loopy belief propagation

(1.8 hours to learn)


The sum-product and max-product algorithms give exact answers for tree graphical models, but if we apply the same update rules on a general graph, it often gives pretty reasonable results. This is known as loopy belief propagation, and it is a widely used approximate inference algorithm in coding theory and low level vision.


This concept has the prerequisites:

Core resources (read/watch one of the following)


Supplemental resources (the following are optional, but you may find them useful)


Coursera: Probabilistic Graphical Models (2013)


See also