| Paper | Code | Human World Record Breakthrough | ModelName | ReleaseDate |
|---|---|---|---|---|
| Generalized Data Distribution Iteration | 22 | GDI-H3 | 2022-06-07 | |
| Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model | ✓ Link | 19 | Muzero | 2019-11-19 |
| Agent57: Outperforming the Atari Human Benchmark | ✓ Link | 18 | Agent57 | 2020-03-30 |
| Go-Explore: a New Approach for Hard-Exploration Problems | ✓ Link | 17 | Go-Explore | 2019-01-30 |
| Generalized Data Distribution Iteration | 17 | GDI-I3 | 2022-06-07 | |
| Recurrent Experience Replay in Distributed Reinforcement Learning | ✓ Link | 15 | R2D2 | 2019-05-01 |
| Never Give Up: Learning Directed Exploration Strategies | ✓ Link | 8 | NGU | 2020-02-14 |
| Muesli: Combining Improvements in Policy Optimization | ✓ Link | 5 | Muesli | 2021-04-13 |
| Rainbow: Combining Improvements in Deep Reinforcement Learning | ✓ Link | 4 | Rainbow | 2017-10-06 |