Multi-Objective Offline Policy Evaluation on MIMIC-IV (test)
0.643FQEPEDA DT
Evaluation Results
| Method | Links | |
|---|---|---|
| PEDA DTPreferences=[0.1, 0.9]2025.12 | 0.643 | |
| OPL-MT-MNARInfo=MNAR + Text2026.04 | 0.634 | |
| OPL-MT-MNAR (Low-Freq. Text)Info=Rad. + Micro.2026.04 | 0.634 | |
| OPL-MT-MNARInfo=All2026.04 | 0.634 | |
| PEDA DTPreferences=[0.3, 0.7]2025.12 | 0.615 | |
| PEDA DTPreferences=[0.2, 0.8]2025.12 | 0.612 | |
| PEDA DTPreferences=[0.4, 0.6]2025.12 | 0.608 | |
| OPL-MT-MNAR (Structured Only)Info=None2026.04 | 0.606 | |
| BCPreferences=[0.0, 1.0]2025.12 | 0.601 | |
| BCPreferences=[0.1, 0.9]2025.12 | 0.594 | |
| MedDreamerInfo=Model-based2026.04 | 0.591 | |
| CQLPreferences=[0.0, 1.0]2025.12 | 0.59 | |
| PEDA DTPreferences=[0.0, 1.0]2025.12 | 0.588 | |
| BCPreferences=[0.2, 0.8]2025.12 | 0.587 | |
| CQLPreferences=[0.1, 0.9]2025.12 | 0.581 | |
| BCPreferences=[0.3, 0.7]2025.12 | 0.579 | |
| PEDA DTPreferences=[0.5, 0.5]2025.12 | 0.575 | |
| DDQNPreferences=[1.0, 0.0]2025.12 | 0.574 | |
| PEDA DTPreferences=[0.7, 0.3]2025.12 | 0.573 | |
| DDQNPreferences=[0.9, 0.1]2025.12 | 0.573 | |
| CQLPreferences=[0.2, 0.8]2025.12 | 0.572 | |
| BCPreferences=[0.4, 0.6]2025.12 | 0.572 | |
| PEDA DTPreferences=[0.6, 0.4]2025.12 | 0.572 | |
| DDQNPreferences=[0.7, 0.3]2025.12 | 0.572 | |
| DDQNPreferences=[0.8, 0.2]2025.12 | 0.572 | |
| DDQNPreferences=[0.5, 0.5]2025.12 | 0.571 | |
| DDQNPreferences=[0.6, 0.4]2025.12 | 0.571 | |
| DDQNPreferences=[0.4, 0.6]2025.12 | 0.57 | |
| DDQNPreferences=[0.2, 0.8]2025.12 | 0.569 | |
| DDQNPreferences=[0.3, 0.7]2025.12 | 0.569 | |
| DDQNPreferences=[0.1, 0.9]2025.12 | 0.568 | |
| DDQNPreferences=[0.0, 1.0]2025.12 | 0.567 | |
| BCPreferences=[0.5, 0.5]2025.12 | 0.565 | |
| CQLPreferences=[0.3, 0.7]2025.12 | 0.564 | |
| BCPreferences=[0.6, 0.4]2025.12 | 0.558 | |
| CQLPreferences=[0.4, 0.6]2025.12 | 0.555 | |
| PEDA DTPreferences=[0.8, 0.2]2025.12 | 0.555 | |
| BCPreferences=[0.7, 0.3]2025.12 | 0.55 | |
| CQLPreferences=[0.5, 0.5]2025.12 | 0.546 | |
| BCPreferences=[0.8, 0.2]2025.12 | 0.543 | |
| DDPG with Clinician SupervisionInfo=Model-free2026.04 | 0.538 | |
| CQLPreferences=[0.6, 0.4]2025.12 | 0.537 | |
| PEDA DTPreferences=[0.9, 0.1]2025.12 | 0.537 | |
| BCPreferences=[0.9, 0.1]2025.12 | 0.536 | |
| PEDA DTPreferences=[1.0, 0.0]2025.12 | 0.529 | |
| BCPreferences=[1.0, 0.0]2025.12 | 0.529 | |
| CQLPreferences=[0.7, 0.3]2025.12 | 0.528 | |
| Clinician (Behavior)Info=Behavior2026.04 | 0.521 | |
| CQLPreferences=[0.8, 0.2]2025.12 | 0.52 | |
| CQLPreferences=[0.9, 0.1]2025.12 | 0.511 | |
| SBCQInfo=Model-free2026.04 | 0.508 | |
| CQLPreferences=[1.0, 0.0]2025.12 | 0.502 | |
| AI ClinicianInfo=Model-free2026.04 | 0.491 | |
| Continuous State-Space DDQNInfo=Model-free2026.04 | 0.483 | |
| BCQInfo=Model-free2026.04 | 0.458 | |
| CQLInfo=Model-free2026.04 | 0.418 | |
| Adaptive CPQLPreferences=[0.3, 0.7]2025.12 | 0.415 | |
| Adaptive CPQLPreferences=[0.8, 0.2]2025.12 | 0.381 | |
| Conditioned CPQLPreferences=[1.0, 0.0]2025.12 | 0.352 | |
| Conditioned CPQLPreferences=[0.4, 0.6]2025.12 | 0.349 | |
| Adaptive CPQLPreferences=[0.0, 1.0]2025.12 | 0.342 | |
| Conditioned CPQLPreferences=[0.3, 0.7]2025.12 | 0.341 | |
| Adaptive CPQLPreferences=[0.2, 0.8]2025.12 | 0.337 | |
| Adaptive CPQLPreferences=[1.0, 0.0]2025.12 | 0.325 | |
| Conditioned CPQLPreferences=[0.5, 0.5]2025.12 | 0.323 | |
| Adaptive CPQLPreferences=[0.1, 0.9]2025.12 | 0.32 | |
| Adaptive CPQLPreferences=[0.4, 0.6]2025.12 | 0.32 | |
| Conditioned CPQLPreferences=[0.2, 0.8]2025.12 | 0.292 | |
| Adaptive CPQLPreferences=[0.7, 0.3]2025.12 | 0.284 | |
| Conditioned CPQLPreferences=[0.6, 0.4]2025.12 | 0.271 | |
| Conditioned CPQLPreferences=[0.7, 0.3]2025.12 | 0.271 | |
| Adaptive CPQLPreferences=[0.6, 0.4]2025.12 | 0.261 | |
| Conditioned CPQLPreferences=[0.0, 1.0]2025.12 | 0.259 | |
| Adaptive CPQLPreferences=[0.5, 0.5]2025.12 | 0.255 | |
| Conditioned CPQLPreferences=[0.8, 0.2]2025.12 | 0.24 | |
| Conditioned CPQLPreferences=[0.1, 0.9]2025.12 | 0.238 | |
| Conditioned CPQLPreferences=[0.9, 0.1]2025.12 | 0.232 | |
| Adaptive CPQLPreferences=[0.9, 0.1]2025.12 | 0.232 |