Multi-strategy differential evolution

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

We propose the Multi-strategy Differential Evolution (MsDE) algorithm to construct and maintain a self-adaptive ensemble of search strategies while solving an optimization problem. The ensemble of strategies is represented as agents that interact with the candidate solutions to improve their fitness. In the proposed algorithm, the performance of each agent is measured so that successful strategies are promoted within the ensemble. We propose two performance measures, and show their effectiveness in selecting successful strategies. We then present three population adaptation mechanisms, based on sampling, clone-best and clone-multiple adaptation schemes. The MsDE with different performance measures and population adaptation schemes is tested on the CEC2013 benchmark functions and compared with basic DE and with Self-Adaptive DE (SaDE). Our results show that MsDE is capable of efficiently adapting the strategies and parameters of DE and providing competitive results with respect to the state-of-the-art
LanguageEnglish
Title of host publicationApplications of Evolutionary Computation
Subtitle of host publication21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings
EditorsK. Sim, P. Kaufmann
Place of PublicationDordrecht
PublisherSpringer
Pages617-633
ISBN (Electronic)978-3-319-77538-8
ISBN (Print)978-3-319-77537-1
DOIs
StatePublished - 4 Mar 2018
Event21st International Conference on the Applications of Evolutionary Computing (EvoApplications 2018) - Parma, Italy
Duration: 4 Apr 20186 Apr 2018
Conference number: 21
http://www.evostar.org/2018/cfp_evoapps.php

Publication series

NameLNCS
Volume10784

Conference

Conference21st International Conference on the Applications of Evolutionary Computing (EvoApplications 2018)
Abbreviated titleEvoApplications 2018
CountryItaly
CityParma
Period4/04/186/04/18
OtherHeld as part of the EvoStar 2018 event in Parma, Italy, April 2018
Internet address

Fingerprint

Sampling

Cite this

Yaman, A., Iacca, G., Coler, M., Fletcher, G. H. L., & Pechenizkiy, M. (2018). Multi-strategy differential evolution. In K. Sim, & P. Kaufmann (Eds.), Applications of Evolutionary Computation: 21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings (pp. 617-633). (LNCS; Vol. 10784). Dordrecht: Springer. DOI: 10.1007/978-3-319-77538-8_42
Yaman, A. ; Iacca, G. ; Coler, M. ; Fletcher, G.H.L. ; Pechenizkiy, M./ Multi-strategy differential evolution. Applications of Evolutionary Computation: 21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings. editor / K. Sim ; P. Kaufmann. Dordrecht : Springer, 2018. pp. 617-633 (LNCS).
@inproceedings{21527a8c0e6f4132ae38e73b60229dbf,
title = "Multi-strategy differential evolution",
abstract = "We propose the Multi-strategy Differential Evolution (MsDE) algorithm to construct and maintain a self-adaptive ensemble of search strategies while solving an optimization problem. The ensemble of strategies is represented as agents that interact with the candidate solutions to improve their fitness. In the proposed algorithm, the performance of each agent is measured so that successful strategies are promoted within the ensemble. We propose two performance measures, and show their effectiveness in selecting successful strategies. We then present three population adaptation mechanisms, based on sampling, clone-best and clone-multiple adaptation schemes. The MsDE with different performance measures and population adaptation schemes is tested on the CEC2013 benchmark functions and compared with basic DE and with Self-Adaptive DE (SaDE). Our results show that MsDE is capable of efficiently adapting the strategies and parameters of DE and providing competitive results with respect to the state-of-the-art",
author = "A. Yaman and G. Iacca and M. Coler and G.H.L. Fletcher and M. Pechenizkiy",
year = "2018",
month = "3",
day = "4",
doi = "10.1007/978-3-319-77538-8_42",
language = "English",
isbn = "978-3-319-77537-1",
series = "LNCS",
publisher = "Springer",
pages = "617--633",
editor = "K. Sim and P. Kaufmann",
booktitle = "Applications of Evolutionary Computation",
address = "Germany",

}

Yaman, A, Iacca, G, Coler, M, Fletcher, GHL & Pechenizkiy, M 2018, Multi-strategy differential evolution. in K Sim & P Kaufmann (eds), Applications of Evolutionary Computation: 21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings. LNCS, vol. 10784, Springer, Dordrecht, pp. 617-633, 21st International Conference on the Applications of Evolutionary Computing (EvoApplications 2018), Parma, Italy, 4/04/18. DOI: 10.1007/978-3-319-77538-8_42

Multi-strategy differential evolution. / Yaman, A.; Iacca, G.; Coler, M.; Fletcher, G.H.L.; Pechenizkiy, M.

Applications of Evolutionary Computation: 21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings. ed. / K. Sim; P. Kaufmann. Dordrecht : Springer, 2018. p. 617-633 (LNCS; Vol. 10784).

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Multi-strategy differential evolution

AU - Yaman,A.

AU - Iacca,G.

AU - Coler,M.

AU - Fletcher,G.H.L.

AU - Pechenizkiy,M.

PY - 2018/3/4

Y1 - 2018/3/4

N2 - We propose the Multi-strategy Differential Evolution (MsDE) algorithm to construct and maintain a self-adaptive ensemble of search strategies while solving an optimization problem. The ensemble of strategies is represented as agents that interact with the candidate solutions to improve their fitness. In the proposed algorithm, the performance of each agent is measured so that successful strategies are promoted within the ensemble. We propose two performance measures, and show their effectiveness in selecting successful strategies. We then present three population adaptation mechanisms, based on sampling, clone-best and clone-multiple adaptation schemes. The MsDE with different performance measures and population adaptation schemes is tested on the CEC2013 benchmark functions and compared with basic DE and with Self-Adaptive DE (SaDE). Our results show that MsDE is capable of efficiently adapting the strategies and parameters of DE and providing competitive results with respect to the state-of-the-art

AB - We propose the Multi-strategy Differential Evolution (MsDE) algorithm to construct and maintain a self-adaptive ensemble of search strategies while solving an optimization problem. The ensemble of strategies is represented as agents that interact with the candidate solutions to improve their fitness. In the proposed algorithm, the performance of each agent is measured so that successful strategies are promoted within the ensemble. We propose two performance measures, and show their effectiveness in selecting successful strategies. We then present three population adaptation mechanisms, based on sampling, clone-best and clone-multiple adaptation schemes. The MsDE with different performance measures and population adaptation schemes is tested on the CEC2013 benchmark functions and compared with basic DE and with Self-Adaptive DE (SaDE). Our results show that MsDE is capable of efficiently adapting the strategies and parameters of DE and providing competitive results with respect to the state-of-the-art

U2 - 10.1007/978-3-319-77538-8_42

DO - 10.1007/978-3-319-77538-8_42

M3 - Conference contribution

SN - 978-3-319-77537-1

T3 - LNCS

SP - 617

EP - 633

BT - Applications of Evolutionary Computation

PB - Springer

CY - Dordrecht

ER -

Yaman A, Iacca G, Coler M, Fletcher GHL, Pechenizkiy M. Multi-strategy differential evolution. In Sim K, Kaufmann P, editors, Applications of Evolutionary Computation: 21st International Conference, EvoApplications 2018, Parma, Italy, April 4-6, 2018, Proceedings. Dordrecht: Springer. 2018. p. 617-633. (LNCS). Available from, DOI: 10.1007/978-3-319-77538-8_42