• DocumentCode
    3547595
  • Title

    Hardware-based support vector machine classification in logarithmic number systems

  • Author

    Khan, Faisal M. ; Arnold, Mark G. ; Pottenger, William M.

  • Author_Institution
    Dept. of Comput. Sci. & Eng., Lehigh Univ., Bethlehem, PA, USA
  • fYear
    2005
  • fDate
    23-26 May 2005
  • Firstpage
    5154
  • Abstract
    Support vector machines are emerging as a powerful machine-learning tool. Logarithmic number systems (LNS) utilize the property of logarithmic compression for numerical operations. We present an implementation of a digital support vector machine (SVM) classifier using LNS in which, when compared with other implementations, considerable hardware savings are achieved with no significant loss in classification accuracy.
  • Keywords
    data compression; learning (artificial intelligence); pattern classification; support vector machines; LNS; classification accuracy; digital SVM classifier; hardware savings; hardware-based support vector machine classification; logarithmic compression; logarithmic number systems; machine-learning tool; numerical operations; Application software; Computer science; Event detection; Hardware; Kernel; Machine learning; Neural networks; Power engineering and energy; Support vector machine classification; Support vector machines;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems, 2005. ISCAS 2005. IEEE International Symposium on
  • Print_ISBN
    0-7803-8834-8
  • Type

    conf

  • DOI
    10.1109/ISCAS.2005.1465795
  • Filename
    1465795