Quantitative convergence bounds for unadjusted kinetic Langevin and Hamiltonian Monte Carlo

Add to your list(s) Download to your calendar using vCal

• Pierre Monmarché (Sorbonne Université)
• Thursday 28 November 2024, 11:25-12:15
• Seminar Room 1, Newton Institute.

If you have a question about this talk, please contact nobody.

SSDW04 - Monte Carlo sampling: beyond the diffusive regime

Splitting schemes for the Hamiltonian and (underdamped) Langevin dynamics, which are kinetic (possibly non-reversible) processes, are widely used in Markov Chain Monte Carlo methods. We will discuss how, for Gaussian distributions, a very precise optimization of the parameters can be conducted, revealing how inertia induces a diffusive-to-ballistic speed-up for ill-conditioned targets. Motivated by this, we will present non-asymptotic efficiency bounds for this family of MCMC samplers that cover non-convex potentials and mean-field interacting particles.

This talk is part of the Isaac Newton Institute Seminar Series series.

This talk is included in these lists:

• All CMS events
• Featured lists
• INI info aggregator
• Isaac Newton Institute Seminar Series
• School of Physical Sciences
• Seminar Room 1, Newton Institute
• bld31

Note that ex-directory lists are not shown.