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

Alan McLachlan, David Lowe

    Research output: Chapter in Book/Published conference outputChapter


    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.
    Original languageEnglish
    Title of host publicationEMCSR 1996 13th European meeting on cybernetics and systems research: April 9-12 1996 at the University of Vienna, Austria
    EditorsR. Trappl
    Place of PublicationVienna
    PublisherAustrian Society for Cybernetic Studies
    Number of pages6
    Publication statusPublished - 10 Apr 1996
    EventCybernetics and Systems '96 -
    Duration: 10 Apr 199610 Apr 1996


    OtherCybernetics and Systems '96


    • adaptable network algorithms
    • non-stationary time
    • electricity load data
    • Kalman filter
    • dynamic model-order increment procedure
    • resource allocating RBF network


    Dive into the research topics of 'Tracking of non-stationary time-series using resource allocating RBF networks'. Together they form a unique fingerprint.

    Cite this