Safe Reinforcement Learning on Safety Gymnasium Hopper (C=25) benchmark leaderboard · SOTA2 Research

Safe Reinforcement Learning on Safety Gymnasium Hopper (C=25)

1,691.92ReturnCSPO

Evaluation Results

Method	Links
CSPO2026.06		1,691.92	7.1
CUP2026.06		1,661.6	20.5
CPO2026.06		1,639.22	28.1
FOCOPS2026.06		1,613.64	19.6
C-TRPO2026.06		1,597.48	23.2
CPPOPID2026.06		1,579.19	21.4
EPO2026.06		1,340.49	12.1
P3O2026.06		1,255.34	24.4
APPO2026.06		1,229.86	23.6
PCPO2026.06		1,112.55	22.5
PPO-Lag2026.06		607.2	22.59
IPO2026.06		185.98	27.1