Convergence in total variation for the kinetic Langevin algorithm
We prove non-asymptotic total variation estimates for the kinetic Langevin algorithm in high dimension when the target measure satisfies a Poincaré inequality and has gradient Lipschitz potential. The main point is that the estimate improves significantly upon the corresponding bound for the non-kin...
Saved in:
| Published in: | Mathematical statistics and learning (Online) Vol. 8; no. 1; pp. 71 - 104 |
|---|---|
| Main Author: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
21.08.2025
|
| ISSN: | 2520-2316, 2520-2324 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | We prove non-asymptotic total variation estimates for the kinetic Langevin algorithm in high dimension when the target measure satisfies a Poincaré inequality and has gradient Lipschitz potential. The main point is that the estimate improves significantly upon the corresponding bound for the non-kinetic version of the algorithm, due to Dalalyan. In particular, the dimension dependence drops from O(n) to O(\sqrt{n}) . |
|---|---|
| ISSN: | 2520-2316 2520-2324 |
| DOI: | 10.4171/msl/49 |