Latest News and Events

The SAMSI-FODAVA Workshop on Interactive Visualization and Analysis of Massive Data will be held on December 10-12, 2012.
Posted: October 02, 2012
The FODAVA Annual Meeting will immediately follow (Dec 12-13) the SAMSI/FODAVA joint workshop at the same location.
Posted: September 05, 2012
Many of the modern data sets such as text and image data can be represented in high-dimensional vector spaces and have benefited from computational methods that utilize advanced techniques from num
Posted: June 30, 2012

Affine-invariant Principal Components

Santosh Vempala

Principal component analysis (PCA) is widely used to identify important
directions or subspaces. For a given data set or distribution, it can be
identified with a particular affine transformation. In this talk, we discuss
a notion of principal components which is affine-invariant. We apply this
extension of PCA to a classical problem from statistics, namely unraveling
a mixture of arbitrary Gaussian distributions in high-dimensional space
given unlabeled samples from the mixture. We will discuss the method in
the context of other known methods for dimension reduction and show how
it complements them.