Capacity and VC-dimension of multilayer network with higher order input transformation

Author

Kowalczyk, Adam ; Szymanski, Jacek

Author_Institution

Telecom Australia Res. Labs., Clayton, Vic., Australia

fYear

1994

fDate

29 Nov-2 Dec 1994

Firstpage

1

Lastpage

5

Abstract

The paper investigates a multilayer perceptron built of linear threshold units superposed on a higher order transformation of input, a type of structure considered for optical implementation. Theoretical estimates of its separating capacity, VC-dimension and probability of implementation of a random dichotomy are given. It is shown that the capacity is limited by the number of neurons in the first hidden layer and that for a sufficiently large network it can be lower than the VC-dimension

Keywords

multilayer perceptrons; optical neural nets; probability; VC-dimension; hidden layer; higher order input transformation; higher order transformation; linear threshold units; multilayer network capacity; multilayer perceptron; optical implementation; random dichotomy; separating capacity; Holographic optical components; Holography; Neural networks; Neurons; Nonhomogeneous media; Nonlinear optics; Optical computing; Optical network units; Polynomials; Zinc;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Information Systems,1994. Proceedings of the 1994 Second Australian and New Zealand Conference on

Conference_Location

Brisbane, Qld.

Print_ISBN

0-7803-2404-8

Type

conf

DOI

10.1109/ANZIIS.1994.396961

Filename

396961