Principal Components: Foundations, Applications, and Computation

CS Dept. Head Candidate

Jesse L. Barlow
Department of Computer Science and Engineering
The Pennsylvania State University
University Park, PA

Monday, March 17, 2003
11 a.m. - 12 noon
Fuller Labs 320

Abstract

Principal components are an important statistical tool in a number of scientific disciplines. In computer science, they arise in latent semantic indexing, face recognition in vision, and hyperspectral imaging. They have also been used in oceanography, meterology, and in geostatistics. Principal components are recovered from computing the leading singular values and vectors of a factor of the covariance matrix.

This presentation will discuss the motivation behind principal components, their application to various scientific disciplines, and how they are computed. Ongoing work on methods for computing and updating principal component is discussed. A new fast algorithm for computing principal components is introduced.

Host

Computer Science Dept Head Search Committee

Refreshments will be served.

Maintained by webmaster@wpi.edu
Last modified: Sep 27, 2006, 16:05 EDT