Metraipolis › archive › dc_1571815bf15a

Bridging Compressed Communication and Local Steps via Spectral Method in Federated Learning — arXiv:2610.23434

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.23434
License (as archived)
CC0-1.0
Archived
2026-10-19T09:10:16.404Z
SHA-256 of the archived text
fbf270b458054c83289fe5c3e432a778ce3127844fa2bfff1cb08d0f5c62e28c
Stable URL of this page
https://metraipolis.com/documento/dc_1571815bf15a

How to cite

Bridging Compressed Communication and Local Steps via Spectral Method in Federated Learning — arXiv:2610.23434 — arXiv (en), https://arxiv.org/abs/2610.23434 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1571815bf15a (Metraipolis archive dc_1571815bf15a, sha256 fbf270b458054c83289fe5c3e432a778ce3127844fa2bfff1cb08d0f5c62e28c)
BibTeX
@misc{dc1571815bf15a,
  title = {Bridging Compressed Communication and Local Steps via Spectral Method in Federated Learning — arXiv:2610.23434},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23434},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1571815bf15a; sha256 fbf270b458054c83289fe5c3e432a778ce3127844fa2bfff1cb08d0f5c62e28c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23434. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Federated Learning (FL) enables collaborative training of complex models across large-scale networks of edge devices while keeping data localized. However, data heterogeneity and high communication overhead remain fundamental bottlenecks. Recently, the spectral optimizer Muon has demonstrated remarkable advantages over Adam-like methods in centralized training. While recent efforts have attempted to transition Muon to FL, naive integrations into traditional architectures fail short. Crucially, existing approaches do not support the the communication compression and local steps --- the two foundational techniques of efficient FL. To address this limitation, we introduce a novel spectral FL method that utilizes both local updates and compression. We establish rigorous convergence guarantees under mild assumptions, notably eliminating the restrictive bounded heterogeneity condition. Empirical evaluation on standard image classification benchmarks validates the efficiency and robustness of our approach.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.23434 — 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 2026-10-19T09:06:29.549Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).