Dai, Hongsheng and Fan, Xiequan and Lu, Jianya (2026) Self-normalized Cramer-type Moderate Deviation of Stochastic Gradient Langevin Dynamics. Journal of Applied Probability. pp. 1-28. DOI https://doi.org/10.1017/jpr.2025.10065
Dai, Hongsheng and Fan, Xiequan and Lu, Jianya (2026) Self-normalized Cramer-type Moderate Deviation of Stochastic Gradient Langevin Dynamics. Journal of Applied Probability. pp. 1-28. DOI https://doi.org/10.1017/jpr.2025.10065
Dai, Hongsheng and Fan, Xiequan and Lu, Jianya (2026) Self-normalized Cramer-type Moderate Deviation of Stochastic Gradient Langevin Dynamics. Journal of Applied Probability. pp. 1-28. DOI https://doi.org/10.1017/jpr.2025.10065
Abstract
In this paper, we study the self-normalized Cramér-type moderate deviation of the empirical measure of the stochastic gradient Langevin dynamics (SGLD). Consequently, we also derive the Berry–Esseen bound for the SGLD. Our approach is by constructing a stochastic differential equation to approximate the SGLD and then applying Stein’s method to decompose the empirical measure into a martingale difference series sum and a negligible remainder term.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | Self-normalized Cramér-type moderate deviation; stochastic gradient Langevin dynamics; Stein’s method; Diffusion approximation; Berry–Esseen bound |
| Divisions: | Faculty of Science and Health Faculty of Science and Health > Mathematics, Statistics and Actuarial Science, School of |
| SWORD Depositor: | Unnamed user with email elements@essex.ac.uk |
| Depositing User: | Unnamed user with email elements@essex.ac.uk |
| Date Deposited: | 13 Apr 2026 14:27 |
| Last Modified: | 13 Apr 2026 14:28 |
| URI: | http://repository.essex.ac.uk/id/eprint/42187 |
Available files
Filename: div-class-title-self-normalized-cramer-type-moderate-deviation-of-stochastic-gradient-langevin-dynamics-div.pdf
Licence: Creative Commons: Attribution 4.0