Tag
This paper proposes OSPDIM, a source-free online unsupervised domain adaptation framework for EEG-based BCIs that corrects geometric misalignment caused by class-imbalanced label shifts on the Riemannian manifold, outperforming standard alignment methods in online scenarios.
This paper proposes a locality-aware private class identification approach and a reliable optimal transport-based method (ReOT) to address domain adaptation challenges under extreme label shift, particularly distinguishing shared from private classes.