Bounds on the number of hidden neurons in multilayer perceptrons

Author

Huang, Shih-Chi ; Huang, Yih-Fang

Author_Institution

Dept. of Electr. Eng., Notre Dame Univ., IN, USA

Volume

2

Issue

1

fYear

1991

fDate

1/1/1991 12:00:00 AM

Firstpage

47

Lastpage

55

Abstract

Fundamental issues concerning the capability of multilayer perceptrons with one hidden layer are investigated. The studies are focused on realizations of functions which map from a finite subset of Eⁿ into E^d. Real-valued and binary-valued functions are considered. In particular, a least upper bound is derived for the number of hidden neurons needed to realize an arbitrary function which maps from a finite subset of Eⁿ into E^d. A nontrivial lower bound is also obtained for realizations of injective functions. This result can be applied in studies of pattern recognition and database retrieval. An upper bound is given for realizing binary-valued functions that are related to pattern-classification problems

Keywords

neural nets; binary-valued functions; database retrieval; hidden neurons; injective functions; least upper bound; lower bound; multilayer perceptrons; pattern recognition; pattern-classification problems; real-valued functions; Backpropagation algorithms; Databases; Information retrieval; Multi-layer neural network; Multilayer perceptrons; Neural networks; Neurons; Pattern classification; Pattern recognition; Upper bound;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/72.80290

Filename

80290