dr Mateusz Wiliński, UW

Detectability of Macroscopic Structures in Directed Networks

Identifying communities in networks is one of the funding problems in Network Science. We study the limits of this task. Specifically, we characterise the impact of introducing link directions and how it depends on their asymmetry. To this end, building on the stochastic block model, we construct a class of hardly detectable directed networks. We find closed form solutions by using, statistical physics inspired, cavity method and show an existence of a phase transition, which depends on the assortativity and asymmetry of the network.