Abstract
This paper presents a cellular version of Differential Evolution (DE) algorithm. The notion behind the geographical distribution of DE population with local interaction is to study the influence of slow diffusion of information throughout the population. The study was carried out using the compact configuration of neighborhood from which all the auxiliary parents for DE recombination were selected. The empirical study was carried out using a standard benchmark suite consisting of 10 functions. The results show that the structured population with local interaction improves the convergence characteristics of DE and the performance improvement was also verified using scalability study. A brief comparison with cellular GA was also included. © 2010 Springer-Verlag.
Cite
CITATION STYLE
Noman, N., & Iba, H. (2010). Cellular differential evolution algorithm. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6464 LNAI, pp. 293–302). https://doi.org/10.1007/978-3-642-17432-2_30
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
