4.2 Article

L0-regularization for high-dimensional regression with corrupted data

期刊

出版社

TAYLOR & FRANCIS INC
DOI: 10.1080/03610926.2022.2076125

关键词

Measurement errors; L-0-regularization; polynomial algorithm; nearest positive semi-definite matrix projection; model selection

资金

  1. National Natural Science Foundation of China [12101584, 72071187, 11671374, 71731010, 71921001]
  2. China Postdoctoral Science Foundation [2021TQ0326, 2021M703100]
  3. Fundamental Research Funds for the Central Universities [WK2040000047, WK3470000017, WK2040000027]
  4. Hefei Postdoctoral Research Project Funds in 2021
  5. Anhui Postdoctoral Research Project Funds in 2021

向作者/读者索取更多资源

This article discusses the widespread issue of corrupted data in many contemporary applications. It proposes a sparse modeling method based on L-0 regularization and efficiently solves the regularization problem using projection techniques. It proves the statistical properties of the proposed method under certain conditions and demonstrates its effectiveness through simulation studies.
Corrupted data appears widely in many contemporary applications including voting behavior, high-throughput sequencing and sensor networks. In this article, we consider the sparse modeling via L-0-regularization under the framework of high-dimensional measurement error models. By utilizing the techniques of the nearest positive semi-definite matrix projection, the resulting regularization problem can be efficiently solved through a polynomial algorithm. Under some interpretable conditions, we prove that the proposed estimator can enjoy comprehensive statistical properties including the model selection consistency and the oracle inequalities. In particular, the nonoptimality of the logarithmic factor of dimensionality will be showed in the oracle inequalities. We demonstrate the effectiveness of the proposed method by simulation studies.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.2
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据