Generative Network Automata: A Generalized Framework for Modeling Adaptive Network Dynamics Using Graph Rewritings

摘要: A variety of modeling frameworks have been proposed and utilized in complex systems studies, including dynamical models that describe state transitions on a system fixed topology, self-organizing network topological transformations with little attention paid to changes. Earlier typically assumed are caused by exogenous factors, such as preferential attachment new nodes stochastic or targeted removal existing nodes. However, many real-world exhibit both transition topology transformation simultaneously, they evolve largely autonomously based the system’s own states topologies. Here we show that, using concept graph rewriting, autonomous can be seamlessly integrated represented unified computational framework. We call this novel framework “Generative Network Automata (GNA)”. In chapter, introduce basic concepts GNA, its working definition, generality represent other models, some our latest results extensive experiments exhaustively swept over possible rewriting rules simple binary-state GNA. The revealed several distinct types GNA dynamics.