Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (2): 29-31.DOI: 10.3778/j.issn.1002-8331.2011.02.009

• 研究、探讨 • Previous Articles     Next Articles

Optimal-structure determination of power-activation feed-forward neural net

ZHANG Yunong1,GUO Dongsheng1,TAN Ning2   

  1. 1.School of Information Science and Technology,Sun Yat-Sen University,Guangzhou 510275,China
    2.School of Software,Sun Yat-Sen University,Guangzhou 510275,China
  • Received:2009-09-28 Revised:2010-05-31 Online:2011-01-11 Published:2011-01-11
  • Contact: ZHANG Yunong


张雨浓1,郭东生1,谭 宁2   

  1. 1.中山大学 信息科学与技术学院,广州 510275
    2.中山大学 软件学院,广州 510275
  • 通讯作者: 张雨浓

Abstract: For a kind of feed-forward neural net with hidden-neurons’ activation-functions being a sequence of power functions,an optimal network-structure determination algorithm is proposed based on the weights-direct-determination method.Computer simulation and verification results indicate that the algorithm can determine the optimal number of hidden-layer neurons automatically,quickly and effectively,which achieves the best approximation ability of the neural net and thus realizes the network-structure optimization.

Key words: power series, feed-forward neural net, hidden-layer neurons, structure optimization, weights direct determination

摘要: 针对一种以幂函数序列为各隐神经元激励函数的前向神经网络,提出了一种基于权值直接确定方法的网络最优结构确定算法。计算机仿真与验证结果表明,该算法能自动、快速、有效地确定网络的最优隐神经元数,达到网络的最佳逼近能力,从而实现网络结构的最优化。

关键词: 幂级数, 前向神经网络, 隐神经元数, 结构最优化, 权值直接确定法

CLC Number: