Share Email Print
cover

Proceedings Paper

Optical neural networks for classification into arbitrary classes
Author(s): H. H. Arsenault; S. Chang; Philippe Gagne; Oscar Gualdron Gonzalez
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Some important concepts of neural networks are similarity, generalization, invariance and training. Some neural networks are supposed to be able to classify objects according to hidden similarities. All of those concepts are put into question by the consideration first put forward by Watenabe that from a purely logical point of view, similarity is a purely arbitrary concept. It can be shown that this implies that the notion of invariance is also arbitrary , that so-called hidden similarities and generalization cannot exist without some external criteria. Such criteria are either implicit in the training algorithms or must be imposed explicitly. This imposes severe limitations on what neural networks can accomplish. However there are some positive implications; neural networks can be designed to classify objects into arbitrary classes. Applications to optical neural networks and examples will be presented.

Paper Details

Date Published: 9 August 1996
PDF: 15 pages
Proc. SPIE 10287, Optoelectronic Devices and Systems for Processing: A Critical Review, 1028706 (9 August 1996); doi: 10.1117/12.259697
Show Author Affiliations
H. H. Arsenault, Univ. Laval (Canada)
S. Chang, Univ. Laval (Canada)
Philippe Gagne, Univ. Laval (Canada)
Oscar Gualdron Gonzalez, Univ. de Santander (Colombia)


Published in SPIE Proceedings Vol. 10287:
Optoelectronic Devices and Systems for Processing: A Critical Review
Bahram Javidi; Kristina M. Johnson, Editor(s)

© SPIE. Terms of Use
Back to Top