ResearchBenchmarksReinforcement Learning on AntMaze medium-play D4RLFollow673Average Episodic ReturnGC-Oracle214.36333.43452.5571.57Mar 10, 2024Evaluation ResultsMethodMethodLinksAverage Episodic ReturnGC-Oracle2024.03673DiSPO2024.03624COMBO2024.03397GC-IQL2024.03390USFA2024.03370RaMP2024.03271FB2024.03264MOPO2024.03232