arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1206.0773math.STcs.ITmath.ITstat.MLstat.TH
收藏

基于谱扫描统计的图的变点检测

Changepoint Detection over Graphs with the Spectral Scan Statistic

James Sharpnack, Alessandro Rinaldo, Aarti Singh

摘要:

我们考虑的变点检测问题的决定,噪声测量的基础上,是否一个未知的信号在一个给定的图是恒定的,而是分段恒定的两个连接的诱导子图相对较低的切割尺寸。我们分析了相应的广义似然比(GLR)统计量,并将其与图中稀疏割问题联系起来。我们开发了一个易于处理的放松的GLR统计的基础上的组合拉普拉斯的图,我们称之为谱扫描统计,并分析其性质。我们展示了它的性能作为一个测试程序直接依赖于图的频谱,并使用这个结果来明确推导出它的渐近性质的一些重要的图拓扑结构。最后,我们从理论上和模拟表明,频谱扫描统计可以优于朴素的测试程序的基础上边缘阈值和$χ^2$测试。

英文摘要:

We consider the change-point detection problem of deciding, based on noisy measurements, whether an unknown signal over a given graph is constant or is instead piecewise constant over two connected induced subgraphs of relatively low cut size. We analyze the corresponding generalized likelihood ratio (GLR) statistics and relate it to the problem of finding a sparsest cut in a graph. We develop a tractable relaxation of the GLR statistic based on the combinatorial Laplacian of the graph, which we call the spectral scan statistic, and analyze its properties. We show how its performance as a testing procedure depends directly on the spectrum of the graph, and use this result to explicitly derive its asymptotic properties on few significant graph topologies. Finally, we demonstrate both theoretically and by simulations that the spectral scan statistic can outperform naive testing procedures based on edge thresholding and $χ^2$ testing.

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。

↑