Metraipolis › archive › dc_f7d41226692a

Convergence of Stochastic Gradient Methods under Heavy-Tailed Noise and Hölder Smoothness — arXiv:2609.12785

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.12785
License (as archived)
CC0-1.0
Archived
2026-10-16T03:15:49.773Z
SHA-256 of the archived text
369fe06f4fa2d57a5d91e9e79cace4506c6d84e88099810a271040fd639ace7a
Stable URL of this page
https://metraipolis.com/documento/dc_f7d41226692a

How to cite

Convergence of Stochastic Gradient Methods under Heavy-Tailed Noise and Hölder Smoothness — arXiv:2609.12785 — arXiv (en), https://arxiv.org/abs/2609.12785 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f7d41226692a (Metraipolis archive dc_f7d41226692a, sha256 369fe06f4fa2d57a5d91e9e79cace4506c6d84e88099810a271040fd639ace7a)
BibTeX
@misc{dcf7d41226692a,
  title = {Convergence of Stochastic Gradient Methods under Heavy-Tailed Noise and Hölder Smoothness — arXiv:2609.12785},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.12785},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f7d41226692a; sha256 369fe06f4fa2d57a5d91e9e79cace4506c6d84e88099810a271040fd639ace7a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.12785. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Classical convergence guarantees for stochastic gradient methods typically assume Lipschitz-smooth objectives and finite-variance gradient noise, both frequently violated in practice. In contrast, we study nonconvex stochastic optimization under the joint relaxation of these assumptions: objectives with $(L,s)$-Hölder continuous gradients, $s\in(0,1]$, and gradient noise satisfying only a bounded $\alpha$-th moment condition for $\alpha\in(1,2]$. We establish three convergence results. Firstly, that standard SGD converges at rate $O(T^{-s/(1+s)})$ whenever $\alpha\ge1+s$, extending the classical nonconvex SGD rate to heavy-tailed noise and Hölder smoothness simultaneously. Secondly, we analyze $\delta$-regularized gradient clipping ($\delta$-GClip), a provable trainer of wide and deep nets, and establish a stationarity rate of $O(T^{-2s(\alpha-1)/[(1+s)(2\alpha-1)]})$ under the same condition. Thirdly, we analyze standard gradient clipping (G-Clip) and show that it recovers the above rate for $\alpha\ge1+s$ while in the very heavy-tailed regime $\alpha<1+s$, it has a convergence rate $O(T^{-2s(\alpha-1)/[(\alpha-1)+s(2\alpha-1)]})$ --- the first convergence guarantee in this regime for any stochastic gradient based method.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-14 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Optimization and Control (math.OC); Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2609.12785 — 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 2026-10-16T03:05:44.063Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).