Self-adaptive crossover in genetic programming: The case of the tartarus problem

Thomas D. Griffiths*, Anikó Ekárt

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The runtime performance of many evolutionary algorithms depends heavily on their parameter values, many of which are problem specific. Previous work has shown that the modification of parameter values at runtime can lead to significant improvements in performance. In this paper we discuss both the ‘when’ and ‘how’ aspects of implementing self-adaptation in a Genetic Programming system, focusing on the crossover operator. We perform experiments on Tartarus Problem instances and find that the runtime modification of crossover parameters at the individual level, rather than population level, generate solutions with superior performance, compared to traditional crossover methods.

Original languageEnglish
Title of host publicationParallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings
PublisherSpringer
Pages236-246
Number of pages11
ISBN (Print)9783319992525
DOIs
Publication statusPublished - 4 Jun 2018
Event15th International Conference on Parallel Problem Solving from Nature, PPSN 2018 - Coimbra, Portugal
Duration: 8 Sep 201812 Sep 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11101 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference15th International Conference on Parallel Problem Solving from Nature, PPSN 2018
CountryPortugal
CityCoimbra
Period8/09/1812/09/18

Fingerprint

Genetic programming
Genetic Programming
Evolutionary algorithms
Crossover
Self-adaptation
Crossover Operator
Experiments
Evolutionary Algorithms
Experiment

Keywords

  • Crossover
  • Self-adaption
  • Tartarus problem

Cite this

Griffiths, T. D., & Ekárt, A. (2018). Self-adaptive crossover in genetic programming: The case of the tartarus problem. In Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings (pp. 236-246). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11101 LNCS). Springer. https://doi.org/10.1007/978-3-319-99253-2_19
Griffiths, Thomas D. ; Ekárt, Anikó. / Self-adaptive crossover in genetic programming : The case of the tartarus problem. Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings. Springer, 2018. pp. 236-246 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e05a22a5851c444680cba7b31e12f2cd,
title = "Self-adaptive crossover in genetic programming: The case of the tartarus problem",
abstract = "The runtime performance of many evolutionary algorithms depends heavily on their parameter values, many of which are problem specific. Previous work has shown that the modification of parameter values at runtime can lead to significant improvements in performance. In this paper we discuss both the ‘when’ and ‘how’ aspects of implementing self-adaptation in a Genetic Programming system, focusing on the crossover operator. We perform experiments on Tartarus Problem instances and find that the runtime modification of crossover parameters at the individual level, rather than population level, generate solutions with superior performance, compared to traditional crossover methods.",
keywords = "Crossover, Self-adaption, Tartarus problem",
author = "Griffiths, {Thomas D.} and Anik{\'o} Ek{\'a}rt",
year = "2018",
month = "6",
day = "4",
doi = "10.1007/978-3-319-99253-2_19",
language = "English",
isbn = "9783319992525",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer",
pages = "236--246",
booktitle = "Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings",
address = "Germany",

}

Griffiths, TD & Ekárt, A 2018, Self-adaptive crossover in genetic programming: The case of the tartarus problem. in Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11101 LNCS, Springer, pp. 236-246, 15th International Conference on Parallel Problem Solving from Nature, PPSN 2018, Coimbra, Portugal, 8/09/18. https://doi.org/10.1007/978-3-319-99253-2_19

Self-adaptive crossover in genetic programming : The case of the tartarus problem. / Griffiths, Thomas D.; Ekárt, Anikó.

Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings. Springer, 2018. p. 236-246 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11101 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Self-adaptive crossover in genetic programming

T2 - The case of the tartarus problem

AU - Griffiths, Thomas D.

AU - Ekárt, Anikó

PY - 2018/6/4

Y1 - 2018/6/4

N2 - The runtime performance of many evolutionary algorithms depends heavily on their parameter values, many of which are problem specific. Previous work has shown that the modification of parameter values at runtime can lead to significant improvements in performance. In this paper we discuss both the ‘when’ and ‘how’ aspects of implementing self-adaptation in a Genetic Programming system, focusing on the crossover operator. We perform experiments on Tartarus Problem instances and find that the runtime modification of crossover parameters at the individual level, rather than population level, generate solutions with superior performance, compared to traditional crossover methods.

AB - The runtime performance of many evolutionary algorithms depends heavily on their parameter values, many of which are problem specific. Previous work has shown that the modification of parameter values at runtime can lead to significant improvements in performance. In this paper we discuss both the ‘when’ and ‘how’ aspects of implementing self-adaptation in a Genetic Programming system, focusing on the crossover operator. We perform experiments on Tartarus Problem instances and find that the runtime modification of crossover parameters at the individual level, rather than population level, generate solutions with superior performance, compared to traditional crossover methods.

KW - Crossover

KW - Self-adaption

KW - Tartarus problem

UR - http://www.scopus.com/inward/record.url?scp=85053618355&partnerID=8YFLogxK

UR - https://link.springer.com/chapter/10.1007%2F978-3-319-99253-2_19

U2 - 10.1007/978-3-319-99253-2_19

DO - 10.1007/978-3-319-99253-2_19

M3 - Conference contribution

AN - SCOPUS:85053618355

SN - 9783319992525

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 236

EP - 246

BT - Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings

PB - Springer

ER -

Griffiths TD, Ekárt A. Self-adaptive crossover in genetic programming: The case of the tartarus problem. In Parallel Problem Solving from Nature – PPSN XV - 15th International Conference, 2018, Proceedings. Springer. 2018. p. 236-246. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-99253-2_19