东北大学学报:自然科学版 ›› 2020, Vol. 41 ›› Issue (2): 176-182.DOI: 10.12068/j.issn.1005-3026.2020.02.005

• 信息与控制 • 上一篇    下一篇

基于参数优化的RBF神经网络结构设计算法

翟莹莹, 左丽, 张恩德   

  1. (东北大学 计算机科学与工程学院, 辽宁 沈阳110169)
  • 收稿日期:2019-03-11 修回日期:2019-03-11 出版日期:2020-02-15 发布日期:2020-03-06
  • 通讯作者: 翟莹莹
  • 作者简介:翟莹莹(1984-),女,黑龙江望奎人,东北大学讲师,博士.冯明杰(1971-), 男, 河南禹州人, 东北大学副教授; 王恩刚(1962-), 男, 辽宁沈阳人, 东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(51704063,51834004,51774076); 中央高校基本科研业务费专项资金资助项目(N171603015,N181604016); “十三五”国家重点研发计划项目(2017YFB0304200).

Algorithm for Structure Design of RBF Neural Network Based on Parameter Optimization

ZHAI Ying-ying, ZUO Li, ZHANG En-de   

  1. School of Computer Science & Engineering, Northeastern University, Shenyang 110169, China;
  • Received:2019-03-11 Revised:2019-03-11 Online:2020-02-15 Published:2020-03-06
  • Contact: ZUO Li
  • About author:-
  • Supported by:
    -

摘要: 针对径向基函数(RBF)神经网络的结构优化问题,提出了一种基于参数优化的RBF神经网络优化算法.首先,改进K-means++算法,使得聚类算法更精确,为RBF神经网络的隐含层节点找到一个合适的初始中心;然后,考虑数据分布和缩放因子选择的影响,采用方差度量法计算隐含层神经元基函数的宽度;最后,修正网络参数,提高网络的非线性逼近能力.实验结果表明,本文提出的基于参数优化的RBF神经网络具有良好的逼近效果和泛化能力.

关键词: 径向基函数, 神经网络, K-means++算法, 核函数, 数据密度估计

Abstract: An algorithm based on parameter optimization(KV-RBF)is proposed for optimizing RBF(radial basis function) neural network structure. Firstly, the K-means++ algorithm, which makes the clustering algorithm more accurate, is improved to find an appropriate initial center for the hidden layer node of RBF neural network. Then, considering the influence of data distribution and scaling factor selection, the width of neuron basis function in the hidden layer is calculated by variance measurement. Finally, the network parameters are modified to improve the nonlinear approximation ability of the network. The experimental results show that the proposed RBF neural network based on parameter optimization has good approximation effect and generalization ability.

Key words: radial basis function(RBF), neural network, K-means++ algorithm, kernel function, data density estimation

中图分类号: