List your company

Constrained Reinforcement Learning on MuJoCo HalfCheetah (eval)

2,951.7RewardFOCOPS
2,349.3322,505.7162,662.12,818.484Apr 17, 2025

Evaluation results

MethodLinks
FOCOPSKnowledge Level=Oracle2025.042,951.73.1—
TRPO-SauteKnowledge Level=Oracle2025.042,448.616.7—
CUPKnowledge Level=Oracle2025.042,441.47.6—
TraCeSKnowledge Level=Zero Knowledge2025.042,372.522.42,749