Publication Details

Comparison of Parallel Linear Genetic Programming Implementations

GROCHOL, D.; SEKANINA, L. Comparison of Parallel Linear Genetic Programming Implementations. In Recent Advances in Soft Computing: Proceedings of the 22nd International Conference on Soft Computing (MENDEL 2016) held in Brno, Czech Republic, at June 8-10, 2016. Cham: Springer International Publishing, 2017. p. 64-76. ISBN: 978-3-319-58088-3.
Czech title
Porovnání paralelních implementací lineárního genetického programování
Type
conference paper
Language
English
Authors
Grochol David, Ing., Ph.D.
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY)
Keywords

Linear genetic programming, parallel implementation, island model, hash function, symbolic regresion

Abstract

Linear genetic programming (LGP) represents candidate programs as sequences of instructions for a register machine. In order to accelerate the evaluation time of candidate programs and reduce the overall time of evolution, we propose parallel implementations of LGP suitable for current multi-core processors. The implementations are based on a parallel evaluation of candidate programs and the island model of parallel evolutionary algorithm in which subpopulations are evolved independently, but some genetic material can be exchanged by means of migration. Proposed implementations are evaluated using three symbolic regression problems and hash function design problem.

Published
2017
Pages
64–76
Proceedings
Recent Advances in Soft Computing: Proceedings of the 22nd International Conference on Soft Computing (MENDEL 2016) held in Brno, Czech Republic, at June 8-10, 2016
ISBN
978-3-319-58088-3
Publisher
Springer International Publishing
Place
Cham
DOI
EID Scopus
BibTeX
@inproceedings{BUT144385,
  author="David {Grochol} and Lukáš {Sekanina}",
  title="Comparison of Parallel Linear Genetic Programming Implementations",
  booktitle="Recent Advances in Soft Computing: Proceedings of the 22nd International Conference on Soft Computing (MENDEL 2016) held in Brno, Czech Republic, at June 8-10, 2016",
  year="2017",
  pages="64--76",
  publisher="Springer International Publishing",
  address="Cham",
  doi="10.1007/978-3-319-58088-3\{_}7",
  isbn="978-3-319-58088-3",
  url="https://www.fit.vut.cz/research/publication/10997/"
}
Files
Back to top