Self-adapting Bci Based on Unsupervised Learning
نویسنده
چکیده
This paper adopts a simple but effective unsupervised method for incrementally updating the means and variances that define LDA and Bayesian classifiers for real-time BCI. The method is evaluated using asynchronous BCI data from three subjects. Experimental results show that the proposed selfadaptation approach is stable and able to improve BCI performance consistently. This paper also aims to clarify the confusion between BCI that is capable of online training using true class labels and self-adapting BCI that is able to adapt without knowing true labels.
منابع مشابه
Co-adaptivity in Unsupervised Adaptive Brain-Computer Interfacing: a Simulation Approach
A Brain-Computer Interface (BCI) allows a user to control a computer by pure brain activity. Due to the nonstationarity of the recorded brain signals, the BCI performance tends to decrease over time. Recently, adaption of the BCI has been proposed as a means to counter non-stationarity and help to stabilize the BCI performance. Since most adaption methods for BCI are analysed in an offline sett...
متن کاملLearning from label proportions in brain-computer interfaces: Online unsupervised learning with guarantees
OBJECTIVE Using traditional approaches, a brain-computer interface (BCI) requires the collection of calibration data for new subjects prior to online use. Calibration time can be reduced or eliminated e.g., by subject-to-subject transfer of a pre-trained classifier or unsupervised adaptive classification methods which learn from scratch and adapt over time. While such heuristics work well in pr...
متن کاملHangman BCI: An unsupervised adaptive self-paced Brain-Computer Interface for playing games
This paper presents a novel user interface suitable for adaptive Brain Computer Interface (BCI) system. A customized self-paced BCI architecture is introduced where the system combines onset detection system along with an adaptive classifier working in parallel. An unsupervised adaptive method based on sequential expectation maximization for Gaussian mixture model is employed with new timing sc...
متن کاملAdaptive Methods Exploiting the Time Structure in EEG for Self-paced Brain-Computer Interfaces
Although self-paced Brain-Computer Interface (BCI) is desirable from the users’ point of view, it brings about great technological challenges. This thesis aims to provide novel solutions to several important aspects in developing a realistic self-paced Electroencephalogram (EEG) based BCI. A Sequential Floating Forward Search (SFFS) based method is developed for feature selection, followed by t...
متن کاملFeature Dimensionality Reduction by Manifold Learning in Brain-computer Interface Design
Unsupervised manifold learning for dimensionality reduction has drawn much attention in recent years. This paper applies two manifold learning methods for the first time to feature dimensionality reduction in brain-computer interface (BCI) design, and compares them with principal component analysis (PCA) and supervised PCA that is mathematically equivalent to the common spatial patterns (CSP) m...
متن کامل