Unadjusted Langevin algorithm for sampling a mixture of weakly smooth potentials

Abstract

Discretization of continuous-time diffusion processes is a widely recognized method for sampling. However, it seems to be a considerable restriction when the potentials are often required to be smooth (gradient Lipschitz). This paper studies the problem of sampling through Euler discretization, where the potential function is assumed to be a mixture of weakly smooth distributions and satisfies weakly dissipative. We establish the convergence in Kullback–Leibler (KL) divergence with the number of iterations to reach ϵ-neighborhood of a target distribution in only polynomial dependence on the dimension. We relax the degenerated convex at infinity conditions of (Erdogdu and Hosseinzadeh (2020)) and prove convergence guarantees under Poincaré inequality or non-strongly convex outside the ball. In addition, we also provide convergence in ${\mathit{L}}_{\mathit{\beta }}$-Wasserstein metric for the smoothing potential.