Applying Continuous Action Reinforcement Learning Automata(CARLA) to Global Training of Hidden Markov Models.
Jahanshah KabudianMohammad Reza MeybodiMohammad Mehdi HomayounpourPublished in: ITCC (2) (2004)
Keyphrases
- hidden markov models
- learning automata
- reinforcement learning
- continuous action
- discriminative training
- baum welch
- continuous state
- policy search
- hidden state
- speech recognition
- conditional random fields
- markov models
- sequence classification
- genetic algorithm
- markov model
- sequential data
- supervised learning
- state space
- state transition
- neural network
- continuous hidden markov models
- hidden states
- minimum classification error
- action space
- partially observable markov decision processes
- learning algorithm
- machine learning
- transition probabilities
- markov decision processes