Skip to content

Exact simulation of diffusions and improved algorithms for log-concave sampling

Fan Chen, Sinho Chewi, Alexander Rakhlin, Matthew S. Zhang

cs.DSarXiv:2608.05022

Abstract

We study exact simulation of diffusions via rejection sampling on path space using unbiased estimators of the density ratio obtained from Girsanov's theorem. When applied to the underdamped Langevin diffusion, it yields an algorithm for sampling from a strongly log-concave and log-smooth distribution with condition number κ, in dimension d, to accuracy in Rényi divergence, in O(κ2/3 d1/3\,polylog(1/)) queries. Under a third derivative bound, the dimension dependence improves to d1/5. This improves substantially over the prior state-of-the-art complexity of O(κd1/2\,polylog(1/)) for the Metropolis-adjusted Langevin algorithm, and over the d1/4 dimension dependence of Metropolized Hamiltonian Monte Carlo under the same third derivative bound. We also present applications to the mirror Langevin diffusion, and for obtaining Fisher information bounds in the non-log-concave case.

Create a lesson