Tracking of non-stationary time-series using resource allocating RBF networks

McLachlan, Alan and Lowe, David Tracking of non-stationary time-series using resource allocating RBF networks. IN: EMCSR 1996 13th European meeting on cybernetics and systems research: April 9-12 1996 at the University of Vienna, Austria. Trappl, R. (ed.) Vienna: Austrian Society for Cybernetic Studies.


In this paper, we discuss some practical implications for implementing adaptable network algorithms applied to non-stationary time series problems. Using electricity load data and training with the extended Kalman filter, we demonstrate that the dynamic model-order increment procedure of the resource allocating RBF network (RAN) is highly sensitive to the parameters of the novelty criterion. We investigate the use of system noise and forgetting factors for increasing the plasticity of the Kalman filter training algorithm, and discuss the consequences for on-line model order selection. We also find that a recently-proposed alternative novelty criterion, found to be more robust in stationary environments, does not fare so well in the non-stationary case due to the need for filter adaptability during training.

Divisions: Engineering & Applied Sciences > Mathematics
Engineering & Applied Sciences > Non-linearity and complexity research group
Event Title: Cybernetics and Systems '96
Event Type: Other
Event Dates: 1996-04-10 - 1996-04-10
Uncontrolled Keywords: adaptable network algorithms,non-stationary time,electricity load data,Kalman filter,dynamic model-order increment procedure,resource allocating RBF network


Full text not available from this repository.

Export / Share Citation


Additional statistics for this record