Please register for this event under the link provided on the right. We will send the link to registered attendees 1 hour before the event starts.
Please note that this talk is part of the departmental colloquium series, featuring 2 talks, each 30 minutes long, between 2:00pm-3:00pm. The two talks may be swapped, so it is possible that this talk will only start at 2:30pm.
ABSTRACT / Stochastic Block Models (SBMs) are state-of-the-art methods for modeling network structure. However, they contain several simplifying assumptions that are not likely to be valid in a variety of empirical settings. Currently, neither the extent of these potential discrepancies in empirical data nor the consequences that SBM modeling inconsistencies can introduce to analyses are well understood. In this talk, we will present some advances in large-scale assessment of SBMs via posterior predictive checks. In particular, we will show how to conduct this kind of studies, which challenges we face, and which patterns arise when testing the capacity of these models in reproducing structural descriptors of empirical networks drawn from several domains.
BIO / Felipe is interested in the foundations of network science and data analysis. His current research focuses the large-scale study of a kind of models of network structure, namely Stochastic Block Models (SBMs), in realistic settings.