Abstract
This paper analyses the statistical convergence properties of the modified NLMS rules which were formulated in an attempt to produce more robust and faster converging training algorithms. However, the statistical analysis described in this paper leads us to the conjecture that the standard NLMS rule is the only unconditionally stable modified NLMS training algorithm, and that the optimal value of the learning rate and region of convergence for the modified NLMS rules is generally less than for the standard NLMS rule.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
E.D. Aved’yan Bestimmung der Parameter linearer Modelle stationärer und instationärer Strecken, Messen, Steuern, Regeln, Vol.14:9 (1971) 348–350.
E.D. Aved’yan, Learning Systems, Springer Verlag, London, UK, (1995).
M. Brown and C. Harris, Neurofuzzy Adaptive Modelling and Control, Prentice Hall, Hemel-Hempstead, UK, (1994).
M. Brown and P.E. An and C.J. Harris, A Stability Analysis of the Modified NLMS Rules, submitted to IEEE Trans. Signal Processing, (1995).
S.C. Douglas, A Family of Normalised LMS Algorithms, IEEE Signal Processing Letters, Vol.1:3 (1994) pp49–51.
S. Kaczmarz, Angenäherte Auflösung von Systemen Linearer Gleichungen, Bull. Int. Acad. Pol. Sci. Lett and Cl. Sci. Math. Nat. Ser. A., (1937) pp355–357.
J.I. Nagumo and A. Noda, A Learning Method for System Identification, IEEE Trans. Automatic Control, Vol.12:3 (1967) pp282–287.
P.C. Parks and J. Militzer, Convergence Properties of Associative Memory Storage for Learning Control Systems, Automation and Remote Control, Vol. 50:2 (1989) 254–286.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media New York
About this chapter
Cite this chapter
Aved’yan, E.D., Brown, M., Harris, C.J. (1997). A Statistical Analysis of the Modified NLMS Rules. In: Ellacott, S.W., Mason, J.C., Anderson, I.J. (eds) Mathematics of Neural Networks. Operations Research/Computer Science Interfaces Series, vol 8. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-6099-9_10
Download citation
DOI: https://doi.org/10.1007/978-1-4615-6099-9_10
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-7794-8
Online ISBN: 978-1-4615-6099-9
eBook Packages: Springer Book Archive