CVSL Logo
FrancaisHome
AboutPeopleResearchPublicationsEventsProfile
About
Publications

 

 

 

CERVIM

REPARTI

MIVIM

Resource-Aware Parameterizations of EDA


Sylvain Gelly, Olivier Teytaud and Christian Gagné


Abstract - This paper presents a framework for the theoretical analysis of Estimation of Distribution Algorithms (EDA). Using this framework, derived from the VC-theory, we propose non-asymptotic bounds which depend on: 1) the population size, 2) the selection rate, 3) the families of distributions used for the modelling, 4) the dimension, and 5) the number of iterations. To validate these results, optimization algorithms are applied to a context where bounds on resources are crucial, namely Design of Experiments, that is a black-box optimization with very few fitness-values evaluations.

download document

Bibtex:

@inproceedings{and 621,
    author    = { Sylvain Gelly and Olivier Teytaud and Christian Gagné },
    title     = { Resource-Aware Parameterizations of EDA },
    booktitle = { Proc. of the 2006 IEEE Congress on Evolutionary Computation (IEEE-CEC 2006) },
    year      = { 2006 },
    month     = { July 16-21 },
    location  = { Vancouver (BC), Canada }
}

Last modification: 2006/05/23 by cgagne

     
 
Home | About | People | Research | Publications | Events | Profile
 
   

©2002-. Computer Vision and Systems Laboratory. All rights reserved