Title:The continuum limit of the Kuramoto model on sparse directed graphs

Abstract:A system of coupled phase oscillators on a convergent family of graphs is analyzed in this work. We consider coupled systems on directed and undirected, deterministic and random, dense and sparse graphs of unbounded degree. When the size of the graph in the sequence tends to infinity, we derive the continuum limit of discrete models in the form of a nonlinear diffusion equation. The kernel of the integral operator describing the nonlocal diffusion is given by an integrable function, representing the limit of the graph sequence. We show that the solutions of the initial value problem for discrete models converge to solutions of the IVP for continuous equation on finite time intervals. For coupled systems on random graphs, we prove the averaging principle, which allows to substitute the dynamical system on a random graph by the deterministic problem on a complete weighted graph. The latter model is obtained from the original one by averaging the vector field over all possible realizations of the random graph model. Our analysis covers the Kuramoto model of coupled phase oscillators on a variety of graphs including directed and undirected (sparse) Erdos-Renyi, small-world, and power law graphs.