Policy search with cross-entropy optimization of basis functions.
Lucian BusoniuDamien ErnstBart De SchutterRobert BabuskaPublished in: ADPRL (2009)
Keyphrases
- basis functions
- cross entropy
- policy search
- error function
- linear combination
- radial basis function
- log likelihood
- state space
- reinforcement learning
- maximum likelihood
- evaluation metrics
- activation function
- reinforcement learning algorithms
- learning algorithm
- markov chain
- machine learning
- function approximation
- reward function
- search algorithm