Multi-Agent Reinforcement Learning on SMAC 3svs3z to 3svs4z
0.62Environment Steps to 80% RewardASALT
Evaluation results
| Method | Links | |
|---|---|---|
| ASALTScale=10^5 environment steps2026.06 | 0.62 | |
| MALTScale=10^5 environment steps2026.06 | 1.17 | |
| EPCScale=10^5 environment steps2026.06 | 1.29 | |
| Distilled PolicyScale=10^5 environment steps2026.06 | 2.59 | |
| Fine-tuneScale=10^5 environment steps2026.06 | 2.91 | |
| PSMARLScale=10^5 environment steps2026.06 | 3.2 | |
| BaselineScale=10^5 environment steps2026.06 | 3.72 |