ResearchBenchmarksOffline Reinforcement Learning on DMControl reacher-easy (mixed)Follow75.61Normalized ScoreIQL + S2P61.965265.507669.0572.5924Sep 30, 2022Evaluation ResultsMethodMethodLinksNormalized ScoreIQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0975.61CQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0975.53CQLS2P Augmentation=falseS2P Augmentation=false2022.0970.37IQLS2P Augmentation=falseS2P Augmentation=false2022.0966.88SLAC-off + S2PS2P Augmentation=trueS2P Augmentation=true2022.0963.59SLAC-offS2P Augmentation=falseS2P Augmentation=false2022.0962.49