Abstract
A theoretical framework for investigation of approximation capabilities of feed-forward networks is presented in the context of nonlinear approximation theory. Some recent results on universal approximation property and estimates of network complexity, measured by the number of hidden units, are described.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Barron, A. R. (1992). Neural net approximation. InProceedings of the 7th Yale Workshop on Adaptive and Learning Systems (pp. 69–72).
Barron, A.R. (1993). Universal approximation bounds for superposition of a sigmoidal function.IEEE Transactions on Information Theory, 39, 930–945.
Carroll, S.M. & Dickinson, B. W. (1989). Construction of neural nets using the Radon transform. InProceedings of IJCNN’89 (pp.I. 607–611).New York: IEEE Press.
Courant, R. & Hilbert, D. (1962).Methods of Mathematical Physics, vol.2. New York: Wiley.
Cybenko, G. (1989). Approximation by superpositions of a sigmoidal function.Mathematics of Control and Signals Systems, 2, 303–314.
Darken, C., Donahue, M., Gurvits, L. & Sontag, E. (1993). Rate of approximation results motivated by robust neural network learning. InProceedings of the 6th Annual ACM Conference on Computational Learning Theory (pp.303–309). New York: ACM.
Girosi, F. & Poggio, T. (1990). Networks and the best approximation property. InBiological Cybernetics, 63, 169–176.
Girosi, F., & Anzellotti, G. (1993). Rates of convergence for radial basis function and neural networks. InArtificial Neural Networks for Speech and Vision(pp.97–113). London: Chapman & Hall.
Girosi, F. (1995). Approximation error bounds that use VC-bounds. InProceedings of ICANN’95(pp. 295–302). Paris: EC2 & Cie.
Gurvits, L. & Koiran, P. (1997). Approximation and learning of convex superpositions.Journal of Computer and System Sciences, 55, 161–170.
Ito, Y. (1992). Finite mapping by neural networks and truth functions.Mathematical Scientist, 17, 69–77.
Jones, L. K. (1992). A simple lemma on greedy approximation in Hilbert space and convergence rates for projection pursuit regression and neural network training.Annals of Statistics, 20, 608–613.
Kainen, P. C., Kůrkova, V. & Vogt, A. (1999). Approximation by neural networks is not continuous. Neurocomputing, 29, 47–56.
Kainen, P. C., Kůrková, V. & Vogt, A. (2000). Best approximation by Heaviside perceptron networks.Neural Networks, 13, 645–647.
Kainen, P. C; Kůrková, V. & Vogt, A. (2001). Continuity of approximation by neural networks inL p -spaces.Annals of Operational Research, 101, 143–147.
Kůrková, V. (1992). Kolmogorov’s theorem and multilayer neural networks.Neural Networks, 5, 501–506.
Kůrková, V. (1997). Dimension-independent rates of approximation by neural networks. InComputer-Intensive Methods in Control and Signal Processing: Curse of Dimensionality (Eds. Warwick, K., Kárný, M.) (pp. 261–270). Boston: Birkhauser.
Kůrková, V. (1998). Incremental approximation by neural networks. InComplexity: Neural Network Approach. (Eds. Warwick, K., Kárný, M., Kůrková, V.) (pp. 177–188). London: Springer.
Kůrková, V., Kainen, P. C. & Kreinovich, V. (1997). Estimates of the numberof hidden unitsand variation with respect to half-spaces. Neural Networks, 10,1061-1068.
Kůrková, V. & Sanquineti, M. (2001). Bounds on rates of variable-basis and neural network approximation.IEEE Trans. on Information Theory, 47, 2659–2665.
Kůrková, V., Savický, P. & Hlaváčková, K. (1998). Representations and rates of approximation of real-valued Boolean functions by neural networks.Neural Networks, 11,651–659.
Leshno, M., Lin, V. Y, Pinkus, A. & Schocken, S. (1993). Multilayer feedforward networks with a non-polynomial activation can approximate any function.Neural Networks, 6, 861–867.
Makovoz, Y. (1998). Uniform approximation by neural networks.Journal of Approximation Theory, 95, 215–228.
Mhaskar, H. N. (1995). Versatile Gaussian networks. InProceedings of IEEE Workshop of Nonlinear Image Processing (pp. 70–73).
Micchelli, C. A. (1986). Interpolation of scattered data: distance matrices and conditionally positive definite functions.Constructive Approximation, 2, 11–22.
Park, J., & Sandberg, I. W. (1993). Approximation and radial-basis-function networks.Neural Computation, 5, 305–316.
Pinkus, A. (1986). n -Width in Approximation Theory. Berlin: Springer.
Pisier, G. (1981). Remarques sur un resultat non publié de B. Maurey. InSeminaire d’Analyse Fonctionelle I., n.12.
Sejnowski, T. J. & Rosenberg, C. (1987). Parallel networks that learn to pronounce English text.Complex Systems 1, 145–168.
Stinchcombe, M. & White, H. (1990). Approximating and learning unknown mappings using multilayer networks with bounded weights. InProceedings of IJCNN’90(pp. III. 7–16).New York: IEEE Press.
Zemanian, A. H. (1987).Distribution Theory and Transform Analysis. New York: Dover.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag London
About this chapter
Cite this chapter
Kůrková, V. (2002). Universality and Complexity of Approximation of Multivariable Functions by Feedforward Networks. In: Roy, R., Köppen, M., Ovaska, S., Furuhashi, T., Hoffmann, F. (eds) Soft Computing and Industry. Springer, London. https://doi.org/10.1007/978-1-4471-0123-9_2
Download citation
DOI: https://doi.org/10.1007/978-1-4471-0123-9_2
Publisher Name: Springer, London
Print ISBN: 978-1-4471-1101-6
Online ISBN: 978-1-4471-0123-9
eBook Packages: Springer Book Archive