Emergence of diversity and its benefits for crossover in genetic algorithms

Duc Cuong Dang; Tobias Friedrich; Timo Kötzing; Martin S. Krejca; Per Kristian Lehre; Pietro S. Oliveto; Dirk Sudholt; Andrew M. Sutton

doi:10.1007/978-3-319-45823-6_83

Emergence of diversity and its benefits for crossover in genetic algorithms

Duc Cuong Dang, Tobias Friedrich, Timo Kötzing, Martin S. Krejca, Per Kristian Lehre, Pietro S. Oliveto, Dirk Sudholt, Andrew M. Sutton

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

18 Scopus citations

Abstract

Population diversity is essential for avoiding premature convergence in Genetic Algorithms (GAs) and for the effective use of crossover. Yet the dynamics of how diversity emerges in populations are not well understood.We use rigorous runtime analysis to gain insight into population dynamics and GA performance for a standard (μ+1) GA and the Jump_k test function. By studying the stochastic process underlying the size of the largest collection of identical genotypes we show that the interplay of crossover followed by mutation may serve as a catalyst leading to a sudden burst of diversity. This leads to improvements of the expected optimisation time of order Ω(n/ log n) compared to mutationonly algorithms like the (1+1) EA.

Original language	English (US)
Title of host publication	Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings
Editors	Emma Hart, Ben Paechter, Julia Handl, Manuel López-Ibáñez, Peter R. Lewis, Gabriela Ochoa
Publisher	Springer Verlag
Pages	890-900
Number of pages	11
ISBN (Print)	9783319458229
DOIs	https://doi.org/10.1007/978-3-319-45823-6_83
State	Published - 2016
Externally published	Yes
Event	14th International Conference on Parallel Problem Solving from Nature, PPSN 2016 - Edinburgh, United Kingdom Duration: Sep 17 2016 → Sep 21 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9921 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	14th International Conference on Parallel Problem Solving from Nature, PPSN 2016
Country/Territory	United Kingdom
City	Edinburgh
Period	9/17/16 → 9/21/16

Bibliographical note

Publisher Copyright:
© Springer International Publishing AG 2016.

Keywords

Crossover
Diversity
Genetic algorithms
Runtime analysis
Theory

Access

10.1007/978-3-319-45823-6_83

http://eprints.whiterose.ac.uk/101311/

OpenUrl availability

Full text

Cite this

Dang, D. C., Friedrich, T., Kötzing, T., Krejca, M. S., Lehre, P. K., Oliveto, P. S., Sudholt, D., & Sutton, A. M. (2016). Emergence of diversity and its benefits for crossover in genetic algorithms. In E. Hart, B. Paechter, J. Handl, M. López-Ibáñez, P. R. Lewis, & G. Ochoa (Eds.), Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings (pp. 890-900). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9921 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-45823-6_83

Emergence of diversity and its benefits for crossover in genetic algorithms. / Dang, Duc Cuong; Friedrich, Tobias; Kötzing, Timo et al.
Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings. ed. / Emma Hart; Ben Paechter; Julia Handl; Manuel López-Ibáñez; Peter R. Lewis; Gabriela Ochoa. Springer Verlag, 2016. p. 890-900 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9921 LNCS).

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

Dang, DC, Friedrich, T, Kötzing, T, Krejca, MS, Lehre, PK, Oliveto, PS, Sudholt, D & Sutton, AM 2016, Emergence of diversity and its benefits for crossover in genetic algorithms. in E Hart, B Paechter, J Handl, M López-Ibáñez, PR Lewis & G Ochoa (eds), Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9921 LNCS, Springer Verlag, pp. 890-900, 14th International Conference on Parallel Problem Solving from Nature, PPSN 2016, Edinburgh, United Kingdom, 9/17/16. https://doi.org/10.1007/978-3-319-45823-6_83

Dang DC, Friedrich T, Kötzing T, Krejca MS, Lehre PK, Oliveto PS et al. Emergence of diversity and its benefits for crossover in genetic algorithms. In Hart E, Paechter B, Handl J, López-Ibáñez M, Lewis PR, Ochoa G, editors, Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings. Springer Verlag. 2016. p. 890-900. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-45823-6_83

