|
|
|
|
The DLR Hierarchy of Approximate Inference
Michal Rosen-Zvi, Michael Jordan, Alan Yuille
Abstract:
We propose a hierarchy for approximate inference
based on the Dobrushin, Lanford,
Ruelle (DLR) equations. This hierarchy includes
existing algorithms, such as belief
propagation, and also motivates novel algorithms
such as factorized neighbors (FN) algorithms
and variants of mean eld (MF) algorithms.
In particular, we show that extrema
of the Bethe free energy correspond to
approximate solutions of the DLR equations.
In addition, we demonstrate a close connection
between these approximate algorithms
and Gibbs sampling. Finally, we compare
and contrast various of the algorithms in the
DLR hierarchy on spin-glass problems. The
experiments show that algorithms higher up
in the hierarchy give more accurate results
when they converge but tend to be less stable.
Keywords:
Pages: 493-500
PS Link:
PDF Link: /papers/05/p493-rosen-zvi.pdf
BibTex:
@INPROCEEDINGS{Rosen-Zvi05,
AUTHOR = "Michal Rosen-Zvi
and Michael Jordan and Alan Yuille",
TITLE = "The DLR Hierarchy of Approximate Inference",
BOOKTITLE = "Proceedings of the Proceedings of the Twenty-First Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-05)",
PUBLISHER = "AUAI Press",
ADDRESS = "Arlington, Virginia",
YEAR = "2005",
PAGES = "493-500"
}
|
|
