High-Order Langevin Monte Carlo Algorithms
Main:2 Pages
3 Figures
1 Tables
Appendix:71 Pages
Abstract
Langevin algorithms are popular Markov chain Monte Carlo (MCMC) methods for large-scale sampling problems that often arise in data science. We propose Monte Carlo algorithms based on the discretizations of -th order Langevin dynamics for any . Our design of -th order Langevin Monte Carlo (LMC) algorithms is by combining splitting and accurate integration methods. We obtain Wasserstein convergence guarantees for sampling from distributions with log-concave and smooth densities. Specifically, the mixing time of the -th order LMC algorithm scales as for , which has a better dependence on the dimension and the accuracy level as grows. Numerical experiments illustrate the efficiency of our proposed algorithms.
View on arXivComments on this paper
