More Information
Monthly
288 pp. per issue
6 x 9, illustrated
ISSN
0899-7667
E-ISSN
1530-888X
2014 Impact factor:
2.21
List of All Issues » Issue TOC

Neural Computation

July 1, 1996, Vol. 8, No. 5, Pages 939-949
(doi: 10.1162/neco.1996.8.5.939)
© 1996 Massachusetts Institute of Technology
Effects of Nonlinear Synapses on the Performance of Multilayer Neural Networks
Article PDF (493.85 KB)
G. Dündar
Electrical, Computer, and Systems Engineering Department, Center for Integrated Electronics and Electronics Manufacturing, Rensselaer Polytechnic Institute, Troy, NY 12180 USA
F-C. Hsu
Electrical, Computer, and Systems Engineering Department, Center for Integrated Electronics and Electronics Manufacturing, Rensselaer Polytechnic Institute, Troy, NY 12180 USA
K. Rose
Electrical, Computer, and Systems Engineering Department, Center for Integrated Electronics and Electronics Manufacturing, Rensselaer Polytechnic Institute, Troy, NY 12180 USA
Abstract

The problems arising from the use of nonlinear multipliers in multilayer neural network synapse structures are discussed. The errors arising from the neglect of nonlinearities are shown and the effect of training in eliminating these errors is discussed. A method for predicting the final errors resulting from nonlinearities is described. Our approximate results are compared with the results from circuit simulations of an actual multiplier circuit.