| Sigmoid传输函数与三层前馈神经网络的映射能力 |
| |
| 引用本文： | 李勇,郭平. Sigmoid传输函数与三层前馈神经网络的映射能力[J]. 北京师范大学学报(自然科学版), 2001, 37(5): 569-573 |
| |
| 作者姓名： | 李勇 郭平 |
| |
| 作者单位： | 1. 北京师范大学数学系
2. 北京师范大学信息学院 |
| |
| 基金项目： | 国家自然科学基金;10071009; |
| |
| 摘 要： | 对于具有n个隐含单元的三层前馈神经网络，证明了隐含层的Sigmoid传输函数可以把输入层数据的维数提升到n维，这意味着有n个隐神经元的前馈三层神经网络具有把低维空间映射到高维空间的能力。
|
| 关 键 词： | 神经网络 矩阵 Sigmoid传输函数 映射能力 |
| 修稿时间： | 2001-03-01 |
CAPABILITIES OF A THREE-LAYER FEEDFORWARD NEURAL NETWORK WITH NONLINEAR ACTIVATE FUNCTION |
| |
| Li Yong ) Guo Ping ). CAPABILITIES OF A THREE-LAYER FEEDFORWARD NEURAL NETWORK WITH NONLINEAR ACTIVATE FUNCTION[J]. Journal of Beijing Normal University(Natural Science), 2001, 37(5): 569-573 |
| |
| Authors: | Li Yong ) Guo Ping ) |
| |
| Affiliation: | Li Yong 1) Guo Ping 2) |
| |
| Abstract: | For a three-layer feedforward neural network with n hidden units, we prove that the sigmoid functions of a hidden layer of the network can raise the dimension of the input space up to the number of the hidden units. This result means that a three-layer feedforward neural network with n hidden units has the capability of mapping lower dimension space into n-dimension space. |
| |
| Keywords: | neural network Sigmoid function matrix |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
