Unsupervised learning in a BCI chess application using label proportions and expectation-maximization

References

• Wolpaw JR , Birbaumer N , McFarland DJ , et al. Brain–computer interfaces for communication and control. Clin Neurophysiol. 2002;113(6):767–791.
   PubMed Web of Science ®Google Scholar
• Kindermans P-J , Schreuder M , Schrauwen B , et al. True zero-training brain-computer interfacing – an online study. Plos One. 2014; 9(7):e102504.
   PubMed Web of Science ®Google Scholar
• Shenoy P , Krauledat M , Blankertz B , et al. Towards adaptive classification for BCI. J Neural Eng. 2006;3(1):R13–R23.
   PubMed Web of Science ®Google Scholar
• Nicolas-Alonso LF , Gomez-Gil J. Brain computer interfaces, a review. Sensors. 2012;12(2):1211–1279.
   PubMed Web of Science ®Google Scholar
• Iturrate I , Chavarriaga R , Montesano L , et al. Teaching brain-machine interfaces as an alternative paradigm to neuroprosthetics control. Sci Rep. 2015;5(13893):1–10.
   Google Scholar
• Zeyl T , Yin E , Keightley M , et al. Partially supervised P300 speller adaptation for eventual stimulus timing optimization: target confidence is superior to error-related potential score as an uncertain label. J Neural Eng. 2016;13(2):026008.
   PubMed Web of Science ®Google Scholar
• Zander TO , Krol LR , Birbaumer NP , et al. Neuroadaptive technology enables implicit cursor control based on medial prefrontal cortex activity. Proc Nat Acad Sci. 2016;113(52):14 898–14 903.
   Web of Science ®Google Scholar
• Chavarriaga R , Iturrate I , Millán JDR . Robust, accurate spelling based on error-related potentials. Proceedings of the 6th International Brain-Computer Interface Meeting; Asilomar, USA: Verlag der Technischen Universität Graz; 2016. p. 15.
   Google Scholar
• Thielen J , Broek PVD , Farquhar J , et al. Broad-band visually evoked potentials: re(con)volution in brain-computer interfacing. Plos One. 2015;10(7):e0133797.
   PubMed Web of Science ®Google Scholar
• Kindermans P-J , Verstraeten D , Schrauwen B . A bayesian model for exploiting application constraints to enable unsupervised training of a P300-based BCI. Plos One. 2012;7(4):e33758.
   PubMed Web of Science ®Google Scholar
• Hübner D , Verhoeven T , Schmid K , et al. Learning from label proportions in brain-computer interfaces: online unsupervised learning with guarantees. Plos One. 2017;12(4):e0175856.
   PubMed Web of Science ®Google Scholar
• Verhoeven T , Hübner D , Tangermann M , et al. Improving zero-training brain-computer interfaces by mixing model estimators. J Neural Eng. 2017;14(3):036021.
   PubMed Web of Science ®Google Scholar
• Hübner D , Verhoeven T , Müller K-R , et al. Unsupervised learning for brain-computer interfaces based on event-related potentials: review and online comparison. IEEE Comput Intel Mag. 2018;13(2):66–77.
   Web of Science ®Google Scholar
• Kindermans P-J , Tangermann M , Müller K-R , et al. Integrating dynamic stopping, transfer learning and language models in an adaptive zero-training ERP speller. J Neural Eng. 2014;11(3):035005.
   PubMed Web of Science ®Google Scholar
• Hübner D , Kindermans P-J , Verhoeven T , et al. Improving learning from label proportions by reducing the feature dimensionality. In: Müller-Putz GR , Steyrl D , Wriessnegger SC , et al. editors. Proceedings of the 7th International Brain-Computer Interface Meeting 2017: from Vision to Reality; Graz: Verlag der Technischen Universität Graz; 2017. p. 186–191.
   Google Scholar
• Hübner D , Schall A , Tangermann M , Two player online brain-controlled chess. Proc. 41th Int. Conf. of the IEEE Eng. in Medicine and Biology Soc. (EMBC); Berlin: IEEE; 2019. p. 3019–3021.
   Google Scholar
• Blankertz B , Tangermann M , Vidaurre C , et al. The Berlin brain-computer interface: non-medical uses of BCI technology. Front Neurosci. 2010;4:198.
   PubMed Web of Science ®Google Scholar
• Blankertz B , Lemm S , Treder M , et al. Single-trial analysis and classification of ERP components, a tutorial. NeuroImage. 2011;56(2):814–825.
