EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

(Exhaustive) symbolic regression and model selection by minimum description length

Harry Desmond iD

Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · 09.04.2026 · 384 (2317)

DOI: 10.1098/rsta.2024.0584 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, online erschienen am 09.04.2026. Verfasst von Harry Desmond. Beteiligte Einrichtung: Institute of Cosmology and Gravitation, University of Portsmouth, Portsmouth, UK. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Royal Society.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Symbolic regression (SR) is the machine learning (ML) method for learning functions from data. After a brief overview of the SR landscape, I will describe the two main challenges that traditional algorithms face: they have an unknown (and probably significant) probability of failing to find any given good function, and they suffer from ambiguity and poorly justified assumptions in their function-selection procedure. To address these, I propose an exhaustive search and model selection by the minimum description length (MDL) principle, which allows accuracy and complexity to be directly traded off by measuring each in units of information. I showcase the resulting publicly available Exhaustive Symbolic Regression (ESR) algorithm on three open problems in astrophysics: the expansion history of the universe, the effective behaviour of gravity in galaxies and the potential of the inflaton field. In each case, the algorithm identifies many functions superior to the literature standards. This general-purpose methodology should find widespread utility in science and beyond. This article is part of the discussion meeting issue ‘Symbolic regression in the physical sciences’.

Schlagwörter

Stichwörter der Autor:innen

machine learningminimum description lengthsymbolic regression

Autor:innen und Institutionen

  • Institute of Cosmology and Gravitation, University of Portsmouth, Portsmouth, UK.

ORCID-Profile: Harry Desmond

Förderung

  • Royal Society

Zitieren

APA

Desmond, H. (2026). (Exhaustive) symbolic regression and model selection by minimum description length. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 384(2317). https://doi.org/10.1098/rsta.2024.0584

Vancouver

Desmond H. (Exhaustive) symbolic regression and model selection by minimum description length. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2026;384(2317). doi:10.1098/rsta.2024.0584

BibTeX
@article{Desmond2026,
  author = {Desmond, Harry},
  title = {(Exhaustive) symbolic regression and model selection by minimum description length},
  journal = {Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences},
  year = {2026},
  volume = {384},
  number = {2317},
  doi = {10.1098/rsta.2024.0584}
}
Kostenlos starten →