List your company
ResearchFormerly WizWand

Safe Reinforcement Learning on Safety Gymnasium HalfCheetah C=25

2,870.7ReturnFOCOPS
1,271.73121,686.84812,101.9652,517.0819Jun 12, 2026
Updated 4mo ago

Evaluation results

MethodLinks
FOCOPS2026.062,870.720.3
CPPOPID2026.062,797.5224.2
C-TRPO2026.062,748.815.3
CPO2026.062,464.819.7
EPO2026.062,102.1110.9
CSPO2026.062,007.915.5
PPO-Lag2026.061,841.3824.6
IPO2026.061,825.6523.8
APPO2026.061,771.3223.1
P3O2026.061,745.8423.4
CUP2026.061,718.1921.3
PCPO2026.061,333.2318.3