TY - JOUR
T1 - Message passing on networks with loops
JF - Proceedings of the National Academy of Sciences
JO - Proc Natl Acad Sci USA
SP - 23398
LP - 23403
DO - 10.1073/pnas.1914893116
VL - 116
IS - 47
AU - Cantwell, George T.
AU - Newman, M. E. J.
Y1 - 2019/11/19
UR - http://www.pnas.org/content/116/47/23398.abstract
N2 - Message passing, a celebrated family of methods for performing calculations on networks, has led to many important results in physics, statistics, computer science, and other areas. The technique allows one to divide large network calculations into manageable pieces and hence solve them either analytically or numerically. However, the method has a substantial and widely recognized shortcoming, namely that it works poorly on networks that contain short loops. Unfortunately, most real-world networks contain many such loops, which limits the applicability of the method. In this paper we give a solution for this problem, demonstrating how message passing can be extended to any network, regardless of structure, allowing it to become a general tool for the quantitative study of network phenomena.Message passing is a fundamental technique for performing calculations on networks and graphs with applications in physics, computer science, statistics, and machine learning, including Bayesian inference, spin models, satisfiability, graph partitioning, network epidemiology, and the calculation of matrix eigenvalues. Despite its wide use, however, it has long been recognized that the method has a fundamental flaw: It works poorly on networks that contain short loops. Loops introduce correlations that can cause the method to give inaccurate answers or to fail completely in the worst cases. Unfortunately, most real-world networks contain many short loops, which limits the usefulness of the message-passing approach. In this paper we demonstrate how to rectify this shortcoming and create message-passing methods that work on any network. We give 2 example applications, one to the percolation properties of networks and the other to the calculation of the spectra of sparse matrices.
ER -