On the Sample Complexity of Actor-Critic Method for Reinforcement Learning with Function Approximation.
Harshat KumarAlec KoppelAlejandro RibeiroPublished in: CoRR (2019)
Keyphrases
- function approximation
- reinforcement learning
- model free
- actor critic
- function approximators
- machine learning
- policy gradient
- temporal difference
- gradient method
- support vector machine
- cost function
- support vector machine svm
- reinforcement learning algorithms
- basis functions
- radial basis function
- optimal policy
- dynamic programming
- learning process
- decision trees
- learning algorithm