Authors: Todorović, Branimir
Stanković, Miomir 
Moraga, Claudio
Title: On-line learning in recurrent neural networks using nonlinear Kalman filters
Journal: Proceedings of the 3rd IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2003
First page: 802
Last page: 805
Issue Date: 1-Jan-2003
Rank: M33
ISBN: 0-7803-8292-7
DOI: 10.1109/ISSPIT.2003.1341242
The extended Kalman filter has been successfully applied to the feedforward and the recurrent neural network training. Recently introduced derivative-free filters (unscented Kalman filter and divided difference filter) outperform the extended Kalman filter in nonlinear state estimation. In the parameter estimation of the feedforward neural networks UKF and DDF are comparable or slightly better than EKF, with a significant advantage that they do not demand calculation of the neural network Jacobian. In this paper, we consider the application of EKF, UKF and DDF to the recurrent neural network training. The class of non-linear autoregressive recurrent neural networks with exogenous inputs is chosen as a basic architecture due to its powerful representational capabilities.
Keywords: Electronic mail | Feedforward neural networks | Filters | Intelligent networks | Jacobian matrices | Neural networks | Neurons | Parameter estimation | Recurrent neural networks | State estimation
Publisher: IEEE

Show full item record


checked on Jul 23, 2024

Page view(s)

checked on May 9, 2024

Google ScholarTM




Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.