Skip to main content
SpringerLink
Log in
Book cover

International Conference on Parallel Problem Solving from Nature

PPSN 1990: Parallel Problem Solving from Nature pp 274–283Cite as

Performance evaluation of evolutionarily created neural network topologies

  • Neural Networks
  • Conference paper
  • First Online:

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 496))

Abstract

The paper presents a method for building-up application adapted topologies for Backpropagation-trained neural networks. We investigate the influence of topology on classification performance and speed by comparing fixed and evolutionary created networks which were trained to recognize handwritten digits.

This is a preview of subscription content, log in via an institution.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. György Barna, Kimuno Kaski, Choosing Optimal Network Structure, proceedings of the international Neural Network Conference, Paris 1990

    Google Scholar 

  2. James L. McClelland, David E. Rumelhart, Explorations in Parallel Distributed processing,Vol. 1–3, MIT Press, 3rd revision, 1988

    Google Scholar 

  3. Nigel Dodd, Optimisation of Network Structure Using GeneticTechniques, proceedings of the International Neural Network Conference, Paris 1990

    Google Scholar 

  4. David H. Hubel, Eye, Brain and Vision,Scientific American Library, 1988

    Google Scholar 

  5. Hahn-Ming Lee, Ching-Chi Hsu, Training of a Neural Network with Topology Generation for the Classification Problem, proceedings of the International Neural Network Conference, Paris 1990

    Google Scholar 

  6. Ian D. Longstaff, John F. Cross, A pattern recognition approach to understanding the multilayer perceptron, Pattern Recognition Letters 5 (1987), 315–319, Elsevier North-Holland

    Article  Google Scholar 

  7. Fernando M. Silva, Luis B. Almeida, Speeding up Backpropagation, Advanced Neural Computers, Ed. R. Eckmiller, Elsevier North-Holland, 1990

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institute of Physics, University of Koblenz, Rheinau 3-4, D-5400, Koblenz

    Wolfram Schiffmann, Merten Joost & Randolf Werner

Authors
  1. Wolfram Schiffmann
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Merten Joost
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Randolf Werner
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Hans-Paul Schwefel Reinhard Männer

Rights and permissions

Reprints and permissions

Copyright information

© 1991 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Schiffmann, W., Joost, M., Werner, R. (1991). Performance evaluation of evolutionarily created neural network topologies. In: Schwefel, HP., Männer, R. (eds) Parallel Problem Solving from Nature. PPSN 1990. Lecture Notes in Computer Science, vol 496. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0029764

Download citation

  • DOI: https://doi.org/10.1007/BFb0029764

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-54148-6

  • Online ISBN: 978-3-540-70652-6

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation