A map approach for ℓq-norm regularized sparse parameter estimation using the EM algorithm.
Rodrigo CarvajalJuan C. AgüeroBoris I. GodoyDimitrios KatselisPublished in: MLSP (2015)
Keyphrases
- parameter estimation
- em algorithm
- maximum a posteriori
- expectation maximization
- maximum likelihood
- estimation problems
- mixture model
- map estimation
- maximum likelihood estimation
- generative model
- gaussian mixture model
- hyperparameters
- objective function
- posterior distribution
- expectation maximisation
- likelihood function
- least squares
- incomplete data
- probabilistic model
- approximate inference
- hidden variables
- log likelihood
- bayesian framework
- gaussian mixture
- probability density function
- parameter learning
- parameter estimation algorithm
- gibbs sampling
- maximum likelihood estimates
- structure learning
- density estimation
- unsupervised learning
- bayesian model selection
- markov chain monte carlo
- parameter space
- high dimensional