• DocumentCode
    1205677
  • Title

    A metric entropy bound is not sufficient for learnability

  • Author

    Kulkarni, Sanjeev R. ; Richardson, Tom ; Zeitouni, O.

  • Volume
    40
  • Issue
    3
  • fYear
    1994
  • fDate
    5/1/1994 12:00:00 AM
  • Firstpage
    883
  • Lastpage
    885
  • Abstract
    The authors prove by means of a counterexample that it is not sufficient, for probably approximately correct (PAC) learning under a class of distributions, to have a uniform bound on the metric entropy of the class of concepts to be learned. This settles a conjecture of Benedek and Itai (1991)
  • Keywords
    entropy; information theory; learning (artificial intelligence); probability; learnability; metric entropy bound; probably approximately correct learning; uniform bound; Algebra; Entropy; Extraterrestrial measurements; Intelligent control; Mathematics; Notice of Violation; Random variables; Sufficient conditions;
  • fLanguage
    English
  • Journal_Title
    Information Theory, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9448
  • Type

    jour

  • DOI
    10.1109/18.335898
  • Filename
    335898
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=1205677