ResearchBenchmarksLocomotion on HalfCheetah non-stationary MuJoCoFollow15,369Final Training RewardBAPR4,443.87,280.1510,116.512,952.85May 15, 2026Evaluation ResultsMethodMethodLinksFinal Training RewardBAPRIterations=2000Iterations=20002026.0515,369RE-SACIterations=2000Iterations=20002026.0513,133ESCPIterations=2000Iterations=20002026.054,864