THE OPTIMIZED INTERNAL REPRESENTATION OF MULTILAYER CLASSIFIER NETWORKS PERFORMS NONLINEAR DISCRIMINANT-ANALYSIS

被引:76
作者
WEBB, AR
LOWE, D
机构
[1] Royal Signals and Radar Establishment, Great Malvern
关键词
Adaptive layered networks; Learning; Nonlinear discriminant analysis; Nonlinear optimisation; Pattern classification;
D O I
10.1016/0893-6080(90)90019-H
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper illustrates why a nonlinear adaptive feed-forward layered network with linear output units can perform well as a pattern classification device. The central result is that minimising the error at the output of the network is equivalent to maximising a particular norm, the network discriminant function, at the output of the hidden units. The first part of the network is explicitly performing a nonlinear transformation of the data into a space in which the classes may be more easily separated. The specific nature of this transformation is constrained to maximise the network discriminant function. If the targets are appropriately chosen, this discriminant function relates the pseudo-inverse of the total covariance matrix and the weighted between-class covariance matrix of the hidden unit patterns. Numerical simulations are presented to illustrate the results. © 1990.
引用
收藏
页码:367 / 375
页数:9
相关论文
共 22 条
[1]  
[Anonymous], 1974, CLASSIFICATION ESTIM
[2]  
[Anonymous], 1965, J SOC IND APPL MAT B, DOI [10.1137/0702016, DOI 10.1137/0702016]
[3]  
BOULARD H, 1987, AUTO ASS MULTILAYER
[4]  
Broomhead D. S., 1988, COMPLEX SYSTEMS, P269
[5]   QUASI-NEWTON METHODS AND THEIR APPLICATION TO FUNCTION MINIMISATION [J].
BROYDEN, CG .
MATHEMATICS OF COMPUTATION, 1967, 21 (99) :368-&
[6]  
Byrne G., 1973, NUMERICAL SOLUTIONS
[7]  
Denker J., 1987, Complex Systems, V1, P877
[8]  
Duda R. O., 1973, PATTERN CLASSIFICATI, V3
[9]  
FUKUNAGA K, 1972, INTRO STATISTICAL PA
[10]  
GALLINARI P, 1988, IEEE ANN INT C NEURA, pI391