GeMM

Contexte :

A Brain-Computer Interface based on EEG is a system to send commands to a machine without using the motor system: only the brain signals are used. A major problem of current BCI systems is the need of a calibration before each use of the system.

Contributions de l'étudiant :

The goal of this internship is to device, implement and test a new transfer learning method in order to build a general machine-learning model (GeMM) that can be used 'out of the shelf', that is, without any training. The findings will be tested on a large corpus of data, including a number of open access BCI databases available at GIPSA-lab and elsewhere. The student will learn how to handle, pre-process and process a large amount of EEG-based BCI data. He will investigate domain adaptation methods, implying the acquisition of skills in linear algebra and signal processing methods. Finally, the student will learn how to employ employ machine learning methods in the field of BCI, both shallow (Elastic-Net Logistic Regression) and deep (Riemannian neural networks).