ResearchBenchmarksOffline multitask Reinforcement Learning on D4RL Antmaze large-playFollow306Average Episodic ReturnDiSPO120.88168.94217265.06Mar 10, 2024Evaluation ResultsMethodMethodLinksAverage Episodic ReturnDiSPO2024.03306USFA2024.03250COMBO2024.03248GC-IQL2024.03229FB2024.03165RaMP2024.03134MOPO2024.03128