Safe Reinforcement Learning on Safety Gymnasium HalfCheetah C=25 benchmark leaderboard · SOTA2 Research

Safe Reinforcement Learning on Safety Gymnasium HalfCheetah C=25

2,870.7ReturnFOCOPS

Evaluation Results

Method	Links
FOCOPS2026.06		2,870.7	20.3
CPPOPID2026.06		2,797.52	24.2
C-TRPO2026.06		2,748.8	15.3
CPO2026.06		2,464.8	19.7
EPO2026.06		2,102.11	10.9
CSPO2026.06		2,007.9	15.5
PPO-Lag2026.06		1,841.38	24.6
IPO2026.06		1,825.65	23.8
APPO2026.06		1,771.32	23.1
P3O2026.06		1,745.84	23.4
CUP2026.06		1,718.19	21.3
PCPO2026.06		1,333.23	18.3