Leveraged Weighted Loss for Partial Label Learning
This work addresses the lack of theoretical understanding and empirical guidance in partial label learning, a weakly supervised learning domain, though it appears incremental as it builds on existing methodologies.
The paper tackles the problem of partial label learning by proposing a new family of loss functions called Leveraged Weighted (LW) loss, which introduces a leverage parameter to balance losses on partial and non-partial labels, and demonstrates its effectiveness with improved performance on benchmark and real datasets compared to state-of-the-art methods.
As an important branch of weakly supervised learning, partial label learning deals with data where each instance is assigned with a set of candidate labels, whereas only one of them is true. Despite many methodology studies on learning from partial labels, there still lacks theoretical understandings of their risk consistent properties under relatively weak assumptions, especially on the link between theoretical results and the empirical choice of parameters. In this paper, we propose a family of loss functions named \textit{Leveraged Weighted} (LW) loss, which for the first time introduces the leverage parameter $β$ to consider the trade-off between losses on partial labels and non-partial ones. From the theoretical side, we derive a generalized result of risk consistency for the LW loss in learning from partial labels, based on which we provide guidance to the choice of the leverage parameter $β$. In experiments, we verify the theoretical guidance, and show the high effectiveness of our proposed LW loss on both benchmark and real datasets compared with other state-of-the-art partial label learning algorithms.