FedBS improves federated EEG motor-imagery classification by keeping batch-normalization local and using sharpness-aware minimization, outperforming six FL baselines and typically centralized training.
Brain–computer interfaces for communication an d control,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Federated Motor Imagery Classification for Privacy-Preserving Brain-Computer Interfaces
FedBS improves federated EEG motor-imagery classification by keeping batch-normalization local and using sharpness-aware minimization, outperforming six FL baselines and typically centralized training.