ResearchBenchmarksOffline Reinforcement Learning on D4RL Antmaze large-diverse v2Follow53.3Averaged ScoreDTAMP22.130.238.346.4Dec 6, 2023Evaluation ResultsMethodMethodLinksAveraged ScorePerformance Diff vs Single-GoalDTAMPGoal Setting=multi-goalGoal Setting=multi-goal2023.1253.3-31.7IQL+HERGoal Setting=multi-goa...Goal Setting=multi-goal, Hindsight Experience Replay=true2023.1242-11.6ContRLGoal Setting=multi-goalGoal Setting=multi-goal2023.1226.7-50.6TT+IQLGoal Setting=multi-goalGoal Setting=multi-goal2023.1223.3-64.5