Abstract
This paper covers the emphasis of a lecture given previously [1], concerning the development of neural networks (NN) from the transversal filters of the late 1940’s to the 1980’s rebirth of interest.
Multiple layers of linear cells are equivalent to two layers. Two layers of cells cannot handle exceptions of the type needed for XOR. One of the key differences of modem NN algorithms is the nonlinear (sigmoid) response of the cells. Unlike a threshold, this permits “learning”without human intervention. Multiple layered nonlinear networks can distinguish as many classes as desired.
There are numerous techniques aimed at human brain-like behavior. Whether brain-like or not is of little interest to a problem solver. However, a few of these methods are useful in the practical sense. Emphasis here will be on the extension of related methods rather than completeness.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Penn, T. C., “SELF ORGANIZATION: Adaptive Filters to Neural Networks”, Lecture at Universität der Bundeswehr, Munich, May 19, 1989.
Linsker, Ralph, “From basic network principles to neural architecture: Emergence of spatial-opponent cells”, Proc. Natl. Acad. Sei. USA, vol 83, pp 7508–7512, Oct 1986.
Linsker, Ralph, “From basic network principles to neural architecture: Emergence of orientation-selective cells”, op. cit. pp 8390–8394, Nov 1986.
Linsker, Ralph, “From basic network principles to neural architecture: Emergence of orientation colums”, op. cit. pp 9779–8783, Nov 1986.
Bellman, Richard, “ADAPTIVE CONTROL PROCESSES:A GUIDED TOUR”, Princeton, N.J., Princeton University Press, 1961.
Widrow, B and Hoff, M.E., “Adaptive Switching Circuits”, 1960 IRE WESCON Conv. Record, Part 4, pp 96–104, Aug 1960.
Rosenblatt, Frank, “PRINCIPLES OF NEURODYNAMICS, PERCEPTRONS AND TH THORY OF BRAIN MECHANISMS”, Washington, D.C., Spartan Books, 6411 Chillum Place NW, 616 pages, 1962.
Rumelhart, D. E., Hinton, G. E. and Williams, R. J., “Learning Internal Representations by Error Propagation”, “PARLEL DISTRIBUTED PROCESSING” vol 1, The MIT Press, Cambridge, Mass, pp 318–462, 1986.
Werbos, P. J., “Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences”, Ph.D. thesis, Harvard University, Cambridge, MA, 1974.
Kohonen, Tuevo, “SELF-ORGANIZATION AND ASSOCIATIVE MEMORY”, Springer-Verlag, Berlin, 255 pages, 1984.
Sejnowski, T. J. and Rosenberg, C. R., “Parallel Networks that Learn to Pronounce English Text”, COMPLEX SYSTEMS, 1, pp 145–168.
Gorman, Paul, private communication but the work has been published.
Lapedes, Alan and Färber, Robert, Preprint LA-UR-87-2662, submitted to the IEEE for publication, Jul 1987. Reproduced here with permission from Alan Lapedes.
Kanerva, Pentti, “SPARSE DISTRIBUTED MEMORY”, MIT Press, Cambridge, MA, 145 pages, 1988.
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1992 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Penn, T.C.C. (1992). SELF ORGANIZATION: Adaptive Filters to Neural Networks. In: Niegel, W., Molzberger, P. (eds) Aspekte der Selbstorganisation. Informatik-Fachberichte, vol 304. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-77485-0_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-77485-0_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-55428-8
Online ISBN: 978-3-642-77485-0
eBook Packages: Springer Book Archive