arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1306.1185stat.MLcs.LGmath.OC
收藏

多类全变分聚类

Multiclass Total Variation Clustering

City University of Hong Kong(香港城市大学);University of California Riverside(加州大学河滨分校);University of San Francisco(旧金山大学);University of California Los Angeles(加州大学洛杉矶分校)

Xavier Bresson, Thomas Laurent, David Uminsky, James H. von Brecht

摘要:

最近,图像处理文献中的思想激发了一组新的聚类算法,这些算法依赖于总变差的概念。虽然这些算法在二分区任务中表现良好,但它们的递归扩展在多类聚类任务中效果不佳。本文提出了一个通用的框架,多类全变差聚类,不依赖于递归。结果大大优于以前的全变分算法和比较以及与国家的最先进的NMF方法。

英文摘要:

Ideas from the image processing literature have recently motivated a new set of clustering algorithms that rely on the concept of total variation. While these algorithms perform well for bi-partitioning tasks, their recursive extensions yield unimpressive results for multiclass clustering tasks. This paper presents a general framework for multiclass total variation clustering that does not rely on recursion. The results greatly outperform previous total variation algorithms and compare well with state-of-the-art NMF approaches.

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。

↑