ELLIS Unit London and UCL AI Centre hold Seminar on Bayesian computation
17 March 2025 - 17 March 2025 Seminar Hybrid
17 March 2025
-
Join ELLIS Unit London and the UCL AI Centre for their Seminar on March 17, 2025 (13:00-14:00 GMT) featuring a talk by Prof. Nicolas Chopin (ENSAE, Paris) on "Saddlepoint Monte Carlo and its applications in ecological inference". The talk will explore advanced Bayesian computation techniques for handling intractable likelihoods and their impact on political and environmental data analysis. The event is hybrid.
We would like to invite you to a talk by Nicolas Chopin, Professor of Statistics at ENSAE, Paris, entitled "Saddlepoint Monte Carlo and its application to exact ecological inference". This is part of the UCL AI Centre seminar series and run jointly with ELLIS.
Date: Mon 17th March 2025
Time: 12:30 - lunch; 1-2 - talk
Format: Hybrid - in person and Teams
Location: Function Space, AI Centre, 90 High Holborn, WC1V 6LJ
Sign up: Please sign up here
and see abstract and speaker's bio below. The talk will be recorded and available on the AI Centre YouTube channel.
Abstract:
Assuming X is a random vector and A a non-invertible matrix, one sometimes need to perform inference while only having access to samples of Y = AX. The corresponding likelihood is typically intractable. One may still be able to perform exact Bayesian inference using a pseudo-marginal sampler, but this requires an unbiased estimator of the intractable likelihood.
We propose saddlepoint Monte Carlo, a method for obtaining an unbiased estimate of the density of Y with very low variance, for any model belonging to an exponential family. Our method relies on importance sampling of the characteristic function, with insights brought by the standard saddlepoint approximation scheme with exponential tilting. We show that saddlepoint Monte Carlo makes it possible to perform exact inference on particularly challenging problems and datasets. We focus on the ecological inference problem, where one observes only aggregates at a fine level. We present in particular a study of the carryover of votes between the two rounds of various French elections, using the finest available data (number of votes for each candidate in about 60,000 polling stations over most of the French territory).
We show that existing, popular approximate methods for ecological inference can lead to substantial bias, which saddlepoint Monte Carlo is immune from. We also present original results for the 2024 legislative elections on political centre-to-left and left-to-centre conversion rates when the far-right is present in the second round. Finally, we discuss other exciting applications for saddlepoint Monte Carlo, such as dealing with aggregate data in privacy or inverse problems.
Speaker bio:
Nicolas Chopin (PhD, Université Pierre et Marie Curie, Paris, 2003) has been Professor of Statistics at ENSAE, Paris, since 2006. He was previously a lecturer at Bristol University (UK).
Nicolas Chopin is a fellow of the IMS, and a current or former associate editor for Annals of Statistics, Biometrika, Journal of the Royal Statistical Society, Statistics and Computing, and Statistical Methods & Applications. He has served as a member (2013-14) and secretary (2015-16) of the research section committee of the Royal Statistical Society. He received a Savage Award for his doctoral dissertation in 2002.
His research interests include computational statistics, Bayesian inference, and probabilistic machine learning.
Nicolas Chopin is a fellow of the IMS, and a current or former associate editor for Annals of Statistics, Biometrika, Journal of the Royal Statistical Society, Statistics and Computing, and Statistical Methods & Applications. He has served as a member (2013-14) and secretary (2015-16) of the research section committee of the Royal Statistical Society. He received a Savage Award for his doctoral dissertation in 2002.
His research interests include computational statistics, Bayesian inference, and probabilistic machine learning.