A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness

Alexandre Mayer

doi:10.1145/3067695.3075968

A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

Résumé

We present a Genetic Algorithm that we developed in order to address computationally expensive optimization problems. In order to accelerate this algorithm, we establish, generation after generation, quadratic approximations of the fitness in the close neighborhood of the best-so-far individual. We then inject in the population an individual that corresponds to the optimum of this approximation. We also introduce a modified mutation operator that acts on randomly-shifted Gray codes. We show that these techniques lead to the global optimum of typical benchmark problems in 5, 10 and 20 dimensions with a probability of success in one run of the order
of 95-97% and an average number of fitness evaluations of the order of 400−750×n, where n refers to the dimension of the problem.

langue originale	Anglais
titre	GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference
rédacteurs en chef	Peter A. N. Bosman
Editeur	ACM Press
Pages	195-196
ISBN (imprimé)	978-1-4503-4920-8
Les DOIs	https://doi.org/10.1145/3067695.3075968
Etat de la publication	Publié - 2017
Evénement	The Genetic and Evolutionary Computation Conference 2017 - Berlin, Allemagne Durée: 15 juil. 2017 → 19 juil. 2017

Série de publications

Nom	Proceedings of the Genetic and Evolutionary Computation Conference Companion

Une conférence

Une conférence	The Genetic and Evolutionary Computation Conference 2017
Titre abrégé	GECCO 2017
Pays/Territoire	Allemagne
La ville	Berlin
période	15/07/17 → 19/07/17

mots-clés

genetic algorithm

Accès au document

10.1145/3067695.3075968

Document sous embargo

paper95
Version finale publiée, 78,7 KB
Licence: Non spécifié

consortium des équipements de calcul intensif
Champagne, B. (Co-investigateur)
1/01/11 → 31/12/22
Projet: Recherche

Plateforme Technologique Calcul Intensif
Champagne, B. (!!Manager)
Plateforme technologique Calcul intensif
Equipement/installations: Plateforme technolgique

Contient cette citation

Mayer, A. (2017). A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness. Dans P. A. N. Bosman (Ed.), GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference (p. 195-196). (Proceedings of the Genetic and Evolutionary Computation Conference Companion). ACM Press. https://doi.org/10.1145/3067695.3075968

@inproceedings{080943b138224f59a638592e495a4bf6,

title = "A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness",

abstract = "We present a Genetic Algorithm that we developed in order to address computationally expensive optimization problems. In order to accelerate this algorithm, we establish, generation after generation, quadratic approximations of the fitness in the close neighborhood of the best-so-far individual. We then inject in the population an individual that corresponds to the optimum of this approximation. We also introduce a modified mutation operator that acts on randomly-shifted Gray codes. We show that these techniques lead to the global optimum of typical benchmark problems in 5, 10 and 20 dimensions with a probability of success in one run of the orderof 95-97% and an average number of fitness evaluations of the order of 400−750×n, where n refers to the dimension of the problem.",

keywords = "genetic algorithm",

author = "Alexandre Mayer",

year = "2017",

doi = "10.1145/3067695.3075968",

language = "English",

isbn = "978-1-4503-4920-8",

series = "Proceedings of the Genetic and Evolutionary Computation Conference Companion",

publisher = "ACM Press",

pages = "195--196",

editor = "Bosman, {Peter A. N.}",

booktitle = "GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference",

address = "United States",

note = "The Genetic and Evolutionary Computation Conference 2017 ; Conference date: 15-07-2017 Through 19-07-2017",

}

Mayer, A 2017, A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness. Dans PAN Bosman (Ed.), GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference. Proceedings of the Genetic and Evolutionary Computation Conference Companion, ACM Press, p. 195-196, The Genetic and Evolutionary Computation Conference 2017, Berlin, Allemagne, 15/07/17. https://doi.org/10.1145/3067695.3075968

A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness. / Mayer, Alexandre.
GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference. Ed. / Peter A. N. Bosman. ACM Press, 2017. p. 195-196 (Proceedings of the Genetic and Evolutionary Computation Conference Companion).

Résultats de recherche: Contribution dans un livre/un catalogue/un rapport/dans les actes d'une conférence › Article dans les actes d'une conférence/un colloque

TY - GEN

T1 - A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness

AU - Mayer, Alexandre

PY - 2017

Y1 - 2017

N2 - We present a Genetic Algorithm that we developed in order to address computationally expensive optimization problems. In order to accelerate this algorithm, we establish, generation after generation, quadratic approximations of the fitness in the close neighborhood of the best-so-far individual. We then inject in the population an individual that corresponds to the optimum of this approximation. We also introduce a modified mutation operator that acts on randomly-shifted Gray codes. We show that these techniques lead to the global optimum of typical benchmark problems in 5, 10 and 20 dimensions with a probability of success in one run of the orderof 95-97% and an average number of fitness evaluations of the order of 400−750×n, where n refers to the dimension of the problem.

AB - We present a Genetic Algorithm that we developed in order to address computationally expensive optimization problems. In order to accelerate this algorithm, we establish, generation after generation, quadratic approximations of the fitness in the close neighborhood of the best-so-far individual. We then inject in the population an individual that corresponds to the optimum of this approximation. We also introduce a modified mutation operator that acts on randomly-shifted Gray codes. We show that these techniques lead to the global optimum of typical benchmark problems in 5, 10 and 20 dimensions with a probability of success in one run of the orderof 95-97% and an average number of fitness evaluations of the order of 400−750×n, where n refers to the dimension of the problem.

KW - genetic algorithm

U2 - 10.1145/3067695.3075968

DO - 10.1145/3067695.3075968

M3 - Conference contribution

SN - 978-1-4503-4920-8

T3 - Proceedings of the Genetic and Evolutionary Computation Conference Companion

SP - 195

EP - 196

BT - GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference

A2 - Bosman, Peter A. N.

PB - ACM Press

T2 - The Genetic and Evolutionary Computation Conference 2017

Y2 - 15 July 2017 through 19 July 2017

ER -

Mayer A. A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness. Dans Bosman PAN, Editeur, GECCO '17: Proceedings of the Genetic and Evolutionary Computation Conference. ACM Press. 2017. p. 195-196. (Proceedings of the Genetic and Evolutionary Computation Conference Companion). doi: 10.1145/3067695.3075968

A Genetic Algorithm with randomly shifted Gray codes and local optimizations based on quadratic approximations of the fitness

Résumé

Série de publications

Une conférence

mots-clés

Accès au document

Document sous embargo

Empreinte digitale

Projets

consortium des équipements de calcul intensif

Équipement

Plateforme Technologique Calcul Intensif

Contient cette citation