An EM algorithm for GMM parameter estimation in the presence of censored and dropped data with potential application for indoor positioning.
Trung Kien VuManh Kha HoangHung Lan LePublished in: ICT Express (2019)
Keyphrases
- em algorithm
- parameter estimation
- expectation maximization
- gaussian mixture model
- maximum likelihood
- incomplete data
- mixture model
- generative model
- gaussian mixture
- maximum likelihood estimation
- parameter estimates
- likelihood function
- hyperparameters
- missing data
- probability density function
- model selection
- data analysis
- parameter estimation algorithm
- approximate inference
- least squares
- probability distribution
- mixture components
- gibbs sampling
- probabilistic model
- bayesian model selection
- parameter learning
- maximum a posteriori
- markov random field