{"id":"arxiv/2608.04284","url":"https://arxiv.org/abs/2608.04284","tags":["stat.CO","math.PR"],"score":0,"title":"Diffeomorphic Markov Chain Monte Carlo: fast mixing for heavy-tailed distributions","domain":"stat.CO","source":"arxiv","authors":["Miha Brešar","Aleksandar Mijatović"],"reviews":0,"subject":"stat.CO","summary":"We introduce a new class of uniformly ergodic MCMC algorithms, termed Diffeomorphic Contraction Sampler (DCS), and provide fast non-asymptotic mixing guarantees for DCS targeting distributions on $\\R^d$ with arbitrarily heavy polynomial tails. DCS provides a solution to a well-known problem for MCMC samplers, which typically struggle with the combination of unbounded high-dimensional state space and vanishing gradients. The DCS pulls back a target on $\\R^d$ onto a Euclidean ball $B(R)\\subset\\R^d$ and then samples from the transformed density on the convex set $B(R)$ via algorithms such as…","category":"research","comments":0,"disputed":false,"division":"stat","mentions":0,"discussed":0,"published":"2026-08-04","reactions":{}}