Journal Homepage
 Library Recommendation Card
 Free Sample
 Subscribe
 Buy Book Version
 Buy E-book Version
 Download journal pdf
 Suggest an Update




 

Graphical Models, Exponential Families, and Variational Inference

Foundations and Trends® in
Machine Learning

Volume 1 Issue 1–2
DOI: 10.1561/2200000001

Graphical Models, Exponential Families, and Variational Inference

Martin J. Wainwright
Department of Statistics, and Department of Electrical Engineering and Computer Science, University of California, Berkeley 94720, USA, wainwrig@stat.berkeley.edu

Michael I. Jordan
Department of Statistics, and Department of Electrical Engineering and Computer Science, University of California, Berkeley 94720, USA, jordan@stat.berkeley.edu

SUGGESTED CITATION:
Martin J. Wainwright and Michael I. Jordan (2008) "Graphical Models, Exponential Families, and Variational Inference",
Foundations and Trends® in Machine Learning: Vol. 1: No 1–2, pp 1-305.
http:/dx.doi.org/10.1561/2200000001

Abstract

The formalism of probabilistic graphical models provides a unifying framework for capturing complex dependencies among random variables, and building large-scale multivariate statistical models. Graphical models have become a focus of research in many statistical, computational and mathematical fields, including bioinformatics, communication theory, statistical physics, combinatorial optimization, signal and image processing, information retrieval and statistical machine learning. Many problems that arise in specific instances -- including the key problems of computing marginals and modes of probability distributions -- are best studied in the general setting. Working with exponential family representations, and exploiting the conjugate duality between the cumulant function and the entropy for exponential families, we develop general variational representations of the problems of computing likelihoods, marginal probabilities and most probable configurations. We describe how a wide variety of algorithms -- among them sum-product, cluster variational methods, expectation-propagation, mean field methods, max-product and linear programming relaxation, as well as conic programming relaxations -- can all be understood in terms of exact or approximate forms of these variational representations. The variational approach provides a complementary alternative to Markov chain Monte Carlo as a general source of approximation methods for inference in large-scale statistical models.

Next
Open Journal PDF Buy Book Version
ISBN: 978-1-60198-184-4
List Price $ 125.00 , € 125.00 , £ 99.00
Buy E-book Version
ISBN: 978-1-60198-185-1
List Price $ 200.00 , € 200.00