Fundamental limits to learning closed-form mathematical models from data

Oscar Fajardo-Fontiveros, Ignasi Reichardt, Harry R. De Los Ríos, Jordi Duch, Marta Sales-Pardo () and Roger Guimerà ()
Additional contact information
Oscar Fajardo-Fontiveros: Universitat Rovira i Virgili
Ignasi Reichardt: Universitat Rovira i Virgili
Harry R. De Los Ríos: Universitat Rovira i Virgili
Jordi Duch: Universitat Rovira i Virgili
Marta Sales-Pardo: Universitat Rovira i Virgili
Roger Guimerà: Universitat Rovira i Virgili

Nature Communications, 2023, vol. 14, issue 1, 1-9

Abstract: Abstract Given a finite and noisy dataset generated with a closed-form mathematical model, when is it possible to learn the true generating model from the data alone? This is the question we investigate here. We show that this model-learning problem displays a transition from a low-noise phase in which the true model can be learned, to a phase in which the observation noise is too high for the true model to be learned by any method. Both in the low-noise phase and in the high-noise phase, probabilistic model selection leads to optimal generalization to unseen data. This is in contrast to standard machine learning approaches, including artificial neural networks, which in this particular problem are limited, in the low-noise phase, by their ability to interpolate. In the transition region between the learnable and unlearnable phases, generalization is hard for all approaches including probabilistic model selection.

Date: 2023
References: View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.nature.com/articles/s41467-023-36657-z Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:14:y:2023:i:1:d:10.1038_s41467-023-36657-z

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-023-36657-z

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().