ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo hopper-medium-replay v2Follow109.6Normalized Average ReturnVIPO-MOBILE28.27249.38670.591.614Apr 16, 2025Evaluation ResultsMethodMethodLinksNormalized Average ReturnVIPO-MOBILE2025.04109.6MOBILE2025.04103.9MOPO*Retrained on v2=trueRetrained on v2=true2025.04103.5VIPO-MOPO2025.04103DQL2025.04101.3RAMBO2025.0499.5IQL2025.0494.7MOReL2025.0493.6COMBO2025.0489.5CQL2025.0486.3TD3+BC2025.0431.4