Kernel dynamic policy programming: Practical reinforcement learning for high-dimensional robots.
Yunduan CuiTakamitsu MatsubaraKenji SugimotoPublished in: Humanoids (2016)
Keyphrases
- reinforcement learning
- high dimensional
- optimal policy
- mobile robot
- policy search
- feature space
- kernel function
- function approximation
- state space
- input space
- machine learning
- multi robot
- markov decision processes
- dynamic environments
- low dimensional
- robot control
- action space
- policy evaluation
- cooperative
- real robot
- computer programming
- real world
- robotic control
- model free
- multi agent
- learning algorithm
- autonomous robots
- temporal difference
- robotic systems
- partially observable
- kernel methods
- programming language
- dynamic programming
- control policy
- control policies
- support vector
- actor critic
- hands on guide
- agent learns
- state and action spaces