Goal-navigation on Multi Rooms custom (Room 5)
-26ReturnROVER
Evaluation Results
| Method | Links | |
|---|---|---|
| ROVERAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -26 | |
| CICAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -143.4 | |
| DDPG (online)Training protocol=50k interactions2026.06 | -221.3 | |
| ROVERAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -221.7 | |
| RandomAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -300 | |
| APTAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -300 | |
| SMMAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -300 | |
| CICAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -300 | |
| RNDAlgorithm=CQL, Training protocol=15k gradient updates2026.06 | -300 | |
| RandomAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -300 | |
| APTAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -300 | |
| SMMAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -300 | |
| RNDAlgorithm=DDPG, Training protocol=15k gradient updates2026.06 | -300 |