Problem
Research questions and friction points this paper is trying to address.
Federated Learning
EEG Decoding
Inter-subject Variability
Personalised Federated Learning
Riemannian Geometry
Personalised federated learning for Riemannian and Euclidean EEG decoding
🤖 AI Summary
This study addresses the underfitting of shared EEG decoders in federated learning caused by inter-subject variability. To this end, we propose a personalized federated learning framework that, for the first time, introduces personalization into Riemannian geometric networks. By integrating SPDNet with EEGNet, the method constructs a "shared backbone–private classification head" architecture, effectively accommodating individual differences while preserving data privacy. Experimental results demonstrate that the proposed framework outperforms both standard federated learning and centralized training in classification accuracy across three motor imagery datasets. Furthermore, it achieves faster convergence and lower communication overhead, highlighting its practical advantages for privacy-preserving, subject-adaptive 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.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Personalised federated learning
Riemannian geometry
SPDNet
EEG decoding
Motor imagery
🔎 Similar Papers
No similar papers found.
