Reinforcement Learning on Atari 2600 (test) (9 Game Subset)
16,218Alien ScoreMuesli
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Mueslitraining_frames=40M2021.04 | 16,218 | 524 | 143,898 | 2,344 | 10,919 | 15,195 | 19,244 | 30,937 | 142,431 | |
| ATreeC-1training_frames=40M, variant=12021.04 | 3,448 | 1,578 | 102,546 | 678 | 1,035 | 8,227 | 4,866 | 25,159 | 1,734 | |
| ATreeC-2training_frames=40M, variant=22021.04 | 2,813 | 1,566 | 110,712 | 649 | 281 | 8,134 | 4,450 | 25,459 | 2,176 | |
| TreeQN-2training_frames=40M, variant=22021.04 | 2,497 | 1,170 | 104,932 | 825 | 581 | 11,035 | 3,277 | 15,970 | 8,241 | |
| TreeQN-1training_frames=40M, variant=12021.04 | 2,321 | 1,030 | 107,983 | 800 | 2,254 | 10,836 | 3,030 | 15,688 | 9,302 |