Reinforcement Learning for POMDP Using State Classification.
Le Tien DungTakashi KomedaMotoki TakagiPublished in: MLMTA (2007)
Keyphrases
- reinforcement learning
- state space
- partially observable
- machine learning
- hidden state
- markov decision processes
- support vector machine
- image classification
- supervised learning
- pattern recognition
- classification accuracy
- markov decision process
- classification algorithm
- action space
- feature extraction
- text classification
- support vector machine svm
- optimal policy
- feature space
- function approximation
- classification method
- belief state
- model free reinforcement learning
- decision trees
- feature selection
- continuous state
- model free
- support vector
- neural network
- feature set
- multi agent
- control policy
- dynamic programming
- transition model
- feature vectors