EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mean-Square Convergence of Particle Swarm Optimization via Stochastic Momentum Analysis

Boris Budak, Georgii Vorontsov

Mathematics · 12.06.2026 · 14 (12) · 2107

DOI: 10.3390/math14122107 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mathematics, online erschienen am 12.06.2026. Verfasst von Boris Budak und 1 weiteren Autor:innen. Die Arbeit nennt 14 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Commission on Science, Technology and Innovation of the Shenzhen Municipality.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We analyze the standard multi-particle particle swarm optimization (PSO) algorithm with global-best (all-to-all) topology and constant hyperparameters on smooth strongly convex objectives. By rewriting the PSO velocity recursion as a stochastic heavy-ball method acting on a time-varying quadratic surrogate defined by the personal and global bests, and by applying a Lyapunov drift argument in the style of stochastic momentum analyses, we obtain mean-square convergence of particle positions to the unique minimizer and convergence of the best-so-far objective gaps. The deterministic PSO obtained by fixing the random coefficients at their mean values appears as a noise-free special case of the same Lyapunov framework.

Förderung

  • Commission on Science, Technology and Innovation of the Shenzhen Municipality

Zitieren

APA

Budak, B., & Vorontsov, G. (2026). Mean-Square Convergence of Particle Swarm Optimization via Stochastic Momentum Analysis. Mathematics, 14(12), 2107. https://doi.org/10.3390/math14122107

Vancouver

Budak B, Vorontsov G. Mean-Square Convergence of Particle Swarm Optimization via Stochastic Momentum Analysis. Mathematics. 2026;14(12):2107. doi:10.3390/math14122107

BibTeX
@article{Budak2026,
  author = {Budak, Boris and Vorontsov, Georgii},
  title = {Mean-Square Convergence of Particle Swarm Optimization via Stochastic Momentum Analysis},
  journal = {Mathematics},
  year = {2026},
  volume = {14},
  number = {12},
  pages = {2107},
  doi = {10.3390/math14122107}
}
Kostenlos starten →