Publication Details

Indirectly Encoded Fitness Predictors Coevolved with Cartesian Programs

DRAHOŠOVÁ, M.; HULVA, J.; SEKANINA, L. Indirectly Encoded Fitness Predictors Coevolved with Cartesian Programs. In Genetic Programming. Lecture Notes in Computer Science. Berlin: Springer International Publishing, 2015. p. 113-125. ISBN: 978-3-319-16500-4.
Czech title
Koevoluce nepřímo kódovaných prediktorů fitness a kartézských programů
Type
conference paper
Language
English
Authors
Drahošová Michaela, Ing., Ph.D. (DCSY)
Hulva Jiří, Ing.
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY)
URL
Keywords

coevolution, cartesian genetic programming, fitness prediction, symbolic regression

Abstract

We investigate coevolutionary Cartesian genetic programming that coevolves fitness predictors in order to diminish the number of target objective vector (TOV) evaluations, needed to obtain a satisfactory solution, to reduce the computational cost of evolution. This paper introduces the use of coevolution of fitness predictors in CGP with a new type of indirectly encoded predictors. Indirectly encoded predictors are operated using the CGP and provide a variable number of TOVs used for solution evaluation during the coevolution. It is shown in 5 symbolic regression problems that the proposed predictors are able to adapt the size of TOVs array in response to a particular training data set.

Published
2015
Pages
113–125
Proceedings
Genetic Programming
Series
Lecture Notes in Computer Science
Volume
9025
ISBN
978-3-319-16500-4
Publisher
Springer International Publishing
Place
Berlin
DOI
10.1007/978-3-319-16501-1_10
UT WoS
000361758600010
EID Scopus
2-s2.0-84925120847
BibTeX 
@inproceedings{BUT119803,
  author="Michaela {Drahošová} and Jiří {Hulva} and Lukáš {Sekanina}",
  title="Indirectly Encoded Fitness Predictors Coevolved with Cartesian Programs",
  booktitle="Genetic Programming",
  year="2015",
  series="Lecture Notes in Computer Science",
  volume="9025",
  pages="113--125",
  publisher="Springer International Publishing",
  address="Berlin",
  doi="10.1007/978-3-319-16501-1\{_}10",
  isbn="978-3-319-16500-4",
  url="http://dx.doi.org/10.1007/978-3-319-16501-1_10"
}
Back to top