ResearchBenchmarksOffline Reinforcement Learning on D4RL Antmaze-umaze 1M (full)Follow85.5Normalized ScoreIQL53.36461.70770.0578.393Jun 7, 2023Evaluation ResultsMethodMethodLinksNormalized ScoreIQLRatio=1, Size=1MRatio=1, Size=1M2023.0685.5CQLRatio=1, Size=1MRatio=1, Size=1M2023.0684.8TD3+BCRatio=1, Size=1MRatio=1, Size=1M2023.0678.6BCRatio=1, Size=1MRatio=1, Size=1M2023.0654.6