Accuracy effects in pattern recognition neural nets

Author

Casasent, David

Author_Institution

Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA, USA

Volume

3

fYear

1992

fDate

7-11 Jun 1992

Firstpage

643

Abstract

Various errors, including analog accuracy, nonlinearities, and noise, are present in all neural networks. The author considers their effects in training and testing on two different pattern recognition neural nets. He shows that the neural nets considered allow some such effects to be included inherently in the neural net synthesis algorithm and that the effect of the other error sources can be trained out by proper selection of neural net design parameters. Multiclass distortion-invariant pattern recognition neural nets are considered. The results are applicable to analog VLSI and optical neural nets

Keywords

learning (artificial intelligence); neural nets; pattern recognition; analog VLSI; analog accuracy; multi-class distortion-invariant neural nets; noise; nonlinearities; optical neural nets; pattern recognition neural nets; testing; training; Acoustic noise; Algorithm design and analysis; Analog computers; Neural networks; Neurons; Optical computing; Optical distortion; Optical noise; Pattern recognition; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1992. IJCNN., International Joint Conference on

Conference_Location

Baltimore, MD

Print_ISBN

0-7803-0559-0

Type

conf

DOI

10.1109/IJCNN.1992.227101

Filename

227101