Transfer Learning Through Maximum Classification Discrepancy for EEG Signal Processing in the Motor Imagery Paradigm

• Grant number: 25/06583-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2025
• End date: June 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Denis Gustavo Fantinato
• Grantee: Gabriel da Costa
• Host Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Company: Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Elétrica e de Computação (FEEC)
• Associated research grant: 20/09838-0 - BI0S - Brazilian Institute of Data Science, AP.PCPE

Abstract

Electroencephalographic (EEG) data are recordings of the brain's electrical activity and have a wide range of applications, from medical diagnostics and assistive technologies to rehabilitation and entertainment devices. In this context, with the advancement of signal processing and machine learning techniques, one system that stands out is the direct communication pathway between the brain and a computer, known as a Brain-Computer Interface (BCI). However, the considerable variability observed in user patterns within BCI systems, as well as their use in increasingly sophisticated applications, makes the effective use of this interface a challenging task. In this regard, the present project aims to perform an in-depth study of Deep Learning and Transfer Learning methods for processing EEG signals. Specifically, the focus will be on the EEGNet architecture, a deep learning-based artificial neural network, and its combination with the Maximum Classification Discrepancy method. A comparative analysis of the main methods will be carried out using publicly available EEG datasets.