Personalised federated learning for Riemannian and Euclidean EEG decoding
Abstract
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.
Related
- A converse bound for dGH vs. dH for Euclidean space and Riemannian manifolds
- Efficient Optimization on the Euclidean Sphere: Riemannian Gradient Alignment as a Unifying Principle
- Mapping Artificial Intelligence for Personalised Learning based on Learning Style: Scopus-Based Bibliometric Analysis (2006-2026)
- Progressive-Resolution Secure Aggregation for Federated Learning
- Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation
BibTeX
@misc{pautrel2026personalised,
title = {{Personalised federated learning for Riemannian and Euclidean EEG decoding}},
author = {Thibault Pautrel and Florent Bouchard and Ammar Mian and Guillaume Ginolhac},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.29037},
eprint = {2609.29037},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.29037},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Personalised federated learning for Riemannian and Euclidean EEG decoding
AU - Thibault Pautrel
AU - Florent Bouchard
AU - Ammar Mian
AU - Guillaume Ginolhac
PY - 2026
DA - 2026/09/24
PB - arXiv
DO - 10.48550/arxiv.2609.29037
UR - https://arxiv.org/abs/2609.29037
AB - 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.
ER -