arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.04000cs.LG
收藏

学习Sinkhorn分歧用于有监督的变点检测

Learning Sinkhorn divergences for supervised change point detection

Georgia Tech(佐治亚理工学院);Washington University in St. Louis(圣路易斯华盛顿大学)

Nauman Ahad, Eva L. Dyer, Keith B. Hengen, Yao Xie, Mark A. Davenport

摘要:

许多现代应用程序需要检测复杂的连续数据中的变化点。大多数现有的变点检测方法都是无监督的,因此,缺乏有关我们想要检测什么类型的变化或某些类型的变化是否可以安全忽略的任何信息。这通常会导致较差的变化检测性能。我们提出了一种新的变点检测框架,使用真正的变点实例作为监督学习的地面指标,使Sinkhorn分歧,然后可以使用在滑动窗口上的双样本测试,以在线方式检测变点。我们的方法可以用来学习一个稀疏的度量,这可以是有用的特征选择和解释在高维变化点检测设置。模拟以及现实世界的序列上的实验表明,我们提出的方法可以大大提高变化点检测性能比现有的无监督变化点检测方法,只使用少数标记的变化点实例。

英文摘要:

Many modern applications require detecting change points in complex sequential data. Most existing methods for change point detection are unsupervised and, as a consequence, lack any information regarding what kind of changes we want to detect or if some kinds of changes are safe to ignore. This often results in poor change detection performance. We present a novel change point detection framework that uses true change point instances as supervision for learning a ground metric such that Sinkhorn divergences can be then used in two-sample tests on sliding windows to detect change points in an online manner. Our method can be used to learn a sparse metric which can be useful for both feature selection and interpretation in high-dimensional change point detection settings. Experiments on simulated as well as real world sequences show that our proposed method can substantially improve change point detection performance over existing unsupervised change point detection methods using only few labeled change point instances.

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。

↑