Exact Hessian calculation in feedforward FIR neural networks

Author

Cholewo, Tomasz J. ; Zurada, Jacek M.

Author_Institution

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

Volume

2

fYear

1998

fDate

4-9 May 1998

Firstpage

1074

Abstract

FIR neural networks are feedforward neural networks with regular scalar synapses replaced by linear finite impulse response filters. This paper introduces the second order temporal backpropagation algorithm which enables the exact calculation of the second order error derivatives for a FIR neural network. This method is based on the error gradient calculation method first proposed by Wan (1993) and referred to as temporal backpropagation. A reduced FIR synapse model obtained by ignoring unnecessary time lags is proposed to reduce the number of network parameters

Keywords

FIR filters; Hessian matrices; backpropagation; feedforward neural nets; filtering theory; FIR filters; error gradient calculation method; exact Hessian calculation; feedforward FIR neural networks; linear finite impulse response filters; reduced FIR synapse model; regular scalar synapses; second order error derivatives; second order temporal backpropagation algorithm; Backpropagation algorithms; Difference equations; Ear; Feedforward neural networks; Finite impulse response filter; Intelligent networks; Multi-layer neural network; Neural networks; Neurons; Nonlinear filters;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.685921

Filename

685921