Application of Particle Swarm Algorithm to Optimization of BP Neural Network

摘要：

To alleviate the drawback of easily getting into the local minimum appearing in the BP neural network, this paper applies the particle swarm algorithm to the optimization of BP neural network. Primarily, we utilize the particle swarm algorithm to get the optimal weight values as well as the optimal threshold values. Then, by assigning the optimal values to the corresponding parameters, we can optimize the BP neural network. Moreover, the performance of the optimized BP neural network is measured by means of the numerical method. The results demonstrate that, compared with the original BP neural network, the optimized BP neural network through the particle swarm algorithm can quickly converge to the preset target and hold the advantages of high calculation accuracy and quick convergence speed.

关键词： BP neural network particle swarm algorithm optimization

作者: Ling Zhang

作者单位: School of Information and Control Engineering, Weifang University, Weifang, Shandong, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 576-578

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）