Learning from Label Proportions by Learning with Label Noise
Learning from Label Proportions by Learning with Label Noise
复制标题
DOI：
10.48550/arxiv.2203.02496
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Jianxin Zhang;Yutong Wang;C. Scott
中科院分区：
文献类型：
--
作者：
Jianxin Zhang;Yutong Wang;C. Scott
从标签比例中学习（LLP）是一个弱监督分类问题，其中数据点被分组到包中，并且观察每个包中的标签比例，而不是实例级标签。任务是学习一个分类器来预测未来单个实例的单个标签。先前针对多类数据的LLP工作尚未开发出理论基础的算法。在这项工作中，我们提供了一种基于减少带有标签噪声的学习的理论基础的LLP方法，使用\citet{Patrini2017MakingDN}的前向校正（FC）损失。我们为我们的方法建立了一个超额风险界和泛化误差分析，同时也扩展了FC损失理论，这可能是一个独立的兴趣。与领先的现有方法相比，我们的方法在跨多个数据集和架构的深度学习场景中展示了改进的经验性能。
Learning from label proportions (LLP) is a weakly supervised classification problem where data points are grouped into bags, and the label proportions within each bag are observed instead of the instance-level labels. The task is to learn a classifier to predict the individual labels of future individual instances. Prior work on LLP for multi-class data has yet to develop a theoretically grounded algorithm. In this work, we provide a theoretically grounded approach to LLP based on a reduction to learning with label noise, using the forward correction (FC) loss of \citet{Patrini2017MakingDN}. We establish an excess risk bound and generalization error analysis for our approach, while also extending the theory of the FC loss which may be of independent interest. Our approach demonstrates improved empirical performance in deep learning scenarios across multiple datasets and architectures, compared to the leading existing methods.