Metraipolis › archive › dc_223f98aafcd6

ExaTune: Exhaustive Hyperparameter Landscape Exploration at Scale — arXiv:2611.03867

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03867
License (as archived)
CC0-1.0
Archived
2026-11-04T18:53:49.532Z
SHA-256 of the archived text
04fe991cdad8993ffd2d7de4701598778b4e28430bdc840ddd759f2e59dd3f8e
Stable URL of this page
https://metraipolis.com/documento/dc_223f98aafcd6

How to cite

ExaTune: Exhaustive Hyperparameter Landscape Exploration at Scale — arXiv:2611.03867 — arXiv (en), https://arxiv.org/abs/2611.03867 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_223f98aafcd6 (Metraipolis archive dc_223f98aafcd6, sha256 04fe991cdad8993ffd2d7de4701598778b4e28430bdc840ddd759f2e59dd3f8e)
BibTeX
@misc{dc223f98aafcd6,
  title = {ExaTune: Exhaustive Hyperparameter Landscape Exploration at Scale — arXiv:2611.03867},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03867},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_223f98aafcd6; sha256 04fe991cdad8993ffd2d7de4701598778b4e28430bdc840ddd759f2e59dd3f8e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03867. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Hyperparameter optimization (HPO) returns a single optimal configuration, systematically discarding the explored performance landscape that encodes how results depend on configuration choices. Fitness landscape analysis comprises a robust collection of methods from evolutionary computation offering ruggedness indices, autocorrelation functions, and fitness-distance correlation that have never been applied to machine learning hyperparameter spaces at scale, given that fine-grained ground-truth enumeration data has not existed. We present ExaTune, an open-source HPC system to produce a new artifact class: the hyperparameter landscape map. ExaTune exhaustively evaluates every point on a discretized hyperparameter grid via $K$-fold cross-validation, aggregates results into reproducible parquet benchmarks, and applies a suite of five formalized landscape metrics: spatial autocorrelation (Moran's $I$), ruggedness, local optima density, fitness-distance correlation (FDC), and variance-based parameter importance. We release a benchmark collection of 8 complete landscape maps (totalling 285,080 exhaustively evaluated configurations) spanning five model families (SVMs, random forests, XGBoost, ridge regression, and $k$-nearest neighbors) across six classification and two regression benchmarks, hosted on HuggingFace. Empirical analysis demonstrates that Bayesian optimization, Optuna, Hyperband, and Random Search at standard trial budgets systematically miss instability regions, secondary optima, and parameter interaction effects that are visible in complete landscapes; information that is structurally unavailable to any optimizer by design.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.03867 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T18:35:37.401Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).