• DocumentCode
    3184326
  • Title

    Lower Bounds on the Vapnik-Chervonenkis Dimension of Convex Polytope Classifiers

  • Author

    Takács, Gábor ; Pataki, Béa

  • Author_Institution
    Budapest Univ. of Technol. & Econ., Budapest
  • fYear
    2007
  • fDate
    June 29 2007-July 2 2007
  • Firstpage
    145
  • Lastpage
    148
  • Abstract
    In statistical learning theory, the Vapnik-Chervonenkis (VC) dimension is an important combinatorial property of classifier families. In this paper we examine the case of convex polytope classification, i.e. when the separation of the two classes is done by a convex surface consisting of linear segments. We collect the known facts about the VC dimension of convex polytope classifiers with n facets in Rd and present two new lower bounds (one for the general case and one for the special case d = 4).
  • Keywords
    combinatorial mathematics; learning (artificial intelligence); pattern classification; statistical analysis; VC dimension; Vapnik-Chervonenkis dimension; classifier family combinatorial property; convex polytope classification; lower bounds; statistical learning theory; Classification tree analysis; Decision trees; Error analysis; Error probability; Information systems; Labeling; Nearest neighbor searches; Statistical learning; Vectors; Virtual colonoscopy;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Engineering Systems, 2007. INES 2007. 11th International Conference on
  • Conference_Location
    Budapest
  • Print_ISBN
    1-4244-1147-5
  • Electronic_ISBN
    1-4244-1148-3
  • Type

    conf

  • DOI
    10.1109/INES.2007.4283688
  • Filename
    4283688
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3184326