{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,28]],"date-time":"2025-09-28T04:06:55Z","timestamp":1759032415722},"reference-count":47,"publisher":"MIT Press - Journals","issue":"2","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Neural Computation"],"published-print":{"date-parts":[[2014,2]]},"abstract":"<jats:p> Electroencephalographic signals are known to be nonstationary and easily affected by artifacts; therefore, their analysis requires methods that can deal with noise. In this work, we present a way to robustify the popular common spatial patterns (CSP) algorithm under a maxmin approach. In contrast to standard CSP that maximizes the variance ratio between two conditions based on a single estimate of the class covariance matrices, we propose to robustly compute spatial filters by maximizing the minimum variance ratio within a prefixed set of covariance matrices called the tolerance set. We show that this kind of maxmin optimization makes CSP robust to outliers and reduces its tendency to overfit. We also present a data-driven approach to construct a tolerance set that captures the variability of the covariance matrices over time and shows its ability to reduce the nonstationarity of the extracted features and significantly improve classification accuracy. We test the spatial filters derived with this approach and compare them to standard CSP and a state-of-the-art method on a real-world brain-computer interface (BCI) data set in which we expect substantial fluctuations caused by environmental differences. Finally we investigate the advantages and limitations of the maxmin approach with simulations. <\/jats:p>","DOI":"10.1162\/neco_a_00544","type":"journal-article","created":{"date-parts":[[2013,11,9]],"date-time":"2013-11-09T01:04:52Z","timestamp":1383959092000},"page":"349-376","source":"Crossref","is-referenced-by-count":29,"title":["Robust Common Spatial Filters with a Maxmin Approach"],"prefix":"10.1162","volume":"26","author":[{"given":"Motoaki","family":"Kawanabe","sequence":"first","affiliation":[{"name":"ATR Brain Information Communication Research Laboratory Group, Soraku-gun, Kyoto 619-0288, Japan"}]},{"given":"Wojciech","family":"Samek","sequence":"additional","affiliation":[{"name":"Department of Machine Learning Group, Berlin Institute of Technology, 10587 Berlin, Germany"}]},{"given":"Klaus-Robert","family":"M\u00fcller","sequence":"additional","affiliation":[{"name":"Department of Machine Learning Group, Berlin Institute of Technology, 10587 Berlin, Germany, and Department of Brain and Cognitive Engineering, Korea University, Seoul 136-713, Korea"}]},{"given":"Carmen","family":"Vidaurre","sequence":"additional","affiliation":[{"name":"Department of Machine Learning Group, Berlin Institute of Technology, 10587 Berlin, Germany"}]}],"member":"281","reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2013.2239310"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2007.01.051"},{"key":"B3","volume-title":"Advances in neural information processing systems, 20","author":"Blankertz B.","year":"2008"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2008.923152"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2010.03.022"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2008.4408441"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1109\/TNSRE.2006.875555"},{"key":"B9","volume-title":"Handbook of brain theory and neural networks","author":"del R. Mill\u00e1n J.","year":"2003"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7493.001.0001"},{"key":"B11","first-page":"1097","volume":"8","author":"Dyrholm M.","year":"2007","journal-title":"Journal of Machine Learning Research"},{"key":"B12","volume-title":"Introduction to statistical pattern recognition","author":"Fukunaga K.","year":"1990"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2010.04.265"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1109\/LSP.2009.2022557"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-03889-1_75"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1109\/IEMBS.2009.5334786"},{"key":"B17","first-page":"659","volume-title":"Advances in neural information processing systems","volume":"18","author":"Kim S.-J.","year":"2006"},{"key":"B19","first-page":"801","volume":"5","author":"Laub J.","year":"2004","journal-title":"Journal of Machine Learning"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1016\/S0047-259X(03)00096-4"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2010.11.004"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.1162\/neco.2006.18.11.2730"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2010.5495183"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2082539"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2082540"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1109\/TNSRE.2009.2015197"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2013.2249294"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1109\/72.914517"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jneumeth.2007.09.022"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1109\/MIS.2008.41"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1016\/0024-3795(94)00348-3"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2005.05.032"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1016\/S1388-2457(99)00141-8"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2013.2253608"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1109\/EMBC.2012.6346563"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/9\/2\/026013"},{"key":"B38","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2009.08.006"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/3\/1\/R02"},{"key":"B40","volume-title":"Machine learning in non-stationary environments: Introduction to covariate shift adaptation","author":"Sugiyama M.","year":"2011"},{"key":"B41","first-page":"985","volume":"8","author":"Sugiyama M.","year":"2007","journal-title":"Journal of Machine Learning Research"},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2009.07.045"},{"key":"B43","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2093133"},{"key":"B44","doi-asserted-by":"publisher","DOI":"10.1162\/NECO_a_00089"},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2006.873542"},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2006.888836"},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.103.214101"},{"key":"B47","doi-asserted-by":"publisher","DOI":"10.1109\/IEMBS.2010.5626537"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.1109\/IEMBS.2007.4353477"},{"key":"B50","doi-asserted-by":"publisher","DOI":"10.1016\/S1388-2457(02)00057-3"}],"container-title":["Neural Computation"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/NECO_a_00544","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:40:11Z","timestamp":1615585211000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/neco\/article\/26\/2\/349-376\/7950"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2014,2]]},"references-count":47,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2014,2]]}},"alternative-id":["10.1162\/NECO_a_00544"],"URL":"https:\/\/doi.org\/10.1162\/neco_a_00544","relation":{},"ISSN":["0899-7667","1530-888X"],"issn-type":[{"value":"0899-7667","type":"print"},{"value":"1530-888X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2014,2]]}}}