Dang, Duc Cuong ; Friedrich, Tobias ; Kötzing, Timo et al. / Emergence of diversity and its benefits for crossover in genetic algorithms. Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings. editor / Emma Hart ; Ben Paechter ; Julia Handl ; Manuel López-Ibáñez ; Peter R. Lewis ; Gabriela Ochoa. Springer Verlag, 2016. pp. 890-900 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{2e76e27ff47d420095a5cf3555323568,

title = "Emergence of diversity and its benefits for crossover in genetic algorithms",

abstract = "Population diversity is essential for avoiding premature convergence in Genetic Algorithms (GAs) and for the effective use of crossover. Yet the dynamics of how diversity emerges in populations are not well understood.We use rigorous runtime analysis to gain insight into population dynamics and GA performance for a standard (μ+1) GA and the Jumpk test function. By studying the stochastic process underlying the size of the largest collection of identical genotypes we show that the interplay of crossover followed by mutation may serve as a catalyst leading to a sudden burst of diversity. This leads to improvements of the expected optimisation time of order Ω(n/ log n) compared to mutationonly algorithms like the (1+1) EA.",

keywords = "Crossover, Diversity, Genetic algorithms, Runtime analysis, Theory",

author = "Dang, {Duc Cuong} and Tobias Friedrich and Timo K{\"o}tzing and Krejca, {Martin S.} and Lehre, {Per Kristian} and Oliveto, {Pietro S.} and Dirk Sudholt and Sutton, {Andrew M.}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing AG 2016.; 14th International Conference on Parallel Problem Solving from Nature, PPSN 2016 ; Conference date: 17-09-2016 Through 21-09-2016",

year = "2016",

doi = "10.1007/978-3-319-45823-6_83",

language = "English (US)",

isbn = "9783319458229",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "890--900",

editor = "Emma Hart and Ben Paechter and Julia Handl and Manuel L{\'o}pez-Ib{\'a}{\~n}ez and Lewis, {Peter R.} and Gabriela Ochoa",

booktitle = "Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings",

}

TY - GEN

T1 - Emergence of diversity and its benefits for crossover in genetic algorithms

AU - Dang, Duc Cuong

AU - Friedrich, Tobias

AU - Kötzing, Timo

AU - Krejca, Martin S.

AU - Lehre, Per Kristian

AU - Oliveto, Pietro S.

AU - Sudholt, Dirk

AU - Sutton, Andrew M.

N1 - Publisher Copyright: © Springer International Publishing AG 2016.

PY - 2016

Y1 - 2016

N2 - Population diversity is essential for avoiding premature convergence in Genetic Algorithms (GAs) and for the effective use of crossover. Yet the dynamics of how diversity emerges in populations are not well understood.We use rigorous runtime analysis to gain insight into population dynamics and GA performance for a standard (μ+1) GA and the Jumpk test function. By studying the stochastic process underlying the size of the largest collection of identical genotypes we show that the interplay of crossover followed by mutation may serve as a catalyst leading to a sudden burst of diversity. This leads to improvements of the expected optimisation time of order Ω(n/ log n) compared to mutationonly algorithms like the (1+1) EA.

AB - Population diversity is essential for avoiding premature convergence in Genetic Algorithms (GAs) and for the effective use of crossover. Yet the dynamics of how diversity emerges in populations are not well understood.We use rigorous runtime analysis to gain insight into population dynamics and GA performance for a standard (μ+1) GA and the Jumpk test function. By studying the stochastic process underlying the size of the largest collection of identical genotypes we show that the interplay of crossover followed by mutation may serve as a catalyst leading to a sudden burst of diversity. This leads to improvements of the expected optimisation time of order Ω(n/ log n) compared to mutationonly algorithms like the (1+1) EA.

KW - Crossover

KW - Diversity

KW - Genetic algorithms

KW - Runtime analysis

KW - Theory

UR - http://www.scopus.com/inward/record.url?scp=84988514114&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988514114&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-45823-6_83

DO - 10.1007/978-3-319-45823-6_83

M3 - Conference contribution

AN - SCOPUS:84988514114

SN - 9783319458229

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

SP - 890

EP - 900

BT - Parallel Problem Solving from Nature - 14th International Conference, PPSN 2016, Proceedings

A2 - Hart, Emma

A2 - Paechter, Ben

A2 - Handl, Julia

A2 - López-Ibáñez, Manuel

A2 - Lewis, Peter R.

A2 - Ochoa, Gabriela

PB - Springer Verlag

T2 - 14th International Conference on Parallel Problem Solving from Nature, PPSN 2016

Y2 - 17 September 2016 through 21 September 2016

ER -

Emergence of diversity and its benefits for crossover in genetic algorithms

Abstract

Publication series

Other

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this