Zeroth-Order Non-Log-Concave Sampling with Variance Reduction and Applications to Inverse Problems

ArXi:2605.30573v1 Announce Type: new Sampling from high-dimensional, non-log-concave distributions with unnormalized densities remains a fundamental challenge in machine learning, particularly in black-box settings where gradient information is inaccessible or computationally prohibitive. While Langevin dynamics provides a principled framework for sampling when gradients are accessible, its extension to the black-box settings suffers from high variance and lacks non-asymptotic convergence guarantees for non-log-concave sampling.