Learning with Proper Partial Labels
This work addresses the challenge of weakly-supervised learning with inexact labels for machine learning practitioners, offering a more general and theoretically sound approach compared to existing methods.
The paper tackles the problem of partial-label learning, where training examples have sets of candidate labels instead of a single true label, by proposing a proper partial-label learning framework that requires weaker distributional assumptions and includes previous settings as special cases, resulting in a unified unbiased estimator with proven risk consistency and error bounds validated through experiments.
Partial-label learning is a kind of weakly-supervised learning with inexact labels, where for each training example, we are given a set of candidate labels instead of only one true label. Recently, various approaches on partial-label learning have been proposed under different generation models of candidate label sets. However, these methods require relatively strong distributional assumptions on the generation models. When the assumptions do not hold, the performance of the methods is not guaranteed theoretically. In this paper, we propose the notion of properness on partial labels. We show that this proper partial-label learning framework requires a weaker distributional assumption and includes many previous partial-label learning settings as special cases. We then derive a unified unbiased estimator of the classification risk. We prove that our estimator is risk-consistent, and we also establish an estimation error bound. Finally, we validate the effectiveness of our algorithm through experiments.