Machine Learning Reading Group

Organizer: Jesse Davis (jdavis at cs)
Meeting Time: 4:30-5:30 pm on Wednesday
Location: TBD
Mailing List: mlread at cs dot washington dot edu

Format

Each week one person will be responsible for picking a paper for the group to read. This individual will also lead the group discussion. Everyone who attends the meeting should read the paper and contribute to the discussion.

The theme of the reading group this quarter is inference. Discussion leaders should try to pick papers that will have wide appeal. Good strategies are to either pick a survey paper (or tutorial) or a popular/hot paper from a recent conference.

The discussion leader should plan on presenting a 15-20 minute overview of the paper. The presentation does not need to be formal, i.e., no slides are needed. If we are discussing a conference paper, a good tactic is for the discussion leader to ask the author of the paper for the slides from the conference talk.

Here is a link to last semester's papers.

Calendar

  Discussion Leader Paper
April 1st Shawn Ling Gating
April 8th Daniel Lowd Chavira and Darwiche, "Compiling Bayesian networks using variable elimination," IJCAI-07.
April 15th No Class Go to Russ Griener's Talk
April 22nd Cynthia Matusek Guiding Inference with Policy Search Reinforcement Learning
April 29th Fei Wu V. Ganapathi, D. Vickery, J. Ducki and D. Koller," Constrained approximate maximum entropy learning," In Proceedings of the Twenty-fourth Conference on Uncertainty in AI (UAI), 2008.
May 6th Andrey Kolobov Tutorial on MCMC
May 13th Andrew Guillory Jarvis Haupt, Rui Castro and Robert Nowak, "Distilled sensing: selective sampling for sparse signal recovery"
May 20th Alan Ritter Jordan et al, "An introduction to variational methods for graphical models."
May 27th Paul Rosenbloom Towards a Graphical Implementation Level for Cognitive Architecture
June 3rd No Class  

Possible Papers

  • J. Mooij and H. Kappen, "Bounds on marginal probability distributions," In Neural Information Processing Systems (NIPS) 22, 2008. (2 votes)
  • David Sontag, Talya Meltzer, Amir Globerson, Tommi Jaakkola, Yair Weiss: Tightening LP Relaxations for MAP using Message Passing , UAI 2008. (2 votes)
  • Sontag et al., "Clusters and coarse partitions in LP relaxations," In NIPS 2008.
  • Minka and Winn, "Gates," In NIPS 2008.
  • Wexler and Meeks, "MAS," In NIPS 2008.
  • M. Chertkov,"Exactness of belief propagation for some graphical models with loops," Journal of Statistical Mechanics: Theory and Experiment, 2008.
  • D. Sontaga and T. Jaakkola, "New Outer Bounds on the Marginal Polytope," In Neural Information Processing Systems (NIPS) 21, 2007.
  • D. Malioutov, J. Johnson and A. Willsky "Walk-sums and Belief Propogation in Gaussian Graphical Models," Journal of Machine Learning Research, vol. 7, pp. 2031-2064, 2006.
  • L. El Ghaoui, "A Convex upper bound on the log-partition function for graphical models," Technical Report, University of California-Berkeley, 2007.
  • Tamir Hazan and Amnon Shashua: Convergent Message-Passing Algorithms for Inference over General Graphs with Convex Free Energies, UAI 2008
  • T Finley and T Joachims. Training Structural SVMs when Exact Inference is Intractable, ICML 2008
  • Chieu, H. L.; Lee, W. S.; Teh, Y. W. Cooled and relaxed survey propagation for MRFs NIPS 07.
  • Choi, A.; Darwiche, A. Focusing Generalizations of Belief Propagation on Targeted Queries, AAAI 08.
  • Gogate, V.; Dechter, R. AND/OR Importance Sampling, UAI 08.