| 一种新颖的神经网络稳健估计方法 |
| |
| 引用本文： | 刘光远,邱玉辉,廖晓峰,覃朝玲.一种新颖的神经网络稳健估计方法[J].计算机研究与发展,1999,36(5):567-571. |
| |
| 作者姓名： | 刘光远 邱玉辉 廖晓峰 覃朝玲 |
| |
| 作者单位： | [1]西南师范大学计算机科学系 [2]成都电子科技大学光电系 |
| |
| 基金项目： | 重庆市科学技术委员会研究基金 |
| |
| 摘 要： | 当神经网络应用于实际工程问题时，网络的训练数据集成或多或少都有噪声或异常值掺入其中，为了使网络具有更好的稳健性，文中根据稳健统计学原理，针对前馈神经网络（ＦＮＮ）提出了一种稳健估计（ＲＥ）函数作为新的网络目标函数。
|
| 关 键 词： | 稳健估计 前馈神经网络 函数逼近 神经网络 |
A NOVEL ROBUST ESTIMATION METHOD
FOR NEURAL NETWORK |
| |
| LIU Guang\|Yuan,QIU Yu\|Hui,LIAO Xiao\|Feng,and QIN Chao\|Ling.A NOVEL ROBUST ESTIMATION METHOD
FOR NEURAL NETWORK[J].Journal of Computer Research and Development,1999,36(5):567-571. |
| |
| Authors: | LIU Guang\|Yuan QIU Yu\|Hui LIAO Xiao\|Feng and QIN Chao\|Ling |
| |
| Affiliation: | LIU Guang\|Yuan,QIU Yu\|Hui,LIAO Xiao\|Feng*,and QIN Chao\|Ling |
| |
| Abstract: | When neural network (NN) is applied to practical engineering, there are more or less some noises or outliers in training data sets. To make NN hold more robust, a novel robust estimation function is proposed to serve as new target function of feedforward neural network (FNN) according to the theorem of robust statistics in the paper. The results of the theoretical analysis and simulation for function approximation problem show superiority of the method, which not only holds basically the merit of the least square estimate (LSE) method but also has robustness against outliers. |
| |
| Keywords: | robust estimation feedforward NN function approximation outliers |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
