01 2.75847167e-01 -6.83531761e-02
-8.28680217e-01 -8.40837359e-01 -1.00892031e+00 1.68566778e-01
4.22860414e-01 -8.97178471e-01 1.29855251e+00 -8.88765574e-01
-8.83812070e-01 1.53862298e+00 -1.90086976e-01 -3.01732749e-01
4.34739649e-01 1.00569136e-01 1.12507880e-01 5.02266347e-01
4.98068124e-01 1.07144272e+00 1.06789577e+00 -1.39651775e+00
-8.98035347e-01 -2.33951628e-01 2.38620952e-01 3.19559485e-01
-5.12055270e-02 6.49362728e-02 -9.27112401e-01 -5.77413917e-01
2.23232210e-01 -6.86122596e-01 -8.64890069e-02 5.73217094e-01
-3.20544541e-01 -1.65768757e-01 6.14948988e-01 -5.19672275e-01
6.41412556e-01 -2.18674183e+00 4.41941917e-02 1.22885786e-01
2.63374776e-01 -1.18219048e-01 -4.68001664e-01 -1.59777571e-02
1.10467955e-01 -1.43383250e-01 -4.00441438e-01 -7.21542954e-01
-4.52321097e-02 2.58369178e-01 -4.27924424e-01 5.25953352e-01
8.54882777e-01 1.25772524e+00 -1.14394951e+00 -8.85268092e-01
2.61675835e-01 1.26186252e-01 -7.15927780e-01 2.78328419e-01
-5.66997051e-01 2.19688356e-01 -2.86179990e-01 7.53082871e-01
5.29263496e-01 -6.33314908e-01 -2.84494668e-01 -2.27191985e-01
-1.14825390e-01 2.16356769e-01 -9.21482265e-01 1.99704647e+00
-1.91760570e-01 5.95271230e-01 1.86488375e-01 -9.17543173e-01
6.81246877e-01 5.00222184e-02 4.28430557e-01 -8.38613927e-01
-5.02079492e-03 2.06481367e-01 -3.32579792e-01 -4.77396935e-01
4.23501015e-01 -2.87077993e-01 -1.59983605e-01 3.06962073e-01
2.61209548e-01 -4.77630824e-01 8.85260403e-02 3.95379782e-01
1.05582643e+00 2.42152929e-01 6.73002750e-02 -2.03415364e-01
5.97726762e-01 -1.42064989e-02 4.18839246e-01 9.09377694e-01
-2.21670449e-01 1.01610923e+00 5.82489133e-01 -3.02322060e-01
-1.23203707e+00 -1.54111016e+00 -2.11604685e-01 1.20335686e+00
4.87320304e-01 -3.10651273e-01 -6.51307702e-01 -8.81529987e-01
1.17546609e-02 1.70058340e-01 -5.38352847e-01 1.46045253e-01
-5.34268081e-01 -2.02635482e-01 9.54772756e-02 5.46901345e-01
2.43075699e-01 -9.35729742e-01 -5.01231790e-01 3.54680866e-02
-1.78165331e-01 -1.44774055e+00 -7.78157175e-01 1.80778295e-01
-7.02361226e-01 -9.48030829e-01 -7.97261775e-01 -1.23012257e+00
1.02843797e+00 3.62382263e-01 1.56348836e+00 1.94437265e-01
-4.70874190e-01 4.82301086e-01 -1.37293741e-01 -2.73033619e-01
-1.57083809e-01 -1.81470826e-01 -2.86507010e-01 -5.14705293e-02
1.31820247e-01 -2.58320719e-01 -8.31577837e-01 2.32421190e-01
-7.34910905e-01 2.24572018e-01 9.13283527e-01 1.03229558e+00
1.01940644e+00 -3.28316599e-01 1.35849699e-01 -5.82085907e-01
-3.64253707e-02 -1.16218261e-01 -8.17606747e-01 3.67819130e-01
-2.88287878e-01 1.65763229e-01 9.20448303e-02 -3.34316999e-01
-5.71251035e-01 4.13295060e-01 -1.44488335e-01 -4.13509816e-01
1.67502031e-01 1.97830305e-01 5.13064414e-02 -2.04887956e-01
5.21117389e-01 2.81969786e-01 -1.11177051e-03 -9.53303576e-02
4.22820032e-01 2.65158772e-01 9.00052726e-01 -5.68230331e-01
9.98817563e-01 6.81599677e-01 -9.86500978e-02 -5.75024486e-01
-1.36630917e+00 -8.84943306e-01 -6.30489469e-01 -2.59717345e-01
1.12039745e+00 -1.19409871e+00 -4.18397218e-01 -1.07724434e-02
-1.43293595e+00 -1.31606579e-01 -5.67944109e-01 3.01171958e-01
-7.81922102e-01 2.65536606e-01 -5.66164732e-01 -5.72131813e-01
-2.22941518e-01 -1.18621707e+00 1.77497685e+00 -2.99393591e-02
6.57739937e-02 -7.76980102e-01 -1.77925944e-01 4.58531886e-01
5.07872440e-02 3.94835137e-02 7.66544104e-01 -2.47703120e-01
-1.06991053e+00 -3.17956954e-01 -7.31664121e-01 3.80513430e-01
-2.48250574e-01 -4.12241936e-01 -1.02631664e+00 -4.11977202e-01
1.06289305e-01 -4.17605609e-01 9.18007672e-01 3.41414869e-01
9.32855368e-01 -1.24086745e-01 -2.53273398e-01 5.71560860e-01
1.50863051e+00 -4.27555054e-01 3.61763686e-01 1.69583395e-01
8.16852450e-01 5.81617594e-01 5.63978434e-01 4.01494466e-02
4.20061141e-01 8.78899395e-01 4.39912945e-01 -6.25827432e-01
-3.24951112e-01 -5.08433998e-01 3.75663280e-01 2.48876154e-01
4.07539546e-01 7.09542856e-02 -8.98416817e-01 6.96131051e-01
-1.96763563e+00 -8.97627771e-01 -4.38268855e-02 2.07172894e+00
8.52521718e-01 4.60616320e-01 1.75851628e-01 -2.32215956e-01
6.18369639e-01 9.02349278e-02 -4.64107603e-01 1.33820146e-01
-5.40243648e-02 2.23731458e-01 5.05477250e-01 7.48345077e-01
-1.20059347e+00 1.00482404e+00 6.17760706e+00 6.17049098e-01
-9.17583823e-01 2.52941936e-01 5.20582795e-01 -1.05056085e-01
-3.39539498e-01 2.01188587e-02 -9.40346420e-01 2.64169484e-01
3.01879525e-01 2.02561572e-01 6.75327927e-02 8.39253902e-01
1.39614642e-01 -2.24733993e-01 -1.49197006e+00 1.18171561e+00
4.11789060e-01 -1.46136415e+00 -4.76407111e-02 -1.24725863e-01
7.73212254e-01 2.00599387e-01 2.65106678e-01 1.29401937e-01
-3.81985344e-02 -7.54926920e-01 1.18196189e+00 3.14031750e-01
7.70618558e-01 -3.07210505e-01 5.81014216e-01 1.78861558e-01
-1.22979879e+00 6.49197772e-02 -2.59555131e-01 2.53964458e-02
1.47652641e-01 5.70133030e-01 -9.01781380e-01 1.89543635e-01
5.71703374e-01 6.44461691e-01 -6.68822944e-01 8.88851821e-01
-5.16290009e-01 2.73728430e-01 -1.71148986e-01 1.32533908e-01
3.82753551e-01 1.37298957e-01 6.20572925e-01 1.35661209e+00
4.35530953e-02 -1.49440527e-01 6.79446280e-01 1.23956668e+00
-3.29931043e-02 -1.10141881e-01 -5.14008522e-01 2.27000132e-01
3.30510318e-01 1.23690534e+00 -9.85196829e-01 -2.77172863e-01
-7.22030699e-01 1.10850322e+00 4.06926900e-01 2.80585349e-01
-7.27938652e-01 1.77212935e-02 3.54527652e-01 4.60001707e-01
5.40591598e-01 -2.50716209e-01 -4.22045141e-01 -1.18286324e+00
4.68845427e-01 -3.64122719e-01 1.47600070e-01 -9.17653322e-01
-1.38656330e+00 2.10145697e-01 -3.81092019e-02 -1.22312689e+00
-9.72092822e-02 -7.28833437e-01 -5.00964403e-01 8.18427861e-01
-1.71114552e+00 -1.53868139e+00 -2.29201302e-01 6.19535506e-01
5.93785405e-01 1.37208387e-01 5.82954109e-01 1.65085077e-01
-2.34980479e-01 6.39122069e-01 -3.60726893e-01 2.71273643e-01
5.97195029e-01 -1.36170340e+00 4.06344235e-01 9.26629126e-01
6.50745690e-01 4.94023412e-01 7.09478736e-01 -2.91954517e-01
-1.13880336e+00 -1.00389910e+00 9.16772425e-01 -1.02760160e+00
7.88087964e-01 -9.15893555e-01 -7.85769343e-01 4.33414340e-01
-6.25959784e-02 4.44444180e-01 2.74365306e-01 7.18912482e-02
-5.55414379e-01 -1.11518070e-01 -8.88328135e-01 4.48448569e-01
8.51011634e-01 -1.08161652e+00 -6.23909414e-01 6.52473569e-01
8.02889764e-01 -5.97756982e-01 -5.08434892e-01 3.30454230e-01
5.39195597e-01 -8.89927685e-01 1.18641222e+00 -2.28003219e-01
4.78783131e-01 -5.25906384e-01 -1.53668776e-01 -5.91357946e-01
-2.48354912e-01 -1.56388506e-01 -8.19186792e-02 1.00135767e+00
6.88539445e-01 -1.77977607e-01 1.12578821e+00 4.70759362e-01
-1.58457547e-01 -9.00907755e-01 -9.01608825e-01 -7.72418678e-01
-2.45358631e-01 -3.64119262e-01 1.24256648e-02 5.88626146e-01
-9.11585912e-02 5.49971402e-01 -3.42190228e-02 3.43431860e-01
8.96128774e-01 2.20383063e-01 5.90301454e-01 -8.62521946e-01
-7.64500678e-01 -3.65108162e-01 -6.53209388e-01 -1.26561642e+00
1.86784238e-01 -9.03607428e-01 5.54233074e-01 -1.57674325e+00
7.37266421e-01 -2.73063570e-01 -3.32417667e-01 6.01286709e-01
-3.80657971e-01 6.84947908e-01 3.38889658e-01 3.62394601e-01
-9.20067012e-01 2.80367851e-01 1.29418945e+00 -3.13873649e-01
-1.03447970e-03 -2.64457345e-01 -5.26980102e-01 7.36914814e-01
2.98860252e-01 -6.66641116e-01 -4.33322757e-01 -5.50983250e-01
2.66279489e-01 7.79152960e-02 9.45127726e-01 -8.02662194e-01
4.99089211e-01 6.88666776e-02 2.57648945e-01 -8.50692511e-01
4.23809558e-01 -6.36657774e-01 -5.21469653e-01 3.70160252e-01
-3.73783410e-01 5.81051446e-02 1.18737519e-01 8.71957958e-01
-2.83765286e-01 -2.51065671e-01 7.81937361e-01 -2.83587605e-01
-8.02114069e-01 4.03200209e-01 2.06672072e-01 2.03187969e-02
1.04055154e+00 -3.74255091e-01 -2.42075130e-01 1.72161628e-02
-6.74116910e-01 2.96796978e-01 5.76969743e-01 3.06451499e-01
7.98051834e-01 -1.09605527e+00 -7.20420182e-01 2.85567373e-01
4.88634139e-01 4.47536856e-01 2.79339887e-02 8.19526374e-01
-2.87472129e-01 3.96449924e-01 1.21010594e-01 -1.17454338e+00
-1.28948891e+00 4.20132786e-01 2.84582198e-01 -2.85581321e-01
-7.41445005e-01 1.38579428e+00 6.68504238e-01 -2.31106713e-01
3.82959753e-01 -3.94508898e-01 3.13735098e-01 -1.67183027e-01
4.29343581e-01 -3.81792754e-01 4.60251234e-02 -5.96147895e-01
-4.37854797e-01 7.66025007e-01 -2.88169205e-01 -4.57382977e-01
1.08253670e+00 1.76227372e-02 -2.23495215e-02 4.03198302e-01
1.23050988e+00 -1.02655008e-01 -1.58892822e+00 -4.68208164e-01
4.77860449e-03 -5.51673412e-01 8.51615816e-02 -7.46481180e-01
-6.88737869e-01 8.38198781e-01 6.49349511e-01 -1.11477874e-01
9.09799278e-01 6.34026349e-01 6.99934125e-01 2.86472201e-01
3.82016063e-01 -8.76862884e-01 7.59972930e-01 4.15407389e-01
6.35029316e-01 -1.59114933e+00 -2.64308248e-02 -5.03159285e-01
-3.51648659e-01 6.77443504e-01 4.91753131e-01 -3.42943072e-01
2.48582259e-01 2.94402182e-01 -1.40622243e-01 -4.92533445e-01
-4.76672888e-01 -4.49979901e-01 7.00485170e-01 4.91107613e-01
3.71270806e-01 -6.90222010e-02 -2.61240005e-02 2.55353421e-01
6.07772171e-02 -3.01196307e-01 -3.99352424e-02 9.18058157e-01
-6.99936867e-01 -1.12945557e+00 -2.78990239e-01 3.47314477e-01
-2.80687451e-01 -3.99792314e-01 -3.13744605e-01 6.47220075e-01
2.07436025e-01 6.48776174e-01 3.54004622e-01 -1.69730887e-01
3.97615612e-01 -1.77965626e-01 7.63435900e-01 -1.16139162e+00
-4.24193084e-01 4.38852347e-02 -2.80748516e-01 -6.20738506e-01
-4.13563222e-01 -5.73573172e-01 -1.34510529e+00 2.71045387e-01
-5.17388225e-01 4.46737930e-03 6.91749275e-01 1.00787556e+00
1.17607817e-01 3.12189966e-01 5.70450783e-01 -9.12683129e-01
-4.87171650e-01 -5.34225523e-01 -4.78900403e-01 5.53283811e-01
6.47688150e-01 -5.66834688e-01 -1.85527712e-01 1.89499557e-01] | [10.328133583068848, 1.439555048942566] |
ee2d56e4-6e72-4fa8-aa7f-9333f92942f0 | action-understanding-with-multiple-classes-of | 1704.08723 | null | http://arxiv.org/abs/1704.08723v1 | http://arxiv.org/pdf/1704.08723v1.pdf | Action Understanding with Multiple Classes of Actors | Despite the rapid progress, existing works on action understanding focus
strictly on one type of action agent, which we call actor---a human adult,
ignoring the diversity of actions performed by other actors. To overcome this
narrow viewpoint, our paper marks the first effort in the computer vision
community to jointly consider algorithmic understanding of various types of
actors undergoing various actions. To begin with, we collect a large annotated
Actor-Action Dataset (A2D) that consists of 3782 short videos and 31 temporally
untrimmed long videos. We formulate the general actor-action understanding
problem and instantiate it at various granularities: video-level single- and
multiple-label actor-action recognition, and pixel-level actor-action
segmentation. We propose and examine a comprehensive set of graphical models
that consider the various types of interplay among actors and actions. Our
findings have led us to conclusive evidence that the joint modeling of actor
and action improves performance over modeling each of them independently, and
further improvement can be obtained by considering the multi-scale natural in
video understanding. Hence, our paper concludes the argument of the value of
explicit consideration of various actors in comprehensive action understanding
and provides a dataset and a benchmark for later works exploring this new
problem. | ['Chenliang Xu', 'Caiming Xiong', 'Jason J. Corso'] | 2017-04-27 | null | null | null | null | ['action-understanding'] | ['computer-vision'] | [ 7.33215690e-01 2.60585636e-01 -5.34872055e-01 -4.33092177e-01
-4.10519779e-01 -6.21900797e-01 9.19875681e-01 -2.74117708e-01
-2.44196191e-01 3.99310559e-01 5.76293230e-01 -4.99412715e-02
-1.78098232e-01 -1.49196431e-01 -7.38993645e-01 -7.08204031e-01
-1.23992674e-01 4.02357638e-01 2.94946015e-01 2.17410251e-01
1.08447656e-01 5.49498498e-01 -1.49204242e+00 4.99058276e-01
3.84525597e-01 6.10718846e-01 -2.22558439e-01 9.31553006e-01
2.66221255e-01 1.50902677e+00 -4.89976406e-01 -4.00752127e-01
4.55820084e-01 -7.29856074e-01 -1.17290926e+00 9.00575459e-01
6.36843145e-01 -7.19609082e-01 -6.30921721e-01 8.09665322e-01
2.71599203e-01 2.11183414e-01 6.62273765e-01 -1.64294875e+00
-3.51679862e-01 7.38024235e-01 -6.79985404e-01 4.64389324e-01
6.80403590e-01 4.60342735e-01 1.06660497e+00 -3.71109694e-01
8.79827321e-01 1.58957136e+00 3.56715590e-01 5.72602630e-01
-7.33120322e-01 -2.91999549e-01 7.18472481e-01 4.87220615e-01
-8.33117127e-01 -4.78844285e-01 8.42250288e-01 -6.09026372e-01
9.73897278e-01 1.25013098e-01 9.91610110e-01 1.37197292e+00
-1.62810028e-01 1.67507207e+00 8.70184541e-01 -3.91618669e-01
-5.96085601e-02 -4.68671232e-01 1.61329299e-01 7.17423618e-01
-3.69804591e-04 -6.00586161e-02 -5.72911859e-01 1.47249654e-01
1.06965029e+00 5.74329309e-02 -4.43160674e-03 -3.44287187e-01
-1.60863137e+00 4.53295648e-01 -1.12608448e-01 3.32572609e-01
-4.25597906e-01 6.54349685e-01 5.76615930e-01 1.06293820e-01
3.52895886e-01 2.26843581e-01 -4.60262060e-01 -6.12425983e-01
-7.80073822e-01 2.03348786e-01 7.03350008e-01 8.36386383e-01
3.73185694e-01 9.57104266e-02 -1.05589181e-01 3.76718670e-01
2.53217787e-01 2.11601838e-01 -3.19397002e-02 -1.78176868e+00
4.12215054e-01 6.80051267e-01 2.04872668e-01 -9.58316684e-01
-3.65263224e-01 2.18088120e-01 -4.88985866e-01 5.62618673e-02
8.40689600e-01 -1.58398032e-01 -8.39506209e-01 1.60789084e+00
6.14976287e-01 7.45043278e-01 -9.21008829e-03 7.13978052e-01
6.82158411e-01 4.63339686e-01 3.56222481e-01 -3.96270990e-01
1.39323437e+00 -1.53953767e+00 -9.63195801e-01 -1.90846771e-01
8.17165494e-01 -3.91368657e-01 5.81647933e-01 4.37854052e-01
-1.15306294e+00 -6.97370231e-01 -6.91532075e-01 -4.62941118e-02
-2.66572416e-01 1.64513171e-01 1.01859558e+00 6.11856163e-01
-9.14488137e-01 3.57634574e-01 -1.03467357e+00 -5.63447297e-01
6.57576442e-01 2.76001304e-01 -5.53236008e-01 -1.58084527e-01
-9.18287933e-01 6.49159789e-01 3.15198511e-01 8.05125386e-02
-1.35007691e+00 -4.33052719e-01 -8.11801672e-01 -3.34959567e-01
8.85222495e-01 -6.17530346e-01 1.31040382e+00 -1.67970240e+00
-1.33118916e+00 1.01550877e+00 -1.05169535e-01 -4.50830370e-01
7.84517050e-01 -5.09038568e-01 -3.02851498e-01 6.79103792e-01
-3.43294963e-02 7.65149295e-01 7.62067795e-01 -1.18719471e+00
-6.57329023e-01 -4.83277470e-01 7.45714068e-01 4.11612511e-01
1.63115822e-02 4.44240183e-01 -8.39231074e-01 -7.95414031e-01
-2.21229717e-01 -1.13369274e+00 -2.27467239e-01 -5.69023192e-03
-3.26807946e-01 -3.83442432e-01 7.14717329e-01 -5.48114359e-01
1.30233264e+00 -1.99237883e+00 4.98912543e-01 -4.02987212e-01
3.83521289e-01 3.26600373e-01 -2.90811390e-01 4.76792425e-01
-3.05949837e-01 1.07988566e-01 4.33933772e-02 -2.95728534e-01
-4.75293323e-02 4.81734037e-01 5.40884063e-02 7.26665795e-01
7.66908601e-02 1.12955785e+00 -1.12537372e+00 -8.74501586e-01
5.09058774e-01 2.79073566e-01 -3.39457154e-01 2.08099186e-02
-3.70096445e-01 7.38602281e-01 -1.04203546e+00 1.01458824e+00
2.45690867e-01 -3.94916236e-01 2.52216965e-01 4.32967255e-03
1.16704807e-01 -3.16161931e-01 -1.36236155e+00 1.76565993e+00
6.12765215e-02 7.96030521e-01 1.00281678e-01 -1.17871380e+00
1.90260828e-01 4.97288436e-01 1.20784593e+00 -3.26368034e-01
1.03673793e-01 -2.69792169e-01 1.52584016e-01 -1.01268566e+00
1.85196295e-01 4.84786630e-02 -1.27659962e-01 6.67501986e-01
8.10501724e-03 9.21692625e-02 4.87609208e-01 4.31915492e-01
1.25264668e+00 5.33149540e-01 5.87049007e-01 7.73364082e-02
6.37220263e-01 -6.60785735e-02 4.13394123e-01 1.01221490e+00
-7.75917232e-01 4.41378534e-01 7.58373499e-01 -6.72892511e-01
-7.68337369e-01 -6.74244761e-01 2.38163993e-01 1.40481651e+00
2.90102333e-01 -3.29163909e-01 -9.04655695e-01 -1.08096731e+00
-1.13204651e-01 2.98532516e-01 -8.90796840e-01 3.76827091e-01
-8.44484627e-01 -5.17020583e-01 7.87538350e-01 9.67374384e-01
5.41252971e-01 -1.21298707e+00 -8.27014387e-01 7.89614990e-02
-2.83639401e-01 -1.42096519e+00 -3.38044941e-01 -2.09124923e-01
-8.40677917e-01 -1.65499592e+00 -6.37235343e-01 -3.52899343e-01
5.05861998e-01 3.40160340e-01 1.18822873e+00 2.18916014e-01
-3.06411862e-01 1.31183362e+00 -6.88949227e-01 -3.15823019e-01
-3.75632465e-01 -4.71948057e-01 3.45972590e-02 4.46057796e-01
3.41809034e-01 -3.46703082e-01 -6.37317896e-01 5.52646875e-01
-9.73221600e-01 1.78489402e-01 5.47873914e-01 3.37535232e-01
3.56133759e-01 2.26560906e-02 2.34897077e-01 -8.60731363e-01
6.78873137e-02 -4.23381031e-01 -2.00474754e-01 5.99879920e-01
5.66253699e-02 -1.60734072e-01 2.26389527e-01 -5.63801646e-01
-1.33673763e+00 5.49043596e-01 2.11474895e-01 -4.93769854e-01
-6.01960599e-01 -9.99625847e-02 -3.42604697e-01 5.51254116e-03
1.45991936e-01 1.23184793e-01 -2.91872501e-01 -3.62802923e-01
7.29215205e-01 2.98367321e-01 3.95668924e-01 -6.11647487e-01
3.41770142e-01 8.52221370e-01 5.11368550e-03 -9.43271995e-01
-9.19400811e-01 -7.12830245e-01 -1.21299171e+00 -9.09991801e-01
1.44953716e+00 -9.87943113e-01 -8.35751772e-01 8.25253248e-01
-1.25097728e+00 -5.75468600e-01 -3.11676681e-01 4.99971300e-01
-1.00446153e+00 8.08008969e-01 -5.62836409e-01 -9.12866473e-01
4.31164563e-01 -1.19552279e+00 1.29614580e+00 2.07299106e-02
-2.49887690e-01 -1.32625139e+00 3.01591810e-02 1.04617000e+00
-3.17146182e-01 6.92561626e-01 3.86237442e-01 -6.91849530e-01
-8.04989278e-01 -1.42255187e-01 -1.46145284e-01 3.62826735e-01
2.89289020e-02 3.24356854e-01 -7.43270338e-01 9.72725973e-02
-1.36477247e-01 -4.78776455e-01 9.05236363e-01 5.85805893e-01
1.10710490e+00 3.72329215e-03 -3.34557831e-01 1.96092218e-01
1.01752472e+00 4.38343614e-01 6.06549025e-01 6.76531643e-02
9.09146965e-01 6.66557252e-01 9.06026721e-01 5.66502571e-01
3.47335905e-01 7.91085005e-01 5.86250782e-01 -1.18534915e-01
-3.23400706e-01 -1.64367393e-01 5.73967457e-01 4.14005429e-01
-8.34416330e-01 -6.23682320e-01 -8.36195469e-01 6.14704311e-01
-2.19996691e+00 -1.38475752e+00 -3.30807030e-01 1.48599827e+00
3.62018287e-01 -1.27100900e-01 4.67777759e-01 1.93459019e-01
5.37379265e-01 8.10654819e-01 -6.21585846e-01 -5.31127304e-03
1.43175982e-02 -3.44798535e-01 3.37182432e-01 3.82481426e-01
-1.43748868e+00 1.00986695e+00 7.45390558e+00 8.53936374e-01
-4.64524299e-01 1.98574793e-02 6.16590500e-01 -4.92769778e-02
1.73801899e-01 5.67320623e-02 -6.42474592e-01 1.51046872e-01
6.65574491e-01 1.50225863e-01 2.30953142e-01 7.76734531e-01
4.83870625e-01 -4.57943767e-01 -1.46562886e+00 8.27152908e-01
4.19084489e-01 -1.06257999e+00 5.77747859e-02 1.96403079e-02
8.30983877e-01 -3.80237162e-01 -2.76807398e-01 5.98136969e-02
4.53300208e-01 -9.22898650e-01 7.39542365e-01 6.21615350e-01
3.00736457e-01 -1.93137541e-01 3.98532420e-01 3.44338506e-01
-1.42164171e+00 -2.46332839e-01 2.82333761e-01 -4.32458639e-01
4.97623652e-01 -2.71043833e-02 -2.61725932e-01 5.69022059e-01
3.56491119e-01 1.36515057e+00 -4.96392012e-01 4.97707158e-01
-3.83587271e-01 5.19348860e-01 -5.76328151e-02 2.84647733e-01
5.01411378e-01 -2.81141818e-01 4.42137510e-01 1.19551682e+00
-2.09784791e-01 7.00888395e-01 4.75478023e-01 1.82598412e-01
1.43785179e-01 -1.06665522e-01 -6.09578252e-01 -4.80669945e-01
-5.86290620e-02 9.34888363e-01 -1.14254808e+00 -7.65840352e-01
-8.29282045e-01 1.03161919e+00 5.18105552e-02 4.53806967e-01
-1.18434870e+00 4.39605385e-01 6.37010038e-01 1.14070289e-01
3.10756207e-01 -3.89793068e-01 -3.53051089e-02 -1.35150123e+00
-1.92605913e-01 -1.12809026e+00 7.13226140e-01 -7.25046992e-01
-7.30775297e-01 2.88944785e-02 5.56711495e-01 -1.04391921e+00
-2.53006756e-01 -6.83575630e-01 -3.57932955e-01 -8.86911899e-02
-1.11925340e+00 -1.37434459e+00 -2.07698047e-01 7.90010154e-01
1.31784654e+00 1.06468879e-01 2.22750157e-01 3.36623192e-01
-6.83501840e-01 7.70541281e-02 -2.16513678e-01 2.76589125e-01
3.69281679e-01 -1.11563110e+00 1.22704573e-01 9.39109445e-01
7.29856253e-01 2.49489442e-01 6.41689718e-01 -6.92688465e-01
-1.43418789e+00 -7.81246185e-01 6.83869660e-01 -1.12382030e+00
7.63617992e-01 -7.81669468e-02 -4.57951009e-01 1.26572347e+00
2.29289562e-01 9.60261002e-02 4.82707620e-01 -1.68151245e-01
-6.87052161e-02 3.68816674e-01 -7.35971153e-01 4.72280085e-01
1.59783363e+00 -4.18578118e-01 -7.39305973e-01 5.52460849e-01
5.31633675e-01 -1.82893112e-01 -9.28429604e-01 4.12718058e-01
6.99619412e-01 -1.25967860e+00 1.24567389e+00 -1.07993305e+00
6.60951555e-01 -1.91771939e-01 -5.01154996e-02 -4.91155595e-01
6.08913191e-02 -5.03662050e-01 -4.34233248e-01 9.66308832e-01
-2.04214051e-01 1.42207528e-02 8.59393656e-01 9.03691113e-01
2.46880949e-02 -7.48510420e-01 -8.29420805e-01 -7.09381342e-01
-3.36731315e-01 -8.05194676e-01 1.79136656e-02 9.01718259e-01
-1.20738305e-01 -7.57549629e-02 -8.90181184e-01 -5.00327675e-04
4.86292601e-01 -5.04646590e-03 1.08206952e+00 -8.96158099e-01
-4.08245742e-01 -3.29297096e-01 -5.01416862e-01 -1.52309406e+00
4.87367570e-01 -3.80475491e-01 -1.58582240e-01 -1.46596003e+00
4.23525423e-01 1.25339642e-01 -9.93823335e-02 4.97862846e-01
-1.64402917e-01 2.50279546e-01 3.33993763e-01 1.37682959e-01
-1.38687479e+00 3.75257313e-01 1.61283612e+00 -1.68276340e-01
3.42182890e-02 1.04174510e-01 -2.76525021e-01 1.36528039e+00
1.21006653e-01 -3.46192956e-01 -6.21711433e-01 -4.44086313e-01
-1.06846362e-01 3.71276349e-01 5.18239260e-01 -7.76477396e-01
8.77831355e-02 -6.52584970e-01 1.39126673e-01 -4.00549054e-01
4.43322867e-01 -8.76230896e-01 2.22308487e-01 3.31016272e-01
-5.78946531e-01 -1.09390914e-01 -1.85167283e-01 1.06493783e+00
-3.02259684e-01 -5.61753772e-02 5.97917438e-01 -5.23013234e-01
-1.27453339e+00 3.06574136e-01 -5.75665653e-01 2.06014171e-01
1.66063857e+00 -5.90447187e-01 3.44608678e-03 -5.70011914e-01
-1.17001808e+00 2.56815583e-01 3.57574373e-01 5.70497930e-01
2.90246308e-01 -1.04685903e+00 -5.77880859e-01 -1.69840947e-01
7.17576407e-03 -4.54101354e-01 6.85069978e-01 1.05035651e+00
-4.50650990e-01 3.84830505e-01 -2.09643528e-01 -5.10800838e-01
-1.76142418e+00 6.03230357e-01 2.34804720e-01 -4.43885952e-01
-7.11688936e-01 8.62947285e-01 4.92137998e-01 1.08278081e-01
4.45280313e-01 -3.09165776e-01 -3.56302083e-01 2.83086061e-01
5.05092502e-01 7.93348193e-01 -7.79907823e-01 -1.15797162e+00
-4.38136965e-01 6.67115748e-01 2.60279208e-01 4.42559505e-03
1.18578196e+00 -2.78429836e-01 9.19136181e-02 6.12566769e-01
8.42681766e-01 -3.18715662e-01 -1.74548936e+00 -9.91182849e-02
-5.78034334e-02 -7.07618952e-01 -4.41158473e-01 -6.10476851e-01
-1.19031334e+00 8.37510824e-01 2.22874999e-01 2.60655880e-01
1.18619084e+00 3.53686005e-01 4.62356269e-01 3.24298799e-01
2.43494809e-01 -1.26019752e+00 7.92602718e-01 3.59649152e-01
7.05328882e-01 -1.44292152e+00 4.10222352e-01 -4.63620871e-01
-9.64214325e-01 1.09367681e+00 5.81567287e-01 -1.56298876e-02
3.78800660e-01 7.07633197e-02 7.28081688e-02 -4.44515020e-01
-6.73238635e-01 -3.86271924e-01 2.25604281e-01 4.87203687e-01
1.56373426e-01 -2.17122342e-02 -2.29376599e-01 2.19994769e-01
6.74934030e-01 1.57416210e-01 5.64048231e-01 1.00061989e+00
-2.12288454e-01 -1.13448024e+00 -1.92735747e-01 1.50702819e-01
-5.63566804e-01 2.90276051e-01 -6.69864178e-01 1.21220887e+00
3.71326238e-01 1.05557632e+00 -1.34765223e-01 -1.01973020e-01
1.55734226e-01 1.12292342e-01 6.70662880e-01 -3.73333961e-01
-5.06750524e-01 3.03172600e-02 3.05264562e-01 -9.91486609e-01
-1.49816442e+00 -1.07544899e+00 -1.11222363e+00 2.87481770e-02
-8.01628605e-02 -3.92936498e-01 2.87528902e-01 1.34647679e+00
7.39044026e-02 4.86189246e-01 3.38754267e-01 -1.06589842e+00
-1.51443541e-01 -6.66020572e-01 -7.77152896e-01 8.01667988e-01
3.18199933e-01 -8.27212334e-01 -4.41914797e-01 6.83007360e-01] | [8.38583755493164, 0.5241627097129822] |
b3b2ccab-d400-4442-aa2c-ac271b94114c | geofault-a-well-founded-fault-ontology-for | 2302.07059 | null | https://arxiv.org/abs/2302.07059v1 | https://arxiv.org/pdf/2302.07059v1.pdf | GeoFault: A well-founded fault ontology for interoperability in geological modeling | Geological modeling currently uses various computer-based applications. Data harmonization at the semantic level by means of ontologies is essential for making these applications interoperable. Since geo-modeling is currently part of multidisciplinary projects, semantic harmonization is required to model not only geological knowledge but also to integrate other domain knowledge at a general level. For this reason, the domain ontologies used for describing geological knowledge must be based on a sound ontology background to ensure the described geological knowledge is integratable. This paper presents a domain ontology: GeoFault, resting on the Basic Formal Ontology BFO (Arp et al., 2015) and the GeoCore ontology (Garcia et al., 2020). It models the knowledge related to geological faults. Faults are essential to various industries but are complex to model. They can be described as thin deformed rock volumes or as spatial arrangements resulting from the different displacements of geological blocks. At a broader scale, faults are currently described as mere surfaces, which are the components of complex fault arrays. The reference to the BFO and GeoCore package allows assigning these various fault elements to define ontology classes and their logical linkage within a consistent ontology framework. The GeoFault ontology covers the core knowledge of faults 'strico sensu,' excluding ductile shear deformations. This considered vocabulary is essentially descriptive and related to regional to outcrop scales, excluding microscopic, orogenic, and tectonic plate structures. The ontology is molded in OWL 2, validated by competency questions with two use cases, and tested using an in-house ontology-driven data entry application. The work of GeoFault provides a solid framework for disambiguating fault knowledge and a foundation of fault data integration for the applications and the users. | ['Martin Giese', 'Mara Abel', 'Anita Torabi', 'Michel Perrin', 'Yuanwei Qu'] | 2023-02-14 | null | null | null | null | ['data-integration'] | ['knowledge-base'] | [-4.21461821e-01 6.49820745e-01 1.76821932e-01 -3.01125795e-01
-9.01390761e-02 -5.17421424e-01 6.86162055e-01 3.35568100e-01
-1.18719697e-01 7.90049434e-01 3.49205732e-01 -3.68617028e-01
-7.33742118e-01 -1.58553219e+00 -5.84400415e-01 -3.82908702e-01
-3.97491813e-01 8.21654737e-01 9.15233791e-01 -1.01501167e+00
4.00829047e-01 5.65314233e-01 -2.14122033e+00 3.01647693e-01
1.04298913e+00 7.93592811e-01 4.91674870e-01 2.20542401e-01
-3.91043991e-01 6.89237893e-01 -5.46256542e-01 9.13800970e-02
-1.64649785e-01 2.03460038e-01 -1.41251135e+00 -3.60497415e-01
1.27245456e-01 5.36101572e-02 1.34547725e-01 1.04858482e+00
2.77106255e-01 -3.39544117e-02 6.29656792e-01 -8.56987238e-01
-6.75413609e-01 8.28275919e-01 6.06953204e-01 -1.76475614e-01
5.54360986e-01 -3.05996358e-01 6.30089462e-01 -7.76102662e-01
8.66866171e-01 1.43792546e+00 5.34687340e-01 1.18391536e-01
-7.39066899e-01 -1.94488108e-01 -1.75747171e-01 5.13512850e-01
-1.35167134e+00 -6.73441142e-02 6.77766725e-02 -8.78879488e-01
1.01394010e+00 4.02388692e-01 9.30930972e-01 3.09811383e-01
4.71245319e-01 -3.40191633e-01 1.08022487e+00 -5.72544515e-01
7.23498583e-01 -7.97255710e-02 2.65292376e-01 3.36870968e-01
