FREDIS: A Fusion Framework of Refinement and Disambiguation for Unreliable Partial Label Learning

To reduce annotation difficulty, Partial label learning (PLL) uses ambiguous annotations with candidate labels instead of the exact correct label. It assumes the candidate label set contains the correct label, inducing disambiguation, which is commonly adopted in PLL methods. However, this assumption is impractical as no one could guarantee the existence of the correct label in the candidate label set under real-world scenarios. Therefore, Unreliable Partial Label Learning (UPLL) is investigated where the correct label of each example may not exist in the candidate label set. In this paper, we propose a fusion framework of refinement and disambiguation named FREDIS to handle the UPLL problem. Specifically, with theoretical guarantees, not only does disambiguation move incorrect labels from candidate labels to non-candidate labels but also refinement, an opposite procedure, moves correct labels from non-candidate labels to candidate labels. Besides, we prove that the classifier trained by our framework could eventually approximate the Bayes optimal classifier. Extensive experiments on widely used benchmark datasets validate the effectiveness of our proposed framework. Source code is available at https: //github.com/palm-ml/fredis .