Variational(Gradient)EstimateoftheScoreFunctioninEnergy-basedLatentVariableModelsFanBao12KunXu1ChongxuanLi1LanqingHong2JunZhu1BoZhang1Abstractsivepower(Salakhutdinov&Hinton,2009;Baoetal.,2020)andto...
ThreeOperatorSplittingwithaNonconvexLossFunctionAlpYurtsever1VarunMangalick1SuvritSra1Abstract2007),incertaintransportandassignmentproblems(Koop-mans&Beckmann,1957;Peyre´etal.,2019),amongcount-Wec...
SafeReinforcementLearningwithLinearFunctionApproximationSanaeAmani1ChristosThrampoulidis2LinF.Yang1Abstractactionmayleadtocatastrophicresults.Thus,safetyinRLhasbecomeaseriousissuethatrestrictstheap...
Risk-SensitiveReinforcementLearningwithFunctionApproximation:ADebiasingApproachYingjieFei1ZhuoranYang2ZhaoranWang1Abstractrisk-seekingobjectiveandβ<0inducesarisk-averseone.ItcanalsobeseenthatVβte...
RandomizedAlgorithmsforSubmodularFunctionMaximizationwithak-SystemConstraintShuangCui1KaiHan1TianshuaiZhu1JingTang2BenweiWu1HeHuang3Abstractsourcing(Singlaetal.,2016;Hanetal.,2018a),cluster-ing(Gom...
Neural-Pull:LearningSignedDistanceFunctionsfromPointCloudsbyLearningtoPullSpaceontoSurfacesBaoruiMa1ZhizhongHan2Yu-ShenLiu1MatthiasZwicker3Abstract2020;Takikawaetal.,2021;Marteletal.,2021;Oechsleet...
HowDoesLossFunctionAffectGeneralizationPerformanceofDeepLearning?ApplicationtoHumanAgeEstimationAliAkbari1MuhammadAwais1ManijehBashar2JosefKittler1Abstractconditions,cameraquality,headpose,makeupap...
FunctionContrastiveLearningofTransferableMeta-RepresentationsMuhammadWaleedGondal1ShrutiJoshi1NasimRahaman12StefanBauer13ManuelWu¨thrich1BernhardScho¨lkopf1AbstractrelyonthenumberofsamplesNbeingl...
DFACFramework:FactorizingtheValueFunctionviaQuantileMixtureforMulti-AgentDistributionalQ-LearningWei-FangSun123Cheng-KuangLee2Chun-YiLee1Abstractoptimizetheoverallrewardsineachepisode.Nevertheless,...
DecomposableSubmodularFunctionMinimizationviaMaximumFlowKyriakosAxiotis1AdamKarczmarz2AnishMukherjee2PiotrSankowski342AdrianVladu56Abstractimagesegmentation(Aroraetal.,2012;Shanuetal.,2016),cluster...
BesovFunctionApproximationandBinaryClassificationonLow-DimensionalManifoldsUsingConvolutionalResidualNetworksHaoLiu1MinshuoChen2TuoZhao2WenjingLiao3AbstractThesuccessofdeeplearningclearlydemonstrat...
Average-RewardOff-PolicyPolicyEvaluationwithFunctionApproximationShangtongZhang1YiWan2RichardS.Sutton2ShimonWhiteson1Abstractwhichaimtogenerateapolicythatmaximizestherewardratebyiterativelyimprovin...
Black-boxdensityFunctionestimationusingrecursivepartitioningErikBodin1ZhenwenDai2NeillD.F.Campbell3CarlHenrikEk4Abstractispreventingtractablecomputationsisthatthedensityfunc-tiondoesnothaveanamenab...
ProvablyConvergentTwo-TimescaleOff-PolicyActor-CriticwithFunctionApproximationShangtongZhang1BoLiu2HengshuaiYao3ShimonWhiteson1Abstractatwo-timescaleconvergentanalysisunderFunctionapproxi-mation(Ko...
PolynomialTensorSketchforElement-wiseFunctionofLow-RankMatrixInsuHan1HaimAvron2JinwooShin31Abstractside,weobtainao(n2)-timeapproximationschemeoff(A)x≈TUTVxforanarbitraryvectorx∈RndueThispaperstud...
Minimax-OptimalOff-PolicyEvaluationwithLinearFunctionApproximationYaqiDuan1ZeyuJia2MengdiWang34Abstractvalue)tobeearnedbyanewpolicybasedonloggedhistory.Thispaperstudiesthestatisticaltheoryofoff-Int...
LossFunctionSearchforFaceRecognitionXiaoboWang1ShuoWang1ChengChi2ShifengZhang2TaoMei1AbstractGenerally,theCNNsareequippedwithclassificationlossFunctions(Liuetal.,2017;Wangetal.,2018f;Chenetal.,Infa...
IdentifyingRewardFunctionsusingAnchorActionsSinongGeng1HoussamNassif2CarlosA.Manzanares2A.MaxReppen3RonnieSircar3AbstractwithfirmprofitFunctions(Abbring,2010;AguirregabiriaandNevo,2013).Weproposear...
TheValueFunctionPolytopeinReinforcementLearningRobertDadashi1AdrienAliTa¨ıga12NicolasLeRoux1DaleSchuurmans13MarcG.Bellemare1AbstractLinetheorem.Weshowthatpoliciesthatagreeonallbutonestategenerate...
SortingOutLipschitzFunctionApproximationCemAnil12JamesLucas12RogerGrosse12AbstractExistingapproachestoenforceLipschitzconstraintsfallintotwocategories:regularizationandarchitecturalcon-Trainingneur...