Multi-Agent Reinforcement Learning on SMAC 8m
100Win RateDG-MAPPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DG-MAPPODifficulty=Easy, Sight Range=92026.01 | 100 | 1.4 | |
| HAPPODifficulty=Easy, Sight Range=92026.01 | 97.5 | 1.1 | |
| MAT-DecDifficulty=Easy, Sight Range=92026.01 | 97.2 | 2.5 | |
| MAPPODifficulty=Easy, Sight Range=92026.01 | 96.8 | 2.9 | |
| SeqComm-DFLTraining Iterations=10M2026.04 | 59 | — | |
| SeqCommTraining Iterations=10M2026.04 | 45 | — | |
| OMDTraining Iterations=10M2026.04 | 31 | — |