Brain-computer interfacing in discriminative and stationary subspaces.

Wojciech Samek, Klaus Robert Muller, Motoaki Kawanabe, Carmen Vidaurre

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

The non-stationary nature of neurophysiological measurements, e.g. EEG, makes classification of motion intentions a demanding task. Variations in the underlying brain processes often lead to significant and unexpected changes in the feature distribution resulting in decreased classification accuracy in Brain Computer Interfacing (BCI). Several methods were developed to tackle this problem by either adapting to these changes or extracting features that are invariant. Recently, a method called Stationary Subspace Analysis (SSA) was proposed and applied to BCI data. It diminishes the influence of non-stationary changes as learning and classification is performed in a stationary subspace of the data which can be extracted by SSA. In this paper we extend this method in two ways. First we propose a variant of SSA that allows to extract stationary subspaces from labeled data without disregarding class-related variations or treating class-differences as non-stationarities. Second we propose a discriminant variant of SSA that trades-off stationarity and discriminativity, thus it allows to extract stationary subspaces without losing relevant information. We show that learning in a discriminative and stationary subspace is advantageous for BCI application and outperforms the standard SSA method.

Original languageEnglish
Pages (from-to)2873-2876
Number of pages4
JournalConference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
Volume2012
Publication statusPublished - 2012 Dec 1

Fingerprint

Brain
Learning
Computer applications
Electroencephalography

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering
  • Health Informatics

Cite this

Brain-computer interfacing in discriminative and stationary subspaces. / Samek, Wojciech; Muller, Klaus Robert; Kawanabe, Motoaki; Vidaurre, Carmen.

In: Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, Vol. 2012, 01.12.2012, p. 2873-2876.

Research output: Contribution to journalArticle

@article{f104cfa57c664412a98543766e8dfeb2,
title = "Brain-computer interfacing in discriminative and stationary subspaces.",
abstract = "The non-stationary nature of neurophysiological measurements, e.g. EEG, makes classification of motion intentions a demanding task. Variations in the underlying brain processes often lead to significant and unexpected changes in the feature distribution resulting in decreased classification accuracy in Brain Computer Interfacing (BCI). Several methods were developed to tackle this problem by either adapting to these changes or extracting features that are invariant. Recently, a method called Stationary Subspace Analysis (SSA) was proposed and applied to BCI data. It diminishes the influence of non-stationary changes as learning and classification is performed in a stationary subspace of the data which can be extracted by SSA. In this paper we extend this method in two ways. First we propose a variant of SSA that allows to extract stationary subspaces from labeled data without disregarding class-related variations or treating class-differences as non-stationarities. Second we propose a discriminant variant of SSA that trades-off stationarity and discriminativity, thus it allows to extract stationary subspaces without losing relevant information. We show that learning in a discriminative and stationary subspace is advantageous for BCI application and outperforms the standard SSA method.",
author = "Wojciech Samek and Muller, {Klaus Robert} and Motoaki Kawanabe and Carmen Vidaurre",
year = "2012",
month = "12",
day = "1",
language = "English",
volume = "2012",
pages = "2873--2876",
journal = "The BMJ",
issn = "0730-6512",
publisher = "Kluwer Academic Publishers",

}

TY - JOUR

T1 - Brain-computer interfacing in discriminative and stationary subspaces.

AU - Samek, Wojciech

AU - Muller, Klaus Robert

AU - Kawanabe, Motoaki

AU - Vidaurre, Carmen

PY - 2012/12/1

Y1 - 2012/12/1

N2 - The non-stationary nature of neurophysiological measurements, e.g. EEG, makes classification of motion intentions a demanding task. Variations in the underlying brain processes often lead to significant and unexpected changes in the feature distribution resulting in decreased classification accuracy in Brain Computer Interfacing (BCI). Several methods were developed to tackle this problem by either adapting to these changes or extracting features that are invariant. Recently, a method called Stationary Subspace Analysis (SSA) was proposed and applied to BCI data. It diminishes the influence of non-stationary changes as learning and classification is performed in a stationary subspace of the data which can be extracted by SSA. In this paper we extend this method in two ways. First we propose a variant of SSA that allows to extract stationary subspaces from labeled data without disregarding class-related variations or treating class-differences as non-stationarities. Second we propose a discriminant variant of SSA that trades-off stationarity and discriminativity, thus it allows to extract stationary subspaces without losing relevant information. We show that learning in a discriminative and stationary subspace is advantageous for BCI application and outperforms the standard SSA method.

AB - The non-stationary nature of neurophysiological measurements, e.g. EEG, makes classification of motion intentions a demanding task. Variations in the underlying brain processes often lead to significant and unexpected changes in the feature distribution resulting in decreased classification accuracy in Brain Computer Interfacing (BCI). Several methods were developed to tackle this problem by either adapting to these changes or extracting features that are invariant. Recently, a method called Stationary Subspace Analysis (SSA) was proposed and applied to BCI data. It diminishes the influence of non-stationary changes as learning and classification is performed in a stationary subspace of the data which can be extracted by SSA. In this paper we extend this method in two ways. First we propose a variant of SSA that allows to extract stationary subspaces from labeled data without disregarding class-related variations or treating class-differences as non-stationarities. Second we propose a discriminant variant of SSA that trades-off stationarity and discriminativity, thus it allows to extract stationary subspaces without losing relevant information. We show that learning in a discriminative and stationary subspace is advantageous for BCI application and outperforms the standard SSA method.

UR - http://www.scopus.com/inward/record.url?scp=84882949036&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84882949036&partnerID=8YFLogxK

M3 - Article

C2 - 23366524

AN - SCOPUS:84882949036

VL - 2012

SP - 2873

EP - 2876

JO - The BMJ

JF - The BMJ

SN - 0730-6512

ER -