Estimation of Distribution Algorithms Based on the Beta Distribution for Bounded Search Spaces.

This work presents a metaheuristic based on the use of the beta distribution as a search distribution for solving numerical optimization problems in search spaces defined on two sided intervals. The innovation of this work lies on the efficiency of the proposed method to estimate the parameters of the beta distribution with a minimal cost for each decision variable by using the method of moments. The numerical experiments provided evidence that applying the method of moments for parameter estimation and the beta distribution as a search distribution generates competitive results.