ResearchBenchmarksOffline Reinforcement Learning on D4RL AntMaze Medium-Diverse v2 (test)Follow80.2Normalized ScoreA2PO-3.20818.44640.161.754Mar 12, 2024Evaluation ResultsMethodMethodLinksNormalized ScoreA2PO2024.0380.2Diffusion-QL2024.0378.6IQL2024.0370EQL2024.0362.2CQL2024.0353.7LAPO2024.0330.2CQL+AW2024.0324TD3+BC2024.033AWAC2024.030.7BC2024.030BCQ2024.030MOPO2024.030