This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 320
Type: Invited
Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
Sponsor: IMS
Abstract - #306033
Title: A Selective Overview and New Perspectives on Nonparametric Classification
Author(s): Richard Samworth*+
Companies: University of Cambridge
Address: Centre for Mathematical Sciences, Cambridge, International, CB3 0WB, United Kingdom
Keywords: Nonparametric classification ; $k$-nearest neighbor classifier ; Bagged nearest neighbor classifier ; Log-concave density estimation
Abstract:

There are many different types of nonparametric classifiers. Some are based on generative models, others on discriminative models, for instance. Classifiers may be very simple, such as the $k$-nearest neighbor classifier (which discards considerable information in the data) or more sophisticated. Most methods involve the choice of at least one tuning parameter, and sometimes more than one, particularly in multidimensional contexts. I will present a selective overview of different nonparametric classifiers and their relative merits, and discuss new fully automatic nonparametric classifiers (with no tuning parameters to choose) based on estimating the underlying densities under shape constraints such as log-concavity.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.