3v1 with keeper (3v2) on Google Research Football (GRF) (test)
0.025Mean Extrinsic RewardsELIGN_adv
Evaluation results
| Method | Links | ||
|---|---|---|---|
| ELIGN_advPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.025 | 0.001 | |
| CURIO_selfPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.024 | 0.004 | |
| ELIGN_teamPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.022 | 0.001 | |
| CURIO_teamPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.021 | 0.002 | |
| SPARSEPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.02 | 0.001 | |
| ELIGN_selfPartial observability=true, Decentralized training=true, Agt# v Adv#=3v22022.10 | 0.003 | 0.018 |