# DEDPUL: Difference-of-Estimated-Densities-based Positive-Unlabeled Learning

## TL;DR

DEDPUL introduces a novel, efficient method for positive-unlabeled learning that estimates class proportions and classifies data without relying on neural networks or prior proportion knowledge, outperforming existing methods.

## Contribution

The paper presents DEDPUL, a new approach that estimates class proportions and classifies unlabeled data without needing known proportions or neural network classifiers.

## Key findings

- DEDPUL outperforms state-of-the-art methods in proportion estimation.
- DEDPUL achieves higher classification accuracy on PU learning tasks.
- The method is computationally efficient and classifier-agnostic.

## Abstract

Positive-Unlabeled (PU) learning is an analog to supervised binary classification for the case when only the positive sample is clean, while the negative sample is contaminated with latent instances of positive class and hence can be considered as an unlabeled mixture. The objectives are to classify the unlabeled sample and train an unbiased PN classifier, which generally requires to identify the mixing proportions of positives and negatives first. Recently, unbiased risk estimation framework has achieved state-of-the-art performance in PU learning. This approach, however, exhibits two major bottlenecks. First, the mixing proportions are assumed to be identified, i.e. known in the domain or estimated with additional methods. Second, the approach relies on the classifier being a neural network. In this paper, we propose DEDPUL, a method that solves PU Learning without the aforementioned issues. The mechanism behind DEDPUL is to apply a computationally cheap post-processing procedure to the predictions of any classifier trained to distinguish positive and unlabeled data. Instead of assuming the proportions to be identified, DEDPUL estimates them alongside with classifying unlabeled sample. Experiments show that DEDPUL outperforms the current state-of-the-art in both proportion estimation and PU Classification.