Efficient Training for Positive Unlabeled Learning

Abstract

Positive unlabeled (PU) learning is useful in various practical situations, where there is a need to learn a classifier for a class of interest from an unlabeled data set, which may contain anomalies as well as samples from unknown classes. The learning task can be formulated as an optimization problem under the framework of statistical learning theory. Recent studies have theoretically analyzed its properties and generalization performance, nevertheless, little effort has been made to consider the problem of scalability, especially when large sets of unlabeled data are available. In this work we propose a novel scalable PU learning algorithm that is theoretically proven to provide the optimal solution, while showing superior computational and memory performance. Experimental evaluation confirms the theoretical evidence and shows that the proposed method can be successfully applied to a large variety of real-world problems involving PU learning.

Submitted to IEEE TPAMI

References in corpus (6)

• Semi-Supervised Learning with Deep Generative Models
• One-Class Classification: Taxonomy of Study and Review of Techniques
• PU Learning for Matrix Completion
• Class-prior Estimation for Learning from Positive and Unlabeled Data
• Theoretical Comparisons of Positive-Unlabeled Learning against Positive-Negative Learning
• Semi-Supervised Classification Based on Classification from Positive and Unlabeled Data

Cited by in corpus (10)

• Positive-Unlabeled Learning with Non-Negative Risk Estimator
• Toward Deep Supervised Anomaly Detection: Reinforcement Learning from Partially Labeled Anomaly Data
• Simplify and Robustify Negative Sampling for Implicit Collaborative Filtering
• Principled analytic classifier for positive-unlabeled learning via weighted integral probability metric
• Multi-Complementary and Unlabeled Learning for Arbitrary Losses and Models
• Deep Anomaly Detection with Deviation Networks
• A method on selecting reliable samples based on fuzziness in positive and unlabeled learning
• Multi-Class Classification from Single-Class Data with Confidences
• Learning from Similarity-Confidence Data
• Generating Relevant Counter-Examples from a Positive Unlabeled Dataset for Image Classification