The Open Cybernetics & Systemics Journal

2014, 8 : 519-524
Published online 2014 December 31. DOI: 10.2174/1874110X01408010519
Publisher ID: TOCSJ-8-519

The Algorithms Optimization of Artificial Neural Network Based on Particle Swarm

Yang Xin-quan
Computer and Information Engineering Department, HeZe University of Heze, Shandong, China.

ABSTRACT

As a new kind of swarm intelligence algorithm, particle swarm optimization (PSO) algorithm can be calculated conveniently to achieve fast convergence and good convergence performance advantages. However, it shows shortcoming of falling into local extreme point. In this paper, a harmony search algorithm was used to improve PSO. Harmony Search Algorithm, as a new optimization algorithm, presents a good global search performance. By examining four standard test functions, the accuracy of convergence speed or convergence using improved PSO harmony search algorithm was validated.

Keywords:

Neural network, optimization, particle swarm optimization algorithm.