Recursive training of neural networks for classification

被引：2

作者：

Aladjem, M ^{[1
]}

机构：

[1] Ben Gurion Univ Negev, Dept Elect & Comp Engn, IL-84105 Beer Sheva, Israel

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 2000年 / 11卷 / 02期

关键词：

autoassociative network; linear and nonlinear classification functions; neural networks for classifications; projection pursuit; structure removal;

D O I：

10.1109/72.839018

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A method for recursive training of neural networks for classification is proposed. It searches for the discriminant functions corresponding to several small Local minima of the error function. The novelty of the proposed method lies in the transformation of the data into new training data with a deflated minimum of the error function and iteration to obtain the nest solution. A simulation study and a character recognition application indicate that the proposed method has the potential to escape from local minima and to direct the local optimizer to new solutions.

引用

页码：496 / 503

页数：8

共 10 条

[1]

Aladjem M, 1999, LECT NOTES COMPUT SC, V1642, P223

[2] Linear discriminant analysis for two classes via removal of classification structure [J].