Partial Multi-label Learning (PML) is a type of weakly supervised learning where each training instance corresponds to a set of candidate labels, among which only some are true. In this paper, we introduce \our{}, a novel probabilistic approach to this problem that extends the binary cross entropy to the PML setup. In contrast to existing methods, it does not require suboptimal disambiguation and, as such, can be applied to any deep architecture. Furthermore, experiments conducted on artificial and real-world datasets indicate that \our{} outperforms existing approaches, especially for high noise in a candidate set.

翻译：[translated abstract in Chinese] 部分多标签学习（PML）是一种弱监督学习范式，其中每个训练实例对应一组候选标签，但仅有部分标签为真实标签。本文提出了一种新颖的概率方法\our{}，将二元交叉熵扩展至PML场景。与现有方法相比，该方法无需执行次优的去歧义处理，因此可直接适用于任意深度架构。此外，在人工数据集与真实数据集上的实验表明，\our{}在候选标签集噪声较高的场景下尤其优于现有方法。