9.37400281e-01 -3.00815135e-01 -4.80558455e-01 -7.65008852e-02
5.71288884e-01 -2.64513880e-01 -4.44643706e-01 2.04335712e-02
-5.00917017e-01 4.97427821e-01 2.88612485e-01 8.69701803e-01
-1.37820199e-01 3.03145200e-01 1.91575631e-01 2.09434241e-01
6.50696084e-02 3.25881302e-01 -4.06841695e-01 -1.73155174e-01
-4.77875978e-01 5.66365063e-01 1.27758968e+00 6.67810917e-01
1.19642997e+00 2.36296535e-01 8.91118705e-01 7.13782787e-01
1.06294823e+00 4.67580557e-01 4.50435489e-01 -9.99267578e-01
5.31442389e-02 1.00799930e+00 1.68530241e-01 -7.33357728e-01
-3.73954743e-01 5.12302965e-02 -1.41920477e-01 6.52849972e-01
7.07876086e-02 3.54072958e-01 -9.35998678e-01 1.11002028e+00
3.60528857e-01 -2.16940135e-01 6.22876346e-01 5.22714317e-01
1.13433278e+00 2.28661910e-01 3.40906471e-01 5.73466361e-01
1.79478383e+00 7.12662265e-02 -7.46438503e-01 -1.21232308e-01
9.77231324e-01 -4.22907948e-01 8.67111683e-01 1.19628116e-01
-8.07111323e-01 -1.39967203e-01 -1.24224293e+00 6.05146214e-02
-1.18751335e+00 -6.54184639e-01 4.55924779e-01 8.63450348e-01
-1.19175267e+00 6.81132495e-01 -6.76227450e-01 -7.10381985e-01
9.41013023e-02 1.19142629e-01 -4.88982707e-01 3.45082730e-02
-2.15892148e+00 1.52231252e+00 1.00355065e+00 1.36290878e-01
-6.96960986e-01 -6.21398509e-01 -1.04442430e+00 4.26274724e-02
1.51441962e-01 -4.62032437e-01 6.95665419e-01 -2.61826098e-01
-9.69911277e-01 9.27626550e-01 3.07429016e-01 -5.12569070e-01
5.86864650e-01 -1.62138581e-01 -9.81342435e-01 1.17488496e-01
6.46296382e-01 3.71339694e-02 -3.63694429e-01 -1.36925650e+00
-7.36705005e-01 -2.54404217e-01 4.18229729e-01 6.33019432e-02
4.98175398e-02 -7.41961822e-02 -3.46075632e-02 -2.78023630e-01
3.33082736e-01 -3.61419678e-01 -1.66727319e-01 -4.82678056e-01
2.66562611e-01 -1.33851707e-01 9.27669287e-01 -8.66975546e-01
1.32911742e+00 -1.87390625e+00 -1.83213949e-01 5.31333268e-01
-9.92438644e-02 -3.19350995e-02 7.66523480e-01 9.36132610e-01
4.47437391e-02 5.98823905e-01 -7.16392040e-01 6.29835248e-01
2.53370017e-01 8.02214146e-01 -6.80164471e-02 4.36940938e-02
-2.76545674e-01 2.21941367e-01 -8.78183663e-01 -3.43030721e-01
3.15746397e-01 3.53556246e-01 -3.35530013e-01 -4.02323008e-01
-3.08687955e-01 1.23014443e-01 -5.15274167e-01 9.24221277e-01
8.67365181e-01 1.98359132e-01 5.43603301e-01 -3.94918360e-02
-5.44850945e-01 1.04750507e-01 -1.54334557e+00 1.49796844e+00
-5.60497344e-01 1.24673016e-01 9.45397615e-02 -8.27729881e-01
1.29271638e+00 7.59928167e-01 2.74637908e-01 -5.22719026e-01
-1.41965881e-01 1.14034629e+00 -3.71589065e-02 -1.02306712e+00
5.83898306e-01 -2.30241582e-01 -1.75505936e-01 -1.31279394e-01
5.33493310e-02 -6.43030822e-01 4.97938186e-01 5.54967672e-02
8.37846994e-01 6.16561413e-01 4.01924670e-01 -1.34964204e+00
7.02597439e-01 2.83315837e-01 4.26864713e-01 3.49379182e-01
4.25369233e-01 1.30812958e-01 3.81723315e-01 -6.57681882e-01
-9.05422151e-01 -1.19122970e+00 -9.24734354e-01 5.07153928e-01
3.42266113e-01 -3.95203084e-01 -8.24401796e-01 4.07371074e-02
2.02060178e-01 5.32381594e-01 -7.60498941e-01 1.34749427e-01
-2.95827270e-01 -7.94796646e-01 7.21836865e-01 2.40121499e-01
5.85503578e-01 -9.08440053e-01 -1.07508445e+00 6.00379467e-01
-7.46618062e-02 -7.34505057e-01 8.95452559e-01 8.82095918e-02
-8.25786114e-01 -1.44963443e+00 5.47589026e-02 -5.41691065e-01
3.05033147e-01 -4.57596958e-01 1.01874518e+00 5.24878204e-01
-1.39257222e-01 2.93372571e-01 -7.98829794e-01 -5.37450910e-01
-5.74479043e-01 -1.15616471e-01 -2.17291757e-01 -3.53665650e-01
1.54435620e-01 -6.83807552e-01 -1.69235304e-01 3.78609300e-01
-1.50935352e+00 -2.81593710e-01 4.15854231e-02 2.44126290e-01
1.56826094e-01 1.83480710e-01 8.56522202e-01 -7.50137806e-01
3.48474860e-01 -8.26893389e-01 -4.02391464e-01 4.33326304e-01
-4.16197449e-01 2.36047748e-02 1.91416889e-01 4.11789149e-01
-1.11339748e+00 -6.25075638e-01 -5.25249422e-01 5.63491046e-01
-3.90558481e-01 1.17294812e+00 -8.27087224e-01 -1.03496805e-01
7.34617710e-01 -2.09619835e-01 -1.67886734e-01 -4.84004468e-01
-1.23394705e-01 7.99429476e-01 4.46611911e-01 -1.22232521e+00
4.09609824e-01 7.88573325e-01 1.44794390e-01 -1.12571692e+00
-9.01836827e-02 -1.30357310e-01 -6.87119365e-01 -6.79247260e-01
1.08948731e+00 -6.97881401e-01 -5.21286905e-01 2.17736155e-01
-9.32049870e-01 -3.70540172e-01 -3.29474360e-01 4.67444569e-01
-3.78816754e-01 1.75478190e-01 -1.37114048e-01 -8.41165483e-01
4.43795435e-02 -1.10494614e+00 5.98803759e-01 1.11385621e-01
-3.52553785e-01 -1.44468176e+00 2.05566093e-01 9.44054499e-02
4.52816188e-01 8.00943613e-01 9.57410455e-01 -4.39595699e-01
-3.86726826e-01 1.63451076e-01 2.99271613e-01 2.81398445e-02
4.87634748e-01 1.47804573e-01 -7.84762502e-01 2.21362725e-01
-2.18094483e-01 2.14450628e-01 5.31410336e-01 -1.60461694e-01
3.34784955e-01 -7.19463527e-02 -1.89743653e-01 2.40302578e-01
2.07556605e+00 6.77916944e-01 1.30083704e+00 1.30511117e+00
3.72953743e-01 1.11236084e+00 6.70518160e-01 3.65556389e-01
5.83995700e-01 4.80545878e-01 7.78356552e-01 3.12744260e-01
8.00020695e-02 4.11450684e-01 3.43369134e-02 4.13355321e-01
-9.65314090e-01 2.32931182e-01 -1.77742946e+00 7.88000107e-01
-1.88633776e+00 -1.17242622e+00 -7.96107590e-01 2.07953882e+00
5.76738358e-01 8.92122090e-02 -5.73517978e-01 2.68357873e-01
6.26243472e-01 -1.32939041e-01 1.89796224e-01 -4.68973726e-01
-4.72450793e-01 5.21500111e-01 4.22609061e-01 9.87013042e-01
-6.97874546e-01 9.32526231e-01 5.70677805e+00 3.56643111e-01
-9.42236841e-01 3.21225375e-01 -1.90969169e-01 7.40422785e-01
-7.46092498e-01 6.71055198e-01 -6.74134135e-01 4.88887101e-01
9.82277811e-01 -3.49926174e-01 1.01572417e-01 6.32032216e-01
4.55500722e-01 -4.18961853e-01 -3.52854252e-01 1.59422889e-01
-2.20824867e-01 -1.83360624e+00 1.56612366e-01 2.36385971e-01
4.99864995e-01 1.66519731e-01 -9.45052803e-01 -1.88942194e-01
6.73150182e-01 -9.52091873e-01 1.37593842e+00 9.91275489e-01
6.45541430e-01 -4.64453429e-01 1.00484645e+00 -3.44347715e-01
-1.60321498e+00 3.90441231e-02 -2.37559363e-01 -2.37553537e-01
5.19614339e-01 5.38580179e-01 -1.76874146e-01 1.55312836e+00
1.44485950e+00 5.28203428e-01 -4.09382939e-01 7.59984553e-01
-2.86797345e-01 2.48772174e-01 -3.59951347e-01 4.11949575e-01
1.66870937e-01 -5.34673929e-01 2.08517462e-01 1.00951767e+00
6.93607628e-01 4.53597168e-03 -1.02159373e-01 5.09523630e-01
5.22108078e-01 3.37188303e-01 -4.92294580e-01 3.72071534e-01
9.63160872e-01 6.93865538e-01 -7.17806816e-01 -5.50186574e-01
-3.32033396e-01 9.39710140e-02 -2.35212877e-01 2.07004949e-01
-5.06057084e-01 -6.13736033e-01 6.93068385e-01 7.20153987e-01
-5.66191971e-02 -2.13326085e-02 -2.79024869e-01 -8.24482739e-01
-2.49769557e-02 -4.87392813e-01 3.21325332e-01 -8.48295212e-01
-7.88414955e-01 4.28221762e-01 4.85181868e-01 -1.06594932e+00
-6.66190609e-02 -8.43664944e-01 -5.04799187e-01 9.09472227e-01
-1.29875720e+00 -1.52038395e+00 -7.05944240e-01 3.01815033e-01
-3.00701350e-01 1.59424350e-01 1.01934636e+00 5.17805338e-01
6.05706908e-02 -5.54897964e-01 2.72212923e-01 -1.62359819e-01
4.82439190e-01 -1.38151848e+00 4.51817364e-02 6.34839296e-01
-8.36433530e-01 4.06223089e-01 8.69033337e-01 -1.09905827e+00
-7.33796835e-01 -9.70996678e-01 1.00134480e+00 -1.48151338e-01
9.75221038e-01 2.35777758e-02 -1.47120440e+00 7.32220769e-01
1.02080494e-01 -1.48423821e-01 6.94802582e-01 -3.23973984e-01
-8.91844705e-02 -2.38808617e-02 -1.30830407e+00 5.53708315e-01
1.06533456e+00 -4.35538948e-01 -1.10522270e+00 6.69399649e-02
1.24804184e-01 -2.57076859e-01 -1.94468963e+00 6.68771327e-01
7.46016383e-01 -1.15175295e+00 8.50437641e-01 -1.67977110e-01
3.14348608e-01 -7.68223226e-01 -6.06018245e-01 -8.68039608e-01
-2.16649361e-02 -2.31095348e-02 4.79091495e-01 1.20963717e+00
1.89803109e-01 -1.38103545e+00 2.84079671e-01 5.35915792e-01
-9.70172167e-01 -2.28020817e-01 -1.23474896e+00 -9.81597185e-01
4.40382034e-01 -4.78491008e-01 1.40239787e+00 1.28800118e+00
2.87215084e-01 -5.89161873e-01 3.77388388e-01 5.03186464e-01
4.89043266e-01 -3.98845226e-01 4.40264434e-01 -1.97850156e+00
1.03455625e-01 -4.92357731e-01 -9.57544088e-01 1.38694361e-01
-1.28738537e-01 -8.31308842e-01 -1.65421814e-01 -2.33859730e+00
-7.31633186e-01 -8.41337860e-01 1.61826223e-01 5.28761566e-01
4.24416304e-01 -2.54355520e-02 -9.49387476e-02 5.34724355e-01
3.38474602e-01 2.32890218e-01 7.91176856e-01 8.67395774e-02
6.58338144e-02 -9.05141950e-01 -3.85965228e-01 8.59330118e-01
8.89264703e-01 -4.82399434e-01 -3.32283713e-02 -3.44040960e-01
8.04882050e-01 -3.18297982e-01 4.12721276e-01 -1.37859750e+00
3.12469844e-02 -4.16103125e-01 -4.51730162e-01 -3.32260221e-01
-2.36837915e-03 -1.07782054e+00 1.21348453e+00 1.04401839e+00
5.93804121e-01 -2.82865137e-01 1.71042532e-01 2.53125280e-02
-5.68299830e-01 -9.04125571e-01 7.35966206e-01 -4.77372438e-01
-1.27488828e+00 -1.55469939e-01 -5.85032284e-01 -9.58989412e-02
1.10154951e+00 -8.87082040e-01 -5.23099363e-01 2.54268825e-01
-1.18782341e+00 1.43571198e-01 1.10015285e+00 1.25528559e-01
1.78870127e-01 -1.20647907e+00 -5.80446899e-01 -8.97460058e-02
3.24839264e-01 2.79690892e-01 6.69023931e-01 3.21278244e-01
-1.55250466e+00 2.83910364e-01 -6.73108339e-01 -2.45698780e-01
-6.67369604e-01 -3.25176418e-01 9.44859505e-01 3.14625120e-03
-3.60081732e-01 1.28213018e-01 -8.62239003e-02 -6.39418006e-01
-5.61046124e-01 -4.19603407e-01 -7.94292688e-01 3.25890988e-01
3.86302322e-01 6.30498886e-01 4.00052994e-01 -9.43344176e-01
-5.69804132e-01 8.65674138e-01 8.19363177e-01 -2.30867967e-01
1.55337465e+00 -1.59755461e-02 -7.33183205e-01 6.16942048e-01
4.02174681e-01 9.74695385e-02 -6.47322237e-01 2.94828773e-01
6.46811962e-01 -3.68858814e-01 -4.29136723e-01 -7.16750026e-01
-6.04870319e-01 3.86101663e-01 4.15831059e-01 5.40552080e-01
5.82189620e-01 2.44120836e-01 -1.83072194e-01 1.90740049e-01
8.91286016e-01 -1.30361688e+00 -5.93789756e-01 6.74826205e-01
1.25069201e+00 -4.30451900e-01 7.33071789e-02 -8.14095557e-01
-1.55169247e-02 1.46482372e+00 4.82198149e-01 2.57419683e-02
1.13776076e+00 4.79541451e-01 1.37647420e-01 -6.56430483e-01
-1.44716635e-01 -4.50358182e-01 -1.68330982e-01 7.28471100e-01
4.72164750e-01 1.06296465e-01 -9.26005244e-01 5.08034468e-01
-3.68366867e-01 1.79942995e-01 8.89193237e-01 1.55581462e+00
-1.12317312e+00 -1.43946230e+00 -1.13322401e+00 2.55203277e-01
-3.70338738e-01 7.34208971e-02 -7.54202679e-02 1.29987431e+00
8.95688117e-01 6.84234977e-01 2.56119877e-01 1.02463536e-01
6.57218814e-01 7.83547238e-02 2.26964921e-01 -8.26017201e-01
-6.12012744e-01 -4.11084890e-01 5.90248585e-01 1.90251917e-02
-7.29633808e-01 -5.49653709e-01 -1.93584454e+00 -3.14021975e-01
-1.60134938e-02 7.00326920e-01 7.70729363e-01 1.29577971e+00
-2.35018064e-03 5.95894694e-01 -3.09207678e-01 -7.71660984e-01
2.08145618e-01 -1.03540051e+00 -1.14226198e+00 3.82865459e-01
-5.61708748e-01 -1.10704172e+00 -3.41851413e-01 3.49058688e-01] | [9.164803504943848, 7.988325595855713] |
be60cdbf-b5be-486d-a79c-8cd47488c9df | virapart-a-text-refinement-framework-for-asr | 2110.09086 | null | https://arxiv.org/abs/2110.09086v3 | https://arxiv.org/pdf/2110.09086v3.pdf | ViraPart: A Text Refinement Framework for Automatic Speech Recognition and Natural Language Processing Tasks in Persian | The Persian language is an inflectional subject-object-verb language. This fact makes Persian a more uncertain language. However, using techniques such as Zero-Width Non-Joiner (ZWNJ) recognition, punctuation restoration, and Persian Ezafe construction will lead us to a more understandable and precise language. In most of the works in Persian, these techniques are addressed individually. Despite that, we believe that for text refinement in Persian, all of these tasks are necessary. In this work, we proposed a ViraPart framework that uses embedded ParsBERT in its core for text clarifications. First, used the BERT variant for Persian following by a classifier layer for classification procedures. Next, we combined models outputs to output cleartext. In the end, the proposed model for ZWNJ recognition, punctuation restoration, and Persian Ezafe construction performs the averaged F1 macro scores of 96.90%, 92.13%, and 98.50%, respectively. Experimental results show that our proposed approach is very effective in text refinement for the Persian language. | ['Saeed Bibak', 'Hamed Babaei Giglou', 'Saman Jamalabbasi', 'Milad Molazadeh', 'Narges Farokhshad'] | 2021-10-18 | null | null | null | null | ['punctuation-restoration'] | ['natural-language-processing'] | [ 2.44307548e-01 2.06976622e-01 2.24507555e-01 -1.96018592e-01
-5.43604136e-01 -6.75299525e-01 6.40049160e-01 3.43101472e-01
-4.70151842e-01 1.05078697e+00 1.71151876e-01 -4.14047927e-01
-2.19316974e-01 -8.12679112e-01 -1.95908979e-01 -4.69124913e-01
3.68185729e-01 7.15523124e-01 3.93464625e-01 -4.08470362e-01
8.72273922e-01 4.26730871e-01 -1.36375821e+00 2.45857403e-01
1.22243786e+00 4.11325991e-01 4.66080993e-01 5.36005259e-01
-6.61723435e-01 6.88803017e-01 -1.00618887e+00 -5.93434453e-01
-5.18282354e-02 -1.09713823e-02 -1.17418551e+00 -1.80259168e-01
1.21845277e-02 1.63940698e-01 2.69604564e-01 9.84828234e-01
2.51411200e-01 3.11092697e-02 1.00803149e+00 -7.42777467e-01
-5.56155324e-01 1.52983725e+00 -6.35025859e-01 6.22515799e-03
3.02994370e-01 -3.68658006e-01 1.03102398e+00 -8.23240161e-01
4.76257682e-01 1.51914179e+00 5.32959044e-01 2.81558692e-01
-1.01786733e+00 -7.62442946e-01 -1.17317133e-01 3.60386103e-01
-1.47309268e+00 -2.04779059e-01 6.50555551e-01 -3.22371632e-01
1.13947833e+00 6.58132911e-01 4.45252538e-01 4.88374293e-01
5.61864078e-01 9.14960444e-01 1.31298447e+00 -7.65963912e-01
-1.40777826e-01 1.02243848e-01 6.61026001e-01 1.51363447e-01
4.60545629e-01 -4.26600248e-01 -2.91501343e-01 2.77345508e-01
2.06230149e-01 -2.20763624e-01 -2.06815943e-01 5.20865083e-01
-8.40973794e-01 4.80093658e-01 -1.37024865e-01 7.89757133e-01
-2.40698501e-01 -5.47389388e-01 4.12697792e-01 -1.63400605e-01
5.89223504e-02 5.11699617e-01 -6.09376729e-01 -2.80024678e-01
-1.13135338e+00 3.36296529e-01 7.89687991e-01 1.01480103e+00
5.02085805e-01 9.69132483e-02 -1.16605885e-01 9.63684380e-01
3.13849181e-01 7.34928310e-01 6.85177028e-01 -5.14527678e-01
5.77618361e-01 8.83812845e-01 1.40005648e-01 -8.33916426e-01
-3.53790581e-01 -2.30236456e-01 -9.76037741e-01 1.63613245e-01
1.22036479e-01 1.61833093e-01 -9.49812710e-01 1.20000005e+00
9.97161642e-02 -4.98824626e-01 3.32082421e-01 2.51917690e-01
7.08710015e-01 9.13745582e-01 3.44278723e-01 -3.43575537e-01
1.73361957e+00 -7.10897267e-01 -9.69420373e-01 -1.83992147e-01
4.87868398e-01 -1.45024037e+00 1.23825312e+00 7.98929811e-01
-1.00407028e+00 -5.54997742e-01 -1.17066073e+00 -1.94057167e-01
-4.07144397e-01 3.50646675e-01 -1.61863305e-02 7.98662841e-01
-6.08905911e-01 6.58129156e-01 -7.40166306e-01 -6.05415523e-01
-1.12076953e-01 2.26021320e-01 -1.51923463e-01 3.20463866e-01
-1.02485156e+00 9.69104886e-01 1.19371617e+00 9.35157984e-02
4.37614620e-02 -1.54452994e-01 -5.85534573e-01 2.57374913e-01
1.74494952e-01 -1.70882434e-01 1.35709453e+00 -5.30737340e-01
-1.35670674e+00 7.98647463e-01 -5.67822978e-02 -5.65937042e-01
3.41755569e-01 -4.67652410e-01 -4.31554496e-01 -2.41024762e-01
2.68764198e-02 4.66838092e-01 6.23428881e-01 -1.11825359e+00
-1.11919832e+00 -6.02814674e-01 -3.31267029e-01 2.18324766e-01
6.18797988e-02 1.94077313e-01 -1.96128622e-01 -8.78458679e-01
3.92130703e-01 -6.10187471e-01 3.86541903e-01 -8.85563970e-01
-5.93274772e-01 -8.02257061e-01 8.83289814e-01 -1.33146942e+00
1.65001357e+00 -2.14877772e+00 -1.74048856e-01 3.42740625e-01
1.49091259e-02 6.15896821e-01 1.05476059e-01 4.09632057e-01
-1.06973581e-01 4.30405736e-01 -4.44363892e-01 -1.18537202e-01
1.60826087e-01 3.07607204e-01 -3.36785018e-01 -1.09130768e-02
1.56842247e-01 6.32703125e-01 -3.51556152e-01 -7.52600491e-01
3.63407791e-01 3.23431075e-01 -3.20852637e-01 5.04592210e-02
-7.06979334e-02 -1.46532297e-01 -1.09711163e-01 6.28549695e-01
1.02884185e+00 6.27810538e-01 3.43511403e-01 -1.63957104e-01
-5.16119361e-01 4.42758799e-01 -1.67017949e+00 9.70425069e-01
-3.23907882e-01 2.31784612e-01 -2.57600266e-02 -8.78056347e-01
1.20968044e+00 9.89618301e-02 3.27789113e-02 -6.12255037e-01
3.24096262e-01 3.53175372e-01 -4.18987311e-02 -4.66259092e-01
1.08511496e+00 -8.43929406e-03 -1.59800604e-01 1.96647346e-01
-1.25273675e-01 -4.83429819e-01 5.98468840e-01 1.84517249e-01
5.81994176e-01 2.62745172e-01 9.50188160e-01 -3.98062676e-01
1.01522410e+00 5.16105862e-03 7.01374650e-01 4.47931051e-01
1.60834372e-01 3.85927916e-01 3.01740468e-01 -9.22234654e-02
-1.02213240e+00 -1.10873640e+00 -3.76446754e-01 8.00778866e-01
-1.77229047e-01 -9.78730083e-01 -1.01515865e+00 -5.17690420e-01
-3.29337418e-01 1.46904016e+00 3.72205637e-02 2.71340698e-01
-9.78242576e-01 -6.12310112e-01 7.37832487e-01 3.07195574e-01
7.30794191e-01 -1.41979885e+00 -3.20633948e-01 4.09978062e-01
-4.11741942e-01 -7.04571962e-01 -2.51712888e-01 2.37334460e-01
-9.19403911e-01 -1.15996456e+00 -4.12095696e-01 -1.07692087e+00
1.53751805e-01 -1.55444844e-02 8.97530675e-01 9.39148739e-02
5.64435683e-02 -9.20302272e-02 -6.72524631e-01 -6.29133523e-01
-6.16260350e-01 2.68402994e-01 -3.06684852e-01 -6.01105690e-01
5.10940552e-01 -4.60709631e-01 -5.63413948e-02 8.22459012e-02
-1.05471468e+00 -1.59753077e-02 7.09216893e-01 5.19225895e-01
5.67550957e-01 3.73132855e-01 2.80351102e-01 -1.20602524e+00
7.99021244e-01 7.21998960e-02 -5.63142002e-01 2.69589782e-01
-1.00648677e+00 3.42459470e-01 8.04348707e-01 -4.38471362e-02
-1.40465093e+00 -1.55768925e-02 -5.50441027e-01 6.13304138e-01
-3.87625456e-01 4.38947588e-01 -5.38389742e-01 4.99644250e-01
6.50211036e-01 4.73712444e-01 -3.22638035e-01 -6.61188185e-01
2.67103851e-01 1.27341723e+00 6.86999321e-01 -4.86777157e-01
7.46111453e-01 -2.86783010e-01 -2.48097792e-01 -1.23965037e+00
-3.90722424e-01 -3.75617772e-01 -6.39101267e-01 1.16822049e-01
7.88200855e-01 -4.13308412e-01 -7.70641565e-01 5.53601384e-01
-1.43585694e+00 1.95923477e-01 -2.44450629e-01 1.44276321e-01
-2.74648249e-01 9.76070821e-01 -5.14641941e-01 -1.06203103e+00
-1.05158925e+00 -8.74780595e-01 8.38658214e-01 3.90946597e-01
-7.74442136e-01 -5.71640849e-01 -1.32922575e-01 4.29389358e-01
9.63594019e-02 -4.80673313e-01 1.34879780e+00 -9.26717162e-01
-1.60316542e-01 1.07345976e-01 -2.19242901e-01 4.87526506e-01
1.78016827e-01 2.42151812e-01 -6.77687466e-01 2.17244759e-01
1.30830184e-01 1.29034445e-01 5.79009056e-01 2.00103775e-01
9.64741170e-01 -6.19616508e-01 -1.80182382e-01 3.56121302e-01
1.14232349e+00 7.28967965e-01 1.13769019e+00 5.49760044e-01
3.81273687e-01 4.85496938e-01 9.25715506e-01 5.38081884e-01
5.19838929e-01 2.76471466e-01 1.20430641e-01 3.04772913e-01
-2.08529457e-01 -1.30309209e-01 4.84918803e-01 1.04237604e+00
-4.73355390e-02 -3.86685431e-01 -1.09301353e+00 4.90464985e-01
-1.56047070e+00 -8.89520586e-01 -5.37649989e-01 1.95568204e+00
1.06094825e+00 3.30943614e-01 -1.67107001e-01 9.49006557e-01
7.24614918e-01 4.39510047e-02 2.66691238e-01 -8.01452637e-01
-3.96881133e-01 6.49955869e-01 3.41325551e-01 7.80855417e-01
-1.12612975e+00 1.39220977e+00 5.14761829e+00 1.07486129e+00
-9.41376686e-01 -2.37521276e-01 1.54273853e-01 6.84494913e-01
-2.46173993e-01 4.15127948e-02 -1.11929250e+00 3.21338415e-01
6.81131124e-01 -2.40318224e-01 3.81769031e-01 5.21149278e-01
2.29382604e-01 -3.91053915e-01 -5.44115186e-01 1.29058778e+00
1.95831597e-01 -7.48211503e-01 4.11578625e-01 -5.69051921e-01
3.08153003e-01 -4.54575807e-01 -3.80484998e-01 4.60991234e-01
2.52408266e-01 -9.69260037e-01 9.52223241e-01 4.68525589e-01
3.44609141e-01 -1.03118312e+00 9.41909790e-01 5.76475561e-01
-1.19892812e+00 1.42448530e-01 -3.77473146e-01 9.96927395e-02
-5.47816083e-02 4.21476036e-01 -1.20689595e+00 6.18966460e-01
5.26402950e-01 4.66633499e-01 -7.61984825e-01 1.06005239e+00
-5.79811096e-01 8.53776693e-01 -2.94806451e-01 -3.31894249e-01
-1.50524661e-01 -3.64057839e-01 6.64442599e-01 1.60206175e+00
4.40546006e-01 2.58467615e-01 -3.91121209e-02 6.46195590e-01
3.31706911e-01 8.07575881e-01 -1.12103909e-01 -1.25915006e-01
5.16540349e-01 1.17061353e+00 -8.15501809e-01 -3.07820380e-01
1.36680946e-01 9.84186828e-01 1.29624456e-01 7.73582235e-02
-4.60521936e-01 -8.68768692e-01 1.68920845e-01 9.53532010e-02
9.23117697e-02 -3.37350696e-01 -8.18932474e-01 -9.37706709e-01
4.82205674e-02 -1.03525651e+00 6.54839456e-01 -7.53182828e-01
-1.07640564e+00 6.44716382e-01 -7.05423206e-02 -8.44863296e-01
-1.75656989e-01 -7.38709331e-01 -4.75282997e-01 1.03308070e+00
-1.23695815e+00 -1.23815727e+00 1.85329933e-02 3.82070899e-01
9.83589590e-01 -1.24192774e-01 9.90649462e-01 5.45336306e-02
-4.95842844e-01 3.04210842e-01 1.29107907e-01 2.00137347e-01
8.22370529e-01 -1.52904189e+00 1.23236388e-01 1.05152011e+00
1.34614453e-01 6.49269521e-01 8.99014831e-01 -9.49897707e-01
-9.71650779e-01 -9.13803935e-01 1.60828841e+00 -2.67945260e-01
4.34335560e-01 -1.60811082e-01 -9.72019494e-01 7.62758672e-01
5.21805346e-01 -1.13599896e+00 2.61225432e-01 7.52006769e-02
6.41810969e-02 -2.64510304e-01 -1.09366536e+00 9.02901053e-01
5.72425306e-01 2.15607602e-02 -1.39743578e+00 -7.72700086e-02
4.21779871e-01 -2.89352477e-01 -5.15648007e-01 9.26568359e-02
2.47189805e-01 -6.91235781e-01 5.35606802e-01 -1.23831794e-01
1.48604363e-01 -6.35069311e-01 -1.89332709e-01 -1.22625864e+00
-3.44325215e-01 -5.37768424e-01 5.60732901e-01 1.77106619e+00
6.29592478e-01 -4.49722618e-01 3.86701733e-01 2.40535766e-01
-2.98773259e-01 9.13868695e-02 -6.76487088e-01 -4.17289793e-01
1.99790746e-01 -6.72415376e-01 5.71748972e-01 7.41225123e-01
2.65464962e-01 8.43120515e-01 -1.40122622e-01 3.44669037e-02
3.35170180e-01 -7.36011714e-02 4.52788562e-01 -1.52613795e+00
7.55203664e-02 -4.52672899e-01 -2.84191877e-01 -7.26172805e-01
1.04360461e-01 -1.08269691e+00 7.47838393e-02 -1.66434467e+00
-1.00587006e-03 -3.25275272e-01 1.06804706e-01 6.42792284e-01
-3.25446934e-01 -1.50780663e-01 4.16852772e-01 9.45221782e-02
-1.04264565e-01 4.14564282e-01 8.61857951e-01 -4.11276221e-01
-4.73231316e-01 3.09795029e-02 -5.81153393e-01 8.32174122e-01
1.25367260e+00 -5.67875624e-01 -1.42952368e-01 -1.40921146e-01
1.18234493e-01 -2.42233247e-01 -2.41762340e-01 -1.12467611e+00
-3.94141153e-02 -3.99325550e-01 3.77700388e-01 -1.09031761e+00
-1.13317752e-02 -7.92388976e-01 4.01796214e-02 4.66319084e-01
1.23460047e-01 3.22305828e-01 3.82609725e-01 -1.92263111e-01
-1.18535377e-01 -5.03193974e-01 8.17330956e-01 7.07951114e-02
-7.84538269e-01 -3.62845927e-01 -8.11077952e-01 6.81397617e-02
8.94632339e-01 -3.76281470e-01 -1.66378528e-01 1.10681660e-01
-4.77601141e-01 1.10062874e-05 1.40463002e-02 3.00053686e-01
5.72589576e-01 -8.42132568e-01 -7.94759035e-01 1.47124365e-01
-6.73206225e-02 2.25097649e-02 6.62065297e-02 5.45585275e-01
-9.32723463e-01 6.51868582e-01 -3.84249985e-01 -3.26404274e-01
-1.70045877e+00 3.76071870e-01 1.38058065e-04 -2.51176745e-01
-4.16486800e-01 2.82821745e-01 -2.27206275e-01 -6.43132031e-01
2.15744555e-01 -6.46413207e-01 -8.86411011e-01 1.98674053e-01
4.78111267e-01 5.64338624e-01 1.92970902e-01 -6.98163986e-01
-2.50289828e-01 4.55799699e-01 -1.62229687e-01 -2.83674091e-01
1.10474837e+00 -2.82544285e-01 -5.94393969e-01 3.40804845e-01
2.45269641e-01 7.85458505e-01 -2.17851281e-01 5.44317178e-02
6.36925757e-01 1.26228072e-02 -3.32638115e-01 -1.18033731e+00
-2.71730185e-01 7.91845024e-01 4.02674288e-01 2.13345826e-01
1.18571675e+00 -3.05857301e-01 7.86694348e-01 5.37067533e-01
8.47102404e-02 -1.34457135e+00 -7.78678954e-01 1.23913276e+00
9.90441442e-01 -6.38921261e-01 -1.89728051e-01 -5.90153396e-01
-6.08079135e-01 1.44663739e+00 6.51308835e-01 2.20972568e-01
4.96268958e-01 4.99153048e-01 1.04637913e-01 2.58464783e-01
-4.27570045e-01 -4.32174392e-02 1.82984412e-01 6.67296529e-01
9.37349439e-01 4.55312759e-01 -1.40224302e+00 1.15178847e+00
-1.09892237e+00 -1.72010407e-01 6.95721924e-01 8.76190245e-01
-9.48289990e-01 -1.49975526e+00 -9.13871884e-01 2.85306007e-01
-5.77092826e-01 -4.01644409e-01 -7.05374658e-01 9.25847411e-01
2.69661874e-01 1.12495100e+00 -7.01206326e-02 -3.64225149e-01
4.92936373e-01 4.37297314e-01 3.53145778e-01 -5.22133768e-01
-7.96806931e-01 2.61403799e-01 2.46736348e-01 -1.94807146e-02
-3.92822996e-02 -6.11450732e-01 -1.78073692e+00 -6.27691209e-01
-1.19743586e-01 5.03170729e-01 5.66207409e-01 1.03332126e+00
-1.05151519e-01 2.76204139e-01 9.37047750e-02 -2.17043310e-01
-6.35886669e-01 -1.44485819e+00 -9.72298622e-01 2.33342335e-01
-2.00533494e-01 -1.79409131e-01 -7.77284577e-02 2.68875062e-01] | [10.539507865905762, 10.21861457824707] |
ce1a946b-fdb6-4467-8c44-9e61907ebdfb | onepose-one-shot-object-pose-estimation | 2205.12257 | null | https://arxiv.org/abs/2205.12257v1 | https://arxiv.org/pdf/2205.12257v1.pdf | OnePose: One-Shot Object Pose Estimation without CAD Models | We propose a new method named OnePose for object pose estimation. Unlike existing instance-level or category-level methods, OnePose does not rely on CAD models and can handle objects in arbitrary categories without instance- or category-specific network training. OnePose draws the idea from visual localization and only requires a simple RGB video scan of the object to build a sparse SfM model of the object. Then, this model is registered to new query images with a generic feature matching network. To mitigate the slow runtime of existing visual localization methods, we propose a new graph attention network that directly matches 2D interest points in the query image with the 3D points in the SfM model, resulting in efficient and robust pose estimation. Combined with a feature-based pose tracker, OnePose is able to stably detect and track 6D poses of everyday household objects in real-time. We also collected a large-scale dataset that consists of 450 sequences of 150 objects. | ['Xiaowei Zhou', 'Guofeng Zhang', 'Hongcheng Zhao', 'Xingyi He', 'Siyu Zhang', 'ZiHao Wang', 'Jiaming Sun'] | 2022-05-24 | null | http://openaccess.thecvf.com//content/CVPR2022/html/Sun_OnePose_One-Shot_Object_Pose_Estimation_Without_CAD_Models_CVPR_2022_paper.html | http://openaccess.thecvf.com//content/CVPR2022/papers/Sun_OnePose_One-Shot_Object_Pose_Estimation_Without_CAD_Models_CVPR_2022_paper.pdf | cvpr-2022-1 | ['6d-pose-estimation-1'] | ['computer-vision'] | [-8.65854919e-02 5.45052905e-03 -1.90141827e-01 -3.92671138e-01
-5.55766046e-01 -5.57044923e-01 2.79448241e-01 3.73092256e-02
-1.95306256e-01 5.30357398e-02 -2.37920254e-01 1.79760441e-01
2.12828238e-02 -5.94205022e-01 -1.16216648e+00 -3.18179876e-01
6.26984285e-03 9.00675893e-01 5.21475136e-01 3.31032813e-01
1.11486472e-01 9.69490170e-01 -1.62849414e+00 -1.79573864e-01
3.79401088e-01 1.39377224e+00 5.17695546e-01 5.32945633e-01
9.23795477e-02 3.25624973e-01 -4.67033952e-01 -2.23776340e-01
3.81758302e-01 6.68461621e-02 -3.17863673e-01 4.14475828e-01
1.16170490e+00 -4.50906456e-01 -4.96160597e-01 9.39171731e-01
5.00136495e-01 1.77115053e-01 3.78015071e-01 -1.65827692e+00
-5.27094126e-01 2.11860240e-01 -6.27767801e-01 -2.80840397e-01
