Action-Value coverage estimation on RiverSwim mostly-right target policy T=50
0.523Q-Value Estimate (s=1, a=0)
Model-based bootstrap with percentile CI
Updated 21d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| 0.523 | 0.526 | 0.518 | ||
| 0.507 | 0.508 | 0.483 | ||
2026.05 | 0.503 | 0.499 | 0.462 | |
| 0.495 | 0.497 | 0.525 | ||
2026.05 | 0.487 | 0.488 | 0.491 | |
| 0.48 | 0.485 | 0.477 | ||
| 0.474 | 0.473 | 0.484 | ||
| 0.47 | 0.465 | 0.447 | ||
| 0.465 | 0.453 | 0.411 | ||
| 0.444 | 0.449 | 0.452 | ||
| 0.434 | 0.445 | 0.036 | ||
| 0.427 | 0.406 | 0.422 | ||
2026.05 | 0.332 | 0.324 | 0.208 | |
| 0.304 | 0.31 | 0.005 | ||
| 0.271 | 0.224 | 0.232 | ||
| 0.24 | 0.237 | 0.077 | ||
| 0.229 | 0.205 | 0.309 | ||
2026.05 | 0.195 | 0.188 | 0.119 | |
| 0 | 0 | 0 | ||
| 0 | 0 | 0 |