ARepresentationLearningPerspectiveontheImportanceofTrain-ValidationSplittinginMeta-LearningNikunjSaunshi1ArushiGupta1WeiHu1Abstractfrommany“train”taskstolearnausefulpriorthatcanhelpsolvenew“test...
ANewRepresentationofSuccessorFeaturesforTransferacrossDissimilarEnvironmentsMajidAbdolshah1HungLe1ThommenKarimpanalGeorge1SunilGupta1SantuRana1SvethaVenkatesh1Abstractintoindependentsub-domains.How...
UnderstandingContrastiveRepresentationLearningthroughAlignmentandUniformityontheHypersphereTongzhouWang1PhillipIsola1AbstractContrastiveRepresentationlearninghasbeenout-Alignment:Similarsampleshave...
T-Basis:aCompactRepresentationforNeuralNetworksAntonObukhov1MaximRakhuba1StamatiosGeorgoulis1MenelaosKanakis1DengxinDai1LucVanGool12Abstractthatcontainmillionsoftrainableparametersandconsumelotsofm...
SemiparametricNonlinearBipartiteGraphRepresentationLearningwithProvableGuaranteesSenNa1YuweiLuo1ZhuoranYang2ZhaoranWang3MladenKolar4Abstract1.IntroductionGraphRepresentationlearningisaubiquitoustas...
RobustGraphRepresentationLearningviaNeuralSparsificationChengZheng1BoZong2WeiCheng2DongjinSong2JingchaoNi2WenchaoYu2HaifengChen2WeiWang1Abstracttioninsocialnetworks(Akogluetal.,2015),recommen-datio...
RepresentationLearningviaAdversarially-ContrastiveOptimalTransportAnoopCherian1ShuchinAeron2Abstractspatio-temporalsettings.Thereareseveralpracticalissuesthatleadtothisdifficulty,notably(i)mostofth...
ProvableRepresentationLearningforImitationLearningviaBi-levelOptimizationSanjeevArora12SimonS.Du2ShamKakade3YupingLuo1NikunjSaunshi1AbstractMarkovdecisionprocesses(MDPs)thatsharethesamestateandacti...
NestedSubspaceArrangementforRepresentationofRelationalDataNozomiHata1ShizuoKaji2AkihiroYoshida1KatsukiFujisawa2Abstracttion,multiplication,differentiation,metric,andtopologysothatvariousoperations,...
InterferometricGraphTransform:aDeepUnsupervisedGraphRepresentationEdouardOyallon1Abstract(Defferrardetal.,2016),whichheavilyrelyonspectralRepresentationsestimatedfromagivenLaplacianoperator.Wepropo...
InformativeDropoutforRobustRepresentationLearning:AShape-biasPerspectiveBaifengShi1DinghuaiZhang2QiDai3ZhanxingZhu245YadongMu6JingdongWang3AbstractFigure1.Comparisonofdifferentshape-biasedmethods.(...
EducatingTextAutoencoders:LatentRepresentationGuidanceviaDenoisingTianxiaoShen1JonasMueller2ReginaBarzilay1TommiJaakkola1AbstractFigure1.LatentRepresentationslearnedbyAAEandDAAEwhenmappingclustered...
DistanceMetricLearningwithJointRepresentationDiversificationXuChu12YangLin12YashaWang23XitingWang4HailongYu12XinGao12QiTong25AbstractFigure1.Thearchitecturesofembeddinglossmethodsandclassifi-cation...
CURL:ContrastiveUnsupervisedRepresentationsforReinforcementLearningMichaelLaskin⇤1AravindSrinivas⇤1PieterAbbeel1AbstractFigure1.ContrastiveUnsupervisedRepresentationsforReinforcementLearning(CURL...
ConvexRepresentationLearningforGeneralizedInvarianceinSemi-Inner-ProductSpaceYingyiMa1VigneshGanapathiraman1YaoliangYu2XinhuaZhang1Abstractshipsbetweenclasseswhichholdforanyinput.Commonexamplesincl...
ContrastiveMulti-ViewRepresentationLearningonGraphsKavehHassani1AmirHoseinKhasahmadi12Abstractworks(Kipf&Welling,2017),molecules(Duvenaudetal.,2015),andknowledgegraphs(Vivona&Hassani,2019).Weintrod...
ChannelEquilibriumNetworksforLearningDeepRepresentationWenqiShao1ShitaoTang2XingangPan3PingTan2XiaogangWang1PingLuo4Abstractareimportantcomponentsforawiderangeoftaskssuchasimageclassification(Ioffe...
AutomaticShortcutRemovalforSelf-SupervisedRepresentationLearningMatthiasMinderer12OlivierBachem1NeilHoulsby1MichaelTschannen1AbstractInputimageLensoutputDifferenceInself-supervisedvisualrepresentat...
AdaptiveAdversarialMulti-taskRepresentationLearningYurenMao1WeiweiLiu2XueminLin1AbstractrestrictthehypothesisclassbyenforcingallthetaskstoshareanidenticaldistributionintheRepresentationspace.Advers...
Multi-ObjectRepresentationLearningwithIterativeVariationalInferenceKlausGreff12RaphaëlLopezKaufman3RishabhKabra3NickWatters3ChrisBurgess3DanielZoran3LoicMatthey3MatthewBotvinick3AlexanderLerchner3...