Advertisement

A Generative Learning Algorithm that uses Structural Knowledge of the Input Domain yields a better Multi-layer Perceptron

  • Elena Pérez-Miñana
Conference paper
Part of the Perspectives in Neural Computing book series (PERSPECT.NEURAL)

Abstract

Many classifier applications have been developed using the Multi-layer perceptron (MLP) model as representation form. The main difficulty found in designing an architecture based on the model has been, for the most part, induced by a lack of understanding of what each of an MLP’s network components embodies. Expressing the input domain to a classification task in terms of a subspace in R N , the problem to solve consists of computing an appropriate segmentation of the domain so that every input point will be assigned to a region of the space into which only points of the same class have fallen. This can be achieved with an MLP network if every weight vector is computed as the normal to each of the surfaces in the input domain that will induce the same sort of partitioning that is engendered by the classification criteria associated to the problem for which the network has been built. As the Delaunay Triangulation (DT) of a set of points is a geometric structure in which everything one would ever want to know about the proximity of the points from which it was derived is recorded, it provides an ideal source of information for computing the number and form of those weight vectors, enabling the possibility of building an initial maximal network architecture for a particular problem.

Keywords

Weight Vector Classification Task Voronoi Diagram Delaunay Triangulation Hide Unit 
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]
    N.K. Bose, A.K. Garga, Neural Network Design using Voronoi Diagrams, IEEE Transactions on Neural Networks 1993; 4, 5: 778–787.CrossRefGoogle Scholar
  2. [2]
    A. Bowyer, J. Woodwark, Introduction to Computing with Geometry, Winchester: Information Geometers 1993.Google Scholar
  3. [3]
    N. Burgess, S. Di Zenzo, M. Notturno Granieri, The Generalisation of a Constructive Algorithm in Pattern Classification Problems, International Journal of Neural Systems 1992; 3: 1–6.CrossRefGoogle Scholar
  4. [4]
    G. Cybenko, Continuous valued neural networks with two hidden layers are sufficient, technical report, Department of Computer Science, Tufts University, USA 1988.Google Scholar
  5. [5]
    J. Hertz, A. Krogh, R. Palmer, Introduction to the Theory of Neural Computation, Addison-Wesley 1991.Google Scholar
  6. [6]
    A. Okabe, B. Boots, K. Sugihara, Spatial Tessellations: Concepts and Applications of Voronoi Diagrams, Wiley series in Probability and Statistics, John Wiley and Sons 1992.MATHGoogle Scholar
  7. [7]
    J. O’Rourke, Computational Geometry in C, Cambridge University Press 1994.Google Scholar
  8. [8]
    E. Pérez-Mii ana, Learning Nature of the Feedforward Neural Networks, PhD thesis, Department of Artificial Intelligence, University of Edinburgh 1997.Google Scholar
  9. [9]
    L. Prechelt, Probenl: A Set of Benchmarks and Benchmarking Rules for Neural Network Training Algorithms, Fakultät für Informatik Universität Karlsruhe 1994.Google Scholar

Copyright information

© Springer-Verlag London Limited 1998

Authors and Affiliations

  • Elena Pérez-Miñana
    • 1
  1. 1.Department of Artificial IntelligenceUniversity of Edinburgh EdinburghScotlandUK

Personalised recommendations