Mirror Langevin Monte Carlo: the Case Under IsoperimetryDownload PDF

May 21, 2021 (edited Oct 25, 2021)NeurIPS 2021 PosterReaders: Everyone
  • Keywords: Langevin Dynamics, Mirror Descent, Discretization of SDE
  • TL;DR: We give non-asymptotic convergence guarantee for several discretizations of Mirror-Langevin SDE under weak smoothness/convexity conditions.
  • Abstract: Motivated by the connection between sampling and optimization, we study a mirror descent analogue of Langevin dynamics and analyze three different discretization schemes, giving nonasymptotic convergence rate under functional inequalities such as Log-Sobolev in the corresponding metric. Compared to the Euclidean setting, the result reveals intricate relationship between the underlying geometry and the target distribution and suggests that care might need to be taken in order for the discretized algorithm to achieve vanishing bias with diminishing stepsize for sampling from potentials under weaker smoothness/convexity regularity conditions.
  • Supplementary Material: pdf
  • Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.
  • Code: zip
11 Replies