Safe Reinforcement Learning on Hopper-Velocity benchmark leaderboard · SOTA2 Research

Safe Reinforcement Learning on Hopper-Velocity

1,554.56RewardRCPO

Evaluation Results

Method	Links
RCPO2026.01		1,554.56	37.53	979.3
TRPO-ADRC2026.01		1,470.61	10.67	1,000
CUP2026.01		1,326.84	26.9	854.4
CPPO-ADRC2026.01		1,322.83	10.43	910.2
RCPO-ADRC2026.01		1,248.94	9.27	818.1
IPO2026.01		1,216.01	27.57	797.33
PDO2026.01		1,177.19	18.03	797.57