English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Individual differences of Electroencephalogram (EEG) could cause the domainshift which would significantly degrade the performance of cross-subjectstrategy. The domain adversarial neural networks (DANN), where theclassification loss and domain loss jointly update the parameters of featureextractor, are adopted to deal with the domain shift. However, limited EEG dataquantity and strong individual difference are challenges for the DANN withcumbersome feature extractor. In this work, we propose knowledge distillation(KD) based lightweight DANN to enhance cross-subject EEG-based emotionrecognition. Specifically, the teacher model with strong context learningability is utilized to learn complex temporal dynamics and spatial correlationsof EEG, and robust lightweight student model is guided by the teacher model tolearn more difficult domain-invariant features. In the feature-based KDframework, a transformer-based hierarchical temporalspatial learning model isserved as the teacher model. The student model, which is composed of Bi-LSTMunits, is a lightweight version of the teacher model. Hence, the student modelcould be supervised to mimic the robust feature representations of teachermodel by leveraging complementary latent temporal features and spatialfeatures. In the DANN-based cross-subject emotion recognition, we combine theobtained student model and a lightweight temporal-spatial feature interactionmodule as the feature extractor. And the feature aggregation is fed to theemotion classifier and domain classifier for domain-invariant feature learning.To verify the effectiveness of the proposed method, we conduct thesubject-independent experiments on the public dataset DEAP with arousal andvalence classification. The outstanding performance and t-SNE visualization oflatent features verify the advantage and effectiveness of the proposed method.

A Lightweight Domain Adversarial Neural Network Based on Knowledge Distillation for EEG-based Cross-subject Emotion Recognition