Metraipolis › archive › dc_2cf856965716

Personalised federated learning for Riemannian and Euclidean EEG decoding — arXiv:2609.29037

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29037
License (as archived)
CC0-1.0
Archived
2026-10-08T05:26:27.780Z
SHA-256 of the archived text
4490fa7cbd3fa7b5e4fbd335637b7c8eabe3d74dd0eb54b4e6d76a74f849e1ca
Stable URL of this page
https://metraipolis.com/documento/dc_2cf856965716

How to cite

Personalised federated learning for Riemannian and Euclidean EEG decoding — arXiv:2609.29037 — arXiv (en), https://arxiv.org/abs/2609.29037 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2cf856965716 (Metraipolis archive dc_2cf856965716, sha256 4490fa7cbd3fa7b5e4fbd335637b7c8eabe3d74dd0eb54b4e6d76a74f849e1ca)
BibTeX
@misc{dc2cf856965716,
  title = {Personalised federated learning for Riemannian and Euclidean EEG decoding — arXiv:2609.29037},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29037},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2cf856965716; sha256 4490fa7cbd3fa7b5e4fbd335637b7c8eabe3d74dd0eb54b4e6d76a74f849e1ca},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29037. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Federated learning (FL) lets EEG decoders learn from recordings of several subjects without pooling them. We consider two light EEG decoders, the Riemannian SPDNet and the Euclidean EEGNet. Both split into a trunk, which builds a latent representation, and a head, which classifies it. Inter-subject variability, however, makes a single shared FL model a poor fit for each subject. Personalised FL addresses this: all subjects learn a common trunk, and each subject keeps its own head. We adapt it for SPDNet and study its effects against standard FL and centralised training, with EEGNet as a Euclidean baseline. Experiments cover three motor-imagery datasets that span diverse regimes in channels, subjects and classes. We observe that personalised SPDNet reaches higher accuracy than both standard FL and centralised training, while converging in fewer rounds and communicating fewer parameters than standard FL. It also outperforms every EEGNet configuration on two of the three datasets, although centralised EEGNet outperforms centralised SPDNet.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.LG).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.29037 — 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 2026-10-08T04:50:39.576Z 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).