EUVIMEDEUVIMED entdecken
Crossref

A Unified Convergence Analysis for Semi-Decentralized Learning: Sampled-to-Sampled vs. Sampled-to-All Communication

Angelo Rodio, Giovanni Neglia, Zheng Chen, Erik G. Larsson

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (30) · 25151-25159

DOI: 10.1609/aaai.v40i30.39705 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Angelo Rodio und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In semi-decentralized federated learning, devices primarily rely on device-to-device communication but occasionally interact with a central server. Periodically, a sampled subset of devices uploads their local models to the server, which computes an aggregate model. The server can then either (i) share this aggregate model only with the sampled clients (sampled-to-sampled, S2S) or (ii) broadcast it to all clients (sampled-to-all, S2A). Despite their practical significance, a rigorous theoretical and empirical comparison of these two strategies remains absent. We address this gap by analyzing S2S and S2A within a unified convergence framework that accounts for key system parameters: sampling rate, server aggregation frequency, and network connectivity. Our results, both analytical and experimental, reveal distinct regimes where one strategy outperforms the other, depending primarily on the degree of data heterogeneity across devices. These insights lead to concrete design guidelines for practical semi-decentralized FL deployments.

Zitieren

APA

Rodio, A., Neglia, G., Chen, Z., & Larsson, E. G. (2026). A Unified Convergence Analysis for Semi-Decentralized Learning: Sampled-to-Sampled vs. Sampled-to-All Communication. Proceedings of the AAAI Conference on Artificial Intelligence, 40(30), 25151-25159. https://doi.org/10.1609/aaai.v40i30.39705

Vancouver

Rodio A, Neglia G, Chen Z, Larsson EG. A Unified Convergence Analysis for Semi-Decentralized Learning: Sampled-to-Sampled vs. Sampled-to-All Communication. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(30):25151-25159. doi:10.1609/aaai.v40i30.39705

BibTeX
@article{Rodio2026,
  author = {Rodio, Angelo and Neglia, Giovanni and Chen, Zheng and Larsson, Erik G.},
  title = {A Unified Convergence Analysis for Semi-Decentralized Learning: Sampled-to-Sampled vs. Sampled-to-All Communication},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {30},
  pages = {25151--25159},
  doi = {10.1609/aaai.v40i30.39705}
}
Kostenlos starten →