Federated Distributionally Robust Optimization With Non-Convex Objectives: Algorithm and Analysis
IEEE Transactions on Mobile Computing · 2026-01 · 25 (1) · 1219-1235
DOI: 10.1109/tmc.2025.3602796 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2026-01. Verfasst von Yang Jiao und 2 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 94 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Asiainfo Technologies, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Distributionally Robust Optimization (DRO), which aims to find an optimal decision that minimizes the worst case cost over the ambiguity set of probability distribution, has been widely applied in diverse applications, e.g., network behavior analysis, risk management, etc. Nevertheless, prevailing DRO techniques encounter three primary challenges in distributed environments: 1) addressing asynchronous updating efficiently; 2) leveraging the prior distribution effectively; 3) appropriately adjusting the degree of robustness based on varying scenarios. To this end, we propose an asynchronous distributed algorithm, named A synchronous S ingle-loo P alternat I ve g R adient proj E ction (ASPIRE) algorithm with the it E rative A ctive S E t method (EASE) to tackle the federated distributionally robust optimization (FDRO) problem. In addition, a new uncertainty set, i.e., constrained $D$ D -norm uncertainty set, is developed to effectively leverage the prior distribution and flexibly control the degree of robustness. We further enhance the proposed framework by integrating various uncertainty sets and conducting a comprehensive theoretical analysis of the computational complexity associated with each uncertainty set. To expedite convergence speed, we also introduce ASPIRE-ADP, a method that can dynamically adjust the number of active workers. Finally, our theoretical analysis elucidates that the proposed algorithm is guaranteed to converge and the iteration complexity and communication complexity are also analyzed. Extensive empirical studies on real-world datasets validate that the proposed method excels not only in achieving fast convergence and robustness against data heterogeneity and malicious attacks but also in effectively managing the trade-off between robustness and performance.
Förderung
- National Natural Science Foundation of China
- Asiainfo Technologies
- Fundamental Research Funds for the Central Universities
- Fundamental Research Funds of Shanghai Jiading District
Zitieren
APA
Jiao, Y., Yang, K., & Song, D. (2026). Federated Distributionally Robust Optimization With Non-Convex Objectives: Algorithm and Analysis. IEEE Transactions on Mobile Computing, 25(1), 1219-1235. https://doi.org/10.1109/tmc.2025.3602796
Vancouver
Jiao Y, Yang K, Song D. Federated Distributionally Robust Optimization With Non-Convex Objectives: Algorithm and Analysis. IEEE Transactions on Mobile Computing. 2026;25(1):1219-1235. doi:10.1109/tmc.2025.3602796
BibTeX
@article{Jiao2026,
author = {Jiao, Yang and Yang, Kai and Song, Dongjin},
title = {Federated Distributionally Robust Optimization With Non-Convex Objectives: Algorithm and Analysis},
journal = {IEEE Transactions on Mobile Computing},
year = {2026},
volume = {25},
number = {1},
pages = {1219--1235},
doi = {10.1109/tmc.2025.3602796}
}
EUVIMED