BOUNDS ON THE NUMBER OF HIDDEN UNITS IN BINARY-VALUED 3-LAYER NEURAL NETWORKS

被引：39

作者：

ARAI, M

机构：

来源：

NEURAL NETWORKS | 1993年 / 6卷 / 06期

关键词：

3-LAYER NEURAL NETWORKS; HIDDEN UNIT NUMBER; ARBITRARY SEPARABILITY; 2-CATEGORY CLASSIFICATION; BINARY UNITS; HYPERPLANE; HYPERCUBE; CLASSIFICATION PROBLEMS;

D O I：

10.1016/S0893-6080(05)80130-3

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

For three-layer artificial neural networks (TANs) that take binary values, the number of hidden units is considered regarding two problems: One is to find the necessary and sufficient number to make mapping between the binary output values of TANs and learning patterns (inputs) arbitrary, and the other is to get the sufficient number for two-category classification (TCC) problems. We show that for the former I - 1 hidden units are necessary and sufficient for I learning patterns and that for the latter about I/3 hidden units are sufficient. These results mean that we can reduce the necessary number of hidden units by taking into account the features of learning pattern distributions.

引用

页码：855 / 860

页数：6

共 8 条

[1]

ANDERSON DE, 1987, NEURAL INFORMATION P

[2]

ARAI M, 1989, P INT C NEURAL NETWO, P419

[3]

Baum E. B., 1988, Journal of Complexity, V4, P193, DOI 10.1016/0885-064X(88)90020-9

[4] GEOMETRICAL AND STATISTICAL PROPERTIES OF SYSTEMS OF LINEAR INEQUALITIES WITH APPLICATIONS IN PATTERN RECOGNITION [J].

COVER, TM .

IEEE TRANSACTIONS ON ELECTRONIC COMPUTERS, 1965, EC14 (03) :326-&

[5] ANALYSIS OF HIDDEN UNITS IN A LAYERED NETWORK TRAINED TO CLASSIFY SONAR TARGETS [J].

GORMAN, RP ;

SEJNOWSKI, TJ .

NEURAL NETWORKS, 1988, 1 (01) :75-89

[6] BOUNDS ON THE NUMBER OF HIDDEN NEURONS IN MULTILAYER PERCEPTRONS [J].

HUANG, SC ;

HUANG, YF .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1991, 2 (01) :47-55

[7]

Nilsson N., 1965, LEARNING MACHINES

[8]

Rockafellar R.T., 1970, CONVEX ANAL

← 1 →