Learnability and the Vapnik-Chervonenkis dimension

  title={Learnability and the Vapnik-Chervonenkis dimension},
  author={A. Blumer and A. Ehrenfeucht and D. Haussler and Manfred K. Warmuth},
  journal={J. ACM},
  • A. Blumer, A. Ehrenfeucht, +1 author Manfred K. Warmuth
  • Published 1989
  • Mathematics, Computer Science
  • J. ACM
  • Valiant's learnability model is extended to learning classes of concepts defined by regions in Euclidean space En. The methods in this paper lead to a unified treatment of some of Valiant's results, along with previous results on distribution-free convergence of certain pattern recognition algorithms. It is shown that the essential condition for distribution-free learnability is finiteness of the Vapnik-Chervonenkis dimension, a simple combinatorial parameter of the class of concepts to be… CONTINUE READING
    1,861 Citations

    Topics from this paper

    Complexity Dimensions and Learnability
    • 3
    • PDF
    Deciding the Vapnik-Cervonenkis dimension is ...-complete
    • 13
    • PDF
    Results on learnability and the Vapnick-Chervonenkis dimension
    • 29
    • PDF
    Deciding the Vapnik-Červonenkis Dimension in ∑p3-Complete
    • M. Schaefer
    • Computer Science, Mathematics
    • J. Comput. Syst. Sci.
    • 1999
    • 19
    Lower Bounds on the Vapnik-Chervonenkis Dimension of Convex Polytope Classifiers
    • G. Takács, B. Pataki
    • Mathematics
    • 2007 11th International Conference on Intelligent Engineering Systems
    • 2007
    • 6
    Complexity of Computing Vapnik-Chervonenkis Dimension
    • 8
    • PDF
    Vapnik-Chervonenkis Dimension Bounds for Two- and Three-Layer Networks
    • P. Bartlett
    • Mathematics, Computer Science
    • Neural Computation
    • 1993
    • 20


    A theory of the learnable
    • 1,787
    • Highly Influential
    • PDF
    Convergence of stochastic processes
    • 2,031
    • Highly Influential