ResearchBenchmarksOffline Reinforcement Learning on AntMaze Large DiverseFollow47.5ScoreIQL-1.910.92523.7536.575Jun 6, 2022Evaluation ResultsMethodMethodLinksScoreIQL2022.0647.5ROMI+BCQ2022.0641.2RORL2022.0641IQL+smoothing2022.0640CQL2022.0614.9AWAC2022.061BC2022.060TD3+BC2022.060