Objective Improvement in Information-Geometric Optimization - Archive ouverte HAL Access content directly
Conference Papers Year : 2013

Objective Improvement in Information-Geometric Optimization

(1) , (1)
1
Youhei Akimoto
  • Function : Author
  • PersonId : 924013
Yann Ollivier
  • Function : Author
  • PersonId : 883809

Abstract

Information-Geometric Optimization (IGO) is a unified framework of stochastic algorithms for optimization problems. Given a family of probability distributions, IGO turns the original optimization problem into a new maximization problem on the parameter space of the probability distributions. IGO updates the parameter of the probability distribution along the natural gradient, taken with respect to the Fisher metric on the parameter manifold, aiming at maximizing an adaptive transform of the objective function. IGO recovers several known algorithms as particular instances: for the family of Bernoulli distributions IGO recovers PBIL, for the family of Gaussian distributions the pure rank-mu CMA-ES update is recovered, and for exponential families in expectation parametrization the cross-entropy/ML method is recovered. This article provides a theoretical justification for the IGO framework, by proving that any step size not greater than 1 guarantees monotone improvement over the course of optimization, in terms of q-quantile values of the objective function f. The range of admissible step sizes is independent of f and its domain. We extend the result to cover the case of different step sizes for blocks of the parameters in the IGO algorithm. Moreover, we prove that expected fitness improves over time when fitness-proportional selection is applied, in which case the RPP algorithm is recovered.
Fichier principal
Vignette du fichier
sigproc-sp.pdf (237.32 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00752489 , version 1 (15-11-2012)
hal-00752489 , version 2 (29-11-2012)
hal-00752489 , version 3 (07-03-2013)

Identifiers

Cite

Youhei Akimoto, Yann Ollivier. Objective Improvement in Information-Geometric Optimization. Foundations of Genetic Algorithms XII, Jan 2013, Adelaide, Australia. ⟨hal-00752489v3⟩
309 View
266 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More