Double Ramp Loss Based Reject Option Classifier

Abstract

We consider the problem of learning reject option classifiers. The goodness of a reject option classifier is quantified using $0-d-1$ loss function wherein a loss $d\in (0,.5)$ is assigned for rejection. In this paper, we propose {\em double ramp loss} function which gives a continuous upper bound for $(0-d-1)$ loss. Our approach is based on minimizing regularized risk under the double ramp loss using {\em difference of convex (DC) programming}. We show the effectiveness of our approach through experiments on synthetic and benchmark datasets. Our approach performs better than the state of the art reject option classification approaches.

References in corpus (6)

• Scikit-learn: Machine Learning in Python
• Identifying Mislabeled Training Data
• Noise Tolerance under Risk Minimization
• Making Risk Minimization Tolerant to Label Noise
• Loss factorization, weakly supervised learning and label noise robustness
• Lasso type classifiers with a reject option

Cited by in corpus (5)

• Active learning for data streams: a survey
• Distributed Deep Forest and its Application to Automatic Detection of Cash-out Fraud
• Contrastive Learning Improves Model Robustness Under Label Noise
• Trained Without My Consent: Detecting Code Inclusion In Language Models Trained on Code
• Do We Really Need Gold Samples for Sample Weighting Under Label Noise?