Policy Optimization on Office World MAP4
5,630Average Training StepsQR-MAXRM
Evaluation Results
| Method | Links | |
|---|---|---|
| QR-MAXRMExp=EXP6, Map Size=15x152025.12 | 5,630 | |
| QR-MAXExp=EXP6, Map Size=15x152025.12 | 20,150 | |
| UCBVI-sBExp=EXP6, Map Size=15x152025.12 | 80,160 | |
| R-MAXExp=EXP6, Map Size=15x152025.12 | 82,000 | |
| UCBVI-BExp=EXP6, Map Size=15x152025.12 | 90,030 | |
| UCBVI-HExp=EXP6, Map Size=15x152025.12 | 101,120 | |
| QRMExp=EXP6, Map Size=15x152025.12 | 9,910,340 |