Compatible natural gradient policy search.
Joni PajarinenHong Linh ThaiRiad AkrourJan PetersGerhard NeumannPublished in: Mach. Learn. (2019)
Keyphrases
- policy search
- natural gradient
- policy gradient
- reinforcement learning
- gradient method
- optimal control
- function approximation
- reinforcement learning algorithms
- independent component analysis
- learning rate
- blind source separation
- single agent
- average reward
- dynamic programming
- reinforcement learning methods
- multi agent
- approximation methods
- real valued
- markov decision processes