Browse › Playing Games › Atari Games › Atari 2600 Video Pinball
Atari 2600 Video Pinball Benchmark (Atari Games)
The Atari 2600 Games task (and dataset) involves training an agent to achieve high game scores.
The archive carries no text for this table; the description above is the archive's text for the task Atari Games. archive 2025-07-28
Over time archive 2025-07-28
The chart needs JavaScript; the table below carries every value.
Direction inferred from the metric name, not from the archive: Score (higher is better). Points are placed at the row's paper date; 41 of 42 rows carry one.
Results archive 2025-07-28
Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.
| Paper | Code | Ran Syntology | Report | |||||
|---|---|---|---|---|---|---|---|---|
| 1 | R2D2 | 999383.2 | – | Paper | Code | 2019 | linked, not harvested | report |
| 2 | Agent57 | 992340.74 | – | Paper | Code | 2020 | 0 of 9 ran · 9 unverified | report |
| 3 | MuZero | 981791.88 | – | Paper | Code | 2019 | 43 of 64 ran · 21 unverified | report |
| 4 | GDI-H3 | 978190 | – | Paper | – | 2022 | no code linked | report |
| 5 | C51 noop | 949604.0 | – | Paper | Code | 2017 | 0 of 1 ran · 1 unverified | report |
| 6 | GDI-I3 | 925830 | – | Paper | – | 2022 | no code linked | report |
| 7 | NoisyNet-Dueling | 870954 | – | Paper | Code | 2017 | 1 of 3 ran · 2 unverified | report |
| 8 | MuZero (Res2 Adam) | 865543.44 | – | Paper | Code | 2021 | linked, not harvested | report |
| 9 | Bootstrapped DQN | 811610 | – | Paper | Code | 2016 | 1 of 2 ran · 1 unverified | report |
| 10 | QR-DQN-1 | 705662 | – | Paper | Code | 2017 | 2 of 2 ran · 0 unverified | report |
| 11 | IQN | 698045 | – | Paper | Code | 2018 | 0 of 2 ran · 2 unverified | report |
| 12 | ASL DDQN | 626794 | – | Paper | Code | 2023 | linked, not harvested | report |
| 13 | IMPALA (deep) | 572898.27 | – | Paper | Code | 2018 | 16 of 34 ran · 18 unverified | report |
| 14 | Ape-X | 565163.2 | – | Paper | Code | 2018 | 0 of 15 ran · 15 unverified | report |
| 15 | Advantage Learning | 543504 | – | Paper | Code | 2015 | linked, not harvested | report |
| 16 | DNA | 505392 | – | Paper | Code | 2022 | linked, not harvested | report |
| 17 | Prior+Duel noop | 479197.0 | – | Paper | Code | 2015 | 5 of 11 ran · 6 unverified | report |
| 18 | A3C LSTM hs | 470310.5 | – | Paper | Code | 2016 | 38 of 95 ran · 57 unverified | report |
| 19 | A2C + SIL | 461522.4 | – | Paper | Code | 2018 | 1 of 6 ran · 5 unverified | report |
| 20 | Prior+Duel hs | 447408.6 | – | Paper | Code | 2015 | 55 of 106 ran · 51 unverified | report |
| 21 | DDQN (tuned) hs | 367823.7 | – | Paper | Code | 2015 | 55 of 106 ran · 51 unverified | report |
| 22 | A3C FF hs | 331628.1 | – | Paper | Code | 2016 | 38 of 95 ran · 57 unverified | report |
| 23 | DDQN (tuned) noop | 309941.9 | – | Paper | Code | 2015 | 5 of 11 ran · 6 unverified | report |
| 24 | Prior hs | 295972.8 | – | Paper | Code | 2015 | 78 of 111 ran · 33 unverified | report |
| 25 | Prior noop | 282007.3 | – | Paper | Code | 2015 | 78 of 111 ran · 33 unverified | report |
| 26 | UCT | 254748 | – | Paper | Code | 2012 | 2 of 2 ran · 0 unverified | report |
| 27 | DQN noop | 196760.4 | – | Paper | Code | 2015 | 55 of 106 ran · 51 unverified | report |
| 28 | A3C FF (1 day) hs | 185852.6 | – | Paper | Code | 2016 | 38 of 95 ran · 57 unverified | report |
| 29 | DQN hs | 154414.1 | – | Paper | Code | 2015 | 55 of 106 ran · 51 unverified | report |
| 30 | Rational DQN Average | 149712 | – | Paper | Code | 2021 | 6 of 6 ran · 0 unverified | report |
| 31 | Gorila | 112093.4 | – | Paper | Code | 2015 | linked, not harvested | report |
| 32 | Duel hs | 110976.2 | – | Paper | Code | 2015 | 5 of 11 ran · 6 unverified | report |
| 33 | Duel noop | 98209.5 | – | Paper | Code | 2015 | 5 of 11 ran · 6 unverified | report |
| 34 | Recurrent Rational DQN Average | 86942 | – | Paper | Code | 2021 | 6 of 6 ran · 0 unverified | report |
| 35 | DDQN+Pop-Art noop | 56287.0 | – | Paper | – | 2016 | no code linked | report |
| 36 | Nature DQN | 42684.0 | – | Paper | Code | 2015 | linked, not harvested | report |
| 37 | DreamerV2 | 41860 | – | Paper | Code | 2020 | 3 of 15 ran · 12 unverified | report |
| 38 | POP3D | 37780.7 | – | Paper | Code | 2018 | linked, not harvested | report |
| 39 | CGP | 33752.4 | – | Paper | Code | 2018 | 0 of 9 ran · 9 unverified | report |
| 40 | ES FF (1 hour) noop | 22834.8 | – | Paper | Code | 2017 | 7 of 29 ran · 22 unverified | report |
| 41 | SARSA | 19761.0 | – | – | – | not matched | report | |
| 42 | Best Learner | 16871.3 | – | Paper | Code | 2012 | 2 of 2 ran · 0 unverified | report |
All 42 rows shown. 41 link to a paper page on this site; 0 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28
Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 30 rows have a graph line, from 19 distinct papers; 25 rows (14 papers) have at least one sample that ran. Counting each paper once: Syntology ran 258 of 522 samples; 264 unverified. Separately, 199 of those 522 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections