ResearchBenchmarksOffline multitask Reinforcement Learning on Hopper standOffline multitask Reinforcement Learning on Hopper standFollow800Average Episodic ReturnDiSPO233.2380.35527.5674.65Mar 10, 2024Evaluation resultsMethodMethodLinksAverage Episodic ReturnDiSPO2024.03800MOPO2024.03800USFA2024.03685FB2024.03670COMBO2024.03600RaMP2024.03255