Application of reinforcement learning based on on-line EM algorithm to balancing of acrobot.
Junichiro YoshimotoShin IshiiMasa-aki SatoPublished in: Systems and Computers in Japan (2001)
Keyphrases
- em algorithm
- expectation maximization
- reinforcement learning
- maximum likelihood
- gaussian mixture model
- parameter estimation
- maximum likelihood estimation
- generative model
- gaussian mixture
- incomplete data
- mixture model
- likelihood function
- expectation maximisation
- probability density function
- log likelihood
- gibbs sampling
- parameter learning
- image processing
- maximum a posteriori
- machine learning
- hidden variables
- particle filter
- state space
- latent dirichlet