Improving (1+1) covariance matrix adaptation evolution strategy: a simple yet efficient approach

Fabio Caraffini, Giovanni Iacca, Anil Yaman

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

Abstract

In recent years, part of the meta-heuristic optimisation research community has called for a simplification of the algorithmic design: indeed, while most of the state-of-The-Art algorithms are characterised by a high level of complexity, complex algorithms are hard to understand and therefore tune for specific real-world applications. Here, we follow this reductionist approach by combining straightforwardly two methods recently proposed in the literature, namely the Re-sampling Inheritance Search (RIS) and the (1+1) Covariance Matrix Adaptation Evolution Strategy (CMA-ES). We propose an RI-(1+1)-CMA-ES algorithm that on the one hand improves upon the original (1+1)-CMA-ES, on the other it keeps the original spirit of simplicity at the basis of RIS. We show with an extensive experimental campaign that the proposed algorithm efficiently solves a large number of benchmark functions and is competitive with several modern optimisation algorithms much more complex in terms of algorithmic design.

LanguageEnglish
Title of host publicationProceedings LeGO 2018 : 14th International Global Optimization Workshop
EditorsAndre H. Deutz, Sander C. Hille, Yaroslav D. Sergeyev, Michael T. M. Emmerich
Place of PublicationMaryland
PublisherAmerican Institute of Physics
Number of pages4
ISBN (Electronic)9780735417984
DOIs
StatePublished - 12 Feb 2019
Event14th International Global Optimization Workshop, LeGO 2018 - Leiden, Netherlands
Duration: 18 Sep 201821 Sep 2018

Publication series

NameAIP Conference Proceedings
Volume2070

Conference

Conference14th International Global Optimization Workshop, LeGO 2018
CountryNetherlands
CityLeiden
Period18/09/1821/09/18

Fingerprint

sampling
optimization
simplification

Cite this

Caraffini, F., Iacca, G., & Yaman, A. (2019). Improving (1+1) covariance matrix adaptation evolution strategy: a simple yet efficient approach. In A. H. Deutz, S. C. Hille, Y. D. Sergeyev, & M. T. M. Emmerich (Eds.), Proceedings LeGO 2018 : 14th International Global Optimization Workshop [20004] (AIP Conference Proceedings; Vol. 2070). Maryland: American Institute of Physics. DOI: 10.1063/1.5089971
Caraffini, Fabio ; Iacca, Giovanni ; Yaman, Anil. / Improving (1+1) covariance matrix adaptation evolution strategy : a simple yet efficient approach. Proceedings LeGO 2018 : 14th International Global Optimization Workshop. editor / Andre H. Deutz ; Sander C. Hille ; Yaroslav D. Sergeyev ; Michael T. M. Emmerich. Maryland : American Institute of Physics, 2019. (AIP Conference Proceedings).
@inproceedings{d3ae70867717426ea9fae64f6b13e517,
title = "Improving (1+1) covariance matrix adaptation evolution strategy: a simple yet efficient approach",
abstract = "In recent years, part of the meta-heuristic optimisation research community has called for a simplification of the algorithmic design: indeed, while most of the state-of-The-Art algorithms are characterised by a high level of complexity, complex algorithms are hard to understand and therefore tune for specific real-world applications. Here, we follow this reductionist approach by combining straightforwardly two methods recently proposed in the literature, namely the Re-sampling Inheritance Search (RIS) and the (1+1) Covariance Matrix Adaptation Evolution Strategy (CMA-ES). We propose an RI-(1+1)-CMA-ES algorithm that on the one hand improves upon the original (1+1)-CMA-ES, on the other it keeps the original spirit of simplicity at the basis of RIS. We show with an extensive experimental campaign that the proposed algorithm efficiently solves a large number of benchmark functions and is competitive with several modern optimisation algorithms much more complex in terms of algorithmic design.",
author = "Fabio Caraffini and Giovanni Iacca and Anil Yaman",
year = "2019",
month = "2",
day = "12",
doi = "10.1063/1.5089971",
language = "English",
series = "AIP Conference Proceedings",
publisher = "American Institute of Physics",
editor = "Deutz, {Andre H.} and Hille, {Sander C.} and Sergeyev, {Yaroslav D.} and Emmerich, {Michael T. M.}",
booktitle = "Proceedings LeGO 2018 : 14th International Global Optimization Workshop",
address = "United States",

