Cross task neural architecture search for EEG signal recognition

Electroencephalograms (EEGs) are brain dynamics measured outside of the brain, which have been widely utilized in non-invasive brain-computer interface applications. Recently, various neural network approaches have been proposed to improve the accuracy of EEG signal recognition. However, these approaches severely rely on manually designed network structures for different tasks which normally are not sharing the same empirical design cross-task-wise. In this paper, we propose a cross-task neural architecture search (CTNAS-EEG) framework for EEG signal recognition, which can automatically design the network structure across tasks and improve the recognition accuracy of EEG signals. Specifically, a compatible search space for cross-task searching and an efficient constrained searching method is pro-posed to overcome challenges brought by EEG signals. By unifying structure search on different EEG tasks, this work is the first to explore and analyze the searched structure difference in cross-task-wise. Moreover, by introducing architecture search, this work is the first to analyze model performance by cus-tomizing model structure for each human subject. Detailed experimental results suggest that the pro-posed CTNAS-EEG could reach state-of-the-art performance on different EEG tasks, such as Motor Imagery (MI) and Emotion recognition. Extensive experiments and detailed analysis are provided as a good reference for follow-up researchers.(c) 2023 Elsevier B.V. All rights reserved.

Automated summary

In this paper, a cross-task neural architecture search framework called CTNAS-EEG is proposed for EEG signal recognition, which can automatically design the network structure and improve the accuracy of EEG signal recognition. The framework explores and analyzes the differences in searched structures across different EEG tasks, and analyzes model performance by customizing the model structure for each human subject. The experimental results show that the proposed framework achieves state-of-the-art performance on various EEG tasks such as Motor Imagery and Emotion recognition.