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

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

91.4ReturnCSPO

Evaluation Results

Method	Links
CSPO2026.06		91.4	4.8
APPO2026.06		74.72	2.6
C-TRPO2026.06		68.2	16.5
CPPOPID2026.06		67.12	24.7
PPO-Lag2026.06		64.26	30.5
FOCOPS2026.06		49.48	30.2
IPO2026.06		45.32	25.6
CPO2026.06		44.32	25
CUP2026.06		42.39	22.1
EPO2026.06		41.85	19.2
P3O2026.06		40.94	22.5
PCPO2026.06		32.53	37