ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo halfcheetah-medium-expert v2Follow110Normalized ReturnVIPO-MOBILE51.03266.34181.6596.959Apr 16, 2025Evaluation ResultsMethodMethodLinksNormalized ReturnVIPO-MOBILE2025.04110MOBILE2025.04108.2TD3+BC2025.0497.9DQL2025.0496.8RAMBO2025.0495.4CQL2025.0495VIPO-MOPO2025.0493.8MOPO*Retrained on v2=trueRetrained on v2=true2025.0490.8COMBO2025.0490IQL2025.0486.7MOReL2025.0453.3