Multiple kernel learning for brain-computer interfacing

Wojciech Samek, Alexander Binder, Klaus Muller

Research output: Chapter in Book/Report/Conference proceedingConference contribution

10 Citations (Scopus)

Abstract

Combining information from different sources is a common way to improve classification accuracy in Brain-Computer Interfacing (BCI). For instance, in small sample settings it is useful to integrate data from other subjects or sessions in order to improve the estimation quality of the spatial filters or the classifier. Since data from different subjects may show large variability, it is crucial to weight the contributions according to importance. Many multi-subject learning algorithms determine the optimal weighting in a separate step by using heuristics, however, without ensuring that the selected weights are optimal with respect to classification. In this work we apply Multiple Kernel Learning (MKL) to this problem. MKL has been widely used for feature fusion in computer vision and allows to simultaneously learn the classifier and the optimal weighting. We compare the MKL method to two baseline approaches and investigate the reasons for performance improvement.

Original languageEnglish
Title of host publicationProceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Pages7048-7051
Number of pages4
DOIs
Publication statusPublished - 2013 Oct 31
Event2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013 - Osaka, Japan
Duration: 2013 Jul 32013 Jul 7

Other

Other2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013
CountryJapan
CityOsaka
Period13/7/313/7/7

Fingerprint

Brain
Classifiers
Learning
Learning algorithms
Computer vision
Fusion reactions
Weights and Measures

ASJC Scopus subject areas

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

Cite this

Samek, W., Binder, A., & Muller, K. (2013). Multiple kernel learning for brain-computer interfacing. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (pp. 7048-7051). [6611181] https://doi.org/10.1109/EMBC.2013.6611181

Multiple kernel learning for brain-computer interfacing. / Samek, Wojciech; Binder, Alexander; Muller, Klaus.

Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2013. p. 7048-7051 6611181.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Samek, W, Binder, A & Muller, K 2013, Multiple kernel learning for brain-computer interfacing. in Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS., 6611181, pp. 7048-7051, 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013, Osaka, Japan, 13/7/3. https://doi.org/10.1109/EMBC.2013.6611181
Samek W, Binder A, Muller K. Multiple kernel learning for brain-computer interfacing. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2013. p. 7048-7051. 6611181 https://doi.org/10.1109/EMBC.2013.6611181
Samek, Wojciech ; Binder, Alexander ; Muller, Klaus. / Multiple kernel learning for brain-computer interfacing. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2013. pp. 7048-7051
@inproceedings{4f789b656f1f4035aac8c0654ccb6fe6,
title = "Multiple kernel learning for brain-computer interfacing",
abstract = "Combining information from different sources is a common way to improve classification accuracy in Brain-Computer Interfacing (BCI). For instance, in small sample settings it is useful to integrate data from other subjects or sessions in order to improve the estimation quality of the spatial filters or the classifier. Since data from different subjects may show large variability, it is crucial to weight the contributions according to importance. Many multi-subject learning algorithms determine the optimal weighting in a separate step by using heuristics, however, without ensuring that the selected weights are optimal with respect to classification. In this work we apply Multiple Kernel Learning (MKL) to this problem. MKL has been widely used for feature fusion in computer vision and allows to simultaneously learn the classifier and the optimal weighting. We compare the MKL method to two baseline approaches and investigate the reasons for performance improvement.",
author = "Wojciech Samek and Alexander Binder and Klaus Muller",
year = "2013",
month = "10",
day = "31",
doi = "10.1109/EMBC.2013.6611181",
language = "English",
isbn = "9781457702167",
pages = "7048--7051",
booktitle = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

}

TY - GEN

T1 - Multiple kernel learning for brain-computer interfacing

AU - Samek, Wojciech

AU - Binder, Alexander

AU - Muller, Klaus

PY - 2013/10/31

Y1 - 2013/10/31

N2 - Combining information from different sources is a common way to improve classification accuracy in Brain-Computer Interfacing (BCI). For instance, in small sample settings it is useful to integrate data from other subjects or sessions in order to improve the estimation quality of the spatial filters or the classifier. Since data from different subjects may show large variability, it is crucial to weight the contributions according to importance. Many multi-subject learning algorithms determine the optimal weighting in a separate step by using heuristics, however, without ensuring that the selected weights are optimal with respect to classification. In this work we apply Multiple Kernel Learning (MKL) to this problem. MKL has been widely used for feature fusion in computer vision and allows to simultaneously learn the classifier and the optimal weighting. We compare the MKL method to two baseline approaches and investigate the reasons for performance improvement.

AB - Combining information from different sources is a common way to improve classification accuracy in Brain-Computer Interfacing (BCI). For instance, in small sample settings it is useful to integrate data from other subjects or sessions in order to improve the estimation quality of the spatial filters or the classifier. Since data from different subjects may show large variability, it is crucial to weight the contributions according to importance. Many multi-subject learning algorithms determine the optimal weighting in a separate step by using heuristics, however, without ensuring that the selected weights are optimal with respect to classification. In this work we apply Multiple Kernel Learning (MKL) to this problem. MKL has been widely used for feature fusion in computer vision and allows to simultaneously learn the classifier and the optimal weighting. We compare the MKL method to two baseline approaches and investigate the reasons for performance improvement.

UR - http://www.scopus.com/inward/record.url?scp=84886483003&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84886483003&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2013.6611181

DO - 10.1109/EMBC.2013.6611181

M3 - Conference contribution

C2 - 24111368

AN - SCOPUS:84886483003

SN - 9781457702167

SP - 7048

EP - 7051

BT - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

ER -