Overcooked-AI
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multi-agent coordination | Overcooked-AI Pipeline 2-agent | Mean Reward307.7 | 12 | |
| Multi-agent coordination | Overcooked-AI | Coordination Ring Score122.3 | 10 | |
| Multi-agent coordination | Overcooked-AI Forced Coordination (4-agent) | Mean Reward387.5 | 6 | |
| Multi-agent coordination | Overcooked-AI Asymmetric Advantages 4-agent | Mean Reward253.4 | 6 | |
| Multi-agent coordination | Overcooked-AI Cramped Room (3-agent) | Mean Reward504.3 | 6 | |
| Multi-agent coordination | Overcooked-AI Open Room (3-agent) | Mean Reward379.8 | 6 | |
| Multi-agent coordination | Overcooked-AI Forced Coordination (3-agent) | Mean Reward337.6 | 6 | |
| Multi-agent coordination | Overcooked-AI Asymmetric Advantages (3-agent) | Mean Reward200.8 | 6 | |
| Multi-agent coordination | Overcooked-AI Coordination Ring 2-agent | Mean Reward276.2 | 6 | |
| Multi-agent coordination | Overcooked-AI Cramped Room (2-agent) | Mean Reward314 | 6 | |
| Multi-agent coordination | Overcooked-AI Asymmetric Advantages (2-agent) | Mean Reward596 | 6 | |
| Human-AI Collaboration | Overcooked-AI (Evaluation partner population (novel AI behaviors)) | Cramped Room Score165.8 | 4 | |
| Cooperative Multi-Agent Coordination | Overcooked-AI Forced Coordination | Total Mean Reward44.25 | 4 | |
| Cooperative Multi-Agent Coordination | Overcooked-AI Counter Circuit | Total Mean Reward60.42 | 4 | |
| Cooperative Multi-Agent Coordination | Overcooked-AI Coordination Ring | Total Mean Reward115.42 | 4 | |
| Cooperative Multi-Agent Coordination | Overcooked-AI Asymmetric Advantages | Mean Reward140.63 | 4 | |
| Cooperative Multi-Agent Coordination | Overcooked-AI Cramped Room | Total Mean Reward172.96 | 4 | |
| Cooperative Multi-Agent Reinforcement Learning | Overcooked-AI Forced Coordination (test) | Mean Reward35.43 | 4 | |
| Cooperative Multi-Agent Reinforcement Learning | Overcooked-AI Counter Circuit (test) | Mean Reward43.65 | 4 | |
| Cooperative Multi-Agent Reinforcement Learning | Overcooked-AI Coordination Ring (test) | Mean Reward97.35 | 4 | |
| Cooperative Multi-Agent Reinforcement Learning | Overcooked-AI Asymmetric Advantages (test) | Mean Reward124.39 | 4 | |
| Cooperative Multi-Agent Reinforcement Learning | Overcooked-AI Cramped Room (test) | Mean Reward148.76 | 4 | |
| Cooperative Cooking | Overcooked-AI Cramped Room | Total Mean Reward150 | 4 | |
| Cooperative Cooking | Overcooked-AI Asymmetric Advantages | J Score381 | 3 | |
| Cooperative Cooking | Overcooked-AI Coordination Ring | J-Score153 | 3 |
Showing 25 of 26 rows