• DocumentCode
    166222
  • Title

    Belief propagation and learning in convolution multi-layer factor graphs

  • Author

    Palmieri, F.A.N. ; Buonanno, A.

  • Author_Institution
    Dipt. di Ing. Ind. e dell´Inf., Seconda Univ. di Napoli (SUN), Aversa, Italy
  • fYear
    2014
  • fDate
    26-28 May 2014
  • Firstpage
    1
  • Lastpage
    6
  • Abstract
    In modeling time series, convolution multi-layer graphs are able to capture long-term dependence at a gradually increasing scale. We present an approach to learn a layered factor graph architecture starting from a stationary latent models for each layer. Simulations of belief propagation are reported for a three-layer graph on a small data set of characters.
  • Keywords
    belief maintenance; graph theory; time series; belief propagation; convolution multilayer factor graphs; layered factor graph architecture; long-term dependence; stationary latent models; three-layer graph; time series modeling; Approximation methods; Bayes methods; Belief propagation; Computer architecture; Convolution; Hidden Markov models; Tin; Belief Propagation; Deep Belief Graphs; Factor Graphs;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Cognitive Information Processing (CIP), 2014 4th International Workshop on
  • Conference_Location
    Copenhagen
  • Type

    conf

  • DOI
    10.1109/CIP.2014.6844500
  • Filename
    6844500
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=166222