ImportAnceSamplingPolicyEvaluationwithAnEstimatedBehaviorPolicyJosiahP.HAnna1ScottNiekum1PeterStone1Abstractdeterminetheexpectedreturn–sumofrewards–thatAnevaluationpolicy,πe,willobtainwhendeploy...
HOList:AnEnvironmentforMachineLearningofHigher-OrderTheoremProvingKshitijBAnsal1SarahLoos1MarkusRabe1ChristiAnSzegedy1StewartWilcox1AbstractriSpeech(PAnayotovetal.,2015)forspeechrecognition,theNet...
FastDirectSearchinAnOptimallyCompressedContinuousTargetSpaceforEfficientMulti-LabelActiveLearningWeishiShi1QiYu1AbstractretrievalAndorgAnization.UsersfromQ&Awebsites,suchasstackoverflowAndQuora,are...
ELFOpenGo:AnAnalysisAndOpenReimplementationofAlphaZeroYuAndongTiAn1JerryMa1QuchengGong1ShubhoSengupta1ZhuoyuAnChen1JamesPinkerton1C.LawrenceZitnick1AbstractHowever,theseadvAncesinplayingabilitycome...
CapsAndRuns:AnImprovedMethodforApproximatelyOptimalAlgorithmConfigurationGelle´rtWeisz1Andra´sGyo¨rgy12CsabaSzepesva´ri13AbstractlectingthesolverconfigurationinAnapplication-specificway.Theprob...
AnalyzingFederatedLearningthroughAnAdversarialLensArjunNitinBhagoji1SupriyoChakraborty2PrateekMittal1SeraphinCalo2AbstractthetrainingofAneuralnetworkmodelisdistributedbe-tweenmultipleagents.Ineachr...
AnInvestigationofModel-FreePlAnningArthurGuez1MehdiMirza1KarolGregor1RishabhKabra1SébastienRacAnière1ThéophAneWeber1DavidRaposo1AdamSAntoro1LaurentOrseau1TomEccles1GregWayne1DavidSilver1TimothyL...
AnOptimalPrivateStochastic-MABAlgorithmBasedonAnOptimalPrivateStoppingRuleTouqirSajed1OrSheffet1Abstractexperimentsinmedicine(Robbins,1952),hasapplicationsinfieldssuchasrAnking(Kvetonetal.,2015),re...
AnInvestigationintoNeuralNetOptimizationviaHessiAnEigenvalueDensityBehroozGhorbAni12ShAnkarKrishnAn2YingXiao2Abstractetal.,2016;2017;Yaoetal.,2018).Intheabsenceofsuchconcreteinformationabouttheeige...
AnInstabilityinVariationalInferenceforTopicModelsBehroozGhorbAni1HamidJavadi2AndreaMontAnari13AbstractNotethatwabelongstothesimplexP1(k)={w∈Rk≥0:w,1k=1}.ItiscommontoassumethatitspriorisNaivemeAn...
SAFFRON:AnAdaptiveAlgorithmforOnlineControloftheFalseDiscoveryRateAadityaRamdas1TijAnaZrnic2MartinJ.Wainwright1MichaelI.JordAn1Abstract1.IntroductionIntheonlinefalsediscoveryrate(FDR)problem,Itisno...
RacingThompson:AnEfficientAlgorithmforThompsonSamplingwithNon-conjugatePriorsYichiZhou1JunZhu1JingweZhuo1AbstractAsoneofthemostimportAntproblemsinlearningAnddecision-makinginunknownenvironments,MAB...
PixelSNAIL:AnImprovedAutoregressiveGenerativeModelXiChen12NikhilMishra12MostafaRohAninejad12PieterAbbeel12Abstractallowmodelingcomplexdependencies.ComparedtoGAns(Goodfellowetal.,2014),neuralautoreg...
NonparametricvariableimportAnceusingAnaugmentedneuralnetworkwithmulti-tasklearningJeAnFeng1BriAnD.Williamson1MarcoCarone12NoahSimon1AbstractrequiresarigorousdefinitionofAnestimablevariableim-portAn...
LinearSpectralEstimatorsAndAnApplicationtoPhaseRetrievalRaminaGhods1AndrewS.LAn2TomGoldstein3ChristophStuder1AbstractexistingresultsonphaseretrievalthatassumerAndomnessinthemeasurementmatrixA,wefoc...
Let’sbeHonest:AnOptimalNo-RegretFrameworkforZero-SumGamesEhsAnAsadiKAngarshahi1Ya-PingHsieh1MehmetFatihSahin1VolkAnCevher1AbstractresultingdynamicsisofgreatinterestinoptimizationAndbehavioralecono...
LearningtoExplain:AnInformation-TheoreticPerspectiveonModelInterpretationJiAnboChen12LeSong34MartinJ.Wainwright15MichaelI.JordAn1Abstractisfeatureselection,whichselectsasubsetoffeaturesthatareusefu...
GraphicalNonconvexOptimizationviaAnAdaptiveConvexRelaxationQiAngSun1KeAnMingTAn2HAnLiu3TongZhAng3AbstractdiagonalmatrixwithdiagonalelementsofΣ∗.ItiswellknownthatthejthAndkthvariablesareconditiona...
CRAFTML,AnEfficientClustering-basedRAndomForestforExtremeMulti-labelLearningWissamSiblini12PascaleKuntz1FrAnkMeyer2Abstractmal/dualconversionorsparsification(Yenetal.,2016)Andparallelizationonsuper...
AnOptimalControlApproachtoDeepLearningAndApplicationstoDiscrete-WeightNeuralNetworksQiAnxiaoLi1ShujiHao1Abstractversionsoftheabovealgorithms.Morebroadly,necessaryconditionsforoptimalitycAnbederived...