Speaker: Chien-I Liao
Title: On generalization bounds, projection profile, and margin distribution (based on a paper by Ashutosh Garg, Sariel Har-Peled and Dan Roth)
Date: 24th January, 2005

We study generalization properties of linear learning algorithms and develop a data dependent approach that is used to derive generalization bounds that depend on the margin distribution. Our method uses random projection techniques to allow the use of existing VC dimension bounds in the effective, lower, dimension of the data. Our bounds are tighter than existing bounds and (sometimes) give informative generalization bounds for real world, high dimensional problems.

Link to the paper: http://cs.nyu.edu/~cil217/ML/MarginDistribution.pdf