Reinforcement Learning on Atari 2600 Freeway ALE (test)
34ScoreA2C+SIL
Evaluation Results
| Method | Links | |
|---|---|---|
| A2C+SILSelf-Imitation Learning=true2018.06 | 34 | |
| Prior. DuelTraining frames=150M, Seeds=32017.03 | 33 | |
| TRPO-AE-SimHashExploration=SimHash2018.06 | 33 | |
| PPO2023.06 | 32.5 | |
| Reactor-PixelCNNExploration=PixelCNN2018.06 | 32 | |
| Expert2023.06 | 31.9 | |
| DQN-CTSTraining frames=150M, Seeds=32017.03 | 31.7 | |
| DQN-PixelCNNTraining frames=150M, Seeds=32017.03 | 31.7 | |
| DQNTraining frames=150M, Seeds=32017.03 | 30.8 | |
| A3C-CTSTraining frames=150M, Seeds=32017.03 | 30.48 | |
| A3C-CTSExploration=CTS2018.06 | 30 | |
| STG2023.06 | 21.8 | |
| ELE2023.06 | 2.7 | |
| GAIfO2023.06 | 0.6 |