Advertisement

A hybrid evolutionary learning system for synthesizing neural network pattern recognition systems

  • Devert Wicker
  • Mateen M. Rizki
  • Louis A. Tamburino
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1447)

Abstract

An approach is introduced for developing neural network pattern recognition systems using a hybrid evolutionary learning system for pattern recognition (HELPR) concept. A genetic algorithm is used to assemble detectors and pattern recognition systems while traditional weight training methods are used to determine weights. The results show that this novel approach develops simpler neural topologies than cascade correlation and can do so using very simple training metrics.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliography

  1. Duda, R.O. and P.E. Hart, 1973. Pattern Classification and Scene Analysis, Wiley.Google Scholar
  2. Fahlman, Scott E. and Christian Lebiere, 1990. “The Cascade-Correlation Learning Architecture,” CMU-CS-90-100.Google Scholar
  3. Fisher, R. A., 1936. “The use of multiple measures in taxonomic problems,” Ann. Eugenics, 7, Part II, pp. 179–188.Google Scholar
  4. Fujita, Osamu, 1992. “Optimization of the Hidden Unit Function in Feedforward Neural Networks,” Neural Networks, Vol. 5, pp. 755–764.Google Scholar
  5. Goldberg, D. E., 1989. Genetic Algorithms in Search, Optimization, and Machine Learning, Reading, MA: AddisonWesley.Google Scholar
  6. Hush, Don R. and Bill G. Home, 1993. “Progress in Supervised Neural Networks,” IEEE SP.Google Scholar
  7. Merz, C.J., & Murphy, P.M. (1996). UCI Repository of machine learning databases [http://www.ics.uci.edu/∼mlearn/MLRepository.html]. Irvine, CA: University of California, Department of Information and Computer Science.Google Scholar
  8. Rizki, M. M., L. A. Tamburino, and M. A. Zmuda, 1993. “Evolving multi-resolution feature detectors.” In Proceedings of the Second Annual Conference on Evolutionary Learning, eds. D. B. Fogel and W. Atmar, La Jolla, CA: Evolutionary Programming Society, pp. 57–66.Google Scholar
  9. Rizki, M. M., L. A. Tamburino, and M. A. Zmuda, 1994. “EMORPH: A two phased learning system for evolving morphological classification systems.” In Proceedings of the Third Annual Conference on Evolutionary Learning, eds. A. V. Sebald and L. J. Fogel, River Edge, NJ: World Scientific, pp. 60–67.Google Scholar
  10. Rizki, M. M., L. A. Tamburino, and M. A. Zmuda, 1995. “Evolution of morphological recognition systems.” In Evolutionary Programming IV, eds. J. R. McDonnell, R. G. Reynolds, and D. B. Fogel, MIT Press, Cambridge, MA, pp. 95–106.Google Scholar
  11. Tamburino, L. A. and M. M. Rizki, 1992. “Performance driven autonomous design of pattern recognition systems”, Intern. J. of App. Art. Intel., 6, pp. 59–77.Google Scholar
  12. Therrien, Charles W., 1989. Decision Estimation and Classification, Wiley.Google Scholar
  13. Wicker, D.W., M.M. Rizki, and L. A. Tamburino, 1996. “HELPR: A hybrid evolutionary system for pattern recognition,” ADPE'96.Google Scholar
  14. Zmuda, M. A., M. M. Rizki, and L. A. Tamburino, 1992. “Automatic generation of morphological sequences,” SPIE Conference on Image Algebra and Morphological Image Processing III, pp. 106–118.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Devert Wicker
    • 1
  • Mateen M. Rizki
    • 2
  • Louis A. Tamburino
    • 1
  1. 1.Air Force Research Laboratory (AFRL/SNAT)Wright-Patterson Air Force BaseDayton
  2. 2.Department of Computer Science and Engr.Wright State UniversityDayton

Personalised recommendations