• DocumentCode
    2603189
  • Title

    Unsupervised learning in constrained linear networks

  • Author

    Palmieri, F. Rancesco ; Zhu, Jie

  • Author_Institution
    Dept. of Electr. & Syst. Eng., Connecticut Univ., Storrs, CT, USA
  • fYear
    1991
  • fDate
    4-5 Apr 1991
  • Firstpage
    9
  • Lastpage
    10
  • Abstract
    Constrained linear architectures which learn in unsupervised mode according to Hebb´s rule to minimize the output energy are analyzed. Under which conditions such networks act as decorrelating (square-root) filters is investigated. An algorithm is proposed which is almost optimum since the inputs of each filter (which are also the outputs of the net) become more and more decorrelated as the algorithm progresses. The search essentially approaches Newton´s algorithm. The results of a simulation are shown
  • Keywords
    learning systems; neural nets; Hebb´s rule; Newton´s algorithm; constrained linear architectures; constrained linear networks; decorrelating filters; output energy minimization; square-root filters; unsupervised learning; Convergence; Decorrelation; Ducts; Filters; Hebbian theory; Intelligent networks; Power engineering and energy; Systems engineering and theory; Unsupervised learning; Vectors;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Bioengineering Conference, 1991., Proceedings of the 1991 IEEE Seventeenth Annual Northeast
  • Conference_Location
    Hartford, CT
  • Print_ISBN
    0-7803-0030-0
  • Type

    conf

  • DOI
    10.1109/NEBC.1991.154555
  • Filename
    154555