Classification with a Reject Option using a Hinge Loss

@article{Bartlett2008ClassificationWA,
  title={Classification with a Reject Option using a Hinge Loss},
  author={Peter L. Bartlett and Marten H. Wegkamp},
  journal={Journal of Machine Learning Research},
  year={2008},
  volume={9},
  pages={1823-1840}
}
We consider the problem of binary classification where the classifier can, for a particular cost, choose not to classify an observation. Just as in the conventional classification problem, minimization of the sample average of the cost is a difficult optimization problem. As an alternative, we propose the optimization of a certain convex loss function φ, analogous to the hinge loss used in support vector machines (SVMs). Its convexity ensures that the sample average of this surrogate loss can… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 161 CITATIONS, ESTIMATED 29% COVERAGE

On Reject and Refine Options in Multicategory Classification

VIEW 12 EXCERPTS
CITES BACKGROUND & METHODS
HIGHLY INFLUENCED

Support vector machines with a reject option

VIEW 6 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Active Learning for Binary Classification with Abstention

Shubhanshu Shekhar, Mohammad Ghavamzadeh, Tara Javidi
  • ArXiv
  • 2019
VIEW 8 EXCERPTS
HIGHLY INFLUENCED

Binary Classification with Bounded Abstention Rate

VIEW 11 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Stochastic Negative Mining for Learning with Large Output Spaces

  • AISTATS
  • 2018
VIEW 5 EXCERPTS
CITES BACKGROUND
HIGHLY INFLUENCED

Boosting with Abstention

VIEW 5 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

Learning with Rejection

VIEW 7 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

Agnostic Pointwise-Competitive Selective Classification

  • J. Artif. Intell. Res.
  • 2015
VIEW 13 EXCERPTS
CITES BACKGROUND, METHODS & RESULTS
HIGHLY INFLUENCED

Design and Analysis of Consistent Algorithms for Multiclass Learning Problems

VIEW 8 EXCERPTS
CITES METHODS & BACKGROUND
HIGHLY INFLUENCED

FILTER CITATIONS BY YEAR

2007
2019

CITATION STATISTICS

  • 20 Highly Influenced Citations

  • Averaged 22 Citations per year over the last 3 years

  • 7% Increase in citations per year in 2018 over 2017

Similar Papers

Loading similar papers…