arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.05627stat.MLcs.LG
收藏

谱聚类的性能保证

A Performance Guarantee for Spectral Clustering

University of California Los Angeles(加利福尼亚大学洛杉矶分校);University of California Davis(加利福尼亚大学戴维斯分校);University of California at Davis(加利福尼亚大学戴维斯分校)

March Boedihardjo, Shaofeng Deng, Thomas Strohmer

摘要:

两步谱聚类方法是一种广泛使用的图划分方法,它由Laplacian特征映射和舍入步骤组成。它可以被看作是对NP难最小比率切割问题的一种自然松弛。在本文中,我们研究的中心问题:当是谱聚类能够找到全球解决方案的最小比率削减问题?首先,我们提供了一个条件,自然取决于一个给定的分区的集群内和集群间的连通性,我们可以证明,这个分区是最小比率削减问题的解决方案。然后,我们开发了一个确定的2到无穷大范数扰动界的不变子空间的图拉普拉斯对应的$k$最小特征值。最后结合这两个结果,给出了保证谱聚类输出最小比割问题全局解的条件,作为谱聚类性能的保证。

英文摘要:

The two-step spectral clustering method, which consists of the Laplacian eigenmap and a rounding step, is a widely used method for graph partitioning. It can be seen as a natural relaxation to the NP-hard minimum ratio cut problem. In this paper we study the central question: when is spectral clustering able to find the global solution to the minimum ratio cut problem? First we provide a condition that naturally depends on the intra- and inter-cluster connectivities of a given partition under which we may certify that this partition is the solution to the minimum ratio cut problem. Then we develop a deterministic two-to-infinity norm perturbation bound for the the invariant subspace of the graph Laplacian that corresponds to the $k$ smallest eigenvalues. Finally by combining these two results we give a condition under which spectral clustering is guaranteed to output the global solution to the minimum ratio cut problem, which serves as a performance guarantee for spectral clustering.

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。

↑