Gaussian inference in loopy graphical models1 Conference Paper

abstract

• We show precisely that message passing for inference in Gaussian graphical models on singly connected graphs is just a distributed implementation of Gaussian elimination without any need for backsubstitution. This observation allows us to generalize the procedure to arbitrary loopy Gaussian graphical models. We thus construct a message passing algorithm that is guaranteed to converge in finite time, and solve the inference problem exactly. The complexity of this algorithm grows gradually with the "distance" of the graph to a tree. This algorithm can be implemented in a distributed environment as, for example, in sensor networks.

name of conference

• 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475)

authors

• Kumar, Panganamala

published proceedings

• 42nd IEEE International Conference on Decision and Control (IEEE Cat. No.03CH37475)

author list (cited authors)

• Plarre, K., & Kumar, P. R.

citation count

• 0

complete list of authors

• Plarre, Kurt||Kumar, PR

publication date

• January 2003

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• IEEE Conference on Decision and Control  Journal

Digital Object Identifier (DOI)

• 10.1109/cdc.2003.1271921

International Standard Book Number (ISBN) 10

• 0780379241

start page

• 5747

end page

• 5752

volume

• 6

URL

• http://dx.doi.org/10.1109/cdc.2003.1271921