Why one must use reweighting in Estimation Of Distribution Algorithms

2009
Teytaud, Fabien
Teytaud, Olivier

Abstract: Résumé : We study the update of the distribution in Estimation of Distribution Algorithms, and show that a simple modification leads to unbiased estimates of the optimum. The simple modification (based on a proper reweighting of estimates) leads to a strongly improved behavior in front of premature convergence.