arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2302.07540stat.ML
收藏

标签在半监督学习中有信息吗?--估计和利用缺失数据机制

Are labels informative in semi-supervised learning? -- Estimating and leveraging the missing-data mechanism

Aude Sportisse, Hugo Schmutz, Olivier Humbert, Charles Bouveyron, Pierre-Alexandre Mattei

摘要:

半监督学习是一种利用未标记数据来改进机器学习模型的强大技术,但它可能会受到“信息”标签的影响,当某些类比其他类更有可能被标记时,就会出现这种情况。在缺失数据文献中,此类标签称为非随机缺失。在本文中,我们提出了一种新的方法来解决这个问题,估计的缺失数据机制,并使用逆倾向加权去偏任何SSL算法,包括那些使用数据增强。我们还提出了一个似然比检验,以评估是否标签确实是信息。最后,我们证明了所提出的方法在不同的数据集上的性能,特别是在两个医疗数据集上,我们设计了伪现实的缺失数据场景。

英文摘要:

Semi-supervised learning is a powerful technique for leveraging unlabeled data to improve machine learning models, but it can be affected by the presence of ``informative'' labels, which occur when some classes are more likely to be labeled than others. In the missing data literature, such labels are called missing not at random. In this paper, we propose a novel approach to address this issue by estimating the missing-data mechanism and using inverse propensity weighting to debias any SSL algorithm, including those using data augmentation. We also propose a likelihood ratio test to assess whether or not labels are indeed informative. Finally, we demonstrate the performance of the proposed methods on different datasets, in particular on two medical datasets for which we design pseudo-realistic missing data scenarios.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