Student Seminar Series
DATE: Wednesday November 2nd, 2016, 12:00pm
LOCATION: MSB 1147 (Colloquium Room).
SPEAKERS: Justin Wang, Dept Statistics UC Davis
TITLE: “Techniques for Nonlinear Dimension Reduction”
ABSTRACT: Dimension reduction is a class of statistical methods that aims to reduce observations in a high-dimensional space to a lower dimensional space. In other words, the goal is to extract a set of features that best capture the essence ofp the original data. The most well-known technique for dimension reduction is principal components analysis, which projects high-dimensional observations into a lower dimensional linear subspace. Principal components analysis, however, is limited by the requirement that the data lies on or near a lower dimensional linear subspace. Nonlinear dimension reduction is a subclass of dimension reduction techniques that do not require the linearity assumption. In this talk we will discuss two distinct techniques for dimension reduction: deep autoencoders and locally linear embedding.
This seminar series is organized by PhD Students Irene Kim and Clark Fitzgerald.