Enhancing exploration and exploitation of NSGA-II with GP and PDL

Peter David Shannon; Chrystopher L. Nehaniv; Somnuk Phon-Amnuaisuk

doi:10.1007/978-3-319-61824-1_38

Enhancing exploration and exploitation of NSGA-II with GP and PDL

Peter David Shannon, Chrystopher L. Nehaniv, Somnuk Phon-Amnuaisuk

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

1 Citation (Scopus)

Abstract

In this paper, we show that NSGA-II can be applied to GP and the Process Description Language (PDL) and describe two modifications to NSGA-II. The first modification removes individuals which have the same behaviour from GP populations. It selects for de-duplication by taking the result of each objective fitness function together to make a comparison. NSGA-II is designed to expand its Pareto front of solutions by favouring individuals who have the highest or lowest value (boundary points) in a front, for any objective. The second modification enhances exploitation by preferring individuals who occupy an extreme position for most objective fitness functions. The results show, for the first time, that NSGA-II can be used with PDL and GP to successfully solve a robot control problem and that the suggested modifications offer significant improvements over an algorithm used previously with GP and PDL and unmodified NSGA-II for our test problem.

Original language	English
Title of host publication	Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings
Publisher	Springer Nature Link
Pages	349-361
Number of pages	13
Volume	10385 LNCS
ISBN (Print)	9783319618234
DOIs	https://doi.org/10.1007/978-3-319-61824-1_38
Publication status	E-pub ahead of print - 24 Jun 2017
Event	8th International Conference on Swarm Intelligence, ICSI 2017 - Fukuoka, Japan Duration: 27 Jul 2017 → 1 Aug 2017

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10385 LNCS
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Conference

Conference	8th International Conference on Swarm Intelligence, ICSI 2017
Country/Territory	Japan
City	Fukuoka
Period	27/07/17 → 1/08/17

Keywords

Exploration and exploitation
Genetic programming
NSGA-II
Process description language

Access to Document

10.1007/978-3-319-61824-1_38

Cite this

Shannon, P. D., Nehaniv, C. L., & Phon-Amnuaisuk, S. (2017). Enhancing exploration and exploitation of NSGA-II with GP and PDL. In Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings (Vol. 10385 LNCS, pp. 349-361). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10385 LNCS). Springer Nature Link. Advance online publication. https://doi.org/10.1007/978-3-319-61824-1_38

Shannon, Peter David ; Nehaniv, Chrystopher L. ; Phon-Amnuaisuk, Somnuk. / Enhancing exploration and exploitation of NSGA-II with GP and PDL. Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings. Vol. 10385 LNCS Springer Nature Link, 2017. pp. 349-361 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{9139aad88403453b9110994916f11f4a,

title = "Enhancing exploration and exploitation of NSGA-II with GP and PDL",

abstract = "In this paper, we show that NSGA-II can be applied to GP and the Process Description Language (PDL) and describe two modifications to NSGA-II. The first modification removes individuals which have the same behaviour from GP populations. It selects for de-duplication by taking the result of each objective fitness function together to make a comparison. NSGA-II is designed to expand its Pareto front of solutions by favouring individuals who have the highest or lowest value (boundary points) in a front, for any objective. The second modification enhances exploitation by preferring individuals who occupy an extreme position for most objective fitness functions. The results show, for the first time, that NSGA-II can be used with PDL and GP to successfully solve a robot control problem and that the suggested modifications offer significant improvements over an algorithm used previously with GP and PDL and unmodified NSGA-II for our test problem.",

keywords = "Exploration and exploitation, Genetic programming, NSGA-II, Process description language",

author = "Shannon, {Peter David} and Nehaniv, {Chrystopher L.} and Somnuk Phon-Amnuaisuk",

year = "2017",

month = jun,

day = "24",

doi = "10.1007/978-3-319-61824-1_38",

language = "English",

isbn = "9783319618234",

volume = "10385 LNCS",

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

publisher = "Springer Nature Link",

pages = "349--361",

booktitle = "Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings",

address = "Netherlands",

note = "8th International Conference on Swarm Intelligence, ICSI 2017 ; Conference date: 27-07-2017 Through 01-08-2017",

}

Shannon, PD, Nehaniv, CL & Phon-Amnuaisuk, S 2017, Enhancing exploration and exploitation of NSGA-II with GP and PDL. in Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings. vol. 10385 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10385 LNCS, Springer Nature Link, pp. 349-361, 8th International Conference on Swarm Intelligence, ICSI 2017, Fukuoka, Japan, 27/07/17. https://doi.org/10.1007/978-3-319-61824-1_38

Enhancing exploration and exploitation of NSGA-II with GP and PDL. / Shannon, Peter David; Nehaniv, Chrystopher L.; Phon-Amnuaisuk, Somnuk.
Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings. Vol. 10385 LNCS Springer Nature Link, 2017. p. 349-361 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10385 LNCS).

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

TY - GEN

T1 - Enhancing exploration and exploitation of NSGA-II with GP and PDL

AU - Shannon, Peter David

AU - Nehaniv, Chrystopher L.

AU - Phon-Amnuaisuk, Somnuk

PY - 2017/6/24

Y1 - 2017/6/24

N2 - In this paper, we show that NSGA-II can be applied to GP and the Process Description Language (PDL) and describe two modifications to NSGA-II. The first modification removes individuals which have the same behaviour from GP populations. It selects for de-duplication by taking the result of each objective fitness function together to make a comparison. NSGA-II is designed to expand its Pareto front of solutions by favouring individuals who have the highest or lowest value (boundary points) in a front, for any objective. The second modification enhances exploitation by preferring individuals who occupy an extreme position for most objective fitness functions. The results show, for the first time, that NSGA-II can be used with PDL and GP to successfully solve a robot control problem and that the suggested modifications offer significant improvements over an algorithm used previously with GP and PDL and unmodified NSGA-II for our test problem.

AB - In this paper, we show that NSGA-II can be applied to GP and the Process Description Language (PDL) and describe two modifications to NSGA-II. The first modification removes individuals which have the same behaviour from GP populations. It selects for de-duplication by taking the result of each objective fitness function together to make a comparison. NSGA-II is designed to expand its Pareto front of solutions by favouring individuals who have the highest or lowest value (boundary points) in a front, for any objective. The second modification enhances exploitation by preferring individuals who occupy an extreme position for most objective fitness functions. The results show, for the first time, that NSGA-II can be used with PDL and GP to successfully solve a robot control problem and that the suggested modifications offer significant improvements over an algorithm used previously with GP and PDL and unmodified NSGA-II for our test problem.

KW - Exploration and exploitation

KW - Genetic programming

KW - NSGA-II

KW - Process description language

UR - http://www.scopus.com/inward/record.url?scp=85026769534&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-61824-1_38

DO - 10.1007/978-3-319-61824-1_38

M3 - Conference contribution

AN - SCOPUS:85026769534

SN - 9783319618234

VL - 10385 LNCS

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

SP - 349

EP - 361

BT - Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings

PB - Springer Nature Link

T2 - 8th International Conference on Swarm Intelligence, ICSI 2017

Y2 - 27 July 2017 through 1 August 2017

ER -

Shannon PD, Nehaniv CL, Phon-Amnuaisuk S. Enhancing exploration and exploitation of NSGA-II with GP and PDL. In Advances in Swarm Intelligence - 8th International Conference, ICSI 2017, Proceedings. Vol. 10385 LNCS. Springer Nature Link. 2017. p. 349-361. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2017 Jun 24. doi: 10.1007/978-3-319-61824-1_38

Enhancing exploration and exploitation of NSGA-II with GP and PDL

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this