Bibliographic record
Abstract
I describe two machine-learning models for elite classical chess: a result prediction model that outputs well-calibrated win/draw/loss probabilities for the current position, and a move prediction model that ranks legal candidate moves by how likely a strong human would play each one. Both are gradient-boosted decision-tree ensembles (LightGBM) built on features from Stockfish evaluations, an upstream human-imitation policy network, and a range of position- and game-level signals. The models are trained on ~464k classical chess games from The Week in Chess in which both players are rated 2400 Elo or higher, and evaluated on an 82k-game evaluation set held out from the training pool. On the result task the production model (which does not see the rating gap between the two players) reaches an expected calibration error of 0.002 on 6.78M held-out positions. On the move task the production model reaches 61.9% top-1 / 88.4% top-3 accuracy, versus 55.8% / 80.6% for an engine-best-move baseline evaluated on the same positions. Both models are deployed on chessds.com in two latency tiers (around 200 ms and 1 s per position on a single CPU core). I report scaling behavior, sliced metrics for both tasks, and a short ladder of toy baselines that situate the headline numbers against simpler alternatives.v2 (May 2026). Both models retrained on a substantially larger and cleaner dataset: all issues of The Week in Chess (≈464k training / 82k evaluation games, up from 266k/47k), with removal of mislabeled fast-time-control events and duplicate games. All figures, tables, and metrics reflect the current production models; scaling curves regenerated under production settings. Headline results improved (result-model ECE ≈0.002 on 6.78M held-out positions; move model 61.9% top-1 / 88.4% top-3). No methodological or structural changes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".