Multi-Agent Reinforcement Learning on MPE PP_3/1
202.9Average Cumulative RewardWALL
Evaluation results
| Method | Links | |
|---|---|---|
| WALLAttack Method=Natural2025.02 | 202.9 | |
| WALLAttack Method=Radom Attack2025.02 | 202.9 | |
| RANDOMAttack Method=Natural2025.02 | 178.3 | |
| ROMANCEAttack Method=Natural2025.02 | 175 | |
| ROMANCEAttack Method=Radom Attack2025.02 | 173.4 | |
| WALLAttack Method=Wolfpack Adversarial Attack2025.02 | 171.5 | |
| RANDOMAttack Method=Radom Attack2025.02 | 170.1 | |
| ROMANCEAttack Method=EGA2025.02 | 167.1 | |
| WALLAttack Method=EGA2025.02 | 166.5 | |
| Vanilla QMIXAttack Method=Natural2025.02 | 165.4 | |
| ROMANCEAttack Method=Wolfpack Adversarial Attack2025.02 | 159.3 | |
| Vanilla QMIXAttack Method=Radom Attack2025.02 | 158.9 | |
| RANDOMAttack Method=EGA2025.02 | 157.6 | |
| RANDOMAttack Method=Wolfpack Adversarial Attack2025.02 | 157.5 | |
| Vanilla QMIXAttack Method=EGA2025.02 | 153.7 | |
| Vanilla QMIXAttack Method=Wolfpack Adversarial Attack2025.02 | 135.7 |