ASSESSING ROBUSTNESS OF CLASSIFICATION USING ANGULAR BREAKDOWN POINT.

TitleASSESSING ROBUSTNESS OF CLASSIFICATION USING ANGULAR BREAKDOWN POINT.
Publication TypeJournal Article
Year of Publication2018
AuthorsZhao, Junlong, Guan Yu, and Yufeng Liu
JournalAnn Stat
Volume46
Issue6B
Pagination3362-3389
Date Published2018 Dec
ISSN0090-5364
Abstract

Robustness is a desirable property for many statistical techniques. As an important measure of robustness, breakdown point has been widely used for regression problems and many other settings. Despite the existing development, we observe that the standard breakdown point criterion is not directly applicable for many classification problems. In this paper, we propose a new breakdown point criterion, namely angular breakdown point, to better quantify the robustness of different classification methods. Using this new breakdown point criterion, we study the robustness of binary large margin classification techniques, although the idea is applicable to general classification methods. Both bounded and unbounded loss functions with linear and kernel learning are considered. These studies provide useful insights on the robustness of different classification methods. Numerical results further confirm our theoretical findings.

DOI10.1214/17-AOS1661
Alternate JournalAnn Stat
Original PublicationAssessing robustness of classification using angular breakdown point.
PubMed ID30294050
PubMed Central IDPMC6168219
Grant ListP01 CA142538 / CA / NCI NIH HHS / United States
R01 GM126550 / GM / NIGMS NIH HHS / United States
Project: