GMAC:ADistributionalPerspectiveonActor-CriticFrameworkDanielWontaeNam1YounghoonKim1ChanY.Park1Abstract(a)TheobservationinputInthispaper,wedeviseadistributionalframe-(b)Theevaluatedvaluedistribution...
Class2Simi:ANoiseReductionPerspectiveonLearningwithNoisyLabelsSonghuaWu1XiaoboXia1TongliangLiu1BoHan2MingmingGong3NannanWang4HaifengLiu5GangNiu6Abstractdealwiththelabelnoiseprobleminpointwisemanner...
ACE:ExplainingclusterfromanadversarialPerspectiveYangYoungLu1TimothyC.Yu2GiancarloBonora1WilliamStaffordNoble13Abstractcorrespondtofeaturesofthosecells(geneexpressionval-ues,methylationevents,etc.)...
AFunctionalPerspectiveonLearningSymmetricFunctionswithNeuralNetworksAaronZweig1JoanBruna12AbstractSuchsymmetricfunctionsappearnaturallyacrossseveraldomains,includingparticlephysics,computergraphics...
AStatisticalPerspectiveonDistillationAdityaKrishnaMenon1AnkitSinghRawat1SashankJ.Reddi1SeungyeonKim1SanjivKumar1Abstractetal.,2019).OnecommonlyacceptedintuitionfromHintonetal.(2015)isthattheteacher...
ARepresentationLearningPerspectiveontheImportanceofTrain-ValidationSplittinginMeta-LearningNikunjSaunshi1ArushiGupta1WeiHu1Abstractfrommany“train”taskstolearnausefulpriorthatcanhelpsolvenew“test...
TowardsAdaptiveResidualNetworkTraining:ANeural-ODEPerspectiveChengyuDong1LiyuanLiu2ZichaoLi1JingboShang1Abstractisnotnew(Changetal.,2017;Wenetal.,2019),butthedynamicsofgrowingispoorlyunderstood.Inp...
MinimaxParetoFairness:AMultiObjectivePerspectiveNataliaMartinez1MartinBertran1GuillermoSapiro1Abstractgorithmswhoseoutputsareindependentofthesensitiveattributegiventheobjective’sgroundtruth(e.g.,E...
AnOptimisticPerspectiveonOfflineReinforcementLearningRishabhAgarwal1DaleSchuurmans12MohammadNorouzi1Abstractunsafe,orrequireahigh-fidelitysimulatorthatisoftendiffi-culttobuild(Dulac-Arnoldetal.,201...
VariationalAnnealingofGANs:ALangevinPerspectiveChenyangTao1ShuyangDai1LiqunChen1KeBai1JunyaChen12ChangLiu13RuiyiZhang1GeorgiyBobashev4LawrenceCarin1Abstractapre-specifiedsimpledistributionq(z)throu...
ProbabilityFunctionalDescent:AUnifyingPerspectiveonGANs,VariationalInference,andReinforcementLearningCaseyChu1JoseBlanchet2PeterGlynn2Abstractreinforcementlearning.Thegoalofthispaperistoprovideauni...
NeuralNetworkAttributions:ACausalPerspectiveAdityaChattopadhyay1PiyushiManupriya2AnirbanSarkar2VineethNBalasubramanian2Abstractwayforanewparadigm,“explainablemachinelearning”.Weproposeanewattribu...
AKernelPerspectiveforRegularizingDeepNeuralNetworksAlbertoBietti1GrégoireMialon12DexiongChen1JulienMairal1Abstractsuchkernels,theRKHScontainsindeeddeepconvolutionalnetworkssimilartogenericones—up...
ADynamicalSystemsPerspectiveonNesterovAccelerationMichaelMuehlebach1MichaelI.Jordan1Abstractbeenmanyattemptstounderstandandcharacterizethephe-nomenon.Bubecketal.(2015)suggestamodificationofWepresen...
ADeepReinforcementLearningPerspectiveonInternetCongestionControlNathanJay1NogaH.Rotman2P.BrightenGodfrey1MichaelSchapira2AvivTamar3AbstractFigure1:MultipletrafficflowssharingalinkWepresentandinvest...
LearningtoExplain:AnInformation-TheoreticPerspectiveonModelInterpretationJianboChen12LeSong34MartinJ.Wainwright15MichaelI.Jordan1Abstractisfeatureselection,whichselectsasubsetoffeaturesthatareusefu...
ADistributionalPerspectiveonReinforcementLearningMarcG.Bellemare1WillDabney1Re´miMunos1Abstractmentlearning.Specifically,themainobjectofourstudyistherandomreturnZwhoseexpectationisthevalueQ.ThisIn...
SketchedRidgeRegression:OptimizationPerspective,StatisticalPerspective,andModelAveragingShusenWang1AlexGittens2MichaelW.Mahoney1Abstractlemwhilegreatlymitigatingthestatisticalrisksincurredbysketchi...