ResearchBenchmarksOffline Reinforcement Learning on D4RL Medium v2Follow48.3Normalized Score (HalfCheetah)TD3+BC42.37243.91145.4546.989Jun 12, 2021Evaluation ResultsMethodMethodLinksNormalized Score (HalfCheetah)Normalized Score (Hopper)Normalized Score (Walker2d)TD3+BCseeds=5, hyperparamete...seeds=5, hyperparameter tuning=none, evaluation protocol=average over final 10 evaluations2021.0648.359.383.7DTseeds=3seeds=32021.0642.667.674