A Fast Iterative Algorithm for High-dimensional Differential Network

TL;DR

This paper introduces a fast, linear-complexity iterative algorithm for recovering differential networks in high-dimensional data, effectively capturing changes in conditional correlations with superior performance on simulated and real datasets.

Contribution

The paper presents a novel, efficient algorithm with linear complexity for differential network estimation in high-dimensional settings, outperforming existing methods.

Findings

Algorithm has linear complexity in sample size and parameters

Outperforms existing methods on simulated data

Effective on real high-dimensional datasets

Abstract

Differential network is an important tool to capture the changes of 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 computation complexity of our algorithm is linear in the sample size and the number of parameters, which is optimal in the sense 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 data sets show that the proposed algorithm outperforms other existing methods.