Reducing Overparameterization of Symbolic Regression Models with Equality Saturation

Fabricio Olivetti De Franca, Gabriel Kronberger

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

Abstract

Overparameterized models in regression analysis are often harder to interpret and can be harder to fit because of ill-conditioning. Genetic programming is prone to overparameterized models as it evolves the structure of the model without taking the location of parameters into account. One way to alleviate this is rewriting the expression and merging the redundant fitting parameters. In this paper we propose the use of equality saturation to alleviate overparameterization. We first notice that all the tested GP implementations suffer from overparameterization to different extents and then show that equality saturation together with a small set of rewriting rules is capable of reducing the number of fitting parameters to a minimum with a high probability. Compared to one of the few available alternatives, Sympy, it produces much better and consistent results. These results lead to different possible future investigations such as the simplification of expressions during the evolutionary process, and improvement of the interpretability of symbolic models.

Original languageEnglish
Title of host publicationGECCO 2023 - Proceedings of the 2023 Genetic and Evolutionary Computation Conference
PublisherAssociation for Computing Machinery, Inc
Pages1064-1072
Number of pages9
ISBN (Electronic)9798400701191
DOIs
Publication statusPublished - 15 Jul 2023
Event2023 Genetic and Evolutionary Computation Conference, GECCO 2023 - Lisbon, Portugal
Duration: 15 Jul 202319 Jul 2023

Publication series

NameGECCO 2023 - Proceedings of the 2023 Genetic and Evolutionary Computation Conference

Conference

Conference2023 Genetic and Evolutionary Computation Conference, GECCO 2023
Country/TerritoryPortugal
CityLisbon
Period15.07.202319.07.2023

Keywords

  • equality saturation
  • genetic programming
  • simplification
  • symbolic regression

Fingerprint

Dive into the research topics of 'Reducing Overparameterization of Symbolic Regression Models with Equality Saturation'. Together they form a unique fingerprint.

Cite this