Learning A Disentangling Representation For PU Learning

TL;DR

This paper introduces a neural network-based representation learning method for PU learning that effectively separates positive and negative data clusters, improving classification performance in high-dimensional settings.

Contribution

It proposes a novel clustering-based representation learning approach with vector quantization for PU learning, supported by theoretical justification and experimental validation.

Findings

Outperforms current state-of-the-art methods on simulated data

Effectively separates positive and negative clusters in high-dimensional data

Provides theoretical insights into the clustering approach

Abstract

In this paper, we address the problem of learning a binary (positive vs. negative) classifier given Positive and Unlabeled data commonly referred to as PU learning. Although rudimentary techniques like clustering, out-of-distribution detection, or positive density estimation can be used to solve the problem in low-dimensional settings, their efficacy progressively deteriorates with higher dimensions due to the increasing complexities in the data distribution. In this paper we propose to learn a neural network-based data representation using a loss function that can be used to project the unlabeled data into two (positive and negative) clusters that can be easily identified using simple clustering techniques, effectively emulating the phenomenon observed in low-dimensional settings. We adopt a vector quantization technique for the learned representations to amplify the separation between the learned unlabeled data clusters. We conduct experiments on simulated PU data that demonstrate the improved performance of our proposed method compared to the current state-of-the-art approaches. We also provide some theoretical justification for our two cluster-based approach and our algorithmic choices.