Zero Forcing on Iterated Graph Models

Abstract: Modeling how information travels throughout a network has vast applications across social sciences, cybersecurity, and graph-based neural networks. In this paper, we consider the zero forcing model for information diffusion on iterative deterministic complex network models. In particular, we continue the exploration of the Iterative Local Transitive (ILT) model and the Iterative Local Anti-Transitive (ILAT) model, both introduced by Bonato et. al. in 2009 and 2017, respectively. These models use ideas from Structural Balance Theory to generate edges through a notion of cloning where the friend of my friend is my friend'' and anticloning wherethe enemy of my enemy is my friend.'' Zero forcing, introduced independently by Burgarth and Giovanetti and a special working group at AIM in 2007 and 2008, begins with some set of forced vertices, the remaining are unforced. If a forced vertex has a single unforced neighbor, that neighbor becomes forced. The minimum number of vertices in a starting set required to guarantee all vertices eventually become forced is the zero forcing number of the graph. The maximum number of vertices in a starting set such that the graph cannot become fully forced is the failed zero forcing number of a graph. In this paper we provide bounds for both the zero forcing and failed zero forcing number of graphs created using both the ILT and ILAT models. In particular, we show that in the case of ILAT graphs the failed zero forcing number can only take on one of four values, dependent on the number of vertices of the graph. Finally, we briefly consider another information diffusion model, graph burning, on a more general iterated graph model.