Hanabi Evaluation on Full-scale (2-players)
24Average ReturnIPPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| IPPOEnvironment steps=10B, Evaluation games=10k, Number of seeds=>= 32021.03 | 24 | 24.19 | |
| MAPPOEnvironment steps=10B, Evaluation games=10k, Number of seeds=>= 32021.03 | 23.89 | 24.23 | |
| SADEnvironment steps=10B, Evaluation games=10k, Number of seeds=>= 32021.03 | 23.87 | 24.01 | |
| VDNEnvironment steps=10B, Evaluation games=10k, Number of seeds=>= 32021.03 | 23.83 | 23.96 |