Multi-Agent Reinforcement Learning on Simple Spread N=6
0.0107CollisionsMAPPO
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MAPPOlocal_ratio=0.02026.06 | 0.0107 | 0.0007 | 2.82 | |
| MAPPOlocal_ratio=0.32026.06 | 0.0174 | 0.0012 | 2.88 | |
| MAPPOlocal_ratio=0.62026.06 | 0.0183 | 0.0012 | 7.23 | |
| MAPPO-Lagcost_limit=0.12026.06 | 0.0188 | 0.0013 | 4.48 | |
| QMIXlocal_ratio=0.02026.06 | 0.0217 | 0.0014 | 1.85 | |
| QMIXlocal_ratio=0.32026.06 | 0.0243 | 0.0016 | 1.8 | |
| MAPPO-Lagcost_limit=1.02026.06 | 0.031 | 0.0021 | 7.37 | |
| QMIXlocal_ratio=0.62026.06 | 0.0313 | 0.0021 | 2.88 | |
| IQLlocal_ratio=0.62026.06 | 0.0802 | 0.0053 | 12.87 | |
| IQLlocal_ratio=0.32026.06 | 0.096 | 0.0064 | 14.21 | |
| IQLlocal_ratio=0.02026.06 | 0.1058 | 0.0071 | 15.83 | |
| DCGlocal_ratio=0.02026.06 | 0.1425 | 0.0095 | 24.17 | |
| DCGlocal_ratio=0.32026.06 | 0.1458 | 0.0097 | 24 | |
| DCGlocal_ratio=0.62026.06 | 0.1544 | 0.0103 | 24.17 | |
| CG-CMARLlambda=10.00, Pareto-optimal=true2026.06 | 1.2 | 0.08 | 7.33 | |
| CG-CMARLlambda=5.00, Pareto-optimal=true2026.06 | 1.53 | 0.1 | 11.75 | |
| CG-CMARLlambda=2.00, Pareto-optimal=true2026.06 | 1.72 | 0.11 | 22.38 | |
| CG-CMARLlambda=1.00, Pareto-optimal=true2026.06 | 2.89 | 0.19 | 28.67 | |
| CG-CMARLlambda=0.50, Pareto-optimal=true2026.06 | 2.99 | 0.2 | 33.21 | |
| CG-CMARLlambda=0.10, Pareto-optimal=true2026.06 | 3.9 | 0.26 | 34.83 | |
| CG-CMARLlambda=0.00, Pareto-optimal=false2026.06 | 4.13 | 0.28 | 33.29 | |
| CG-CMARLlambda=0.20, Pareto-optimal=false2026.06 | 4.28 | 0.29 | 33.92 | |
| CG-CMARLlambda=0.05, Pareto-optimal=false2026.06 | 4.57 | 0.3 | 33.75 |