EEG-based emotion recognition systems face significant challenges when deployed across different domains due to limited labeled data and individual signal variability. We introduce …
We introduce a predictive coding self-supervised learning (PC-SSL) approach to overcome the limitations of traditional supervised methods hindered by the high cost of labeling EEG …
Vision transformers have shown tremendous success in numerous computer vision applications; however, they have not been exploited for stress assessment using physiological signals …
We propose DynaGuide, a generalizable dynamic guidance framework that enables zero-shot guided unsupervised semantic segmentation without requiring labeled training data. The …
We propose an enhanced unsupervised domain adaptation framework for EEG-based emotion recognition that achieves robust cross-dataset generalization without requiring target-domain …