GDDN: Graph Domain Disentanglement Network for Generalizable EEG Emotion Recognition

• Published in: IEEE transactions on affective computing Vol. 15; no. 3; pp. 1739 - 1753
• Main Authors: Chen, Bianna, Chen, C. L. Philip, Zhang, Tong
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.07.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Affective computing; Brain modeling; Computational modeling; Disentanglement learning; EEG connectivity; Electroencephalography; Emotion recognition; Emotional factors; Emotions; generalizability; Graph representations; Graph theory; Graphical representations; inter-subject variability; Modules; Predictive models; Training
• ISSN: 1949-3045; 1949-3045
• DOI: 10.1109/TAFFC.2024.3371540

Cross-subject EEG emotion recognition suffers a major setback due to high inter-subject variability in emotional responses. Many prior studies have endeavored to alleviate the inter-subject discrepancies of EEG feature distributions, ignoring the variable EEG connectivity and prediction deviation caused by individual differences, which may cause poor generalization to the unseen subject. This article proposes a graph domain disentanglement network (GDDN) to generalize EEG emotion recognition across subjects in terms of EEG connectivity, representation, and prediction. More specifically, a graph domain disentanglement module is proposed to extract common-specific characteristics on both EEG graph connectivity and graph representation, enabling a more comprehensive network transferability to the unseen individual. Meanwhile, to strengthen stable emotion prediction capability, a domain-adaptive classifier aggregation module is developed to facilitate adaptive emotional prediction for the unseen individual conditioned on the domain weights of the input individuals. Finally, an auxiliary supervision module is imposed to alleviate the domain discrepancy and reduce information loss during the disentanglement learning. Extensive experiments on three public EEG emotion datasets, i.e., SEED, SEED-IV, and MPED, validate the superior generalizability of GDDN compared with the state-of-the-art methods.