Estimation of Distribution Algorithms Based on the Beta Distribution for Bounded Search Spaces
| Autor: | Arturo Hernández-Aguirre, José Antonio Guerrero-Díaz de León, Rogelio Salinas-Gutiérrez, Ángel Eduardo Muñoz-Zavala |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Mathematical optimization
Optimization problem Estimation theory Asymptotic distribution 02 engineering and technology Method of moments (statistics) 01 natural sciences 010104 statistics & probability Estimation of distribution algorithm Bounded function 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing 0101 mathematics Metaheuristic Beta distribution Mathematics |
| Zdroj: | Advances in Soft Computing ISBN: 9783319624273 MICAI (2) |
| DOI: | 10.1007/978-3-319-62428-0_13 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|