Dr. Subhajit Datta, IIT Kanpur

Description
Title: On perfect classification for Gaussian processes

Abstract: We study the problem of discriminating Gaussian processes by analyzing the behavior of the underlying probability measures in an infinite-dimensional space. Motivated by singularity of a certain class of Gaussian measures, we first propose a data based transformation for the training data. For a J class classification problem, this transformation induces complete separation among the associated Gaussian processes. The misclassification probability of a component-wise classifier when applied on this transformed data asymptotically converges to zero. In finite samples, the empirical classifier is constructed and related theoretical properties are studied.

This is a joint work with Juan A. Cuesta-Albertos.
Description
Ramanujan Hall
Date
Tue, March 14, 2017
Start Time
2:00pm IST
Priority
5-Medium
Access
Public
Created by
DEFAULT ADMINISTRATOR
Updated
Wed, March 15, 2017 9:35pm IST