Advertisement

Methods of Designing and System Analysis of Fast Neural Networks and Linear Tunable Transformations

  • A. Yu. Dorogov
Conference paper

Abstract

This paper discusses the paradigm of fast neural networks (FNN). System invariants of fast transformations are represented. Formal linguistic methods of structure and topologies designing of FNN and linear tunable transformations are developed. The methods of tuning FNN for realization of spectral transformations, regular fractal, optimum filters are considered. The questions of using FNN in quantum calculations are investigated. The method of separating capacity estimation for weakly-connected feed-forward neural networks is offered. The dependence of amount of recognized patterns on neural network freedom degrees is obtained. The experimental results are represented.

Keywords

Neural Network Fast Fourier Transformation Fractal Filter Output Field Fast Fourier Transformation Algorithm 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Dorogov, A.Yu., Structure Synthesis of Fast Neural Networks, in Neurocomputers Design and Application, New York, 2000, vol. 1, no. 1, pp. 1–18.Google Scholar
  2. [2]
    Good I.J. The Interaction Algorithm and Practical Fourier Analysis // Journal of Royal Statistical Society. Ser.B.- 1958.- Vol.20.- No.2.- P.361-372.Google Scholar
  3. [3]
    Andrews, H.C. and Caspari, K.L., A General Technique for Spectral Analysis, IEEE Trans. Comput., 1970, vol. C-19, no. 1, pp. 16–25.CrossRefGoogle Scholar
  4. [4]
    Labunets, V.G., A Unified Approach to Fast Transformation Algorithms, in Primenenie ortogonal’nykh metodov pri obrabotke signalov i analize sistem(Application of Orthogonal Methods to Signal Processing and Systems Analysis), Sverdlovsk: UPI, 1980, pp. 4–14. (in Russian).Google Scholar
  5. [5]
    Solodovnikov, A.I., Kanatov, I.I., and Spivakovskiy, A.M., Orthogonal Bases Synthesis Based on the Generalized Spectral Kernel, in Voprosy teorii sistem avtomaticheskogo upravleniya(Theoretical Issues of Automatic Control Systems), Leningrad: LGU, 1976, vol. 2, pp. 99-112. (in Russian).Google Scholar
  6. [6]
    Cooley J.W., and J.W. Tukey, An Algorithm for the Machine Computation of Complex Fourier Series // Math.Comp. – 1965. – N.19. – P.297-301.Google Scholar
  7. [7]
    Walsh J.I., A closed set of normal orthogonal functions,- “Amer. J. Math.” 1923, v. 45, p. 5-24.Google Scholar
  8. [8]
    Trakhtman, A.M. and Trakhtman, V.A., Osnovy teorii diskretnykh signalov na konechnykh intervalakh(Theory of Discrete Signals on Finite Intervals), Moscow: Sovetskoe Radio, 1975. (in Russian).Google Scholar
  9. [9]
    Yaroslavskii, L.P., Vvedenie v tsifrovuyu obrabotku izobrazhenii(Introduction to Image Processing), Moscow: Sovetskoe Radio, 1979. (in Russian).Google Scholar
  10. [10]
    Dorogov A.Yu. Implementation of Spectral Transformation in the Class of Fast Neural Networks Programming and Computer Software. Kluwer Academic / Plenum Publishers.- Vol. 29, No 4, 2003.- 13-26p.Google Scholar
  11. [11]
    Stark J. Iterated function systems as neural networks // Neural Networks.- 1991.-V.4.- pp. 679-690.Google Scholar
  12. [12]
    Karhunen N. 1947, English translation by Selin I., On Linear Methods in Probability Theory, The Rand Corporation, Doc, T-13, August 1960.Google Scholar
  13. [13]
    Loeve M. Fonctions Aleatories de Seconde Ordre, Hermann, Paris, 1948.Google Scholar
  14. [14]
    Feynman R. Quantum mechanical computers. Optics News 11, 1985. Also in Foundations of Physics, 16(6) : 507–531, 1986.Google Scholar
  15. [15]
    Shor P. W. Algorithms for quantum computation: Discrete log and factoring. In Proceedings of the 35th Annual Symposium on Foundations of Computer Science (Nov. 1994). pp. 124–134, 1994. Institute of Electrical and Electronic Engineers Computer Society Press, ftp://netlib.att.com/netlib/att/math/shor/quantum.algorithms.ps.Z.Google Scholar
  16. [16]
    Dorogov A.Yu. Structural Synthesis of Modular Weakly Connected Neural Networks. I. Methodology of Structural Synthesis of Modular Neural Networks // Cybernetics and Systems Analysis 37 (2): March - April, 2001.- pp.175-181.Google Scholar
  17. [17]
    Dorogov A.Yu. Structural Synthesis of Modular Weakly Connected Neural Networks. II. Nuclear Neural Networks // Cybernetics and Systems Analysis 37 (4): July - August, 2001.- pp.470-477.Google Scholar
  18. [18]
    Cover T.M. Geometrical and Statistical Properties of System of Liner Inequalities with Applications in Pattern Recognition // IEEE Trans. Electronic Computers, 1965, EC-14, 3, pp. 326-334.Google Scholar
  19. [19]
    Tou .J.T, Gonzaleez R.C. Pattern Recognition Principles.- Addison-Wesley Publishing Company Inc., London, Amsterdam.- 1974.- 411p.Google Scholar
  20. [20]
    Vapnik V.N. The Nature of Statistic Learning Theory.- Springer-Verlag, New York Inc. 1995.- 188p.Google Scholar
  21. [21]
    Borisovich U.G., Bliznyakov U.G., Izrailevich Y.A., Fomenko T.N. Introduction in Topology.- Moscow, Nauka Fizmatlit 1995.-416p. (in Russian).Google Scholar
  22. [22]
    Dorogov A. Yu, Shestopalov M.Yu Separating Capacity of Weakly-Connected Module Neural Networks, Optical Memory & Neural Networks (Information Optics), Vol. 13, No. 3, 2004, p.145-152.Google Scholar
  23. [23]
    Dorogov A.Yu., Alekseev A.A. Plasticity of Multilayer Weakly-Connected Neural Networks // Neurocomputers Design and Application (Moscow).- No 11, 2001.- pp.22-40. (in Russian).Google Scholar
  24. [24]
    Dorogov A.Yu. Fast Neural Networks.- Saint Petersburg: Publishing of Saint Petersburg State University.- 2002.- 80p. (in Russian).Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2007

Authors and Affiliations

  • A. Yu. Dorogov
    • 1
  1. 1.Saint Petersburg State Electrotechnical University ‘‘LETI’’Saint Petersburg,197376Russia

Personalised recommendations