Review for NeurIPS paper: Semi-Supervised Partial Label Learning via Confidence-Rated Margin Maximization

Summary and Contributions: This paper studies an interesting problem setting called semi-supervised partial label learning. To solve this problem, this paper adopts a two-stage method. For the first stage, label propagation is used to produce labeling confidence for partial label examples. For the second stage, a maximum margin formulation is introduced to jointly enable the induction of the predictive model and the estimation of labeling confidence over unlabeled data. Experiments have validated the effectiveness of the proposed method.