Novel Genetic Algorithm Crossover Approaches for Time-Series Problems

Paul M. Godley, Julie Cowie, David E. Cairns

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)

Abstract

Genetic Algorithms (GAs) are a commonly used stochastic search
heuristic which have been applied to a plethora of problem domains.
GAs work on a population of chromosomes (an encoding of a solution
to the problem at hand) and breed solutions from fit parents to
hopefully produce fitter children through a process of crossover and
mutation. This work discusses two novel crossover approaches for GAs
when applied to the optimisation of time-series problems, with particular
application to bio-control schedules.
Original languageEnglish
Title of host publicationSLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms
EditorsEnda Ridge, Thomas Stützle, Mauro Birattari, Holger H. Hoos
Place of PublicationBelgium
Pages47-51
Number of pages5
Publication statusPublished - 2007

Publication series

NameIRIDIA – Technical Report Series
PublisherIRIDIA
ISSN (Print)1781-3794

Fingerprint

Time series
Genetic algorithms
Biocontrol
Chromosomes

Keywords

  • genetic algorithims
  • chromosomes
  • crossover approaches
  • bio-control schedules

Cite this

Godley, P. M., Cowie, J., & Cairns, D. E. (2007). Novel Genetic Algorithm Crossover Approaches for Time-Series Problems. In E. Ridge, T. Stützle, M. Birattari, & H. H. Hoos (Eds.), SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms (pp. 47-51). (IRIDIA – Technical Report Series). Belgium.
Godley, Paul M. ; Cowie, Julie ; Cairns, David E. / Novel Genetic Algorithm Crossover Approaches for Time-Series Problems. SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms. editor / Enda Ridge ; Thomas Stützle ; Mauro Birattari ; Holger H. Hoos. Belgium, 2007. pp. 47-51 (IRIDIA – Technical Report Series).
@inbook{e75da3506fc7429fa7482675e72aa363,
title = "Novel Genetic Algorithm Crossover Approaches for Time-Series Problems",
abstract = "Genetic Algorithms (GAs) are a commonly used stochastic searchheuristic which have been applied to a plethora of problem domains.GAs work on a population of chromosomes (an encoding of a solutionto the problem at hand) and breed solutions from fit parents tohopefully produce fitter children through a process of crossover andmutation. This work discusses two novel crossover approaches for GAswhen applied to the optimisation of time-series problems, with particularapplication to bio-control schedules.",
keywords = "genetic algorithims, chromosomes, crossover approaches, bio-control schedules",
author = "Godley, {Paul M.} and Julie Cowie and Cairns, {David E.}",
year = "2007",
language = "English",
series = "IRIDIA – Technical Report Series",
publisher = "IRIDIA",
pages = "47--51",
editor = "Ridge, {Enda } and Thomas St{\"u}tzle and Birattari, {Mauro } and Hoos, {Holger H. }",
booktitle = "SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms",

}

Godley, PM, Cowie, J & Cairns, DE 2007, Novel Genetic Algorithm Crossover Approaches for Time-Series Problems. in E Ridge, T Stützle, M Birattari & HH Hoos (eds), SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms. IRIDIA – Technical Report Series, Belgium, pp. 47-51.

Novel Genetic Algorithm Crossover Approaches for Time-Series Problems. / Godley, Paul M.; Cowie, Julie; Cairns, David E.

SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms. ed. / Enda Ridge; Thomas Stützle; Mauro Birattari; Holger H. Hoos. Belgium, 2007. p. 47-51 (IRIDIA – Technical Report Series).

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)

TY - CHAP

T1 - Novel Genetic Algorithm Crossover Approaches for Time-Series Problems

AU - Godley, Paul M.

AU - Cowie, Julie

AU - Cairns, David E.

PY - 2007

Y1 - 2007

N2 - Genetic Algorithms (GAs) are a commonly used stochastic searchheuristic which have been applied to a plethora of problem domains.GAs work on a population of chromosomes (an encoding of a solutionto the problem at hand) and breed solutions from fit parents tohopefully produce fitter children through a process of crossover andmutation. This work discusses two novel crossover approaches for GAswhen applied to the optimisation of time-series problems, with particularapplication to bio-control schedules.

AB - Genetic Algorithms (GAs) are a commonly used stochastic searchheuristic which have been applied to a plethora of problem domains.GAs work on a population of chromosomes (an encoding of a solutionto the problem at hand) and breed solutions from fit parents tohopefully produce fitter children through a process of crossover andmutation. This work discusses two novel crossover approaches for GAswhen applied to the optimisation of time-series problems, with particularapplication to bio-control schedules.

KW - genetic algorithims

KW - chromosomes

KW - crossover approaches

KW - bio-control schedules

M3 - Chapter (peer-reviewed)

T3 - IRIDIA – Technical Report Series

SP - 47

EP - 51

BT - SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms

A2 - Ridge, Enda

A2 - Stützle, Thomas

A2 - Birattari, Mauro

A2 - Hoos, Holger H.

CY - Belgium

ER -

Godley PM, Cowie J, Cairns DE. Novel Genetic Algorithm Crossover Approaches for Time-Series Problems. In Ridge E, Stützle T, Birattari M, Hoos HH, editors, SLS-DS 2007 Doctoral Symposium on Engineering Stochastic Local Search Algorithms. Belgium. 2007. p. 47-51. (IRIDIA – Technical Report Series).