Data augmentation in Riemannian space for Brain-Computer Interfaces

Abstract : Brain-Computer Interfaces (BCI) try to interpret brain signals, such as EEG, to issue some command or to detect various cognitive state of subject. A strong limitation is that BCI tasks require a high concentration of the user, de facto limiting the length of experiment and the size of the dataset. Furthermore, several BCI paradigms depend on rare events, as for event-related potentials , also reducing the number of training examples available. A common strategy in machine learning when dealing with scarse data is called data augmentation; new samples are generated by applying chosen transformations on original dataset. In this contribution, we propose a scheme to adapt data augmentation in EEG-based BCI with a Riemannian standpoint: geometrical properties of EEG covariance matrix are taken into account to generate new training samples. Neural network are good candidates to benefit from such training scheme and a simple multi-layer perceptron offer good results. Experimental validation is conducted on two datasets: an SSVEP experiment with few training samples in each class and an error potential experiment with unbalanced classes (NER Kaggle competition).
Type de document :
Communication dans un congrès
ICML Workshop on Statistics, Machine Learning and Neuroscience (Stamlins 2015), Jul 2015, Lille, France. 2015
Liste complète des métadonnées

Littérature citée [28 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01225255
Contributeur : Bertrand Thirion <>
Soumis le : lundi 16 novembre 2015 - 08:49:02
Dernière modification le : lundi 9 avril 2018 - 12:22:35
Document(s) archivé(s) le : vendredi 28 avril 2017 - 06:28:10

Fichier

paper_209.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01225255, version 1

Citation

Emmanuel Kalunga, Sylvain Chevallier, Quentin Barthélemy. Data augmentation in Riemannian space for Brain-Computer Interfaces. ICML Workshop on Statistics, Machine Learning and Neuroscience (Stamlins 2015), Jul 2015, Lille, France. 2015. 〈hal-01225255〉

Partager

Métriques

Consultations de la notice

321

Téléchargements de fichiers

423