Efficient learning in sparsely connected Boltzmann machines

  • Marcel J. Nijman
  • Hilbert J. Kappen
Oral Presentations: Theory Theory II: Learning
First Online:
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1112)

Abstract

We present a heuristical procedure for efficient estimation of the partition function in the Boltzmann distribution. The resulting speed-up is of immediate relevance for the speed-up of Boltzmann Machine learning rules, especially for networks with a sparse connectivity.

Keywords

Partition Function Bayesian Network Gibbs Sampling Learning Rule Boltzmann Machine 
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.
This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    D.H. Ackley, G.E. Hinton, and T.J. Sejnowski. A learning algorithm for Boltzmann Machines. Cognitive Science, 9:147–169, 1985.Google Scholar
  2. 2.
    M.R. Garey and D.S. Johnson. Computers and Intractability: A Guide to the Theory of NP-Completeness. Freeman, San Francisco, 1979.Google Scholar
  3. 3.
    C. Itzykson and J. Drouffe. Statistical Field Theory. Cambrigde University Press, Cambridge, 1991.Google Scholar
  4. 4.
    B. Kappen and M.J. Nijman. Radial basis Boltzmann machines and learning with missing values. In Proceedings of the World Congress on Neural Networks, volume I, pages 72–75. Lawrence Erlbaum Associates, Mahwah, NJ, 1995.Google Scholar
  5. 5.
    R.M. Karp. Reducibility among combinatorial problems. In R.E. Miller and J.W. Thatcher, editors, Complexity of Computer Computations, pages 85–103. Plenum Press, 1972.Google Scholar
  6. 6.
    S. Kullback. Information theory and statistics. Wiley, N.Y., 1959.Google Scholar
  7. 7.
    J. Pearl. Probabilistic reasoning using graphs. In B. Bouchon and R.R. Yager, editors, Uncertainty in Knowledge-Based Systems, pages 200–202. Springer-Verlag, 1987.Google Scholar
  8. 8.
    L. Saul and M. Jordan. Learning in Boltzmann trees. Neural Computation, 6(6):1174–1184, 1994.Google Scholar
  9. 9.
    D.M.J. Tax and H.J. Kappen. Learning structure with many-take-all networks. In proceedings of ICANN, 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1996

Authors and Affiliations

  • Marcel J. Nijman
    • 1
    • 2
  • Hilbert J. Kappen
    • 1
    • 2
  1. 1.Real World Computing Partnership Novel Functions, Dept. of Medical Physics and BiophysicsUniversity of NijmegenEZ NijmegenThe Netherlands
  2. 2.Dutch Foundation for Neural Networks Laboratory Dept. of Medical Physics and BiophysicsUniversity of NijmegenEZ NijmegenThe Netherlands

Personalised recommendations