Continuous Control on Safety-Gymnasium SafetyCarGoal1 (generalization)
26.8Episodic RewardCPO-FOAM
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CPO-FOAM2026.04 | 26.8 | 10.2 | |
| Lagrangian PPO2026.04 | 22.1 | 48.5 | |
| Vanilla CPO2026.04 | 20.2 | 25.4 | |
| RCPO2026.04 | 19.8 | 41.2 | |
| PID-Lagrangian2026.04 | 19.5 | 20.8 | |
| IPO2026.04 | 18.5 | 33.6 |