Contrastive Self-Supervised Learning for Motor Imagery: impact of the embedding size - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Contrastive Self-Supervised Learning for Motor Imagery: impact of the embedding size

Résumé

Due to the intra- and inter-individual variability of the electroencephalography (EEG) signals, brain-computer interfaces (BCI) require a daily user-specific calibration. This offline calibration step is necessary to set feature extraction, classification and pre-processing parameters. Yet, it is time consuming and might cause fatigue before the actual use of the BCI. Our goal is to reduce this time with a self-supervised classification method that achieves good detections with minimal calibration trials, for use in a motor imagery (MI)-based BCI that aims to enhance the rehabilitation of stroke patients. To process a small amount of labeled data, self-supervised learning (SSL) is currently the state-of-the-art method in the fields of vision and natural language processing [1], which makes it interesting to explore for EEG data.
Fichier principal
Vignette du fichier
BCIMeeting2023_MarissensCueva.pdf (1.04 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04126477 , version 1 (13-06-2023)

Identifiants

  • HAL Id : hal-04126477 , version 1

Citer

Valérie Marissens Cueva, Laurent Bougrain. Contrastive Self-Supervised Learning for Motor Imagery: impact of the embedding size. 10th International BCI Meeting Balancing Innovation and Translation, BCI Society, Jun 2023, Bruxelles, Belgium. ⟨hal-04126477⟩
58 Consultations
39 Téléchargements

Partager

Gmail Facebook X LinkedIn More