Multi-class Classification from Multiple Unlabeled Datasets with Partial Risk Regularization
This addresses the challenge of costly or privacy-restricted labeling in supervised learning, offering a novel approach for training classifiers from unlabeled data, though it is incremental in building upon existing risk estimation techniques.
The paper tackles the problem of learning a multi-class classifier without any labeled data by using multiple unlabeled datasets with known class priors, proposing a partial risk regularization method that mitigates overfitting and outperforms state-of-the-art methods in experiments.
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.