• DocumentCode
    3774915
  • Title

    Deep learning: Architectures, algorithms, applications

  • Author

    Roland Memisevic

  • Author_Institution
    University of Montreal
  • fYear
    2015
  • Firstpage
    1
  • Lastpage
    127
  • Abstract
    This article consists of a collection of slides from the author´s conference presentation. Some of the topics covered include: Machine learning 101: Neural nets, backprop, RNNs; Applications; Structured prediction; Unsupervised learning; "Neural Programs"; Architecture exploration; Towards hardware-friendlier DL; and Software.
  • Keywords
    "Machine learning","Unsupervised learning","Artificial neural networks","Recurrent neural networks","Learning (artificial intelligence)"
  • Publisher
    ieee
  • Conference_Titel
    Hot Chips 27 Symposium (HCS), 2015 IEEE
  • Type

    conf

  • DOI
    10.1109/HOTCHIPS.2015.7477319
  • Filename
    7477319