On the On-line Learning Algorithms for EEG Signal Classification in Brain Computer Interfaces

• Published in: Fuzzy Systems and Knowledge Discovery pp. 638 - 647
• Main Authors: Sun, Shiliang, Zhang, Changshui, Lu, Naijiang
• Format: Book Chapter Conference Proceeding
• Language: English
• Published: Berlin, Heidelberg Springer Berlin Heidelberg 2005; Springer
• Series: Lecture Notes in Computer Science
• Subjects: Applied sciences; Brain Computer Interface; Computer science; control theory; systems; Exact sciences and technology; Gaussian Mixture Model; Gaussian Probability Density Function; Little Mean Square; Little Mean Square Algorithm; Bayes estimation; Human; Brain; Density estimation; Gradient; Central nervous system; Updating; Electroencephalography; Encephalon; Time varying system; Least squares method; Classification; Occupation time; Decorrelation; Learning algorithm; Artificial intelligence
• ISBN: 3540283315; 9783540283317; 3540283129; 9783540283126
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/11540007_79

The on-line update of classifiers is an important concern for categorizing the time-varying neurophysiological signals used in brain computer interfaces, e.g. classification of electroencephalographic (EEG) signals. However, up to the present there is not much work dealing with this issue. In this paper, we propose to use the idea of gradient decorrelation to develop the existent basic Least Mean Square (LMS) algorithm for the on-line learning of Bayesian classifiers employed in brain computer interfaces. Under the framework of Gaussian mixture model, we give the detailed representation of Decorrelated Least Mean Square (DLMS) algorithm for updating Bayesian classifiers. Experimental results of off-line analysis for classification of real EEG signals show the superiority of the on-line Bayesian classifier using DLMS algorithm to that using LMS algorithm.