Deep Transfer Learning for Subject-Independent ERP-based BCIs

Abstract

Designing subject-independent Brain-Computer In- terfaces remains to be an open question for developing systems that can capture the inter-subject intrinsic brain features and classify them with reasonable accuracy. This paper presents the application of the state-of-the-art deep transfer learning archi- tectures on classifying ERP signals. We report 66.87%, 67.64%, 65.58%, and 71.93% test classification accuracy for DenseNet121, DenseNet201, Xception, and VGG-16 models, respectively. The experimental results demonstrate the viability of our approach in subject independent ERP-signals classification and suggest the better performance of models with fewer layers in classifying ERP signals.