The performance effects of interaction frequency in parallel cooperative coevolution

Xiaofen LU; Stefan MENZEL; Ke TANG; Xin YAO

doi:10.1007/978-3-319-13563-2_8

The performance effects of interaction frequency in parallel cooperative coevolution

Xiaofen LU, Stefan MENZEL, Ke TANG, Xin YAO

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

1 Citation (Scopus)

Abstract

Cooperative coevolution (CC) employs a divide-and-conquer paradigm for tackling complex optimization problems. Its performance is influenced by many design decisions. Therefore, to beneficially use it, it is important to acquire some knowledge of the effects of different design settings on the performance of CC. In this paper, we investigate experimentally the performance effects of interaction frequency in parallel CC. The experimental results show that it is overall best for subpopulations to interact with each other as frequently as possible when communication cost is ignored; when communication cost is considered, the best interaction frequency varies from problem to problem and a dynamic change of it is desirable during the optimization process. © 2014, Springer International Publishing Switzerland.

Original language	English
Title of host publication	Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings
Editors	Grant DICK, Will N. BROWNE, Peter WHIGHAM, Mengjie ZHANG, Thu Bui LAM, Hisao ISHIBUCHI, Yaochu JIN, Xiaodong LI, Yuhui SHI, Pramod SINGH, Kay Chen TAN, Ke TANG
Publisher	Springer, Cham
Pages	82-93
Number of pages	12
ISBN (Electronic)	9783319135632
ISBN (Print)	9783319135625
DOIs	https://doi.org/10.1007/978-3-319-13563-2_8
Publication status	Published - 2014
Externally published	Yes
Event	10th International Conference on Simulated Evolution and Learning, SEAL 2014 - Dunedin, New Zealand Duration: 15 Dec 2014 → 18 Dec 2014

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer Cham
Volume	8886
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	10th International Conference on Simulated Evolution and Learning, SEAL 2014
Country/Territory	New Zealand
City	Dunedin
Period	15/12/14 → 18/12/14

Funding

This work was supported by Honda Research Institute Europe (HRI-EU) and European Commission FP7 grants (Nos. 257906 and 247619).

Keywords

Cooperative Coevolution
Interaction Frequency

Access to Document

10.1007/978-3-319-13563-2_8

Cite this

LU, X., MENZEL, S., TANG, K., & YAO, X. (2014). The performance effects of interaction frequency in parallel cooperative coevolution. In G. DICK, W. N. BROWNE, P. WHIGHAM, M. ZHANG, T. B. LAM, H. ISHIBUCHI, Y. JIN, X. LI, Y. SHI, P. SINGH, K. C. TAN, & K. TANG (Eds.), Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings (pp. 82-93). (Lecture Notes in Computer Science ; Vol. 8886). Springer, Cham. https://doi.org/10.1007/978-3-319-13563-2_8

LU, Xiaofen ; MENZEL, Stefan ; TANG, Ke et al. / The performance effects of interaction frequency in parallel cooperative coevolution. Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings. editor / Grant DICK ; Will N. BROWNE ; Peter WHIGHAM ; Mengjie ZHANG ; Thu Bui LAM ; Hisao ISHIBUCHI ; Yaochu JIN ; Xiaodong LI ; Yuhui SHI ; Pramod SINGH ; Kay Chen TAN ; Ke TANG. Springer, Cham, 2014. pp. 82-93 (Lecture Notes in Computer Science ).

@inproceedings{d0cde25066d4421f9dadf7571d9933af,

title = "The performance effects of interaction frequency in parallel cooperative coevolution",

abstract = "Cooperative coevolution (CC) employs a divide-and-conquer paradigm for tackling complex optimization problems. Its performance is influenced by many design decisions. Therefore, to beneficially use it, it is important to acquire some knowledge of the effects of different design settings on the performance of CC. In this paper, we investigate experimentally the performance effects of interaction frequency in parallel CC. The experimental results show that it is overall best for subpopulations to interact with each other as frequently as possible when communication cost is ignored; when communication cost is considered, the best interaction frequency varies from problem to problem and a dynamic change of it is desirable during the optimization process. {\textcopyright} 2014, Springer International Publishing Switzerland.",

keywords = "Cooperative Coevolution, Interaction Frequency",

author = "Xiaofen LU and Stefan MENZEL and Ke TANG and Xin YAO",

year = "2014",

doi = "10.1007/978-3-319-13563-2_8",

language = "English",

isbn = "9783319135625",

series = "Lecture Notes in Computer Science ",

publisher = "Springer, Cham",

pages = "82--93",

editor = "Grant DICK and BROWNE, {Will N.} and Peter WHIGHAM and Mengjie ZHANG and LAM, {Thu Bui} and Hisao ISHIBUCHI and Yaochu JIN and Xiaodong LI and Yuhui SHI and Pramod SINGH and TAN, {Kay Chen} and Ke TANG",

booktitle = "Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings",

note = "10th International Conference on Simulated Evolution and Learning, SEAL 2014 ; Conference date: 15-12-2014 Through 18-12-2014",

