arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.07921cs.LGstat.ML
收藏

使用部分标签信息的半监督对比学习

Semi-supervised Contrastive Learning Using Partial Label Information

Vanderbilt University(范德堡大学);Vanderbilt University Medical Center(范德堡大学医学中心)

Colin B. Hansen, Vishwesh Nath, Diego A. Mesa, Yuankai Huo, Bennett A. Landman, Thomas A. Lasko

AI总结:

本文提出一种名为 Nullspace Tuning 的半监督对比学习方法,利用无标签样本中的部分标签信息(即已知子集具有相同标签),通过鼓励模型对同类样本给出相同预测来优化线性模型的零空间,在多个公共数据集上使测试误差显著降低。

摘要:

在半监督学习中,来自无标签样本的信息被用来改进从有标签样本中学到的模型。在某些学习问题中,可以从原本无标签的样本中推断出部分标签信息,并利用它来进一步改进模型。具体而言,当已知训练样本的子集具有相同标签时,即使标签本身缺失,也存在部分标签信息。通过对比学习目标鼓励模型对所有此类样本给出相同的标签,我们有可能提升其性能。我们将这种鼓励称为 Nullspace Tuning,因为具有相同标签的任意一对样本之间的差向量应位于线性模型的零空间中。在本文中,我们使用一个严谨的比较框架在特征明确的公共数据集上研究了使用部分标签信息的益处。我们表明,部分标签提供的额外信息将测试误差比优秀的半监督方法降低了通常 2 倍,最佳情况下可达 5.5 倍。我们还表明,将 Nullspace Tuning 添加到较新的且最先进的 MixMatch 方法中,可使其测试误差降低高达 1.8 倍。

英文摘要:

In semi-supervised learning, information from unlabeled examples is used to improve the model learned from labeled examples. In some learning problems, partial label information can be inferred from otherwise unlabeled examples and used to further improve the model. In particular, partial label information exists when subsets of training examples are known to have the same label, even though the label itself is missing. By encouraging the model to give the same label to all such examples through contrastive learning objectives, we can potentially improve its performance. We call this encouragement Nullspace Tuning because the difference vector between any pair of examples with the same label should lie in the nullspace of a linear model. In this paper, we investigate the benefit of using partial label information using a careful comparison framework over well-characterized public datasets. We show that the additional information provided by partial labels reduces test error over good semi-supervised methods usually by a factor of 2, up to a factor of 5.5 in the best case. We also show that adding Nullspace Tuning to the newer and state-of-the-art MixMatch method decreases its test error by up to a factor of 1.8.

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。

↑