Shape-Constrained Symbolic Regression with NSGA-III.

Christian Haider, Gabriel Kronberger

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

Abstract

Shape-constrained symbolic regression (SCSR) allows to include prior knowledge into data-based modeling. This inclusion allows to ensure that certain expected behavior is reflected by the resulting models. This specific behavior is defined by constraints which restrict the functional form e.g. monotonicity, concavity or model image boundaries. That allows finding more robust and reliable models in the specific case of highly noisy data or in extrapolation domains. This paper presents a multi-objective approach to minimize the approximation error as well as the constraint violations. Explicitly the two algorithms NSGA-II and NSGA-III are implemented and compared against each other in terms of model quality and runtime. Both algorithms are executed on a selected set of benchmark instances from physics textbooks. The results indicate that both algorithms are able to generate mostly feasible solutions and NSGA-III provides slight improvements in terms of model quality. Moreover, an improvement in runtime can be observed when using NSGA-III.

Original languageEnglish
Title of host publicationComputer Aided Systems Theory – EUROCAST 2022 - 18th International Conference, Revised Selected Papers
EditorsRoberto Moreno-Díaz, Franz Pichler, Alexis Quesada-Arencibia
PublisherSpringer
Pages164-172
Number of pages9
ISBN (Print)9783031253119
DOIs
Publication statusPublished - 2022
Event18th International Conference on Computer Aided Systems Theory, EUROCAST 2022 - Las Palmas de Gran Canaria, Spain
Duration: 20 Feb 202225 Feb 2022

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume13789 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th International Conference on Computer Aided Systems Theory, EUROCAST 2022
Country/TerritorySpain
CityLas Palmas de Gran Canaria
Period20.02.202225.02.2022

Keywords

  • Many-objective optimization
  • Shape-constraints
  • Symbolic regression

Fingerprint

Dive into the research topics of 'Shape-Constrained Symbolic Regression with NSGA-III.'. Together they form a unique fingerprint.

Cite this