arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1108.3755math.STstat.TH
收藏

Lasso的新误差分析

New Error Analysis for Lasso

Junlong Zhao, Chenlei Leng

摘要:

Lasso是高维线性回归中参数估计和变量选择的重要方法之一。其成功的核心是有吸引力的收敛速度的结果,即使当$p$,问题的维度,是远远大于样本大小$n$。特别是,Bickel等人(2009)表明,根据$\ell_1$范数,对于稀疏指数$s$,该比率为$s\sqrt{(\log p)/n}$。本文利用模型的分布信息,得到了一个新的收敛速度界。在正态或亚高斯假设下,对于某些设计矩阵,速率可以提高到接近$s/\sqrt{n}$。我们进一步概述了一个一般的分区技术,有助于获得更清晰的收敛速度的套索。该结果适用于许多适合于高维数据分析的协方差阵。

英文摘要:

The Lasso is one of the most important approaches for parameter estimation and variable selection in high dimensional linear regression. At the heart of its success is the attractive rate of convergence result even when $p$, the dimension of the problem, is much larger than the sample size $n$. In particular, Bickel et al. (2009) showed that this rate, in terms of the $\ell_1$ norm, is of the order $s\sqrt{(\log p)/n}$ for a sparsity index $s$. In this paper, we obtain a new bound on the convergence rate by taking advantage of the distributional information of the model. Under the normality or sub-Gaussian assumption, the rate can be improved to nearly $s/\sqrt{n}$ for certain design matrices. We further outline a general partitioning technique that helps to derive sharper convergence rate for the Lasso. The result is applicable to many covariance matrices suitable for high-dimensional data analysis.

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。

↑