MULTITASK LEARNING APPLIED TO SPATIAL FILTERING IN MOTOR IMAGERY BCI - A Preliminary Offline Study

Dieter Devlaminck, Bart Wyns, Georges Otte, Patrick Santens

2011

Abstract

Motor imagery based brain-computer interfaces (BCI) commonly use the common spatial pattern filter (CSP) as preprocessing step before feature extraction and classifiction. The CSP method is a supervised algorithm and therefore needs subject specific training data for calibration, which is very time consuming to collect. Instead of letting all that data and effort go to waste, the data of other subjects could be used to further improve results for new subjects. This problem setting is often encountered in multitask learning, from which we will borrow some ideas and apply it to the preprocessing phase. This paper outlines the details of the multitask CSP algorithm and shows some results on data from the third BCI competition. In some of the subjects a clear improvement can be seen by using information of other subjects, while in some subjects the algorithm determines that a specific model is the best. We also compare the use of a global filter, which is constructed only with data of other subjects, with the case where we ommit any form of spatial filtering. Here, the global filter seems to boost performance in four of the five subjects.

Paper Citation

in Harvard Style

Devlaminck D., Wyns B., Otte G. and Santens P. (2011). MULTITASK LEARNING APPLIED TO SPATIAL FILTERING IN MOTOR IMAGERY BCI - A Preliminary Offline Study . In Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011) ISBN 978-989-8425-35-5, pages 378-382. DOI: 10.5220/0003335403780382

in Bibtex Style

@conference{biosignals11,
author={Dieter Devlaminck and Bart Wyns and Georges Otte and Patrick Santens},
title={MULTITASK LEARNING APPLIED TO SPATIAL FILTERING IN MOTOR IMAGERY BCI - A Preliminary Offline Study},
booktitle={Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011)},
year={2011},
pages={378-382},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003335403780382},
isbn={978-989-8425-35-5},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Bio-inspired Systems and Signal Processing - Volume 1: BIOSIGNALS, (BIOSTEC 2011)
TI - MULTITASK LEARNING APPLIED TO SPATIAL FILTERING IN MOTOR IMAGERY BCI - A Preliminary Offline Study
SN - 978-989-8425-35-5
AU - Devlaminck D.
AU - Wyns B.
AU - Otte G.
AU - Santens P.
PY - 2011
SP - 378
EP - 382
DO - 10.5220/0003335403780382