A Unified Modeling of Neural Networks Architectures

  • S. Y. Rung
  • J. N. Hwang
Conference paper
Part of the NATO ASI Series book series (volume 66)

Abstract

Although neural networks can ultimately be used for many applications, their suitability for a specific application depends on the acquisition/representation, performance vs. training data, response time, classification accuracy, fault tolerance, generality, adaptability, computational efficiency, size and power requirement. In order to deal with such a multiple-spectrum consideration, there is a need of unified examination of the theoretical foundations of neural network modeling. This can lead to more effective simulation and implementation tools. For this purpose, the paper proposes a unified modeling formulation for a wide variety of artificial neural networks (ANNs): single layer feedback networks, competitive learning networks, multilayer feed-forward networks, as well as some probabilistic models. The existing connectionist neural networks are parameterized by nonlinear activation function, weight measure function, weight updating formula, back-propagation, and iteration index (for retrieving phase) and recursion index (for learning phase). Based on the formulation, new models may be derived and one such example is discussed in the paper. The formulation also leads to a basic structure for a universal simulation tool and neurocomputer architecture.

Keywords

Manifold Radar Mellon 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    D. H. Ackley, G. E. Hinton, and T. J. Sejnowski. A learning algorithm for Boltzmann machines. Cognitive Science, 9:147–169, 1985.CrossRefGoogle Scholar
  2. [2]
    H.M. Ahmed. Alternative arithmetic unit architectures for VLSI digital signal processors. In VLSI and Modern Signal Processing, chapter 16, pages 277–306. Prentice Hall, Inc., Englewood Cliffs, NJ, 1985.Google Scholar
  3. [3]
    J. A. Anderson. Neurocomputing — Paper Collections MIT Press, 1988.Google Scholar
  4. [4]
    L. E. Baum and G. R. Sell. Growth transformations for functions on manifolds. Pacific Journal of Mathematics, 27(2):211–227, 1968.MathSciNetMATHGoogle Scholar
  5. [5]
    L.E. Baum and J. A. Eagon. An inequality with applications to statistical estimation for probabilistic function of Markov processes and a model for ecology. Amer. Math. Soc. Bull, 73:360–363, May 1967.MathSciNetMATHCrossRefGoogle Scholar
  6. [6]
    L.E. Baum, T. Petrie, G. Soules, and N. Weiss. A maximization technique occurring in the statistical analysis of+ probabilistic functions of Markov chains. Ann. Math. Statistic, 41:164–171, 1970.MathSciNetMATHCrossRefGoogle Scholar
  7. [7]
    G. A. Carpenter and S. Grossberg. ART2: Self-organization of stable category recognition codes for analog input patterns. In Proc. IEEE ICNN’87, San Diego, pages II 727- II 736, 1987.Google Scholar
  8. [8]
    Gail A. Carpenter and Stephen Grossberg. ART2: Self-organized of stable category recognition codes for analog input patterns. Applied Optics, 26(23):4919–4930, December 1987.CrossRefGoogle Scholar
  9. [9]
    K. Fukushima. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics, 36:193–202, April 1980.Google Scholar
  10. [10]
    S. Grossberg. Adaptive pattern classification and universal recoding: Part 1. parallel development and coding of neural feature detectors. Biological Cybernetics, 23:121–134, 1976.MathSciNetMATHCrossRefGoogle Scholar
  11. [11]
    D. O. Hebb. The Organization of Behavior Wiley Inc., New York, 1949.Google Scholar
  12. [12]
    G. E. Hinton. Connectionist learning procedure. Technical Report CMU-CS-87–115, Carnegie Mellon University, September 1987.Google Scholar
  13. [13]
    J. J. Hopfield. Neural network and physical systems with emergent collective computational abilities. In Proc. Nat’l. Acad. Sci. USA, volume 79, pages 2554–2558, 1982.Google Scholar
  14. [14]
