EUVIMEDEUVIMED entdecken
Crossref

Distributed Online Stochastic Convex-Concave Optimization: Dynamic Regret Analyses Under Single and Multiple Consensus Steps

Wentao Zhang, Baoyong Zhang, Deming Yuan, Shengyuan Xu, Vincent K. N. Lau

IEEE Transactions on Signal Processing · 2026 · 74 · 2258-2272

DOI: 10.1109/tsp.2026.3694033 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Signal Processing, 2026. Verfasst von Wentao Zhang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Research Grants Council under the Areas of Excellence Scheme.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper considers the distributed online convex-concave optimization with constraint sets over a multiagent network, in which each agent autonomously generates a series of decision pairs through a designable mechanism to cooperatively minimize the global loss function. To this end, under no-Euclidean distance metrics, we propose a distributed online stochastic mirror descent convex-concave optimization algorithm with time-varying predictive mappings. Taking dynamic saddle point regret as a performance metric, it is proved that the proposed algorithm achieves the regret upper-bound in $\boldsymbol{\mathcal{O}}\mathbf{(}\mathbf{max}\mathbf{\{}\boldsymbol{T^{\theta_{1}}},\boldsymbol{T^{\theta_{2}}}\mathbf{(1+}\boldsymbol{V_{T})}\mathbf{\})}$ for the general convex-concave loss function, where $\boldsymbol{\theta}_{\mathbf 1}\boldsymbol{,\theta}_{\mathbf 2}\boldsymbol{\in}\mathbf{(0,1)}$ are the tuning parameters, $\boldsymbol{T}$ is the total iteration time, and $\boldsymbol{V_{T}}$ is the path-variation. Surely, this algorithm guarantees the sublinear convergence, provided that $\boldsymbol{V_{T}}$ is sublinear. Moreover, aiming to achieve better convergence, we further investigate a variant of this algorithm by employing the multiple consensus technique. The obtained results show that the appropriate setting can effectively tighten the regret bound to a certain extent. Finally, the efficacy of the proposed algorithms is validated and compared through the simulation example of a target tracking problem.

Förderung

  • National Natural Science Foundation of China
  • Research Grants Council under the Areas of Excellence Scheme

Zitieren

APA

Zhang, W., Zhang, B., Yuan, D., Xu, S., & Lau, V. K. N. (2026). Distributed Online Stochastic Convex-Concave Optimization: Dynamic Regret Analyses Under Single and Multiple Consensus Steps. IEEE Transactions on Signal Processing, 74, 2258-2272. https://doi.org/10.1109/tsp.2026.3694033

Vancouver

Zhang W, Zhang B, Yuan D, Xu S, Lau VKN. Distributed Online Stochastic Convex-Concave Optimization: Dynamic Regret Analyses Under Single and Multiple Consensus Steps. IEEE Transactions on Signal Processing. 2026;74:2258-2272. doi:10.1109/tsp.2026.3694033

BibTeX
@article{Zhang2026,
  author = {Zhang, Wentao and Zhang, Baoyong and Yuan, Deming and Xu, Shengyuan and Lau, Vincent K. N.},
  title = {Distributed Online Stochastic Convex-Concave Optimization: Dynamic Regret Analyses Under Single and Multiple Consensus Steps},
  journal = {IEEE Transactions on Signal Processing},
  year = {2026},
  volume = {74},
  pages = {2258--2272},
  doi = {10.1109/tsp.2026.3694033}
}
Kostenlos starten →