Robert Paluch presented a talk on "Fast and accurate detection of spread source in large complex networks" co-authored by Xiaoyan Lu, Krzysztof Suchecki, Boleslaw Szymanski, Janusz Holyst in Cambridge, MA in July 2018.

Robert Paluch presented a talk on "Fast and accurate detection of spread source in large complex networks" co-authored by Xiaoyan Lu, Krzysztof Suchecki, Boleslaw Szymanski, Janusz Holyst at the 9th International Conference on Complex Systems in Cambridge, MA, July 22-27, 2018. The talks focuses on spread over complex networks which is a ubiquitous process with increasingly wide applications. Locating spread sources is often important, e.g. finding the patient one in epidemics, or source of rumor spreading in social network. The talk introduces a new approach in which observers with low quality information (i.e. with large spread encounter times) are ignored and potential sources are selected based on the likelihood gradient from high quality observers. The resulting Gradient Maximum Likelihood Algorithm (GMLA) reduces this complexity and on synthetic networks and real Gnutella network with limitation that id’s of spreaders are unknown to observers demonstrate that for scale-free networks with such limitation yields high quality localization results. The paper on which the talk is based is available a the link below.