Reinforcement Learning on 18 Confounded Environments Aggregate
1.93Normalized MeanRecurrent SAC
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Recurrent SAC2026.02 | 1.93 | 0.89 | 0.88 | |
| Causal PBRS2026.02 | 1.29 | 1.09 | 1.09 | |
| SAC Baseline2026.02 | 1 | 1 | 1 | |
| T-REX2026.02 | 0.96 | 0.98 | 0.96 | |
| CQL2026.02 | 0.85 | 0.85 | 0.81 |