Enhancing evolutionary algorithms through recombination and parallelism

Authors
Gallard, Raúl Hector; Esquivel, Susana Cecilia
Publication Year
2001
Language
English
Format
article
Status
Published version
Description
Evolutionary computation (EC) has been recently recognized as a research field, which studies a new type of algorithms: Evolutionary Algorithms (EAs). These algorithms process populations of solutions as opposed to most traditional approaches which improve a single solution. All these algorithms share common features: reproduction, random variation, competition and selection of individuals. During our research it was evident that some components of EAs should be re-examined. Hence, specific topics such as multiple crossovers per couple and its enhancements, multiplicity of parents and crossovers and their application to single and multiple criteria optimization problems, adaptability, and parallel genetic algorithms, were proposed and investigated carefully. This paper show the most relevant and recent enhancements on recombination for a genetic-algorithm-based EA and migration control strategies for parallel genetic algorithms. Details of implementation and results are discussed.
Facultad de Informática
Subject
Ciencias Informáticas
Access level
Open access
License
Creative Commons Attribution-NonCommercial 3.0 Unported (CC BY-NC 3.0)
Repository
SEDICI (UNLP)
Institution
Universidad Nacional de La Plata
OAI Identifier
oai:sedici.unlp.edu.ar:10915/9423