Title :
Research of a genetic algorithms designer
Author :
Komartsova, L.G.
Author_Institution :
Kaluga Branch, Moscow Bauman State Tech. Univ., Russia
Abstract :
Basic problems of genetic algorithm (GA) usage effectiveness increase during the solution of optimization, search and adaptation problems are investigated in this study. The greatest influence on the speed of convergence and the probability of the best solution finding is made by the selective types of the genetic operators and other parameters, which ensure the greatest effectiveness of GA work. For the solution of these problems the paper proposes to use a Designer allowing to automatically select the GA parameters. With the aid of the GA parameters automatic selection at the solution of the continuous and combinatory optimization problems it becomes possible to decrease the number of iterations and to increase the probability of extreme finding.
Keywords :
convergence; genetic algorithms; search problems; adaptation problems; convergence; extreme finding; genetic algorithm designer; genetic operators; optimization; parameter selection; probability; search; Algorithm design and analysis; Biological cells; Computational complexity; Computer networks; Computer science education; Design optimization; Genetic algorithms; Iterative decoding; Neural networks; Neurons;
Conference_Titel :
Artificial Intelligence Systems, 2002. (ICAIS 2002). 2002 IEEE International Conference on
Print_ISBN :
0-7695-1733-1
DOI :
10.1109/ICAIS.2002.1048136