ResearchBenchmarksOffline Reinforcement Learning on D4RL mujoco-halfcheetah (med-replay)Follow68.2Normalized ScoreMOPO-0.4417.3835.253.02Oct 1, 2021Evaluation ResultsMethodMethodLinksNormalized ScoreMOPO2021.1068.2Repb-SDE2021.1065.5COMBO2021.1055.1CQL2021.1047.1ROMI-CQL2021.1046.4ROMI-BCQ2021.1043.3BCQ2021.1041.1MOREL2021.1040.2BAIL2021.1038.9BEAR2021.1035.7BC2021.1025.6BRAC-P2021.102.2BRAC-v2021.102.2