7.65172601e-01 1.73219204e-01 6.12081826e-01 -1.12605262e+00
8.14846992e-01 1.40566587e+00 9.33313251e-01 3.62799615e-01
-1.23690951e+00 -5.63209176e-01 4.31746870e-01 2.80911177e-01
-1.62562990e+00 -2.84332424e-01 9.19441342e-01 -2.32871592e-01
1.05126619e+00 7.30903447e-02 1.04770565e+00 1.00807512e+00
-1.88140154e-01 6.60665929e-01 6.48685813e-01 -2.63963461e-01
2.64053285e-01 -9.11763459e-02 -6.71008900e-02 8.89898062e-01
3.47047836e-01 -2.37119481e-01 -4.89470929e-01 -1.94705930e-02
1.20384395e+00 5.56505203e-01 -1.36147708e-01 -1.49999571e+00
-1.48668742e+00 5.11440456e-01 1.08140481e+00 3.47649902e-02
-3.97450179e-01 7.53351867e-01 1.38032204e-02 -2.72177577e-01
1.59551531e-01 1.09278448e-01 -4.89988059e-01 1.75747141e-01
-6.54077888e-01 2.69426461e-02 6.83356881e-01 1.54839540e+00
7.60879099e-01 -8.24898779e-02 -8.23122561e-02 3.57488751e-01
6.57856703e-01 1.00221682e+00 5.68387024e-02 -1.16556740e+00
3.88351828e-01 7.72786021e-01 2.00624362e-01 -1.28802061e+00
-4.96279061e-01 -4.17912543e-01 -3.97774667e-01 2.03768294e-02
3.13518584e-01 4.97803986e-01 -1.06701887e+00 1.48974013e+00
7.34690309e-01 3.65824014e-01 -5.96167982e-01 1.17505717e+00
9.00237381e-01 3.31450343e-01 -1.40191630e-01 1.63798735e-01
1.26111066e+00 -1.06788588e+00 -3.23895574e-01 -4.21141624e-01
2.10995406e-01 -4.22960430e-01 8.85508418e-01 1.37734607e-01
-1.00945151e+00 -7.29924202e-01 -1.01475465e+00 -2.69603074e-01
-4.28560257e-01 2.51137435e-01 7.86208570e-01 4.32412744e-01
-1.07442319e+00 4.40665215e-01 -9.78733659e-01 -7.21649349e-01
5.08488715e-01 6.32873952e-01 -6.08667791e-01 -2.70174921e-01
-4.06167924e-01 8.89852405e-01 3.34241837e-01 2.31534258e-01
-1.16786218e+00 -5.17263532e-01 -1.12997890e+00 -1.52057454e-01
6.90801680e-01 -9.41087484e-01 1.11893594e+00 -5.54220736e-01
-1.39014721e+00 8.31842899e-01 -2.46485695e-01 -8.69175792e-02
5.18147886e-01 -5.00516653e-01 1.36676356e-01 2.32399240e-01
3.26177418e-01 7.04029858e-01 1.05492544e+00 -1.53942275e+00
-3.20138127e-01 -6.84050977e-01 8.18614885e-02 2.28291169e-01
8.04968849e-02 -2.44440302e-01 -1.17095625e+00 -1.66456565e-01
8.63522887e-01 -8.54773223e-01 -1.98036239e-01 7.27273583e-01
-4.20000672e-01 -2.01054633e-01 9.49277759e-01 -4.20167744e-01
4.01868761e-01 -1.98098481e+00 1.73882559e-01 3.19745302e-01
3.74769747e-01 7.39954924e-03 -2.38834202e-01 7.36715570e-02
1.07120551e-01 -2.66006887e-01 4.14043128e-01 -6.96976125e-01
2.52216309e-01 1.36599556e-01 5.57002239e-02 7.58883357e-01
-4.02881391e-02 1.20543277e+00 -1.04789937e+00 -4.63328689e-01
5.81800997e-01 7.15078175e-01 -5.75365424e-01 2.34456003e-01
-4.20372695e-01 1.98594302e-01 -4.60973471e-01 1.03170109e+00
7.97764361e-01 -6.09204173e-01 1.71327814e-02 -6.24992728e-01
1.88440353e-01 1.00681491e-01 -1.27447498e+00 2.27262568e+00
-3.24021816e-01 2.40736142e-01 -2.64190836e-03 -7.27661848e-01
8.07460904e-01 -9.11623389e-02 6.08505070e-01 -4.60701674e-01
1.57016739e-01 4.83475886e-02 -5.65727830e-01 -2.45774642e-01
2.80079216e-01 5.36424160e-01 -4.88005280e-02 1.30182743e-01
3.68963212e-01 -3.34738374e-01 -2.23227501e-01 9.93163139e-03
9.34664011e-01 6.75375879e-01 2.51530379e-01 1.36063978e-01
2.59740174e-01 -2.09941372e-01 2.97807366e-01 7.48883009e-01
-1.99015304e-01 8.04292202e-01 -7.34155104e-02 -5.94215155e-01
-1.01464474e+00 -1.46893036e+00 1.14181817e-01 1.03608406e+00
6.12084210e-01 -3.84475589e-01 -5.11616707e-01 -7.46695161e-01
4.63637054e-01 2.19906956e-01 -4.87644732e-01 3.95171754e-02
-5.21867335e-01 1.01455718e-01 -5.03662489e-02 8.28027904e-01
5.40001512e-01 -8.44929576e-01 -8.30678999e-01 1.14101330e-02
-4.52517979e-02 -1.20617199e+00 -6.87628865e-01 2.49952778e-01
-7.75411665e-01 -1.32209742e+00 -5.72070956e-01 -9.09869909e-01
9.87556398e-01 4.87362713e-01 1.09711564e+00 -7.42183253e-02
-3.07100236e-01 8.35953891e-01 -1.65722579e-01 -2.58043468e-01
1.98487639e-01 1.77256558e-02 2.48172894e-01 -4.54243273e-02
4.62483019e-01 -4.16311324e-01 -6.13220632e-01 4.19390976e-01
-9.11827907e-02 -5.61503768e-02 5.42535365e-01 3.53675455e-01
8.84435713e-01 -5.01387835e-01 9.33396630e-03 -1.75536275e-02
-1.53805032e-01 -6.48760051e-02 -8.91049504e-01 2.70154089e-01
-9.43796039e-02 -2.50556827e-01 9.50072855e-02 -7.88581491e-01
-3.66927773e-01 7.37127721e-01 2.82787740e-01 -1.18977213e+00
-1.90608487e-01 5.62392361e-02 -4.08529997e-01 -4.03072298e-01
4.40287411e-01 1.08397290e-01 -6.44121990e-02 -6.19848907e-01
6.06227219e-01 2.69383907e-01 7.73372531e-01 -3.16572666e-01
1.14181924e+00 4.23571795e-01 3.76936258e-03 -5.85834146e-01
-7.73396552e-01 -5.91244221e-01 -1.12074649e+00 -4.44348574e-01
8.59846771e-01 -1.11079729e+00 -1.16198361e+00 4.25986767e-01
-1.33989704e+00 -2.85311878e-01 -3.32770944e-01 6.06603622e-01
-8.36462677e-01 1.45983055e-01 -3.34174901e-01 -6.23664558e-01
-1.31630346e-01 -9.31505919e-01 1.75628090e+00 1.31977499e-01
-5.99969849e-02 -5.74484408e-01 -4.23124999e-01 1.20691396e-01
1.66057423e-01 4.11606938e-01 2.47506902e-01 -3.11632723e-01
-1.21600580e+00 -4.96204108e-01 -2.77550489e-01 -1.23677172e-01
1.37891367e-01 -2.75248349e-01 -8.30383003e-01 -4.62949216e-01
-1.08374648e-01 -3.10909510e-01 4.16003525e-01 3.38132560e-01
1.10357809e+00 -7.68082738e-02 -8.01538646e-01 8.96840334e-01
1.44878709e+00 8.68192501e-03 2.04441816e-01 2.63337851e-01
1.22703600e+00 2.29081541e-01 6.75293744e-01 2.58041978e-01
5.41426003e-01 1.01475108e+00 8.75160158e-01 3.22268158e-02
-1.69653654e-01 -5.96798360e-01 2.18157157e-01 5.22707999e-01
-3.60949337e-02 8.94369334e-02 -8.17155659e-01 3.97230923e-01
-1.80550480e+00 -6.47868156e-01 -3.11680231e-02 2.21423316e+00
3.44160736e-01 -5.23353182e-02 1.56226873e-01 -2.19558209e-01
7.95066059e-01 -6.45493045e-02 -8.58012557e-01 4.01144207e-01
3.28606099e-01 -1.65159144e-02 6.33815765e-01 2.47690216e-01
-1.24920690e+00 9.81640577e-01 6.22834253e+00 3.58045071e-01
-9.01301384e-01 4.15747025e-04 6.30331859e-02 -2.45420530e-01
1.36550367e-01 -1.89783230e-01 -8.80345225e-01 2.05478013e-01
3.22981447e-01 1.08750232e-01 3.53745997e-01 1.55445588e+00
-1.82895958e-01 -6.39620349e-02 -1.52558637e+00 1.48789001e+00
3.87825817e-01 -9.82388198e-01 -1.83463097e-01 6.01011179e-02
4.76170242e-01 1.06510550e-01 -1.11853220e-01 1.12721860e-01
1.35839373e-01 -8.43992710e-01 1.13394761e+00 6.07888043e-01
8.02521527e-01 -5.89647233e-01 4.90496218e-01 3.50367218e-01
-1.51292050e+00 1.51362445e-04 -5.96030295e-01 3.86167616e-02
1.23429924e-01 2.30838522e-01 -9.04983699e-01 3.86072040e-01
1.02218115e+00 8.33571732e-01 -8.56059194e-01 1.33073628e+00
-2.60033756e-01 -1.41881615e-01 -6.92586482e-01 -8.01902413e-02
-1.64915800e-01 2.94783823e-02 3.84513468e-01 6.44935548e-01
5.31701744e-01 -3.02320957e-01 5.88844419e-01 1.01512802e+00
-1.30353555e-01 -3.04450810e-01 -6.91112936e-01 2.63753235e-01
7.18111157e-01 1.36436498e+00 -9.43096101e-01 -4.31724221e-01
-1.86606020e-01 1.25005412e+00 3.43316734e-01 2.74233401e-01
-9.98707175e-01 -9.09353495e-02 5.30567884e-01 1.95465028e-01
8.24255168e-01 -5.09743631e-01 1.66684121e-01 -1.21641850e+00
3.48718226e-01 -6.52536750e-01 -1.09735340e-01 -1.29461324e+00
-1.14192855e+00 2.85582751e-01 6.33639023e-02 -1.15494967e+00
-2.31743962e-01 -7.06900775e-01 -7.65943304e-02 6.70238912e-01
-1.00641119e+00 -1.62700009e+00 -9.12583292e-01 7.75458932e-01
3.42845380e-01 2.72079021e-01 6.49030745e-01 1.13242038e-01
-1.61140069e-01 3.97393733e-01 -3.05781305e-01 2.38201618e-01
5.22503614e-01 -1.16750050e+00 6.63179219e-01 5.26536167e-01
3.34334940e-01 7.51949728e-01 3.85510921e-01 -7.59175539e-01
-1.78690124e+00 -1.27508688e+00 6.40808403e-01 -1.10115695e+00
4.39058214e-01 -9.95567501e-01 -5.10003746e-01 9.71491694e-01
-3.15516561e-01 4.91105855e-01 -1.37778390e-02 -1.20103799e-01
-4.93571967e-01 -2.39483654e-01 -1.09630334e+00 3.25479180e-01
1.70070601e+00 -5.90858400e-01 -5.23483932e-01 4.91683334e-01
9.92410481e-01 -9.54496682e-01 -7.81996787e-01 3.81461173e-01
6.26167655e-01 -6.83173835e-01 1.43970740e+00 -2.73988962e-01
-2.87008107e-01 -6.50945425e-01 -2.95432299e-01 -9.07525957e-01
-4.82235461e-01 -1.51273444e-01 -5.78065634e-01 9.33456957e-01
-2.06115857e-01 -3.84335518e-01 1.05041122e+00 5.72048187e-01
-7.36357942e-02 -5.01623571e-01 -9.64200974e-01 -1.01474369e+00
-6.30090594e-01 -5.74132740e-01 7.57685900e-01 5.32531083e-01
-5.23986220e-01 7.86356702e-02 -1.63349330e-01 5.63949585e-01
9.99201953e-01 3.35499376e-01 1.13454080e+00 -1.27007806e+00
-1.66049302e-01 -1.03428893e-01 -9.78415072e-01 -1.37810326e+00
1.39692854e-02 -7.25612521e-01 2.69774646e-01 -1.71608317e+00
1.03192382e-01 -3.16512257e-01 -4.99870330e-02 6.41448319e-01
1.73340470e-01 5.29537439e-01 4.40927178e-01 5.45747206e-02
-1.05736589e+00 4.93666857e-01 1.07776022e+00 -2.74493217e-01
-5.40106744e-02 -3.03838439e-02 -3.03087056e-01 7.88713992e-01
3.29444021e-01 -4.19875532e-01 -2.74132580e-01 -3.73994023e-01
-5.94634525e-02 -1.01570368e-01 1.01275766e+00 -1.33318031e+00
2.96232939e-01 3.53466794e-02 9.83859420e-01 -1.18812478e+00
6.04432106e-01 -1.24054432e+00 3.43366235e-01 4.96260166e-01
7.72870257e-02 4.88279760e-02 3.65102589e-02 6.85208440e-01
3.65088105e-01 2.52191368e-02 4.54276800e-01 -3.28258216e-01
-9.59645987e-01 7.07982063e-01 2.54099369e-01 -4.75295216e-01
1.25967646e+00 -5.26784897e-01 -4.19096164e-02 -2.85492003e-01
-8.76250505e-01 3.16233575e-01 9.75069046e-01 8.29451382e-01
7.49365747e-01 -1.65943158e+00 -8.78987312e-02 3.39134961e-01
2.73900002e-01 5.41976988e-01 -4.13030982e-02 7.20277488e-01
-5.90047896e-01 4.17414993e-01 -3.52161401e-03 -1.26294637e+00
-1.12075448e+00 1.08365154e+00 4.57490027e-01 2.96440035e-01
-6.67445660e-01 8.18457544e-01 1.86611190e-01 -6.91486955e-01
5.74655175e-01 -6.07926726e-01 1.85961708e-01 -1.63121551e-01
3.63547772e-01 2.12932050e-01 -3.50537859e-02 -8.94389391e-01
-8.24017525e-01 1.14719880e+00 3.35124046e-01 1.39071375e-01
1.31067777e+00 -1.54065922e-01 2.33570114e-02 5.97988009e-01
1.28278959e+00 -2.76255250e-01 -1.67582679e+00 -2.53347486e-01
-1.03725106e-01 -7.18693852e-01 -2.14707211e-01 -5.13947964e-01
-1.09355509e+00 6.98951662e-01 8.60656917e-01 -2.46902347e-01
7.58741796e-01 3.84456426e-01 2.82398522e-01 6.97483659e-01
1.09540951e+00 -9.04666245e-01 6.17349982e-01 4.00280952e-01
1.06635296e+00 -1.20928216e+00 4.05952074e-02 -5.13239563e-01
-1.06400587e-01 9.59089220e-01 8.62689018e-01 -2.09570363e-01
3.67540270e-01 6.43424317e-02 -8.05756003e-02 -3.16120058e-01
-9.60693955e-02 -2.98728764e-01 4.94774610e-01 1.14331186e+00
-1.69811547e-01 -2.28556618e-01 5.77907741e-01 1.50149122e-01
-5.41768335e-02 -3.64046283e-02 -1.30387127e-01 9.85888600e-01
-4.24262404e-01 -6.47538126e-01 -5.65998733e-01 1.51292101e-01
2.94870168e-01 2.72496015e-01 -3.31186503e-01 9.54530299e-01
1.97796404e-01 6.25062406e-01 2.78778464e-01 -3.94155800e-01
5.31846702e-01 3.19145881e-02 8.58994246e-01 -6.68322325e-01
-1.84036329e-01 1.81617737e-01 -3.45522314e-01 -1.30981481e+00
-6.12153471e-01 -6.54620707e-01 -1.15956044e+00 6.80826008e-02
-5.40210843e-01 -3.47123057e-01 8.93038392e-01 6.43474817e-01
5.09797812e-01 2.49379665e-01 3.79756957e-01 -1.74333537e+00
-4.04569566e-01 -8.16610515e-01 -4.12000418e-01 4.64842081e-01
3.28692168e-01 -9.76768792e-01 -1.01393543e-01 -4.73430119e-02] | [7.48704719543457, -2.586268901824951] |
72d850a9-a19b-4aae-94e1-0a64f9f31dd4 | domain-knowledge-informed-self-supervised | 2202.14019 | null | https://arxiv.org/abs/2202.14019v2 | https://arxiv.org/pdf/2202.14019v2.pdf | Domain Knowledge-Informed Self-Supervised Representations for Workout Form Assessment | Maintaining proper form while exercising is important for preventing injuries and maximizing muscle mass gains. Detecting errors in workout form naturally requires estimating human's body pose. However, off-the-shelf pose estimators struggle to perform well on the videos recorded in gym scenarios due to factors such as camera angles, occlusion from gym equipment, illumination, and clothing. To aggravate the problem, the errors to be detected in the workouts are very subtle. To that end, we propose to learn exercise-oriented image and video representations from unlabeled samples such that a small dataset annotated by experts suffices for supervised error detection. In particular, our domain knowledge-informed self-supervised approaches (pose contrastive learning and motion disentangling) exploit the harmonic motion of the exercise actions, and capitalize on the large variances in camera angles, clothes, and illumination to learn powerful representations. To facilitate our self-supervised pretraining, and supervised finetuning, we curated a new exercise dataset, \emph{Fitness-AQA} (\url{https://github.com/ParitoshParmar/Fitness-AQA}), comprising of three exercises: BackSquat, BarbellRow, and OverheadPress. It has been annotated by expert trainers for multiple crucial and typically occurring exercise errors. Experimental results show that our self-supervised representations outperform off-the-shelf 2D- and 3D-pose estimators and several other baselines. We also show that our approaches can be applied to other domains/tasks such as pose estimation and dive quality assessment. | ['Helge Rhodin', 'Amol Gharat', 'Paritosh Parmar'] | 2022-02-28 | null | null | null | null | ['action-quality-assessment', 'action-understanding', 'action-analysis', 'action-assessment', 'pose-contrastive-learning', 'motion-disentanglement', '3d-human-action-recognition'] | ['computer-vision', 'computer-vision', 'computer-vision', 'computer-vision', 'computer-vision', 'computer-vision', 'computer-vision'] | [ 3.10481042e-01 8.33425373e-02 -5.05947113e-01 -2.56975353e-01
-7.83528507e-01 -5.28429091e-01 -9.37671587e-02 -2.60570198e-01
-3.24853659e-01 5.58798909e-01 3.61811250e-01 3.07839632e-01
-3.12114656e-01 -3.34599435e-01 -8.52670014e-01 -3.66201550e-01
-2.06585571e-01 3.22175980e-01 8.90686214e-02 -2.78466940e-01
7.88943097e-02 1.48197174e-01 -1.46488607e+00 1.20384291e-01
7.58564651e-01 7.54718006e-01 9.92565230e-02 9.64142501e-01
5.79658926e-01 5.94377339e-01 -5.60493112e-01 -3.70440096e-01
4.97799456e-01 -7.27893353e-01 -6.08566284e-01 4.86281127e-01
8.91829014e-01 -4.98946398e-01 -6.66293204e-01 6.22456312e-01
9.38674331e-01 5.56894481e-01 3.96029919e-01 -1.05940092e+00
-3.67587239e-01 1.00681208e-01 -7.36601532e-01 4.22860652e-01
5.36327124e-01 6.28658533e-01 6.06634080e-01 -6.19637370e-01
4.04626846e-01 9.34971571e-01 1.24555373e+00 6.80809557e-01
-1.05673981e+00 -6.70306981e-01 7.00872540e-02 1.48145586e-01
-1.32298982e+00 -4.26185846e-01 1.03362226e+00 -5.71524382e-01
3.48823726e-01 3.49128306e-01 1.01337433e+00 1.44549406e+00
6.41163737e-02 8.73827636e-01 1.05855572e+00 -3.13998044e-01
-1.03614479e-03 -3.15743119e-01 -1.11008920e-01 9.75767553e-01
2.57196099e-01 8.73693004e-02 -8.82012606e-01 3.08976769e-01
1.09342527e+00 5.67116104e-02 -3.78109157e-01 -6.14603102e-01
-1.13296568e+00 4.36373591e-01 3.42336863e-01 -9.94869992e-02
-1.78135037e-01 4.36487079e-01 4.11151677e-01 7.04359934e-02
3.52657259e-01 6.68774247e-01 -5.89423299e-01 -6.82770431e-01
-8.33930910e-01 3.87331426e-01 5.42635739e-01 9.83111084e-01
4.16136414e-01 6.77036494e-02 -1.18238166e-01 7.49608576e-01
-1.41851693e-01 2.66500056e-01 5.90215743e-01 -1.29921389e+00
6.09498858e-01 6.12588346e-01 1.45595446e-01 -1.11715078e+00
-5.75943589e-01 -4.18177485e-01 -3.32099587e-01 1.33642823e-01
8.13226640e-01 -3.52979600e-01 -7.14302421e-01 1.45013690e+00
5.73040605e-01 1.32565275e-01 -5.69180548e-01 1.41478169e+00
7.18927443e-01 -6.94386140e-02 -1.32198706e-01 7.36687481e-02
1.34349406e+00 -1.13628900e+00 -6.71799302e-01 -5.78900576e-01
6.37122512e-01 -5.25285900e-01 1.28541934e+00 5.51396012e-01
-1.19840264e+00 -1.14757442e+00 -1.11878431e+00 -1.08504407e-01
1.84961930e-01 5.79393685e-01 5.03640592e-01 7.88209736e-01
-1.64431676e-01 1.05272591e+00 -1.18740308e+00 -2.32961103e-01
2.74266958e-01 3.84175867e-01 -4.72695291e-01 -2.75200028e-02
-7.83393204e-01 7.00644910e-01 1.69289503e-02 4.36348289e-01
-9.01258111e-01 -8.00422251e-01 -1.11114430e+00 -4.57213312e-01
1.04421723e+00 -6.18776143e-01 1.13570523e+00 -1.02129829e+00
-1.62277842e+00 9.44308102e-01 3.27670515e-01 -5.92259355e-02
8.53088379e-01 -1.10818446e+00 -1.40601113e-01 4.44390327e-01
-4.99826521e-02 1.98108122e-01 9.05894876e-01 -8.76326084e-01
5.29542044e-02 -6.13030970e-01 1.09295554e-01 6.33055210e-01
-3.23145598e-01 -3.27059776e-01 -5.63655078e-01 -7.82684028e-01
8.56883153e-02 -1.22022939e+00 -4.81752381e-02 5.29375851e-01
-3.88889253e-01 1.47775471e-01 4.17052984e-01 -1.02770972e+00
1.03800941e+00 -2.26747870e+00 2.93783307e-01 -6.63129753e-03
1.56349987e-01 1.39041498e-01 2.24391863e-01 1.27046198e-01
-7.13025182e-02 -1.83273807e-01 1.96554922e-02 -1.91910818e-01
-9.85357240e-02 3.40682328e-01 3.72035027e-01 8.53478312e-01
4.21966426e-02 7.41191149e-01 -1.10521817e+00 -7.04044938e-01
4.90788013e-01 2.83165365e-01 -6.04869246e-01 3.75737548e-01
-1.87845044e-02 7.89625347e-01 -3.70664567e-01 7.69546866e-01
8.66296291e-02 -4.32090983e-02 4.91199195e-01 -4.90078986e-01
4.14676726e-01 -2.04103324e-03 -1.56770039e+00 2.31603503e+00
-2.60912299e-01 4.56558138e-01 4.53647710e-02 -1.10644019e+00
6.81980014e-01 1.08882904e-01 7.66536474e-01 -4.23077464e-01
1.32630408e-01 5.43274023e-02 -6.04415312e-02 -9.63990510e-01
3.36906284e-01 2.76868958e-02 -9.19001773e-02 9.95037258e-02
2.19451427e-01 -1.49190426e-01 1.81395531e-01 -1.49572805e-01
9.97187734e-01 9.42748845e-01 2.58264184e-01 -1.61413267e-01
3.30092050e-02 2.57892366e-02 6.78484261e-01 3.41337085e-01
-5.45590401e-01 1.08723831e+00 1.49730846e-01 -5.18017054e-01
-7.48078704e-01 -1.03307903e+00 3.57519001e-01 1.18719375e+00
1.06313065e-01 -6.35255218e-01 -7.77913451e-01 -8.33820403e-01
8.40672255e-02 -1.63952522e-02 -6.85014904e-01 -5.07642746e-01
-9.28489447e-01 -3.27160746e-01 5.73127747e-01 1.19812179e+00
3.47511232e-01 -5.11692286e-01 -7.37239540e-01 -5.45381987e-03
-3.44780147e-01 -9.40546751e-01 -9.34405208e-01 -4.09854867e-04
-9.44409966e-01 -1.61099696e+00 -7.26793945e-01 -4.49240535e-01
7.32998669e-01 1.20615393e-01 1.09310889e+00 1.25778718e-02
-7.50199497e-01 7.46622145e-01 -2.84729868e-01 -2.32476473e-01
5.98763935e-02 -7.54487514e-02 4.95386243e-01 -2.05285072e-01
-2.70139687e-02 -6.40158415e-01 -1.07179701e+00 6.02485418e-01
-5.23194790e-01 -2.23909602e-01 3.59168768e-01 8.53118539e-01
7.60681629e-01 -1.42304767e-02 -5.21765463e-02 -6.12923861e-01
3.22265446e-01 1.62922759e-02 -4.06687856e-02 -2.16835067e-02
-2.87716657e-01 -3.99296105e-01 1.69852018e-01 -8.82311344e-01
-8.17315519e-01 3.54116350e-01 -1.46101303e-02 -6.89495802e-01
-1.84076756e-01 4.03660536e-02 -1.96237057e-01 -1.00068497e-02
1.14161515e+00 -9.12220255e-02 3.07746261e-01 -6.09900296e-01
2.24132746e-01 2.58597940e-01 9.98140335e-01 -8.42295587e-01
9.22729135e-01 2.97498822e-01 1.79758817e-02 -7.54657567e-01
-1.26139557e+00 -5.27385533e-01 -9.16205347e-01 -5.81955791e-01
8.33997846e-01 -1.21888900e+00 -7.90799141e-01 3.08614612e-01
-5.11460185e-01 -4.92342353e-01 -7.41449773e-01 9.75962937e-01
-8.27385783e-01 5.70035577e-01 -5.62163174e-01 -9.35736775e-01
-6.40173107e-02 -7.06648707e-01 1.20766115e+00 4.18409079e-01
-6.36053443e-01 -7.95388758e-01 7.96330348e-02 1.02504015e+00
-1.25450447e-01 5.98164737e-01 1.75351024e-01 -3.11767220e-01
-3.24179083e-02 -2.89497435e-01 3.50296110e-01 7.04704046e-01
2.46817976e-01 -3.63210201e-01 -8.69417548e-01 -5.68946779e-01
-2.26681307e-01 -8.49465251e-01 3.46678823e-01 4.17708457e-01
1.39923668e+00 -2.43260950e-01 1.41014168e-02 7.22450495e-01
9.05140936e-01 -5.72251856e-01 4.87838238e-01 3.21117520e-01
1.03218162e+00 7.50092685e-01 9.23662066e-01 4.91103470e-01
2.78618008e-01 7.33629048e-01 3.68560404e-01 -3.09360206e-01
-3.68222713e-01 -5.07976770e-01 3.55430096e-01 6.71188712e-01
-8.21462035e-01 3.34187776e-01 -6.57162607e-01 4.60619628e-01
-1.57954490e+00 -8.34039450e-01 -1.82567611e-01 2.22717214e+00
1.07852483e+00 1.98040113e-01 5.54096878e-01 2.89052725e-01
3.80275935e-01 1.40054926e-01 -7.75173903e-01 8.62385631e-02
3.02506298e-01 1.87021166e-01 5.47015011e-01 2.42092013e-02
-1.11997318e+00 3.48763883e-01 5.95308161e+00 5.30591011e-01
-6.03241920e-01 8.98892954e-02 3.49591941e-01 -5.59313357e-01
2.31604233e-01 -2.63317883e-01 -5.81593573e-01 4.01362926e-01
6.21761978e-01 3.37500066e-01 2.83207148e-01 9.73131657e-01
2.84104139e-01 -1.12447701e-01 -1.22503150e+00 1.02965057e+00
3.57944280e-01 -7.08149314e-01 -8.17458630e-01 -1.11216597e-01
6.57648504e-01 -3.82759720e-01 -1.83682501e-01 5.44234812e-01
3.17022228e-03 -8.63292217e-01 6.19490921e-01 5.60791492e-01
8.80093575e-01 -4.50351119e-01 4.55339313e-01 3.80752534e-01
-1.05763841e+00 -1.43734410e-01 -8.33361223e-02 -3.13625276e-01
-1.09405860e-01 4.05845702e-01 -2.53377378e-01 6.10947788e-01
8.32368433e-01 7.46856213e-01 -6.66901410e-01 8.60335052e-01
-3.68202478e-01 7.14159787e-01 -3.46686661e-01 3.88800144e-01
-3.03668201e-01 1.24406312e-02 4.41347599e-01 8.04903209e-01
7.06957746e-03 1.34199291e-01 6.55551434e-01 5.80233157e-01
-6.02705730e-03 2.92668846e-02 -4.73134667e-01 8.57222639e-03
-8.36748481e-02 1.15623617e+00 -3.48868400e-01 -4.32454012e-02
-1.66898236e-01 1.16100705e+00 1.20715328e-01 2.30801880e-01
-9.70605791e-01 -1.47481352e-01 7.50925004e-01 6.41517997e-01
8.08197260e-02 -4.69632685e-01 -6.41536266e-02 -1.17467332e+00
4.77279365e-01 -1.27787828e+00 5.64465702e-01 -8.35851133e-01
-1.19490588e+00 -7.35345483e-02 2.88808167e-01 -1.41043901e+00
-2.78443962e-01 -6.55582786e-01 -3.56343389e-01 4.72888827e-01
-1.14008856e+00 -9.59422410e-01 -6.77072346e-01 6.68469429e-01
9.92778063e-01 3.76601934e-01 6.60292327e-01 3.93780380e-01
-8.22089434e-01 6.36980712e-01 -4.44228172e-01 3.44690979e-01
9.38417256e-01 -1.26074767e+00 7.19544291e-03 6.56452060e-01
1.44552603e-01 5.26419759e-01 6.92263126e-01 -6.43536389e-01
-1.76856720e+00 -7.34251738e-01 1.90838113e-01 -7.57723927e-01
4.13302302e-01 -3.45978022e-01 -5.80286324e-01 7.58945227e-01
-3.09294701e-01 5.25276065e-01 7.95173228e-01 2.11473092e-01
-7.61107281e-02 -2.46069908e-01 -8.51859152e-01 5.33760130e-01
1.49876440e+00 -4.11942333e-01 -6.84700191e-01 7.13585794e-01
3.26476961e-01 -1.19425023e+00 -1.13209903e+00 4.69131559e-01
9.59365368e-01 -6.92451715e-01 1.08690405e+00 -8.59600663e-01
5.88418067e-01 -1.48469314e-01 1.18203714e-01 -1.08172047e+00
-1.27243295e-01 -8.68345261e-01 -5.39570630e-01 8.84764969e-01
-1.24207601e-01 9.70678851e-02 1.15055597e+00 7.61395454e-01
-8.29312429e-02 -8.18885565e-01 -6.96821570e-01 -9.58233833e-01
-2.63450921e-01 -4.80687112e-01 6.99216127e-02 8.99536908e-01
-7.20476210e-02 1.30485967e-01 -9.90252197e-01 1.41339555e-01
5.97345173e-01 1.19507916e-01 1.29999995e+00 -1.06596899e+00
-8.34385395e-01 1.44758061e-01 -5.41698754e-01 -1.17743015e+00
-3.55304629e-01 -2.08228394e-01 2.24189267e-01 -1.16812110e+00
5.96962199e-02 3.76916304e-02 -5.97046278e-02 5.30882001e-01
-4.16911155e-01 5.94130874e-01 2.11743891e-01 8.91576111e-02
-6.57537758e-01 3.85325938e-01 1.62936115e+00 -3.37978452e-02
-2.62442261e-01 2.68949002e-01 -5.39018512e-01 1.01067781e+00
7.26190805e-01 -4.70380902e-01 -5.43644071e-01 -4.67708707e-01
8.28981325e-02 5.22722416e-02 5.81396461e-01 -1.26424932e+00
-1.87085122e-01 -1.61963925e-01 7.35091567e-01 -2.50577360e-01
5.48440874e-01 -5.51478207e-01 1.52379841e-01 4.51867431e-01
-4.46169466e-01 -1.29987923e-02 1.19191445e-01 6.89350605e-01
1.82399720e-01 -1.53008565e-01 4.08203870e-01 -5.89693367e-01
-5.61028242e-01 2.09088102e-01 -6.18774034e-02 4.77604598e-01
9.82500732e-01 -7.38177896e-01 -4.15521748e-02 -3.61632973e-01
-1.00122237e+00 3.24707210e-01 3.54221046e-01 5.24732351e-01
4.90997881e-01 -1.18285978e+00 -4.38895524e-01 1.35120109e-01
2.04100087e-01 1.47265270e-01 4.81128156e-01 1.25922942e+00
-5.29944301e-01 -3.25572670e-01 -3.17448735e-01 -7.66081452e-01
-1.22478712e+00 2.03971267e-01 3.71501923e-01 -2.04964891e-01
-9.45744216e-01 7.65743852e-01 -2.18017146e-01 -4.51706976e-01
4.92802203e-01 -4.12588567e-01 1.22385323e-01 -3.06364626e-01
3.52656335e-01 7.83289671e-01 4.10800660e-03 -2.94466913e-01
-2.28053689e-01 7.34200597e-01 2.66420066e-01 2.97005653e-01
1.15828085e+00 -1.55684292e-01 7.52177358e-01 5.28563082e-01
8.84388626e-01 2.25910217e-01 -1.82450294e+00 -7.19821360e-03
-3.64574313e-01 -7.68184483e-01 -1.66640192e-01 -6.83845699e-01
-1.15197313e+00 8.46168756e-01 8.85044813e-01 -3.36278230e-01
1.12333024e+00 -2.12858737e-01 8.15953791e-01 2.05706581e-01
2.97544181e-01 -1.55870426e+00 7.41316974e-01 -1.89802662e-01
7.78953314e-01 -1.27633023e+00 4.87428546e-01 -5.12379348e-01
-8.26510608e-01 8.86679232e-01 1.09482551e+00 -2.77249247e-01
2.82059014e-01 1.24038883e-01 3.29510756e-02 -3.18201751e-01
-4.18485962e-02 -3.03408355e-01 6.83891714e-01 8.14026177e-01
4.57774937e-01 -2.79582618e-03 -2.78535128e-01 7.19363093e-01
-8.51935968e-02 2.02658087e-01 2.20897555e-01 1.27657318e+00
-7.47755095e-02 -8.48369360e-01 -4.76483643e-01 2.87285119e-01
-5.29858589e-01 4.10230935e-01 -3.70772660e-01 1.04930985e+00
3.76368701e-01 6.33382022e-01 -2.65313715e-01 -4.85204518e-01
8.66725266e-01 6.75852550e-03 8.68856490e-01 -7.97414482e-01
-6.57622933e-01 1.04933992e-01 4.21029657e-01 -9.91551638e-01
-4.98323798e-01 -6.17824793e-01 -1.02710164e+00 -3.78299132e-02
-4.46604669e-01 -2.05222607e-01 2.74268687e-01 7.20546484e-01
1.49064228e-01 7.90472746e-01 3.81784678e-01 -1.14804804e+00
-8.75754118e-01 -1.00415969e+00 -6.64280117e-01 8.45189512e-01
1.18740641e-01 -9.82944846e-01 -2.02529535e-01 3.27322453e-01] | [7.216451168060303, -0.6972991228103638] |
b2a4e037-ab0b-4e2f-a7ad-5f6e3d17e44b | attacking-graph-classification-via-bayesian | null | null | https://openreview.net/forum?id=7oziDfK4Fs | https://openreview.net/pdf?id=7oziDfK4Fs | Attacking Graph Classification via Bayesian Optimisation | Graph neural networks have been shown to be vulnerable to adversarial attacks. While the majority of the literature focuses on such vulnerability in node-level classification tasks, little effort has been dedicated to attacks on graph-level classification, an important problem with numerous real-life applications such as biochemistry and social network analysis. The few existing methods often require unrealistic setups, such as access to internal information of the victim models, or an impractically-large number of queries. We present a novel Bayesian optimisation-based attack method for graph classification models. Our method is black-box, query-efficient and parsimonious with respect to the perturbation applied. We empirically validate the effectiveness and flexibility of the proposed method and analyse patterns behind the adversarial samples produced, which may shed further light on the adversarial robustness of graph classification models. | ['Xiaowen Dong', 'Michael Osborne', 'Arno Blaas', 'Binxin Ru', 'Henry Kenlay', 'Xingchen Wan'] | 2021-06-18 | null | null | null | icml-workshop-aml-2021-7 | ['bayesian-optimisation'] | ['methodology'] | [ 5.89627445e-01 1.96393073e-01 -8.84942617e-03 1.30639508e-01
