Reinforcement Learning on Atari 2600 Solaris
2,279.4Average ScoreDQN MMC PIXELCNN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DQN MMC PIXELCNNTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 2,279.4 | — | |
| DQN MMCTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 2,244.6 | 378.8 | |
| DQN MMC + SRTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 1,890.1 | 163.1 | |
| RUDDERDelay (frames)=122, Delay-event=navigate map, Training game frames=200M, Number of random seeds=3, Trials per seed=10, Frame skip=42018.06 | 1,827 | — | |
| RNDTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 1,270.3 | 291 | |
| DQN MMC CTSTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 1,147.1 | — | |
| DQNTraining frames=100 million, Sticky actions (s)=0.25, Frame skip=52018.07 | 783.4 | 55.3 | |
| PPODelay (frames)=122, Delay-event=navigate map, Training game frames=200M, Number of random seeds=3, Trials per seed=10, Frame skip=42018.06 | 616 | — |