{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T05:43:45Z","timestamp":1790142225506,"version":"4.0.1"},"reference-count":33,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2017,4,6]],"date-time":"2017-04-06T00:00:00Z","timestamp":1491436800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2017,4,6]],"date-time":"2017-04-06T00:00:00Z","timestamp":1491436800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"bwHPC initiative","award":["INST 39\/963-1 FUGG"],"award-info":[{"award-number":["INST 39\/963-1 FUGG"]}]},{"name":"European Unions Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement","award":["657679"],"award-info":[{"award-number":["657679"]}]},{"DOI":"10.13039\/501100001659","name":"Deutsche Forschungsgemeinschaft","doi-asserted-by":"crossref","award":["EXC 1086"],"award-info":[{"award-number":["EXC 1086"]}],"id":[{"id":"10.13039\/501100001659","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Special Research Fund from Ghent University"},{"name":"BK21 program funded by Korean National Research Foundation","award":["2012-005741"],"award-info":[{"award-number":["2012-005741"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["J. Neural Eng."],"published-print":{"date-parts":[[2017,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:italic>Objective<\/jats:italic>\n                    . Brain-computer interfaces (BCI) based on event-related potentials (ERP) incorporate a decoder to classify recorded brain signals and subsequently select a control signal that drives a computer application. Standard supervised BCI decoders require a tedious calibration procedure prior to every session. Several unsupervised classification methods have been proposed that tune the decoder during actual use and as such omit this calibration. Each of these methods has its own strengths and weaknesses. Our aim is to improve overall accuracy of ERP-based BCIs without calibration.\n                    <jats:italic>Approach<\/jats:italic>\n                    . We consider two approaches for unsupervised classification of ERP signals. Learning from label proportions (LLP) was recently shown to be guaranteed to converge to a supervised decoder when enough data is available. In contrast, the formerly proposed expectation maximization (EM) based decoding for ERP\u2013BCI does not have this guarantee. However, while this decoder has high variance due to random initialization of its parameters, it obtains a higher accuracy faster than LLP when the initialization is good. We introduce a method to optimally combine these two unsupervised decoding methods, letting one method\u2019s strengths compensate for the weaknesses of the other and vice versa. The new method is compared to the aforementioned methods in a resimulation of an experiment with a visual speller.\n                    <jats:italic>Main results<\/jats:italic>\n                    . Analysis of the experimental results shows that the new method exceeds the performance of the previous unsupervised classification approaches in terms of ERP classification accuracy and symbol selection accuracy during the spelling experiment. Furthermore, the method shows less dependency on random initialization of model parameters and is consequently more reliable.\n                    <jats:italic>Significance<\/jats:italic>\n                    . Improving the accuracy and subsequent reliability of calibrationless BCIs makes these systems more appealing for frequent use.\n                  <\/jats:p>","DOI":"10.1088\/1741-2552\/aa6639","type":"journal-article","created":{"date-parts":[[2017,3,13]],"date-time":"2017-03-13T07:15:12Z","timestamp":1489389312000},"page":"036021","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":23,"title":["Improving zero-training brain-computer interfaces by mixing model estimators"],"prefix":"10.1088","volume":"14","author":[{"given":"T","family":"Verhoeven","sequence":"first","affiliation":[{"name":"Ghent University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"D","family":"H\u00fcbner","sequence":"additional","affiliation":[{"name":"Albert-Ludwigs-University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"M","family":"Tangermann","sequence":"additional","affiliation":[{"name":"Albert-Ludwigs-University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"K R","family":"M\u00fcller","sequence":"additional","affiliation":[{"name":"Korea University"},{"name":"Berlin Institute of Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J","family":"Dambre","sequence":"additional","affiliation":[{"name":"Ghent University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"P J","family":"Kindermans","sequence":"additional","affiliation":[{"name":"Berlin Institute of Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2017,4,6]]},"reference":[{"key":"jneaa6639bib001","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1146\/annurev.bb.02.060173.001105","type":"journal-article","article-title":"Toward direct brain-computer communication","volume":"2","author":"Vidal","year":"1973","journal-title":"Ann. Rev. Biophys. Bioeng."