An Alternative to Random Generation of the Initial Population for Genetic Algorithms

Shane Lee; Hefin Rowlands

An Alternative to Random Generation of the Initial Population for Genetic Algorithms

Shane Lee, Hefin Rowlands

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Cyfraniad i gynhadledd › adolygiad gan gymheiriaid

Crynodeb

An alternative to the traditional random method of generating the initial population of genetic algorithms is introduced. The orthogonal arrays of the Taguchi method are used to create a population of chromosomes. The performance of these populations is compared empirically with populations created in the conventional random way. Just one test function is consider here which adequately illustrates the strengths and weaknesses of the new techniques The addition of Gaussian noise to the new initialization routine is shown to further improve the results.

Iaith wreiddiol	Saesneg
Teitl	Proceedings 20th European Conference on Modelling and Simulation
Golygyddion	Wolfgang Borutzky, Alessandra Orsoni, Richard Zobel
Cyhoeddwr	ECMS
Nifer y tudalennau	5
ISBN (Argraffiad)	0-9553018-0-7
Statws	Cyhoeddwyd - 2006
Digwyddiad	20th European Conference on Modelling and Simulation - Bonn, Yr Almaen Hyd: 28 Mai 2006 → 31 Mai 2006

Cynhadledd

Cynhadledd	20th European Conference on Modelling and Simulation
Teitl cryno	ECMS 2006
Gwlad/Tiriogaeth	Yr Almaen
Dinas	Bonn
Cyfnod	28/05/06 → 31/05/06

Dyfynnu hyn

@inproceedings{375dd1039d064d4ba15926802873668e,

title = "An Alternative to Random Generation of the Initial Population for Genetic Algorithms",

abstract = "An alternative to the traditional random method of generating the initial population of genetic algorithms is introduced. The orthogonal arrays of the Taguchi method are used to create a population of chromosomes. The performance of these populations is compared empirically with populations created in the conventional random way. Just one test function is consider here which adequately illustrates the strengths and weaknesses of the new techniques The addition of Gaussian noise to the new initialization routine is shown to further improve the results. ",

author = "Shane Lee and Hefin Rowlands",

year = "2006",

language = "English",

isbn = "0-9553018-0-7",

editor = "Wolfgang Borutzky and Alessandra Orsoni and Richard Zobel",

booktitle = "Proceedings 20th European Conference on Modelling and Simulation",

publisher = "ECMS",

note = "20th European Conference on Modelling and Simulation , ECMS 2006 ; Conference date: 28-05-2006 Through 31-05-2006",

}

An Alternative to Random Generation of the Initial Population for Genetic Algorithms. / Lee, Shane; Rowlands, Hefin.

Proceedings 20th European Conference on Modelling and Simulation . gol. / Wolfgang Borutzky; Alessandra Orsoni; Richard Zobel. ECMS, 2006.

Allbwn ymchwil: Pennod mewn Llyfr/Adroddiad/Trafodion Cynhadledd › Cyfraniad i gynhadledd › adolygiad gan gymheiriaid

TY - GEN

T1 - An Alternative to Random Generation of the Initial Population for Genetic Algorithms

AU - Lee, Shane

AU - Rowlands, Hefin

PY - 2006

Y1 - 2006

N2 - An alternative to the traditional random method of generating the initial population of genetic algorithms is introduced. The orthogonal arrays of the Taguchi method are used to create a population of chromosomes. The performance of these populations is compared empirically with populations created in the conventional random way. Just one test function is consider here which adequately illustrates the strengths and weaknesses of the new techniques The addition of Gaussian noise to the new initialization routine is shown to further improve the results.

AB - An alternative to the traditional random method of generating the initial population of genetic algorithms is introduced. The orthogonal arrays of the Taguchi method are used to create a population of chromosomes. The performance of these populations is compared empirically with populations created in the conventional random way. Just one test function is consider here which adequately illustrates the strengths and weaknesses of the new techniques The addition of Gaussian noise to the new initialization routine is shown to further improve the results.

M3 - Conference contribution

SN - 0-9553018-0-7

BT - Proceedings 20th European Conference on Modelling and Simulation

A2 - Borutzky, Wolfgang

A2 - Orsoni, Alessandra

A2 - Zobel, Richard

PB - ECMS

T2 - 20th European Conference on Modelling and Simulation

Y2 - 28 May 2006 through 31 May 2006

ER -

An Alternative to Random Generation of the Initial Population for Genetic Algorithms

Crynodeb

Cynhadledd

Ôl bys

Dyfynnu hyn