Bayesian common spatial patterns with Dirichlet process priors for multi-subject EEG classification.
Hyohyeong KangSeungjin ChoiPublished in: IJCNN (2012)
Keyphrases
- dirichlet process
- motor imagery
- nonparametric bayesian
- variational bayesian
- gaussian processes
- eeg signals
- posterior distribution
- brain computer interface
- single trial
- mixture model
- bayesian model
- multi task learning
- eeg data
- training set
- bayesian framework
- bayesian methods
- feature space
- latent topics
- support vector
- variational inference
- sampling algorithm
- spatial filters
- posterior probability
- unsupervised learning
- model selection
- text classification
- support vector machine
- feature selection
- multi task
- machine learning
- high order
- maximum likelihood
- image classification
- active learning
- training data
- decision trees
- learning algorithm