Analog implementation of neural networks

Author

Zurada, Jacek M.

Author_Institution

Dept. of Electr. Eng., Louisville Univ., KY, USA

Volume

8

Issue

5

fYear

1992

Firstpage

36

Lastpage

41

Abstract

Analog silicon-based neural hardware, which represents a large category among special-purpose analog and digital neurocomputers, and neural processing algorithms are reviewed. Artificial neural networks usually contain a large number of synaptic connections and many fewer processing neurons. The central problem in implementing artificial neural networks-making weights that are continuously adjustable, preferably in response to an analog control signal-is discussed. A simple integrated-circuit analog multiplier built from all-MOS components for use in electrically tunable synapses is described.<>

Keywords

MOS integrated circuits; analogue computer circuits; multiplying circuits; neural nets; MOS components; adjustable weights; analog control signal; analogue implementation; electrically tunable synapses; integrated-circuit analog multiplier; neural networks; neural processing algorithms; neurocomputers; synaptic connections; Arithmetic; Artificial neural networks; Computer networks; Computer vision; Concurrent computing; Embedded computing; Neural network hardware; Neural networks; Neurons; Very large scale integration;

fLanguage

English

Journal_Title

Circuits and Devices Magazine, IEEE

Publisher

ieee

ISSN

8755-3996

Type

jour

DOI

10.1109/101.158511

Filename

158511