}

Caraffini, F, Iacca, G & Yaman, A 2019, Improving (1+1) covariance matrix adaptation evolution strategy: a simple yet efficient approach. in AH Deutz, SC Hille, YD Sergeyev & MTM Emmerich (eds), Proceedings LeGO 2018 : 14th International Global Optimization Workshop., 20004, AIP Conference Proceedings, vol. 2070, American Institute of Physics, Maryland, 14th International Global Optimization Workshop, LeGO 2018, Leiden, Netherlands, 18/09/18. DOI: 10.1063/1.5089971

Improving (1+1) covariance matrix adaptation evolution strategy : a simple yet efficient approach. / Caraffini, Fabio; Iacca, Giovanni; Yaman, Anil.

Proceedings LeGO 2018 : 14th International Global Optimization Workshop. ed. / Andre H. Deutz; Sander C. Hille; Yaroslav D. Sergeyev; Michael T. M. Emmerich. Maryland : American Institute of Physics, 2019. 20004 (AIP Conference Proceedings; Vol. 2070).

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

TY - GEN

T1 - Improving (1+1) covariance matrix adaptation evolution strategy

T2 - a simple yet efficient approach

AU - Caraffini,Fabio

AU - Iacca,Giovanni

AU - Yaman,Anil

PY - 2019/2/12

Y1 - 2019/2/12

N2 - In recent years, part of the meta-heuristic optimisation research community has called for a simplification of the algorithmic design: indeed, while most of the state-of-The-Art algorithms are characterised by a high level of complexity, complex algorithms are hard to understand and therefore tune for specific real-world applications. Here, we follow this reductionist approach by combining straightforwardly two methods recently proposed in the literature, namely the Re-sampling Inheritance Search (RIS) and the (1+1) Covariance Matrix Adaptation Evolution Strategy (CMA-ES). We propose an RI-(1+1)-CMA-ES algorithm that on the one hand improves upon the original (1+1)-CMA-ES, on the other it keeps the original spirit of simplicity at the basis of RIS. We show with an extensive experimental campaign that the proposed algorithm efficiently solves a large number of benchmark functions and is competitive with several modern optimisation algorithms much more complex in terms of algorithmic design.

AB - In recent years, part of the meta-heuristic optimisation research community has called for a simplification of the algorithmic design: indeed, while most of the state-of-The-Art algorithms are characterised by a high level of complexity, complex algorithms are hard to understand and therefore tune for specific real-world applications. Here, we follow this reductionist approach by combining straightforwardly two methods recently proposed in the literature, namely the Re-sampling Inheritance Search (RIS) and the (1+1) Covariance Matrix Adaptation Evolution Strategy (CMA-ES). We propose an RI-(1+1)-CMA-ES algorithm that on the one hand improves upon the original (1+1)-CMA-ES, on the other it keeps the original spirit of simplicity at the basis of RIS. We show with an extensive experimental campaign that the proposed algorithm efficiently solves a large number of benchmark functions and is competitive with several modern optimisation algorithms much more complex in terms of algorithmic design.

UR - http://www.scopus.com/inward/record.url?scp=85061895804&partnerID=8YFLogxK

U2 - 10.1063/1.5089971

DO - 10.1063/1.5089971

M3 - Conference contribution

T3 - AIP Conference Proceedings

BT - Proceedings LeGO 2018 : 14th International Global Optimization Workshop

PB - American Institute of Physics

CY - Maryland

ER -

Caraffini F, Iacca G, Yaman A. Improving (1+1) covariance matrix adaptation evolution strategy: a simple yet efficient approach. In Deutz AH, Hille SC, Sergeyev YD, Emmerich MTM, editors, Proceedings LeGO 2018 : 14th International Global Optimization Workshop. Maryland: American Institute of Physics. 2019. 20004. (AIP Conference Proceedings). Available from, DOI: 10.1063/1.5089971