VariationalEmpowermentasRepresentationLearningforGoal-BasedReinforcementLearningJongwookChoi†1ArchitSharma†2HonglakLee13SergeyLevine45ShixiangShaneGu4Abstract1IntroductionLearningtoreachgoalstate...
VariationalAuto-RegressiveGaussianProcessesforContinualLearningSanyamKapoor1TheofanisKaraletsos2ThangD.Bui3Abstract(a)VAR-GP(ours)Throughsequentialconstructionofposteriorson(b)VCL(withcoresetsize10...
UnsupervisedSkillDiscoverywithBottleneckOptionLearningJaekyeomKim1SeohongPark1GunheeKim1Abstractlearnedskillscanencouragetheexplorationforencounter-ingrewards,notonlybyprovidingusefulprimitivesfort...
UnsupervisedLearningofVisual3DKeypointsforControlBoyuanChen1PieterAbbeel1DeepakPathak2AbstractFigure1.Weproposeanend-to-endframeworkforunsupervisedLearningof3Dkeypointsfrommulti-viewimages.Thesekey...
UnsupervisedRepresentationLearningviaNeuralActivationCodingYookoonPark1SanghoLee2GunheeKim2DavidM.Blei1Abstract(a)Atinitialization(b)AfterNACtrainingWepresentneuralactivationcoding(NAC)asaFigure1.D...
UniSpeech:UnifiedSpeechRepresentationLearningwithLabeledandUnlabeledDataChengyiWang1YuWu2YaoQian2KenichiKumatani2ShujieLiu2FuruWei2MichaelZeng2XuedongHuang2Abstractmajorityofthenearly7000languagess...
UnICORNN:ArecurrentmodelforLearningverylongtimedependenciesT.KonstantinRusch1SiddharthaMishra1AbstractsuchasinLSTMs(Hochreiter&Schmidhuber,1997)andGRUs(Choetal.,2014),wheretheadditivestructureofThe...
UnderstandingSelf-SupervisedLearningDynamicswithoutContrastivePairsYuandongTian1XinleiChen1SuryaGanguli12Abstractquiringexpensivetargetlabels(Devlinetal.,2018).Manystate-of-the-artSSLmethodsincompu...
UncertaintyWeightedActor-CriticforOfflineReinforcementLearningYueWu12ShuangfeiZhai1NitishSrivastava1JoshuaSusskind1JianZhang1RuslanSalakhutdinov2HanlinGoh1Abstractleveragingpriorexperience(Langeeta...
TowardsUnderstandingLearninginNeuralNetworkswithLinearTeachersRoeiSarussi1AlonBrutzkus1AmirGloberson1Abstractwillseparatethedata.Whichofthesewillbefoundbygradientdescent?Cananeuralnetworkminimizing...
TowardsDomain-AgnosticContrastiveLearningVikasVerma12Minh-ThangLuong1KenjiKawaguchi3HieuPham1QuocV.Le1Abstract(Dai&Le,2015;Howard&Ruder,2018;Petersetal.,2018;Radfordetal.,2019;Clarketal.,2020),ands...
TowardUnderstandingtheFeatureLearningProcessofSelf-supervisedContrastiveLearningZixinWen1YuanzhiLi2AbstractcanevenoutperformthoselearnedbysupervisedLearninginseveraldownstreamtasks.Theremakablepote...
TowardsBetterLaplacianRepresentationinReinforcementLearningwithGeneralizedGraphDrawingKaixinWang1KuangqiZhou1QixinZhang2JieShao3BryanHooi1JiashiFeng1AbstractFigure1.VisualizationofenvironmentandLap...
TheImpactofRecordLinkageonLearningfromFeaturePartitionedDataRichardNock1StephenHardy2WilkoHenecka2HamishIvey-Law3JakubNabaglo3GiorgioPatrini4GuillaumeSmith2BrianThorne5Abstract"whenisaglobaltrained...
TFix:LearningtoFixCodingErrorswithaText-to-TextTransformerBerkayBerabi12JingxuanHe1VeselinRaychev12MartinVechev1Abstractsuchasvariablemisuses(Allamanisetal.,2018)andintegertypeerrors(Coker&Hafiz,20...
Tesseract:TensorisedActorsforMulti-AgentReinforcementLearningAnujMahajan1MikayelSamvelyan2LeiMao3ViktorMakoviychuk3AnimeshGarg3JeanKossaifi3ShimonWhiteson1YukeZhu3AnimashreeAnandkumar3Abstractarise...
TensorProgramsIV:FeatureLearninginInfinite-WidthNeuralNetworksGregYang1EdwardJ.Hu23AbstractFigure1.PCAofWord2VecembeddingsoftopUScitiesandstates,forNTK,width-64,andwidth-∞featureLearningnetworksAs...
TemporallyCorrelatedTaskSchedulingforSequenceLearningXueqingWu1LewenWang2YingceXia2WeiqingLiu2LijunWu2ShufangXie2TaoQin2Tie-YanLiu2Abstractneousmachinetranslation,weneedtobegintranslationbeforeread...
TemporalDifferenceLearningasGradientSplittingRuiLiu1AlexOlshevsky2AbstractTDusesdifferencesinpredictionsoversuccessivetimestepstodrivetheLearningprocess,withthepredictionatTemporaldifferencelearnin...