Offline Reinforcement Learning Policy Evaluation on MIMIC-III and MIMIC-IV
0.87FQE ScoreT-CQL
Evaluation results
| Method | Links | ||
|---|---|---|---|
| T-CQL2026.03 | 0.87 | 0.16 | |
| Deepvent (CQL)2026.03 | 0.79 | 0.19 | |
| BCQ2026.03 | 0.63 | 0.24 | |
| Clinician2026.03 | 0.46 | 0.27 |
| Method | Links | ||
|---|---|---|---|
| T-CQL2026.03 | 0.87 | 0.16 | |
| Deepvent (CQL)2026.03 | 0.79 | 0.19 | |
| BCQ2026.03 | 0.63 | 0.24 | |
| Clinician2026.03 | 0.46 | 0.27 |