Convergence Study in Extended Kalman Filter-Based Training of Recurrent Neural Networks

Authors:
Xiaoyu Wang; Yong Huang
Affiliations:
Emerging Pathogens Inst., Univ. of Florida, Gainesville, FL, USA;-
Venue:
IEEE Transactions on Neural Networks
Year:
2011

Citing 0
Cited 5

Discrete-time inverse optimal neural control for synchronous generators

Engineering Applications of Artificial Intelligence
Extended Kalman filter-based Elman networks for industrial time series prediction with GPU acceleration

Neurocomputing
Gauss-Newton filtering incorporating Levenberg-Marquardt methods for tracking

Digital Signal Processing
Evolving intelligent algorithms for the modelling of brain and eye signals

Applied Soft Computing
Evolving intelligent system for the modelling of nonlinear systems with dead-zone input

Applied Soft Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recurrent neural network (RNN) has emerged as a promising tool in modeling nonlinear dynamical systems, but the training convergence is still of concern. This paper aims to develop an effective extended Kalman filter-based RNN training approach with a controllable training convergence. The training convergence problem during extended Kalman filter-based RNN training has been proposed and studied by adapting two artificial training noise parameters: the covariance of measurement noise (R) and the covariance of process noise (Q) of Kalman filter. The R and Q adaption laws have been developed using the Lyapunov method and the maximum likelihood method, respectively. The effectiveness of the proposed adaption laws has been tested using a nonlinear dynamical benchmark system and further applied in cutting tool wear modeling. The results show that the R adaption law can effectively avoid the divergence problem and ensure the training convergence, whereas the Q adaption law helps improve the training convergence speed.