DocumentCode
285412
Title
Stationary points of single-layer feedback neural networks
Author
Zurada, Jacek M. ; Kang, Min J.
Author_Institution
Dept. of Electr. Eng., Louisville Univ., KY, USA
Volume
1
fYear
1992
fDate
10-13 May 1992
Firstpage
57
Abstract
The properties of stationary points of single-layer fully coupled neural networks are investigated. The propositions are formulated and proved on the basis of a study of the energy function. Networks with both infinite gain (discrete update) and finite gain (continuous update) are discussed. The study provides considerable insight into the time-domain performance of the networks
Keywords
recurrent neural nets; continuous update; discrete update; energy function; finite gain; fully coupled neural networks; infinite gain; single-layer feedback neural networks; stationary points; time-domain performance; Capacitance; Eigenvalues and eigenfunctions; Equations; Hypercubes; Memory; Neural networks; Neurofeedback; Neurons; Time domain analysis; Voltage;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 1992. ISCAS '92. Proceedings., 1992 IEEE International Symposium on
Conference_Location
San Diego, CA
Print_ISBN
0-7803-0593-0
Type
conf
DOI
10.1109/ISCAS.1992.230015
Filename
230015
Link To Document