arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.04895cs.LGstat.ML
收藏

任意形状高斯混合模型的核集

Coresets for Gaussian Mixture Models of Any Shape

University of Haifa(海法大学)

Dan Feldman, Zahi Kfir, Xuan Wu

摘要:

对于给定的$n$点集合$D$,$\varepsilon$ -核心集通常是一个小的加权集合,这样对于给定的查询族,查询核心集\emph{provably}产生原始(完整)数据集的$(1+\varepsilon)$ -因子近似.使用现有技术,可以在例如网络、GPU或云上并行地为流式传输、动态(插入/删除)和分布式数据维护核心集。我们建议的第一coresets近似的负对数似然$k$ -高斯混合模型(GMM)的任意形状(其协方差矩阵的特征值之间的比率)。例如,对于其坐标在$[-n^{100},n^{100}]$中为整数的任何输入集$D$和任何固定的$k,d\geq 1$,核心集大小为$(\log n)^{O(1)}/\varepsilon^2$,并且可以以高概率在$n$中在时间上接近线性地计算。然后,通过学习小的核心集,可以快速近似最优GMM。以前的结果[NIPS'11,JMLR'18]表明,对于半特殊单位高斯的情况,其中它们的对应特征值是在$\frac{1}{2π}$到$2π$之间的常数。我们的主要技术是一个减少之间的核心$k$ -甘精胰岛素和投影聚类问题。我们实现了我们的算法,并提供了开放的代码,和实验结果。由于我们的核心集是通用的,对GARCH没有特殊的依赖,我们希望它们对许多其他函数有用。

英文摘要:

An $\varepsilon$-coreset for a given set $D$ of $n$ points, is usually a small weighted set, such that querying the coreset \emph{provably} yields a $(1+\varepsilon)$-factor approximation to the original (full) dataset, for a given family of queries. Using existing techniques, coresets can be maintained for streaming, dynamic (insertion/deletions), and distributed data in parallel, e.g. on a network, GPU or cloud. We suggest the first coresets that approximate the negative log-likelihood for $k$-Gaussians Mixture Models (GMM) of arbitrary shapes (ratio between eigenvalues of their covariance matrices). For example, for any input set $D$ whose coordinates are integers in $[-n^{100},n^{100}]$ and any fixed $k,d\geq 1$, the coreset size is $(\log n)^{O(1)}/\varepsilon^2$, and can be computed in time near-linear in $n$, with high probability. The optimal GMM may then be approximated quickly by learning the small coreset. Previous results [NIPS'11, JMLR'18] suggested such small coresets for the case of semi-speherical unit Gaussians, i.e., where their corresponding eigenvalues are constants between $\frac{1}{2π}$ to $2π$. Our main technique is a reduction between coresets for $k$-GMMs and projective clustering problems. We implemented our algorithms, and provide open code, and experimental results. Since our coresets are generic, with no special dependency on GMMs, we hope that they will be useful for many other functions.

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。

↑