ICA Mixture Models for Unsupervised Classification of Non-Gaussian Classes and Automatic Context Switching in Blind Signal Separation.
Te-Won LeeMichael S. LewickiTerrence J. SejnowskiPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2000)
Keyphrases
- mixture model
- blind signal separation
- unsupervised classification
- independent component analysis
- unsupervised learning
- mixing matrix
- blind source separation
- gaussian mixture model
- em algorithm
- supervised classification
- principal component analysis
- independent components
- expectation maximization
- generative model
- source signals
- factor analysis
- maximum likelihood
- probabilistic model
- data clustering
- signal processing
- model selection
- source separation
- clustering ensemble
- supervised learning
- machine learning
- information retrieval
- reinforcement learning
- hyperspectral images
- face recognition
- clustering method
- semi supervised learning
- feature extraction
- neural network
- language model
- dimensionality reduction