Number of the records: 1  

Diffusion estimation of mixture models with local and global parameters

  1. 1.
    0461646 - ÚTIA 2017 RIV ES eng C - Conference Paper (international conference)
    Dedecius, Kamil - Sečkárová, Vladimíra
    Diffusion estimation of mixture models with local and global parameters.
    Proceedings of the 2016 IEEE Workshop on Statistical Signal Processing. Palma de Mallorca, Španělsko: IEEE, 2016, s. 362-366. ISBN 978-1-4673-7802-4.
    [2016 IEEE Statistical Signal Processing Workshop. Palma de Mallorca (ES), 26.06.2016-29.06.2016]
    R&D Projects: GA ČR(CZ) GP14-06678P; GA ČR GA13-13502S
    Institutional support: RVO:67985556
    Keywords : diffusion estimation * distributed estimation * exponential family
    Subject RIV: BD - Theory of Information
    http://library.utia.cas.cz/separaty/2016/AS/dedecius-0461646.pdf

    The state-of-art methods for distributed estimation of mixtures assume the existence of a common mixture model. In many practical situations, this assumption may be too restrictive, as a subset of parameters may be purely local, e.g., if the numbers of observable components differ across the network. To reflect this issue, we propose a new online Bayesian method for simultaneous estimation of local parameters, and diffusion estimation of global parameters. The algorithm consists of two steps. First, the nodes perform local estimation from own observations by means of factorized prior/posterior distributions. Second, a diffusion optimization step is used to merge the nodes' global parameters estimates. A simulation example demonstrates improved performance in estimation of both parameters sets.
    Permanent Link: http://hdl.handle.net/11104/0261345

     
     
Number of the records: 1  

  This site uses cookies to make them easier to browse. Learn more about how we use cookies.