This talk will be presented by Mateusz Wilinski (Los Alamos National Laboratory) on November 19 at 3pm via Zoom.
To join the online talk, click here for the Zoom link.
Title: Learning network structure from noisy and partially unobserved spreading dynamics
Spreading processes play an increasingly important role in modeling for diffusion networks, information propagation, marketing and opinion setting. We address the problem of learning of a spreading model – including the spreading network structure – such that the predictions generated from this model are accurate and could be subsequently used for the optimization, and control of diffusion dynamics. We focus on a challenging setting where full observations of the dynamics are not available or noisy, and standard approaches such as maximum likelihood quickly become intractable for large network instances. We introduce a computationally efficient algorithm, based on a scalable dynamic message-passing approach, which is able to learn both parameters of the effective spreading model and network structure, given only limited information on the activation times of nodes in the network. The popular Independent Cascade model is used to illustrate our approach. We show that tractable inference from the learned model generates a better prediction of marginal probabilities compared to the original model. We develop a systematic procedure for learning a mixture of models, which further improves the prediction quality.
Mateusz obtained his PhD in Physics from the University of Warsaw. He later worked as a Postdoc in the Quantitative Finance Research Group of Scuola Normale Superiore di Pisa. Currently, he is a Postdoc in Plasma Physics and Applied Mathematics Group at the Los Alamos National Laboratory. He has broad scientific interests, which include subjects from statistical physics, network science and machine learning. More specifically, he is interested in belief propagation techniques for network problems, epidemic spreading modelling, financial networks etc.