TY - GEN
T1 - Sampling from Gaussian graphical models using subgraph perturbations
AU - Liu, Ying
AU - Kosut, Oliver
AU - Willsky, Alan S.
PY - 2013
Y1 - 2013
N2 - The problem of efficiently drawing samples from a Gaussian graphical model or Gaussian Markov random field is studied. We introduce the subgraph perturbation sampling algorithm, which makes use of any pre-existing tractable inference algorithm for a subgraph by perturbing this algorithm so as to yield asymptotically exact samples for the intended distribution. The subgraph can have any structure for which efficient inference algorithms exist: for example, tree-structured, low tree-width, or having a small feedback vertex set. The experimental results demonstrate that this subgraph perturbation algorithm efficiently yields accurate samples for many graph topologies.
AB - The problem of efficiently drawing samples from a Gaussian graphical model or Gaussian Markov random field is studied. We introduce the subgraph perturbation sampling algorithm, which makes use of any pre-existing tractable inference algorithm for a subgraph by perturbing this algorithm so as to yield asymptotically exact samples for the intended distribution. The subgraph can have any structure for which efficient inference algorithms exist: for example, tree-structured, low tree-width, or having a small feedback vertex set. The experimental results demonstrate that this subgraph perturbation algorithm efficiently yields accurate samples for many graph topologies.
UR - http://www.scopus.com/inward/record.url?scp=84890407746&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84890407746&partnerID=8YFLogxK
U2 - 10.1109/ISIT.2013.6620676
DO - 10.1109/ISIT.2013.6620676
M3 - Conference contribution
AN - SCOPUS:84890407746
SN - 9781479904464
T3 - IEEE International Symposium on Information Theory - Proceedings
SP - 2498
EP - 2502
BT - 2013 IEEE International Symposium on Information Theory, ISIT 2013
T2 - 2013 IEEE International Symposium on Information Theory, ISIT 2013
Y2 - 7 July 2013 through 12 July 2013
ER -