ResearchBenchmarksOffline Reinforcement Learning on DMControl cheetah-run (mixed)Follow93.16Normalized ScoreCQL + S2P13.568834.231954.89575.5581Sep 30, 2022Evaluation ResultsMethodMethodLinksNormalized ScoreCQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0993.16CQLS2P Augmentation=falseS2P Augmentation=false2022.0992.63IQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0988.53IQLS2P Augmentation=falseS2P Augmentation=false2022.0941.68SLAC-off + S2PS2P Augmentation=trueS2P Augmentation=true2022.0926.39SLAC-offS2P Augmentation=falseS2P Augmentation=false2022.0916.63