Artículo
Unsupervised Common Spatial Patterns
Autor/es | Martín Clemente, Rubén
Olías Sánchez, Francisco Javier Cruces Álvarez, Sergio Antonio Antonio Zarzoso, Vicente |
Departamento | Universidad de Sevilla. Departamento de Teoría de la Señal y Comunicaciones |
Fecha de publicación | 2019 |
Fecha de depósito | 2022-03-28 |
Publicado en |
|
Resumen | 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 direc tions ... 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 direc tions maximizing or minimizing the variance ratio between the two classes. The present contribution proves that kurto sis 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. |
Agencias financiadoras | European Commission (EC). Fondo Europeo de Desarrollo Regional (FEDER) Laboratorio I3S |
Identificador del proyecto | TEC2017-82807-P |
Cita | Martín Clemente, R., Olías Sánchez, F.J., Cruces Álvarez, S. y Antonio Zarzoso, V. (2019). Unsupervised Common Spatial Patterns. IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING, 27 (10), 2135-2144. |
Ficheros | Tamaño | Formato | Ver | Descripción |
---|---|---|---|---|
Unsupervised_Common_Spatial_Pa ... | 1.644Mb | [PDF] | Ver/ | |