Advertisement

Rejection of incorrect answers from a neural net classifier

  • F. J. Śmieja
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 686)

Abstract

The notion of approximator rejection is described, and applied to a neural network. For a real world classification problem the residual error is shown to decrease with the inverse exponential of the fraction of patterns rejected. The trade-off of “good” patterns rejected and “bad” patterns rejected is shown to increase approximately linearly with rejection rate. A compromise is therefore necessary between trade-off/rejection rate and residual error. A meta-level solution is proposed for removal of the residual error, through use of a modular system of parallel approximators.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    U. Beyer and F. J. Śmieja. Quantitative aspects of data-driven information processing. Technical Report #812, Gesellschaft für Mathematik und Datenverarbeitung, St Augustin, Germany, March 1993.Google Scholar
  2. [2]
    S. Hubrig-Schaumburg. Handwritten character recognition using a reflective modular neural network system. Master's thesis, Bonn University, Germany, 1992.Google Scholar
  3. [3]
    R. P. Lippmann. An introduction to computing with neural nets. IEEE ASSP Magazine, April 1987.Google Scholar
  4. [4]
    H. Mühlenbein. Editorial. Parallel Computing, 14(3):247–248, August 1990. special edition on neural networks.Google Scholar
  5. [5]
    D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representations by error propagation. Nature, 323(533), 1986.Google Scholar
  6. [6]
    O. G. Selfridge. Pandemonium: a paradigm for learning. In The Mechanisation of Thought Processes: Proceedings of a Symposium Held at the National Physical Laboratory, November 1958, pages 511–527, London: HMSO, 1958.Google Scholar
  7. [7]
    F. J. Śmieja. Neural network constructive algorithms: Trading generalization for learning efficiency? Circuits, Systems and Signal Processing, 12(2):331–374, 1993.Google Scholar
  8. [8]
    F. J. Śmieja and H. Mühlenbein. The geometry of multilayer perceptron solutions. Parallel Computing, 14:261–275, 1990.Google Scholar
  9. [9]
    F. J. Śmieja and H. Mühlenbein. Reflective modular neural network systems. Technical Report #633, GMD, Sankt Augustin, Germany, February 1992.Google Scholar
  10. [10]
    F. Weber. Self-reflective exploration of the kinematics of a two-joint robot arm. Diplomarbeit, University of Bonn, Germany, 1992. in German.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • F. J. Śmieja
    • 1
  1. 1.German National Research Centre for Computer Science (GMD)St. Augustin 1Germany

Personalised recommendations