ResearchBenchmarksOffline multitask Reinforcement Learning on D4RL Antmaze umaze-diverseFollow577Average Episodic ReturnGC-IQL441.8476.9512547.1Mar 10, 2024Evaluation ResultsMethodMethodLinksAverage Episodic ReturnGC-IQL2024.03577DiSPO2024.03568COMBO2024.03547FB2024.03474MOPO2024.03467RaMP2024.03460USFA2024.03447