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 2602.07252stat.ME
收藏

超越欧几里得摘要:面向分布值数据的在线变点检测

Beyond Euclidean Summaries: Online Change Point Detection for Distribution-Valued Data

University of Cincinnati(辛辛那提大学);University at Buffalo(布法罗纽约州立大学)

Yingyan Zeng, Yujing Huang, Xiaoyu Chen

AI总结:

提出一种基于2-Wasserstein空间的分布值变点检测框架,通过切空间线性化将经典多元监控统计量适配到分布流形,实现对复杂分布变化的快速检测。

摘要:

现有的在线变点检测方法依赖于固定维度的欧几里得摘要,隐含假设分布变化能被基于矩或基于特征的表示很好地捕捉。它们可能掩盖分布形状或几何上的重要变化。我们提出了一种内在的分布值变点检测框架,将流式批处理数据视为2-Wasserstein空间上的随机过程。我们的方法通过将每个经验分布映射到相对于变化前Fr\'echet重心的切空间,检测该过程规律的变化,从而得到2-Wasserstein空间的参考中心局部线性化。这种表示通过将经典多元监控统计量适配到切场,使得顺序检测器得以实现。我们提供了理论保证,并通过合成和真实实验证明,与基于矩和无模型基线相比,我们的方法在匹配的$\mathrm{ARL}_0$下以更短的检测延迟检测到复杂的分布变化。代码可在https://github.com/yyzeng43/IDD-icml获取。

英文摘要:

Existing online change-point detection (CPD) methods rely on fixed-dimensional Euclidean summaries, implicitly assuming that distributional changes are well captured by moment-based or feature-based representations. They can obscure important changes in distributional shape or geometry. We propose an intrinsic distribution-valued CPD framework that treats streaming batch data as a stochastic process on the 2-Wasserstein space. Our method detects changes in the law of this process by mapping each empirical distribution to a tangent space relative to a pre-change Fréchet barycenter, yielding a reference-centered local linearization of 2-Wasserstein space. This representation enables sequential detectors by adapting classical multivariate monitoring statistics to tangent fields. We provide theoretical guarantees and demonstrate, via synthetic and real-world experiments, that our approach detects complex distributional shifts with reduced detection delay at matched $\mathrm{ARL}_0$ compared with moments-based and model-free baselines. The code is available at https://github.com/yyzeng43/IDD-icml .

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。

↑