Multi-Agent Reinforcement Learning on Ising Model Medium N=10,000
94.5Normalized ReturnMF-Diffuser
Evaluation results
| Method | Links | |
|---|---|---|
| MF-Diffuser2026.05 | 94.5 | |
| MF-CDMs-RL2026.05 | 85.4 | |
| MFQ-Offline2026.05 | 84.6 | |
| Oryx2026.05 | 84 | |
| DoF2026.05 | 78.2 | |
| Indep. Diffuser2026.05 | 73.5 | |
| MA-TD3+BC2026.05 | 16.2 | |
| OMAR2026.05 | 14.5 | |
| MADiffattention=chunked-attention (max 12-agent windows)2026.05 | 12.5 | |
| Joint Diffuserattention=chunked-attention (block size 32)2026.05 | 8.1 |