ResearchBenchmarksOffline Reinforcement Learning on D4RL Walker2d random v0Follow412ReturnGELATO32.4130.95229.5328.05Feb 22, 2021Evaluation ResultsMethodMethodLinksReturnGELATOdistance=metricdistance=metric2021.02412MOPOvariant=bootstrapvariant=bootstrap2021.02396GELATOdistance=l2distance=l22021.02388MBPO2021.02251SAC2021.02120Imitation2021.0247