Learning the relative composition of EEG signals using pairwise relative shift pretraining
This work addresses the need for efficient EEG representation learning in clinical applications like sleep staging and seizure detection, offering a novel approach that improves performance over current methods.
The paper tackled the problem of learning EEG representations from unlabeled data by introducing PARS pretraining, a method that predicts relative temporal shifts between EEG window pairs, and demonstrated that it outperforms existing strategies in label-efficient and transfer learning settings.
Self-supervised learning (SSL) offers a promising approach for learning electroencephalography (EEG) representations from unlabeled data, reducing the need for expensive annotations for clinical applications like sleep staging and seizure detection. While current EEG SSL methods predominantly use masked reconstruction strategies like masked autoencoders (MAE) that capture local temporal patterns, position prediction pretraining remains underexplored despite its potential to learn long-range dependencies in neural signals. We introduce PAirwise Relative Shift or PARS pretraining, a novel pretext task that predicts relative temporal shifts between randomly sampled EEG window pairs. Unlike reconstruction-based methods that focus on local pattern recovery, PARS encourages encoders to capture relative temporal composition and long-range dependencies inherent in neural signals. Through comprehensive evaluation on various EEG decoding tasks, we demonstrate that PARS-pretrained transformers consistently outperform existing pretraining strategies in label-efficient and transfer learning settings, establishing a new paradigm for self-supervised EEG representation learning.