| 多层前馈神经网络隐单元数目上界的证明 |
| |
| 引用本文： | 刘英敏,吴沧浦.多层前馈神经网络隐单元数目上界的证明[J].北京理工大学学报,2000,20(1):73-76. |
| |
| 作者姓名： | 刘英敏 吴沧浦 |
| |
| 作者单位： | 北京理工大学,自动控制系,北京,100081 |
| |
| 摘 要： | 研究前馈神经网络隐单元数目的上蜀和如何利用样本集的特性减少所需的隐单元个数。利用Ｓｉｇｍｏｉｄ函数的两端极限特性，使每个隐单元表示１～２个样本。在样本集具有局部单调性的情况下，可以用有〖（ｐ－１）／２〗个隐单元的３层前馈神经网络以任意小的误差表示ｐ个目标值。对一般的样本集，所需的隐单元数为〖（ｐ－１）／２〗～（ｐ－１）个。
|
| 关 键 词： | 前馈神经网络 隐单元数 上界 泛化 激活函数 |
| 修稿时间： | 1999-02-10 |
A Proof of the Upper Bounds of the Number of Hidden Units in Multistory Feedforward Neural Networks |
| |
| LIU Ying min,WU Cang pu.A Proof of the Upper Bounds of the Number of Hidden Units in Multistory Feedforward Neural Networks[J].Journal of Beijing Institute of Technology(Natural Science Edition),2000,20(1):73-76. |
| |
| Authors: | LIU Ying min WU Cang pu |
| |
| Abstract: | To investigate the upper bound of the number of the hidden units in feedforward nerual networks and how to utilize the property of the patterns set to reduce the needed number. The property of the two unequal limits at infinities of Sigmoid function was used to make one hidden unit to represent one to two patterns. When the patterns set possesses the property of local monotone, the upper bound is BHDG1,K*2W]( p -1)/2BHDG1,WK*2]. In general, the upper bound ranges from BHDG1,K*2W]( p -1)/2BHDG1,WK*2] to ( p -1). |
| |
| Keywords: | feedforward nerual networks number of hidden units upper bound generalization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《北京理工大学学报》浏览原始摘要信息 |
| 点击此处可从《北京理工大学学报》下载免费的PDF全文 |
|
