Unadjusted Langevin Algorithm for Non-convex Weakly Smooth Potentials

Abstract

Discretization of continuous-time diffusion processes is a widely recognized method for sampling. However, the canonical Euler Maruyama discretization of the Langevin diffusion process, referred as unadjusted Langevin algorithm (ULA), studied mostly in the context of smooth (gradient Lipschitz) and strongly log-concave densities, is a considerable hindrance for its deployment in many sciences, including statistics and machine learning. In this paper, we establish several theoretical contributions to the literature on such sampling methods for non-convex distributions. Particularly, we introduce a new mixture weakly smooth condition, under which we prove that ULA will converge with additional log-Sobolev inequality. We also show that ULA for smoothing potential will converge in \(L_{2}\)-Wasserstein distance. Moreover, using convexification of nonconvex domain (Ma et al. in Proc Natl Acad Sci 116(42):20881–20885, 2019) in combination with regularization, we establish the convergence in Kullback–Leibler divergence with the number of iterations to reach \(\epsilon \)-neighborhood of a target distribution in only polynomial dependence on the dimension. We relax the conditions of Vempala and Wibisono (Advances in Neural Information Processing Systems, 2019) and prove convergence guarantees under isoperimetry, and non-strongly convex at infinity.