Advertisement

On the use of formal specifications in the design and simulation of artificial neural networks

  • Patrícia Duarte de Lima Machado
  • Silvio Lemos Meira
Applications I
Part of the Lecture Notes in Computer Science book series (LNCS, volume 967)

Abstract

This paper presents an experience on the use of objectoriented formal specifications in the process of software development for artificial neural networks. A formal specification of artificial neural networks using the MooZ language is presented. This specification of class hierarchies shows the gradual inclusion of neural network concepts such that new models or paradigms are easily incorporated by reusing previous definitions. EASY, a neural network simulation environment which was developed using the formal specification, is also described.

Keywords

Neural networks formal specifications object orientation neural network simulation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    G. A. Carpenter and S. Grossberg. A massively parallel architecture for a selforganizing neural pattern recognition machine. Computer Vision, Graphics, and Image Processing, 37:54–115, 1987.Google Scholar
  2. 2.
    E. Fiesler. Neural network classification and formalization. In John Fulcher, editor, Computer Standards & Interfaces, special issue on Neural Network Standards, volume 16. Elsevier Science Publishers B. V., Amsterdam, The Netherlands, 1994. ISSN 0920-5489.Google Scholar
  3. 3.
    L. Fuentes, J. F. Aldana, and J. M. Troya. Urano: An object-oriented artificial neural network simulation tool. In J. Mira, J. Cabestany, and A. Prieto, editors, New Trends in Neural Computation: International Workshop on Artificial Neural Networks — IWANN, pages 364–369, Sitges, Spain, june 1993.Google Scholar
  4. 4.
    R. Hecht-Nielsen. Neurocomputing. Addison-Wesley Publishing Company, Inc, 1990.Google Scholar
  5. 5.
    J. R. Jang. Anfis: Adaptative-network-based fuzzy inference system. IEEE Trans. on Systems, Man, and Cybernetics, May 1993.Google Scholar
  6. 6.
    P. D. L. Machado. EASY — an environment for simulating artificial neural networks. Master's thesis, Federal University of Pernambuco, Informatics Departament, Recife-PE, Brazil, 1994. In Portuguese.Google Scholar
  7. 7.
    P. D. L. Machado, E. C. D. B. Carvalho Filho, S. R. L. Meira, and H. M. Gomes. EASY — an [E]nvironment for [A]rtificial neural [SY]stems simulation. In Proceedings of Fourth Irish Neural Network Conference — INNC'94, Dublin, Ireland, September 1994.Google Scholar
  8. 8.
    P. D. L. Machado, E. C. D. B. Carvalho Filho, S. R. L. Meira, and H. M. Gomes. EASY — an environment for modelling, simulation and analysis of artificial neural networks. In Proceedings of I Brazilian Symposium on Neural Networks, Caxambu, Minas Gerais, Brazil, August 1994. In Portuguese.Google Scholar
  9. 9.
    P. D. L. Machado, S. R. L. Meira, E. C. D. B. Carvalho Filho, and H. M. Gomes. A formal object-oriented specification of artificial neural networks. In Proceedings of VIII Brazilian Symposium on Software Engineering, Curitiba, PR, Brazil, October 1994. In Portuguese.Google Scholar
  10. 10.
    P. D. L. Machado, S. R. L. Meira, E. C. D. B. Carvalho Filho, and H. M. Gomes. Specifying artificial neural networks in MooZ. In Proceedings of XX Conferencia Latino Americana de Informatica — PANEL'94, Mexico, September 1994. In Portuguese.Google Scholar
  11. 11.
    W. S. McCulloch and W. H. Pitts. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys, 5:115–133, 1943. formal neuron.Google Scholar
  12. 12.
    S. R. L. Meira and A. L. C. Cavalcanti. Modular Object-Oriented Z Specifications. In Prof. C. J. van Rijsbergen, editor, Workshop on Computing Series, pages 173–192, Oxford-UK, December 1990. Springer-Verlag.Google Scholar
  13. 13.
    S. R. L. Meira and A. L. C. Cavalcanti. The MooZ Specification Language. Technical report, Universidade Federal de Pernambuco, Departamento de Informática, Recife-PE, 1992. Available via anonymous ftp from rosa.cr-pe.rnp.br, file pub/MooZ/MooZ.ps.Z.Google Scholar
  14. 14.
    B. Meyer. On formalism in specifications. IEEE Software, pages 6–26, january 1985.Google Scholar
  15. 15.
    J. M. J. Murre. Neurosimulators. In M. A. Arbib, editor, HandBook of Brain Research and Neural Networks. MIT Press, MRC APU, Cambridge, 1994.Google Scholar
  16. 16.
    H. R. Myler, A. R. Weeks, R. K. Gillis, and G. W. Hall. Object-oriented neural simulation tools for a hypercube parallel machine. Neurocomputing, 4(5):235–248, 1992.Google Scholar
  17. 17.
    D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representations by error propagation. In Parallel Distributed Processing, chapter 8, pages 318–362. MIT Press, Cambridge, 1986.Google Scholar
  18. 18.
    D. A. Santos. A formal model for the specification of neural networks — MOFEU. Master's thesis, Federal University of Pernambuco, Informatics Departament, Recife-PE, Brazil, 1994. In Portuguese.Google Scholar
  19. 19.
    P. K. Simpson. Artificial Neural Systems: Foundations, Paradigms, Aplications and Implementations. Pergamon Press, Inc, 1990.Google Scholar
  20. 20.
    L. S. Smith. A framework for neural net specification. IEEE Transactions on Software Engineering, 18(7):601–612, july 1992.Google Scholar
  21. 21.
    J. M. Spivey. The Z Notation: A Reference Manual Prentice Hall, 1989.Google Scholar
  22. 22.
    ParcPlace Systems. Visual Works TM Release 1.0 — User's Guide. 999 E. Arques Ave., Sunnyvale, California 94086-4593, 1992.Google Scholar
  23. 23.
    P. D. Wasserman. Neural Computing: Theory and Practice. ANZA Research, Inc, New York, 1989.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1995

Authors and Affiliations

  • Patrícia Duarte de Lima Machado
    • 1
  • Silvio Lemos Meira
    • 1
  1. 1.Department of InformaticsFederal University of PernambucoRecife, PEBrazil

Personalised recommendations