A modular analog chip for feed-forward networks with on-chip learning

Author

Oh, Hwa-Joon ; A. Salam, Fathi

Author_Institution

Dept. of Electr. Eng., Michigan State Univ., East Lansing, MI, USA

fYear

1993

fDate

16-18 Aug 1993

Firstpage

766

Abstract

A feedforward artificial neural network (ANN) with learning capability in modular design is presented using CMOS circuits. We employ a modified error backpropagation continuous-time learning rule. A (nonlinear) analog Gilbert multiplier is used as a synapse and a wide-range transconductance amplifier is used as soma. For learning circuits, the, same multiplier is used for updating the weights. Test results demonstrate the successful operation of the chip. Finally, a modular chip design for a large scale implementation of feedforward ANN with learning is described

Keywords

CMOS analogue integrated circuits; analogue processing circuits; backpropagation; feedforward neural nets; large scale integration; neural chips; CMOS circuits; Gilbert multiplier; continuous-time learning rule; error backpropagation; feedforward artificial neural network; feedforward networks; large scale implementation; learning capability; modular analog chip; on-chip learning; soma; synapse; wide-range transconductance amplifier; Artificial neural networks; Circuit testing; Circuits and systems; Electronic mail; Equations; Feedforward systems; Laboratories; Large-scale systems; Network-on-a-chip; Neurons;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1993., Proceedings of the 36th Midwest Symposium on

Conference_Location

Detroit, MI

Print_ISBN

0-7803-1760-2

Type

conf

DOI

10.1109/MWSCAS.1993.342934

Filename

342934