ResearchBenchmarksOffline Reinforcement Learning on D4RL Gym MuJoCo walker2d-medium-expert v2Follow117.7Normalized Average ReturnVIPO-MOBILE54.2670.7387.2103.67Apr 16, 2025Evaluation ResultsMethodMethodLinksNormalized Average ReturnVIPO-MOBILE2025.04117.7MOBILE2025.04115.2MOPO*Retrained on v2=trueRetrained on v2=true2025.04112.9VIPO-MOPO2025.04111.3DQL2025.04110.1IQL2025.04109.6CQL2025.04109.1TD3+BC2025.04105.7COMBO2025.04103.3MOReL2025.0495.6RAMBO2025.0456.7