Meta-Gradient Reinforcement Learning.
Zhongwen XuHado van HasseltDavid SilverPublished in: NeurIPS (2018)
Keyphrases
- reinforcement learning
- policy gradient
- function approximation
- reinforcement learning algorithms
- state space
- model free
- temporal difference
- robotic control
- dynamic programming
- meta level
- optimal policy
- edge detection
- optimal control
- image gradient
- markov decision processes
- transfer learning
- data sets
- gradient orientation
- gradient information
- temporal difference learning
- database
- evaluation function
- supervised learning
- multi agent
- case study
- data mining
- real time