Publication Details

Plastic Fitness Predictors Coevolved with Cartesian Programs

WIGLASZ, M.; DRAHOŠOVÁ, M. Plastic Fitness Predictors Coevolved with Cartesian Programs. In 19th European Conference on Genetic programming. Lecture Notes in Computer Science. Berlin: Springer International Publishing, 2016. p. 164-179. ISBN: 978-3-319-30667-4.
Czech title
Koevoluce plastických prediktorů fitness a kartézských programů
Type
conference paper
Language
English
Authors
Wiglasz Michal, Ing.
Drahošová Michaela, Ing., Ph.D. (DCSY)
Keywords


fitness predictors, cartesian genetic programming, coevolution, phenotypic
plasticity

Abstract

Coevolution of fitness predictors, which are a small sample of all training data
for a particular task, was successfully used to reduce the computational cost of
the design performed by cartesian genetic programming. However, it is necessary
to specify the most advantageous number of fitness cases in predictors, which
differs from task to task. This paper proposes to introduce a new type of
directly encoded fitness predictors inspired by the principles of phenotypic
plasticity. The size of the coevolved fitness predictor is adapted in response to
the phase of learning that the program evolution goes through. It is shown in 5
symbolic regression tasks that the proposed algorithm is able to adapt the number
of fitness cases in predictors in response to the solved task and the program
evolution flow.

Published
2016
Pages
164–179
Proceedings
19th European Conference on Genetic programming
Series
Lecture Notes in Computer Science
Volume
9594
Conference
19th European Conference on Genetic Programming, Porto, PT
ISBN
978-3-319-30667-4
Publisher
Springer International Publishing
Place
Berlin
DOI
UT WoS
000894258400011
EID Scopus
BibTeX
@inproceedings{BUT130922,
  author="Michal {Wiglasz} and Michaela {Drahošová}",
  title="Plastic Fitness Predictors Coevolved with Cartesian Programs",
  booktitle="19th European Conference on Genetic programming",
  year="2016",
  series="Lecture Notes in Computer Science",
  volume="9594",
  pages="164--179",
  publisher="Springer International Publishing",
  address="Berlin",
  doi="10.1007/978-3-319-30668-1\{_}11",
  isbn="978-3-319-30667-4",
  url="https://www.fit.vut.cz/research/publication/11001/"
}
Files
Back to top