Séminaire de Probabilités et Statistique :

Le 26 septembre 2022 à 13:45 - UM - Bât 09 - Salle de conférence (1er étage)

Présentée par Michel Manon - LMBP, Université Clermont - Auvergne, CNRS

Reshaping MCMC Sampling with piecewise deterministic Markov processes

During this talk, I will discuss the main concepts and ideas that have been underlying and pushing the development of MCMC sampling through piecewise deterministic Markov processes (PDMP). First, I will introduce PDMP, which are composed of a random sequence of ballistic trajectories separated by Poisson jumps, and their associated formalism. I will then present the evolution and generalisation of PDMP-based MCMC methods since their first implementations in multiparticle systems. The general progress line can be framed into a constant effort of replacing the usually-enforced time reversibility, as done in detailed-balanced schemes, by symmetries of the sampled probability distribution itself. I will finally discuss the question of ergodicity of PDMP in multiparticle systems.

Séminaire en salle 109, pas de retransmission sur zoom.