Consistency of plug-in confidence sets for classification in semi-supervised learning

DOI10.1080/10485252.2019.1689241MaRDI QIDQ5221297zbMATHOpenAlexFDO

Authors Christophe Denis, Mohamed Hebiri

Publication date 25 March 2020

Published in Journal of Nonparametric Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1507.07235

zbMATH Keywords

classificationconfidence setsclassification with reject optionplug-in confidence sets

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Learning and adaptive systems in artificial intelligence (68T05) Nonparametric tolerance and confidence regions (62G15) Inference from stochastic processes and prediction (62M20)

Abstract: Confident prediction is highly relevant in machine learning; for example, in applications such as medical diagnoses, wrong prediction can be fatal. For classification, there already exist procedures that allow to not classify data when the confidence in their prediction is weak. This approach is known as classification with reject option. In the present paper, we provide new methodology for this approach. Predicting a new instance via a confidence set, we ensure an exact control of the probability of classification. Moreover, we show that this methodology is easily implementable and entails attractive theoretical and numerical properties.

Recommendations

• ANALYSIS OF CLASSIFICATION WITH A REJECT OPTION
• Classification with confidence
• Learning with Rejection
• On semi-supervised learning
• Confidence sets with expected sizes for multiclass classification

Cites work

• scientific article; zbMATH DE number 1332320 (Why is no real title available?)
• scientific article; zbMATH DE number 2168212 (Why is no real title available?)
• scientific article; zbMATH DE number 835699 (Why is no real title available?)
• A distribution-free theory of nonparametric regression
• Agnostic pointwise-competitive selective classification
• Asymptotic Statistics
• Classification with a reject option using a hinge loss
• Classification with confidence
• Classification with reject option
• Distribution-Free Prediction Sets
• Fast learning rates for plug-in classifiers
• Knows what it knows: a framework for self-aware learning
• On optimum recognition error and reject tradeoff
• On the foundations of noise-free selective classification
• Optimal aggregation of classifiers in statistical learning.
• Support vector machines with a reject option
• The tight constant in the Dvoretzky-Kiefer-Wolfowitz inequality

Cited in

(6)

• Error rate control for classification rules in multiclass mixture models
• Confidence sets with expected sizes for multiclass classification
• scientific article; zbMATH DE number 7370598 (Why is no real title available?)
• Variance function estimation in regression model via aggregation procedures
• Set-Valued Support Vector Machine with Bounded Error Rates
• Classification with confidence

This page was built for publication: Consistency of plug-in confidence sets for classification in semi-supervised learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5221297)