Graphical Models Reading Group

This is the wiki page for topics to be discussed in the Graphical Models Reading Group, starting Spring 2013.

We will hold informal meetings to discuss papers regarding Graphical Models. The field is very expansive, and as such paper topics may include exact/approximate inference techniques, variational methods, structure/parameter learning, interesting applications, and so on. Each weekly meeting will have a discussion leader who will both propose the paper to be discussed and get the discussion ball rolling for the meeting.

For questions, or other, please contact either:

• John Halloran - halloj3 [at] ee.wash....edu
• Scott Wisdom - swisdom [at] ee.wash....edu

Announcements

Wiki created, first meeting quickly approaching.

Meeting Schedule

First meeting to take place between May1-3. If you are interested in coming, please fill out the following doodle poll so that we can decide on a meeting time and room: http://doodle.com/7u5ztt639683z5p6

we'll be discussing a recent paper by Po-Ling Loh and Martin J Wainwright titled "Structure estimation for discrete graphical models: Generalized covariance matrices and their inverses" from NIPS 2012, where they prove conditions under which the (generalized)information matrix between jointly distributed discrete random variables denotes the edges in the corresponding graphical model of the random variables.

The paper is available [| here].

Email list

You can subscribe here: https://mailman.cs.washington.edu/mailman/listinfo/graphicalmodels-rg

Prior Meetings

Date	Paper	Authors	Venue	Leader	Info