ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo walker2d-medium-replay v2Follow98.4Normalized ReturnVIPO-MOBILE22.27242.03661.881.564Apr 16, 2025Evaluation ResultsMethodMethodLinksNormalized ReturnVIPO-MOBILE2025.0498.4DQL2025.0495.5MOBILE2025.0489.9RAMBO2025.0489.2VIPO-MOPO2025.0488.2MOPO*Retrained on v2=trueRetrained on v2=true2025.0485.6CQL2025.0476.8IQL2025.0473.9COMBO2025.0456MOReL2025.0449.8TD3+BC2025.0425.2