Metraipolis › archive › dc_ee4ab05daed6

When Is Personalized Federated Learning Worth Its Deployment Cost? — arXiv:2611.02146

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02146
License (as archived)
CC0-1.0
Archived
2026-11-03T11:21:24.796Z
SHA-256 of the archived text
c8959a648f6d65ab62bfec1b4d03e5e1bbc0e44c64c31d1584c1d6efb95866ec
Stable URL of this page
https://metraipolis.com/documento/dc_ee4ab05daed6

How to cite

When Is Personalized Federated Learning Worth Its Deployment Cost? — arXiv:2611.02146 — arXiv (en), https://arxiv.org/abs/2611.02146 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ee4ab05daed6 (Metraipolis archive dc_ee4ab05daed6, sha256 c8959a648f6d65ab62bfec1b4d03e5e1bbc0e44c64c31d1584c1d6efb95866ec)
BibTeX
@misc{dcee4ab05daed6,
  title = {When Is Personalized Federated Learning Worth Its Deployment Cost? — arXiv:2611.02146},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02146},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ee4ab05daed6; sha256 c8959a648f6d65ab62bfec1b4d03e5e1bbc0e44c64c31d1584c1d6efb95866ec},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02146. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Personalized federated learning (PFL) may improve client-level performance, but deeper personalization increases deployment cost. We formulate personalization-depth selection as a utility-based empirical decision problem in which client utility equals Macro-F1 minus a cost-sensitivity weight times deployment cost, over six personalization depths labeled P0 through P5. We compare fixed policies with an offline per-client Oracle reference and report allocation maps over Dirichlet data heterogeneity and cost sensitivity. Although evaluated on this hierarchy, the formulation can accommodate any finite candidate set. On CIFAR-10, CIFAR-100, and a PTB-XL subset, personalization shows empirical, setting-dependent decision regimes rather than a universal benefit: higher cost sensitivity shifts Oracle mass toward the global-only depth P0; low cost sensitivity can favor deeper actions. Patterns are grid observations (three seeds per cell), not theoretical boundaries. Best-uniform is competitive with Oracle assignment (6.5 percent plus or minus 7.8 percent relative regret; 95 percent confidence interval 5.9 to 7.1), whereas random and personalize-all incur much larger regret (189.3 percent plus or minus 367.1 percent and 409.0 percent plus or minus 803.2 percent). Under the same utility, Ditto full-model personalization is substantially suboptimal once cost is counted. The contribution is an offline characterization of when personalization depth is worth its cost, not a new PFL optimizer or deployment-time scheduler.

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/2611.02146 — 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 2026-11-03T11:02:37.588Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).