Thumbnail

On Averaging the Best Samples in Evolutionary Computation

Meunier, Laurent; Chevaleyre, Yann; Rapin, J.; Royer, Clément; Teytaud, O. (2020), On Averaging the Best Samples in Evolutionary Computation, in Bäck, Thomas; Preuss, Mike; Deutz, André, Parallel Problem Solving from Nature – PPSN XVI, Proceedings, Part I, Springer : Berlin Heidelberg, p. 661-674. 10.1007/978-3-030-58115-2_46

View/Open
Type
Communication / Conférence
Date
2020
Conference title
16th International Conference, PPSN 2020
Conference date
2020-09
Conference city
Leiden
Conference country
Netherlands
Book title
Parallel Problem Solving from Nature – PPSN XVI, Proceedings, Part I
Book author
Bäck, Thomas; Preuss, Mike; Deutz, André
Publisher
Springer
Published in
Berlin Heidelberg
ISBN
978-3-030-58115-2
Number of pages
735
Pages
661-674
Publication identifier
Metadata
Show full item record
Author(s)
Meunier, Laurent
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Chevaleyre, Yann
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Rapin, J.
Royer, Clément
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Teytaud, O.
Abstract (EN)
Choosing the right selection rate is a long standing issue in evolutionary computation. In the continuous unconstrained case, we prove mathematically that a single parent μ=1 leads to a sub-optimal simple regret in the case of the sphere function. We provide a theoretically-based selection rate μ/λ that leads to better progress rates. With our choice of selection rate, we get a provable regret of order O(λ−1) which has to be compared with O(λ−2/d) in the case where μ=1 . We complete our study with experiments to confirm our theoretical claims.
Subjects / Keywords
evolutionary computation