A Learning Algorithm of Artificial Neural Network Based on GA - PSO

Author

Du, Shiqiang ; Li, Wanshe ; Cao, Kai

Author_Institution

Coll. of Mathematic & Inf. Sci., Shaanxi Normal Univ., Xi´´an

Volume

1

fYear

0

fDate

0-0 0

Firstpage

3633

Lastpage

3637

Abstract

In order to get over the insufficiency of back-propagation (BP) algorithm, after analyses of genetic algorithm (GA) and particle swarm optimization (PSO), a GA-PSO algorithm is proposed. In GA-PSO, individuals in a new generation are created, not only by crossover and mutation operation in GA, but also by PSO, based on redefined local optimization swarm. So it can both avoid local minimum and has good global search capacity. The performance of GA-PSO is compared to both GA and PSO in artificial neural networks weight training, demonstrating its superiority

Keywords

backpropagation; genetic algorithms; neural nets; particle swarm optimisation; artificial neural network weight training; backpropagation; crossover operation; genetic algorithm; global search capacity; learning algorithm; mutation operation; particle swarm optimization; Algorithm design and analysis; Artificial neural networks; Educational institutions; Genetic algorithms; Genetic mutations; Information analysis; Information science; Mathematics; Multi-layer neural network; Particle swarm optimization; BP algorithm; GA-PSO algorithm; genetic algorithm; particle swarm optimization;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2006. WCICA 2006. The Sixth World Congress on

Conference_Location

Dalian

Print_ISBN

1-4244-0332-4

Type

conf

DOI

10.1109/WCICA.2006.1713047

Filename

1713047