Reinforcement Learning on Cat Feeder
79.46Net Feeding RequestsAll-Pay
Evaluation Results
| Method | Links | |
|---|---|---|
| All-Pay2026.04 | 79.46 | |
| All-PayObservation Scope=Local2026.04 | 69.73 | |
| Winner-PaysObservation Scope=Local2026.04 | 67.21 | |
| Winner-PaysAttention Pooling=No2026.04 | 58.41 | |
| Winner-Pays2026.04 | 51.17 | |
| All-PayAttention Pooling=No2026.04 | 48.78 | |
| Single-Policy PPOReward Shaping Type=No Reward Shaping2026.04 | 48.75 | |
| Single-Policy PPOReward Shaping Type=Expiry Shaping2026.04 | 44.34 | |
| Single-Policy PPOReward Shaping Type=Nearest Shaping2026.04 | 22.17 | |
| DWN2026.04 | 2.28 |