A Bayesian Approach to Blind Separation of Mixed Discrete Sources by Gibbs Sampling.
Hang ZhangFanglin GuYi XiaoPublished in: UIC (2011)
Keyphrases
- blind separation
- gibbs sampling
- blind source separation
- parameter estimation
- topic models
- em algorithm
- approximate inference
- markov chain
- latent dirichlet allocation
- belief networks
- expectation maximization
- independent component analysis
- markov chain monte carlo
- exact inference
- graphical models
- source signals
- computer vision
- latent variables
- artificial neural networks
- prior knowledge
- probability distribution
- hyperparameters
- belief propagation
- non stationary
- state space
- generative model
- maximum likelihood