Reinforcement learning based on local state feature learning and policy adjustment.
Ya-Ping LinXue-Yong LiPublished in: Inf. Sci. (2003)
Keyphrases
- reinforcement learning
- partially observable
- learning problems
- learning process
- state space
- policy search
- learning algorithm
- state action
- active learning
- supervised learning
- online learning
- learning tasks
- machine learning
- action space
- function approximation
- learning systems
- knowledge acquisition
- optimal policy
- markov decision process
- transfer learning
- action selection
- robot control
- feature vectors
- feature selection
- learning agents
- markov decision problems
- continuous state
- autonomous learning
- neural network
- agent learns
- eligibility traces