Advertisement

Version-space induction with multiple concept languages

  • Claudio Carpineto
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 549)

Abstract

The version space approach suffers from two main problems, i. e. inability of inducing concepts consistent with data due to use of a restricted hypothesis space and lack of computational efficiency. In this paper we investigate the use of multiple concept languages in a version space approach. We define a graph of languages ordered by the size of their associated concept sets, and provide a procedure for efficiently inducing version spaces while shifting from small to larger concept languages. We show how this framework can help overcome the two above-mentioned limitations. Also, compared to other work on language shift, our approach suggests an alternative strategy for searching the space of new concepts, which is not based on constructive operators.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [Bund85]
    Bundy, A., Silver, B., Plummer, D. (1985). An analytical comparison of some rule-learning problems. Artificial Intelligence, 27.Google Scholar
  2. [Carp91]
    Carpineto, C. (1991). Analytical negative generalization and empirical negative generalization are not cumulative: a case study. In Proceedings of EWSL-91, Lecture Notes on Artificial Intelligence, Springer-Verlag.Google Scholar
  3. [Diet82]
    Dietterich, T.G., London B., Clarkson K., Dromey R.G. (1982). Learning and inductive inference. In Cohen & Feigenbaum (Eds.) The Handbook of Artificial Intelligence, Morgan Kaufmann.Google Scholar
  4. [Haus88]
    Haussler, D. (1988). Quantifying inductive bias: Artificial Intelligence learning algorithms and Valiant's learning framework. Artificial Intelligence, 36.Google Scholar
  5. [Hirs89]
    Hirsh, H. (1989). Combining Empirical and Analytical Learning with Version Spaces. In Proceedings of the sixth international workshop on Machine Learning. Syracuse, Morgan Kaufmann.Google Scholar
  6. [Math89]
    Matheus, C.J., Rendell, L.A. (1989). Constructive induction on decision trees. In Proceedings of the 11th IJCAI, Detroit, Morgan Kaufmann.Google Scholar
  7. [Mich83]
    Michalski, R.S. (1983). A theory and methodology of inductive learning. Artificial Intelligence, 20, 111–161Google Scholar
  8. [Mitc82]
    Mitchell, T.M. (1982). Generalization as Search. Artificial Intelligence, 18, 203–226.Google Scholar
  9. [Paga89]
    Pagallo, G.(1989). Learning DNF by Decision Trees. In Proceedings of the 11th IJCAI, Detroit, Morgan Kaufmann.Google Scholar
  10. [Smit90]
    Smith, Rosenbloom, P., (1990). Incremental Non-Backtracking Focusing: A Poliniomally Bounded Generalization Algorithm for Version Spaces. In Proceedings of the eighth AAAI, Boston, Morgan Kaufmann.Google Scholar
  11. [Subr86]
    Subramanian, D., Feigenbaum, J. (1986). Factorization in Experiment Generation. In Proceedings of the fifth AAAI, Philadelphia, Morgan Kaufmann.Google Scholar
  12. [Utgo86]
    Utgoff, P.E. (1986). Shift of bias for inductive concept learning. In R. Michalski et al. (Eds), Machine Learning II. Morgan Kaufmann, Palo Alto, California.Google Scholar
  13. [Wogu89]
    Wogulis, J., Langley, P. (1989). Improving efficiency by learning intermediate concepts. In Proceedings of the 11th IJCAI, Detroit, Morgan Kaufmann.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • Claudio Carpineto
    • 1
  1. 1.Fondazione Ugo BordoniRome

Personalised recommendations