Welcome to the Machine Learning Reading Group at Brown University. We meet in CIT Room 345 on Fridays at 1pm and are open to all students and faculty.

MLRG meetings and other news are announced via posts to the ML-READING-GROUP list. You can subscribe here. The MLRG leader this semester is David McClosky.

Our topics this semester are log-linear models, belief propagation, gaussian processes, and Variational Bayes. We will also be hosting some invited talks, speakers permitting. If you have an ongoing or just-completed project involving machine learning, please offer to lead a session!

Date Topic Discussion leaders
January 30th Organizational Meeting
Bring papers and topics
February 6th Intro to MaxEnt Principle and Log-linear models
Main reading: Sutton and McCallum 2006 (only pages 1-24)
Supplemental: Lafferty, McCallum, and Pereira, ICML 2001
Excellent slides: Klein and Manning 2003
Matt Lease and David McClosky
February 13th Log-linear models and CRFs
Same readings as previous week
Matt Lease and David McClosky
February 20th Log-linear models: Estimation
Main reading: Berger 1997
Supplemental: Globerson, Koo, Carreras, and Collins, ICML 2007
Slides: Klein and Manning 2003 are still relevant.
Will Headden
February 29th Belief Propagation
Reading: Yedidia, Freeman, and Weiss 2002 (sections 1.2 and 2)
Aggeliki's slides: (PPT) (PDF)
Aggeliki Tsoli
March 7th Belief Propagation
Same paper as previous week (read later sections)
Jesse Butterfield
March 14th No meeting this week due to job talk Job talks
March 21st Nuts and Boltzmann
Tutorials: Scholarpedia and Sam Roweis's notes
Application and extension: Hinton, Osindero, and Teh (2006)
March 21st Principal Geodesic Analysis
Optional readings:
Wu, Smith, and Hancock (2006) (sections 1-3)
Fletcher, Lu, Pizer, and Joshi (2004) (sections II.B-C, III, and IV)
David Hirshberg and Jason Pacheco
April 4th No meeting this week due to job talks Job talks
April 11th Two-view Feature Generation Model for Semi-supervised Learning
Ando and Zhang (ICML 2007)
April 18th Learning Multiscale Representations of Natural Scenes Using Dirichlet Processes
Jyri Kivinen, Erik Sudderth, and Michael Jordan (ICCV 2007)
April 25th Introduction to Variational Bayes
Tutorial by Tommi Jaakkola (sections 4, 5, and 8)
Will Headden
May 2nd Variational Bayes
Beal and Ghahramani (2004) (sections 2, 3.1, 3.4, 3.5)
May 30th Variational Bayes...once more
Tutorial by Tommi Jaakkola (sections 8 and 7)

Conferences and journals:

Some ML deadlines: