Safe Reinforcement Learning on Bullet Safety Gym DroneRun (eval)
459RewardPPO-Lagrangian
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| PPO-LagrangianKnowledge Level=Oracle2025.04 | 459 | 33.2 | — | |
| CPPO-PIDKnowledge Level=Oracle2025.04 | 449.9 | 10.2 | — | |
| C-T BaselineKnowledge Level=Zero Knowledge2025.04 | 446.4 | 18.1 | 6,000 | |
| TraCeSKnowledge Level=Zero Knowledge2025.04 | 440 | 13.6 | 3,494 |