Model card
Disco57.
Google DeepMindopen-sourceDiscoRL — meta-learned RL update rule (discovered by automated search)
RL algorithm discovered by automated meta-learning (DiscoRL). Published in Nature Oct 2025. Trained and evaluated on all 57 Atari games.
§ 01 · Benchmarks
Every benchmark Disco57 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | Atari 2600 | Reinforcement Learning · Atari Games | human-normalized-score | 1386.00 | #5 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where Disco57 actually performs.
§ 04 · Related models
Other Google DeepMind models scored on Codesota.
§ 05 · Sources & freshness
Where these numbers come from.
unknown
1
result
0 of 1 rows marked verified.