Speaker: Misha Belkin

Title: Manifold Methods and Semi-supervised Learning

Abstract: 
The concept of a Riemannian manifold provides a very general
notion of the nonlinear dependency in the data. In my talk I will discuss
how different problems of machine learning may be approached in the
manifold setting. In particular, I will discuss semi-supervised learning
which seems particularly suited for the manifold methods.
I will also present some encouraging experimental results.
This is joint work with Partha Niyogi.