Physics-informed and Unsupervised Riemannian Domain Adaptation for Machine Learning on Heterogeneous EEG Datasets

arXiv:2403.15415v1 Announce Type: cross
Abstract: Combining electroencephalogram (EEG) datasets for supervised machine learning (ML) is challenging due to session, subject, and device variability. ML algorithms typically require identical features at train and test time, complicating analysis due to varying sensor numbers and positions across datasets. Simple channel selection discards valuable data, leading to poorer performance, especially with datasets sharing few channels. To address this, we propose an unsupervised approach leveraging EEG signal physics. We map EEG channels to fixed positions …

abstract algorithms analysis arxiv cs.lg datasets domain domain adaptation eeg eess.sp features machine machine learning ml algorithms numbers physics physics-informed sensor session simple supervised machine learning test train type unsupervised