Positive-Unlabeled Classification under Class-Prior Shift: A Prior-invariant Approach Based on Density Ratio Estimation

TL;DR

This paper introduces a novel positive-unlabeled classification method that remains robust under class-prior shift by using density ratio estimation, eliminating the need for prior knowledge during training.

Contribution

The proposed approach is the first to handle class-prior shift in PU learning without requiring prior information during training, supported by theoretical justification and experimental validation.

Findings

Effective in scenarios with class-prior shift

Does not require class-prior knowledge during training

Outperforms existing PU methods under shift conditions

Abstract

Learning from positive and unlabeled (PU) data is an important problem in various applications. Most of the recent approaches for PU classification assume that the class-prior (the ratio of positive samples) in the training unlabeled dataset is identical to that of the test data, which does not hold in many practical cases. In addition, we usually do not know the class-priors of the training and test data, thus we have no clue on how to train a classifier without them. To address these problems, we propose a novel PU classification method based on density ratio estimation. A notable advantage of our proposed method is that it does not require the class-priors in the training phase; class-prior shift is incorporated only in the test phase. We theoretically justify our proposed method and experimentally demonstrate its effectiveness.