EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

On the Adaptive Penalty Parameter Selection in ADMM

Serena Crisci, Valentina De Simone, Marco Viola

Algorithms · 25.05.2023 · 16 (6) · 264

DOI: 10.3390/a16060264 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Algorithms, online erschienen am 25.05.2023. Verfasst von Serena Crisci und 2 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Italian Ministry of University, Istituto Nazionale di Alta Matematica-Gruppo Nazionale per il Calcolo Scientifico (INdAM-GNCS).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Many data analysis problems can be modeled as a constrained optimization problem characterized by nonsmooth functionals, often because of the presence of ℓ1-regularization terms. One of the most effective ways to solve such problems is through the Alternate Direction Method of Multipliers (ADMM), which has been proved to have good theoretical convergence properties even if the arising subproblems are solved inexactly. Nevertheless, experience shows that the choice of the parameter τ penalizing the constraint violation in the Augmented Lagrangian underlying ADMM affects the method’s performance. To this end, strategies for the adaptive selection of such parameter have been analyzed in the literature and are still of great interest. In this paper, starting from an adaptive spectral strategy recently proposed in the literature, we investigate the use of different strategies based on Barzilai–Borwein-like stepsize rules. We test the effectiveness of the proposed strategies in the solution of real-life consensus logistic regression and portfolio optimization problems.

Förderung

  • Italian Ministry of University
  • Istituto Nazionale di Alta Matematica-Gruppo Nazionale per il Calcolo Scientifico (INdAM-GNCS)

Zitieren

APA

Crisci, S., Simone, V. D., & Viola, M. (2023). On the Adaptive Penalty Parameter Selection in ADMM. Algorithms, 16(6), 264. https://doi.org/10.3390/a16060264

Vancouver

Crisci S, Simone VD, Viola M. On the Adaptive Penalty Parameter Selection in ADMM. Algorithms. 2023;16(6):264. doi:10.3390/a16060264

BibTeX
@article{Crisci2023,
  author = {Crisci, Serena and Simone, Valentina De and Viola, Marco},
  title = {On the Adaptive Penalty Parameter Selection in ADMM},
  journal = {Algorithms},
  year = {2023},
  volume = {16},
  number = {6},
  pages = {264},
  doi = {10.3390/a16060264}
}
Kostenlos starten →