Safe Reinforcement Learning on Bullet Safety Gym
0.73Normalized RewardBCQ-Lag
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BCQ-LagAggregation=Average2026.02 | 0.73 | 3.11 | |
| BC-ALLAggregation=Average2026.02 | 0.64 | 3.36 | |
| RCDTAggregation=Average2026.02 | 0.63 | 0.68 | |
| CDTAggregation=Average2026.02 | 0.61 | 0.77 | |
| TraCAggregation=Average2026.02 | 0.61 | 0.56 | |
| COptiDICEAggregation=Average2026.02 | 0.54 | 2.55 | |
| BC-SafeAggregation=Average2026.02 | 0.52 | 0.82 | |
| BEAR-LagAggregation=Average2026.02 | 0.48 | 3.81 | |
| FISORAggregation=Average2026.02 | 0.39 | 0.03 | |
| CPQAggregation=Average2026.02 | 0.33 | 1.15 |