A fast iterative algorithm for high-dimensional differential network

Zhou Tang, Zhangsheng Yu, Cheng Wang

Research output: Contribution to journalArticle

Abstract

A differential network is an important tool for capturing the changes in conditional correlations under two sample cases. In this paper, we introduce a fast iterative algorithm to recover the differential network for high-dimensional data. The computational complexity of our algorithm is linear in the sample size and the number of parameters, which is optimal in that it is of the same order as computing two sample covariance matrices. The proposed method is appealing for high-dimensional data with a small sample size. The experiments on simulated and real datasets show that the proposed algorithm outperforms other existing methods.

Original languageEnglish (US)
JournalComputational Statistics
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

High-dimensional Data
Iterative Algorithm
Fast Algorithm
High-dimensional
Sample Covariance Matrix
Small Sample Size
Computational Complexity
Sample Size
Covariance matrix
Computing
Computational complexity
Experiment
Experiments
Sample size

Keywords

  • ADMM
  • Differential network
  • Gaussian graphical model
  • High-dimensional data
  • Precision matrix

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Computational Mathematics

Cite this

A fast iterative algorithm for high-dimensional differential network. / Tang, Zhou; Yu, Zhangsheng; Wang, Cheng.

In: Computational Statistics, 01.01.2019.

Research output: Contribution to journalArticle

@article{8321efc1da244d35a5e9616b8a6684d3,
title = "A fast iterative algorithm for high-dimensional differential network",
abstract = "A differential network is an important tool for capturing the changes in conditional correlations under two sample cases. In this paper, we introduce a fast iterative algorithm to recover the differential network for high-dimensional data. The computational complexity of our algorithm is linear in the sample size and the number of parameters, which is optimal in that it is of the same order as computing two sample covariance matrices. The proposed method is appealing for high-dimensional data with a small sample size. The experiments on simulated and real datasets show that the proposed algorithm outperforms other existing methods.",
keywords = "ADMM, Differential network, Gaussian graphical model, High-dimensional data, Precision matrix",
author = "Zhou Tang and Zhangsheng Yu and Cheng Wang",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s00180-019-00915-w",
language = "English (US)",
journal = "Computational Statistics",
issn = "0943-4062",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - A fast iterative algorithm for high-dimensional differential network

AU - Tang, Zhou

AU - Yu, Zhangsheng

AU - Wang, Cheng

PY - 2019/1/1

Y1 - 2019/1/1

N2 - A differential network is an important tool for capturing the changes in conditional correlations under two sample cases. In this paper, we introduce a fast iterative algorithm to recover the differential network for high-dimensional data. The computational complexity of our algorithm is linear in the sample size and the number of parameters, which is optimal in that it is of the same order as computing two sample covariance matrices. The proposed method is appealing for high-dimensional data with a small sample size. The experiments on simulated and real datasets show that the proposed algorithm outperforms other existing methods.

AB - A differential network is an important tool for capturing the changes in conditional correlations under two sample cases. In this paper, we introduce a fast iterative algorithm to recover the differential network for high-dimensional data. The computational complexity of our algorithm is linear in the sample size and the number of parameters, which is optimal in that it is of the same order as computing two sample covariance matrices. The proposed method is appealing for high-dimensional data with a small sample size. The experiments on simulated and real datasets show that the proposed algorithm outperforms other existing methods.

KW - ADMM

KW - Differential network

KW - Gaussian graphical model

KW - High-dimensional data

KW - Precision matrix

UR - http://www.scopus.com/inward/record.url?scp=85071079109&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071079109&partnerID=8YFLogxK

U2 - 10.1007/s00180-019-00915-w

DO - 10.1007/s00180-019-00915-w

M3 - Article

AN - SCOPUS:85071079109

JO - Computational Statistics

JF - Computational Statistics

SN - 0943-4062

ER -