Metraipolis › archive › dc_0202fcaeb953
Scalable Statistical Inference in Stochastic Gradient Descent — arXiv:2608.30845
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.30845
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T09:02:25.977Z
- SHA-256 of the archived text
be1c99b15c9090e9ece7de606297f3f87db3c2c819ceb42b5ebe1fa60a4d3b52- Stable URL of this page
https://metraipolis.com/documento/dc_0202fcaeb953
How to cite
Scalable Statistical Inference in Stochastic Gradient Descent — arXiv:2608.30845 — arXiv (en), https://arxiv.org/abs/2608.30845 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0202fcaeb953 (Metraipolis archive dc_0202fcaeb953, sha256 be1c99b15c9090e9ece7de606297f3f87db3c2c819ceb42b5ebe1fa60a4d3b52)
BibTeX
@misc{dc0202fcaeb953,
title = {Scalable Statistical Inference in Stochastic Gradient Descent — arXiv:2608.30845},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.30845},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0202fcaeb953; sha256 be1c99b15c9090e9ece7de606297f3f87db3c2c819ceb42b5ebe1fa60a4d3b52},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30845. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Constructing confidence regions for stochastic gradient descent (SGD) ideally requires estimating the asymptotic covariance matrix, a severe computational bottleneck in high dimensions. Traditional cancellation-based batch means methods bypass this estimation but require inverting a sample batch covariance matrix. This introduces strict mathematical degeneracy when the parameter dimension exceeds the number of batches. To address this problem, we utilize equal batch size batch means method and propose a simultaneous, marginal-friendly framework. The proposed marginal statistics has a asymptotic Student's $t$-distribution, and eliminates the matrix inversion step, entirely circumventing high-dimensional degeneracy. To achieve valid simultaneous coverage, we present an algorithm utilizing wild bootstrap samples drawn from a statistic as a function of only the diagonals of the variance-covariance estimator, and to further incorporate the contribution of cross-dependencies, we introduce an efficient Quasi-Monte Carlo procedure utilizing a $t$-copula approximation. Additionally, we integrate a Lugsail variance estimator to aggressively correct finite-sample bias and under-coverage. The proposed methodology delivers interpretable, simultaneous hyper-rectangular confidence regions that are statistically robust, memory-efficient, and strictly scalable for high-dimensional inference. The theoretical results are supported by extensive numerical simulation analysis through various aspects of dimension, number of batches and error structure. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-01 in stat.CO). Categorie arXiv: Computation (stat.CO). Pagina del lavoro: https://arxiv.org/abs/2608.30845 — 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 2026-10-17T08:40:03.068Z dalla pagina https://arxiv.org/catchup/stat.CO/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).