
A discrete version of CMA-ES
Benhamou, Eric; Atif, Jamal; Laraki, Rida (2018), A discrete version of CMA-ES. https://basepub.dauphine.fr/handle/123456789/18927
View/ Open
Type
Document de travail / Working paperDate
2018Publisher
Preprint Lamsade
Series title
Preprint LamsadePublished in
Paris
Pages
13
Metadata
Show full item recordAuthor(s)
Benhamou, EricLaboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Atif, Jamal
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Laraki, Rida

Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Abstract (EN)
Modern machine learning uses more and more advanced optimization techniques to find optimal hyper parameters. Whenever the objective function is non-convex, non continuous and with potentially multiple local minima, standard gradient descent optimization methods fail. A last resource and very different method is to assume that the optimum(s), not necessarily unique, is/are distributed according to a distribution and iteratively to adapt the distribution according to tested points. These strategies originated in the early 1960s, named Evolution Strategy (ES) have culminated with the CMA-ES (Covariance Matrix Adaptation) ES. It relies on a multi variate normal distribution and is supposed to be state of the art for general optimization program. However, it is far from being optimal for discrete variables. In this paper, we extend the method to multivariate binomial correlated distributions. For such a distribution, we show that it shares similar features to the multi variate normal: independence and correlation is equivalent and correlation is efficiently modeled by interaction between different variables. We discuss this distribution in the framework of the exponential family. We prove that the model can estimate not only pairwise interactions among the two variables but also is capable of modeling higher order interactions. This allows creating a version of CMA ES that can accomodate efficiently discrete variables. We provide the corresponding algorithm and conclude.Subjects / Keywords
Covariance Matrix Adaptation; Evolution StrategyRelated items
Showing items related by title and author.
-
Benhamou, Éric; Saltiel, David; Ohana, Jean-Jacques; Atif, Jamal; Laraki, Rida Communication / Conférence
-
Benhamou, Eric; Atif, Jamal; Laraki, Rida (2019-01) Document de travail / Working paper
-
Benhamou, Eric; Atif, Jamal; Laraki, Rida (2018) Document de travail / Working paper
-
Benhamou, Éric; Saltiel, David; Laraki, Rida; Atif, Jamal (2020) Document de travail / Working paper
-
Benhamou, Éric; Atif, Jamal; Laraki, Rida; Saltiel, David (2020) Document de travail / Working paper