},{"key":"jneaa6639bib002","doi-asserted-by":"publisher","first-page":"510","DOI":"10.1016\/0013-4694(88)90149-6","type":"journal-article","article-title":"Talking off the top of your head: toward a mental prosthesis utilizing event-related brain potentials","volume":"70","author":"Farwell","year":"1988","journal-title":"Electroencephalogr. Clin. Neurophysiol."},{"key":"jneaa6639bib003","doi-asserted-by":"publisher","first-page":"4185","DOI":"10.1109\/IEMBS.2010.5627379","type":"other","article-title":"Two-dimensional auditory P300 speller with predictive text system","author":"H\u00f6hne","year":"2010"},{"key":"jneaa6639bib004","doi-asserted-by":"publisher","first-page":"19","DOI":"10.3389\/fnins.2010.00019","type":"journal-article","article-title":"A tactile P300 brain-computer interface","volume":"4","author":"Brouwer","year":"2010","journal-title":"Front. Neurosci."},{"key":"jneaa6639bib005","doi-asserted-by":"publisher","first-page":"R1","DOI":"10.1088\/1741-2560\/4\/2\/R01","type":"journal-article","article-title":"A review of classification algorithms for EEG-based brain-computer interfaces","volume":"4","author":"Lotte","year":"2007","journal-title":"J. Neural Eng."},{"key":"jneaa6639bib006","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1016\/j.jneumeth.2007.09.022","type":"journal-article","article-title":"Machine learning for real-time single-trial EEG-analysis: from brain-computer interfacing to mental state monitoring","volume":"167","author":"M\u00fcller","year":"2008","journal-title":"J. Neurosci. Methods"},{"key":"jneaa6639bib007","doi-asserted-by":"publisher","first-page":"387","DOI":"10.1016\/j.neuroimage.2010.11.004","type":"journal-article","article-title":"Introduction to machine learning for brain imaging","volume":"56","author":"Lemm","year":"2011","journal-title":"Neuroimage"},{"key":"jneaa6639bib008","doi-asserted-by":"publisher","first-page":"530","DOI":"10.3389\/fnins.2016.00530","type":"journal-article","article-title":"The Berlin brain-computer interface: progress beyond communication and control","volume":"10","author":"Blankertz","year":"2016","journal-title":"Front. Neurosci."},{"key":"jneaa6639bib009","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1016\/j.biopsycho.2014.07.014","type":"journal-article","article-title":"Effects of mental workload and fatigue on the P300, alpha and theta band power during operation of an ERP (P300) brain-computer interface","volume":"102","author":"K\u00e4thner","year":"2014","journal-title":"Biol. Psychol."},{"key":"jneaa6639bib010","doi-asserted-by":"publisher","first-page":"1305","DOI":"10.1016\/j.neunet.2009.06.003","type":"journal-article","article-title":"Subject-independent mental state classification in single trials","volume":"22","author":"Fazli","year":"2009","journal-title":"Neural Netw."},{"key":"jneaa6639bib011","doi-asserted-by":"publisher","first-page":"e2967","DOI":"10.1371\/journal.pone.0002967","type":"journal-article","article-title":"Towards zero training for brain-computer interfacing","volume":"3","author":"Krauledat","year":"2008","journal-title":"PloS one"},{"key":"jneaa6639bib012","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/8\/2\/025009","type":"journal-article","article-title":"Co-adaptive calibration to improve BCI efficiency","volume":"8","author":"Vidaurre","year":"2011","journal-title":"J. Neural Eng."},{"key":"jneaa6639bib013","doi-asserted-by":"publisher","first-page":"587","DOI":"10.1109\/TBME.2010.2093133","type":"journal-article","article-title":"Toward unsupervised adaptation of LDA for brain-computer interfaces","volume":"58","author":"Vidaurre","year":"2011","journal-title":"IEEE Trans. Biomed. Eng."},{"key":"jneaa6639bib014","doi-asserted-by":"publisher","first-page":"207","DOI":"10.1002\/9781119144977.ch10","type":"journal-article","article-title":"Adaptive methods in machine learning","author":"Clerc","year":"2016","journal-title":"Brain-Computer Interfaces 1: Foundations and Methods"},{"key":"jneaa6639bib015","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0033758","type":"journal-article","article-title":"A bayesian model for exploiting application constraints to enable unsupervised training of a P300-based BCI","volume":"7","author":"Kindermans","year":"2012","journal-title":"PloS one"},{"key":"jneaa6639bib016","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0102504","type":"journal-article","article-title":"True zero-training brain-computer interfacing-an online study","volume":"9","author":"Kindermans","year":"2014","journal-title":"PloS one"},{"key":"jneaa6639bib017","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0175856)","type":"journal-article","article-title":"Learning from label proportions in brain-computer interfaces: online unsupervised learning with guarantees","author":"H\u00fcbner","year":"2017","journal-title":"PloS one"},{"key":"jneaa6639bib018","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0175856","type":"preprint","article-title":"Learning from label proportions in brain-computer interfaces: online unsupervised learning with guarantees","author":"H\u00fcbner","year":"2017"},{"key":"jneaa6639bib019","first-page":"2349","type":"journal-article","article-title":"Estimating labels from label proportions","volume":"10","author":"Quadrianto","year":"2009","journal-title":"J. Mach. Learn. Res."