-3.40363353e-01 -9.34464395e-01 5.57118952e-01 5.35470366e-01
-2.13363424e-01 7.33845592e-01 -3.61335188e-01 -7.70997226e-01
-4.14976001e-01 -1.05259883e+00 -7.03366876e-01 -8.41498315e-01
-5.58526874e-01 4.81592476e-01 4.20940071e-01 -2.23425850e-01
2.56762177e-01 9.18415546e-01 -9.59567547e-01 -3.89567167e-01
5.03225923e-01 6.71806157e-01 -6.25636756e-01 8.85268629e-01
5.47977924e-01 3.60945195e-01 -8.39031518e-01 -9.54113781e-01
4.22484010e-01 -3.54030579e-01 -6.22748673e-01 -3.09241503e-01
1.92299560e-01 3.20332758e-02 -6.78272009e-01 1.44785976e+00
8.18384767e-01 1.52992517e-01 6.62560821e-01 -1.44150317e+00
-2.02494740e-01 5.62403858e-01 -2.96298206e-01 3.94466400e-01
1.54568389e-01 3.51646364e-01 8.43962491e-01 -1.51010649e-02
6.26859903e-01 1.32312524e+00 6.99322224e-01 4.23312545e-01
-1.47686267e+00 -8.14274251e-01 -9.71752778e-02 2.61104852e-01
-1.17391670e+00 -3.99439216e-01 1.00199902e+00 -9.56476405e-02
5.22088945e-01 5.23427129e-01 3.99409056e-01 1.38456595e+00
3.72639716e-01 3.16035777e-01 9.87770736e-01 -1.79974481e-01
4.14180815e-01 5.23353443e-02 -7.75956735e-02 5.47793269e-01
5.98693967e-01 4.37481970e-01 -1.81713313e-01 -6.13018036e-01
5.58793426e-01 -1.84956476e-01 -2.58967698e-01 -5.74566483e-01
-7.38310039e-01 1.11513364e+00 4.66360331e-01 1.13646679e-01
-2.94229239e-01 3.17235559e-01 6.07459009e-01 4.27456230e-01
4.83893752e-01 7.26287127e-01 -2.84960479e-01 1.16446480e-01
-6.26431465e-01 2.20605895e-01 1.05187261e+00 7.26993442e-01
3.21219206e-01 3.68496388e-01 -1.01512456e-02 3.67215812e-01
1.61629781e-01 3.53959143e-01 1.52983190e-02 -4.84576076e-01
2.70247817e-01 3.23533118e-01 -3.49191636e-01 -1.58267474e+00
-3.90757322e-01 -7.17951596e-01 -1.17196417e+00 9.57510173e-02
5.60846984e-01 -2.47240067e-01 -6.56841636e-01 1.81868291e+00
5.77107191e-01 3.46524954e-01 9.52603295e-02 5.16511321e-01
5.86216331e-01 3.05187315e-01 4.44485359e-02 -2.04838261e-01
1.04063094e+00 -3.79641443e-01 -2.02478275e-01 -7.35232830e-02
5.79555571e-01 -4.57450420e-01 6.47282541e-01 2.41924867e-01
-9.64718342e-01 8.11558068e-02 -1.12274408e+00 5.49018741e-01
-4.23571199e-01 -7.83746541e-01 7.96233594e-01 1.32817852e+00
-6.93415821e-01 9.10185397e-01 -6.24229550e-01 -4.38298434e-01
5.50216377e-01 6.95976436e-01 -3.98493290e-01 1.59302913e-02
-1.46045983e+00 6.23960078e-01 4.00445044e-01 1.20524682e-01
-1.13400805e+00 -5.19157708e-01 -6.49435461e-01 1.14474334e-01
8.09290230e-01 -4.83380407e-01 6.91877902e-01 -3.61381680e-01
-1.37736952e+00 4.77307141e-01 3.65761727e-01 -7.56562591e-01
7.31317103e-01 4.50745553e-01 -3.52837443e-01 3.02687764e-01
-5.11902988e-01 7.15075359e-02 9.74375904e-01 -9.38962996e-01
1.89733341e-01 -5.27050257e-01 3.21132094e-01 -1.69413537e-01
-4.57206160e-01 7.06469491e-02 -1.11148141e-01 -9.28399682e-01
-4.45574708e-02 -1.14042473e+00 -5.86682796e-01 -8.71764868e-02
-8.33510816e-01 6.34803101e-02 6.96886837e-01 -4.23827678e-01
1.04163897e+00 -1.95289171e+00 1.93258107e-01 7.01134622e-01
2.65240341e-01 4.96288240e-01 -1.93035588e-01 9.52342570e-01
-1.61858246e-01 5.53996623e-01 -3.57867748e-01 1.42921716e-01
4.73852232e-02 4.61420901e-02 -5.55426955e-01 8.44486117e-01
4.61119153e-02 8.43269050e-01 -8.96975219e-01 -2.51625985e-01
6.13881312e-02 3.56197447e-01 -5.27100086e-01 1.41527504e-01
-2.18962446e-01 4.25313830e-01 -5.16289651e-01 7.44228661e-01
4.41488713e-01 -1.39133781e-01 2.75057286e-01 -3.12975086e-02
8.91045630e-01 3.52149196e-02 -1.07898688e+00 9.85083342e-01
-4.04771864e-02 3.72399479e-01 7.53418431e-02 -1.20701098e+00
7.48946667e-01 2.91243732e-01 2.48491719e-01 -3.47313315e-01
2.87570894e-01 -3.98412533e-02 4.42582935e-01 -2.57182904e-02
1.00044593e-01 -4.17794101e-02 -2.74905741e-01 5.46192467e-01
-1.80948868e-01 -2.11595967e-01 1.13013834e-01 4.29877222e-01
1.49798298e+00 -4.49297160e-01 4.38686430e-01 -5.88810928e-02
4.79495585e-01 -2.46012077e-01 3.26821059e-01 9.88201618e-01
-2.43661076e-01 1.86470032e-01 8.79125774e-01 -3.40928823e-01
-9.36303735e-01 -7.61175811e-01 -1.33959670e-02 6.53223336e-01
1.60776258e-01 -5.52991092e-01 -8.69836986e-01 -9.29222226e-01
1.02799334e-01 5.12457907e-01 -6.37865007e-01 -7.24846244e-01
-3.45006227e-01 -9.44281161e-01 1.20347691e+00 1.00416549e-01
2.77497202e-01 -9.88169134e-01 -2.10823104e-01 1.36375099e-01
3.93025041e-01 -1.08297169e+00 -1.40215248e-01 1.51106849e-01
-9.85934138e-01 -1.28961647e+00 -1.35290280e-01 -1.53758585e-01
6.48729146e-01 -9.94807556e-02 9.67354655e-01 3.78849268e-01
-4.33931470e-01 2.82491088e-01 -1.56979129e-01 -4.59125757e-01
-9.07328963e-01 2.40593091e-01 1.50246263e-01 8.14239159e-02
-4.07717936e-02 -9.71193612e-01 -3.41421276e-01 4.03229028e-01
-1.13515449e+00 -6.78860247e-01 6.11917317e-01 8.14041853e-01
2.74047762e-01 4.27829295e-01 5.71768582e-01 -1.16019297e+00
1.11158013e+00 -6.12607479e-01 -8.50595474e-01 1.63151443e-01
-7.14800000e-01 9.81080607e-02 9.54677701e-01 -7.39547312e-01
-1.19699158e-01 -9.07565281e-02 -1.09180428e-01 -4.58748221e-01
-1.35912254e-01 6.86628759e-01 -3.92073423e-01 -7.10792601e-01
8.81454110e-01 2.41599455e-01 9.00653154e-02 -7.99978226e-02
3.08605343e-01 1.37979791e-01 4.04959172e-01 -5.30953586e-01
1.37619555e+00 8.26146826e-02 8.67374837e-01 -1.12968862e+00
-1.68800399e-01 4.30278182e-02 -3.85841370e-01 -1.93448678e-01
3.22008729e-01 -1.50047004e-01 -1.15239799e+00 5.59702516e-01
-8.49951684e-01 -1.04730070e-01 1.44751370e-01 -4.67736982e-02
-3.86154592e-01 8.92782450e-01 -5.06376505e-01 -1.02536821e+00
-3.93273860e-01 -1.16878927e+00 4.98840928e-01 -1.50972277e-01
-1.49543270e-01 -1.28962541e+00 1.89488083e-02 2.40821332e-01
4.57035571e-01 8.21284175e-01 1.29262686e+00 -1.18067443e+00
-7.39494562e-01 -9.03520346e-01 6.88276393e-03 8.61170143e-02
-1.51318356e-01 1.65531470e-03 -9.31877792e-01 -7.13886499e-01
-7.63137117e-02 -4.61886197e-01 5.97617745e-01 4.66635861e-02
1.17368793e+00 -6.06664836e-01 -3.78157705e-01 6.75589383e-01
1.15451229e+00 1.15214624e-02 5.70136726e-01 6.84017763e-02
7.08391190e-01 6.43598139e-01 1.49079800e-01 1.87182605e-01
-3.26092362e-01 6.73367918e-01 1.01011825e+00 9.39923059e-03
2.97363043e-01 -2.48560250e-01 1.84143841e-01 4.19321775e-01
1.56249449e-01 -6.57697201e-01 -8.21428120e-01 1.90657482e-01
-1.55543292e+00 -9.77813900e-01 6.90550506e-02 2.35016394e+00
7.01952100e-01 5.11532962e-01 2.43784428e-01 5.60147464e-01
7.69768238e-01 4.14467871e-01 -7.97150552e-01 -3.37461263e-01
-6.16041496e-02 2.02795699e-01 7.85529435e-01 4.39739138e-01
-1.02043891e+00 9.38799560e-01 6.86833334e+00 9.88576412e-01
-9.99591529e-01 -2.00032219e-01 5.09614706e-01 1.41812131e-01
-1.01402670e-01 1.53307542e-01 -4.05605376e-01 3.36345017e-01
1.12602437e+00 -2.94786453e-01 7.43673444e-01 6.75944865e-01
-1.05217703e-01 1.84902951e-01 -1.09667826e+00 5.89688957e-01
-4.25727628e-02 -1.22217536e+00 6.27012253e-02 5.20108759e-01
3.12997699e-01 -2.07330152e-01 8.10473710e-02 3.64684947e-02
5.89166284e-01 -1.21755123e+00 1.29372790e-01 1.13523506e-01
5.13657749e-01 -1.09706581e+00 6.27418518e-01 4.66799259e-01
-8.45247209e-01 -3.25622074e-02 -3.47183079e-01 4.62201536e-02
-1.11004174e-01 5.69037914e-01 -1.14013982e+00 6.69722497e-01
3.82600754e-01 3.18398356e-01 -8.24485481e-01 9.14146423e-01
-2.67965555e-01 1.07076049e+00 -4.29731399e-01 -3.76614630e-01
8.16904530e-02 -8.21063295e-02 1.06262577e+00 8.46871257e-01
-1.32029966e-01 -2.56451160e-01 2.02478528e-01 5.81638396e-01
-2.15824932e-01 8.55120085e-03 -1.05411911e+00 -5.76269925e-01
6.36260271e-01 1.13350356e+00 -9.46146607e-01 8.32458138e-02
1.39247417e-01 7.19868779e-01 2.99401164e-01 4.00941402e-01
-6.45137727e-01 -5.87032437e-01 4.83478397e-01 -1.51715074e-02
1.83981404e-01 -2.34404743e-01 -5.93051985e-02 -8.61692131e-01
-1.62156746e-02 -1.28135622e+00 5.93629956e-01 -1.39498934e-01
-1.51939464e+00 5.64429224e-01 5.89889400e-02 -9.61265683e-01
-3.95013362e-01 -6.09493554e-01 -7.26600707e-01 7.27266729e-01
-9.79253173e-01 -1.02822530e+00 1.27913147e-01 6.19903088e-01
-1.53714299e-01 -3.61205131e-01 1.05132258e+00 6.27683476e-02
-8.72361422e-01 1.05689096e+00 1.02084860e-01 2.26736799e-01
4.05375928e-01 -9.70591605e-01 7.31916249e-01 1.19523358e+00
3.07111681e-01 6.96649373e-01 1.00938368e+00 -7.98222721e-01
-1.58936656e+00 -1.06226647e+00 3.36727411e-01 -5.35254419e-01
9.48645651e-01 -6.89330459e-01 -9.29214120e-01 3.37900251e-01
-1.99519277e-01 1.64607793e-01 5.33547580e-01 -9.09409206e-03
-4.28880334e-01 -1.41814515e-01 -1.38675916e+00 8.87684643e-01
9.53442514e-01 -6.53876364e-01 -4.94289510e-02 6.64916277e-01
6.04080260e-01 -2.52908498e-01 -9.89493132e-01 4.48732764e-01
3.59712631e-01 -7.84452140e-01 1.19458771e+00 -1.15013623e+00
3.80809121e-02 -1.28066510e-01 4.46818434e-02 -1.24361444e+00
-1.66553855e-01 -1.04681253e+00 -3.67170095e-01 9.51997161e-01
3.66972327e-01 -9.64661658e-01 9.70546424e-01 3.77753943e-01
3.21402252e-01 -1.00307548e+00 -1.16523755e+00 -9.59313154e-01
7.42372796e-02 -3.50311905e-01 5.53759456e-01 9.96690333e-01
-3.21060628e-01 1.38880879e-01 -5.37876368e-01 6.11363292e-01
7.77687252e-01 -3.58325124e-01 1.04994130e+00 -1.41156721e+00
-5.43841064e-01 -5.40656745e-01 -9.53874528e-01 -1.06549621e-01
2.90158361e-01 -9.66201782e-01 -3.00319165e-01 -1.01408708e+00
-1.55140817e-01 -3.24703664e-01 -2.03996733e-01 3.04386288e-01
-2.51988322e-01 4.98164564e-01 5.96902668e-02 2.10742816e-01
-2.65954614e-01 3.23811918e-01 7.72394180e-01 -2.99291849e-01
6.00797944e-02 3.69569898e-01 -7.15879798e-01 4.91901994e-01
8.64708960e-01 -8.55206251e-01 -6.75325692e-01 1.48487031e-01
4.94949520e-01 -1.70591343e-02 5.27729928e-01 -8.94052505e-01
2.56531209e-01 -1.09526619e-01 1.30835265e-01 -1.50415823e-01
2.65820831e-01 -8.43603015e-01 3.42479914e-01 8.70484531e-01
-3.08864683e-01 1.42139420e-01 9.02911350e-02 1.05653107e+00
1.59145221e-01 -3.10433030e-01 7.91785538e-01 -8.95774141e-02
1.24301147e-02 6.85534358e-01 -3.01060826e-01 1.25485778e-01
1.18404305e+00 -2.07823858e-01 -2.66046166e-01 -5.82481444e-01
-6.28019691e-01 6.81050569e-02 6.60125911e-01 1.70766518e-01
3.88378829e-01 -9.57649946e-01 -5.87905347e-01 8.94887075e-02
3.60363759e-02 -2.44360358e-01 2.39893012e-02 5.58302104e-01
-4.13373470e-01 2.15524748e-01 -1.13313563e-01 -4.58283037e-01
-1.37603259e+00 1.07837903e+00 4.10976350e-01 -5.26308298e-01
-3.94024342e-01 5.80980361e-01 -2.47091129e-01 -3.70845854e-01
4.40439612e-01 3.19078743e-01 1.73239466e-02 -1.41271293e-01
6.51929006e-02 5.10441303e-01 1.08220160e-01 -3.56047571e-01
-3.85505140e-01 2.65538782e-01 -2.58777291e-02 7.27553517e-02
1.09280360e+00 3.77200663e-01 -2.38646150e-01 4.74802777e-02
8.95906031e-01 2.03507096e-02 -7.93667078e-01 -2.23365873e-01
-2.30802912e-02 -4.28309739e-01 -7.72626251e-02 -5.25572181e-01
-1.05818033e+00 9.43471909e-01 3.36816579e-01 7.79926717e-01
9.62667346e-01 -2.90486455e-01 5.15959144e-01 8.29336941e-01
3.95062834e-01 -5.14121234e-01 9.48616937e-02 2.24220559e-01
6.99935257e-01 -9.53901172e-01 2.23309293e-01 -5.50384462e-01
-7.21799508e-02 7.74968445e-01 2.87647814e-01 -2.35149607e-01
6.19110405e-01 -9.47093219e-03 -3.00795585e-01 -1.33937061e-01
-6.84239030e-01 3.64823878e-01 1.87734902e-01 9.17022705e-01
-8.95510465e-02 -1.66328847e-01 -8.34785402e-02 2.75605351e-01
-2.63744801e-01 -6.62777901e-01 5.76575100e-01 9.28193629e-01
3.79553922e-02 -1.20881402e+00 -3.27560276e-01 4.51858789e-01
-7.49385536e-01 -2.14043930e-01 -9.96947408e-01 7.56798685e-01
-5.32066166e-01 9.28875268e-01 -6.54993534e-01 -4.89665031e-01
1.69367775e-01 -2.36935262e-02 3.26957703e-01 -4.12953615e-01
-7.69718945e-01 -3.91496688e-01 2.63143986e-01 -4.70655650e-01
-5.33814840e-02 -5.30324101e-01 -5.79567790e-01 -6.25000000e-01
-4.93117005e-01 1.86568290e-01 5.45425534e-01 8.33475232e-01
5.71549654e-01 4.05386180e-01 7.09433377e-01 -7.59411752e-01
-9.04260218e-01 -8.79011214e-01 -4.81834233e-01 4.22626257e-01
1.98502213e-01 -5.54041386e-01 -6.47819638e-01 -4.91725594e-01] | [5.961103916168213, 7.415004253387451] |
7c2da72b-9747-4b46-9222-e7f99814c734 | using-generic-summarization-to-improve-music | 1503.06666 | null | http://arxiv.org/abs/1503.06666v3 | http://arxiv.org/pdf/1503.06666v3.pdf | Using Generic Summarization to Improve Music Information Retrieval Tasks | In order to satisfy processing time constraints, many MIR tasks process only
a segment of the whole music signal. This practice may lead to decreasing
performance, since the most important information for the tasks may not be in
those processed segments. In this paper, we leverage generic summarization
algorithms, previously applied to text and speech summarization, to summarize
items in music datasets. These algorithms build summaries, that are both
concise and diverse, by selecting appropriate segments from the input signal
which makes them good candidates to summarize music as well. We evaluate the
summarization process on binary and multiclass music genre classification
tasks, by comparing the performance obtained using summarized datasets against
the performances obtained using continuous segments (which is the traditional
method used for addressing the previously mentioned time constraints) and full
songs of the same original dataset. We show that GRASSHOPPER, LexRank, LSA,
MMR, and a Support Sets-based Centrality model improve classification
performance when compared to selected 30-second baselines. We also show that
summarized datasets lead to a classification performance whose difference is
not statistically significant from using full songs. Furthermore, we make an
argument stating the advantages of sharing summarized datasets for future MIR
research. | ['Ricardo Ribeiro', 'David Martins de Matos', 'Francisco Raposo'] | 2015-03-23 | null | null | null | null | ['genre-classification'] | ['computer-vision'] | [ 4.22899306e-01 -2.57160906e-02 -3.53784740e-01 -6.72515035e-02
-1.04752171e+00 -7.99944937e-01 6.45620763e-01 6.84343636e-01
-5.23593903e-01 7.71177888e-01 9.44924235e-01 9.85310748e-02
-6.06313288e-01 -3.72184724e-01 -2.20341325e-01 -5.29916286e-01
-1.79687887e-01 3.22471559e-01 6.97163045e-02 -1.05251074e-01
7.05680192e-01 1.86166614e-01 -1.57705450e+00 6.26041293e-01
8.34399104e-01 8.53006661e-01 6.70172572e-02 8.20014179e-01
-1.59147069e-01 6.58607364e-01 -1.13968980e+00 -2.62284815e-01
1.42712682e-01 -7.56978154e-01 -7.96719134e-01 -2.80490681e-03
6.45281792e-01 -1.63977221e-03 -1.31936267e-01 6.78457856e-01
6.80085778e-01 5.04927278e-01 8.24410260e-01 -8.96303952e-01
-7.70383282e-04 1.50059795e+00 -6.32333338e-01 3.36175650e-01
5.87484658e-01 -4.16965574e-01 1.49692273e+00 -5.25987685e-01
7.49247074e-01 9.29087877e-01 7.13617146e-01 2.26034760e-01
-1.28633642e+00 -4.87115353e-01 2.21993566e-01 7.93537199e-02
-1.13426328e+00 -7.29695976e-01 6.90772176e-01 -2.73830265e-01
9.05178964e-01 8.63422811e-01 7.11738646e-01 9.35304344e-01
-6.94678724e-02 8.44503224e-01 7.02004850e-01 -4.98428077e-01
2.63518870e-01 -1.63641527e-01 7.04720736e-01 -3.05207297e-02
4.69943762e-01 -5.10671437e-01 -1.10574555e+00 -3.40510219e-01
-1.39348298e-01 -4.77961063e-01 -3.92286837e-01 1.45099908e-01
-1.52335572e+00 6.20845079e-01 4.73164879e-02 5.74572325e-01
-5.37187636e-01 7.66201466e-02 8.06123495e-01 5.28885067e-01
7.47962236e-01 1.02410758e+00 -2.68091500e-01 -3.36867332e-01
-1.80919015e+00 3.60257268e-01 9.97331321e-01 6.38087332e-01
1.94010779e-01 1.62197486e-01 -6.95317626e-01 9.34950769e-01
-1.93045884e-01 1.23414844e-01 7.08861232e-01 -1.03727102e+00
5.27842939e-01 2.46854395e-01 -2.08383456e-01 -1.01926184e+00
-5.20761907e-01 -1.00440168e+00 -8.22650671e-01 -3.23396444e-01
2.47119129e-01 -5.60768880e-02 -3.97334129e-01 1.68539762e+00
-1.25608832e-01 -5.91276586e-02 6.62655160e-02 6.41249895e-01
8.52656126e-01 7.87501812e-01 -3.59018415e-01 -7.56128550e-01
1.11201644e+00 -9.50741410e-01 -6.88858390e-01 2.81189512e-02
3.69690925e-01 -9.36519504e-01 9.56677258e-01 1.00864995e+00
-1.03687918e+00 -6.00818455e-01 -1.29002094e+00 1.14963464e-01
7.29189366e-02 4.16171312e-01 4.40085590e-01 5.10426939e-01
-8.08495700e-01 1.03236890e+00 -5.33383608e-01 -4.82603252e-01
2.31396660e-01 2.20010966e-01 -6.93364143e-02 4.51073170e-01
-7.62398243e-01 5.23177624e-01 5.78887522e-01 -4.23638344e-01
-5.03134668e-01 -7.93844700e-01 -2.84200847e-01 3.13436151e-01
3.18395376e-01 -5.89409590e-01 1.31452250e+00 -7.06534803e-01
-1.22650707e+00 4.35608298e-01 -1.48051590e-01 -8.61539185e-01
5.21145165e-01 -4.35757160e-01 -7.65718073e-02 3.01865518e-01
7.50670806e-02 4.33598042e-01 7.39183247e-01 -9.45562124e-01
-8.39968264e-01 -1.18251227e-01 -3.92646611e-01 2.18868896e-01
-4.99688059e-01 -1.07764639e-02 -2.61319935e-01 -1.06250644e+00
1.58510700e-01 -7.84192324e-01 -3.24541703e-02 -9.13975179e-01
-8.50127101e-01 -2.27513626e-01 4.71922159e-01 -7.58287370e-01
1.95334280e+00 -2.18176842e+00 3.32517415e-01 3.15746099e-01
-4.71153706e-02 1.53999645e-02 -3.25561851e-01 8.72159839e-01
-9.45173949e-03 2.86674768e-01 -8.08120817e-02 -4.11642671e-01
2.19260901e-02 -2.34662801e-01 -7.33597517e-01 3.08819890e-01
-3.13780934e-01 6.23704493e-01 -7.67454147e-01 -5.13078153e-01
-1.09230056e-01 -5.36337383e-02 -5.07593155e-01 -2.32434034e-01
-3.05378616e-01 3.96028072e-01 -1.86537668e-01 2.58752316e-01
1.56828776e-01 3.31598967e-02 9.01267901e-02 -1.99984521e-01
-1.53117061e-01 7.77742267e-01 -1.15151250e+00 1.84416544e+00
-1.94512889e-01 9.88890350e-01 -2.82910645e-01 -9.21352386e-01
8.66145015e-01 3.12868565e-01 7.12364674e-01 -4.51184779e-01
-5.39737456e-02 2.41871685e-01 1.11152403e-01 -1.36582386e-02
9.86170411e-01 5.78186437e-02 -2.61687517e-01 8.24640095e-01
-7.87584484e-02 -4.33334708e-01 8.63597929e-01 6.29402220e-01
1.11312473e+00 -2.09859490e-01 4.63648736e-01 -2.53329933e-01
3.40797812e-01 1.70082346e-01 2.90578097e-01 1.08710754e+00
3.48493427e-01 7.54774809e-01 6.96888685e-01 1.86196104e-01
-1.04695284e+00 -7.56814361e-01 1.04833050e-02 1.26986921e+00
-2.31255949e-01 -1.17384267e+00 -6.63378835e-01 -4.64802116e-01
-1.40256926e-01 1.01182544e+00 -3.59430581e-01 -2.05463082e-01
-5.23576438e-01 -6.10617936e-01 8.21299434e-01 8.93359184e-02
-1.16335675e-02 -8.77669573e-01 -7.50799656e-01 3.34202141e-01
-5.75818181e-01 -8.47051322e-01 -5.42319953e-01 3.00197750e-01
-1.01457977e+00 -7.58291960e-01 -7.85729647e-01 -3.42583179e-01
1.09561384e-01 4.29934651e-01 1.04791451e+00 -1.16727531e-01
-2.98335869e-03 3.98861229e-01 -8.19694102e-01 -5.42077124e-01
-4.83110636e-01 7.33694375e-01 1.92876890e-01 -1.07708603e-01
-1.43455252e-01 -7.18406796e-01 -4.09479409e-01 -7.96452612e-02
-8.05238545e-01 1.37305513e-01 4.85517532e-01 6.35045886e-01
4.34273064e-01 9.02798772e-02 1.14258313e+00 -6.98173344e-01
1.07791841e+00 -2.53978461e-01 7.05071613e-02 2.46786755e-02
-6.54731393e-01 -4.18490032e-03 6.54045165e-01 -5.31003714e-01
-6.29512668e-01 -1.16680555e-01 -1.21791689e-02 -8.70143473e-02
2.63021231e-01 8.51198077e-01 3.44178647e-01 5.49348950e-01
9.70300972e-01 3.55730683e-01 -1.30143940e-01 -7.07497001e-01
4.69586164e-01 1.02172685e+00 6.48206413e-01 -4.97800231e-01
4.67379868e-01 2.92153835e-01 -4.39016558e-02 -1.05487037e+00
-1.05925202e+00 -8.57679963e-01 -5.46803892e-01 -1.83026135e-01
2.74029911e-01 -7.77536154e-01 -2.94532806e-01 -1.31859630e-01
-1.01850116e+00 1.11878350e-01 -8.01644921e-01 6.53356314e-01
-6.88331783e-01 7.59807408e-01 -5.15709519e-01 -8.94983947e-01
-7.77762234e-01 -5.37667990e-01 1.07511330e+00 -6.91880211e-02
-1.02275240e+00 -3.95586789e-01 1.07527196e-01 2.31431589e-01
1.20435782e-01 2.11406171e-01 9.43740726e-01 -1.15451813e+00
-1.02522135e-01 -1.68701068e-01 1.87730581e-01 3.42916846e-01
1.97011262e-01 1.13514401e-01 -8.37109804e-01 -3.54026705e-01
-3.23259056e-01 -2.49097392e-01 1.48474586e+00 5.51227510e-01
9.75490451e-01 -4.15731251e-01 -2.47405812e-01 1.80161163e-01
8.03746164e-01 -2.14618836e-02 2.77736634e-01 3.14011484e-01
2.62618393e-01 6.58211410e-01 8.13039839e-01 8.15193474e-01
-1.13504611e-01 6.61814809e-01 -4.32541333e-02 2.61408359e-01
-3.43977898e-01 -2.96374321e-01 5.37192225e-01 1.37464643e+00
-2.72197695e-03 -3.57512444e-01 -5.28887570e-01 6.65825844e-01
-1.98451138e+00 -1.32479668e+00 -3.02449286e-01 2.34548759e+00
1.02521288e+00 1.54370070e-01 5.83090782e-01 9.23394620e-01
4.01608497e-01 3.14164221e-01 -2.87594587e-01 -4.03108090e-01
-4.06736881e-01 3.25187325e-01 1.98556915e-01 1.69447005e-01
-1.05359244e+00 6.48092985e-01 6.49162960e+00 1.20165062e+00
-1.03891885e+00 -1.01853579e-01 2.64373451e-01 -7.30875254e-01
-2.26949424e-01 -7.33490661e-02 -4.34397548e-01 2.74423838e-01
1.12608039e+00 -6.52528942e-01 3.15421700e-01 4.68790352e-01
5.26474297e-01 -3.00415486e-01 -1.29986525e+00 9.58096385e-01
1.80419281e-01 -1.27926433e+00 2.79349416e-01 -6.80069774e-02
8.85951340e-01 -8.08235109e-02 -9.94933993e-02 2.00966239e-01
-1.28838509e-01 -7.62706161e-01 1.13341129e+00 4.77951825e-01
5.44184566e-01 -8.72118533e-01 5.45550108e-01 5.29540479e-01
-1.00571620e+00 -1.17725976e-01 -3.25738758e-01 -1.07355356e-01
4.40100245e-02 8.39567304e-01 -9.30907071e-01 1.05305135e+00
4.07190293e-01 9.38774645e-01 -7.40387201e-01 1.34055007e+00
1.04904085e-01 1.10676146e+00 -3.22925448e-01 -1.40213013e-01
6.96250424e-02 -1.18610159e-01 1.08676565e+00 1.57667959e+00
5.13216317e-01 -2.57209837e-01 3.77927497e-02 3.22291344e-01
3.69825996e-02 6.87120140e-01 -3.21262240e-01 -3.67908537e-01
4.47399050e-01 1.08016956e+00 -1.00501013e+00 -4.59153622e-01
1.12761781e-02 7.82188356e-01 -1.33694902e-01 2.13449240e-01
-4.96385723e-01 -6.03760123e-01 2.20269620e-01 8.35881755e-02
1.25279948e-01 -1.71042189e-01 -5.79110146e-01 -1.10698712e+00
-1.81313276e-01 -1.11974967e+00 6.99626625e-01 -6.10080123e-01
-1.11636949e+00 5.32308042e-01 4.27985489e-02 -1.49469101e+00
-4.96884584e-01 1.65429890e-01 -5.41528404e-01 5.02089143e-01
-9.92800057e-01 -7.19178498e-01 -3.63682397e-02 1.23418488e-01
7.90582895e-01 -3.48382205e-01 6.05567217e-01 2.84369916e-01
-2.63489664e-01 5.38956344e-01 2.87389994e-01 -2.14910343e-01
9.27310228e-01 -1.14574969e+00 2.52145559e-01 7.54483640e-01
9.04140949e-01 5.92962265e-01 9.51278389e-01 -5.79180717e-01
-1.07016790e+00 -8.94708037e-01 9.40579772e-01 -2.50442952e-01
5.45936167e-01 -4.38733511e-02 -6.22145176e-01 3.27705920e-01
4.02125835e-01 -1.07915771e+00 8.47087979e-01 5.06988943e-01
-2.42674857e-01 -1.97733790e-01 -4.82604593e-01 6.20531797e-01
1.07511818e+00 -2.34915525e-01 -1.03037691e+00 1.61983982e-01
7.63227165e-01 -7.67282862e-03 -6.16601408e-01 2.77078867e-01
8.10549796e-01 -8.49304199e-01 8.64472747e-01 -5.10351658e-01
4.38805550e-01 -3.23168188e-01 -2.26539344e-01 -1.60941577e+00
-2.11695462e-01 -8.19336832e-01 -6.03618361e-02 1.44360757e+00
5.49285710e-01 -2.33828530e-01 5.90888143e-01 -3.67853522e-01
-3.35115641e-01 -3.60365987e-01 -8.89693916e-01 -9.00164366e-01
-1.79455087e-01 -6.17441714e-01 3.91284883e-01 7.27976739e-01
4.20278043e-01 8.02416980e-01 -5.32127440e-01 -5.04039168e-01
5.14558911e-01 4.74705130e-01 8.34101379e-01 -1.53830910e+00
-3.45332533e-01 -9.20417845e-01 -1.89679042e-01 -9.19082522e-01
1.08292140e-01 -1.17265666e+00 -4.24900465e-02 -1.82052183e+00
3.58990580e-01 -1.60518050e-01 -4.26568508e-01 3.96201164e-01
8.14716071e-02 3.73002112e-01 5.83146572e-01 5.55902719e-01
-7.58857846e-01 3.12780648e-01 9.87094820e-01 -3.67602557e-01
-6.61056101e-01 2.80497223e-01 -9.87821460e-01 4.37468916e-01
8.22180331e-01 -5.42043388e-01 -5.21857083e-01 -7.66299441e-02
4.61440504e-01 5.44729345e-02 -1.03688397e-01 -1.31988573e+00
3.93045604e-01 1.18077599e-01 2.19032899e-01 -1.05420792e+00
3.31917733e-01 -3.48097831e-01 2.69423723e-01 4.16488826e-01
-7.95076489e-01 -2.49115184e-01 1.92161754e-01 5.49290061e-01
-3.86500359e-01 -3.67754370e-01 4.42896307e-01 1.96269035e-01
-1.39364704e-01 -1.68760836e-01 -3.81293446e-01 3.46995145e-02
5.41406333e-01 -1.96207985e-01 -1.84914470e-01 -7.87321746e-01
-7.28724241e-01 1.65572274e-03 8.87913927e-02 3.77183616e-01
2.91977197e-01 -9.93468702e-01 -1.18229115e+00 -3.45014483e-01
1.50855646e-01 -4.90986466e-01 2.48906277e-02 1.03811920e+00
-2.46049538e-01 6.30241334e-01 9.57628433e-03 -4.92266804e-01
-1.66859472e+00 2.67698526e-01 -2.68835157e-01 -2.81816006e-01
-7.83755422e-01 5.70998311e-01 -1.67757452e-01 1.31879836e-01
5.06578326e-01 -5.86950958e-01 -4.01128739e-01 1.03020275e+00
5.61008215e-01 6.28004670e-01 2.74025828e-01 -4.46013629e-01
-2.76551217e-01 5.02504885e-01 1.05231933e-01 -5.98271847e-01
1.36019397e+00 -2.13525906e-01 -7.44325072e-02 8.20553958e-01
6.91446424e-01 7.74047852e-01 -5.39677858e-01 -1.63891524e-01
4.03893858e-01 -2.22600996e-01 -4.91879089e-03 -9.04317796e-01
-5.96544206e-01 6.93692029e-01 -3.08474582e-02 6.18144691e-01
1.24455178e+00 -2.89581437e-02 5.44270456e-01 4.28413540e-01
2.35013440e-01 -1.12612092e+00 6.90205768e-02 5.32846928e-01
1.08851552e+00 -5.22861540e-01 5.28211653e-01 -2.16152042e-01
-7.41515577e-01 1.11443496e+00 -1.09088235e-01 2.80027911e-02
1.73491955e-01 -6.60809595e-03 -2.17175618e-01 9.83753651e-02
-1.09540355e+00 -3.16068381e-01 6.07388914e-01 2.66769320e-01
5.61993837e-01 4.02968042e-02 -8.95248234e-01 1.01779735e+00
-7.79835641e-01 -3.55621606e-01 7.21054792e-01 5.56571662e-01
-6.96673810e-01 -1.04921412e+00 -4.21298295e-01 9.48238492e-01
-7.17968762e-01 -1.06616504e-01 -9.04802859e-01 5.52211583e-01
-2.68621862e-01 1.34691322e+00 -2.27389988e-02 -4.78220016e-01
3.34220588e-01 1.12171464e-01 3.66993070e-01 -8.34261894e-01
-9.19557571e-01 5.16331255e-01 4.51086879e-01 -8.40189308e-02
-3.78184766e-01 -8.64068329e-01 -1.19322181e+00 -2.12316886e-01
-3.79148185e-01 6.56549275e-01 6.66373670e-01 5.91651976e-01
3.52349967e-01 9.11056519e-01 6.38606608e-01 -8.53062987e-01
-7.58957863e-01 -1.33294022e+00 -5.82870662e-01 2.30906889e-01
2.89811045e-01 -1.66049302e-01 -3.86952430e-01 4.47011366e-02] | [12.530291557312012, 9.533792495727539] |
19b06dd2-9a55-4efb-b6ff-35f34a210a5d | noisy-tensor-ring-approximation-for-computing | 2307.03884 | null | https://arxiv.org/abs/2307.03884v1 | https://arxiv.org/pdf/2307.03884v1.pdf | Noisy Tensor Ring approximation for computing gradients of Variational Quantum Eigensolver for Combinatorial Optimization | Variational Quantum algorithms, especially Quantum Approximate Optimization and Variational Quantum Eigensolver (VQE) have established their potential to provide computational advantage in the realm of combinatorial optimization. However, these algorithms suffer from classically intractable gradients limiting the scalability. This work addresses the scalability challenge for VQE by proposing a classical gradient computation method which utilizes the parameter shift rule but computes the expected values from the circuits using a tensor ring approximation. The parametrized gates from the circuit transform the tensor ring by contracting the matrix along the free edges of the tensor ring. While the single qubit gates do not alter the ring structure, the state transformations from the two qubit rotations are evaluated by truncating the singular values thereby preserving the structure of the tensor ring and reducing the computational complexity. This variation of the Matrix product state approximation grows linearly in number of qubits and the number of two qubit gates as opposed to the exponential growth in the classical simulations, allowing for a faster evaluation of the gradients on classical simulators. | ['Vaneet Aggarwal', 'Utkarsh Priyam', 'Dheeraj Peddireddy'] | 2023-07-08 | null | null | null | null | ['combinatorial-optimization'] | ['methodology'] | [-1.31488489e-02 7.99212530e-02 3.14948052e-01 -2.79046060e-03
