A0220
Title: Metropolis-adjusted subdifferential Langevin algorithm
Authors: Ning Ning - Texas A&M University (United States) [presenting]
Abstract: The Metropolis-adjusted Langevin algorithm (MALA) is a widely used Markov chain Monte Carlo (MCMC) method for sampling from high-dimensional distributions. However, MALA relies on differentiability assumptions that restrict its applicability. The Metropolis-adjusted subdifferential Langevin algorithm (MASLA) is introduced, a generalization of MALA that extends its applicability to distributions whose log-densities are locally Lipschitz, generally non-differentiable, and non-convex. The theoretical foundation of MASLA is established by proving its convergence to a set-valued differential inclusion equation, ensuring well-defined long-run behavior. Furthermore, the performance of MASLA is evaluated by comparing it with other sampling algorithms in settings where they are applicable. Results demonstrate the effectiveness of MASLA in handling a broader class of distributions while maintaining computational efficiency.
