ResearchBenchmarksOffline Reinforcement Learning on DMControl reacher-easy (random)Follow70.45Normalized ScoreIQL + S2P-2.932416.118835.1754.2212Sep 30, 2022Evaluation ResultsMethodMethodLinksNormalized ScoreIQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0970.45IQLS2P Augmentation=falseS2P Augmentation=false2022.0933.75SLAC-offS2P Augmentation=falseS2P Augmentation=false2022.0930.24SLAC-off + S2PS2P Augmentation=trueS2P Augmentation=true2022.0927.62CQLS2P Augmentation=falseS2P Augmentation=false2022.09-0.01CQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.09-0.11