-4.95625883e-01 -9.63756979e-01 5.11284232e-01 4.90546338e-02
-7.44467318e-01 8.08121204e-01 -1.29481986e-01 -5.61808825e-01
-2.05935836e-01 -1.05251634e+00 -6.50176406e-01 -1.16890097e+00
-3.33484888e-01 3.56001347e-01 -4.20630872e-02 -7.23861158e-01
6.24845207e-01 5.30747294e-01 -1.07138538e+00 -1.82997987e-01
6.90919876e-01 6.74280167e-01 -5.28648198e-01 8.53190541e-01
9.85187888e-02 4.21065122e-01 -2.99336195e-01 -4.98385638e-01
6.06156886e-01 -5.24296641e-01 -8.05846334e-01 -5.33223152e-01
2.27281362e-01 -3.97139281e-01 -8.17034185e-01 1.33139706e+00
5.16525626e-01 3.79266918e-01 2.19896242e-01 -1.00240493e+00
5.42870536e-02 7.80776501e-01 3.79417874e-02 -1.44680575e-01
2.15128928e-01 2.81864196e-01 1.44438946e+00 -5.11837840e-01
9.29794133e-01 9.57896054e-01 4.32995319e-01 2.17662141e-01
-1.79693818e+00 -2.78769046e-01 -7.51463830e-01 2.75381953e-01
-1.95487714e+00 -2.73037910e-01 5.80027223e-01 -2.04585746e-01
1.30273867e+00 4.27620769e-01 7.57011056e-01 3.07063699e-01
4.18361843e-01 -1.48245305e-01 1.25031877e+00 -3.47965360e-01
5.95943630e-01 8.90968218e-02 1.63868040e-01 9.27356303e-01
2.77938873e-01 6.55238628e-01 -4.85102057e-01 -5.06252587e-01
4.84710306e-01 -2.60109127e-01 -1.03728533e-01 -6.89495802e-01
-1.43979251e+00 9.83294249e-01 5.12903154e-01 5.28725609e-02
-3.32941234e-01 7.37524867e-01 5.36420822e-01 4.74328309e-01
-3.48010868e-01 7.31017053e-01 -8.41934010e-02 -2.82069266e-01
-9.73999560e-01 5.02579808e-01 1.14816415e+00 5.71284354e-01
1.33720970e+00 -1.18443891e-01 -3.48175436e-01 -2.56935775e-01
7.71010518e-02 8.14090610e-01 -3.68893981e-01 -1.14301872e+00
2.21079677e-01 2.79868513e-01 2.70635396e-01 -6.70840561e-01
-3.47814232e-01 -2.37142473e-01 -7.55904675e-01 3.48631106e-02
5.26113033e-01 -2.58437186e-01 -5.59811473e-01 1.55375600e+00
5.06867051e-01 -2.63298392e-01 -9.73472521e-02 8.47801030e-01
1.93255302e-02 7.44913220e-01 -4.69270915e-01 -2.72127181e-01
1.11628807e+00 -5.79743147e-01 -6.13576889e-01 4.09920543e-01
1.15175378e+00 -7.14763224e-01 4.68368143e-01 2.40505099e-01
-1.01974809e+00 -5.37664704e-02 -1.34444463e+00 -2.29503766e-01
-3.16863447e-01 -4.22929406e-01 8.61939073e-01 8.94896090e-01
-1.19248736e+00 1.16031241e+00 -1.02114403e+00 2.51512825e-01
-2.34282807e-01 8.59260023e-01 -3.84518176e-01 -8.00954551e-02
-1.23626912e+00 9.02982473e-01 4.85346019e-01 4.54446882e-01
-6.59732699e-01 -6.68729067e-01 -4.97985065e-01 2.29142338e-01
2.34158352e-01 -6.84394360e-01 7.00396478e-01 -5.76845789e-03
-1.95997357e+00 4.49331492e-01 -1.61442667e-01 -3.76818508e-01
3.65837127e-01 6.28686666e-01 1.73230425e-01 7.14886338e-02
-1.59723595e-01 2.26092085e-01 4.97740000e-01 -3.09463918e-01
1.20034903e-01 -4.08206463e-01 3.12131703e-01 -3.62628736e-02
1.99333847e-01 -4.38749731e-01 -3.20090465e-02 2.02485397e-01
4.76140231e-01 -1.44868052e+00 -6.29609823e-01 -4.78919506e-01
-5.31187296e-01 1.83217794e-01 2.41728112e-01 -2.05112472e-01
1.17259026e+00 -2.04237437e+00 7.15089142e-01 8.01052272e-01
2.93095946e-01 -2.11319938e-01 1.98890711e-03 1.05982399e+00
1.31319597e-01 2.01850608e-02 -1.02851212e-01 1.93324890e-02
3.97824109e-01 3.64898324e-01 -1.76110953e-01 7.97520280e-01
-1.27496962e-02 9.32069480e-01 -9.28259790e-01 -2.06304803e-01
2.78645474e-02 4.10935223e-01 -1.16310060e+00 -2.34042466e-01
-4.72422838e-02 4.96684760e-01 -5.32151818e-01 -4.83212322e-02
9.62851346e-01 -4.74247970e-02 6.21901929e-01 -5.99051416e-01
-6.30827308e-01 6.64887786e-01 -1.58927226e+00 1.70436156e+00
-2.10420072e-01 2.92038769e-01 3.20526361e-01 -8.74959469e-01
4.44871992e-01 1.23440847e-01 5.48794389e-01 -6.93154216e-01
3.94288510e-01 5.12803912e-01 3.99419695e-01 -1.75326131e-02
8.83926928e-01 -4.22733963e-01 -3.12754124e-01 5.26519060e-01
7.42009580e-02 -8.23867857e-01 6.11258686e-01 5.62852085e-01
1.09342897e+00 2.09582210e-01 -9.36152712e-02 -4.79255438e-01
4.88612503e-01 2.70209461e-01 3.05352271e-01 8.02035034e-01
-9.62205157e-02 -9.86626521e-02 8.80667329e-01 -4.71600085e-01
-1.38074303e+00 -1.06506312e+00 -3.32850069e-01 6.43813074e-01
3.00164938e-01 -1.06088769e+00 -9.04948831e-01 3.92458811e-02
-3.96711193e-02 6.86043918e-01 -4.07141805e-01 -4.30354595e-01
-5.15171528e-01 -1.01478219e+00 5.61277092e-01 -1.17077783e-01
4.18862015e-01 -2.53438026e-01 -3.85582447e-01 2.57120103e-01
8.58170837e-02 -1.09529471e+00 -2.70038038e-01 3.40498596e-01
-8.76950800e-01 -6.66108847e-01 2.91216910e-01 1.53109625e-01
6.86128855e-01 -2.42648304e-01 6.01454020e-01 -2.33538464e-01
-4.87661779e-01 3.50499712e-02 1.03000768e-01 3.82840931e-01
-4.86914426e-01 -5.15938997e-02 1.64004862e-01 -1.37079313e-01
1.30834579e-01 -6.03087127e-01 -5.75237155e-01 -1.33391812e-01
-8.66150677e-01 -2.43620515e-01 3.66279632e-01 1.12839353e+00
7.58348882e-01 8.84111226e-02 -5.23656666e-01 -5.29284060e-01
4.23649251e-01 -3.82456258e-02 -1.23320866e+00 -1.45037934e-01
-6.22951925e-01 1.07407510e+00 6.33245051e-01 9.47304815e-02
-5.00908077e-01 1.54106155e-01 1.99946329e-01 1.32202819e-01
8.17016363e-01 4.53451663e-01 4.63535190e-01 -7.79399574e-01
7.93255568e-01 1.86687127e-01 -1.32610053e-01 -1.27097964e-01
7.65176892e-01 3.10842246e-01 1.86258867e-01 -8.27116907e-01
1.10049140e+00 6.41099691e-01 1.12049961e+00 -6.93422079e-01
-4.76489007e-01 -1.37750119e-01 -6.44000947e-01 2.33351309e-02
9.26003098e-01 -4.82063770e-01 -1.36910367e+00 -1.75136067e-02
-1.06618261e+00 2.50965208e-01 -4.48042154e-01 8.41451466e-01
-3.49402696e-01 5.17229974e-01 -7.48182058e-01 -7.90454030e-01
-2.51568079e-01 -1.62230086e+00 9.43641424e-01 -4.76135015e-02
8.76978487e-02 -5.96963465e-01 3.17783207e-01 1.93464875e-01
5.75374305e-01 1.74723387e-01 9.30553913e-01 -1.26700662e-02
-1.36312830e+00 -5.63547373e-01 -1.92913618e-02 2.39296094e-01
-7.96084166e-01 2.83395555e-02 -6.57856286e-01 -5.72138369e-01
-1.47720680e-01 -1.42571405e-01 5.12885690e-01 -1.22738697e-01
2.43458405e-01 -1.70420259e-02 2.62053180e-02 8.10423732e-01
1.70100021e+00 -3.46031070e-01 5.47212660e-01 -6.35891128e-03
6.71144903e-01 4.44767401e-02 8.06839466e-02 4.54972506e-01
8.87710601e-03 5.85900962e-01 3.03191036e-01 4.63167548e-01
3.74086827e-01 -1.64105460e-01 6.12496853e-01 1.25370169e+00
-4.42220896e-01 5.48591018e-01 -7.72106290e-01 9.63916779e-02
-1.53358102e+00 -8.23098719e-01 -4.09406185e-01 2.60336018e+00
6.92761421e-01 -5.24144573e-03 -3.42468262e-01 -3.03052068e-02
1.86454728e-01 1.51207596e-02 -3.25925976e-01 -1.05360460e+00
1.50309220e-01 8.81246388e-01 1.28163445e+00 8.91363204e-01
-5.22609651e-01 9.50057626e-01 6.70351601e+00 6.40238583e-01
-1.03644443e+00 3.00897896e-01 -2.44935513e-01 -2.94146210e-01
-4.45487678e-01 1.02222264e+00 -6.27432287e-01 7.43904188e-02
1.15423703e+00 -1.21519245e-01 1.28155255e+00 4.73060310e-01
1.43389210e-01 -2.66255289e-01 -1.11578882e+00 8.59101593e-01
-5.53368986e-01 -1.31445348e+00 -1.93274811e-01 4.21013504e-01
9.04216409e-01 4.46684718e-01 -1.08388335e-01 2.96925426e-01
4.89346236e-02 -9.84987378e-01 5.44082701e-01 4.98384029e-01
6.80232584e-01 -7.89947212e-01 7.42205203e-01 8.34621936e-02
-8.13098729e-01 9.68121067e-02 -4.56418753e-01 -3.79100770e-01
1.02406792e-01 4.44739908e-01 -6.74610078e-01 4.52902585e-01
2.36316308e-01 -2.05212221e-01 -1.04654409e-01 4.46950078e-01
-8.64753723e-02 3.86419117e-01 -8.18452597e-01 -4.00006831e-01
6.31173193e-01 -1.24832535e+00 7.56033242e-01 6.52292132e-01
1.05759293e-01 6.23946786e-02 -7.83749372e-02 1.30946720e+00
-8.32046345e-02 1.01858690e-01 -2.02915534e-01 -5.48129678e-01
3.33343506e-01 1.24034977e+00 -4.30400103e-01 -9.85014737e-02
-2.09135696e-01 9.76251304e-01 1.85435981e-01 4.17068958e-01
-7.07854748e-01 -9.52374101e-01 6.87138855e-01 -6.43018112e-02
4.15911436e-01 -8.10475826e-01 -2.54786853e-02 -1.28247666e+00
5.11388555e-02 -6.88174307e-01 -3.13270330e-01 -9.25316960e-02
-4.53602940e-01 1.44537807e-01 -7.78378397e-02 -5.52981317e-01
-4.13219810e-01 -7.79765904e-01 -1.99648514e-01 1.03939831e+00
-1.11212230e+00 -5.81147373e-01 2.55624771e-01 3.31469834e-01
-1.00013900e+00 5.67054033e-01 1.07141626e+00 1.51130974e-01
-5.72629511e-01 5.02946615e-01 8.73539150e-01 -2.72821225e-02
1.49336264e-01 -1.05947101e+00 1.94215298e-01 6.95011973e-01
-8.68802145e-02 1.14540291e+00 9.21554863e-01 -1.66279122e-01
-2.45902944e+00 -3.21261346e-01 7.83965468e-01 -2.15961412e-01
9.45614159e-01 -6.45890653e-01 -2.40553185e-01 4.77614820e-01
4.83044721e-02 8.67659450e-02 4.62921798e-01 1.06327556e-01
-5.52430749e-01 -1.60922229e-01 -1.01069772e+00 6.21605515e-01
7.43811548e-01 -1.21434331e+00 7.29598254e-02 6.70935571e-01
2.09040597e-01 -4.12661999e-01 -1.30449271e+00 -1.25146985e-01
5.58924437e-01 -9.26025689e-01 7.11927295e-01 -5.67279994e-01
-1.24847181e-01 -6.88598096e-01 -3.67346168e-01 -1.05653822e+00
-1.17335290e-01 -1.31753790e+00 1.35084003e-01 3.38263929e-01
4.44434315e-01 -7.23477602e-01 6.59920812e-01 7.35262334e-01
1.35002658e-01 -3.69041532e-01 -1.23915040e+00 -6.27184391e-01
2.37101063e-01 -1.90500796e-01 4.51405525e-01 7.73217738e-01
4.80236083e-01 5.98789215e-01 -1.88681632e-01 2.08031356e-01
7.68085659e-01 2.25568533e-01 6.25248909e-01 -6.25667930e-01
-7.70244122e-01 -4.28244352e-01 -8.76470566e-01 -8.07311177e-01
-2.68530130e-01 -1.26193333e+00 -2.99843788e-01 -7.71621525e-01
2.38578454e-01 -1.88031435e-01 -6.72985017e-02 6.87847510e-02
2.37609133e-01 8.91159549e-02 1.01062402e-01 -5.49569987e-02
-4.64280367e-01 5.29908061e-01 1.33235216e+00 -1.62229389e-02
-1.14188336e-01 -5.00190139e-01 5.90963066e-02 1.16728157e-01
4.04502749e-01 -7.10403740e-01 -1.17157117e-01 -1.99968487e-01
1.04992235e+00 1.20112196e-01 1.83332190e-01 -7.53730536e-01
3.97986561e-01 1.18477173e-01 -3.45574111e-01 -2.10722178e-01
5.11175811e-01 -2.15310648e-01 4.08124894e-01 8.53187621e-01
1.07321464e-01 7.50666782e-02 -7.80631974e-02 3.75656396e-01
-1.10417277e-01 -3.36299390e-01 8.73738170e-01 -1.18728630e-01
-1.43904209e-01 1.76389828e-01 -2.88974136e-01 -1.27952948e-01
7.13968933e-01 9.33907256e-02 -7.13250637e-02 -1.61642164e-01
-7.87289083e-01 -1.63268560e-05 6.15476191e-01 -4.65729803e-01
1.79325398e-02 -1.15986931e+00 -4.41750973e-01 2.66392380e-01
-3.56557935e-01 -2.43874401e-01 3.64657849e-01 1.28122330e+00
-1.10568118e+00 6.08014762e-01 -1.13990523e-01 -4.62009400e-01
-6.92567229e-01 4.27435368e-01 7.22382665e-01 -3.90437663e-01
-8.19342136e-02 7.43696749e-01 -4.20379728e-01 -6.13722444e-01
-3.36295307e-01 -3.23232621e-01 6.30138338e-01 -2.05040887e-01
1.85368717e-01 7.62633204e-01 1.97030962e-01 -6.29862607e-01
-3.89273137e-01 5.47452152e-01 6.31021336e-02 -4.34399933e-01
1.01754463e+00 2.25764766e-01 -9.44720387e-01 -1.88773833e-02
1.63309801e+00 1.82927474e-01 -5.84767401e-01 -1.10222466e-01
-2.39656076e-01 1.06714107e-03 5.59107244e-01 -1.27700642e-01
-5.68872988e-01 9.52626526e-01 5.77967703e-01 6.65373504e-02
5.01205802e-01 -4.29614484e-01 7.66467035e-01 1.12769246e+00
7.49850512e-01 -1.26923931e+00 -5.79354584e-01 6.48655057e-01
5.74187040e-02 -6.60593152e-01 2.83609241e-01 -2.67944068e-01
-1.29175838e-02 1.16211796e+00 -2.71812469e-01 -2.68129736e-01
4.84154046e-01 2.11413041e-01 -4.58128452e-01 -2.74987876e-01
-3.25942069e-01 -6.25216886e-02 -9.01343301e-02 -1.37497157e-01
3.26732606e-01 5.06536901e-01 -7.21273661e-01 -1.13376014e-01
-5.87993681e-01 -3.99252683e-01 7.57205665e-01 8.61099839e-01
-1.10008098e-01 -1.64263844e+00 -3.44538689e-01 4.21366654e-02
-2.04568475e-01 -4.74295169e-01 -2.23186225e-01 6.10486567e-01
-6.31529912e-02 7.57201076e-01 -1.45340621e-01 -4.93115276e-01
2.87167400e-01 3.00339669e-01 1.04824555e+00 -2.78268874e-01
-5.42514741e-01 -2.85404980e-01 3.10876202e-02 -1.06482375e+00
6.44607171e-02 -6.11941874e-01 -1.59368062e+00 -1.03028464e+00
-5.30610740e-01 6.57095730e-01 1.19520271e+00 8.49540949e-01
5.92562973e-01 1.54441312e-01 6.88331127e-01 -6.46216989e-01
-1.51755905e+00 -6.74099565e-01 -8.08407843e-01 2.04262450e-01
2.29990467e-01 -4.36568320e-01 -4.20896053e-01 -7.02321708e-01] | [5.643873691558838, 4.874456405639648] |
f48be3cf-6718-4992-9b50-9c6b2236f451 | improving-neural-morphological-tagging-using | null | null | https://www.researchgate.net/publication/330224906_Improving_neural_morphological_Tagging_using_Language_Models | http://www.dialog-21.ru/media/4530/sorokinaa.pdf | Improving neural morphological Tagging using Language Models | This paper addresses the task of morphological tagging and demonstrates how neural network architectures bene t from using language models for morphological tags. We show that incorporating the probabilities from language model on morphological tags improves the quality of character-based morphological tagging, reducing the error by 10%. | ['Alexey Sorokin'] | 2018-06-02 | null | null | null | dialogue-international-conference-on | ['morphological-tagging'] | ['natural-language-processing'] | [-4.00035363e-03 3.17296147e-01 -2.31868416e-01 -5.41160762e-01
-8.73675108e-01 -8.79909694e-01 4.25844789e-02 5.84014356e-01
-1.05183434e+00 5.39419293e-01 4.08330202e-01 -8.50661278e-01
3.32506120e-01 -7.28454471e-01 -4.23885196e-01 -4.24953699e-01
-2.22927332e-01 2.17981964e-01 3.32767129e-01 1.81391940e-01
1.67391017e-01 6.05676115e-01 -4.82003033e-01 3.47212195e-01
6.02884114e-01 3.49162072e-01 3.00720632e-01 4.90611106e-01
-5.00149190e-01 4.70833868e-01 -9.22935367e-01 -9.67358530e-01
1.30937621e-01 -8.47722590e-02 -6.76888764e-01 -1.04007728e-01
5.95714986e-01 3.41874957e-02 -4.68945652e-01 1.31918025e+00
2.08525673e-01 1.34586185e-01 7.08357513e-01 -3.41327369e-01
-1.16278255e+00 1.26949346e+00 -1.56437680e-01 5.06034553e-01
-4.77835387e-02 -1.73561350e-01 1.38517880e+00 -7.40359783e-01
5.31751454e-01 1.27147388e+00 1.26657891e+00 5.69052398e-01
-1.25458896e+00 -1.84998646e-01 2.09520549e-01 -2.73263007e-01
-1.48432016e+00 -3.40728730e-01 5.85737705e-01 -3.74535590e-01
1.54913402e+00 -9.84270126e-02 5.21238744e-01 4.17028278e-01
2.84346282e-01 8.05689573e-01 9.72091734e-01 -6.33775830e-01
-9.76713076e-02 -7.46440366e-02 3.91824275e-01 1.14269066e+00
4.20309424e-01 6.06595501e-02 -2.59829193e-01 -2.95294464e-01
9.21831667e-01 -6.10220194e-01 3.76278073e-01 1.94239736e-01
-8.22161615e-01 8.68335545e-01 2.73940146e-01 6.38697147e-01
-2.06481859e-01 2.88613796e-01 4.10080045e-01 1.07521176e-01
4.83189493e-01 8.65663409e-01 -8.45888436e-01 7.92410597e-02
-9.12321210e-01 -5.60095727e-01 7.29676545e-01 9.28929985e-01
5.75737178e-01 7.50292301e-01 2.41047189e-01 1.21861255e+00
2.31173500e-01 6.37284756e-01 5.45361519e-01 -8.50029588e-01
1.91962987e-01 4.51064974e-01 -2.18660772e-01 -5.98707438e-01
-5.69504738e-01 -3.47457469e-01 -2.53118873e-01 -1.68948427e-01
4.83725011e-01 -3.49823982e-01 -1.14323580e+00 1.91292036e+00
-1.26303241e-01 -4.00511920e-01 -1.80399921e-02 7.48883188e-02
6.34954095e-01 6.14914060e-01 9.49439645e-01 -1.31755114e-01
1.32713413e+00 -5.86248755e-01 -6.90769017e-01 -5.11318564e-01
8.11416328e-01 -7.29228556e-01 7.57903576e-01 3.84407043e-02
-1.47478223e+00 -5.68968594e-01 -8.82747829e-01 2.14303453e-02
-5.33956587e-01 3.33888948e-01 1.00803280e+00 9.35599864e-01
-1.56711352e+00 6.87784314e-01 -1.07679713e+00 -5.87047935e-01
1.85498565e-01 7.74958134e-01 -5.25663376e-01 5.89771211e-01
-1.00535786e+00 1.20757699e+00 9.28890109e-01 -2.59242598e-02
-3.18105131e-01 -2.00149506e-01 -1.09487939e+00 4.71128970e-02
-2.26128399e-01 -2.61918724e-01 1.44909346e+00 -8.34536254e-01
-1.15051579e+00 1.20044911e+00 -2.58798689e-01 -3.47750604e-01
-4.28554177e-01 2.95410324e-02 -5.55784047e-01 1.42805174e-01
-2.66913444e-01 8.88312578e-01 1.07643321e-01 -1.00699472e+00
-8.90407622e-01 1.02922283e-01 -7.66350627e-02 -9.52492282e-02
-6.04144037e-01 4.78424668e-01 -1.91543028e-01 -9.17783558e-01
2.41345204e-02 -8.18631768e-01 -3.12997639e-01 -4.30639267e-01
-6.65590093e-02 -4.41250831e-01 -7.32144667e-03 -1.27628839e+00
1.10672283e+00 -2.33053803e+00 -3.19272280e-01 -6.28735358e-03
-9.85545367e-02 4.95501697e-01 -5.80788791e-01 2.61121988e-01
-7.49127716e-02 6.65244102e-01 -2.78586954e-01 -1.71058029e-01
1.43021509e-01 7.33373702e-01 -1.31077087e-02 1.62642151e-01
4.94081527e-01 1.11803114e+00 -6.42508209e-01 -6.01908743e-01
-1.49487242e-01 3.05103213e-01 -4.69356179e-01 9.93755274e-03
7.65526146e-02 -3.10348660e-01 1.48752138e-01 5.18181801e-01
4.02794123e-01 2.58843035e-01 7.27673471e-01 5.09245768e-02
-1.77269921e-01 1.05788589e+00 -6.30493164e-01 1.59387374e+00
-5.78037322e-01 4.56454873e-01 2.83671081e-01 -7.28343785e-01
9.19045091e-01 4.50139493e-01 1.53744072e-01 -2.97499985e-01
1.53537199e-01 2.73929119e-01 5.28040886e-01 -7.34729618e-02
5.60452521e-01 -5.80740213e-01 -4.70444530e-01 4.96806681e-01
5.00887692e-01 -9.17929132e-03 1.56728789e-01 -1.32080957e-01
1.10398722e+00 5.49311331e-03 5.87323546e-01 -3.69102448e-01
1.65833488e-01 2.33311906e-01 7.70921946e-01 7.35468209e-01
-9.32361484e-02 -3.38599756e-02 1.61253642e-02 -4.14428711e-01
-1.08762717e+00 -1.23445880e+00 -2.24082604e-01 1.62902975e+00
-5.70697904e-01 -1.84339210e-01 -9.17822540e-01 -9.65327919e-01
3.73980217e-02 8.72083485e-01 -4.54025000e-01 -1.05884388e-01
-1.18836045e+00 -1.13918483e+00 1.15958500e+00 1.08167183e+00
-7.97508061e-02 -1.57069206e+00 -1.00615747e-01 5.04290998e-01
1.54292941e-01 -1.08493054e+00 -4.31559891e-01 1.00265753e+00
-1.30688334e+00 -5.37919700e-01 -2.29631275e-01 -1.49803650e+00
6.28418088e-01 -7.24189878e-02 1.02812231e+00 4.85151261e-01
1.55107290e-01 5.16324006e-02 -3.50862414e-01 -1.34899035e-01
-9.06792164e-01 4.78882313e-01 -7.90018439e-02 -7.18336344e-01
6.68974161e-01 -6.64761841e-01 2.90457219e-01 -2.23691300e-01
-9.31827486e-01 -5.42229593e-01 8.86930227e-01 7.42788136e-01
4.50496674e-01 -9.21465456e-02 3.59050810e-01 -9.76146877e-01
5.85084260e-01 -5.29226661e-02 -4.91766214e-01 1.60118595e-01
-4.62882519e-01 3.25336456e-01 5.33785820e-01 -6.88264966e-01
-1.09069324e+00 3.22567612e-01 -7.22673118e-01 4.38777447e-01
-5.47088444e-01 5.86168110e-01 -2.94959515e-01 -1.91010267e-01
4.27031666e-01 1.56336933e-01 -4.71196741e-01 -9.06929493e-01
5.89992702e-01 5.11641562e-01 7.16351867e-01 -4.96951461e-01
7.07563221e-01 2.04370692e-02 -1.08405516e-01 -6.39351010e-01
-8.55804861e-01 -3.64085793e-01 -1.27518153e+00 4.06987011e-01
9.45879459e-01 -7.17475176e-01 -1.60475597e-01 2.02529207e-01
-1.32413423e+00 -5.19410014e-01 -3.34758908e-01 3.82286668e-01
-4.40270156e-01 6.17198169e-01 -1.54193282e+00 -6.52528405e-01
-2.01782003e-01 -5.48569143e-01 5.24309337e-01 -4.94443849e-02
-3.93304884e-01 -1.55593073e+00 7.55028576e-02 -3.52870941e-01
1.94611534e-01 -4.69756484e-01 1.64006376e+00 -1.12646711e+00
-2.50709020e-02 -2.15874866e-01 6.76155463e-02 2.85566658e-01
1.30105928e-01 -1.72045037e-01 -7.44883776e-01 7.71674365e-02
-2.37233892e-01 -1.92006137e-02 8.89957249e-01 3.55873138e-01
5.32615125e-01 -2.60610759e-01 -1.35536939e-01 5.95991313e-01
1.46873724e+00 5.76557517e-01 1.78769618e-01 2.80974358e-01
7.24374592e-01 5.66958904e-01 6.38289079e-02 -1.83635607e-01
2.66101718e-01 3.58397424e-01 -1.58113912e-01 -1.44526228e-01
-4.21920776e-01 -5.81120014e-01 4.50652450e-01 1.55213106e+00
3.86009067e-01 -3.06861460e-01 -9.66943622e-01 9.44327533e-01
-1.51789653e+00 -6.47380292e-01 -1.45403564e-01 1.80380750e+00
1.22313702e+00 2.35214040e-01 6.54825568e-02 -8.60117674e-02
1.01645219e+00 2.53146049e-02 -4.27096225e-02 -9.82892513e-01
-3.47691923e-01 6.32108390e-01 7.14204907e-01 8.47689390e-01
-1.28619015e+00 1.73766589e+00 8.74257183e+00 7.43243337e-01
-5.80716193e-01 3.92469674e-01 5.68553694e-02 5.38672149e-01
-2.45120898e-01 2.83202603e-02 -9.72062409e-01 5.55610321e-02
1.49661291e+00 1.94283873e-01 3.86659801e-01 6.13056123e-01
-7.71654770e-02 -4.26184037e-04 -9.06442702e-01 2.86304176e-01
1.88004807e-01 -8.41412485e-01 2.97742397e-01 3.31465483e-01
4.82570827e-01 2.32210889e-01 -1.24485858e-01 8.15554187e-02
1.27934849e+00 -6.59732223e-01 9.35552418e-01 1.40061900e-01
7.42523730e-01 -7.98106670e-01 7.71432221e-01 2.16917247e-01
-1.08953166e+00 1.32877707e-01 -9.23638403e-01 -4.48105708e-02
3.51889044e-01 2.86322176e-01 -1.05180502e+00 -6.48759380e-02
-9.41295996e-02 3.45894694e-01 -9.22389030e-01 1.35854518e+00
-9.12555218e-01 1.13613546e+00 -1.65064916e-01 -1.33167118e-01
3.26787055e-01 4.15389501e-02 3.35253000e-01 2.10324025e+00
1.98849067e-01 8.05410463e-03 3.91202927e-01 4.84148651e-01
-2.20019042e-01 3.12197387e-01 -4.41221625e-01 -4.51656729e-01
6.59509063e-01 1.14699817e+00 -1.14363658e+00 -4.31208163e-01
-3.40348274e-01 9.20718312e-01 6.26618207e-01 -7.26025403e-02
-3.13468099e-01 -4.95232821e-01 4.38923091e-01 -2.06868246e-01
5.64372540e-01 -7.79778719e-01 -5.06526887e-01 -7.28595614e-01
-4.38733131e-01 -4.69778746e-01 5.36956608e-01 -7.43847311e-01
-1.43455791e+00 6.53743744e-01 -3.14764738e-01 -2.84386784e-01
-1.77861616e-01 -1.10726535e+00 -2.25978151e-01 9.66024756e-01
-1.07537484e+00 -1.26319242e+00 7.01876283e-01 -8.48532319e-02
3.20351511e-01 -1.09634325e-01 1.28542864e+00 2.27302670e-01
-3.44236761e-01 8.80157411e-01 1.06416615e-02 9.49514866e-01
4.10694420e-01 -1.75130999e+00 1.27284348e+00 1.12487948e+00
8.73551428e-01 9.29227769e-01 1.61721677e-01 -9.06757295e-01
-7.64363229e-01 -1.02439809e+00 1.69472325e+00 -6.49301171e-01
1.07204270e+00 -3.45192343e-01 -9.98656690e-01 7.80878842e-01
4.45143372e-01 -2.83223927e-01 1.04838085e+00 3.11670125e-01
-7.35705018e-01 4.14175421e-01 -1.14957523e+00 4.94119048e-01
1.36718643e+00 -8.98519039e-01 -1.17269361e+00 1.29282951e-01
7.82437742e-01 1.76905870e-01 -9.29346681e-01 -1.15529425e-01
3.66172642e-01 -1.52209729e-01 6.81811512e-01 -1.07519519e+00
-1.76474020e-01 -1.48102731e-01 -2.49849454e-01 -1.58150661e+00
-9.76264060e-01 -3.85554373e-01 3.94343406e-01 1.27073586e+00
1.01778126e+00 -2.88341135e-01 7.64997303e-01 2.72685766e-01
-4.09295946e-01 6.56171367e-02 -7.36487210e-01 -9.81362581e-01
7.39920199e-01 -4.25828278e-01 2.20260471e-01 9.13954139e-01
3.87453318e-01 2.75461257e-01 1.35548890e-01 1.72057658e-01
1.77854031e-01 -4.68976259e-01 -3.66552293e-01 -1.23311031e+00
-4.75235760e-01 -7.03507304e-01 -5.50948977e-01 -1.04563844e+00
6.61354363e-01 -1.38656628e+00 4.69288766e-01 -1.61013401e+00
2.49875426e-01 -3.39378834e-01 -4.75114107e-01 1.11862910e+00
-3.75024587e-01 5.91664553e-01 3.29397112e-01 -1.70663759e-01
-4.83868957e-01 -1.98622704e-01 4.49137181e-01 -1.29424661e-01
1.66769996e-01 -4.81663376e-01 -8.48473847e-01 1.14857864e+00
1.02958620e+00 -1.01995993e+00 3.28945011e-01 -9.21021461e-01
4.94297415e-01 -1.01430684e-01 -3.20888013e-01 -7.83200562e-01
1.80314458e-03 5.86130423e-04 4.50014323e-01 -2.69577742e-01
2.23006174e-01 -5.63878775e-01 -3.33973289e-01 4.86795455e-01
-5.93801439e-01 8.80046964e-01 5.51465631e-01 3.06374669e-01
3.16652395e-02 -8.86554420e-01 8.12924683e-01 -4.67502862e-01
-7.22888589e-01 -2.51499265e-01 -1.15718508e+00 1.19663261e-01
1.29217803e-01 -1.40899047e-01 -2.03864813e-01 9.80241448e-02
-1.25465536e+00 -6.30285561e-01 5.63357413e-01 -6.09929999e-03
1.79768786e-01 -1.17363751e+00 -3.27448130e-01 1.20969780e-01
-2.43133843e-01 -9.05395091e-01 -6.53186023e-01 3.13871145e-01
-4.99141902e-01 5.32720685e-01 -1.89848363e-01 1.84426591e-01
-1.25290835e+00 6.18860722e-01 3.43670666e-01 -2.45390028e-01
-3.47752929e-01 1.36064565e+00 -1.76250979e-01 -6.05062306e-01
2.16627717e-01 -3.10655564e-01 -2.63745755e-01 -1.66561469e-01
2.82928199e-01 1.97096430e-02 1.34534091e-01 -7.54013896e-01
-3.48636478e-01 4.41503435e-01 1.94430090e-02 -5.08253157e-01
1.40281963e+00 1.26941144e-01 -3.43994081e-01 5.32729805e-01
7.45423257e-01 6.83512330e-01 -8.30065072e-01 -2.30527341e-01
7.61641681e-01 1.63923711e-01 -6.83207586e-02 -1.28907406e+00
-8.80542397e-01 1.15028858e+00 1.73336953e-01 1.00105032e-01
6.57261908e-01 8.37139902e-04 9.97832716e-01 7.46388435e-01
4.79407430e-01 -1.06013668e+00 -4.87568527e-01 1.00642323e+00
1.03267916e-01 -4.60114777e-01 -3.08292091e-01 -4.57359165e-01
-4.03419495e-01 9.20750439e-01 3.41855854e-01 -3.60610753e-01
5.88917494e-01 7.59692788e-01 3.91096264e-01 8.15367103e-02
-5.10084689e-01 -4.86799449e-01 1.85976774e-01 8.79777551e-01
8.58849943e-01 4.49991405e-01 -7.14409471e-01 5.18102348e-01
-4.92349595e-01 -7.63587475e-01 4.84797001e-01 8.53553474e-01
-9.09345567e-01 -1.72007000e+00 -1.80702299e-01 3.84287119e-01
-1.08201098e+00 -9.11294281e-01 -9.89885926e-01 7.36178577e-01
4.82158273e-01 9.14260507e-01 2.66553789e-01 -4.01789010e-01
-6.14537746e-02 8.80411208e-01 7.12516069e-01 -1.23323047e+00
-1.09577549e+00 2.61373878e-01 7.21046150e-01 2.04552010e-01
-3.66339743e-01 -8.34082127e-01 -1.40259123e+00 -7.00829849e-02
-3.61772388e-01 5.99806964e-01 6.07729137e-01 8.56347263e-01
-4.60080011e-03 2.29326740e-01 1.04037458e-02 -1.87932059e-01
-5.48987448e-01 -1.16610706e+00 -6.68798804e-01 1.57631919e-01
-2.54985541e-01 -2.14548633e-01 -2.65947521e-01 3.44265819e-01] | [10.352559089660645, 10.027551651000977] |
bea78c5b-1d71-4c5c-b4d1-9b825bce032a | watt-effnet-a-lightweight-and-accurate-model | 2304.10811 | null | https://arxiv.org/abs/2304.10811v2 | https://arxiv.org/pdf/2304.10811v2.pdf | WATT-EffNet: A Lightweight and Accurate Model for Classifying Aerial Disaster Images | Incorporating deep learning (DL) classification models into unmanned aerial vehicles (UAVs) can significantly augment search-and-rescue operations and disaster management efforts. In such critical situations, the UAV's ability to promptly comprehend the crisis and optimally utilize its limited power and processing resources to narrow down search areas is crucial. Therefore, developing an efficient and lightweight method for scene classification is of utmost importance. However, current approaches tend to prioritize accuracy on benchmark datasets at the expense of computational efficiency. To address this shortcoming, we introduce the Wider ATTENTION EfficientNet (WATT-EffNet), a novel method that achieves higher accuracy with a more lightweight architecture compared to the baseline EfficientNet. The WATT-EffNet leverages width-wise incremental feature modules and attention mechanisms over width-wise features to ensure the network structure remains lightweight. We evaluate our method on a UAV-based aerial disaster image classification dataset and demonstrate that it outperforms the baseline by up to 15 times in terms of classification accuracy and 38.3% in terms of computing efficiency as measured by Floating Point Operations per second (FLOPs). Additionally, we conduct an ablation study to investigate the effect of varying the width of WATT-EffNet on accuracy and computational efficiency. Our code is available at \url{https://github.com/TanmDL/WATT-EffNet}. | ['Vu N. Duong', 'Daniel Puiu Poenar', 'Md Meftahul Ferdaus', 'Tanmoy Dam', 'Gao Yu Lee'] | 2023-04-21 | null | null | null | null | ['scene-classification'] | ['computer-vision'] | [-5.70469163e-02 -3.57463419e-01 -9.77729186e-02 -2.03257576e-01
