ResearchBenchmarksReinforcement Learning on Hopper mediumFollow1,676ReturnGELATO270.96635.731,000.51,365.27Feb 22, 2021Evaluation ResultsMethodMethodLinksReturnGELATOvariant=metricvariant=metric2021.021,676GELATOvariant=Lvariant=L2021.021,320Imitation2021.021,234MOPO2021.021,202Data Score2021.021,021MBPO2021.02457SAC2021.02325