Matrix Completion With Covariate Information

Matrix Completion With Covariate Information
复制标题

使用协变量信息进行矩阵补全

DOI:
10.1080/01621459.2017.1389740
复制
发表时间:
2018-06
影响因子:
3.7
通讯作者:
Wong Raymond K W
Wong Raymond K W
中科院分区:
数学1区
文献类型:
--
作者:
Mao Xiaojun;Chen Song Xi;Wong Raymond K W

文献摘要

参考文献

被引文献

相似文献

摘要本文研究了当额外的协变量可用时,从损坏的数据中完成矩阵的问题。尽管在矩阵补全文献中很少考虑,但这些协变量通常为完成高维目标矩阵A0的未观察条目提供有价值的信息。给定协变量矩阵X,其行表示A0的行协变量,我们考虑列空间分解模型A0 = Xβ0 + B0,其中β0是系数矩阵,B0是与X正交的低秩矩阵。列空间。该模型有助于明确区分可解释协变量效应(Xβ0)和灵活隐藏因子效应(B0)。此外,我们的工作允许观察的概率依赖于协变量矩阵,因此,随机丢失的机制是允许的。我们提出了一种新的惩罚估计A0利用Frobenius范数和核范数正则化与一个有效的和可扩展的算法。研究了估计量的渐近收敛速度。通过数值实验和真实的数据应用说明了所提出的方法的经验性能。
ABSTRACT This article investigates the problem of matrix completion from the corrupted data, when the additional covariates are available. Despite being seldomly considered in the matrix completion literature, these covariates often provide valuable information for completing the unobserved entries of the high-dimensional target matrix A0. Given a covariate matrix X with its rows representing the row covariates of A0, we consider a column-space-decomposition model A0 = Xβ0 + B0, where β0 is a coefficient matrix and B0 is a low-rank matrix orthogonal to X in terms of column space. This model facilitates a clear separation between the interpretable covariate effects (Xβ0) and the flexible hidden factor effects (B0). Besides, our work allows the probabilities of observation to depend on the covariate matrix, and hence a missing-at-random mechanism is permitted. We propose a novel penalized estimator for A0 by utilizing both Frobenius-norm and nuclear-norm regularizations with an efficient and scalable algorithm. Asymptotic convergence rates of the proposed estimators are studied. The empirical performance of the proposed methodology is illustrated via both numerical experiments and a real data application.
DOI: 10.5555/1953048.2185803
发表时间: 2009-10
期刊: ArXiv
影响因子: --
作者:
B. Recht
通讯作者: B. Recht
DOI: 10.5555/1756006.1859920
发表时间: 2009-06
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Raghunandan H. Keshavan;A. Montanari;Sewoong Oh
通讯作者: Raghunandan H. Keshavan;A. Montanari;Sewoong Oh
DOI: 10.1017/cbo9781139165495.014
发表时间: 2005-08
影响因子: 6.1
作者:
D. Freedman
通讯作者: D. Freedman
DOI: 10.1145/2184319.2184343
发表时间: 2012-06-01
影响因子: 22.7
作者:
Candes, Emmanuel;Recht, Benjamin
通讯作者: Recht, Benjamin
DOI: --
发表时间: 2012-11
期刊: --
影响因子: --
作者:
Tingni Sun;Cun-Hui Zhang
通讯作者: Tingni Sun;Cun-Hui Zhang