Graph Partition Strategies for Generalized Mean Field Inference

Eric P. Xing and Michael I. Jordan

EECS Department, University of California, Berkeley

Technical Report No. UCB/CSD-03-1274

2003

This publication is archived. It is kept only for reference purposes, so it is no longer being updated and may not meet accessibility standards. If you need this content in a different format, please email webteam@eecs.berkeley.edu.

An autonomous variational inference algorithm for arbitrary graphical model requires the ability to optimize variational approximations over the space of model parameters as well as over the choice of tractable families used for the variational approximation. In this paper, we present a novel combination of graph partitioning algorithms with a generalized mean field (GMF) inference algorithm. This combination optimizes over disjoint clustering of variables and performs inference using those clusters. We explore several graph partition strategies empirically, including several variants of MinCut and MaxCut algorithms. Our results provide rather clear support for a weighted version of MinCut as a useful clustering algorithm for generalized mean field inference.

BibTeX citation:

@techreport{Xing:CSD-03-1274,
    Author= {Xing, Eric P. and Jordan, Michael I.},
    Title= {Graph Partition Strategies for Generalized Mean Field Inference},
    Year= {2003},
    Month= {Jun},
    Url= {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2003/6237.html},
    Number= {UCB/CSD-03-1274},
    Abstract= {An autonomous variational inference algorithm for arbitrary graphical model requires the ability to optimize variational approximations over the space of model parameters as well as over the choice of tractable families used for the variational approximation. In this paper, we present a novel combination of graph partitioning algorithms with a generalized mean field (GMF) inference algorithm. This combination optimizes over disjoint clustering of variables and performs inference using those clusters. We explore several graph partition strategies empirically, including several variants of MinCut and MaxCut algorithms. Our results provide rather clear support for a weighted version of MinCut as a useful clustering algorithm for generalized mean field inference.},
}

EndNote citation:

%0 Report
%A Xing, Eric P. 
%A Jordan, Michael I. 
%T Graph Partition Strategies for Generalized Mean Field Inference
%I EECS Department, University of California, Berkeley
%D 2003
%@ UCB/CSD-03-1274
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2003/6237.html
%F Xing:CSD-03-1274