Alternate Estimation of a Classifier and the Class-Prior from Positive and Unlabeled Data
This work addresses a specific challenge in PU learning for scenarios with limited labeled data, offering an incremental improvement over existing methods.
The paper tackles the problem of learning a binary classifier and estimating the class-prior from positive and unlabeled data (PU learning) by proposing a unified approach that alternates between these tasks, addressing the suboptimality of two-step methods that ignore estimation errors.
We consider a problem of learning a binary classifier only from positive data and unlabeled data (PU learning) and estimating the class-prior in unlabeled data under the case-control scenario. Most of the recent methods of PU learning require an estimate of the class-prior probability in unlabeled data, and it is estimated in advance with another method. However, such a two-step approach which first estimates the class prior and then trains a classifier may not be the optimal approach since the estimation error of the class-prior is not taken into account when a classifier is trained. In this paper, we propose a novel unified approach to estimating the class-prior and training a classifier alternately. Our proposed method is simple to implement and computationally efficient. Through experiments, we demonstrate the practical usefulness of the proposed method.