On the significance of the permutation problem in neuroevolution

Stefan Haflidason; Richard Neville

doi:10.1145/1569901.1570010

On the significance of the permutation problem in neuroevolution

Stefan Haflidason, Richard Neville

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we investigate the impact of the Permutation Problem on a standard Genetic Algorithm evolving neural networks for a difficult control problem. Through the use of Price's equation and an explicit enumeration of permutations within the population we demonstrate that for the given problem and representation the Permutation Problem is not as serious a concern as previously thought. In addition we present the concept of incompatible representations as a more useful guide for what to avoid in the evolution of neural networks. Copyright 2009 ACM.

Original language	English
Title of host publication	Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009\|Proc. Annu. Genetic Evol. Comput. Conf., GECCO
Publisher	Association for Computing Machinery
Pages	787-794
Number of pages	7
ISBN (Print)	9781605583259
DOIs	https://doi.org/10.1145/1569901.1570010
Publication status	Published - 2009
Event	11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009 - Montreal, QC Duration: 1 Jul 2009 → … http://dblp.uni-trier.de/db/conf/gecco/gecco2009.html#HaflidasonN09http://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09

Conference

Conference	11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009
City	Montreal, QC
Period	1/07/09 → …
Internet address	http://dblp.uni-trier.de/db/conf/gecco/gecco2009.html#HaflidasonN09http://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09

Keywords

Genetic algorithms
Neural networks
Permutation problem

Access to Document

10.1145/1569901.1570010

Cite this

@inproceedings{527e4f804e0e4219bacd86cd5a6c616a,

title = "On the significance of the permutation problem in neuroevolution",

abstract = "In this paper we investigate the impact of the Permutation Problem on a standard Genetic Algorithm evolving neural networks for a difficult control problem. Through the use of Price's equation and an explicit enumeration of permutations within the population we demonstrate that for the given problem and representation the Permutation Problem is not as serious a concern as previously thought. In addition we present the concept of incompatible representations as a more useful guide for what to avoid in the evolution of neural networks. Copyright 2009 ACM.",

keywords = "Genetic algorithms, Neural networks, Permutation problem",

author = "Stefan Haflidason and Richard Neville",

note = "100 word commentary: Paper utilised Price{\textquoteright}s equation and explicit enumeration of permutations in order to investigate the significance of the {\textquoteleft}Permutations Problem{\textquoteright} in the evolution of a neural network for a difficult control problem. Paper demonstrated: 1) performance with a standard GA on a difficult control problem is acceptable and 2) the introduction of inversion has a relatively minor negative effect. The empirical results support the claim that the permutations are enumerable. Role: principal academic supervisor of first author who was a PhD student - contribution estimate = 25%. Conference is one of the Top Evolutionary Computation Conferences. Most likely REF score 3*.; 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009 ; Conference date: 01-07-2009",

year = "2009",

doi = "10.1145/1569901.1570010",

language = "English",

isbn = "9781605583259",

pages = "787--794",

booktitle = "Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009|Proc. Annu. Genetic Evol. Comput. Conf., GECCO",

publisher = "Association for Computing Machinery",

address = "United States",

url = "http://dblp.uni-trier.de/db/conf/gecco/gecco2009.html#HaflidasonN09http://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09.xmlhttp://dblp.uni-trier.de/rec/bibtex/conf/gecco/HaflidasonN09",

}

Haflidason, S & Neville, R 2009, On the significance of the permutation problem in neuroevolution. in Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009|Proc. Annu. Genetic Evol. Comput. Conf., GECCO. Association for Computing Machinery, pp. 787-794, 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009, Montreal, QC, 1/07/09. https://doi.org/10.1145/1569901.1570010

On the significance of the permutation problem in neuroevolution. / Haflidason, Stefan; Neville, Richard.

Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009|Proc. Annu. Genetic Evol. Comput. Conf., GECCO. Association for Computing Machinery, 2009. p. 787-794.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - On the significance of the permutation problem in neuroevolution

AU - Haflidason, Stefan

AU - Neville, Richard

N1 - 100 word commentary: Paper utilised Price’s equation and explicit enumeration of permutations in order to investigate the significance of the ‘Permutations Problem’ in the evolution of a neural network for a difficult control problem. Paper demonstrated: 1) performance with a standard GA on a difficult control problem is acceptable and 2) the introduction of inversion has a relatively minor negative effect. The empirical results support the claim that the permutations are enumerable. Role: principal academic supervisor of first author who was a PhD student - contribution estimate = 25%. Conference is one of the Top Evolutionary Computation Conferences. Most likely REF score 3*.

PY - 2009

Y1 - 2009

N2 - In this paper we investigate the impact of the Permutation Problem on a standard Genetic Algorithm evolving neural networks for a difficult control problem. Through the use of Price's equation and an explicit enumeration of permutations within the population we demonstrate that for the given problem and representation the Permutation Problem is not as serious a concern as previously thought. In addition we present the concept of incompatible representations as a more useful guide for what to avoid in the evolution of neural networks. Copyright 2009 ACM.

AB - In this paper we investigate the impact of the Permutation Problem on a standard Genetic Algorithm evolving neural networks for a difficult control problem. Through the use of Price's equation and an explicit enumeration of permutations within the population we demonstrate that for the given problem and representation the Permutation Problem is not as serious a concern as previously thought. In addition we present the concept of incompatible representations as a more useful guide for what to avoid in the evolution of neural networks. Copyright 2009 ACM.

KW - Genetic algorithms

KW - Neural networks

KW - Permutation problem

U2 - 10.1145/1569901.1570010

DO - 10.1145/1569901.1570010

M3 - Conference contribution

SN - 9781605583259

SP - 787

EP - 794

BT - Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009|Proc. Annu. Genetic Evol. Comput. Conf., GECCO

PB - Association for Computing Machinery

T2 - 11th Annual Genetic and Evolutionary Computation Conference, GECCO-2009

Y2 - 1 July 2009

ER -

On the significance of the permutation problem in neuroevolution

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this