State-Space Models: From the EM Algorithm to a Gradient Approach.
Rasmus Kongsgaard OlssonKaare Brandt PetersenTue Lehn-SchiølerPublished in: Neural Comput. (2007)
Keyphrases
- em algorithm
- parameter estimation
- expectation maximization
- state space
- mixture model
- maximum likelihood
- incomplete data
- hidden variables
- gaussian mixture model
- likelihood function
- finite mixture model
- log likelihood function
- expectation maximisation
- generative model
- log likelihood
- update equations
- density estimation
- maximum likelihood estimation
- model selection
- bayesian framework
- parameter learning
- maximum a posteriori
- probabilistic principal component analysis
- gaussian mixture
- probability density function
- least squares
- semi supervised
- high dimensional
- learning algorithm
- particle filter
- probabilistic model
- reinforcement learning