Massachusetts Institute of Technology
Sign in | Create Account

Classification and Dimensionality Reduction Using Geometric Level Sets

Learning classifier decision boundaries and an informative two-dimensional linear subspace jointly through variational level set methods and optimization on the Stiefel manifold. The two classes of measurements, the black + markers and the magenta x markers are eight-dimensional. Only the first two dimensions are informative for classification: the points are separable by an ellipse in those two dimensions. The other six dimensions are independent Gaussian noise. Starting from a random initialization for a two-dimensional subspace, and a poor initialization for a decision boundary, the method converges to the plane of the first two dimensions and an elliptical decision boundary that separates the two classes. The blue line is the decision boundary. The green polygon outlines the projection of an eight-dimensional hypercube onto two dimensions using the current dimensionality reduction mapping. When aligned to two dimensions, the polygon is a square. Visit http://lids.mit.edu/research/project-highlights/classification-and-dimensionality.html for more information.

Comments (1)

insert here

Posted 6 months by Anonymous 00:00:20

You need to log in, in order to post comments. If you don’t have an account yet, sign up now!

SSG Videos

SSG Videos

Category: Research | Updated over 4 years ago

Created
January 11, 2010 16:30
Category
Tags
License
All Rights Reserved (What is this?)
Additional Files


Viewed
17014 times