Emergence of a spectral gap in a class of random matrices associated with split graphs

Abstract

Motivated by the intriguing behavior displayed in a dynamic network that models a population of extreme introverts and extroverts (XIE), we consider the spectral properties of ensembles of random split graph adjacency matrices. We discover that, in general, a gap emerges in the bulk spectrum between -1 and 0 that contains a single eigenvalue. An analytic expression for the bulk distribution is derived and verified with numerical analysis. We also examine their relation to chiral ensembles, which are associated with bipartite graphs.

17 pages, 6 figures

References in corpus (6)

• Synchronization in complex networks
• Graph spectra and the detectability of community structure in networks
• Cavity Approach to the Spectral Density of Sparse Symmetric Random Matrices
• Eigenvalue Separation in Some Random Matrix Models
• Extraordinary variability and sharp transitions in a maximally frustrated dynamic network
• Extreme Thouless effect in a minimal model of dynamic social networks