Loopy belief propagation in the presence of determinism

被引：0
|
作者：
20160501867051
机构：
来源：
(1) University of Texas at Dallas, Richardson; TX; 75080, United States | 1600年 / Amazon; Digile; Facebook; Google; Springer卷 / Microtome Publishing期
关键词：
It is well known that loopy Belief propagation (LBP) performs poorly on probabilistic graphical models (PGMs) with determinism. In this paper; we propose a new method for remedying this problem. The key idea in our method is finding a reparameterization of the graphical model such that LBP; when run on the reparameterization; is likely to have better convergence properties than LBP on the original graphical model. We propose several schemes for finding such reparam-eterizations; all of which leverage unique properties of zeros as well as research on LBP convergence done over the last decade. Our experimental evaluation on a variety of PGMs clearly demonstrates the promise of our method - it often yields accuracy and convergence time improvements of an order of magnitude or more over LBP;
D O I：
暂无
中图分类号：
学科分类号：
摘要：
117374
引用
收藏
相关论文
共 50 条
  • [1] Loopy Belief Propagation in the Presence of Determinism
    Smith, David
    Gogate, Vibhav
    ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 33, 2014, 33 : 895 - 903
  • [2] Using the loopy belief propagation in Siguo
    Xia, Zhengyou
    Zhu, Yongping
    Lu, Hui
    ICGA JOURNAL, 2007, 30 (04) : 209 - 220
  • [3] Sequential clustering by loopy belief propagation
    Ott, T
    Dauwels, J
    Stoop, R
    PROCEEDINGS OF THE 2005 EUROPEAN CONFERENCE ON CIRCUIT THEORY AND DESIGN, VOL 3, 2005, : 297 - 300
  • [4] MESSAGE ERROR ANALYSIS OF LOOPY BELIEF PROPAGATION
    Shi, Xiangqiong
    Schonfeld, Dan
    Tuninetti, Daniela
    2010 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, 2010, : 2078 - 2081
  • [5] Accuracy of Loopy belief propagation in Gaussian models
    Nishiyama, Yu
    Watanabe, Sumio
    NEURAL NETWORKS, 2009, 22 (04) : 385 - 394
  • [6] Loopy belief propagation and probabilistic image processing
    Tanaka, K
    Inoue, JI
    Titterington, DM
    2003 IEEE XIII WORKSHOP ON NEURAL NETWORKS FOR SIGNAL PROCESSING - NNSP'03, 2003, : 329 - 338
  • [7] Foreground detection using loopy belief propagation
    Tu, Gang J.
    Karstoft, Henrik
    Pedersen, Lene J.
    Jorgensen, Erik
    BIOSYSTEMS ENGINEERING, 2013, 116 (01) : 88 - 96
  • [8] EMG signal decomposition by loopy belief propagation
    Koch, VM
    Loeliger, HA
    2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING, 2005, : 397 - 400
  • [9] On the uniqueness of loopy belief propagation fixed points
    Heskes, T
    NEURAL COMPUTATION, 2004, 16 (11) : 2379 - 2413
  • [10] Fast loopy belief propagation for topological SAM
    Pinto Selvatici, Antonio Henrique
    Reali Costa, Anna Helena
    2007 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-9, 2007, : 670 - +
12345