},{"key":"jneaa6639bib020","doi-asserted-by":"publisher","first-page":"740","DOI":"10.1016\/j.neuroimage.2015.09.031","type":"journal-article","article-title":"Analyzing neuroimaging data with subclasses: a shrinkage approach","volume":"124","author":"H\u00f6hne","year":"2016","journal-title":"NeuroImage"},{"key":"jneaa6639bib021","doi-asserted-by":"publisher","first-page":"858","DOI":"10.1080\/01621459.1987.10478509","type":"journal-article","article-title":"A semiparametric approach to density estimation","volume":"82","author":"Olkin","year":"1987","journal-title":"J. Am. Stat. Assoc."},{"key":"jneaa6639bib022","author":"Bishop","year":"2007","edition":"1st edn","type":"book"},{"key":"jneaa6639bib023","doi-asserted-by":"publisher","first-page":"814","DOI":"10.1016\/j.neuroimage.2010.06.048","type":"journal-article","article-title":"Single-trial analysis and classification of ERP components\u2014a tutorial","volume":"56","author":"Blankertz","year":"2011","journal-title":"NeuroImage"},{"key":"jneaa6639bib024","article-title":"How do you like your P300 speller: adaptive, accurate and simple?","author":"Kindermans","year":"2011","type":"other"},{"key":"jneaa6639bib025","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1016\/S0047-259X(03)00096-4","type":"journal-article","article-title":"A well-conditioned estimator for large-dimensional covariance matrices","volume":"88","author":"Ledoit","year":"2004","journal-title":"J. Multivariate Anal."},{"key":"jneaa6639bib026","first-page":"1592","type":"journal-article","article-title":"Covariance shrinkage for autocorrelated data","author":"Bartz","year":"2014","journal-title":"Advances in Neural Information Processing System"},{"key":"jneaa6639bib027","doi-asserted-by":"publisher","first-page":"948","DOI":"10.1214\/aos\/1176342516","type":"journal-article","article-title":"On the asymptotic normality of the maximum-likelihood estimate when sampling from a stable distribution","volume":"1","author":"DuMouchel","year":"1973","journal-title":"Ann. Stat."},{"key":"jneaa6639bib028","first-page":"361","type":"other","article-title":"Estimation with quadratic loss","volume":"vol 1","author":"James","year":"1961"},{"key":"jneaa6639bib029","first-page":"119","type":"journal-article","article-title":"Optimized stimulation events for a visual erp bci","volume":"13","author":"Tangermann","year":"2011","journal-title":"Int. J. Bioelectromagn."},{"key":"jneaa6639bib030","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/12\/6\/066027","type":"journal-article","article-title":"Towards a symbiotic brain-computer interface: exploring the application-decoder interaction","volume":"12","author":"Verhoeven","year":"2015","journal-title":"J. Neural Eng."},{"key":"jneaa6639bib031","doi-asserted-by":"publisher","first-page":"R13","DOI":"10.1088\/1741-2560\/3\/1\/R02","type":"journal-article","article-title":"Towards adaptive classification for BCI","volume":"3","author":"Shenoy","year":"2006","journal-title":"J. Neural Eng."},{"key":"jneaa6639bib032","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.103.214101","type":"journal-article","article-title":"Finding stationary subspaces in multivariate time series","volume":"103","author":"Von B\u00fcnau","year":"2009","journal-title":"Phys. Rev. Lett."},{"key":"jneaa6639bib033","author":"Lehmann","year":"2004","type":"book"}],"container-title":["Journal of Neural Engineering"],"original-title":[],"link":[{"URL":"http:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"http:\/\/stacks.iop.org\/1741-2552\/14\/i=3\/a=036021\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/stacks.iop.org\/1741-2552\/14\/i=3\/a=036021?key=crossref.f53831daf43689043e5372f5bcad0033","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"http:\/\/stacks.iop.org\/1741-2552\/14\/i=3\/a=036021?key=crossref.f53831daf43689043e5372f5bcad0033","content-type":"text\/html","content-version":"vor","intended-application":"similarity-checking"},{"URL":"http:\/\/stacks.iop.org\/1741-2552\/14\/i=3\/a=036021\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"},{"URL":"http:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,25]],"date-time":"2026-07-25T13:17:04Z","timestamp":1784985424000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/aa6639"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,4,6]]},"references-count":33,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2017,4,6]]},"published-print":{"date-parts":[[2017,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/1741-2552\/aa6639","relation":{},"ISSN":["1741-2560","1741-2552"],"issn-type":[{"value":"1741-2560","type":"print"},{"value":"1741-2552","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,4,6]]},"assertion":[{"value":"Improving zero-training brain-computer interfaces by mixing model estimators","name":"article_title","label":"Article Title"},{"value":"Journal of Neural Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2017 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2016-12-17","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2017-03-13","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2017-04-06","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}