Electroencephalography (EEG) is an electrical signal captured by electrodes, representing the brain activity of a subject.

In recent years, Deep Learning techniques have been extensively studied for analyzing such signals in various applications, including Motor Imagery (MI) decoding. MI decoding can be used to control applications, or to aid in patient rehabilitation.

One of the main challenge to apply these algorithms in practice is the cost of model calibration, as they often struggle to generalize to unseen subjects.

In this context, this presentation will introduce some techniques to represent EEG using embeddings, and will discuss the potential benefits of such approach in the context of EEG analysis.

Pay attention :

This seminar will also take place in the form of a video conference

Sandwiches will be provided. Please fill in the form before day D at 09:00 to reserve a sandwich