ResearchBenchmarksOffline Reinforcement Learning on DMControl cartpole-swingup (random)Offline Reinforcement Learning on DMControl cartpole-swingup (random)Follow38.59Normalized ScoreIQL + S2P23.957227.756131.55535.3539Sep 30, 2022Evaluation resultsMethodMethodLinksNormalized ScoreIQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0938.59SLAC-offS2P Augmentation=falseS2P Augmentation=false2022.0935.03CQL + S2PS2P Augmentation=trueS2P Augmentation=true2022.0932.93SLAC-off + S2PS2P Augmentation=trueS2P Augmentation=true2022.0931.01CQLS2P Augmentation=falseS2P Augmentation=false2022.0927.67IQLS2P Augmentation=falseS2P Augmentation=false2022.0924.52