Kernel dynamic policy programming: Applicable reinforcement learning to robot systems with high dimensional states.
Yunduan CuiTakamitsu MatsubaraKenji SugimotoPublished in: Neural Networks (2017)
Keyphrases
- reinforcement learning
- high dimensional
- optimal policy
- perceptual aliasing
- changing environment
- distributed systems
- feature space
- multi robot
- mobile robot
- multi agent
- machine learning
- kernel space
- markov decision problems
- control policies
- kernel principal component analysis
- markov decision process
- robot control
- real robot
- human robot interaction
- human users
- path planning
- support vector
- dynamic environments
- low dimensional
- dimensionality reduction