Multi-agent reinforcement learning on SMAC 5v5 v2
17.4Protoss RewardFULLOBS
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| FULLOBSAlgorithm=MAPPO2026.06 | 17.4 | 243.92 | 16.4 | 187.55 | 14.9 | 138.97 | |
| FULLOBSAlgorithm=IPPO2026.06 | 17.1 | 247.77 | 17.2 | 195.13 | 14.3 | 138.86 | |
| HICOMMAlgorithm=MAPPO2026.06 | 16.7 | 10.98 | 15.6 | 6.83 | 13.9 | 4.25 | |
| HICOMMAlgorithm=IPPO2026.06 | 16.3 | 9.77 | 16.4 | 7.95 | 13.5 | 4.85 | |
| CACOMAlgorithm=MAPPO2026.06 | 16 | 46.74 | 14.7 | 36.97 | 13.1 | 24.2 | |
| CACOMAlgorithm=IPPO2026.06 | 15.6 | 52.67 | 15.6 | 34.84 | 12.3 | 19.76 | |
| T2MACAlgorithm=MAPPO2026.06 | 15.6 | 168.93 | 14.4 | 155.82 | 12.7 | 94.71 | |
| T2MACAlgorithm=IPPO2026.06 | 15.2 | 201.26 | 15.3 | 129.74 | 12.3 | 76.28 | |
| PARTIALOBSAlgorithm=MAPPO2026.06 | 14.8 | 0 | 13.6 | 0 | 12.3 | 0 | |
| PARTIALOBSAlgorithm=IPPO2026.06 | 14.5 | 0 | 14.4 | 0 | 11.4 | 0 |