Publication detail

Adaptive Fitness Predictors in Coevolutionary Cartesian Genetic Programming

DRAHOŠOVÁ, M. SEKANINA, L. WIGLASZ, M.

Original Title

Adaptive Fitness Predictors in Coevolutionary Cartesian Genetic Programming

Type

journal article in Web of Science

Language

English

Original Abstract

In genetic programming (GP), computer programs are often coevolved with training data subsets that are known as fitness predictors. In order to maximize performance of GP, it is important to find the most suitable parameters of coevolution, particularly the fitness predictor size. This is a very time consuming process as the predictor size depends on a given application and many experiments have to be performed to find its suitable size. A new method is proposed which enables us to automatically adapt the predictor and its size for a given problem and thus to reduce not only the time of evolution, but also the time needed to tune the evolutionary algorithm. The method was implemented in the context of Cartesian genetic programming and evaluated using five symbolic regression problems and three image filter design problems. In comparison with three different CGP implementations, the time required by CGP search was reduced while the quality of results remained unaffected.

Keywords

Cartesian genetic programming, coevolutionary algorithms, fitness prediction, symbolic regression, evolutionary design, image processing.

Authors

DRAHOŠOVÁ, M.; SEKANINA, L.; WIGLASZ, M.

Released

3. 9. 2019

ISBN

1063-6560

Periodical

EVOLUTIONARY COMPUTATION

Year of study

27

Number

3

State

United States of America

Pages from

497

Pages to

523

Pages count

27

URL

https://www.fit.vut.cz/research/publication/11206/

BibTex

@article{BUT159961,
  author="Michaela {Drahošová} and Lukáš {Sekanina} and Michal {Wiglasz}",
  title="Adaptive Fitness Predictors in Coevolutionary Cartesian Genetic Programming",
  journal="EVOLUTIONARY COMPUTATION",
  year="2019",
  volume="27",
  number="3",
  pages="497--523",
  doi="10.1162/evco\{_}a\{_}00229",
  issn="1063-6560",
  url="https://www.fit.vut.cz/research/publication/11206/"
}