Reinforcement Learning on Atari 2600 RND-dominating games (test)
435.56PPO-normalized Mean ScoreEIPO-RND
Evaluation results
| Method | Links | ||
|---|---|---|---|
| EIPO-RND2022.11 | 435.56 | 1.13 | |
| Ext-norm RND2022.11 | 427.08 | 1.05 | |
| RND2022.11 | 384.57 | 1.22 | |
| Decay-RND2022.11 | 383.83 | 1.04 | |
| Decoupled-RND2022.11 | 1.54 | 1 |