[Meshfree-methods-seminar] Time and Place

[Qi Ye]

Dear All,

Dr. Jun Zhang will talk in our seminar tormorrow morning at 10:00 AM on Sep. 30th Tuesday. The location is still at AM 222. And the topic is "Laplacian Eigenmaps: a theoretical manifold learning appraoch" and the abstract is as below.

In recent years manifold  methods have become popular in machine learning and have seen numerous applications in data analysis including dimension reduction, visualization, clustering and classification. The unifying assumption of these methods is that the data resides on or near a low dimentional manifold. In the talk, I will introduce Laplacian Eigenmaps, an approach based on the Laplace-Beltrami operator on the underlying manifold, and explain its connections with the prevailing manifold algorithms including Diffusion Maps, Isomap, Hessian Eigenmaps and Locally Linear Embedding. Last, the applications to image processing and  statistical genetics for inferrence of population structure will be presented. 

Looking forward to meeting you tomorrow.

Sincerely,

Qi Ye
Department of Applied Mathematics
Illinois Institute of Technology 
Chicago, IL 60616
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://math.iit.edu/pipermail/meshfree-methods-seminar/attachments/20080929/0a196a38/attachment.htm