BriefGPT.xyz
Jun, 2015
聚类的表示学习:一个统计框架
Representation Learning for Clustering: A Statistical Framework
HTML
PDF
Hassan Ashtiani, Shai Ben-David
TL;DR
本文提出一种协议,将用户提供的较小的数据样本进行聚类,并在此基础上建立一个数据表示方法,通过此方法学习聚类表征,并分析其统计样本复杂度,以及线性嵌入诱导的表征类的VC维,从而可以学习成功地学习具有有限VC维的表征类。
Abstract
We address the problem of communicating domain knowledge from a user to the designer of a
clustering algorithm
. We propose a protocol in which the user provides a clustering of a relatively small random sample of a
data
→