    J. N. Hwang. Algorithms/Applications/Architectures of Artificial Neural Nets PhD thesis, Dept. of Electrical Engineering, University of Southern California, December 1988.Google Scholar
  15. [15]
    J. N. Hwang and S. Y. Kung. A unifying viewpoint between multilayer perceptrons and hidden Markov models. In IEEE Int’l Symposium on Circuits and Systems, ISCAS’89, Portland, pages 770–773, May 1989.Google Scholar
  16. [16]
    B. H. Juang. On the hidden Markov model and dynamic time warping for speech recogintion — a unified view. AT&T Bell Laboratories Technical Journal, 63(7):1213–1243, September 1984.MathSciNetMATHGoogle Scholar
  17. [17]
    T. Kohonen. Self-organized formation of topologically correct feature map. Biological Cybernetics, 43:59–69, 1982.MathSciNetMATHCrossRefGoogle Scholar
  18. [18]
    T. Kohonen. Self-Organization and Associative Memory, Series in Information Science, Vol. 8. Springer-Verlag, New York, 1984.Google Scholar
  19. [19]
    S. Y. Kung. VLSI Array Processors Prentice Hall Inc., N.J., 1988.Google Scholar
  20. [20]
    S. Y. Kung and J. N. Hwang. An algebraic projection analysis for optimal hidden units size and learning rate in back-propagation learning. In IEEE, Int’l Conf on Neural Networks, ICNN’88, San Diego, pages Vol.1: 363–370, July 1988. (Also accepted for publication in Neural Networks.) CrossRefGoogle Scholar
  21. [21]
    S.Y. Kung and J. N. Hwang. A unified systolic architecture for artificial neural networks. Journal of Parallel and Distributed Computing, Special Issue on Neural Networks, 6(2):358–387, April 1989.Google Scholar
  22. [22]
    S. E. Levinson, L. R. Rabiner, and M. M. Sondhi. An introduction to the application of the theory of probabilistic functions of a Markov process to automatic speech recognition. The Bell System Technical Journal, 62:1035–1074, April 1983.MathSciNetMATHGoogle Scholar
  23. [23]
    R. Linsker. Self-organization in a perceptual network. IEEE Computer Magazine, 21:105 – 117, March 1988.Google Scholar
  24. [24]
    J. L. McClelland and D. E. Rumelhart. Distributed memory and the representation of general and specific information. Journal of Experimental Psychology: General, 114:158–188, 1985.CrossRefGoogle Scholar
  25. [25]
    C. Mead and L. Conway. Introduction to VLSI Systems Addison-Wesley, 1980.Google Scholar
  26. [26]
    N. J. Nilsson. Learning Machines McGraw-Hill Book Company, 1965.MATHGoogle Scholar
  27. [26]
    L. R. Rabiner. A tutorial on hidden Markov models ans selected applications in speech recognition. Proceedings IEEE, 77(2):257–286, February 1989.CrossRefGoogle Scholar
  28. [28]
    L. R. Rabiner and B. H. Juang. An introduction to hidden Markov models. IEEE ASSP Magazine, 3(1):4–16, January 1986.CrossRefGoogle Scholar
  29. [29]
    D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representations by error propagation. Parallel Distributed Processing (PDP): Exploration in the Microstructure of Cognition (Vol. 1), chapter 8, pages 318–362. MIT Press, Cambridge, Massachusetts, 1986.Google Scholar
  30. [30]
    D. E. Rumelhart and D. Zipser. Feature discovery by competitive learning. Cognitive Science, 9:75–112, 1985.CrossRefGoogle Scholar
  31. [31]
    P. F. Stebe. Invariant functions of an iterative process for maximization of a polynomial. Pacific Journal of Mathematics, 43(3):765–783, 1972.MathSciNetMATHGoogle Scholar
  32. [32]
    B. Widrow and R. Winter. Neural nets for adaptive filtering and adaptive pattern recognition. IEEE Computer Magazine, 21:25–39, March 1988.Google Scholar
  33. [33]
    G. Widrow and M. E. Hoff. Adaptive switching circuit. IRE Western Electronic Show and Convention: Convention Record, pages 96–104, 1960.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • S. Y. Rung
    • 1
  • J. N. Hwang
    • 2
  1. 1.Department of Electrical EngPrinceton UniversityPrincetonUSA
  2. 2.Department of Electrical EngUniversity of WashingtonSeattleUSA

Personalised recommendations