Metraipolis › archive › dc_96076f41db10

Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis, Causal Investigation, and the Limits of Divergence-Aware Mitigation — arXiv:2609.16977

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.16977
License (as archived): CC0-1.0
Archived: 2026-10-15T19:29:40.223Z
SHA-256 of the archived text: 0f55edb81f37b7d40766f0b0b724cbd2200481f7618400146f8b13760f69f66c
Stable URL of this page: https://metraipolis.com/documento/dc_96076f41db10

How to cite

Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis, Causal Investigation, and the Limits of Divergence-Aware Mitigation — arXiv:2609.16977 — arXiv (en), https://arxiv.org/abs/2609.16977 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_96076f41db10 (Metraipolis archive dc_96076f41db10, sha256 0f55edb81f37b7d40766f0b0b724cbd2200481f7618400146f8b13760f69f66c)

BibTeX

@misc{dc96076f41db10,
  title = {Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis, Causal Investigation, and the Limits of Divergence-Aware Mitigation — arXiv:2609.16977},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.16977},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_96076f41db10; sha256 0f55edb81f37b7d40766f0b0b724cbd2200481f7618400146f8b13760f69f66c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.16977. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Federated learning lets multiple participants train a shared model without pooling raw data, by exchanging locally trained model updates instead. Federated averaging assumes that averaging local models is a reasonable way to solve one shared problem when participants' data are broadly similar. Work on non-IID federated learning has shown that this assumption can withstand differences in label and feature distributions. We ask whether it survives a different strain specific to graph neural networks, where client graphs differ not in label or feature distribution but in structure itself, requiring the same shared weights to operate over fundamentally different topologies. We call the resulting harm structural negative transfer. In a federation of real citation networks and synthetic structural proxies, a structurally atypical client lost more than half its achievable accuracy simply by joining. In an initial six-client federation, two label-free structural statistics computable before training were strongly associated with this harm. Expanding to twenty clients showed that degree divergence remained associated with harm, although more weakly, and survived removal of domain contrast. Spectral divergence did not replicate, which we trace to a confound caused by the composition of the reference pool used for leave-one-out statistics. A causal intervention isolating topology found no significant effect. A degree-normalization mechanism held across twenty-four seeds but did not explain the harm when corrected. The best of five candidate fixes beat a tuned baseline only until a matched, structurally blind control was applied, after which the gain disappeared. What survives is a modest, partially replicated, degree-specific signal that is not yet a validated predictor at scale.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-16 in cs.DC).
Categorie arXiv: Machine Learning (cs.LG) ; Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2609.16977 — 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 2026-10-15T18:35:05.434Z dalla pagina https://arxiv.org/catchup/cs.DC/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).