Accelerated High-Accuracy Sampling from a Warm Start via the Proximal Bouncy Particle Sampler

arXiv:2609.06905v1 Announce Type: cross Abstract: We study the problem of sampling from mu(dx)propto e^{-V(x)},dx on R^d, where V is alpha-strongly convex and eta-smooth, and write kappa:=eta/alpha. We design and analyze the Proximal Bouncy Particle Sampler (Proximal BPS), a new sampler that combines ideas from the proximal sampler and the bouncy particle sampler. From a warm start initialization with O(1) Renyi divergence w.r.t. mu, Proximal BPS returns a sample whose law is arepsilon-close to mu in total variation distance using widetilde O(sqrtkappa,d^{1/4} ,polylog(1/arepsilon)) gradient queries in expectation.

Source: arXiv cs.LG | 2026-09-09