Title
Counterexample-Driven Genetic Programming for Symbolic Regression with Formal Constraints
Authors
[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ P ] employee
Scientific discipline (Law 2.0)
Year of publication
2023
Published in
Article type
scientific article
Publication language
english
Keywords
EN
- Symbolic regression
- Constraints
- Satisfiability Modulo Theories
- Genetic Programming
Abstract
EN In symbolic regression with formal constraints, the conventional formulation of regression problem is extended with desired properties of the target model, like symmetry, monotonicity, or convexity. We present a genetic programming algorithm that solves such problems using a Satisfiability Modulo Theories solver to formally verify the candidate solutions. The essence of the method consists in collecting the counterexamples resulting from model verification and using them to improve search guidance. The method is exact: upon successful termination, the produced model is guaranteed to meet the specified constraints. We compare the effectiveness of the proposed method with standard constraint-agnostic machine learning regression algorithms on a range of benchmarks, and demonstrate that it outperforms them on several performance indicators.
Date of online publication
08.09.2022
License type
CC BY (attribution alone)
Open Access Mode
czasopismo hybrydowe
Open Access Text Version
final author's version
Date of Open Access to the publication
in press
Ministry points / journal
200.0
Impact Factor
16.497 [List 2021]
System created by Poznań University of Technology
and Poznan Supercomputing and Networking Center
Log in through eKonto to add to SIS