AAMAS 2026 Competition Leaderboard#
This leaderboard mirrors the four competition criteria described in Evaluating Your Agent.
Note
Public rankings are populated from organizer-run evaluations. The tables below will be filled as scored submissions are verified and posted.
Adaptability#
un-notify
| Rank | GitHub ID | Approach (3 experiments) | Threshold Success | Convergence Time | Local Stability |
|---|---|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 0.711111 | 0.332233 | 0.000066 |
| 2 | pudovikoff | MLE-Replan/DQN/SAC | 0.633333 | 0.366667 | 0.366732 |
| 3 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 0.666667 | 0.333333 | 0.111177 |
| 4 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 0.666667 | 0.333333 | 0.333400 |
| 5 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 0.666667 | 0.333333 | 0.333400 |
Performance#
| Rank | GitHub ID | Approach (3 experiments) | Average return |
|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 2257.240 |
| 2 | pudovikoff | MLE-Replan/DQN/SAC | 1800.111 |
| 3 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 1734.251 |
| 4 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 568.996 |
| 5 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 567.229 |
| Rank | GitHub ID | Approach (3 experiments) | Average return |
|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 2197.415 |
| 2 | pudovikoff | MLE-Replan/DQN/SAC | 1800.111 |
| 3 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 1734.251 |
| 4 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 568.996 |
| 5 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 567.229 |
Resilience#
fully-notify
| Rank | GitHub ID | Approach (3 experiments) | Immediate Drop | Recovery Time |
|---|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 0.333333 | 0.332727 |
| 2 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 0.333333 | 0.333333 |
| 3 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 0.333333 | 0.333333 |
| 4 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 0.333333 | 0.333333 |
| 5 | pudovikoff | MLE-Replan/DQN/SAC | 0.366667 | 0.366667 |
Efficiency#
| Rank | GitHub ID | Approach (3 experiments) | Efficiency |
|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 0.00001430 |
| 2 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 0.00001862 |
| 3 | pudovikoff | MLE-Replan/DQN/SAC | 0.00005887 |
| 4 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 0.00253916 |
| 5 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 0.08414263 |
| Rank | GitHub ID | Approach (3 experiments) | Efficiency |
|---|---|---|---|
| 1 | Jazhyc | Context-Aware PPO/PPO/PPO | 0.00000918 |
| 2 | Kumar20-21 | Dynamic VI/LQR/QR-SAC | 0.00001913 |
| 3 | pudovikoff | MLE-Replan/DQN/SAC | 0.00005943 |
| 4 | micahr234 | MOUSE+MLP+no history/MOUSE/- | 0.00323941 |
| 5 | micahr234 | MOUSE+sequence model+history/MOUSE/- | 0.08204777 |