Loopy Belief Propagation for Approximate Inference: An Empirical Study

@inproceedings{murphy99loopy, abstract = {Recently, a number of researchers have demonstrated excellent performance by using \"loopy belief propagation\" --- using Pearl's polytree algorithm in a Bayesian network with loops. The most dramatic instance is the near Shannon-limit performance of \"Turbo Codes\" --- error-correcting codes whose decoding algorithm is equivalent to loopy belief propagation. In this paper we ask: is there something special about the error-correcting code context, or does loopy propagation work as an approximate...}, author = {Murphy, Kevin P. and Weiss, Yair and Jordan, Michael I. }, citeulike-article-id = {2568178}, keywords = {bayesiannetworks, beliefpropagation}, pages = {467--475}, posted-at = {2008-03-20 22:31:54}, priority = {1}, title = {Loopy Belief Propagation for Approximate Inference: An Empirical Study}, url = {http://citeseer.ist.psu.edu/114067.html} }

TY - CONF ID - murphy99loopy L3 - citeulike-article-id:2568178 TI - Loopy Belief Propagation for Approximate Inference: An Empirical Study SP - 467 EP - 475 N2 - Recently, a number of researchers have demonstrated excellent performance by using "loopy belief propagation" --- using Pearl's polytree algorithm in a Bayesian network with loops. The most dramatic instance is the near Shannon-limit performance of "Turbo Codes" --- error-correcting codes whose decoding algorithm is equivalent to loopy belief propagation. In this paper we ask: is there something special about the error-correcting code context, or does loopy propagation work as an approximate... KW - bayesiannetworks KW - beliefpropagation AU - Murphy, Kevin AU - Weiss, Yair AU - Jordan, Michael UR - http://citeseer.ist.psu.edu/114067.html ER -