Benchmarking GNN-CMA-ES on the BBOB noiseless testbed

Louis Faury; Clément Calauzènes; Olivier Fercoq

doi:10.1145/3319619.3326856

Benchmarking GNN-CMA-ES on the BBOB noiseless testbed

Louis Faury
, Clément Calauzènes
, Olivier Fercoq

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

Abstract

We evaluate in this paper the GNN-CMA-ES algorithm on the BBOB noiseless testbed. The GNN-CMA-ES algorithm was recently proposed as a plug-in extension to CMA-ES, introducing the possibility to train flexible search distributions, in contrast to standard search distributions (such as the multivariate Gaussian). By comparing GNN-CMA-ES and CMA-ES, we show the benefits of this extension on some unimodal functions as well as on a variety of multimodal functions. We also identify a family of unimodal functions where GNN-CMA-ES can degrade the performances of CMA-ES and discuss the possible reasons behind this behavior.

Original language	English
Title of host publication	GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion
Publisher	Association for Computing Machinery, Inc
Pages	1928-1936
Number of pages	9
ISBN (Electronic)	9781450367486
DOIs	https://doi.org/10.1145/3319619.3326856
Publication status	Published - 13 Jul 2019
Event	2019 Genetic and Evolutionary Computation Conference, GECCO 2019 - Prague, Czech Republic Duration: 13 Jul 2019 → 17 Jul 2019

Publication series

Name	GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion

Conference

Conference	2019 Genetic and Evolutionary Computation Conference, GECCO 2019
Country/Territory	Czech Republic
City	Prague
Period	13/07/19 → 17/07/19

Keywords

Benchmarking
Black-box optimization
Evolutionary Strategies
Generative Neural Networks

Access to Document

10.1145/3319619.3326856

Cite this

Faury, L., Calauzènes, C., & Fercoq, O. (2019). Benchmarking GNN-CMA-ES on the BBOB noiseless testbed. In GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion (pp. 1928-1936). (GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion). Association for Computing Machinery, Inc. https://doi.org/10.1145/3319619.3326856

Faury, Louis ; Calauzènes, Clément ; Fercoq, Olivier. / Benchmarking GNN-CMA-ES on the BBOB noiseless testbed. GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, Inc, 2019. pp. 1928-1936 (GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion).

@inproceedings{0eb244e5df8943de9f08234205ab4dae,

title = "Benchmarking GNN-CMA-ES on the BBOB noiseless testbed",

abstract = "We evaluate in this paper the GNN-CMA-ES algorithm on the BBOB noiseless testbed. The GNN-CMA-ES algorithm was recently proposed as a plug-in extension to CMA-ES, introducing the possibility to train flexible search distributions, in contrast to standard search distributions (such as the multivariate Gaussian). By comparing GNN-CMA-ES and CMA-ES, we show the benefits of this extension on some unimodal functions as well as on a variety of multimodal functions. We also identify a family of unimodal functions where GNN-CMA-ES can degrade the performances of CMA-ES and discuss the possible reasons behind this behavior.",

keywords = "Benchmarking, Black-box optimization, Evolutionary Strategies, Generative Neural Networks",

author = "Louis Faury and Cl{\'e}ment Calauz{\`e}nes and Olivier Fercoq",

note = "Publisher Copyright: {\textcopyright} 2019 Copyright held by the owner/author(s). Publication rights licensed to Association for Computing Machinery.; 2019 Genetic and Evolutionary Computation Conference, GECCO 2019 ; Conference date: 13-07-2019 Through 17-07-2019",

year = "2019",

month = jul,

day = "13",

doi = "10.1145/3319619.3326856",

language = "English",

series = "GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion",

publisher = "Association for Computing Machinery, Inc",

pages = "1928--1936",

booktitle = "GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion",

}

Faury, L, Calauzènes, C & Fercoq, O 2019, Benchmarking GNN-CMA-ES on the BBOB noiseless testbed. in GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion, Association for Computing Machinery, Inc, pp. 1928-1936, 2019 Genetic and Evolutionary Computation Conference, GECCO 2019, Prague, Czech Republic, 13/07/19. https://doi.org/10.1145/3319619.3326856

Benchmarking GNN-CMA-ES on the BBOB noiseless testbed. / Faury, Louis; Calauzènes, Clément; Fercoq, Olivier.
GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, Inc, 2019. p. 1928-1936 (GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion).

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

TY - GEN

T1 - Benchmarking GNN-CMA-ES on the BBOB noiseless testbed

AU - Faury, Louis

AU - Calauzènes, Clément

AU - Fercoq, Olivier

PY - 2019/7/13

Y1 - 2019/7/13

N2 - We evaluate in this paper the GNN-CMA-ES algorithm on the BBOB noiseless testbed. The GNN-CMA-ES algorithm was recently proposed as a plug-in extension to CMA-ES, introducing the possibility to train flexible search distributions, in contrast to standard search distributions (such as the multivariate Gaussian). By comparing GNN-CMA-ES and CMA-ES, we show the benefits of this extension on some unimodal functions as well as on a variety of multimodal functions. We also identify a family of unimodal functions where GNN-CMA-ES can degrade the performances of CMA-ES and discuss the possible reasons behind this behavior.

AB - We evaluate in this paper the GNN-CMA-ES algorithm on the BBOB noiseless testbed. The GNN-CMA-ES algorithm was recently proposed as a plug-in extension to CMA-ES, introducing the possibility to train flexible search distributions, in contrast to standard search distributions (such as the multivariate Gaussian). By comparing GNN-CMA-ES and CMA-ES, we show the benefits of this extension on some unimodal functions as well as on a variety of multimodal functions. We also identify a family of unimodal functions where GNN-CMA-ES can degrade the performances of CMA-ES and discuss the possible reasons behind this behavior.

KW - Benchmarking

KW - Black-box optimization

KW - Evolutionary Strategies

KW - Generative Neural Networks

U2 - 10.1145/3319619.3326856

DO - 10.1145/3319619.3326856

M3 - Conference contribution

AN - SCOPUS:85070628621

T3 - GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion

SP - 1928

EP - 1936

BT - GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion

PB - Association for Computing Machinery, Inc

T2 - 2019 Genetic and Evolutionary Computation Conference, GECCO 2019

Y2 - 13 July 2019 through 17 July 2019

ER -

Faury L, Calauzènes C, Fercoq O. Benchmarking GNN-CMA-ES on the BBOB noiseless testbed. In GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion. Association for Computing Machinery, Inc. 2019. p. 1928-1936. (GECCO 2019 Companion - Proceedings of the 2019 Genetic and Evolutionary Computation Conference Companion). doi: 10.1145/3319619.3326856

Benchmarking GNN-CMA-ES on the BBOB noiseless testbed

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this