Communication-based Multi-agent Reinforcement Learning (com-MARL) on RTGA Clean
50.8Average Winning RateSHINE
Evaluation results
| Method | Links | |
|---|---|---|
| SHINE2026.05 | 50.8 | |
| Original2026.05 | 49.9 | |
| BehaviorGuard2026.05 | 49.4 | |
| Direct retraining2026.05 | 48.9 | |
| BIRD2026.05 | 44.4 |