Identification of nonlinear state space models using an MLP network trained by the EM algorithm.
Aliakbar A. GorjiMohammad Bagher MenhajPublished in: IJCNN (2008)
Keyphrases
- em algorithm
- parameter estimation
- expectation maximization
- mixture model
- state space
- maximum likelihood
- hidden variables
- expectation maximisation
- finite mixture model
- incomplete data
- multilayer perceptron
- gaussian mixture model
- multi layered perceptron
- probabilistic model
- maximum likelihood estimation
- likelihood function
- parameter learning
- multi layer perceptron
- probability density function
- density estimation
- bayesian framework
- statistical model
- prior knowledge
- radial basis function network
- probabilistic principal component analysis
- neural network
- artificial neural networks
- log likelihood
- gaussian mixture
- training set
- reinforcement learning
- penalized likelihood
- image processing