Near-OptimalModel-FreeReinforcementLearninginNon-StationaryEpisodicMDPsWeichaoMao1KaiqingZhang1RuihaoZhu2DavidSimchi-Levi2TamerBas¸ar1Abstractthroughsequentialinteractionswithaninitiallyunknownbut...
ImprovedCorruptionRobustAlgorithmsforEpisodicReinforcementLearningYifangChen1SimonS.Du1KevinJamieson1Abstractstageaccordingtotheunderlyingtransitionfunction.WestudyEpisodicreinforcementlearningunde...
GeneralizableEpisodicMemoryforDeepReinforcementLearningHaoHu1JianingYe2GuangxiangZhu1ZhizhouRen3ChongjieZhang1AbstractDiscreteEpisodicMemoryEpisodicmemory-basedmethodscanrapidlyKeyValuelatchontopas...
DetectingRewardsDeteriorationinEpisodicReinforcementLearningIdoGreenberg1ShieMannor12AbstractRLtasksisthesafetyandreliabilityofthesystem(Dulac-Arnoldetal.,2019;Chanetal.,2020),arisinginbothof-Inman...
BeenThere,DoneThat:Meta-LearningwithEpisodicRecallSamuelRitter12JaneX.Wang1ZebKurth-Nelson13SiddhantM.Jayakumar1CharlesBlundell1RazvanPascanu1MatthewBotvinick14AbstractAssuch,meta-learningresearchh...
NeuralEpisodicControlAlexanderPritzel1BenignoUria1SriramSrinivasan1Adria`Puigdome`nechBadia1OriolVinyals1DemisHassabis1DaanWierstra1CharlesBlundell1Abstractlearningratesmeanthatexperiencecanonlybei...