RUS  ENG
Full version
JOURNALS // Computer Research and Modeling // Archive

Computer Research and Modeling, 2015 Volume 7, Issue 3, Pages 775–780 (Mi crm248)

This article is cited in 2 papers

ÑÒÅÍÄÎÂÛÅ ÄÎÊËÀÄÛ

Non-uniform cellular genetic algorithms

N. M. Ershov

aculty of Computational Mathematics and Cybernetics, Lomonosov Moscow State University, 1-2 Leninskiye Gory, GSP-1, Moscow, 119991, Russia

Abstract: In this paper, we introduce the concept of non-uniform cellular genetic algorithm, in which a number of parameters that affect the operation of genetic operators is dependent on the location of the cells of a given cellular space. The results of numerical comparison of non-uniform cellular genetic algorithms with the standard genetic algorithms, showing the advantages of the proposed approach while minimizing multimodal functions with a large number of local extrema, are presented. The coarse-grained parallel implementation of the non-uniform algorithms using the technology of MPI is considered.

Keywords: evolutionary algorithms, genetic algorithms, parallel computing.

UDC: 004.021

Received: 17.09.2014

DOI: 10.20537/2076-7633-2015-7-3-775-780



© Steklov Math. Inst. of RAS, 2026