Multi-Agent Reinforcement Learning on SMAC Medium v2
37.5Win Rate - Terran (3 Units)STAIRS
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| STAIRSDifficulty=Medium2026.03 | 37.5 | 26.9 | 36.9 | 35.6 | 26.9 | 35 | 38.8 | 35.6 | 37.5 | 38.8 | 35 | 28.1 | 39.4 | 31.3 | 38.1 | 40 | 41.3 | 36.9 | 31.3 | 35.6 | 23.8 | 34.6 | 33.1 | 28.8 | 31.9 | 33.8 | 26.9 | 28.8 | 35 | 31.9 | 23.8 | 28.8 | 30.3 | |
| HiSSDDifficulty=Medium2026.03 | 31.3 | 18.8 | 22.5 | 33.8 | 26.9 | 35.6 | 26.9 | 26.9 | 20.6 | 23.1 | 26.6 | 30.6 | 42.5 | 27.5 | 35 | 35 | 32.5 | 25.6 | 40 | 33.8 | 20 | 32.3 | 28.1 | 15.6 | 20 | 18.1 | 25 | 26.3 | 25 | 23.1 | 25 | 26.3 | 23.3 | |
| ODISDifficulty=Medium2026.03 | 18.1 | 16.3 | 15 | 19.4 | 14.4 | 17.5 | 20 | 12.5 | 6.3 | 7.5 | 14.7 | 14.4 | 9.4 | 11.9 | 13.1 | 10.6 | 8.8 | 15 | 13.1 | 11.3 | 8.8 | 11.6 | 13.1 | 11.3 | 11.9 | 15 | 11.9 | 12.5 | 8.8 | 15 | 11.3 | 11.3 | 12.2 | |
| UPDeT-mDifficulty=Medium2026.03 | 15 | 16.3 | 10.6 | 11.9 | 10.6 | 10.6 | 6.9 | 5 | 2.5 | 3.1 | 9.3 | 16.9 | 13.1 | 10 | 20 | 12.5 | 10.6 | 14.4 | 13.8 | 8.1 | 3.8 | 12.3 | 11.3 | 10.6 | 6.3 | 11.9 | 8.8 | 8.8 | 4.4 | 4.4 | 2.5 | 1.9 | 7.1 |