Fast training of recurrent networks based on the EM algorithm.
Sheng MaChuanyi JiPublished in: IEEE Trans. Neural Networks (1998)
Keyphrases
- em algorithm
- recurrent networks
- expectation maximization
- recurrent neural networks
- mixture model
- maximum likelihood
- parameter estimation
- feed forward
- biologically inspired
- maximum likelihood estimation
- gaussian mixture model
- generative model
- neural network
- gaussian mixture
- expectation maximisation
- incomplete data
- log likelihood
- hidden variables
- maximum a posteriori
- model based clustering
- likelihood function
- mixture modeling
- computer vision
- probability density function
- maximum likelihood estimates
- parameter learning
- density estimation
- pairwise
- log likelihood function
- finite mixture models
- likelihood maximization