-3.63138616e-01 -5.57291031e-01 2.89843023e-01 2.09936723e-01
-7.51739323e-01 3.91506523e-01 8.07426423e-02 -6.78483665e-01
-3.77292514e-01 -8.81385088e-01 -3.79639506e-01 -3.58065993e-01
-3.68172109e-01 -1.75662972e-02 5.35195060e-02 -3.57190788e-01
2.42935345e-02 8.05769086e-01 -1.34257305e+00 -4.75511774e-02
8.10442328e-01 1.30449104e+00 8.95090774e-02 6.57358944e-01
6.96466327e-01 6.82488918e-01 -6.03544116e-01 -7.68577382e-02
7.38322616e-01 1.11988448e-01 -8.93341362e-01 -2.31175736e-01
5.01913786e-01 -8.22033286e-01 -6.08117819e-01 7.39452004e-01
6.51187360e-01 4.00234312e-01 2.96352565e-01 -1.39460921e+00
-1.90621361e-01 1.55246124e-01 -6.20818734e-01 7.99026549e-01
-7.89637119e-02 4.61849689e-01 9.19708252e-01 -6.95966899e-01
1.99000776e-01 9.22361553e-01 9.10066724e-01 5.79026975e-02
-1.06436884e+00 -8.14555764e-01 4.29055333e-01 1.70325294e-01
-1.49251914e+00 -5.39646626e-01 1.95897967e-01 -3.18248153e-01
1.31387496e+00 8.71439278e-02 6.72473431e-01 5.40743172e-01
2.91130036e-01 3.92347127e-01 4.02385384e-01 -5.33012152e-02
1.85965478e-01 -5.82638085e-01 5.59513830e-02 9.37438428e-01
5.89363754e-01 8.97227153e-02 -4.65009451e-01 -6.48303926e-02
6.07785046e-01 1.30592272e-01 -5.36349893e-01 1.49786189e-01
-1.00238729e+00 9.33083892e-01 1.03977740e+00 -3.23095801e-03
-6.82771266e-01 3.97080868e-01 5.34518600e-01 2.45385259e-01
5.56797922e-01 9.23601627e-01 -6.58891082e-01 -1.88337043e-01
-9.03891146e-01 2.59827614e-01 6.02010190e-01 7.76453137e-01
6.64877534e-01 3.99853259e-01 -6.08296469e-02 5.66955864e-01
-2.06216499e-01 5.82622349e-01 8.15436542e-02 -1.00666380e+00
2.42400095e-01 8.44033659e-01 1.43504024e-01 -1.30358171e+00
-6.96680069e-01 -7.32611239e-01 -9.49292660e-01 1.99567333e-01
-2.23873779e-02 -5.23694515e-01 -9.26211476e-01 1.61604059e+00
2.74291307e-01 -6.89085498e-02 -1.18299901e-01 1.07861745e+00
6.49491847e-01 6.22917533e-01 1.52938649e-01 2.06616491e-01
1.19871688e+00 -9.93992150e-01 -1.74035564e-01 -6.03288770e-01
7.42855787e-01 -4.36198235e-01 1.18283248e+00 7.32924491e-02
-6.92315638e-01 -3.40763181e-01 -1.19381666e+00 -4.23038639e-02
-2.97952145e-01 4.54273224e-01 7.49857247e-01 2.01854765e-01
-9.98240292e-01 6.32371783e-01 -8.59601736e-01 -4.99268532e-01
8.41355443e-01 5.57862937e-01 -3.41218501e-01 -2.25097999e-01
-1.01120913e+00 7.35784173e-01 1.65654406e-01 1.44518554e-01
-1.13628244e+00 -8.63500059e-01 -8.10101986e-01 1.94699168e-01
4.63888466e-01 -8.52087855e-01 1.24109340e+00 -6.67669713e-01
-8.35594594e-01 3.56411517e-01 1.43832222e-01 -8.66131961e-01
4.10394482e-02 -5.05780399e-01 8.40697363e-02 2.41572097e-01
2.17735127e-01 1.04095542e+00 7.27848887e-01 -5.56294501e-01
-8.15213084e-01 -2.19874039e-01 6.13957286e-01 2.62669832e-01
-7.14842796e-01 -2.16389373e-01 -1.28806859e-01 -6.19127274e-01
-1.13750160e-01 -9.73541558e-01 -2.89353281e-01 2.33411923e-01
-8.45218971e-02 1.26359193e-02 9.50740635e-01 -4.73190784e-01
1.27860785e+00 -2.20583582e+00 -1.34164467e-02 -1.79821312e-01
5.31024635e-01 5.37657559e-01 -4.18780088e-01 2.73876488e-01
1.75225303e-01 1.88381866e-01 -3.09789389e-01 -3.02358642e-02
-3.02377999e-01 5.90309761e-02 -4.38924164e-01 5.30054212e-01
3.18427563e-01 7.47009635e-01 -7.56261289e-01 -1.15564466e-01
6.66021258e-02 3.57576787e-01 -9.19430614e-01 1.15830354e-01
-6.63836673e-02 1.49399042e-01 -4.47157860e-01 1.08963501e+00
4.80802447e-01 -3.43250602e-01 -1.48936072e-02 -4.44103390e-01
-1.96162954e-01 7.78856948e-02 -5.21901727e-01 1.48006225e+00
-5.71947455e-01 8.45960617e-01 2.70277441e-01 -9.35982049e-01
6.42069399e-01 -1.30922794e-01 4.42471057e-01 -6.12216115e-01
5.13206780e-01 -1.74395973e-03 5.39479926e-02 -1.48493588e-01
7.46660054e-01 3.08226198e-01 -2.00935572e-01 2.98977911e-01
-9.48640406e-02 -3.26296762e-02 1.76524282e-01 1.55708924e-01
1.52388847e+00 -2.35168666e-01 2.25434691e-01 -5.23337126e-01
-3.35715748e-02 5.30859947e-01 5.46215296e-01 5.46318710e-01
-2.69913018e-01 -1.06744833e-01 2.86872238e-01 -1.03016925e+00
-7.46703982e-01 -4.75677609e-01 2.28453968e-02 1.21543586e+00
1.45264328e-01 -6.86397910e-01 -6.70633316e-01 -5.74919164e-01
9.41442177e-02 5.76141775e-01 -4.48999882e-01 -3.79509479e-01
-2.20776349e-01 -8.47810090e-01 8.50703597e-01 5.85236728e-01
1.00895309e+00 -7.59012759e-01 -1.52124107e+00 1.09509369e-02
-1.76144570e-01 -1.23329246e+00 -3.44740689e-01 2.20824555e-01
-7.23937452e-01 -1.25291169e+00 -2.69487947e-01 -3.38800490e-01
7.09503651e-01 7.02854931e-01 9.83608067e-01 4.60505635e-01
-6.25548065e-01 4.01306123e-01 -5.26415884e-01 -4.62989390e-01
5.05607247e-01 6.40721858e-01 2.80715674e-01 -4.20022875e-01
2.72282898e-01 -5.58526576e-01 -7.91443765e-01 1.61944076e-01
-9.34652627e-01 -6.03992827e-02 5.82217753e-01 8.08817506e-01
4.07975167e-01 2.85686135e-01 3.32550168e-01 -2.75445551e-01
7.42414355e-01 -5.44116139e-01 -8.56013298e-01 -1.26991034e-01
-5.91380179e-01 -1.99938014e-01 5.85403025e-01 -2.95114607e-01
-2.86233753e-01 -9.97818168e-03 2.25797445e-02 -6.54029846e-01
8.49499926e-02 7.95606911e-01 2.49248773e-01 -4.92933631e-01
6.80086493e-01 -7.00382143e-02 -8.93967450e-02 -6.52195290e-02
-2.76075117e-02 5.54150760e-01 4.09376979e-01 -2.67053723e-01
6.57949984e-01 3.40613842e-01 1.25309974e-01 -9.41833079e-01
-1.05644703e+00 -1.33161470e-01 -4.35134113e-01 -2.12072209e-01
6.02966428e-01 -1.26316893e+00 -8.05410862e-01 2.92374253e-01
-1.02742481e+00 -7.40406513e-01 -6.79494813e-02 4.55803931e-01
-5.69743589e-02 5.95497154e-03 -4.60921377e-01 -5.46572506e-01
-7.86602676e-01 -1.07428598e+00 1.06118464e+00 1.29704967e-01
-1.76771492e-01 -6.29626989e-01 -3.44205886e-01 1.35492980e-01
7.85636961e-01 3.90505433e-01 8.07972193e-01 -3.37536395e-01
-6.21631742e-01 -2.97036946e-01 -5.42783380e-01 2.71905303e-01
-3.00498884e-02 -2.86385268e-01 -7.31004119e-01 -6.67163253e-01
-2.52035141e-01 -4.40888524e-01 1.07380736e+00 4.31709677e-01
1.20946801e+00 -5.24210989e-01 -2.31756940e-01 9.59447503e-01
1.46673274e+00 3.19398157e-02 3.46850216e-01 4.91999239e-01
4.74754661e-01 2.66696513e-01 7.70469427e-01 8.04432511e-01
3.78026485e-01 5.27919590e-01 9.69898164e-01 -2.18644395e-01
-1.52872987e-02 -5.29222675e-02 2.68135339e-01 2.79122919e-01
-7.08414242e-02 -5.78059494e-01 -1.25978565e+00 5.75836480e-01
-1.64203584e+00 -6.32921576e-01 4.60687995e-01 2.14758134e+00
2.99403280e-01 5.19070104e-02 -1.03602104e-01 9.13678110e-02
3.21040630e-01 2.38917604e-01 -7.62766719e-01 -1.42301768e-01
1.54916838e-01 4.03795093e-02 8.11184227e-01 4.32725191e-01
-1.31691408e+00 1.24543047e+00 5.79170895e+00 6.90244973e-01
-1.33313119e+00 4.79994640e-02 6.67118669e-01 -4.73174363e-01
3.06745827e-01 -1.52256489e-01 -7.85149455e-01 1.59884244e-01
8.67226958e-01 -8.72257426e-02 6.21010661e-01 1.00821137e+00
1.17146157e-01 -2.48308644e-01 -7.36125052e-01 9.37886715e-01
-4.13588770e-02 -1.33255219e+00 8.60584229e-02 1.77913263e-01
2.52953261e-01 4.22984868e-01 7.86245242e-02 2.79722571e-01
2.54052669e-01 -1.04037046e+00 5.59253395e-01 1.82202920e-01
1.08284044e+00 -9.49645162e-01 9.23785150e-01 2.20978811e-01
-1.38965321e+00 -4.90528852e-01 -5.18453658e-01 -5.35101652e-01
-8.08921903e-02 4.45165485e-01 -8.14436793e-01 3.91129136e-01
1.12167454e+00 7.14087069e-01 -5.67892790e-01 9.12984312e-01
-1.12226598e-01 6.70970201e-01 -4.65312153e-01 1.21790417e-01
6.18488252e-01 2.91531742e-01 5.36711693e-01 9.59582269e-01
4.06425625e-01 5.42299867e-01 4.46770370e-01 4.80266631e-01
-2.53225088e-01 -2.82868981e-01 -7.72970915e-01 -1.28252357e-01
7.41492748e-01 1.39937103e+00 -7.33625472e-01 -4.68636677e-02
-6.29821867e-02 7.12220013e-01 3.25092614e-01 7.86073878e-02
-9.10455108e-01 -6.24980569e-01 9.96217191e-01 1.32825598e-01
2.31930047e-01 -6.07286990e-01 -2.04750031e-01 -8.54841650e-01
-2.70272493e-01 -9.08814132e-01 2.66115963e-01 -6.99155986e-01
-9.04767871e-01 1.25361121e+00 -2.17498243e-02 -1.11651337e+00
2.42312849e-01 -4.92823005e-01 -4.01457012e-01 5.25027096e-01
-1.59678054e+00 -1.12721968e+00 -9.44034100e-01 4.95831549e-01
6.86818480e-01 -2.97429144e-01 8.04036200e-01 3.69121969e-01
-8.62660408e-01 6.43670380e-01 -6.24881804e-01 6.50348812e-02
3.50831896e-01 -4.76201564e-01 4.99244153e-01 1.09194970e+00
-1.07038386e-01 4.58666831e-01 4.62701827e-01 -5.47359645e-01
-1.51738572e+00 -1.54088485e+00 5.09801328e-01 -2.88667083e-01
3.96175653e-01 -1.66560784e-01 -4.93176699e-01 5.72869956e-01
1.06259041e-01 1.15397222e-01 5.49931586e-01 7.59030972e-03
-1.30932495e-01 -3.28618854e-01 -1.03427207e+00 6.23066306e-01
1.11167860e+00 -1.93202510e-01 1.47650775e-04 4.89607275e-01
9.77164030e-01 -3.54783624e-01 -7.97082067e-01 7.59923398e-01
3.81524205e-01 -8.25607419e-01 8.08038592e-01 -4.80397731e-01
2.24597424e-01 -3.00024271e-01 -4.20623899e-01 -1.20998824e+00
-6.35137081e-01 -4.57660943e-01 1.47221953e-01 5.30706942e-01
3.21374267e-01 -7.39194870e-01 4.94841039e-01 4.58821833e-01
-4.64554489e-01 -1.08917284e+00 -7.36702204e-01 -7.51479805e-01
-3.22603256e-01 -3.47726583e-01 7.35607088e-01 7.91983664e-01
-3.93981457e-01 3.86962920e-01 -3.18660229e-01 5.12966216e-01
2.80817270e-01 7.18385950e-02 7.70652711e-01 -1.18882143e+00
2.51091927e-01 -2.08200485e-01 -4.31796163e-01 -9.80263531e-01
1.08779259e-01 -6.31947160e-01 2.16138028e-02 -1.49663711e+00
-1.82369664e-01 -6.39022708e-01 -1.27479285e-01 1.22755456e+00
1.63298219e-01 4.39342439e-01 3.48654091e-01 2.97692299e-01
-7.28654325e-01 7.70906985e-01 6.85848594e-01 -1.06179319e-01
-4.24736701e-02 -1.54413775e-01 -7.37649024e-01 7.40970850e-01
1.19659829e+00 -4.56375241e-01 -4.77696866e-01 -1.01411986e+00
1.55743644e-01 -1.97635844e-01 5.60764551e-01 -1.49816930e+00
5.03502071e-01 -1.21607475e-01 3.53461295e-01 -3.97111654e-01
3.87413174e-01 -7.86853075e-01 -1.98047355e-01 7.70314991e-01
-1.18558202e-02 6.70460224e-01 7.59355545e-01 4.59058195e-01
-2.03360468e-01 2.07518980e-01 7.08141387e-01 4.64232340e-02
-8.65786493e-01 7.26112247e-01 -5.33452094e-01 -1.03217766e-01
1.03305185e+00 -8.75928551e-02 -4.89369243e-01 -3.82862061e-01
-1.11417957e-01 3.87542188e-01 5.43377221e-01 2.14541346e-01
7.80176699e-01 -8.88409972e-01 -5.83211362e-01 3.05243045e-01
1.69118252e-02 8.30537602e-02 2.84435242e-01 9.60153699e-01
-9.66720998e-01 5.56156337e-01 -3.68348598e-01 -3.01152468e-01
-1.09086370e+00 3.25333416e-01 4.67822760e-01 -1.10487834e-01
-3.71739835e-01 1.03902054e+00 2.91904937e-02 -2.49605417e-01
2.49303311e-01 -3.18819731e-01 4.59785238e-02 -4.01526168e-02
4.71392184e-01 6.00711048e-01 1.20037094e-01 -3.82354528e-01
-5.73536456e-01 4.50982094e-01 6.81915879e-02 3.32237452e-01
1.46559203e+00 3.26062739e-02 -1.08982451e-01 -4.08146828e-01
7.66409099e-01 -3.78539741e-01 -1.44437611e+00 6.04035892e-02
-2.74466455e-01 -5.94975531e-01 7.38097131e-01 -7.62014508e-01
-1.39182484e+00 7.35168576e-01 6.33729219e-01 1.10393511e-02
1.63104725e+00 -4.21076238e-01 8.91943276e-01 7.82256544e-01
5.38656056e-01 -6.73130810e-01 7.25731626e-02 8.05320501e-01
1.02466691e+00 -1.17189825e+00 2.67792463e-01 -1.65477306e-01
-5.98929465e-01 1.02153587e+00 8.90331745e-01 -3.03832918e-01
6.66204214e-01 4.19117361e-01 -1.23599388e-01 -4.39956069e-01
-9.73137200e-01 -3.15982133e-01 -3.13961180e-03 4.17006314e-01
5.88049069e-02 -1.12584457e-01 1.37608379e-01 5.11530995e-01
-1.68183535e-01 -1.02382354e-01 3.37749630e-01 1.09409928e+00
-5.89655459e-01 -5.38092673e-01 5.38289994e-02 6.32317424e-01
-3.84348512e-01 -4.32563245e-01 -2.63380140e-01 7.67044902e-01
2.10193127e-01 1.02090049e+00 2.21422970e-01 -8.91623497e-01
3.55433673e-01 -4.92328912e-01 2.07481552e-02 -5.03996372e-01
-6.11193597e-01 -4.49492306e-01 2.00263634e-01 -7.93689251e-01
-2.92110980e-01 -1.99663356e-01 -9.77252126e-01 -6.93720400e-01
-3.78871799e-01 -5.79447160e-03 5.31450808e-01 5.03107607e-01
1.00607407e+00 5.82702994e-01 4.13495809e-01 -1.00315011e+00
-4.77013171e-01 -8.48367870e-01 -2.24081367e-01 -4.23172206e-01
3.40745777e-01 -8.30438554e-01 -3.32020819e-01 -2.71555454e-01] | [8.88074016571045, -0.3328262269496918] |
b7dff2ec-18ef-4834-9455-58ce1e3e6f60 | application-of-adversarial-examples-to | 2108.08972 | null | https://arxiv.org/abs/2108.08972v1 | https://arxiv.org/pdf/2108.08972v1.pdf | Application of Adversarial Examples to Physical ECG Signals | This work aims to assess the reality and feasibility of the adversarial attack against cardiac diagnosis system powered by machine learning algorithms. To this end, we introduce adversarial beats, which are adversarial perturbations tailored specifically against electrocardiograms (ECGs) beat-by-beat classification system. We first formulate an algorithm to generate adversarial examples for the ECG classification neural network model, and study its attack success rate. Next, to evaluate its feasibility in a physical environment, we mount a hardware attack by designing a malicious signal generator which injects adversarial beats into ECG sensor readings. To the best of our knowledge, our work is the first in evaluating the proficiency of adversarial examples for ECGs in a physical setup. Our real-world experiments demonstrate that adversarial beats successfully manipulated the diagnosis results 3-5 times out of 40 attempts throughout the course of 2 minutes. Finally, we discuss the overall feasibility and impact of the attack, by clearly defining motives and constraints of expected attackers along with our experimental results. | ['Tatsuya Mori', 'Jun Sakuma', 'Takeshi Sugawara', 'Taiga Ono'] | 2021-08-20 | null | null | null | null | ['ecg-classification'] | ['medical'] | [ 6.46190166e-01 5.77925622e-01 5.17026544e-01 -9.25010536e-03
-8.07994246e-01 -1.07393265e+00 1.47990122e-01 -9.85089168e-02
-1.48658708e-01 5.96387804e-01 -3.00181985e-01 -7.02956915e-01
-3.13547738e-02 -4.83635366e-01 -6.04203403e-01 -6.01134777e-01
-9.05664921e-01 1.25794873e-01 -2.10089147e-01 -1.11115657e-01
-8.09771046e-02 7.89649308e-01 -5.42101562e-01 5.05439527e-02
3.18095714e-01 8.04857254e-01 -9.37206030e-01 1.51909494e+00
1.10525954e+00 9.38909233e-01 -1.56216085e+00 -4.57312912e-01
6.11839652e-01 -6.77292824e-01 -6.97598219e-01 -4.41044390e-01
2.65541494e-01 -6.82230473e-01 -5.93747139e-01 7.34472871e-01
1.25733626e+00 -4.87128794e-01 4.29108202e-01 -1.46790755e+00
1.23596735e-01 9.11857843e-01 4.68912832e-02 3.46266359e-01
6.02642179e-01 6.97330654e-01 2.91955471e-01 -1.17555901e-01
3.44439864e-01 6.19793057e-01 9.62910414e-01 8.84536505e-01
-1.09974146e+00 -7.71087348e-01 -5.73469341e-01 -3.84906560e-01
-1.40748417e+00 -3.79983753e-01 1.08563995e+00 -2.75508285e-01
6.75113380e-01 6.71028197e-01 5.87556958e-01 1.82529712e+00
5.32074869e-01 4.05489236e-01 9.06772554e-01 -3.44490916e-01
4.50849831e-01 3.57943475e-02 -5.69027774e-02 4.18738484e-01
2.55237579e-01 6.14639461e-01 -1.99396878e-01 -7.88275063e-01
8.80520046e-01 -4.17587817e-01 -4.57003415e-01 1.28412442e-02
-1.25600123e+00 5.77343047e-01 5.98764904e-02 5.84121086e-02
-4.42849755e-01 5.43139517e-01 5.94069481e-01 3.57588232e-01
-1.91176713e-01 8.77383530e-01 -4.03628081e-01 -3.98411542e-01
-6.07503474e-01 1.96275085e-01 1.23053467e+00 7.25774884e-01
-4.04767990e-01 6.62719429e-01 -2.19057605e-01 1.60830349e-01
-3.19438204e-02 8.12334061e-01 3.92208278e-01 -8.95780265e-01
3.19079369e-01 -2.08328247e-01 5.79781421e-02 -9.64536190e-01
-5.33265829e-01 -5.10273814e-01 -7.41416514e-01 4.15266693e-01
3.60741347e-01 -9.83816564e-01 -5.43403029e-01 1.52479792e+00
2.33558998e-01 5.93966186e-01 4.68572915e-01 6.42604113e-01
6.23177171e-01 -3.56001444e-02 -2.98815127e-02 -2.40453273e-01
1.17570674e+00 -1.94697663e-01 -4.79862690e-01 1.36156365e-01
3.15364093e-01 -6.83951974e-01 7.30338275e-01 5.98443925e-01
-1.18195975e+00 -6.61607027e-01 -1.30174637e+00 9.43704844e-01
8.12626928e-02 -2.80120373e-01 3.31631541e-01 1.69585550e+00
-6.47604883e-01 7.20949352e-01 -1.00144053e+00 1.08390190e-01
3.40047151e-01 3.94312203e-01 -9.66038033e-02 5.55473447e-01
-1.63376439e+00 6.84602201e-01 4.92317341e-02 -9.93360430e-02
-1.24964058e+00 -7.40597308e-01 -7.84679532e-01 -2.35128790e-01
-8.16495642e-02 -6.51832461e-01 1.20889735e+00 -6.94085479e-01
-1.61479342e+00 8.02545547e-01 5.42048752e-01 -9.72101986e-01
8.16987216e-01 -1.43732384e-01 -8.61077368e-01 4.56657171e-01
-4.73893702e-01 -5.59318252e-02 9.33077753e-01 -1.11925876e+00
3.34064243e-03 7.11779073e-02 3.26098353e-01 -3.20123315e-01
-2.69331038e-01 1.95000209e-02 3.23255956e-01 -9.71698642e-01
-2.89341658e-01 -1.05614209e+00 -3.77748996e-01 -4.84738201e-01
-7.27438629e-01 7.35813856e-01 4.47888911e-01 -2.50496536e-01
1.12161326e+00 -2.22977591e+00 -3.62733424e-01 6.26912534e-01
5.44709027e-01 5.13932467e-01 1.16862528e-01 4.68111545e-01
-4.08381730e-01 3.36347938e-01 -1.27131939e-01 4.31047156e-02
-4.63760272e-02 1.15272149e-01 -7.44081557e-01 7.20274985e-01
1.49303332e-01 1.00804377e+00 -7.51462042e-01 -2.54081070e-01
1.52092114e-01 4.87834841e-01 -4.27353948e-01 5.53821266e-01
4.30660725e-01 7.06356883e-01 -5.26692986e-01 7.07937062e-01
3.63448650e-01 1.51356488e-01 2.29241848e-01 -3.64765674e-01
5.90780854e-01 8.20740126e-04 -1.12812984e+00 1.03822958e+00
-9.52360332e-02 3.89532536e-01 -9.86352935e-02 -8.55460167e-01
7.80379653e-01 8.16945791e-01 6.64172113e-01 -1.99251562e-01
4.71340984e-01 1.80178396e-02 2.36652613e-01 -6.17590904e-01
-2.72487905e-02 -2.32151344e-01 -6.82741463e-01 7.30094790e-01
-1.50846586e-01 -3.48663718e-01 -5.10647893e-01 1.47688583e-01
1.82733274e+00 -1.15031309e-01 5.09118319e-01 -2.38762088e-02
3.65013540e-01 -2.11156443e-01 2.95640290e-01 1.38654196e+00
-6.86521053e-01 6.66384757e-01 4.95533615e-01 -7.26361334e-01
-6.16185129e-01 -1.40023923e+00 1.71691254e-02 2.12690756e-01
1.72248501e-02 -2.37793446e-01 -9.89709258e-01 -1.00286555e+00
-7.51488283e-02 5.43857872e-01 -5.17581224e-01 -5.02376497e-01
-8.22931886e-01 -6.56482816e-01 2.01849461e+00 7.25691319e-01
2.37342685e-01 -1.19172657e+00 -1.17305791e+00 2.92217165e-01
1.02999672e-01 -1.26383758e+00 -1.23187698e-01 2.98165381e-01
-5.19284844e-01 -1.31907976e+00 -2.45770276e-01 -2.78083980e-01
4.96706426e-01 -6.03501678e-01 1.16106474e+00 1.63221553e-01
-8.49169433e-01 7.45600343e-01 -4.09569412e-01 -9.07503188e-01
-1.06938660e+00 -2.56131887e-01 5.45294881e-01 -2.50903845e-01
-2.37867516e-03 -8.00530195e-01 -7.68563867e-01 2.88053066e-01
-7.40643561e-01 -7.17903554e-01 2.62622386e-01 7.27965236e-01
8.05413723e-02 1.30865738e-01 6.25385821e-01 -9.84009266e-01
1.00393260e+00 -3.30916733e-01 -3.63108963e-01 -1.97611913e-01
-3.90970498e-01 -3.92467558e-01 1.06867552e+00 -7.70328999e-01
-2.34510824e-01 1.24388970e-01 -5.20363271e-01 -4.56451505e-01
-4.35896784e-01 1.01227589e-01 -4.62320410e-02 -5.38255572e-01
1.34922767e+00 -2.41327994e-02 -6.01115152e-02 1.95692524e-01
1.22150362e-01 6.38063073e-01 9.42828298e-01 -5.78664184e-01
1.21296799e+00 3.76631975e-01 9.44234505e-02 -4.88945037e-01
-3.31447810e-01 2.14338928e-01 -3.78736377e-01 -3.66860211e-01
6.24199390e-01 -5.82663059e-01 -1.22631109e+00 5.56977689e-01
-9.06489432e-01 -4.81557071e-01 -4.56625640e-01 4.59840447e-01
-5.91489077e-01 4.42419767e-01 -8.69139254e-01 -8.12517762e-01
-7.55511403e-01 -1.07580948e+00 7.39021420e-01 -1.58816800e-01
-6.67054892e-01 -8.69604349e-01 1.27035782e-01 3.22812110e-01
5.70260167e-01 1.21057832e+00 1.70619920e-01 -1.00958252e+00
-2.92133868e-01 -6.94071949e-01 7.04753697e-01 4.17785406e-01
6.31937459e-02 -1.43111467e-01 -1.24454129e+00 -5.73065698e-01
3.93618643e-01 -3.96428227e-01 -2.06319943e-01 9.54365581e-02
1.00333989e+00 -4.11859006e-01 -2.28579298e-01 7.43722558e-01
1.01291859e+00 3.54302227e-01 8.92461300e-01 1.42337352e-01
6.35809541e-01 -7.64099956e-02 4.65389609e-01 5.81312418e-01
-3.98437113e-01 3.38466048e-01 5.42784393e-01 -3.95878226e-01
2.74985522e-01 1.37736633e-01 3.24579537e-01 2.93257475e-01
-4.91542667e-02 -2.47003824e-01 -8.60508084e-01 4.59812135e-02
-1.00426304e+00 -9.79584813e-01 3.94664854e-02 2.10589147e+00
7.01828659e-01 6.29288256e-01 1.12173341e-01 7.25832641e-01
4.66370821e-01 -5.13002612e-02 -6.09022677e-01 -5.98686159e-01
2.70791829e-01 7.54618406e-01 6.24001980e-01 4.10577327e-01
-1.25163352e+00 4.66740787e-01 7.66360807e+00 2.45747969e-01
-1.28568137e+00 -1.33618549e-01 4.73194510e-01 -1.74086660e-01
2.27513924e-01 -3.63609582e-01 -2.70208627e-01 4.72752392e-01
1.22609675e+00 -1.18714198e-02 2.93385714e-01 8.10165584e-01
8.04211199e-02 5.96090674e-01 -1.31490612e+00 8.67769420e-01
2.51211852e-01 -1.08471036e+00 -3.15865189e-01 -1.73338521e-02
1.32382348e-01 -1.41325980e-01 6.65680096e-02 3.10326666e-01
3.53660882e-01 -1.25151098e+00 5.71686089e-01 2.46251523e-01
9.83247638e-01 -9.02845621e-01 8.72159064e-01 3.00934464e-01
-7.67244101e-01 8.56896937e-02 1.81413129e-01 -4.47363183e-02
2.35082954e-01 3.19385618e-01 -1.13270247e+00 4.59045857e-01
3.03856224e-01 -1.07226536e-01 -3.85555059e-01 4.90221739e-01
-2.81773269e-01 1.20508552e+00 -2.76838630e-01 2.18338385e-01
-1.84343725e-01 5.94588101e-01 7.92615891e-01 1.14293003e+00
-5.38404174e-02 3.34653556e-01 2.00280249e-01 6.15114808e-01
1.34149626e-01 -4.60799336e-01 -8.18794072e-01 1.23116121e-01
5.68026841e-01 1.16963220e+00 -4.71577883e-01 -1.46022990e-01
3.43699515e-01 1.01384008e+00 -2.90048569e-01 8.18967745e-02
-1.41372776e+00 -9.00138497e-01 5.70329785e-01 4.16075513e-02
-1.71645775e-01 2.65338197e-02 -4.58670467e-01 -8.97444844e-01
1.68181166e-01 -1.50177658e+00 3.99426162e-01 -5.69380164e-01
-1.21870959e+00 9.07257259e-01 -2.07981482e-01 -1.46218252e+00
-6.95713043e-01 -3.60942155e-01 -7.51749337e-01 8.07236254e-01
-7.02075958e-01 -9.61952984e-01 -3.51664305e-01 8.03761005e-01
-2.13110913e-02 -4.07881498e-01 1.39699352e+00 2.22761884e-01
-3.78764719e-01 1.24960113e+00 -4.94981498e-01 6.61327064e-01
5.75787187e-01 -1.16422856e+00 9.19828117e-01 1.05973399e+00
-1.42842934e-01 8.62566292e-01 1.12665880e+00 -4.79024798e-01
-1.44584012e+00 -1.05740285e+00 1.76512852e-01 -8.21491539e-01
6.41877711e-01 -3.71999651e-01 -6.11124396e-01 6.12267733e-01
3.00903842e-02 4.28903103e-01 9.58944440e-01 -4.16820645e-01
-2.55070329e-01 -2.00649463e-02 -1.63602364e+00 7.56966710e-01
6.35214090e-01 -7.09360361e-01 -5.55427909e-01 4.06222939e-01
5.25884092e-01 -9.51991856e-01 -1.37440467e+00 6.71258628e-01
7.09591866e-01 -8.12278986e-01 1.34642017e+00 -6.38469398e-01
7.52952620e-02 -3.16373289e-01 3.10372692e-02 -1.16657269e+00
4.09308933e-02 -1.45209038e+00 -2.98860729e-01 9.30580378e-01
2.82621205e-01 -9.06162798e-01 7.83886671e-01 3.75187457e-01
1.60871565e-01 -6.75849020e-01 -8.08756590e-01 -7.74046123e-01
7.69244060e-02 -6.65157437e-01 6.04552448e-01 1.13612258e+00
2.49456465e-01 -8.65410119e-02 -6.37910903e-01 8.34310532e-01
6.94945693e-01 -5.70203602e-01 9.93395746e-01 -6.16746664e-01
-1.08199680e+00 -3.65896076e-02 -1.01559162e+00 -2.45202288e-01
-2.65131801e-01 -4.17751849e-01 -2.60192491e-02 -4.59769487e-01
-4.12821978e-01 -3.49240720e-01 -3.07146579e-01 4.18470144e-01
-4.37998414e-01 6.58667564e-01 1.90754235e-01 -7.16114044e-03
-2.20673949e-01 -3.84614676e-01 4.93447900e-01 -1.51969865e-01
-1.54442772e-01 1.82980597e-01 -6.54999375e-01 6.63477540e-01
1.28381920e+00 -6.31755888e-01 -5.08170187e-01 1.92763433e-01
-1.51474178e-01 4.47219402e-01 5.70199311e-01 -1.42835283e+00
6.08437024e-02 2.16582909e-01 5.47169030e-01 -4.96705547e-02
2.13153094e-01 -8.70544374e-01 3.85885090e-01 1.00476694e+00
-3.88532907e-01 3.30374986e-02 2.73443848e-01 1.95960656e-01
1.19314492e-01 1.08298613e-02 7.66519189e-01 9.59408730e-02
2.81450272e-01 8.47058520e-02 -6.63845658e-01 2.22304955e-01
1.42298496e+00 -1.68176353e-01 -2.21806869e-01 -4.29450423e-01
-1.00191772e+00 -8.51122662e-02 2.12822989e-01 7.25222826e-02
6.38425350e-01 -9.91721094e-01 -6.90436900e-01 6.14627123e-01
2.97983624e-02 -5.16803563e-01 7.05692694e-02 5.84205091e-01
-8.34601462e-01 -1.22624226e-01 -2.05860913e-01 -5.91115654e-01
-1.66720772e+00 7.30554223e-01 8.87688577e-01 -3.13407034e-01
-7.93484271e-01 4.45029646e-01 -2.96326518e-01 -1.88322201e-01
4.65086013e-01 -4.58334759e-02 1.24411611e-02 -6.66216850e-01
5.68448842e-01 2.77115464e-01 2.18600422e-01 -6.21212982e-02
-4.58663195e-01 2.82052875e-01 2.65336990e-01 -1.74993098e-01
6.16873264e-01 3.82071555e-01 4.06648546e-01 4.57120612e-02
9.66838539e-01 3.30138147e-01 -7.71479726e-01 4.54348117e-01
-4.99223113e-01 -1.60705194e-01 -5.13770878e-01 -9.64553893e-01
-1.04473305e+00 5.88511229e-01 8.12178135e-01 5.66624999e-01
1.23566782e+00 -4.56723243e-01 8.88677716e-01 4.87787515e-01
5.91544330e-01 -3.69606286e-01 1.22363284e-01 6.36850810e-03
6.42581642e-01 -5.89715660e-01 -7.28893206e-02 -3.01839232e-01
-6.69518232e-01 1.04315913e+00 2.00972438e-01 -4.33811963e-01
7.35670865e-01 9.54017460e-01 7.71291137e-01 -2.37479776e-01
-2.24718064e-01 6.88171685e-01 -2.59441108e-01 1.13969576e+00
1.68831512e-01 3.49452525e-01 1.97241083e-02 5.52119017e-01
-5.86941302e-01 3.80100794e-02 8.68112028e-01 1.25914109e+00
1.99525535e-01 -1.03415561e+00 -7.98056304e-01 1.53543472e-01
-1.13297200e+00 -9.92326997e-03 -4.52985495e-01 6.63785756e-01
-2.01585074e-03 1.14938128e+00 -5.40701151e-01 -6.87858522e-01
6.69697106e-01 -4.69757952e-02 3.73484761e-01 -1.18305609e-01
-1.26716197e+00 -2.59567708e-01 2.01184884e-01 -7.10696518e-01
4.82623912e-02 -4.31550354e-01 -1.10690808e+00 -1.87294304e-01
-6.31210115e-03 4.90210131e-02 4.77080762e-01 5.91108263e-01
2.40582824e-01 1.08387482e+00 1.10158205e+00 -6.59211278e-01
-1.12104785e+00 -7.48760045e-01 -5.40583432e-01 5.24855733e-01
5.20130754e-01 5.47725940e-03 -7.25243032e-01 2.77500063e-01] | [14.353813171386719, 3.113445997238159] |
