Safe Reinforcement Learning on MetaDrive
0.69Normalized RewardRCDT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RCDTAggregation=Average2026.02 | 0.69 | 0.71 | |
| COptiDICEAggregation=Average2026.02 | 0.58 | 2.77 | |
| BCQ-LagAggregation=Average2026.02 | 0.54 | 2.05 | |
| TraCAggregation=Average2026.02 | 0.53 | 0.54 | |
| BC-ALLAggregation=Average2026.02 | 0.42 | 2.06 | |
| CDTAggregation=Average2026.02 | 0.4 | 0.95 | |
| FISORAggregation=Average2026.02 | 0.36 | 0.08 | |
| BC-SafeAggregation=Average2026.02 | 0.18 | 0.58 | |
| BEAR-LagAggregation=Average2026.02 | 0.02 | 0.39 | |
| CPQAggregation=Average2026.02 | -0.06 | 0.06 |