Multi-Agent Reinforcement Learning on Ising Model Med-Replay N=10,000
88.8Normalized ReturnMF-Diffuser
Evaluation Results
| Method | Links | |
|---|---|---|
| MF-Diffuser2026.05 | 88.8 | |
| MFQ-Offline2026.05 | 82.1 | |
| MF-CDMs-RL2026.05 | 81.2 | |
| Oryx2026.05 | 80.2 | |
| DoF2026.05 | 72.5 | |
| Indep. Diffuser2026.05 | 68.4 | |
| MA-TD3+BC2026.05 | 11.8 | |
| OMAR2026.05 | 10.2 | |
| MADiffattention=chunked-attention (max 12-agent windows)2026.05 | 9.2 | |
| Joint Diffuserattention=chunked-attention (block size 32)2026.05 | 5.9 |