Extensions and enhancements of decoupled extended Kalman filter training

We describe here three useful and practical extensions and enhancements of the decoupled extended Kalman filter (DEKF) neural network weight update procedure, which has served as the backbone for much of our applications-oriented research for the last six years. First, we provide a mechanism that co...

Full description

Saved in:
Bibliographic Details
Main Authors: Puskorius, G.V., Feldkamp, L.A.
Format: Conference Proceeding
Language:English
Subjects:
Backpropagation
Cost function
Covariance matrix
Entropy
Equations
Laboratories
Neural networks
Pattern classification
Recurrent neural networks
Spine
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We describe here three useful and practical extensions and enhancements of the decoupled extended Kalman filter (DEKF) neural network weight update procedure, which has served as the backbone for much of our applications-oriented research for the last six years. First, we provide a mechanism that constrains weight values to a pre-specified range during training to allow for fixed-point deployment of trained networks. Second, we examine modifications of DEKF training for alternative cost functions; as an example, we show how to use DEKF training to minimize a measure of relative entropy, rather than mean squared error, for pattern classification problems. Third, we describe an approximation of DEKF training that allows a multiple-output training problem to be treated with single-output training complexity.
DOI:10.1109/ICNN.1997.614185