5c2a5c4c-eafd-43d5-98aa-2060238e4c6c | incorporating-subjectivity-into-gendered | null | null | https://aclanthology.org/2022.gebnlp-1.28 | https://aclanthology.org/2022.gebnlp-1.28.pdf | Incorporating Subjectivity into Gendered Ambiguous Pronoun (GAP) Resolution using Style Transfer | The GAP dataset is a Wikipedia-based evaluation dataset for gender bias detection in coreference resolution, containing mostly objective sentences. Since subjectivity is ubiquitous in our daily texts, it becomes necessary to evaluate models for both subjective and objective instances. In this work, we present a new evaluation dataset for gender bias in coreference resolution, GAP-Subjective, which increases the coverage of the original GAP dataset by including subjective sentences. We outline the methodology used to create this dataset. Firstly, we detect objective sentences and transfer them into their subjective variants using a sequence-to-sequence model. Secondly, we outline the thresholding techniques based on fluency and content preservation to maintain the quality of the sentences. Thirdly, we perform automated and human-based analysis of the style transfer and infer that the transferred sentences are of high quality. Finally, we benchmark both GAP and GAP-Subjective datasets using a BERT-based model and analyze its predictive performance and gender bias. | ['Tanvi Dadu', 'Kartikey Pant'] | null | null | null | null | naacl-gebnlp-2022-7 | ['gender-bias-detection', 'gender-bias-detection'] | ['miscellaneous', 'natural-language-processing'] | [ 2.37613946e-01 4.65414733e-01 -2.90114641e-01 -7.31133342e-01
-9.37162042e-01 -7.90328681e-01 6.55608177e-01 2.48235270e-01
-5.44039488e-01 1.07386136e+00 7.08374679e-01 3.03559434e-02
-2.37293139e-01 -5.45846581e-01 -2.59498745e-01 -4.42490608e-01
4.18027520e-01 9.10294116e-01 1.07903577e-01 -6.44294798e-01
5.63717663e-01 7.53307641e-02 -1.75298846e+00 4.09305871e-01
1.22541726e+00 4.02249038e-01 1.35334119e-01 6.70288622e-01
-1.59261942e-01 6.08374238e-01 -7.80948699e-01 -1.06200063e+00
-2.73285687e-01 -7.19495118e-01 -1.20345080e+00 -2.59567469e-01
7.18518615e-01 2.14460328e-01 1.64297208e-01 1.28252256e+00
9.92116213e-01 1.03939019e-01 6.98517799e-01 -9.08488750e-01
-1.06175512e-01 9.19370592e-01 -2.37093568e-01 1.82379976e-01
8.08703125e-01 -3.24047506e-01 1.04182851e+00 -6.13630474e-01
1.23280752e+00 1.56213737e+00 5.26291430e-01 9.88776445e-01
-1.35482252e+00 -6.43755734e-01 -1.90767258e-01 2.53352523e-01
-9.47784901e-01 -8.67552817e-01 6.96794450e-01 -4.84845191e-01
5.89322448e-01 7.64862537e-01 5.28747976e-01 1.56591380e+00
-1.55605763e-01 6.38453960e-01 1.27904570e+00 -6.28859162e-01
5.28110489e-02 3.80386144e-01 3.72168720e-01 5.27741611e-01
1.72646135e-01 1.08336486e-01 -9.74735141e-01 -1.03234865e-01
4.72209454e-02 -9.24745262e-01 -3.82026076e-01 -2.17796400e-01
-9.90183175e-01 8.22048128e-01 -3.51174891e-01 5.52148104e-01
9.48161539e-03 -5.91864169e-01 8.54094803e-01 4.19954419e-01
4.63980794e-01 6.12852395e-01 -3.09520334e-01 -5.75711906e-01
-1.25753725e+00 3.91279340e-01 9.96813715e-01 8.68056715e-01
2.04519525e-01 -4.73890275e-01 -6.17822468e-01 1.43158507e+00
3.18671502e-02 5.97242296e-01 5.55359364e-01 -1.42724240e+00
6.03397608e-01 4.78829831e-01 1.07540093e-01 -1.14801610e+00
-6.43920481e-01 -2.57962883e-01 -6.46498561e-01 -1.96286097e-01
6.96962297e-01 2.23541021e-01 -1.34616673e-01 2.09065008e+00
-7.80998217e-03 -5.87308288e-01 5.10995463e-03 7.91101038e-01
1.24848723e+00 6.42890707e-02 1.41474098e-01 -8.03297162e-01
1.50413167e+00 -5.92682302e-01 -9.98188138e-01 2.73383819e-02
6.77318931e-01 -8.18021119e-01 1.11903441e+00 4.49798495e-01
-1.21564651e+00 -4.56962675e-01 -8.02786052e-01 -4.74180356e-02
-1.65862620e-01 2.09192529e-01 2.07198456e-01 1.24774659e+00
-8.40553105e-01 6.94460928e-01 -2.75243461e-01 -6.35402858e-01
4.05909084e-02 3.76410186e-01 -2.48432100e-01 3.88010591e-01
-1.39808738e+00 8.75301123e-01 3.97582918e-01 -3.72266829e-01
-2.30394289e-01 -6.16964877e-01 -7.91633904e-01 -2.85314441e-01
1.66860789e-01 -5.93607247e-01 1.29057169e+00 -1.03438199e+00
-1.52737284e+00 1.68692958e+00 -4.30875063e-01 -6.71520531e-02
8.27693403e-01 2.77064294e-02 -5.87508142e-01 1.51869962e-02
3.64222765e-01 6.02814257e-01 3.91836911e-01 -1.26918185e+00
-8.11147273e-01 -5.39285243e-01 -7.68854618e-02 2.23644480e-01
-3.83914769e-01 5.11488974e-01 -4.26658839e-01 -6.58640504e-01
2.20766515e-02 -8.22745323e-01 4.11623746e-01 -6.90026045e-01
-3.83642137e-01 -4.70559269e-01 1.96713597e-01 -9.39922988e-01
1.51889074e+00 -2.10077333e+00 4.97718900e-01 2.05941707e-01
2.21827030e-01 7.29536712e-02 7.04046488e-02 1.22790068e-01
-4.86756116e-03 7.94354528e-02 -2.93425769e-01 -6.25641823e-01
2.99287021e-01 1.03108622e-01 -1.87768281e-01 1.33487135e-01
-2.80656397e-01 4.53711569e-01 -1.14530516e+00 -1.06244195e+00
-5.82389235e-02 -3.24144997e-02 -7.54627228e-01 2.30913743e-01
2.73792576e-02 5.19787073e-01 5.19561470e-02 4.74079043e-01
7.07858443e-01 6.24925733e-01 7.39102662e-01 -3.51460695e-01
-3.39890987e-01 4.79481101e-01 -8.75689149e-01 1.79557741e+00
-5.16642094e-01 7.45010138e-01 2.59602815e-01 -7.49861956e-01
1.13741302e+00 1.61385030e-01 1.97819442e-01 -9.71265733e-01
2.88046390e-01 4.82628703e-01 -1.33297607e-01 -6.18056178e-01
1.20475483e+00 -3.47174436e-01 -4.93802994e-01 2.22717062e-01
3.62206161e-01 -1.64798632e-01 9.50671256e-01 1.09658942e-01
5.94029844e-01 7.38139898e-02 7.37545714e-02 -5.88585377e-01
8.64442706e-01 -1.76575601e-01 9.30779278e-01 6.38404667e-01
-3.26470882e-01 7.93026745e-01 8.93198192e-01 9.32186246e-02
-7.83296764e-01 -9.87018168e-01 -3.47651035e-01 1.42591166e+00
1.18945325e-04 -5.68161547e-01 -1.01105785e+00 -8.06966722e-01
-4.02463168e-01 1.04099083e+00 -5.89723229e-01 -9.83538181e-02
-5.43962777e-01 -9.84846950e-01 8.09709847e-01 2.31254473e-01
5.49142420e-01 -1.06841362e+00 -4.27634656e-01 3.84471286e-03
-9.78574395e-01 -8.89671743e-01 -4.01036173e-01 -3.22285555e-02
-6.82727873e-01 -1.26973438e+00 -3.98896933e-01 -7.61819184e-01
2.13842317e-01 -4.72888798e-01 1.60756564e+00 1.02339990e-01
-3.52582149e-02 1.44663095e-01 -3.87101918e-01 -2.90174991e-01
-8.29301476e-01 2.92159796e-01 1.65514857e-01 -3.18543285e-01
6.18877172e-01 -3.37553382e-01 -2.89084554e-01 2.77854860e-01
-4.30182010e-01 -1.61926866e-01 1.70866564e-01 9.29834425e-01
9.64650959e-02 -2.47690797e-01 3.95112067e-01 -1.06049776e+00
8.62691998e-01 -3.69182117e-02 -1.44900724e-01 1.80788189e-01
-6.80470288e-01 -7.54902363e-02 2.48473495e-01 -8.17324370e-02
-1.44458771e+00 -3.44279766e-01 -3.60261291e-01 2.62940705e-01
-1.08338632e-01 1.56591594e-01 -4.37526792e-01 4.37971324e-01
7.19225466e-01 -3.02325636e-02 -8.56105089e-02 -5.71747780e-01
7.75730163e-02 9.87217963e-01 8.92218292e-01 -9.45374668e-01
2.70545512e-01 9.55509841e-02 -2.32876211e-01 -8.66269171e-01
-9.89706516e-01 -4.25386548e-01 -8.50153148e-01 -4.67637628e-01
8.15165758e-01 -5.31650126e-01 -8.12907577e-01 3.13514322e-01
-1.08034456e+00 -1.67796254e-01 -9.06575993e-02 4.48214293e-01
-8.60376179e-01 4.79370922e-01 -4.38034296e-01 -8.26127648e-01
-4.03481215e-01 -8.64272654e-01 1.02624691e+00 1.48527876e-01
-1.16593170e+00 -1.07947695e+00 3.88623118e-01 6.72289729e-01
-1.10819593e-01 -9.06343199e-03 9.91139710e-01 -6.58108354e-01
5.61799407e-01 3.40403080e-01 -9.25228298e-02 6.25318708e-03
-5.35130948e-02 5.15945777e-02 -8.44347775e-01 -2.45415986e-01
-1.63019165e-01 -2.81069845e-01 9.08938229e-01 2.09095180e-01
7.86753416e-01 -9.42485109e-02 -2.23913565e-01 4.20161903e-01
9.72573996e-01 1.35256955e-02 6.36105835e-01 5.86205423e-01
3.07189465e-01 1.08442104e+00 8.46953034e-01 1.58463106e-01
4.42007750e-01 8.96618724e-01 -2.47734591e-01 1.06608637e-01
-1.22937679e-01 -2.63139904e-01 2.69197375e-01 9.87340868e-01
-4.26894009e-01 -1.33544803e-01 -7.09332824e-01 5.77923954e-01
-1.73799026e+00 -1.32747877e+00 -3.16397369e-01 2.22042894e+00
1.21313012e+00 2.35546350e-01 4.22585785e-01 5.40915132e-01
9.67135489e-01 -1.31102116e-03 1.38517812e-01 -8.11894059e-01
-4.65186507e-01 -5.14111891e-02 8.19093436e-02 5.86688221e-01
-1.01101780e+00 7.50061572e-01 6.69480991e+00 7.08624125e-01
-8.41663659e-01 3.09122633e-02 3.94634038e-01 -1.96341038e-01
-3.25394034e-01 -4.08473015e-01 -7.89399564e-01 5.28718829e-01
9.42033172e-01 -1.38091743e-01 1.77265391e-01 3.77087742e-01
3.29627991e-01 -1.61591053e-01 -1.08637047e+00 9.16310072e-01
3.33414793e-01 -7.18141735e-01 -1.47131234e-01 -2.12668046e-01
4.69683379e-01 -6.01636469e-01 -6.07101992e-02 4.81851965e-01
8.27920064e-02 -7.86372125e-01 1.04793274e+00 5.71228862e-01
8.42209756e-01 -8.68644297e-01 1.07809532e+00 1.33920699e-01
-5.13651669e-01 -1.33589534e-02 -3.07184905e-01 1.95692152e-01
7.71554261e-02 5.90101480e-01 -3.97642076e-01 7.35729992e-01
8.75674367e-01 5.43241143e-01 -7.75355518e-01 5.99223197e-01
-2.25719392e-01 6.19644284e-01 1.21080831e-01 -6.14227727e-02
-4.31742311e-01 -3.25537831e-01 8.68340313e-01 1.53378785e+00
2.22350717e-01 -6.84928000e-02 -3.35577667e-01 7.01448381e-01
-3.40809934e-02 4.87663925e-01 -2.19959602e-01 1.01195715e-01
7.01719820e-01 1.19204009e+00 -6.42610013e-01 -1.75979093e-01
-2.43730813e-01 9.00741339e-01 3.33556950e-01 -7.00471476e-02
-4.17307854e-01 -4.74058211e-01 4.85410690e-01 -2.15736955e-01
-3.67198475e-02 3.47038478e-01 -5.87813497e-01 -9.98006642e-01
-1.85565397e-01 -1.32636940e+00 7.60163009e-01 -4.68884736e-01
-1.03028142e+00 4.50826198e-01 6.51150420e-02 -9.97380018e-01
-3.15572798e-01 -2.87736595e-01 -3.69736761e-01 6.61185145e-01
-8.82596672e-01 -8.34750950e-01 -2.18814149e-01 2.27290094e-01
5.26286483e-01 -3.77778471e-01 9.26520705e-01 2.83002555e-01
-6.34575605e-01 9.49523628e-01 1.07251525e-01 9.61015522e-02
1.30073357e+00 -1.38773894e+00 -1.87002599e-01 3.86571646e-01
-2.79738098e-01 6.83664978e-01 1.29265213e+00 -5.71809053e-01
-6.77229524e-01 -8.25170219e-01 1.60593247e+00 -6.13365710e-01
3.26785833e-01 -3.49453241e-01 -4.78668213e-01 3.39463115e-01
2.84817487e-01 -1.04018223e+00 8.15048695e-01 5.84743977e-01
-2.94423550e-01 1.13442335e-02 -1.42611885e+00 5.30621946e-01
1.37395883e+00 -4.90540355e-01 -1.17640436e+00 -1.59589037e-01
2.33562931e-01 -4.85614628e-01 -9.39321339e-01 3.95680308e-01
6.11280918e-01 -1.37920630e+00 8.05432677e-01 -3.86831254e-01
7.34550536e-01 1.06254414e-01 -2.31562585e-01 -1.54274285e+00
-4.35187757e-01 -5.27651072e-01 2.55632967e-01 1.82842064e+00
4.99237716e-01 -1.72156826e-01 5.22289157e-01 3.66677910e-01
-4.29093055e-02 -8.22743922e-02 -1.01058531e+00 -5.66438973e-01
2.79207677e-01 -1.13756202e-01 6.56451583e-01 1.07347810e+00
5.34080267e-01 5.96166670e-01 -1.99171647e-01 -4.23533797e-01
7.45049357e-01 2.65994310e-01 5.09473145e-01 -1.46947527e+00
3.16065201e-03 -7.13867486e-01 -1.48052767e-01 -3.63383025e-01
5.58948398e-01 -1.07067537e+00 3.76894549e-02 -1.05259001e+00
6.52524948e-01 1.56718075e-01 1.05648592e-01 -5.85164689e-02
-3.08946967e-01 4.58135903e-01 3.96718048e-02 -5.95546216e-02
-6.63502693e-01 4.91788566e-01 9.06168282e-01 -2.12536320e-01
-3.04860443e-01 -1.70405149e-01 -8.20458412e-01 6.67391121e-01
8.88967514e-01 -3.69596243e-01 -6.54961988e-02 9.50212851e-02
5.28393924e-01 -4.86308113e-02 -3.42035550e-03 -5.65284312e-01
-2.24475801e-01 -1.41479135e-01 1.42722592e-01 -6.65126979e-01
6.81183301e-04 -2.07424611e-01 -2.93965906e-01 3.22126925e-01
-5.39358735e-01 -2.10133493e-01 -3.41207124e-02 9.93334055e-02
-2.49255061e-01 -4.27893817e-01 8.29789877e-01 1.91343576e-02
-4.13155437e-01 -3.22662622e-01 -4.99609500e-01 5.07783592e-01
3.80003601e-01 2.29743142e-02 -3.14944446e-01 -3.05629760e-01
-8.32174301e-01 2.61108875e-01 5.92379868e-01 6.40518188e-01
1.05666913e-01 -1.29210007e+00 -9.60732222e-01 -1.28481269e-01
3.92072409e-01 -7.09710836e-01 1.23148866e-01 1.01594031e+00
-2.17834190e-01 4.46377784e-01 -4.14699584e-01 -4.09494013e-01
-1.97849786e+00 3.73738676e-01 2.96503544e-01 -1.59561113e-01
-2.26432942e-02 6.39299035e-01 -1.12595260e-01 -8.63368928e-01
2.54868835e-01 2.18225211e-01 -9.08382058e-01 6.57049596e-01
3.96678776e-01 7.88348138e-01 1.09487310e-01 -8.77068102e-01
-3.80231023e-01 4.20054972e-01 2.09247455e-01 -4.46756810e-01
1.20577073e+00 -5.68243921e-01 -5.85688591e-01 6.68659151e-01
1.04365158e+00 4.75880712e-01 -3.55564266e-01 7.38623068e-02
3.25457096e-01 -4.71630365e-01 -1.11659996e-01 -8.52554739e-01
-6.72811747e-01 4.64224607e-01 6.09927714e-01 1.22214228e-01
1.06266999e+00 1.78873225e-03 3.23937386e-01 1.98000968e-01
2.74215490e-01 -1.80241323e+00 -2.79687136e-01 7.45771766e-01
9.16777432e-01 -1.07509220e+00 -1.39177859e-01 -6.65626585e-01
-4.66838181e-01 1.00454330e+00 4.61437076e-01 5.32957375e-01
2.16763616e-01 -5.65521047e-02 2.74740338e-01 -1.75895080e-01
-6.55947506e-01 -9.32458565e-02 3.40467572e-01 7.62667000e-01
1.06372607e+00 1.84035972e-01 -1.37113690e+00 1.15312111e+00
-1.09759331e+00 -7.40695698e-03 4.57340866e-01 3.90547663e-01
-2.53993571e-01 -1.44840491e+00 -5.28250039e-01 3.05033833e-01
-6.11510456e-01 9.71450210e-02 -1.00660181e+00 4.88001496e-01
1.69940606e-01 1.33540285e+00 6.02760725e-02 -5.93180716e-01
5.01834273e-01 2.77950555e-01 8.18281651e-01 -2.62065351e-01
-7.02949107e-01 -2.49718845e-01 1.05542302e+00 -3.55222404e-01
-8.20965588e-01 -9.98506248e-01 -9.30096328e-01 -5.65978110e-01
-6.65536895e-02 4.39629644e-01 3.18030089e-01 8.21526289e-01
-1.04268111e-01 3.75753492e-01 6.38115048e-01 -4.51892585e-01
-2.64665395e-01 -1.32821584e+00 -6.90681279e-01 8.80026221e-01
-4.57448931e-03 -7.40184784e-01 -4.74591017e-01 4.01923060e-02] | [11.068286895751953, 9.8075590133667] |
05de5389-f5d6-4bba-9d92-01597a40ff18 | katildakat-at-semeval-2021-task-1-lexical | null | null | https://aclanthology.org/2021.semeval-1.91 | https://aclanthology.org/2021.semeval-1.91.pdf | katildakat at SemEval-2021 Task 1: Lexical Complexity Prediction of Single Words and Multi-Word Expressions in English | This paper describes systems submitted to Se- mEval 2021 Task 1: Lexical Complexity Prediction (LCP). We compare a linear and a non-linear regression models trained to work for both tracks of the task. We show that both systems are able to generalize better when supplied with information about complexities of single word and multi-word expression (MWE) targets simultaneously. This approach proved to be the most beneficial for multi-word expression targets. We also demonstrate that some hand-crafted features differ in their importance for the target types. | ['Katja Voskoboinik'] | 2021-08-01 | null | null | null | semeval-2021 | ['lexical-complexity-prediction'] | ['natural-language-processing'] | [ 8.28428864e-02 -2.88871288e-01 -3.43022108e-01 -5.77293277e-01
-1.03450453e+00 -6.54681206e-01 6.52376294e-01 2.72765338e-01
-8.33389640e-01 9.71805394e-01 2.08774731e-01 -5.21884501e-01
4.76409234e-02 -4.68772501e-01 -3.35152477e-01 -3.80335391e-01
-1.49012014e-01 4.76917773e-01 8.76672715e-02 -6.23504519e-01
4.50330883e-01 3.92478615e-01 -1.66529334e+00 8.30046713e-01
4.81748909e-01 7.78788269e-01 3.71332884e-01 1.01373935e+00
-2.58544952e-01 6.39673054e-01 -8.39055896e-01 -4.78715837e-01
1.07324049e-01 -2.67433643e-01 -8.48104358e-01 -3.30782622e-01
4.84861523e-01 5.37932217e-01 1.57688931e-02 9.00708318e-01
6.25433147e-01 -5.63403443e-02 1.19220519e+00 -9.88861918e-01
-1.54841095e-01 6.84279859e-01 -3.06762010e-01 7.69449234e-01
4.98147428e-01 -1.01294719e-01 1.41400933e+00 -1.05195785e+00
6.82161331e-01 1.32236290e+00 8.71259689e-01 5.62339485e-01
-1.13469207e+00 -5.89671016e-01 1.63353190e-01 4.67685908e-01
-1.42639792e+00 -6.18206620e-01 2.92297453e-01 -5.31075776e-01
1.89924598e+00 4.82080877e-01 3.05926263e-01 1.23099625e+00
5.45032084e-01 5.55239022e-01 1.51027894e+00 -8.28603745e-01
1.75255968e-03 3.15009981e-01 3.69648993e-01 5.84691525e-01
1.21679440e-01 2.14387760e-01 -5.79179227e-01 -1.31871909e-01
4.03993241e-02 -9.89644945e-01 -6.93728700e-02 3.25262696e-01
-9.15862143e-01 9.43557203e-01 -3.46861273e-01 8.09467375e-01
-6.68398365e-02 1.08159266e-01 8.18647563e-01 8.11562240e-01
7.86323965e-01 7.84299314e-01 -9.81304288e-01 -5.07371306e-01
-7.59968877e-01 3.59506816e-01 1.07497323e+00 1.03080106e+00
3.87498617e-01 3.44871245e-02 -3.94494265e-01 9.18822169e-01
2.81175468e-02 3.05998594e-01 7.84805477e-01 -2.44745284e-01
5.85052192e-01 3.12057674e-01 -4.14004214e-02 -5.15093625e-01
-9.94982123e-01 -3.59281838e-01 -2.99245775e-01 5.54757975e-02
4.17892069e-01 -1.51835129e-01 -7.34621584e-01 1.88579583e+00
-2.50500292e-01 -4.80293423e-01 -7.09474236e-02 3.86488110e-01
5.33283889e-01 8.59291077e-01 5.57516694e-01 -7.72739053e-01
1.23009431e+00 -9.03913081e-01 -9.22188580e-01 -3.42782676e-01
1.34679520e+00 -1.12311292e+00 1.24370825e+00 6.24820173e-01
-1.20919037e+00 -4.75753278e-01 -1.01635814e+00 1.53212234e-01
-7.84099460e-01 1.77643761e-01 4.45342749e-01 1.21951950e+00
-8.63442957e-01 5.72740197e-01 -3.95783752e-01 -1.69015631e-01
-1.62937507e-01 4.52917963e-01 -1.86568290e-01 3.44978988e-01
-1.49062705e+00 1.69150734e+00 3.35972786e-01 -4.14275497e-01
-5.72483718e-01 -6.48757398e-01 -6.90685391e-01 5.40456250e-02
1.24760114e-01 -2.33073503e-01 1.39204192e+00 -9.19094682e-01
-1.55091393e+00 1.02807653e+00 -8.29630718e-02 -2.76079625e-01
3.69053155e-01 -2.58935928e-01 -6.96744621e-01 -4.89819050e-01
-1.71932667e-01 3.16494763e-01 6.91811860e-01 -7.35032201e-01
-7.59119570e-01 -1.72736064e-01 -2.28223369e-01 3.13680470e-01
-4.03350294e-01 5.69981694e-01 -1.71169322e-02 -8.27218890e-01
-7.06842721e-01 -8.02781284e-01 -1.32591426e-02 -6.69795692e-01
-4.25653160e-01 -6.37231827e-01 3.12528759e-01 -7.91049540e-01
1.96336496e+00 -1.73954988e+00 5.01115680e-01 1.01605788e-01
-1.87260523e-01 2.54931629e-01 -3.75394732e-01 5.22481203e-01
-5.13657987e-01 4.13418919e-01 1.48238853e-01 -3.70854050e-01
2.67560303e-01 1.32553101e-01 1.63846035e-02 3.07724804e-01
3.58988911e-01 7.79331625e-01 -6.07980728e-01 -4.48306888e-01
-6.68416247e-02 9.88598168e-03 -3.29568803e-01 9.79287252e-02
-3.23061258e-01 -1.95861489e-01 -1.53937548e-01 5.67790627e-01
2.70033658e-01 1.87235713e-01 2.76821613e-01 -8.01870003e-02
-2.10616067e-01 5.52680671e-01 -9.08829033e-01 1.13502598e+00
-9.68498230e-01 6.52457893e-01 -2.28819713e-01 -8.54601204e-01
6.56155050e-01 2.30703831e-01 -7.19869509e-02 -6.94139123e-01
3.17634851e-01 3.35171074e-01 3.96081895e-01 -4.29072887e-01
4.22647953e-01 -5.04034877e-01 -3.95962566e-01 1.91307068e-01
2.45041564e-01 -2.50384033e-01 5.37023425e-01 -2.76560038e-01
1.25183785e+00 -3.33148390e-01 9.98891354e-01 -8.14436615e-01
8.15090358e-01 -1.63422137e-01 2.70973295e-01 5.95954359e-01
1.10544395e-02 1.43997639e-01 5.52328408e-01 -3.80347401e-01
-1.20675111e+00 -7.60808885e-01 -2.40670308e-01 1.81099761e+00
-3.15036863e-01 -7.83883512e-01 -5.97868621e-01 -6.93427086e-01
-2.97510415e-01 1.27330744e+00 -6.70795441e-01 -1.34931266e-01
-6.17521822e-01 -9.36882675e-01 7.96085358e-01 5.86206973e-01
-3.45325619e-01 -1.02008986e+00 -4.33932722e-01 2.25694969e-01
-3.25691253e-02 -1.09712958e+00 -5.94897449e-01 8.41442585e-01
-5.68214417e-01 -7.12945819e-01 -3.94138485e-01 -9.10472453e-01
1.99426204e-01 -4.47072923e-01 1.43109334e+00 1.11725383e-01
-5.58441758e-01 1.55722350e-01 -4.76617157e-01 -6.01900041e-01
-8.47865939e-01 5.92390478e-01 2.48265952e-01 -5.29992044e-01
5.43577373e-01 -2.30468735e-01 2.89174110e-01 4.20956552e-01
-5.74376047e-01 -2.73449093e-01 6.45839691e-01 1.01211274e+00
2.42302731e-01 -7.26743117e-02 3.53010118e-01 -8.98613513e-01
1.13032842e+00 -3.66507351e-01 -3.86153281e-01 5.87032855e-01
-8.44564974e-01 3.93552065e-01 5.95386565e-01 -7.95860648e-01
-7.51150727e-01 3.50412935e-01 -5.01849949e-01 5.45741096e-02
-6.77122409e-03 5.02146423e-01 -3.28209102e-01 -9.00822729e-02
9.19378579e-01 1.29914403e-01 -3.46636623e-01 -4.84011233e-01
3.13850731e-01 6.50215566e-01 -1.27352336e-02 -6.94608271e-01
5.02030432e-01 -4.80498016e-01 -6.99072033e-02 -9.97985661e-01
-8.37753296e-01 -5.53286433e-01 -5.78421175e-01 -6.28957525e-02
7.53527522e-01 -7.02805042e-01 -3.27098846e-01 4.34089392e-01
-1.42636263e+00 -3.09106827e-01 -8.91212225e-02 1.95650414e-01
-6.47432685e-01 2.73664087e-01 -5.43542027e-01 -8.81451309e-01
-2.63312072e-01 -9.64775741e-01 1.00290072e+00 -2.42730618e-01
-6.21516764e-01 -1.16161704e+00 3.06890875e-01 -2.54147321e-01
4.60404873e-01 -1.61200121e-01 1.41638863e+00 -1.13067937e+00
3.93403769e-01 -2.10904613e-01 6.28299862e-02 3.42573673e-01
-1.11246370e-01 5.78112192e-02 -8.14660251e-01 -1.58192948e-01
-8.89247097e-03 -6.33132279e-01 8.02825391e-01 2.14756042e-01
9.29738879e-01 -3.36654603e-01 -1.86213166e-01 2.97413826e-01
1.27459407e+00 -4.68622670e-02 4.90805835e-01 3.50818485e-01
1.60835311e-01 7.22357273e-01 6.59010231e-01 3.48520905e-01
2.96064578e-02 1.15713251e+00 2.21122969e-02 2.73526132e-01
-3.34025063e-02 8.41024891e-02 7.67631233e-01 1.14800251e+00
-4.95913718e-03 -5.35008013e-01 -1.11830485e+00 3.88218224e-01
-1.42429411e+00 -6.83750927e-01 -3.97465974e-01 1.75835061e+00
1.10808301e+00 4.63673592e-01 3.44354928e-01 2.93650001e-01
6.28269196e-01 -9.71285719e-03 -4.53341529e-02 -1.07720029e+00
-4.97434676e-01 6.36760652e-01 6.86524928e-01 8.17808330e-01
-9.00490761e-01 1.07866502e+00 7.92184973e+00 1.62275994e+00
-6.93995595e-01 2.95652598e-01 6.49062276e-01 -6.30433336e-02
-2.31779546e-01 -3.36724758e-01 -1.19060981e+00 3.03761005e-01
1.60587716e+00 -4.21083212e-01 3.43168110e-01 8.68677557e-01
-2.06066698e-01 -1.32088035e-01 -1.20556796e+00 8.10113311e-01
1.44809976e-01 -8.40436816e-01 2.66559459e-02 -8.47068876e-02
4.64647979e-01 -1.16924472e-01 -6.06554979e-03 8.13973308e-01
-4.84522106e-03 -1.21023369e+00 6.88069522e-01 2.42485985e-01
8.26515496e-01 -8.75495315e-01 9.11716282e-01 4.17479038e-01
-1.16431451e+00 -2.71033067e-02 -4.71555144e-01 -3.71078819e-01
-6.78403024e-03 5.19440114e-01 -6.36889458e-01 1.86763659e-01
1.78043604e-01 2.39214033e-01 -7.34908164e-01 7.46868312e-01
-1.88552141e-01 6.29536748e-01 -3.30893546e-01 -6.60341561e-01
1.45120546e-01 3.13825458e-01 5.15429139e-01 1.94449401e+00
2.93610096e-01 -7.82318637e-02 -8.95710140e-02 3.69803637e-01
2.42347151e-01 7.95477509e-01 -3.83931428e-01 -2.97708333e-01
3.57404947e-01 1.15836072e+00 -6.16110265e-01 -3.14548343e-01
-3.96093160e-01 9.03162301e-01 4.69126731e-01 -1.88721105e-01
-9.09181476e-01 -4.31357622e-01 6.29399419e-01 -8.32255557e-02
2.50533164e-01 -1.90178663e-01 -4.26609308e-01 -1.00258148e+00
-3.50716025e-01 -9.85795498e-01 4.74115342e-01 -4.36340004e-01
-1.62526107e+00 9.29734528e-01 1.49278596e-01 -9.35831249e-01
-4.07584935e-01 -1.37762940e+00 -4.55952168e-01 1.03418469e+00
-1.34990478e+00 -8.04311574e-01 1.99354723e-01 3.57104093e-01
8.68367374e-01 -4.07974094e-01 1.25286281e+00 2.53262967e-01
-3.32472622e-01 9.46640968e-01 -3.37250754e-02 -4.38762665e-01
7.37822950e-01 -1.51478577e+00 5.53411901e-01 1.99022725e-01
3.17477286e-01 2.97207981e-01 1.07271099e+00 -5.25025547e-01
-1.01003826e+00 -9.19587016e-01 1.55133843e+00 -6.16474986e-01
9.27270234e-01 -4.93107259e-01 -6.00947797e-01 3.29159349e-01
6.69849366e-02 -3.75755996e-01 1.00972641e+00 5.29137433e-01
-5.67795455e-01 4.64692265e-01 -7.11727262e-01 3.68339688e-01
9.25425410e-01 -5.32194078e-01 -8.54297519e-01 6.27944410e-01
6.84271634e-01 -4.33734626e-01 -8.27508390e-01 4.00103509e-01
5.05855978e-01 -7.97154546e-01 7.59105265e-01 -1.08888710e+00
5.57309151e-01 2.82790333e-01 -5.26930213e-01 -1.54777336e+00
-3.34037513e-01 -3.73805195e-01 -5.62728122e-02 6.78016841e-01
9.62191999e-01 -2.07660064e-01 4.34793681e-01 2.19133779e-01
-1.53062016e-01 -8.88249934e-01 -1.18266702e+00 -1.24608803e+00
4.72591251e-01 -6.40191257e-01 -1.59158446e-02 8.29301715e-01
4.53051269e-01 1.00270021e+00 -4.00742054e-01 -1.80111289e-01
-1.06574528e-01 -2.90288925e-01 2.85668910e-01 -1.05616820e+00
-5.09855628e-01 -9.64528203e-01 -6.84012711e-01 -7.06763804e-01
3.65646034e-01 -9.66272354e-01 -1.50081128e-01 -7.05258191e-01
2.47931061e-03 5.08031584e-02 -2.90897399e-01 3.58602703e-01
-3.15740258e-01 5.06268665e-02 2.08072037e-01 -1.75228149e-01
-3.94942284e-01 3.69688928e-01 6.26728714e-01 3.70230563e-02
2.50396337e-02 2.16529638e-01 -4.96462882e-01 5.90426505e-01
8.48409653e-01 -6.97830737e-01 -1.86542436e-01 -1.41731367e-01
5.72098613e-01 -1.05274677e-01 -2.77624369e-01 -8.31882298e-01
-9.14040674e-03 -1.48667142e-01 1.15571029e-01 -5.25548279e-01
2.98692375e-01 -4.81102258e-01 -3.48805226e-02 5.35577595e-01
-6.27803147e-01 5.02114594e-01 6.79218411e-01 3.31725895e-01
6.48414865e-02 -7.68959999e-01 7.95290291e-01 -3.27675231e-02
-4.43231612e-01 -3.21637362e-01 -8.72968078e-01 2.66849756e-01
9.97836530e-01 3.47814523e-02 -1.13443121e-01 -3.21568698e-01
-9.51205373e-01 -9.25435871e-02 5.59635088e-02 4.98214155e-01
2.77082533e-01 -1.02324510e+00 -7.79634416e-01 -1.56559870e-01
4.19515878e-01 -1.10244632e+00 -8.45378116e-02 7.63204515e-01
-3.49221051e-01 6.95083201e-01 -2.94214725e-01 -1.26317382e-01
-1.68236542e+00 6.48762703e-01 3.18529695e-01 -8.07493567e-01
6.35157675e-02 1.06754315e+00 -2.83895433e-01 -3.81272733e-01
3.50372523e-01 -4.06757832e-01 -2.04320565e-01 3.50197941e-01
7.42877781e-01 3.97344440e-01 5.71726739e-01 -6.37577891e-01
-3.14613312e-01 5.53614140e-01 -3.47000599e-01 3.18810642e-02
1.16044223e+00 6.46984577e-02 -2.31696572e-02 8.63308012e-01
1.40228331e+00 2.55908109e-02 -3.13732535e-01 -3.23257029e-01
6.80782259e-01 -1.13531493e-01 1.84540749e-01 -1.14128757e+00
-4.37383980e-01 9.91454005e-01 6.22022212e-01 1.49705201e-01
9.71694469e-01 -1.20507546e-01 3.68369073e-01 7.06225038e-01
2.99984396e-01 -1.48193336e+00 -1.04625463e-01 8.27963948e-01
9.28809822e-01 -9.28456247e-01 1.76560745e-01 -7.59903908e-01
-5.93039215e-01 1.50149322e+00 5.53018332e-01 1.72972411e-01
6.14661217e-01 6.44618928e-01 -1.79131776e-01 -9.53215137e-02
-1.42064393e+00 -1.93741456e-01 3.70620638e-01 5.00626862e-01
7.61034429e-01 2.27620512e-01 -8.96471679e-01 8.88076127e-01
-3.53617460e-01 -6.07929647e-01 4.44799751e-01 8.53974760e-01
-5.74286938e-01 -1.25530457e+00 -1.99673727e-01 6.76480353e-01
-4.76582259e-01 -5.97814679e-01 -7.43199050e-01 1.05754805e+00
-8.81429315e-02 6.16444468e-01 -2.40458116e-01 -7.11754799e-01
5.06860375e-01 6.53804123e-01 7.08126724e-01 -8.47164035e-01
-8.89584064e-01 3.38450149e-02 7.65280545e-01 -4.91711915e-01
-5.21164201e-02 -6.18729711e-01 -7.63665617e-01 -1.61047071e-01
-5.90367854e-01 5.54347448e-02 8.33080769e-01 1.01462173e+00
-2.32301146e-01 5.61251044e-01 5.08341908e-01 -7.23580658e-01
-1.09266794e+00 -1.29165542e+00 -6.93249285e-01 2.43238658e-01
-2.37884015e-01 -7.28260994e-01 -5.62250078e-01 5.05447648e-02] | [10.631973266601562, 10.440585136413574] |
