|
Activity Number:
|
432
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #321411
|
View Presentation
|
|
Title:
|
Nonparametric Classification Using a Forest Dependency Structure
|
|
Author(s):
|
Mary Frances Dorn* and Clifford Spiegelman and Amit Moscovich and Boaz Nadler
|
|
Companies:
|
Texas A&M University and Texas A&M University and Weizmann Institute of Science and Weizmann Institute of Science
|
|
Keywords:
|
classification ;
graphical models ;
forest distributions
|
|
Abstract:
|
We propose a new nonparametric approach for binary classification that exploits the modeling flexibility of sparse graphical models. We assume that each class can be represented by a family of undirected sparse graphical models, specifically a forest-structured distribution. Our procedure requires the nonparametric estimation of only one- and two-dimensional marginal densities to transform the data into a space where a linear classifier is optimal. Experiments with simulated and real data indicate that the proposed method is competitive with popular methods across a range of applications.
|
Authors who are presenting talks have a * after their name.
