Multi-Agent Reinforcement Learning on Ising Model N=10,000 (Mixed)
85.2Normalized ReturnMF-Diffuser
Evaluation Results
| Method | Links | |
|---|---|---|
| MF-Diffuser2026.05 | 85.2 | |
| MF-CDMs-RL2026.05 | 77.5 | |
| MFQ-Offline2026.05 | 76.3 | |
| Oryx2026.05 | 76 | |
| DoF2026.05 | 68.1 | |
| Indep. Diffuser2026.05 | 64.2 | |
| MA-TD3+BC2026.05 | 8.6 | |
| OMAR2026.05 | 7.3 | |
| MADiffattention=chunked-attention (max 12-agent windows)2026.05 | 6.8 | |
| Joint Diffuserattention=chunked-attention (block size 32)2026.05 | 4.2 |