Electronic Journal of Statistics

Lasso type classifiers with a reject option

Marten Wegkamp

Full-text: Open access

Abstract

We consider the problem of binary classification where one can, for a particular cost, choose not to classify an observation. We present a simple proof for the oracle inequality for the excess risk of structural risk minimizers using a lasso type penalty.

Article information

Source
Electron. J. Statist., Volume 1 (2007), 155-168.

Dates
First available in Project Euclid: 16 May 2007

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1179332302

Digital Object Identifier
doi:10.1214/07-EJS058

Mathematical Reviews number (MathSciNet)
MR2312148

Zentralblatt MATH identifier
1320.62153

Subjects
Primary: 62C05: General considerations
Secondary: 62G05: Estimation 62G08: Nonparametric regression

Keywords
Bayes classifiers classification convex surrogate loss empirical risk minimization hinge loss large margin classifiers ℓ_{1} penalties local mutual coherence margin condition reject option support vector machines

Citation

Wegkamp, Marten. Lasso type classifiers with a reject option. Electron. J. Statist. 1 (2007), 155--168. doi:10.1214/07-EJS058. https://projecteuclid.org/euclid.ejs/1179332302


Export citation