Blog

With Tore Selland Kleppe (Stavanger, Norway), released a preprint presenting 2.5/3.5-order L^2-accurate Randomized Runge-Kutta-Nyström methods to approximate the Hamiltonian flow within unadjusted Hamiltonian Monte Carlo and unadjusted kinetic Langevin chains.

Together with Katharina Schuh (TU Wien), released a preprint on a nonlinear generalization of Hamiltonian Monte Carlo for sampling from nonlinear probability measures of mean-field type.

Together with Stefan Oberdörster (Bonn), released a preprint presenting a new tool for total variation mixing time analysis of Metropolis-adjusted Markov kernels via couplings without restrictive assumptions on either the … Read More

Together with Milo Marsden (Stanford), released a preprint that constructs a randomized time integrator for Hamiltonian MCMC and analyzes the L^2-Wasserstein contractivity and asymptotic bias of the corresponding unadjusted Hamiltonian … Read More