arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2101.00575cs.LGmath.STstat.TH
收藏

基于EM和梯度EM的高斯混合模型学习的收敛保证

Improved Convergence Guarantees for Learning Gaussian Mixture Models by EM and Gradient EM

Weizmann Institute of Science(魏茨曼科学研究所)

Nimrod Segol, Boaz Nadler

摘要:

本文研究了具有K个已知权重分量的高斯混合模型的参数估计问题,所有分量都具有单位协方差矩阵。我们做了两个贡献。首先,在人口水平上,我们提出了一个更尖锐的分析EM和梯度EM的局部收敛性,相比以前的作品。假设分离的$Ω(\sqrt{\log K})$,我们证明了这两种方法的收敛到全局最优值从一个初始化区域大于以前的作品。具体来说,每个分量的初始猜测可以远至(几乎)其到最近高斯的距离的一半。这基本上是最大可能的收缩区域。我们的第二个贡献是改进了EM和梯度EM精确估计的样本量要求。在以前的工作中,所需的样本数有一个二次依赖于K个组件之间的最大分离,并由此产生的误差估计线性增加与此最大分离。在这篇手稿中,我们表明,这两个数量只依赖于最大的分离几何。

英文摘要:

We consider the problem of estimating the parameters a Gaussian Mixture Model with K components of known weights, all with an identity covariance matrix. We make two contributions. First, at the population level, we present a sharper analysis of the local convergence of EM and gradient EM, compared to previous works. Assuming a separation of $Ω(\sqrt{\log K})$, we prove convergence of both methods to the global optima from an initialization region larger than those of previous works. Specifically, the initial guess of each component can be as far as (almost) half its distance to the nearest Gaussian. This is essentially the largest possible contraction region. Our second contribution are improved sample size requirements for accurate estimation by EM and gradient EM. In previous works, the required number of samples had a quadratic dependence on the maximal separation between the K components, and the resulting error estimate increased linearly with this maximal separation. In this manuscript we show that both quantities depend only logarithmically on the maximal separation.

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。

↑