Graphical Modeling with Perfect Graphs

Code by Tony Jebara, 2011.

This website is a working repository of some of the functions used to convert graphical models into a maximum weight stable set problem on either a perfect graph or a claw-free graph. Matlab files are provided here for easy dissemination to other researchers and students.

This code is work in progress. It converts a graphical model's potential functions so that they are each a product of a log-sparse potential function and singleton potential functions. This then makes them more likely to form a maximum weight stable set problem on a perfect graph. The conversion code forfactoring a potential function is here:

An additional supporting file is here:

This code is based on the paper (please cite):
Perfect Graphs and Graphical Modeling by Tony Jebara, a chapter in the book Tractability, Cambridge University Press, 2012.

This material is based upon work supported by the National Science Foundation under Grant No. 1117631. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.