Title: How to model very large graphs
Speaker: Professor Santosh S. Vempala, GeorgiaTech Institute of Technology
Abstract:
A wide variety of complex networks (social, biological, information etc.) exhibit local clustering with substantial variation in the clustering coefficient (the probability of neighbors being connected). Existing models of large graphs capture power-law degree distributions and small-world properties, but only limited clustering behavior. We introduce a generalization of the classical Erdõs-Rényi model of random graphs which provably achieves a wide range of desired clustering coefficient, triangle-to-edge and four-cycle-to-edge ratios for any given graph size and edge density. Rather than choosing edges independently at random, in the Random Overlapping Communities model, a graph is generated by choosing a set of random, relatively dense subgraphs ("communities"). We discuss the explanatory power of the model and some of its consequences, including convergence of a sequence of sparse graphs in terms of moments of the graph's spectrum (equivalent to the numbers of closed k-walks) appropriately normalized; such properties provably cannot be captured by popular stochastic block models. We will use the connectome as a running example. This is joint work with Samantha Petti.