Successful network inference from time-series data using Mutual Information Rate

TL;DR

This paper introduces a novel information-theoretic method using Mutual Information Rate to accurately infer network connectivity from noisy, heterogeneous, and multi-scale time-series data, outperforming traditional mutual information approaches.

Contribution

The work derives an analytical expression for MIR applicable to finite, noisy time-series and demonstrates its effectiveness in inferring complex network structures under various challenging conditions.

Findings

Successfully infers connectivity in heterogeneous networks

Effective with different time-series lengths and noise levels

Outperforms mutual information in multi-scale networks

Abstract

This work uses an information-based methodology to infer the connectivity of complex systems from observed time-series data. We first derive analytically an expression for the Mutual Information Rate (MIR), namely, the amount of information exchanged per unit of time, that can be used to estimate the MIR between two finite-length low-resolution noisy time-series, and then apply it after a proper normalization for the identification of the connectivity structure of small networks of interacting dynamical systems. In particular, we show that our methodology successfully infers the connectivity for heterogeneous networks, different time-series lengths or coupling strengths, and even in the presence of additive noise. Finally, we show that our methodology based on MIR successfully infers the connectivity of networks composed of nodes with different time-scale dynamics, where inference based on Mutual Information fails.