S-JEPA: towards seamless cross-dataset transfer through dynamic spatial attention

arXiv:2403.11772v1 Announce Type: new
Abstract: Motivated by the challenge of seamless cross-dataset transfer in EEG signal processing, this article presents an exploratory study on the use of Joint Embedding Predictive Architectures (JEPAs). In recent years, self-supervised learning has emerged as a promising approach for transfer learning in various domains. However, its application to EEG signals remains largely unexplored. In this article, we introduce Signal-JEPA for representing EEG recordings which includes a novel domain-specific spatial block masking strategy and three novel …

abstract architectures article arxiv attention challenge cs.ai cs.lg dataset domains dynamic eeg embedding exploratory however jepa predictive processing self-supervised learning signal spatial study supervised learning through transfer transfer learning type