• Media type: E-Book
  • Title: Stopping rules for accelerated gradient methods with additive noise in gradient
  • Contributor: Vasin, Artem [Author]; Gasnikov, Alexander [Author]; Spokojnyj, Vladimir G. [Author]
  • Corporation: Weierstraß-Institut für Angewandte Analysis und Stochastik
  • Published: Berlin: Weierstraß-Institut für Angewandte Analysis und Stochastik Leibniz-Institut im Forschungsverbund Berlin e.V., 2021
  • Published in: Weierstraß-Institut für Angewandte Analysis und Stochastik: Preprint ; 2812
  • Extent: 1 Online-Ressource (40 Seiten, 1,11 MB); Diagramme
  • Language: English
  • DOI: 10.20347/WIAS.PREPRINT.2812
  • Identifier:
  • Keywords: Forschungsbericht
  • Origination:
  • Footnote: Literaturverzeichnis: Seite 35-38
  • Description: In this article, we investigate an accelerated first-order method, namely, the method of similar triangles, which is optimal in the class of convex (strongly convex) problems with a Lipschitz gradient. The paper considers a model of additive noise in a gradient and a Euclidean proxstructure for not necessarily bounded sets. Convergence estimates are obtained in the case of strong convexity and its absence, and a stopping criterion is proposed for not strongly convex problems.
  • Access State: Open Access