Offline Reinforcement Learning on Battle Medium-Replay N=1000
71.3Normalized ReturnMF-Diffuser
Evaluation Results
| Method | Links | |
|---|---|---|
| MF-DiffuserN=10002026.05 | 71.3 | |
| MF-CDMs-RLN=10002026.05 | 63.5 | |
| OryxN=10002026.05 | 60.2 | |
| MFQ-OfflineN=10002026.05 | 56.3 | |
| MA-TD3+BCN=10002026.05 | 51.2 | |
| DoFN=10002026.05 | 48.2 | |
| OMARN=10002026.05 | 46.7 | |
| Indep. DiffuserN=10002026.05 | 44.1 | |
| MADiffN=10002026.05 | 36.6 | |
| Joint DiffuserN=10002026.05 | 31.4 |