Metraipolis › archive › dc_c52947f03468
On the computational cost of Stochastic Gradient Langevin Dynamics — arXiv:2609.17750
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.17750
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T15:40:06.588Z
- SHA-256 of the archived text
e156055d885747d6bf7842d9f8670a54afa79dc9513cacee04f3bbe22b42b054- Stable URL of this page
https://metraipolis.com/documento/dc_c52947f03468
How to cite
On the computational cost of Stochastic Gradient Langevin Dynamics — arXiv:2609.17750 — arXiv (en), https://arxiv.org/abs/2609.17750 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c52947f03468 (Metraipolis archive dc_c52947f03468, sha256 e156055d885747d6bf7842d9f8670a54afa79dc9513cacee04f3bbe22b42b054)
BibTeX
@misc{dcc52947f03468,
title = {On the computational cost of Stochastic Gradient Langevin Dynamics — arXiv:2609.17750},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.17750},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c52947f03468; sha256 e156055d885747d6bf7842d9f8670a54afa79dc9513cacee04f3bbe22b42b054},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17750. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Stochastic Gradient Langevin Dynamics (SGLD) reduces the cost of Langevin-based sampling by replacing full-dataset drift evaluations with mini-batch approximations, but the resulting subsampling error may offset this computational saving. We study this trade-off for stochastic differential equations with finite-sum drifts and compare the computational cost of SGLD with that of the Euler-Maruyama (EM) method. For a prescribed mean-square accuracy $\varepsilon^2$, we derive complexity estimates that explicitly track the dependence on the dataset size $m$, mini-batch size $s$, and accuracy parameter $\varepsilon$. The resulting comparison reveals distinct parameter regimes in which either method is preferable. In particular, EM can have lower leading-order cost only in a small-data, aggressive-subsampling regime, whereas SGLD is favoured over most of the remaining parameter space. In the practically relevant regime $s \ll m$, the transition between the two methods occurs at the scale $m \asymp \varepsilon^{-1}$. We complement the theoretical analysis with numerical experiments based on a Gaussian Bayesian inference model, which examine the predicted cost regimes together with the underlying discretisation error and variance estimates.
Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-17 in stat.CO).
Categorie arXiv: Numerical Analysis (math.NA) ; Probability (math.PR); Computation (stat.CO).
Pagina del lavoro: https://arxiv.org/abs/2609.17750 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T15:08:06.749Z dalla pagina https://arxiv.org/catchup/stat.CO/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).