Global and partitioned reconstructions of undirected complex networks

TL;DR

This paper introduces global and partitioned optimization methods using compressed sensing to accurately reconstruct undirected complex networks from limited dynamical data, offering new perspectives on network inference.

Contribution

It proposes novel global and partitioned reconstruction approaches for undirected networks based on compressed sensing, improving accuracy with less data.

Findings

Higher reconstruction accuracy with limited data

Effective application to both homogeneous and heterogeneous networks

Provides new insights into network structure inference

Abstract

It is a significant challenge to predict the network topology from a small amount of dynamical observations. Different from the usual framework of the node-based reconstruction, two optimization approaches (i.e., the global and partitioned reconstructions) are proposed to infer the structure of undirected networks from dynamics. These approaches are applied to evolutionary games occurring on both homogeneous and heterogeneous networks via compressed sensing, which can more efficiently achieve higher reconstruction accuracy with relatively small amounts of data. Our approaches provide different perspectives on effectively reconstructing complex networks.