Multi-Agent Reinforcement Learning on SMAC 8mvs8m to 8mvs9m
0.99Steps to 80% RewardASALT
Evaluation results
| Method | Links | |
|---|---|---|
| ASALTScale=10^6 environment steps2026.06 | 0.99 | |
| MALTScale=10^6 environment steps2026.06 | 2.41 | |
| EPCScale=10^6 environment steps2026.06 | 3.02 | |
| Distilled PolicyScale=10^6 environment steps2026.06 | 4.28 | |
| PSMARLScale=10^6 environment steps2026.06 | 5.23 | |
| Fine-tuneScale=10^6 environment steps2026.06 | 5.43 | |
| BaselineScale=10^6 environment steps2026.06 | 5.84 |