CS712 Advanced Topics in Aitificial Intelligence:
Probabilistic Graphical Models
Spring 2010


Information
Syllabus
Assignments
Project
Selected Readings

Tentative Syllabus

Day Topic Reading Optional Reading
3/29/10 Introduction M. Jordan, "Graphical models," Statistical Science, 19:140-155, 2004.
3/31/10 Semantics of Bayesian networks and Markov random fields L 2.1.1; P 3
4/05/10 Computing probabilities: elimination algorithm
4/07/10 Equivalence of chordal graph, decomposability and completeness of minimal separators; Maximum cardinality search algorithm L 2.1.2-4; P 3.2.4 R. Tarjan and M. Yannakakis, "Simple linear time algorithms to test chordality of graphs, test acyclicity of hypergraphs, and selectively reduce acyclic hypergraphs," SIAM J. Comput., 13(3):566--579, 1984.
4/12/10 Heuristics to triangulate a graph; Properties of conditional independence L 3.1; P 3.1.2-4 F. Jensen and F. Jensen. "Optimal junction trees," Proceedings of the Tenth Conference Uncertainty and Artificial Intelligence, 360-366, 1994.
4/14/10 Markov properties on MRFs L 3.2.1
4/19/10 Factorization property on MRFs; Hammersley/Clifford theorem L 3.2.1 P. Clifford, "Markov random fields in statistics," In Disorder in Physical Systems, G. Grimmett and D. Welsh (Eds.), 19?2, Oxford University Press, 1990; G. Grimmett, "A theorem about random fields," Bulletin of the London Mathematical Society 5, 81-84, 1973.
4/21/10 Markov properties on Bayesian networks L 3.2.2
4/26/10 Bayesian networks and directed factorization
4/28/10 Phylogenetic trees and chordal models, mixture Models, inference in HMMs Phylogenetic trees at Wikipedia
5/03/10 Inference on trees, undirected trees and poly trees P 4.2-3 Judea Pearl's classic paper; Belief Propagation at WikiPedia;
5/05/10 Exact inference on junction trees Lauritzen and Spiegelhalter, "Local computation with probabilities on graphical structures and their application to expert systems," J. Roy. Statist. Soc. B, 157?24, 1988.
5/10/10 Message passing protocol and local consistency
5/12/10 Generalized distribution law and Shenoy-Shafer formulation Aji and McEliece, "The generalized distributive law," IEEE Transactions on Information Theory, 46(2):325-343, 2000; Shafer and Shenoy, "Probability propagation," Annals of Mathematics and Artificial Intelligence, 1:(1-4):327--351, 1990; Kschischang and Frey, "Factor graphs and the sum-product algorithm," IEEE Transactions on Information Theory, 47(2):498-519, 2001
5/17/10 Computational complexity, time-space complexity tradeoffs G. Cooper, "The computational complexity of probabilistic inference using Bayesian belief networks," Artificial Intelligence, 42(2-3):393-405, 1990; V. Chandrasekaran, N. Srebro and P. Harsha, "Complexity of inference in graphical models," UAI 2008.
5/19/10 Gaussian graphical models L 5 Weiss and Freeman, "Correctness of belief propagation in Gaussian graphical models of arbitrary topology," Neural Computation, 13:2173?200, 2001; Malioutov, Johnson and Willsky, "Walk-sums and belief propagation in Gaussian graphical models," Journal of Machine Learning Research, 7:2031-2064, 2006
5/24/10 Memorial Day Holiday (University Closed)
5/26/10 Structure learning N. Srebro, "Maximum likelihood bounded tree-width Markov networks," Artificial Intelligence 143(1):123-138, 2003, slides; C. Chow and C. Liu, "Approximating discrete probability distributions with dependence trees," IEEE Transactions on Information Theory, 14(3):462-467, 1968.
5/31/10 Approximate inference: generalized belief propagation, Bethe and Kikuchi free energy, cluster variation method J. Yedidia, W. Freeman and Y. Weiss, "Constructing free-energy approximations and generalized belief propagation algorithms," IEEE Transactions on Information Theory, 51(7):2282-2312, 2005; M. Wainwright and M. Jordan, "A variational principle for graphical models," In New Directions in Statistical Signal Processing: From Systems to Brain. MIT Press.
6/02/10
6/07/08 Project presentation