The paper explores personalised federated learning for EEG decoding using two lightweight models: the Riemannian SPDNet and the Euclidean EEGNet. In the personalised approach, all subjects share a common trunk while keeping individual heads, which improves accuracy and reduces communication compared to standard federated learning and centralised training. Experiments on three motor‑imagery datasets show that personalised SPDNet outperforms both standard FL and centralised training, and beats EEGNet on two datasets, though centralised EEGNet remains superior to centralised SPDNet.
arXiv:2606. 16462v1 Announce Type: cross Abstract: Cross-subject EEG decoding promises more training data, but it also exposes neural networks to strong inter-subject distribution shifts.
By Bruna J. Lopes, Gabriel Schwartz, Sylvain Chevallier, Raphael Y. de Camargo, Bruno Aristimunha
arXiv:2607. 22733v1 Announce Type: cross Abstract: We investigate whether a generative model can supply useful synthetic motor-imagery (MI) electroencephalography (EEG) trials that improve the accuracy of independent downstream classifiers.
By Matei Moldoveanu, Alain Sirois, Claire Ben Ali, Fabien Lotte, Florian Yger
arXiv:2604.22494v2 Announce Type: replace-cross
Abstract: We introduce two federated learning frameworks for the classical SPDnet model operating on symmetric positive definite (SPD) matrices with St...
By Thibault Pautrel, Florent Bouchard, Ammar Mian, Guillaume Ginolhac
arXiv:2601. 17883v3 Announce Type: replace Abstract: Electroencephalography (EEG) foundation models (FMs) have recently emerged as a promising paradigm for brain-computer interfaces, aiming to learn transferable neural representations from large-scale heterogeneous recordings.
By Dingkun Liu, Yuheng Chen, Zhu Chen, Zhenyao Cui, Yaozhi Wen, Jiayu An, Jingwei Luo, Dongrui Wu
BRIDGE-EEG is an efficient multi‑task EEG classification pipeline that leverages self‑supervised pretraining while dramatically reducing model size. It maps heterogeneous EEG recordings to a unified 62‑channel time‑frequency representation, pretrains an SE‑ResNet18 teacher with SimCLR, and distills it into smaller SE‑ResNet8 and SE‑ResNet4 students. The compact models achieve accuracy comparable to or better than larger foundation models on abnormality detection and emotion recognition, and they consume up to three times less energy on edge devices, enabling deployment on wearable hardware.
By Meghna Roy Chowdhury, Chengwei Zhou, Haotian Yu, Gourav Datta, Shreyas Sen