Forward and Backward Bellman Equations Improve the Efficiency of the EM Algorithm for DEC-POMDP.
Takehiro TottoriTetsuya J. KobayashiPublished in: Entropy (2021)
Keyphrases
- em algorithm
- forward and backward
- expectation maximization
- maximum likelihood
- mixture model
- parameter estimation
- maximum likelihood estimation
- generative model
- gaussian mixture model
- expectation maximisation
- likelihood function
- log likelihood
- incomplete data
- mixture modeling
- maximum a posteriori
- gaussian mixture
- model based clustering
- finite mixture model
- log likelihood function
- penalized likelihood
- computational efficiency
- probabilistic model
- high dimensional
- density estimation
- probability density function
- markov chain
- markov random field
- reinforcement learning
- computer vision