ResearchBenchmarksOffline Reinforcement Learning on Stochastic D4RL Hopper Medium MuJoCoOffline Reinforcement Learning on Stochastic D4RL Hopper Medium MuJoCoFollow1,014Mean ReturnCODAC-C872.456909.203945.95982.697Jul 12, 2021Evaluation resultsMethodMethodLinksMean ReturnCVaR (0.1)CODAC-Cobjective=CVaRobjective=CVaR2021.071,014976.4CODAC-CObjective=Risk-sensitiveObjective=Risk-sensitive2021.071,014976.4ORAAC2021.071,007.1767.6ORAAC2021.071,007.1767.6CODAC-Nobjective=risk-neutralobjective=risk-neutral2021.07993.7952.5CODAC-NObjective=Risk-neutralObjective=Risk-neutral2021.07993.7952.5CQL2021.07877.9693CQL2021.07877.9693