EUVIMEDEUVIMED entdecken
Crossref

On the Polyak momentum variants of the greedy deterministic single and multiple row‐action methods

Nian‐Ci Wu, Qian Zuo, Yatian Wang

Numerical Linear Algebra with Applications · 2024-08 · 31 (4)

DOI: 10.1002/nla.2552 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Numerical Linear Algebra with Applications, online erschienen am 14.03.2024. Verfasst von Nian‐Ci Wu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

For solving a consistent system of linear equations, the classical row‐action method, such as Kaczmarz method, is a simple while really effective iteration solver. Based on the greedy index selection strategy and Polyak's heavy‐ball momentum acceleration technique, we propose two deterministic row‐action methods and establish the corresponding convergence theory. We show that our algorithm can linearly converge to a least‐squares solution with minimum Euclidean norm. Several numerical studies have been presented to corroborate our theoretical findings. Real‐world applications, such as data fitting in computer‐aided geometry design, are also presented for illustrative purposes.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Wu, N., Zuo, Q., & Wang, Y. (2024). On the Polyak momentum variants of the greedy deterministic single and multiple row‐action methods. Numerical Linear Algebra with Applications, 31(4). https://doi.org/10.1002/nla.2552

Vancouver

Wu N, Zuo Q, Wang Y. On the Polyak momentum variants of the greedy deterministic single and multiple row‐action methods. Numerical Linear Algebra with Applications. 2024;31(4). doi:10.1002/nla.2552

BibTeX
@article{Wu2024,
  author = {Wu, Nian‐Ci and Zuo, Qian and Wang, Yatian},
  title = {On the Polyak momentum variants of the greedy deterministic single and multiple row‐action methods},
  journal = {Numerical Linear Algebra with Applications},
  year = {2024},
  volume = {31},
  number = {4},
  doi = {10.1002/nla.2552}
}
Kostenlos starten →