MESSAGE ERROR ANALYSIS OF LOOPY BELIEF PROPAGATION

被引：2

作者：

Shi, Xiangqiong ^{[1
]}

Schonfeld, Dan ^{[1
]}

Tuninetti, Daniela ^{[1
]}

机构：

[1] Univ Illinois, Chicago, IL 60637 USA

来源：

2010 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING | 2010年

关键词：

Graphical Model; Bayesian Networks; Markov Random Fields; Loopy Belief Propagation; Error Analysis;

D O I：

10.1109/ICASSP.2010.5495075

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

The loopy belief propagation algorithm (LBP) is known to perform extremely well in many practical problems of probability inference and learning on graphical models, even in presence of multiple loops. Although general necessary conditions for convergence of LBP to a unique fixed point solution are still unknown, various techniques have been explored to understand error propagation when LBP fails to converge. In this paper, we rely on the contractive mapping of message errors to present novel distance bounds between multiple fixed point solutions when LBP does not converge. We give examples of networks where our bounds are tighter than existing ones.

引用

页码：2078 / 2081

页数：4