The Gradient Tracking Is a Distributed Integral Action
IEEE Transactions on Automatic Control · 2023-12 · 68 (12) · 7911-7918
DOI: 10.1109/tac.2023.3248487 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Automatic Control, 2023-12. Verfasst von Ivano Notarnicola und 3 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 37 Literaturangaben.
Abstract (Originalsprache)
We revisit the recent gradient tracking algorithm for distributed consensus optimization from a control theoretic viewpoint. We show that the algorithm can be constructed by solving a servomechanism control problem stemming from the distributed implementation of a centralized gradient method. Moreover, we show that, if expressed in proper coordinates, the gradient tracking embeds an integral action fed by a signal related to the consensus error. Finally, we provide an alternative convergence analysis based on Lyapunov arguments that also proves exponential asymptotic stability of the optimal equilibrium.
Zitieren
APA
Notarnicola, I., Bin, M., Marconi, L., & Notarstefano, G. (2023). The Gradient Tracking Is a Distributed Integral Action. IEEE Transactions on Automatic Control, 68(12), 7911-7918. https://doi.org/10.1109/tac.2023.3248487
Vancouver
Notarnicola I, Bin M, Marconi L, Notarstefano G. The Gradient Tracking Is a Distributed Integral Action. IEEE Transactions on Automatic Control. 2023;68(12):7911-7918. doi:10.1109/tac.2023.3248487
BibTeX
@article{Notarnicola2023,
author = {Notarnicola, Ivano and Bin, Michelangelo and Marconi, Lorenzo and Notarstefano, Giuseppe},
title = {The Gradient Tracking Is a Distributed Integral Action},
journal = {IEEE Transactions on Automatic Control},
year = {2023},
volume = {68},
number = {12},
pages = {7911--7918},
doi = {10.1109/tac.2023.3248487}
}
EUVIMED