ResearchBenchmarksOffline multitask Reinforcement Learning on D4RL Antmaze medium-playFollow624Average Episodic ReturnDiSPO216.32322.16428533.84Mar 10, 2024Evaluation ResultsMethodMethodLinksAverage Episodic ReturnDiSPO2024.03624COMBO2024.03397GC-IQL2024.03390USFA2024.03370RaMP2024.03271FB2024.03264MOPO2024.03232