We will discuss "A Tutorial on Stochastic Approximation Algorithms for Training Restricted Boltzmann Machines and Deep Belief Nets" by Swersky, Chen, Marlin and de Freitas (ITA10: http://www.cs.ubc.ca/~nando/papers/ita2010.pdf). Depending on how much time we take on the first paper, we may also go over derivations for RBM learning methods in "Inductive Principles for Restricted Boltzmann Machine Learning" by Marlin, Swersky, Chen, and de Freitas (AIStats10: http://jmlr.org/proceedings/papers/v9/marlin10a/marlin10a.pdf). 

1. Remember to read this before our meeting  http://www.cs.ubc.ca/~nando/papers/ita2010.pdf
2. Subscribe to the MLRG list: https://mailman.cs.umd.edu/mailman/listinfo/mlrg