Optimal convergence rates of totally asynchronous optimization

Asynchronous optimization algorithms are at the core of modern machine learning and resource allocation systems. However, most convergence results consider bounded information delays and several important algorithms lack guarantees when they operate under total asynchrony. In this paper, we derive e...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the IEEE Conference on Decision & Control S. 6484 - 6490
Hauptverfasser: Wu, Xuyang, Magnusson, Sindri, Reza Feyzmahdavian, Hamid, Johansson, Mikael
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 06.12.2022
Schlagworte:
Computational modeling
Delays
Indexes
Machine learning
Machine learning algorithms
Numerical models
Resource management
ISSN:2576-2370
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Asynchronous optimization algorithms are at the core of modern machine learning and resource allocation systems. However, most convergence results consider bounded information delays and several important algorithms lack guarantees when they operate under total asynchrony. In this paper, we derive explicit convergence rates for the proximal incremental aggregated gradient (PIAG) and the asynchronous block-coordinate descent (Async-BCD) methods under a specific model of total asynchrony, and show that the derived rates are order-optimal. The convergence bounds provide an insightful understanding of how the growth rate of the delays deteriorates the convergence times of the algorithms. Our theoretical findings are demonstrated by a numerical example.
ISSN:2576-2370
DOI:10.1109/CDC51059.2022.9993168