42bcd3ea-0394-4b66-9176-e2f2fa1494f1 | linear-mode-connectivity-in-multitask-and-1 | 2010.04495 | null | https://arxiv.org/abs/2010.04495v1 | https://arxiv.org/pdf/2010.04495v1.pdf | Linear Mode Connectivity in Multitask and Continual Learning | Continual (sequential) training and multitask (simultaneous) training are often attempting to solve the same overall objective: to find a solution that performs well on all considered tasks. The main difference is in the training regimes, where continual learning can only have access to one task at a time, which for neural networks typically leads to catastrophic forgetting. That is, the solution found for a subsequent task does not perform well on the previous ones anymore. However, the relationship between the different minima that the two training regimes arrive at is not well understood. What sets them apart? Is there a local structure that could explain the difference in performance achieved by the two different schemes? Motivated by recent work showing that different minima of the same task are typically connected by very simple curves of low error, we investigate whether multitask and continual solutions are similarly connected. We empirically find that indeed such connectivity can be reliably achieved and, more interestingly, it can be done by a linear path, conditioned on having the same initialization for both. We thoroughly analyze this observation and discuss its significance for the continual learning process. Furthermore, we exploit this finding to propose an effective algorithm that constrains the sequentially learned minima to behave as the multitask solution. We show that our method outperforms several state of the art continual learning algorithms on various vision benchmarks. | ['Hassan Ghasemzadeh', 'Razvan Pascanu', 'Dilan Gorur', 'Mehrdad Farajtabar', 'Seyed Iman Mirzadeh'] | 2020-10-09 | linear-mode-connectivity-in-multitask-and | https://openreview.net/forum?id=Fmg_fQYUejf | https://openreview.net/pdf?id=Fmg_fQYUejf | iclr-2021-1 | ['linear-mode-connectivity'] | ['knowledge-base'] | [ 9.70349684e-02 1.25045404e-01 1.36008129e-01 -5.56326285e-02
-3.24155509e-01 -4.24959809e-01 6.55676246e-01 4.95640874e-01
-6.93056941e-01 1.02640009e+00 -4.23400730e-01 -3.06374848e-01
-6.60786510e-01 -4.59817708e-01 -9.60469425e-01 -1.19922936e+00
-5.34612015e-02 6.40562296e-01 5.34926832e-01 -3.31910610e-01
3.37344021e-01 3.41988057e-01 -1.48410237e+00 1.34114549e-01
8.48516822e-01 7.00561225e-01 4.84845012e-01 4.40747231e-01
-2.78282136e-01 3.18095118e-01 -5.22563219e-01 -4.44873810e-01
2.72475988e-01 -3.89996499e-01 -1.07410860e+00 2.59319216e-01
6.48489833e-01 5.20340145e-01 1.94134295e-01 1.07424736e+00
4.91341613e-02 2.92252153e-01 8.98791909e-01 -1.00269520e+00
-3.17543089e-01 2.60673791e-01 -7.98155665e-01 3.79278898e-01
2.16953158e-02 -1.07529812e-01 7.18320489e-01 -6.30086899e-01
6.46452188e-01 7.33280122e-01 7.89609969e-01 4.39526528e-01
-1.67068696e+00 -1.54650658e-01 2.84322917e-01 1.38997987e-01
-1.33243239e+00 -3.33065510e-01 7.42958724e-01 -4.67829823e-01
7.81966448e-01 3.48872691e-02 7.92360246e-01 9.11578953e-01
5.11793613e-01 2.80845582e-01 1.44407165e+00 -6.43710971e-01
2.04768687e-01 4.29131508e-01 3.49633068e-01 8.78647447e-01
5.12147784e-01 -1.52006209e-01 -6.62337959e-01 1.81994751e-01
4.88380700e-01 -5.50799742e-02 -4.06974941e-01 -6.66582763e-01
-1.02766418e+00 5.34648359e-01 4.90913033e-01 8.41162145e-01
-3.35758716e-01 -9.45633873e-02 2.45665848e-01 8.23009789e-01
2.09875599e-01 4.44599986e-01 -4.81312692e-01 3.26227456e-01
-9.46751356e-01 -1.65048204e-02 7.57780731e-01 3.61079782e-01
1.23940933e+00 -1.45764410e-01 2.50257969e-01 5.37015319e-01
-1.85189947e-01 1.04691796e-02 5.50533354e-01 -6.94507360e-01
1.76729634e-01 3.39163363e-01 1.15139522e-01 -9.35253620e-01
-5.94557643e-01 -7.70225525e-01 -8.64328921e-01 5.12005985e-01
1.09469199e+00 -9.32089053e-03 -6.20805144e-01 1.82157445e+00
2.26174057e-01 2.64980555e-01 1.00194007e-01 7.12448299e-01
2.14053616e-01 3.69801730e-01 7.85945207e-02 -6.08936191e-01
1.00820994e+00 -1.00037253e+00 -3.68942440e-01 -3.15006793e-01
6.42060935e-01 -7.71561265e-01 1.02964175e+00 6.41793489e-01
-9.38810349e-01 -6.55190647e-01 -1.00190365e+00 2.30118558e-01
-2.98301935e-01 -9.85214785e-02 5.22170663e-01 5.04929006e-01
-1.07061589e+00 1.14095700e+00 -6.96777105e-01 -4.95455950e-01
4.95403707e-02 3.45241517e-01 -4.36939508e-01 1.42136604e-01
-7.46710896e-01 1.19114459e+00 4.82043743e-01 5.94866164e-02
-6.83677018e-01 -4.29436117e-01 -2.41098940e-01 8.76920857e-03
5.98391294e-01 -6.45027936e-01 8.78694773e-01 -1.45439613e+00
-1.11000454e+00 1.03263843e+00 -3.69885117e-01 -6.82566464e-01
7.48632312e-01 2.04995591e-02 -4.44499888e-02 -5.26401773e-02
-4.09255363e-02 5.20263612e-01 1.22444892e+00 -1.47191823e+00
-4.40120548e-01 -3.53690326e-01 4.94506694e-02 2.71018833e-01
-6.07757330e-01 -5.10370374e-01 -2.96853036e-01 -4.48968917e-01
3.12279582e-01 -9.20970142e-01 -9.14652497e-02 -1.30739853e-01
-1.63686186e-01 -3.77487928e-01 6.48908257e-01 -1.94435567e-01
8.73695672e-01 -2.06212950e+00 6.53285146e-01 1.80140167e-01
3.99200141e-01 1.59298688e-01 -3.21982093e-02 3.11301440e-01
-2.85796732e-01 5.80967441e-02 -2.11695597e-01 -7.78810859e-01
-3.78458470e-01 3.36317211e-01 -2.91980207e-01 7.17214465e-01
-1.61558896e-01 8.39052677e-01 -7.92685270e-01 -5.19040108e-01
2.41563860e-02 2.96278924e-01 -1.79264650e-01 -2.16233909e-01
-1.60913169e-01 7.81697810e-01 -1.26124784e-01 2.18028110e-02
5.01949370e-01 -2.74639755e-01 2.85274744e-01 -6.18373826e-02
-2.98541903e-01 -7.02563226e-02 -1.20532358e+00 1.66568613e+00
-3.94282341e-01 7.45490193e-01 -1.55686334e-01 -1.56985414e+00
8.08729291e-01 2.96943069e-01 4.39472258e-01 -6.29124105e-01
-7.25829452e-02 5.83368301e-01 1.22804478e-01 -3.48932892e-01
1.43525556e-01 -5.63441515e-01 3.54517877e-01 4.37798619e-01
2.15713754e-01 1.04346782e-01 2.71089077e-01 -1.39470130e-01
9.13886905e-01 -1.03503898e-01 1.16604567e-01 -5.20911813e-01
6.07266784e-01 4.02984209e-02 3.53028178e-01 1.02135766e+00
-5.90198711e-02 5.40798545e-01 6.51745319e-01 -5.97826779e-01
-9.01479304e-01 -9.40257609e-01 -1.95579320e-01 8.44039321e-01
4.12893891e-01 -5.67209944e-02 -5.94160020e-01 -8.34587336e-01
-1.39702573e-01 4.17090714e-01 -9.98749793e-01 -1.96715206e-01
-7.20571458e-01 -7.38967180e-01 1.07452959e-01 -1.05579898e-01
4.98068035e-01 -9.20188248e-01 -8.53262782e-01 2.14738950e-01
-4.90546413e-02 -7.65334249e-01 -2.11349934e-01 6.79336309e-01
-1.31248403e+00 -1.20790255e+00 -9.00840402e-01 -8.22861671e-01
7.02799380e-01 4.72495377e-01 1.19638538e+00 4.84442830e-01
4.81992327e-02 3.54778081e-01 -9.69524309e-02 1.48484791e-02
-4.00987685e-01 4.67913419e-01 2.19690099e-01 1.90993980e-01
-4.59947959e-02 -7.72375166e-01 -4.21650261e-01 4.13866341e-01
-9.85451639e-01 4.97250222e-02 5.42663336e-01 6.91910326e-01
6.05959237e-01 4.00295377e-01 6.61670566e-01 -9.19169962e-01
3.62750471e-01 -4.19867486e-01 -4.35226887e-01 5.26124775e-01
-8.59924912e-01 3.65875423e-01 7.91684687e-01 -5.60946465e-01
-9.52207446e-01 3.31630886e-01 -5.10166921e-02 -3.31689000e-01
-2.50549227e-01 4.01952118e-01 3.78628731e-01 -4.55299556e-01
8.26607823e-01 3.75640452e-01 7.72496387e-02 -5.49212396e-01
7.72726387e-02 8.68137255e-02 3.65846664e-01 -5.54447293e-01
9.30760503e-01 7.82166123e-01 2.85134465e-01 -1.02299261e+00
-1.17669654e+00 -4.48235959e-01 -9.59918082e-01 -4.37938094e-01
7.38746762e-01 -3.01445156e-01 -5.86696327e-01 5.28309882e-01
-1.26619387e+00 -4.90760326e-01 -5.34267247e-01 2.13192433e-01
-5.28737962e-01 4.98518527e-01 -7.74937049e-02 -8.36294293e-01
1.39259472e-01 -9.39437270e-01 5.88881731e-01 2.15495035e-01
-8.38318318e-02 -1.50302052e+00 2.68101633e-01 1.11979179e-01
2.70044774e-01 1.33608595e-01 1.03285444e+00 -4.75354046e-01
-3.14150214e-01 1.00870885e-01 1.05328232e-01 2.49812841e-01
1.77699894e-01 -1.50521323e-01 -6.72653437e-01 -6.67554915e-01
5.13712049e-01 -3.07078212e-01 1.38083923e+00 4.63884771e-01
5.80838382e-01 5.43552674e-02 -5.03033400e-01 4.28661108e-01
1.67830837e+00 -1.14274219e-01 5.76172888e-01 5.92420697e-01
4.11630183e-01 9.57988083e-01 2.72029340e-01 -1.99933231e-01
-8.12043622e-02 7.88485229e-01 4.30594087e-01 -1.19482502e-01
-2.18642697e-01 1.57168016e-01 1.10087551e-01 7.15921581e-01
-2.88327813e-01 1.65799499e-01 -9.03231502e-01 5.95386922e-01
-1.96103263e+00 -7.87476242e-01 -5.11659384e-01 2.63345504e+00
8.69480133e-01 6.60505116e-01 1.31636545e-01 3.06331664e-01
6.17224276e-01 5.71743064e-02 -4.11255151e-01 -1.81064934e-01
-3.45306784e-01 2.05491111e-01 3.74881208e-01 6.47485316e-01
-1.02669001e+00 7.44623005e-01 6.01183462e+00 8.61155450e-01
-1.38099122e+00 5.34387290e-01 5.05650699e-01 -1.81110188e-01
-6.14326783e-02 1.00955382e-01 -6.91063344e-01 2.00340241e-01
7.50205278e-01 -2.71019816e-01 3.64042193e-01 4.20049429e-01
-1.73615310e-02 -5.78140676e-01 -1.11076295e+00 7.02779412e-01
-9.29130167e-02 -1.36774683e+00 -6.66491762e-02 -2.57610008e-02
9.73025620e-01 -7.04649761e-02 2.71535099e-01 6.70748875e-02
-2.16601804e-01 -9.45319414e-01 6.93276644e-01 6.97574735e-01
4.07648891e-01 -4.92861718e-01 4.03254569e-01 7.74071991e-01
-1.11084032e+00 -1.12740755e-01 -4.39210087e-01 -4.28889506e-03
-4.11169225e-04 8.14179003e-01 -3.24045062e-01 7.14888930e-01
5.76921880e-01 5.00197649e-01 -7.69577682e-01 1.40037334e+00
-9.36113521e-02 4.00093496e-01 -3.49322021e-01 1.36555135e-01
3.22616965e-01 -3.07583958e-01 7.46560633e-01 8.78739536e-01
1.64213628e-01 -4.66758251e-01 -3.26330699e-02 5.68461537e-01
2.17684835e-01 1.51283845e-01 -6.94902599e-01 4.40067530e-01
-1.00436255e-01 1.10175431e+00 -1.28923035e+00 -8.69374722e-02
-4.61707711e-01 1.14450657e+00 8.70423615e-01 4.33050632e-01
-6.27261341e-01 -7.48138651e-02 1.94006264e-01 2.86888421e-01
4.22905445e-01 -5.17203510e-01 -3.58997792e-01 -1.11840975e+00
1.91458687e-01 -3.98026526e-01 2.65445650e-01 -2.73173511e-01
-1.14121604e+00 6.47080421e-01 -1.06242783e-01 -8.33106637e-01
2.58816853e-02 -3.88888061e-01 -6.53634667e-01 7.08853483e-01
-1.73485899e+00 -6.63597345e-01 -1.75266370e-01 7.75935411e-01
6.62158191e-01 -3.05016041e-02 6.76154137e-01 2.37679601e-01
-4.17222768e-01 3.09894502e-01 2.87628472e-01 -5.12651503e-01
7.33973384e-01 -1.46335030e+00 -1.43465012e-01 6.90492749e-01
7.49965429e-01 6.03468060e-01 9.07969773e-01 -6.55726552e-01
-1.08673644e+00 -7.68902123e-01 8.79199505e-01 -4.94602978e-01
7.03762114e-01 -1.41817018e-01 -1.31972480e+00 4.14184839e-01
1.98202536e-01 -1.19927473e-01 6.03229292e-02 2.88194805e-01
-1.23480305e-01 -1.35918647e-01 -6.92056894e-01 3.72105032e-01
8.89229596e-01 -1.89451277e-01 -6.63352430e-01 5.19418716e-01
3.25945318e-01 -6.02590702e-02 -3.51238877e-01 1.42667890e-01
3.17321271e-01 -1.44319737e+00 7.68202484e-01 -6.22966230e-01
2.58034050e-01 -1.75038755e-01 3.30073178e-01 -1.39974165e+00
2.71358807e-03 -6.39744282e-01 -1.32322058e-01 8.57024968e-01
4.65930998e-01 -9.74890292e-01 7.77276278e-01 -2.34620683e-02
7.39039481e-02 -9.58592296e-01 -1.39452064e+00 -1.33676803e+00
4.26696479e-01 -1.89318404e-01 -4.37851250e-02 9.35258806e-01
-4.88237172e-01 4.87878948e-01 -3.19258630e-01 7.35053793e-02
7.39880860e-01 2.09871083e-01 4.24850523e-01 -1.68158567e+00
-5.42279661e-01 -5.85718513e-01 -6.44440204e-02 -9.84080970e-01
2.85123110e-01 -9.21023369e-01 -1.20330498e-01 -1.29068863e+00
8.24489519e-02 -7.36921430e-01 -3.91442001e-01 4.20536637e-01
-6.15726821e-02 4.40223992e-01 1.13036640e-01 7.11930394e-01
-6.24797821e-01 2.78112829e-01 1.14543247e+00 3.23879793e-02
-4.95983243e-01 2.51669288e-01 -4.84770268e-01 8.62358510e-01
7.90216267e-01 -7.75721014e-01 -3.04986030e-01 -5.20295560e-01
7.05759466e-01 -1.17938966e-01 3.79150957e-01 -1.27652872e+00
5.67365468e-01 1.87014803e-01 1.26558617e-01 6.89927191e-02
2.78211206e-01 -9.31258082e-01 1.99362099e-01 6.29177511e-01
-2.10356593e-01 -2.15892807e-01 2.35963508e-01 8.47919583e-01
-2.72781169e-03 -9.27760839e-01 9.64123785e-01 -1.38934359e-01
-3.85561168e-01 6.98486296e-03 -4.32204038e-01 -1.80868674e-02
1.13517451e+00 -3.73930186e-01 -1.15676597e-01 -2.19604626e-01
-1.23540092e+00 5.17169647e-02 4.22635168e-01 3.69255394e-02
2.57663190e-01 -9.78381276e-01 -5.33007979e-01 -1.59822315e-01
-3.22448194e-01 -2.35053301e-01 1.07127465e-01 1.44425583e+00
-1.01759322e-01 1.83535129e-01 -1.60294607e-01 -7.43724167e-01
-1.30239463e+00 5.18064141e-01 7.19453931e-01 -4.77015615e-01
-5.56481302e-01 7.12936044e-01 1.02933370e-01 -1.05671003e-01
2.42608443e-01 6.48350343e-02 -2.54139960e-01 4.93964046e-01
1.06625825e-01 2.93870211e-01 3.72254789e-01 -4.46541280e-01
-8.96151215e-02 8.58968973e-01 -3.04532021e-01 -4.01680879e-02
1.41970086e+00 -1.66012481e-01 -2.59572566e-01 9.04696226e-01
1.03247499e+00 -7.05824280e-03 -1.43317652e+00 -2.97369629e-01
2.22867474e-01 -2.43055180e-01 -1.28853127e-01 -3.74903858e-01
-1.10507596e+00 9.36169386e-01 5.62134683e-01 9.37860250e-01
9.73300636e-01 6.01851940e-02 5.31653047e-01 6.32974029e-01
4.27491784e-01 -1.07732904e+00 4.74684507e-01 5.73558688e-01
9.05609250e-01 -1.26789081e+00 -3.20487730e-02 -2.63630748e-01
-4.01035339e-01 1.42692268e+00 2.88697690e-01 -3.41136992e-01
5.62019706e-01 -4.41364497e-02 -2.58427382e-01 -4.01841074e-01
-5.81140935e-01 -2.80164689e-01 2.98464119e-01 4.43156630e-01
1.64002821e-01 -4.06663895e-01 -3.87943029e-01 -8.73889402e-02
-1.87420584e-02 -2.57686853e-01 4.33884680e-01 7.11340606e-01
-6.19456947e-01 -1.18939555e+00 -3.24708849e-01 3.50476503e-01
-3.55068326e-01 6.13433756e-02 -3.23591262e-01 8.41578007e-01
1.37379572e-01 7.24477053e-01 -1.33631257e-02 -2.43107025e-02
2.39850208e-01 3.60892862e-01 8.51319373e-01 -7.18220770e-01
-5.59935272e-01 7.72906318e-02 -3.41827929e-01 -2.08413288e-01
-6.38828278e-01 -8.06669772e-01 -1.14314103e+00 -1.92941666e-01
-4.23875123e-01 2.89756417e-01 5.98569810e-01 1.30039096e+00
-9.97633263e-02 5.85157216e-01 5.16960561e-01 -8.79444718e-01
-4.36222017e-01 -5.15224159e-01 -4.80664492e-01 3.04027677e-01
4.88182843e-01 -8.34503651e-01 -6.33385658e-01 4.51184288e-02] | [9.399831771850586, 2.005295991897583] |
e300e9e3-ba1f-4980-898d-c7c133bbf5fc | batch-normalized-joint-training-for-dnn-based | 1703.08471 | null | http://arxiv.org/abs/1703.08471v1 | http://arxiv.org/pdf/1703.08471v1.pdf | Batch-normalized joint training for DNN-based distant speech recognition | Improving distant speech recognition is a crucial step towards flexible
human-machine interfaces. Current technology, however, still exhibits a lack of
robustness, especially when adverse acoustic conditions are met. Despite the
significant progress made in the last years on both speech enhancement and
speech recognition, one potential limitation of state-of-the-art technology
lies in composing modules that are not well matched because they are not
trained jointly. To address this concern, a promising approach consists in
concatenating a speech enhancement and a speech recognition deep neural network
and to jointly update their parameters as if they were within a single bigger
network. Unfortunately, joint training can be difficult because the output
distribution of the speech enhancement system may change substantially during
the optimization procedure. The speech recognition module would have to deal
with an input distribution that is non-stationary and unnormalized. To mitigate
this issue, we propose a joint training approach based on a fully
batch-normalized architecture. Experiments, conducted using different datasets,
tasks and acoustic conditions, revealed that the proposed framework
significantly overtakes other competitive solutions, especially in challenging
environments. | ['Yoshua Bengio', 'Maurizio Omologo', 'Philemon Brakel', 'Mirco Ravanelli'] | 2017-03-24 | null | null | null | null | ['distant-speech-recognition'] | ['speech'] | [ 6.25437737e-01 5.03046960e-02 3.34904313e-01 -4.29115504e-01
-5.63020945e-01 -3.00906241e-01 4.86025870e-01 -1.66192815e-01
-6.92666352e-01 6.19545579e-01 2.78086156e-01 -2.42538422e-01
-2.40892284e-02 -2.59789079e-01 -5.85077643e-01 -8.44055235e-01
4.79143888e-01 5.77966645e-02 9.76286754e-02 -2.84396797e-01
1.81966822e-03 4.97925878e-01 -1.80079520e+00 -1.96290147e-02
1.05167127e+00 8.70805621e-01 4.83395249e-01 6.56607807e-01
-2.53684402e-01 3.11585277e-01 -8.32087576e-01 -4.29934442e-01
1.75312623e-01 -2.68612742e-01 -4.17007864e-01 3.02696973e-01
3.78158003e-01 -2.18904927e-01 -3.16707522e-01 1.24622595e+00
8.63593102e-01 4.91085678e-01 3.58476758e-01 -8.06513429e-01
-3.62292707e-01 5.23627520e-01 -2.49315456e-01 1.75138354e-01
-4.24995497e-02 4.04885076e-02 5.53370118e-01 -8.41125548e-01
1.86701119e-01 1.06219316e+00 5.34305215e-01 7.40602553e-01
-1.07999277e+00 -4.59308416e-01 4.02199745e-01 5.56222089e-02
-1.14168549e+00 -9.06251967e-01 8.06685627e-01 -9.51845571e-02
1.13266385e+00 2.88418084e-01 2.97165096e-01 1.18488812e+00
-2.06093088e-01 4.92512941e-01 9.60837781e-01 -4.87952858e-01
1.48707151e-01 5.05741119e-01 -5.31570353e-02 1.73839033e-02
-4.42080237e-02 3.49064097e-02 -4.22862351e-01 1.41400009e-01
5.70170581e-01 -2.50648499e-01 -4.45757985e-01 -1.90444678e-01
-8.93813431e-01 2.80217618e-01 4.74859364e-02 9.96916592e-01
-5.37099421e-01 -2.88283914e-01 4.08171326e-01 4.18996334e-01
4.85072285e-01 2.56480664e-01 -3.92482787e-01 -4.48429555e-01
-1.21665609e+00 -2.13619191e-02 7.48462319e-01 4.82249916e-01
4.27903593e-01 4.00574774e-01 -8.09271187e-02 1.26161373e+00
2.83614069e-01 2.14495331e-01 7.40880787e-01 -4.47879016e-01
6.70083165e-01 2.63280332e-01 2.50222888e-02 -7.89633095e-01
-3.71101737e-01 -1.09011745e+00 -9.89889979e-01 3.44151407e-01
3.98120791e-01 -3.60286564e-01 -9.95352447e-01 1.90304470e+00
3.63222778e-01 1.30540997e-01 2.59716183e-01 1.02014959e+00
4.22958970e-01 6.60399318e-01 2.13915348e-01 -1.93010628e-01
1.04084432e+00 -1.07704115e+00 -1.08733058e+00 -4.60127741e-01
1.43426210e-01 -1.14306796e+00 1.03391421e+00 3.58504623e-01
-1.32025802e+00 -9.74653363e-01 -1.22120917e+00 2.07171172e-01
-4.15158153e-01 3.18530798e-01 3.99258323e-02 9.48531389e-01
-1.16249537e+00 5.59957087e-01 -6.79630280e-01 -3.65275383e-01
-6.32488281e-02 5.71622193e-01 -3.24924350e-01 3.05019855e-01
-1.10083437e+00 1.09986615e+00 3.33949059e-01 5.48997283e-01
-4.28742588e-01 -3.73781741e-01 -5.37857115e-01 3.50443155e-01
3.91047448e-01 -5.22843242e-01 1.30959308e+00 -1.19308913e+00
-2.01617670e+00 3.32519740e-01 -1.79360613e-01 -3.15378368e-01
5.58083951e-01 -3.41178000e-01 -8.01838577e-01 -4.23932195e-01
-5.34931421e-01 1.86348483e-01 1.25333023e+00 -1.11126673e+00
-5.62712848e-01 -4.25707549e-01 -2.22843230e-01 4.89897877e-01
-8.49276066e-01 1.99463949e-01 -4.13786143e-01 -7.85087824e-01
1.71253011e-01 -7.59087145e-01 -2.96731979e-01 -3.87279719e-01
-2.05048010e-01 1.20559990e-01 8.56469870e-01 -8.40177238e-01
1.29357088e+00 -2.27610207e+00 2.19751611e-01 1.41223788e-01
-1.88181326e-01 8.89166832e-01 -1.66580841e-01 2.80115426e-01
-2.32395515e-01 -1.54716030e-01 -3.15577596e-01 -8.14796627e-01
8.69515538e-02 7.75307342e-02 -1.20173357e-01 3.02499682e-01
2.37218514e-01 2.46228844e-01 -5.28108478e-01 -9.40221772e-02
4.28711504e-01 9.77939844e-01 -4.11585003e-01 4.65242594e-01
6.25977591e-02 6.84190571e-01 -7.60821477e-02 2.90461063e-01
8.92613471e-01 2.39997491e-01 2.39865750e-01 -2.52406120e-01
-2.60649115e-01 3.75794083e-01 -1.62457407e+00 1.72070217e+00
-6.30004466e-01 5.91627300e-01 6.38101697e-01 -1.36064494e+00
1.01468062e+00 7.14333177e-01 4.13776904e-01 -7.64900506e-01
2.71158129e-01 5.71708918e-01 1.34852082e-01 -6.21970654e-01
5.99540949e-01 -2.98910767e-01 4.67278481e-01 1.39249057e-01
1.35009333e-01 7.90412128e-02 -5.43711707e-03 -3.94441515e-01
8.20828974e-01 -8.75002611e-03 7.71272182e-02 5.90410754e-02
9.12484586e-01 -7.06743836e-01 4.80856150e-01 6.15759611e-01
-3.56581777e-01 6.71626449e-01 5.94391637e-02 1.00771591e-01
-9.67589259e-01 -1.06335127e+00 -3.15266661e-03 1.11925483e+00
-1.29555151e-01 -3.36363055e-02 -9.39912617e-01 -3.45015138e-01
-4.21178937e-01 4.92820501e-01 -1.23324662e-01 -9.96934399e-02
-7.55764365e-01 -7.18439460e-01 5.30147791e-01 3.92331183e-01
5.14802933e-01 -9.37707782e-01 -4.00139034e-01 5.99416316e-01
-9.38595980e-02 -1.33741677e+00 -5.44647694e-01 3.04039389e-01
-8.38018000e-01 -3.95488799e-01 -1.07818425e+00 -8.42898190e-01
5.29868364e-01 3.21805388e-01 7.69123375e-01 4.60441448e-02
8.82334262e-02 1.05310574e-01 -2.34523714e-01 -2.88616300e-01
-6.45909250e-01 2.83352107e-01 1.81779087e-01 3.73491466e-01
4.22968753e-02 -8.16792786e-01 -5.72863758e-01 2.99995750e-01
-1.07853949e+00 -1.33270755e-01 8.53945792e-01 9.60847557e-01
2.25176051e-01 3.52230608e-01 8.20720851e-01 -4.61226374e-01
8.99057269e-01 -2.18018547e-01 -4.05374348e-01 3.50670785e-01
-4.34742123e-01 3.08375824e-02 6.07361853e-01 -5.85358381e-01
-1.63604975e+00 1.49538204e-01 -7.62560308e-01 -6.23166896e-02
-5.03779590e-01 4.52826828e-01 -5.56501806e-01 -1.28490981e-04
4.43489939e-01 2.45204195e-01 6.29164428e-02 -7.12895155e-01
1.31668583e-01 1.16318166e+00 6.55561566e-01 -2.03643084e-01
7.17926741e-01 5.39276749e-02 -3.66365254e-01 -1.17790198e+00
-2.09929094e-01 -5.96431673e-01 -5.64940155e-01 -2.33201504e-01
6.26932263e-01 -6.96525395e-01 -3.44504505e-01 8.52233410e-01
-1.34499228e+00 -1.56514227e-01 -8.52965266e-02 7.45410681e-01
-1.56359330e-01 4.71752703e-01 -2.01043770e-01 -1.05307889e+00
-4.18238759e-01 -1.41834521e+00 6.91250622e-01 4.26924497e-01
-1.26231611e-01 -9.11221802e-01 2.53854692e-03 4.18010473e-01
1.08117425e+00 -3.47661495e-01 5.24979532e-01 -6.57135248e-01
-1.33147791e-01 -2.73215532e-01 -2.47755181e-02 8.56464744e-01
5.11638522e-01 3.21888626e-02 -1.28916585e+00 -3.23062629e-01
4.39533651e-01 2.31735725e-02 6.46375477e-01 3.43884289e-01
9.18231606e-01 -5.86135648e-02 6.27714843e-02 3.69733930e-01
1.02773440e+00 3.25942934e-01 6.87303722e-01 1.32004336e-01
3.01815003e-01 8.22881937e-01 3.71006399e-01 2.88739264e-01
2.96691805e-02 9.97264326e-01 3.68152648e-01 -3.66938472e-01
-3.66626829e-01 1.07905522e-01 5.54229140e-01 1.11859202e+00
-7.66981617e-02 -3.08007717e-01 -6.91823602e-01 3.56891006e-01
-1.67442858e+00 -9.75514770e-01 2.44393140e-01 2.31515980e+00
7.67494142e-01 2.06023827e-01 7.13513568e-02 4.48777378e-01
7.64384687e-01 1.93827495e-01 -5.35478354e-01 -5.57701349e-01
-2.40536407e-01 4.30929005e-01 8.22252259e-02 5.25839508e-01
-1.07127190e+00 6.98208570e-01 5.64771509e+00 7.46119738e-01
-1.52909541e+00 1.54518142e-01 2.99600154e-01 -1.23748749e-01
-1.70787066e-01 -3.79867345e-01 -5.83216548e-01 4.56787676e-01
1.05062425e+00 4.51899990e-02 4.59826887e-01 6.38987720e-01
4.03798431e-01 1.52082175e-01 -6.75432265e-01 1.02943766e+00
6.14713281e-02 -8.39773715e-01 -3.02830070e-01 -7.97935668e-03
5.21872103e-01 6.13378920e-02 3.74992192e-01 3.32358629e-01
-3.62003297e-01 -8.60369503e-01 6.43085003e-01 3.85334045e-01
4.08103704e-01 -5.91436327e-01 7.95405567e-01 3.81229341e-01
-1.15751016e+00 -6.61089793e-02 -1.37856111e-01 -3.35893291e-03
2.14824900e-01 5.90087295e-01 -7.71437526e-01 4.73211020e-01
6.27644300e-01 1.68125495e-01 -2.21896082e-01 1.20343161e+00
2.23898003e-03 4.54232424e-01 -3.73405993e-01 1.22454152e-01
8.92113242e-03 -1.86079755e-01 7.35148609e-01 1.29501224e+00
5.60944855e-01 -2.67674804e-01 -2.80496836e-01 6.24532163e-01
3.59871872e-02 2.35135615e-01 -3.93075913e-01 -1.49973229e-01
3.10092717e-01 1.28266120e+00 -5.09123921e-01 -1.88731849e-01
-5.43591678e-01 1.11480129e+00 1.87132865e-01 6.07456803e-01
-8.40295434e-01 -3.25524271e-01 8.26645255e-01 -9.50999409e-02
3.25085312e-01 -3.91269147e-01 -2.56079614e-01 -1.10050762e+00
4.18924868e-01 -1.12000501e+00 -1.03562713e-01 -4.20349926e-01
-9.46208239e-01 1.08934820e+00 -3.56074840e-01 -1.15030015e+00
-3.48533094e-01 -5.30063868e-01 -4.96956646e-01 1.04591084e+00
-1.56641269e+00 -8.97632957e-01 -1.99367046e-01 5.00609934e-01
7.89614439e-01 -1.94311708e-01 6.76049829e-01 9.69678044e-01
-7.85172522e-01 8.90248418e-01 1.21830232e-01 -4.28528190e-01
7.84952998e-01 -1.00461364e+00 3.93708825e-01 1.13408875e+00
7.30089843e-02 6.02513433e-01 9.45356190e-01 -3.46641093e-01
-1.06603503e+00 -7.83025384e-01 8.21288168e-01 3.51426721e-01
5.90643942e-01 -2.84890622e-01 -1.30863523e+00 4.78833504e-02
4.09844160e-01 -1.88121274e-01 5.24978518e-01 2.49460246e-02
1.70904636e-01 -3.23091656e-01 -8.86790872e-01 6.91302478e-01
9.90013063e-01 -7.19231606e-01 -5.01294315e-01 -1.15512371e-01
4.64329720e-01 -5.76800883e-01 -6.95442855e-01 5.56830764e-01
6.44396663e-01 -9.96293902e-01 7.87429929e-01 -3.45450550e-01
-2.61583686e-01 -2.66088605e-01 -2.07471773e-01 -1.60718060e+00
8.09718966e-02 -7.90596902e-01 -5.39273284e-02 1.58556390e+00
5.70761085e-01 -7.52326846e-01 7.70805120e-01 7.53622651e-01
-4.06383634e-01 -4.79152530e-01 -1.12158573e+00 -8.08692396e-01
-1.85871392e-01 -4.44360435e-01 5.76083779e-01 4.50626761e-01
-1.90915525e-01 2.92192727e-01 -5.62765241e-01 2.86967874e-01
2.57755995e-01 -3.99164021e-01 6.67921484e-01 -1.06469417e+00
-5.28858900e-01 -6.22465968e-01 -1.33795395e-01 -1.07284641e+00
1.55332074e-01 -2.99732476e-01 4.61899787e-01 -1.23805416e+00
-2.53017724e-01 -3.23767513e-01 -4.79082078e-01 2.11901933e-01
-3.07662457e-01 -1.02063023e-01 1.51902214e-01 -8.77782777e-02
-7.46084899e-02 7.96286464e-01 1.03351283e+00 -1.53301150e-01
-4.93493646e-01 4.21737283e-01 -4.98619884e-01 5.91551065e-01
9.06344533e-01 -2.25158513e-01 -5.26772797e-01 -6.28366292e-01
-1.77689552e-01 -4.50983346e-02 5.94808534e-02 -1.36362636e+00
5.48251748e-01 2.18919605e-01 -1.01547688e-01 -3.77113134e-01
6.64397597e-01 -1.18824017e+00 2.91368604e-01 1.94419831e-01
-2.52732128e-01 -2.67404485e-02 4.06451046e-01 3.30803603e-01
-5.39174795e-01 -3.57955992e-01 8.08279514e-01 3.18740129e-01
-5.05644500e-01 -2.95589175e-02 -4.83473569e-01 -3.92210066e-01
6.42983496e-01 -3.61990988e-01 -1.29379332e-01 -3.61012548e-01
-8.28075230e-01 -3.31334382e-01 1.08339779e-01 5.67091167e-01
5.02498865e-01 -1.01752591e+00 -5.82281590e-01 5.26818216e-01
-3.26906919e-01 -2.11986318e-01 7.27431059e-01 8.91469955e-01
4.86587659e-02 3.45306009e-01 -1.33409098e-01 -3.17677259e-01
-1.33060431e+00 2.20815778e-01 5.97275376e-01 -3.09358984e-01
-2.84965724e-01 7.34079897e-01 -5.78888990e-02 -3.89255911e-01
7.39266098e-01 -2.72453874e-01 -3.42291266e-01 8.87191072e-02
5.20488739e-01 3.61325085e-01 6.21653438e-01 -7.85475671e-01
-1.84526786e-01 4.22278821e-01 -3.48261558e-02 -3.82136226e-01
1.32285154e+00 -4.23326045e-01 1.85253128e-01 2.19765142e-01
1.01411378e+00 4.93932255e-02 -1.18178105e+00 -3.10314864e-01
-1.58987671e-01 -2.16867059e-01 2.81958461e-01 -7.14021742e-01
-1.13253260e+00 9.20859516e-01 9.14654553e-01 3.25364590e-01
1.35018253e+00 -4.82463062e-01 7.52758384e-01 2.87660062e-01
-9.23430771e-02 -1.48876154e+00 1.65090635e-02 5.42308629e-01
9.79728937e-01 -1.22006428e+00 -4.32594717e-01 -3.75482291e-01
-3.55892152e-01 1.09388030e+00 5.84231377e-01 3.69680643e-01
6.79022729e-01 3.50721449e-01 2.90600538e-01 2.69596606e-01
-4.31586266e-01 -3.23865950e-01 3.83315653e-01 5.17042339e-01
7.19662428e-01 -2.14426607e-01 -3.08108807e-01 5.05800426e-01
-1.65562227e-01 -8.64946619e-02 1.13492303e-01 8.15532327e-01
-3.69783968e-01 -1.53460991e+00 -5.64922929e-01 1.14117838e-01
-6.10785127e-01 3.33292875e-03 -8.57637301e-02 5.48996449e-01
1.43418089e-01 1.27886045e+00 -1.20358184e-01 -4.78783190e-01
6.22172058e-01 4.45215926e-02 3.14067006e-01 -4.63938087e-01
-8.60132575e-01 3.00661057e-01 7.72429556e-02 -2.31078416e-01
-5.42559206e-01 -5.54740131e-01 -9.99789238e-01 6.24798983e-02
-4.97109175e-01 7.21821487e-02 1.16975427e+00 1.01165569e+00
3.39017183e-01 9.13171530e-01 5.76575577e-01 -1.06192255e+00
-9.06550467e-01 -1.05814910e+00 -4.02092934e-01 3.49135071e-01
4.62989956e-01 -5.92715323e-01 -4.03971583e-01 -6.43989071e-02] | [14.902549743652344, 5.901392936706543] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.