Evaluation of the impact of various modifications to CMA-ES that facilitate its theoretical analysis - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Evaluation of the impact of various modifications to CMA-ES that facilitate its theoretical analysis

Abstract

In this paper we introduce modified versions of CMA-ES with the objective to help to prove convergence of CMA-ES. In order to ensure that the modifications do not alter the performances of the algorithm too much, we benchmark variants of the algorithm derived from them on problems of the bbob test suite. We observe that the main performances losses are observed on ill-conditioned problems, which is probably due to the absence of cumulation in the adaptation of the covariance matrix. However, the versions of CMA-ES presented in this paper have globally similar performances to the original CMA-ES.
Fichier principal
Vignette du fichier
evaluation2023author_version.pdf (2.49 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Licence
Copyright

Dates and versions

hal-04089923 , version 1 (05-05-2023)
hal-04089923 , version 2 (09-05-2023)

Licence

Copyright

Identifiers

Cite

Armand Gissler. Evaluation of the impact of various modifications to CMA-ES that facilitate its theoretical analysis. GECCO 2023 - Genetic and Evolutionary Computation Conference, Jul 2023, Lisbon, Portugal. ⟨10.1145/3583133.3596329⟩. ⟨hal-04089923v2⟩
54 View
115 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More