Reinforcement Learning on Atari 2600 Solaris ALE (test)
5,501.5Game ScoreDQN-PixelCNN
Evaluation results
| Method | Links | |
|---|---|---|
| DQN-PixelCNNTraining frames=150M, Seeds=32017.03 | 5,501.5 | |
| DQNTraining frames=150M, Seeds=32017.03 | 3,482.8 | |
| DQN-CTSTraining frames=150M, Seeds=32017.03 | 2,863.6 | |
| A3C-CTSTraining frames=150M, Seeds=32017.03 | 2,270.15 | |
| Prior. DuelTraining frames=150M, Seeds=32017.03 | 133.4 |