[Colloquium] Sindhwani talk next Tuesday 12:15 at TTI
Meridel Trimble
mtrimble at tti-c.org
Fri Sep 10 16:26:38 CDT 2004
TOYOTA TECHNOLOGICAL INSTITUTE TALK
Speaker: Vikas Sindhwani
Speakers homepage: http://www.cs.uchicago.edu/~vikass
Title: Manifold Regularization : A Geometric Framework for Learning from
Examples.
Time: Tuesday, September 14th, 12:15pm
Place: TTI-C conference room (1427 E. 60th St. 2nd Floor)
Pizza lunch provided
Abstract: We propose a family of learning algorithms based on a new form of
regularization that allows us to exploit the geometry of the marginal
distribution. We focus on a semi-supervised framework that incorporates
labeled and unlabeled data in a general-purpose learner. Some
transductive graph learning algorithms and standard methods including
Support Vector Machines and Regularized Least Squares can be obtained as
special cases. We utilize properties of Reproducing Kernel Hilbert spaces
to prove new Representer theorems that provide theoretical basis for the
algorithms. As a result (in contrast to purely graph based approaches) we
obtain a natural out-of-sample extension to novel examples and so are able
to handle both transductive and truly semi-supervised settings. We
present experimental evidence suggesting that our semi-supervised
algorithms are able to use unlabeled data effectively. Finally, we briefly
discuss unsupervised and fully supervised learning within our general
framework.
This is joint work with Mikhail Belkin and Partha Niyogi.
If you have questions, or would like to meet the speaker, please contact
Meridel at 4-9873 or mtrimble at tti-c.org
More information about the Colloquium
mailing list