MPE
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multi-agent Reinforcement Learning | MPE Cooperative Navigation (CN) v1 (Expert) | Normalized Score126.3 | 19 | |
| Multi-Agent Reinforcement Learning | MPE Speaker-Listener | Return27.9 | 17 | |
| Multi-Agent Reinforcement Learning (Predator-Prey) | MPE PP_9/3 | Average Cumulative Reward802.5 | 16 | |
| Multi-Agent Reinforcement Learning (Predator-Prey) | MPE PP_6/2 | Average Cumulative Reward685.5 | 16 | |
| Multi-Agent Reinforcement Learning (Predator-Prey) | MPE PP_3/1 | Average Cumulative Reward202.9 | 16 | |
| Multi-agent Offline Reinforcement Learning | MPE CN (Medium-replay) | Score95.4 | 16 | |
| Multi-agent Offline Reinforcement Learning | MPE CN (Random) | Score88.3 | 16 | |
| Adversarial Attack | MPE spread | Reward Score-546.99 | 12 | |
| Multi-Agent Reinforcement Learning | MPE Adversary | Return19.1 | 11 | |
| Multi-Agent Reinforcement Learning | MPE Simple Spread | Return-46 | 11 | |
| Multi-agent Reinforcement Learning | MPE Predator-prey (PP) v1 (Expert) | Normalized Score118.2 | 10 | |
| Multi-agent Reinforcement Learning | MPE Predator-prey (PP) v1 (Med-Rep) | Normalized Score71.1 | 10 | |
| World | MPE Random | Normalized Score141.1 | 9 | |
| World | MPE Expert | Normalized Score163.9 | 9 | |
| Predator Prey | MPE Random | Normalized Score133.9 | 9 | |
| Predator Prey | MPE Medium | Normalized Score137.1 | 9 | |
| Predator Prey | MPE Expert | Normalized Score161.4 | 9 | |
| Cooperative Navigation | MPE Random | Normalized Score69.8 | 9 | |
| Cooperative Navigation | MPE Medium | Normalized Score70.1 | 9 | |
| Offline Multi-Agent Reinforcement Learning | MPE World (Random) | Average Normalized Score94.3 | 8 | |
| Multi-agent Reinforcement Learning | MPE Simple-World (Random) | Average Normalized Score15.8 | 6 | |
| Multi-agent Reinforcement Learning | MPE Simple-World Medium-Reply | Average Normalized Score50.9 | 6 | |
| Multi-Agent Reinforcement Learning | MPE Adversary (test) | Final Test Return62.9 | 6 | |
| Multi-Agent Reinforcement Learning | MPE Reference (test) | Final Test Return39.6 | 6 | |
| Attack Detection | MPE reference | F1 Score92 | 5 |
Showing 25 of 36 rows