Metraipolis › archive › dc_853cb794b075

On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression — arXiv:2609.25705

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.25705
License (as archived)
CC0-1.0
Archived
2026-10-08T15:08:26.110Z
SHA-256 of the archived text
5ba3fb9431ffa7b32d05898fb66f439963f4771cd98de71e4bf9461c1b4bb2e3
Stable URL of this page
https://metraipolis.com/documento/dc_853cb794b075

How to cite

On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression — arXiv:2609.25705 — arXiv (en), https://arxiv.org/abs/2609.25705 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_853cb794b075 (Metraipolis archive dc_853cb794b075, sha256 5ba3fb9431ffa7b32d05898fb66f439963f4771cd98de71e4bf9461c1b4bb2e3)
BibTeX
@misc{dc853cb794b075,
  title = {On the Gradient Heterogeneity Dynamics of Adversarially Robust Federated Regression — arXiv:2609.25705},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.25705},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_853cb794b075; sha256 5ba3fb9431ffa7b32d05898fb66f439963f4771cd98de71e4bf9461c1b4bb2e3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.25705. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Federated learning (FL) is intrinsically heterogeneous: honest clients may have different data-generating models. On top of that, adversarial clients can make heterogeneity even more pronounced by sharing arbitrary updates. Existing analyses typically control the interaction between statistical heterogeneity and adversarial behavior through gradient-dissimilarity conditions. However, the underlying bound is imposed a priori and may yield conservative guarantees even for least-squares regression. We instead derive the gradient heterogeneity from the statistical model of linear and nonlinear regression with fresh data samples at every round. Our bounds separate heterogeneity among the honest clients' ground-truth model parameters, finite-sample label noise, and initialization. We then demonstrate that, for any $(f,\kappa)$-robust aggregator with coefficient $\kappa = O(f/n)$, where $f$ is the number of adversarial clients and $n$ the total number of clients (with $f/n < 1/2$), convergence holds after an explicit sample burn-in.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.LG).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.25705 — 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 2026-10-08T14:37:29.544Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).