Multi-agent Reinforcement Learning on Boxpushing
145.35RewardOurs
Evaluation Results
| Method | Links | |
|---|---|---|
| OursHorizon=72023.08 | 145.35 | |
| FM-EHorizon=72023.08 | 138.44 | |
| RNN-EHorizon=72023.08 | 132.7 | |
| OursHorizon=62023.08 | 98.76 | |
| OursHorizon=52023.08 | 98.73 | |
| FM-EHorizon=52023.08 | 96.4 | |
| FM-EHorizon=62023.08 | 94.61 | |
| RNN-EHorizon=62023.08 | 94.36 | |
| OursHorizon=42023.08 | 81.44 | |
| FM-EHorizon=42023.08 | 77.8 | |
| FM-EHorizon=32023.08 | 64.22 | |
| OursHorizon=32023.08 | 62.78 | |
| RNN-EHorizon=52023.08 | 21.78 | |
| RNN-EHorizon=42023.08 | 9.1 | |
| RNN-EHorizon=32023.08 | 8.4 |