Mathematics, Physics & Machine Learning Seminar   RSS

Marcelo Pereyra 11/06/2020, 17:30 — 18:30 Europe/Lisbon — Instituto Superior Técnico
, Heriot-Watt University

Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreau

This talk summarises some new developments in Bayesian statistical methodology for performing inference in high-dimensional inverse problems with an underlying convex geometry. We pay particular attention to problems related to imaging sciences and to new stochastic computation methods that tightly combine proximal convex optimisation and Markov chain Monte Carlo sampling techniques. The new computation methods are illustrated with a range of imaging experiments, where they are used to perform uncertainty quantification analyses, automatically adjust regularisation parameters, and objectively compare alternative models in the absence of ground truth.

The IST seminar series Mathematics, Physics & Machine Learning aims at bringing together mathematicians and physicists interested in machine learning (ML) with  ML and AI experts interested in mathematics and physics, with the goals of introducing innovative mathematics and physics-inspired techniques in ML and, reciprocally, applying ML to problems in mathematics and physics.

Organizers: Cláudia Nunes (DM and CEMAT), Cláudia Soares (DEEC and ISR), Francisco Melo (DEI and INESC-ID), João Seixas (DF and CEFEMA), João Xavier (DEEC and ISR), José Mourão (DM and CAMGSD), Mário Figueiredo (DEEC and IT), Pedro Alexandre Santos (DM and INESC-ID)  and Yasser Omar (DM and IT).
Zoom password: distributed with email announcements or send an email to the organizers asking for it.