Metraipolis › archive › dc_83c24239307c
Improved Convergence of Large Stepsize Gradient Descent for Logistic Regression — arXiv:2610.06675
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.06675
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:31:44.844Z
- SHA-256 of the archived text
81e57b85875c4a224058e972867b544ce423b858faa9b4f1bf6257d3c9f8ef0d- Stable URL of this page
https://metraipolis.com/documento/dc_83c24239307c
How to cite
Improved Convergence of Large Stepsize Gradient Descent for Logistic Regression — arXiv:2610.06675 — arXiv, https://arxiv.org/abs/2610.06675 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_83c24239307c (Metraipolis archive dc_83c24239307c, sha256 81e57b85875c4a224058e972867b544ce423b858faa9b4f1bf6257d3c9f8ef0d)
BibTeX
@misc{dc83c24239307c,
title = {Improved Convergence of Large Stepsize Gradient Descent for Logistic Regression — arXiv:2610.06675},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.06675},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_83c24239307c; sha256 81e57b85875c4a224058e972867b544ce423b858faa9b4f1bf6257d3c9f8ef0d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.06675. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We study gradient descent (GD) with a large constant stepsize for logistic regression on linearly separable data. Existing analysis shows an accelerated rate of $\widetilde{O}(1/\sqrt{\epsilon})$ to reach loss $\epsilon$ with an aggressive stepsize, although the loss may initially oscillate. Tighter control of the oscillatory dynamics has been available only for two-dimensional data. We prove a substantially faster rate in arbitrary dimension: GD with a large stepsize $\eta=1/\epsilon$ reaches loss $\epsilon$ within $O(\ln^{p}(1/\epsilon))$ steps, where $p$ depends only on the margin and the rank of the data. Our proof improves the bound on the transition time of GD from the oscillatory to the stable phase, after which the loss decreases monotonically. We split the oscillatory phase into recursively nested intervals. The margin and the rank bound the nesting depth, and a counting argument bounds the number of intervals at each depth, together yielding the polylogarithmic step complexity.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.06675 — 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 2026-10-06T05:27:05.904Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).