RobustPolicyGradientagainstStrongDataCorruptionXuezhouZhang1YidingChen1JerryZhu1WenSun2Abstracthighlynoisydata,suchasautonomousdriving,quantitativetrading,ormedicaldiagnosis.Westudytheproblemofrobu...
OnRobustMeanEstimationunderCoordinate-levelCorruptionZifanLiu1JonghoPark1TheodorosRekatsinas1ChristosTzamos1Abstractfilteringordown-weightingcorrupteddatavectorstoreducetheirinfluence(Diakonikolase...
OnReinforcementLearningwithAdversarialCorruptionandItsApplicationtoBlockMDPTianhaoWu12YunchangYang3SimonS.Du4LiweiWang35Abstractisvulnerabletocorrupteddatastemmingfrommaliciousentities(Huangetal.,2...
ImprovedCorruptionRobustAlgorithmsforEpisodicReinforcementLearningYifangChen1SimonS.Du1KevinJamieson1Abstractstageaccordingtotheunderlyingtransitionfunction.Westudyepisodicreinforcementlearningunde...