webinar register page

Webinar banner
Scalable computation for Bayesian hierarchical models - Omiros Papaspiliopoulos
The talk describes three MCMC frameworks for scalable Bayesian computation for two of the most canonical hierarchical model structures in applied Statistics: nested multilevel and crossed-effect models. Scalability refers to algorithmic complexity that scales linearly with the number of observations and the number of model parameters. For certain combinations of algorithm and model we establish theoretical guarantees for scalability (and for others the lack thereof) but we also provide numerical evidence for a wide range of cases beyond those the theory is applicable to. The three frameworks are built around Gibbs sampling, sparse linear algebra and belief propagation. We show numerical comparisons against both off-the-self MCMC methods, such as HMC as implemented in STAN, and mean field variational Bayes. We demonstrate the success of the frameworks in two fairly large scale real data applications, one on predicting electoral results from large surveys and another on real estate price prediction at postal code level.

Jan 27, 2021 05:00 PM in Paris

Webinar logo
* Required information

By registering, I agree to the Privacy Statement and Terms of Service.



Omiros Papaspiliopoulos
Professor @Universita' Bocconi
Omiros Papaspiliopoulos is ICREA Research Professor at UPF. He is the Scientific Director of the Barcelona GSE Master's Degree in Data Science. His research has appeared in the top journals in Statistics, including several articles in the Journal of the Royal Statistical Society Series B, Biometrika and the Annals of Statistics. He has been an Associate Editor for the first two journals and a Deputy Editor for Biometrika. He has delivered more than 80 invited talks, and has given courses at ENSAE in Paris, the Berlin Mathematical School, the Department of Mathematics at University of Copenhagen, and the Engineering Department at Osaka University. In 2010 he was awarded the Royal Statistical Society's Guy Medal in Bronze. His research interests include Monte Carlo Methods, Computational Methods, Bayesian Statistics, Stochastic Processes, Machine Learning.