Linear Convergence of Evolution Strategies with Derandomized Sampling Beyond Quasi-Convex Functions

Jérémie Decock 1, 2 Olivier Teytaud 1, 2
1 TAO - Machine Learning and Optimisation
LRI - Laboratoire de Recherche en Informatique, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, CNRS - Centre National de la Recherche Scientifique : UMR8623
Abstract : We study the linear convergence of a simple evolutionary algorithm on non quasi-convex functions on continuous domains. Assumptions include an assumption on the sampling performed by the evolutionary algorithm (supposed to cover efficiently the neighborhood of the current search point), the conditioning of the objective function (so that the probability of improvement is not too low at each time step, given a correct step size), and the unicity of the optimum.
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.inria.fr/hal-00907671
Contributor : Jérémie Decock <>
Submitted on : Thursday, November 21, 2013 - 3:45:47 PM
Last modification on : Thursday, April 5, 2018 - 12:30:12 PM
Long-term archiving on : Saturday, February 22, 2014 - 4:41:18 AM

File

linearConvergence.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00907671, version 1

Collections

Citation

Jérémie Decock, Olivier Teytaud. Linear Convergence of Evolution Strategies with Derandomized Sampling Beyond Quasi-Convex Functions. EA - 11th Biennal International Conference on Artificial Evolution - 2013, Oct 2013, Bordeaux, France. ⟨hal-00907671⟩

Share

Metrics

Record views

428

Files downloads

205