Convergence law for hyper-graphs with prescribed degree sequences

Abstract

We view hyper-graphs as incidence graphs, i.e. bipartite graphs with a set of nodes representing vertices and a set of nodes representing hyper-edges, with two nodes being adjacent if the corresponding vertex belongs to the corresponding hyper-edge. It defines a random hyper-multigraph specified by two distributions, one for the degrees of the vertices, and one for the sizes of the hyper-edges. We develop the logical analysis of this framework and first prove a convergence law for first-order logic, then characterise the limit first-order theories defined by a wide class of degree distributions. Convergence laws of other models follow, and in particular for the classical Erd\H{o}s-R\'enyi graphs and $k$-uniform hyper-graphs.