Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data

Abstract

Most of the semi-supervised classification methods developed so far use unlabeled data for regularization purposes under particular distributional assumptions such as the cluster assumption. In contrast, recently developed methods of classification from positive and unlabeled data (PU classification) use unlabeled data for risk evaluation, i.e., label information is directly extracted from unlabeled data. In this paper, we extend PU classification to also incorporate negative data and propose a novel semi-supervised classification approach. We establish generalization error bounds for our novel methods and show that the bounds decrease with respect to the number of unlabeled data without the distributional assumptions that are required in existing semi-supervised classification methods. Through experiments, we demonstrate the usefulness of the proposed methods.

arXiv commentAccepted to the 34th International Conference on Machine Learning (ICML 2017)

References in corpus (6)

• Laplacian Support Vector Machines Trained in the Primal
• Positive-Unlabeled Learning with Non-Negative Risk Estimator
• PU Learning for Matrix Completion
• Mixture Proportion Estimation via Kernel Embedding of Distributions
• Estimating the class prior and posterior from noisy positives and unlabeled data
• Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning

Cited by in corpus (5)

• Positive-Unlabeled Learning with Non-Negative Risk Estimator
• Efficient Training for Positive Unlabeled Learning
• Semi-Supervised AUC Optimization based on Positive-Unlabeled Learning
• Binary classification with ambiguous training data
• Principled analytic classifier for positive-unlabeled learning via weighted integral probability metric