Multi-class Classification from Multiple Unlabeled Datasets with Partial Risk Regularization

TL;DR

This paper introduces a novel method for multi-class classification using multiple unlabeled datasets with known class priors, employing partial risk regularization to prevent overfitting and improve accuracy.

Contribution

It develops an unbiased risk estimator from unlabeled data with class priors and proposes a partial risk regularization technique to enhance learning stability.

Findings

Effectively mitigates overfitting in unlabeled data classification

Outperforms state-of-the-art methods on multiple unlabeled datasets

Provides theoretical analysis of generalization error

Abstract

Recent years have witnessed a great success of supervised deep learning, where predictive models were trained from a large amount of fully labeled data. However, in practice, labeling such big data can be very costly and may not even be possible for privacy reasons. Therefore, in this paper, we aim to learn an accurate classifier without any class labels. More specifically, we consider the case where multiple sets of unlabeled data and only their class priors, i.e., the proportions of each class, are available. Under this problem setup, we first derive an unbiased estimator of the classification risk that can be estimated from the given unlabeled sets and theoretically analyze the generalization error of the learned classifier. We then find that the classifier obtained as such tends to cause overfitting as its empirical risks go negative during training. To prevent overfitting, we further propose a partial risk regularization that maintains the partial risks with respect to unlabeled datasets and classes to certain levels. Experiments demonstrate that our method effectively mitigates overfitting and outperforms state-of-the-art methods for learning from multiple unlabeled sets.