Semi-Supervised AUC Optimization based on Positive-Unlabeled Learning

Abstract

Maximizing the area under the receiver operating characteristic curve (AUC) is a standard approach to imbalanced classification. So far, various supervised AUC optimization methods have been developed and they are also extended to semi-supervised scenarios to cope with small sample problems. However, existing semi-supervised AUC optimization methods rely on strong distributional assumptions, which are rarely satisfied in real-world problems. In this paper, we propose a novel semi-supervised AUC optimization method that does not require such restrictive assumptions. We first develop an AUC optimization method based only on positive and unlabeled data (PU-AUC) and then extend it to semi-supervised learning by combining it with a supervised AUC optimization method. We theoretically prove that, without the restrictive distributional assumptions, unlabeled data contribute to improving the generalization performance in PU and semi-supervised AUC optimization methods. Finally, we demonstrate the practical usefulness of the proposed methods through experiments.

Fixed typos in Appendix

References in corpus (1)

• Class-prior Estimation for Learning from Positive and Unlabeled Data

Cited by in corpus (7)

• On Symmetric Losses for Learning from Corrupted Labels
• Learning with Multiple Complementary Labels
• An Effective Multi-Resolution Hierarchical Granular Representation based Classifier using General Fuzzy Min-Max Neural Network
• Class-Imbalanced Complementary-Label Learning via Weighted Loss
• Pointwise Binary Classification with Pairwise Confidence Comparisons
• Anomaly Detection with Inexact Labels
• A Symmetric Loss Perspective of Reliable Machine Learning