}

LU, X, MENZEL, S, TANG, K & YAO, X 2014, The performance effects of interaction frequency in parallel cooperative coevolution. in G DICK, WN BROWNE, P WHIGHAM, M ZHANG, TB LAM, H ISHIBUCHI, Y JIN, X LI, Y SHI, P SINGH, KC TAN & K TANG (eds), Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings. Lecture Notes in Computer Science , vol. 8886, Springer, Cham, pp. 82-93, 10th International Conference on Simulated Evolution and Learning, SEAL 2014, Dunedin, New Zealand, 15/12/14. https://doi.org/10.1007/978-3-319-13563-2_8

The performance effects of interaction frequency in parallel cooperative coevolution. / LU, Xiaofen; MENZEL, Stefan; TANG, Ke et al.
Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings. ed. / Grant DICK; Will N. BROWNE; Peter WHIGHAM; Mengjie ZHANG; Thu Bui LAM; Hisao ISHIBUCHI; Yaochu JIN; Xiaodong LI; Yuhui SHI; Pramod SINGH; Kay Chen TAN; Ke TANG. Springer, Cham, 2014. p. 82-93 (Lecture Notes in Computer Science ; Vol. 8886).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - The performance effects of interaction frequency in parallel cooperative coevolution

AU - LU, Xiaofen

AU - MENZEL, Stefan

AU - TANG, Ke

AU - YAO, Xin

PY - 2014

Y1 - 2014

N2 - Cooperative coevolution (CC) employs a divide-and-conquer paradigm for tackling complex optimization problems. Its performance is influenced by many design decisions. Therefore, to beneficially use it, it is important to acquire some knowledge of the effects of different design settings on the performance of CC. In this paper, we investigate experimentally the performance effects of interaction frequency in parallel CC. The experimental results show that it is overall best for subpopulations to interact with each other as frequently as possible when communication cost is ignored; when communication cost is considered, the best interaction frequency varies from problem to problem and a dynamic change of it is desirable during the optimization process. © 2014, Springer International Publishing Switzerland.

AB - Cooperative coevolution (CC) employs a divide-and-conquer paradigm for tackling complex optimization problems. Its performance is influenced by many design decisions. Therefore, to beneficially use it, it is important to acquire some knowledge of the effects of different design settings on the performance of CC. In this paper, we investigate experimentally the performance effects of interaction frequency in parallel CC. The experimental results show that it is overall best for subpopulations to interact with each other as frequently as possible when communication cost is ignored; when communication cost is considered, the best interaction frequency varies from problem to problem and a dynamic change of it is desirable during the optimization process. © 2014, Springer International Publishing Switzerland.

KW - Cooperative Coevolution

KW - Interaction Frequency

UR - http://www.scopus.com/inward/record.url?scp=84921717924&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-13563-2_8

DO - 10.1007/978-3-319-13563-2_8

M3 - Conference paper (refereed)

SN - 9783319135625

T3 - Lecture Notes in Computer Science

SP - 82

EP - 93

BT - Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings

A2 - DICK, Grant

A2 - BROWNE, Will N.

A2 - WHIGHAM, Peter

A2 - ZHANG, Mengjie

A2 - LAM, Thu Bui

A2 - ISHIBUCHI, Hisao

A2 - JIN, Yaochu

A2 - LI, Xiaodong

A2 - SHI, Yuhui

A2 - SINGH, Pramod

A2 - TAN, Kay Chen

A2 - TANG, Ke

PB - Springer, Cham

T2 - 10th International Conference on Simulated Evolution and Learning, SEAL 2014

Y2 - 15 December 2014 through 18 December 2014

ER -

LU X, MENZEL S, TANG K, YAO X. The performance effects of interaction frequency in parallel cooperative coevolution. In DICK G, BROWNE WN, WHIGHAM P, ZHANG M, LAM TB, ISHIBUCHI H, JIN Y, LI X, SHI Y, SINGH P, TAN KC, TANG K, editors, Simulated Evolution and Learning : 10th International Conference, SEAL 2014, Dunedin, New Zealand, December 15-18, Proceedings. Springer, Cham. 2014. p. 82-93. (Lecture Notes in Computer Science ). doi: 10.1007/978-3-319-13563-2_8

The performance effects of interaction frequency in parallel cooperative coevolution

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this