Universal mean-field framework for SIS epidemics on networks, based on graph partitioning and the isoperimetric inequality

Abstract

We propose a new approximation framework that unifies and generalizes a number of existing mean-field approximation methods for the SIS epidemic model on complex networks. We derive the framework, which we call the Universal Mean-Field Framework (UMFF), as a set of approximations of the exact Markovian SIS equations. Our main novelty is that we describe the mean-field approximations from the perspective of the isoperimetric problem, an insight which results in bounds on the UMFF approximation error. These new bounds provide insight in the accuracy of existing mean-field methods, such as the widely-used N-Intertwined Mean-Field Approximation (NIMFA) and Heterogeneous Mean-Field method (HMF). Additionally, the geometric perspective of the isoperimetric problem enables the UMFF approximation accuracy to be related to the regularity notions of Szemer\'edi's regularity lemma, which yields a prediction about the behavior of the SIS process on large graphs.