Unsupervised Common Spatial Patterns - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Neural Systems and Rehabilitation Engineering Année : 2019

Unsupervised Common Spatial Patterns

Résumé

The common spatial pattern (CSP) method is a dimensionality reduction technique widely used in brain-computer interface (BCI) systems. In the two-class CSP problem, training data are linearly projected onto directions maximizing or minimizing the variance ratio between the two classes. The present contribution proves that kurtosis maximization performs CSP in an unsupervised manner, i.e., with no need for labeled data, when the classes follow Gaussian or elliptically symmetric distributions. Numerical analyses on synthetic and real data validate these findings in various experimental conditions, and demonstrate the interest of the proposed unsupervised approach.

Dates et versions

hal-02486801 , version 1 (21-02-2020)

Identifiants

Citer

Rubén Martín-Clemente, Javier Olias, Sergio Cruces, Vicente Zarzoso. Unsupervised Common Spatial Patterns. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2019, 27 (10), pp.2135-2144. ⟨10.1109/TNSRE.2019.2936411⟩. ⟨hal-02486801⟩
52 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More