Metraipolis › archive › dc_efdde0da66bd
SPADE-DFL: Communication-Efficient Decentralized Federated Learning via Derivative-Free Linearized ADMM — arXiv:2609.29446
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.29446
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T05:26:00.388Z
- SHA-256 of the archived text
27a0c7809b6a1bb87131912a3160cd53188e7385b83fc7a8fe48eee55f85869c- Stable URL of this page
https://metraipolis.com/documento/dc_efdde0da66bd
How to cite
SPADE-DFL: Communication-Efficient Decentralized Federated Learning via Derivative-Free Linearized ADMM — arXiv:2609.29446 — arXiv (en), https://arxiv.org/abs/2609.29446 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_efdde0da66bd (Metraipolis archive dc_efdde0da66bd, sha256 27a0c7809b6a1bb87131912a3160cd53188e7385b83fc7a8fe48eee55f85869c)
BibTeX
@misc{dcefdde0da66bd,
title = {SPADE-DFL: Communication-Efficient Decentralized Federated Learning via Derivative-Free Linearized ADMM — arXiv:2609.29446},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.29446},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_efdde0da66bd; sha256 27a0c7809b6a1bb87131912a3160cd53188e7385b83fc7a8fe48eee55f85869c},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29446. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reducing communication in derivative-free decentralized learning requires controlling the disagreement accumulated over multiple local updates. This paper develops SPADE-DFL, a primal--dual method that allows the number of local function-value updates between neighbor exchanges to grow with the computation budget while preserving the nonprivate convergence order. For smooth nonconvex objectives under uniform query-moment bounds, the prescribed nonprivate schedule achieves a time-averaged stationarity and consensus bound of $\mathcal{O}(T^{-1/3})$ using only $\Theta(T^{2/3})$ communication rounds, where $T$ is the number of local updates per client. For private training, the accumulated data-dependent increment is isolated from the graph correction, allowing one protected state per client and round to generate all outgoing messages. We prove client-level differential privacy for the full interactive transcript and quantify the resulting optimization error over a finite horizon. Experiments on four classification tasks show that SPADE-DFL achieves higher mean test accuracy than existing decentralized learning methods.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG) ; Networking and Internet Architecture (cs.NI).
Pagina del lavoro: https://arxiv.org/abs/2609.29446 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T04:50:39.404Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).