ResearchBenchmarksNoisy Action Robust MDP on Hopper high randomness MuJoCoFollow331.69Worst-case ReturnARDT52.2524124.7987197.345269.8913Jul 25, 2024Evaluation ResultsMethodMethodLinksWorst-case ReturnARDTnumber of seeds=5, adv...number of seeds=5, adversarial policies=8, target returns=best2024.07331.69DTnumber of seeds=5, adv...number of seeds=5, adversarial policies=8, target returns=best2024.07298.62ESPERnumber of seeds=5, adv...number of seeds=5, adversarial policies=8, target returns=best2024.0763