Convergence rate analysis of Gaussian belief propagation for Markov networks

被引:5
作者
Zhang, Zhaorong [1 ]
Fu, Minyue [1 ,2 ,3 ]
机构
[1] Univ Newcastle, Sch Elect Engn & Comp Sci, Univ Dr, Callaghan, NSW 2308, Australia
[2] Guangdong Univ Technol, Sch Automat, Guangzhou 510006, Peoples R China
[3] Guangdong Key Lab Intelligent Decis & Cooperat Co, Guangzhou 510006, Peoples R China
基金
中国国家自然科学基金;
关键词
Belief propagation; Density functional theory; Graphical models; Distributed algorithms; SYSTEMS;
D O I
10.1109/JAS.2020.1003105
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Gaussian belief propagation algorithm (GaBP) is one of the most important distributed algorithms in signal processing and statistical learning involving Markov networks. It is well known that the algorithm correctly computes marginal density functions from a high dimensional joint density function over a Markov network in a finite number of iterations when the underlying Gaussian graph is acyclic. It is also known more recently that the algorithm produces correct marginal means asymptotically for cyclic Gaussian graphs under the condition of walk summability (or generalised diagonal dominance). This paper extends this convergence result further by showing that the convergence is exponential under the generalised diagonal dominance condition, and provides a simple bound for the convergence rate. Our results are derived by combining the known walk summability approach for asymptotic convergence analysis with the control systems approach for stability analysis.
引用
收藏
页码:668 / 673
页数:6
相关论文
共 18 条
  • [11] SAAD Y, 2003, ITERATIVE METHODS SP, DOI DOI 10.1137/1.9780898718003
  • [12] Gaussian Belief Propagation Solver for Systems of Linear Equations
    Shental, Ori
    Siegel, Paul H.
    Wolf, Jack K.
    Bickson, Danny
    Dolev, Danny
    [J]. 2008 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY PROCEEDINGS, VOLS 1-6, 2008, : 1863 - +
  • [13] On Convergence Conditions of Gaussian Belief Propagation
    Su, Qinliang
    Wu, Yik-Chung
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2015, 63 (05) : 1144 - 1155
  • [14] Convergence Analysis of the Variance in Gaussian Belief Propagation
    Su, Qinliang
    Wu, Yik-Chung
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2014, 62 (19) : 5119 - 5131
  • [15] Computationally Efficient Sparse Bayesian Learning via Belief Propagation
    Tan, Xing
    Li, Jian
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2010, 58 (04) : 2010 - 2021
  • [16] Correctness of belief propagation in Gaussian graphical models of arbitrary topology
    Weiss, Y
    Freeman, WT
    [J]. NEURAL COMPUTATION, 2001, 13 (10) : 2173 - 2200
  • [17] A Novel Statistical Manifold Algorithm for Position Estimation
    Xia, Bin
    Yuan, Wenhao
    Xie, Nan
    Li, Caihong
    [J]. IEEE-CAA JOURNAL OF AUTOMATICA SINICA, 2019, 6 (06) : 1513 - 1518
  • [18] Networked control systems: a survey of trends and techniques
    Zhang, Xian-Ming
    Han, Qing-Long
    Ge, Xiaohua
    Ding, Derui
    Ding, Lei
    Yue, Dong
    Peng, Chen
    [J]. IEEE-CAA JOURNAL OF AUTOMATICA SINICA, 2020, 7 (01) : 1 - 17