EUVIMEDScopri EUVIMED
Crossref

A multilevel stochastic regularized first-order method with application to finite sum minimization

Filippo Marini, Margherita Porcelli, Elisa Riccietti

Mathematics of Computation · 2026-08-13

DOI: 10.1090/mcom/4232 ↗

At a glance

Journal article in Mathematics of Computation, published online on 2026-08-13. Written by Filippo Marini and 2 co-authors. 43 references. Funded by: Ministero dell’Istruzione, dell’Università e della Ricerca, Istituto Nazionale di Alta Matematica "Francesco Severi", Agence Nationale de la Recherche.

AbstractDati bibliograficiCiteRelated studies

Abstract (original language)

In this paper, we propose a multilevel stochastic framework for the solution of nonconvex unconstrained optimization problems. The proposed approach uses random regularized first-order models that exploit an available hierarchical description of the problem, being either in the classical variable space or in the function space, meaning that different levels of accuracy for the objective function are available. We propose a convergence analysis showing an almost sure global convergence of the method to a first-order stationary point. The numerical behavior is tested on the solution of finite sum minimization problems. Differently from classical deterministic multilevel schemes, our stochastic method does not require the finest approximation to coincide with the original objective function along all the optimization process. This allows for significantly decreasing their cost, for instance in data-fitting problems, where considering all the data at each iteration can be avoided.

Funding

  • Ministero dell’Istruzione, dell’Università e della Ricerca
  • Istituto Nazionale di Alta Matematica "Francesco Severi"
  • Agence Nationale de la Recherche
  • Fondation Simone et Cino Del Duca

Cite

APA

Marini, F., Porcelli, M., & Riccietti, E. (2026). A multilevel stochastic regularized first-order method with application to finite sum minimization. Mathematics of Computation. https://doi.org/10.1090/mcom/4232

Vancouver

Marini F, Porcelli M, Riccietti E. A multilevel stochastic regularized first-order method with application to finite sum minimization. Mathematics of Computation. 2026. doi:10.1090/mcom/4232

BibTeX
@article{Marini2026,
  author = {Marini, Filippo and Porcelli, Margherita and Riccietti, Elisa},
  title = {A multilevel stochastic regularized first-order method with application to finite sum minimization},
  journal = {Mathematics of Computation},
  year = {2026},
  doi = {10.1090/mcom/4232}
}
Start for free →