A new evolutionary computation method

Author

Yan, WeI ; Zhu, Zhaoda

Author_Institution

Dept. of Electron. Eng., Nanjing Univ., China

Volume

2

fYear

1997

fDate

14-18 Jul 1997

Firstpage

803

Abstract

A real-valued genetic algorithm is proposed to the optimization problem with continuous variables. It is composed of a simple and general-purpose dynamic scaled fitness and selection operator, real-valued crossover operator, mutation operators and adaptive probabilities for these operators. The proposed algorithm is tested by two generally used functions and is applied to the training of a neural network for image recognition. Experimental results show that the proposed algorithm is an efficient global optimization algorithm

Keywords

image recognition; learning (artificial intelligence); mathematical operators; neural nets; optimisation; simulated annealing; adaptive operator; adaptive probabilities; continuous variables; dynamic scaled fitness and selection operator; efficient global optimization algorithm; elitist selection strategy; evolutionary computation method; function optimization; image recognition; mutation operators; neural network; optimization problem; real-valued crossover operator; real-valued genetic algorithm; simple general-purpose operator; training; Evolutionary computation; Frequency conversion; Frequency diversity; Genetic algorithms; Genetic mutations; Image recognition; Iterative algorithms; Neural networks; Testing; Wheels;

fLanguage

English

Publisher

ieee

Conference_Titel

Aerospace and Electronics Conference, 1997. NAECON 1997., Proceedings of the IEEE 1997 National

Conference_Location

Dayton, OH

Print_ISBN

0-7803-3725-5

Type

conf

DOI

10.1109/NAECON.1997.622732

Filename

622732