arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2110.03484cs.LGstat.APstat.ML
收藏

通过弱间接监督创建训练集

Creating Training Sets via Weak Indirect Supervision

Microsoft Research Asia(微软亚洲研究院);University of Washington(华盛顿大学);University of Science and Technology of China(中国科学技术大学);Carnegie Mellon University(卡内基梅隆大学);Snorkel AI, Inc.(Snorkel AI公司)

Jieyu Zhang, Bohan Wang, Xiangchen Song, Yujing Wang, Yaming Yang, Jing Bai, Alexander Ratner

摘要:

创建标记的训练集已经成为机器学习的主要障碍之一。为了解决这个问题,最近的\emph{Weak Supervision (WS)}框架综合训练标签从多个潜在的噪声监督源。然而,现有的框架仅限于共享相同的输出空间作为目标任务的监督源。为了扩展可用源的范围,我们制定了弱间接监督(WIS),一个新的研究问题,自动合成训练标签的基础上,间接监督源,具有不同的输出标签空间。为了克服输出空间不匹配的挑战,我们开发了一种概率建模方法PLRM,它使用用户提供的标签关系来建模和利用间接监督源。此外,我们提供了一个理论上的原则测试PLRM的可扩展性看不见的标签,随着推广界。在图像和文本分类任务以及工业广告应用中,我们通过超过基线2%-9%的幅度展示了PLRM的优势。

英文摘要:

Creating labeled training sets has become one of the major roadblocks in machine learning. To address this, recent \emph{Weak Supervision (WS)} frameworks synthesize training labels from multiple potentially noisy supervision sources. However, existing frameworks are restricted to supervision sources that share the same output space as the target task. To extend the scope of usable sources, we formulate Weak Indirect Supervision (WIS), a new research problem for automatically synthesizing training labels based on indirect supervision sources that have different output label spaces. To overcome the challenge of mismatched output spaces, we develop a probabilistic modeling approach, PLRM, which uses user-provided label relations to model and leverage indirect supervision sources. Moreover, we provide a theoretically-principled test of the distinguishability of PLRM for unseen labels, along with a generalization bound. On both image and text classification tasks as well as an industrial advertising application, we demonstrate the advantages of PLRM by outperforming baselines by a margin of 2%-9%.

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。

↑