ResearchBenchmarksReinforcement Learning on AntMaze totalReinforcement Learning on AntMaze totalFollow378.6Initial Score (Offline RL)IQL367.368370.284373.2376.116Feb 13, 2022Evaluation resultsMethodMethodLinksInitial Score (Offline RL)Final Score (1M Online Steps)Score ImprovementIQLEvaluation Protocol=On...Evaluation Protocol=Online Fine-tuning after Offline RL, Online Steps=1M, Number of Seeds=52022.02378.6490.4—SPOTEvaluation Protocol=On...Evaluation Protocol=Online Fine-tuning after Offline RL, Online Steps=1M, Number of Seeds=52022.02367.856978.6