Advertisement

ICANN ’94 pp 573-576 | Cite as

Cascade Correlation convergence theorem

  • Gian Paolo Drago
  • Sandro Ridella

Abstract

In this paper we prove a theorem which guarantees that the Cascade Correlation algorithm has at least a speed of convergence of order O(1/n h ), where n h is the number of hidden neurons, when approximating a function which is series of sigmoids with a finite number of terms.

Keywords

Hide Layer Finite Number Hide Neuron Output Neuron 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]
    Barron A.R., IEEE Trans. I. T., vol.39, no.3, May 1993, pp.930–945.Google Scholar
  2. [2]
    Drago G.P. and Ridella S., Proc. of the ICANN’93, pp.750–754.Google Scholar
  3. [3]
    Drago G.P. and Ridella S., submittedGoogle Scholar
  4. [4]
    Fahlman S.E. and Lebiere C., CMU-CS-90–100, February 14, 1990.Google Scholar
  5. [5]
    Jones L.K., Annals of Statistics, vol.20, no. l, 1992, pp.608–613.Google Scholar

Copyright information

© Springer-Verlag London Limited 1994

Authors and Affiliations

  • Gian Paolo Drago
    • 1
  • Sandro Ridella
    • 2
  1. 1.Istituto per i Circuiti Elettronici - CNRGenovaItaly
  2. 2.DIBE - Universita’ di GenovaGenovaItaly

Personalised recommendations