   PubMed Web of Science ®Google Scholar
• Lotte F , Bougrain L , Cichocki A , et al. A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update. J Neural Eng. 2018;15(3):031005.
   PubMed Web of Science ®Google Scholar
• Bishop CM . Pattern recognition and machine learning. Vol. 128. New York: Springer; 2006.
   Google Scholar
• Ledoit O , Wolf M . A well-conditioned estimator for large-dimensional covariance matrices. J Multivariate Anal. 2004;88(2):365–411.
   Web of Science ®Google Scholar
• Montgomery DC , Peck EA , Vining GG . Introduction to linear regression analysis. Vol. 821. Hoboken (NJ): John Wiley & Sons; 2012.
   Google Scholar
• Verhoeven T , Brain-computer interfaces with machine learning: a symbiotic approach [Ph.D. dissertation]. Ghent, Belgium: Ghent University; 2017.
   Google Scholar
• Ma J , Xu L , Jordan MI . Asymptotic convergence rate of the EM algorithm for gaussian mixtures. Neural Comput. 2000;12(12):2881–2907.
   PubMed Web of Science ®Google Scholar
• Lu S , Guan C , Zhang H . Unsupervised brain computer interface based on intersubject information and online adaptation. IEEE Trans Neural Syst Rehabil Eng. 2009;17(2):135–145.
   PubMed Web of Science ®Google Scholar
• Kindermans P-J , Verstraeten D , Buteneers P , et al. How do you like your P300 speller: adaptive, accurate and simple? In 5th International Brain-Computer Interface Conference (BCI-2011); Graz, Austria; 2011. p. 96–99.
   Google Scholar
• Panicker RC , Puthusserypady S , Sun Y . Adaptation in P300 brain–computer interfaces: A two-classifier cotraining approach. IEEE Trans Biomed Eng. 2010;57(12):2927–2935.
   PubMed Web of Science ®Google Scholar
• Vidaurre C , Kawanabe M , von Bünau P , et al. Toward unsupervised adaptation of LDA for brain–computer interfaces. IEEE Trans Biomed Eng. 2010;58(3):587–597.
   PubMed Web of Science ®Google Scholar
• Dähne S , Höhne J , Tangermann M . Adaptive classification improves control performance in ERP-based BCIs. Proceedings of the 5th International Brain-Computer Interface Conference; Graz, Austria; 2011. pp. 92–95.
   Google Scholar
• Barachant A , Congedo M , A plug&play P300 BCI using information geometry. arXiv preprint:1409.0107, 2014.
   Google Scholar
• Bolagh SNG , Shamsollahi MB , Jutten C , et al. Unsupervised cross-subject BCI learning and classification using Riemannian geometry. 24th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2016); Belgium: Bruges; 2016.
   Google Scholar
• Grizou J , Iturrate I , Montesano L , et al. Calibration-free BCI based control. Twenty-Eighth AAAI Conference on Artificial Intelligence; Quebec, Canada; 2014; p. 1213–1220.
   Google Scholar
• Iturrate I , Grizou J , Omedes J , et al. Exploiting task constraints for self-calibrated brain-machine interface control using error-related potentials. Plos One. 2015;10(7):e0131491.
   PubMed Web of Science ®Google Scholar
• Bensch M , Karim AA , Mellinger J , et al. Nessi: an EEG-controlled web browser for severlely paralyzed patients. Comput Intell Neurosci. 2007;2007:e71863.
   Google Scholar
• Halder S , Pinegger A , Käthner I , et al. Brain-controlled applications using dynamic P300 speller matrices. Artif Intell Med. 2015;63(1):7–17.
   PubMed Web of Science ®Google Scholar
• Congedo M , Goyat M , Tarrin N , et al. “Brain invaders”: a prototype of an open-source p300-based video game working with the openViBE platform. in 5th International Brain-Computer Interface Conference 2011; Graz, Austria; 2011. p. 280–283.
   Google Scholar
• Ruchkin DS . An analysis of average response computations based upon aperiodic stimuli. IEEE Trans Biomed Eng. 1965;BME-12(2):87–94.
   Google Scholar
• Congedo M , Korczowski L , Delorme A , et al. Spatio-temporal common pattern: a companion method for ERP analysis in the time domain. J Neurosci Methods. 2016;267:74–88.
   PubMed Web of Science ®Google Scholar