treeDecomp
The package treeDecomp computes a tree decomposition of a given graphical model (either a Bayesian network or a Weighted CSP).
Author
Description
The library contains the implementations of the tree decomposition of a graphical model. The elimination ordering can be computed either by the min-fill heuristic or by the hypergraph partitioning heuristic. The graphical models supported are: Bayesian networks and Weighted CSPs.
Input format
The belief network has to be specified in the Simple_file_format
The Weighted CSP has to be specified in the WCSP_file_format
For some example problems please see our Repository.
Usage
The algorithm is invoked with the following command line arguments:
treeDecomp -f <networkFile> -h <heuristic> -fj <treeDecompFile>
Parameters
Parameters for the algorithm, which can be specified within the parameter file:
f: (string) specifies the input file name
h: (string) the heuristic to use for finding an variable elimination order by which to construct the tree decomposition. The following values can be used:
minfill: to indicate the min-fill heuristic
hypergraph: to indicate the hypergraph partitioning heuristic
fj: (string) specifies the output file name that contains the tree decomposition. The file will be saved in the Graphviz dot format.
Download
A 32-bit Linux binary and an example parameter file is available here.
If you are interested in obtaining the source code, please contact the program author.
References
Kalev Kask, Rina Dechter, Javier Larrosa and Avi Dechter. Unifying Cluster-Tree Decompositions for Reasoning in Graphical Models. In Artificial Intelligence Journal, 2005. Link