Approximate Bayes Optimal Policy Search using Neural Networks.
Michael CastronovoVincent François-LavetRaphaël FonteneauDamien ErnstAdrien CouëtouxPublished in: ICAART (2) (2017)
Keyphrases
- bayes optimal
- policy search
- neural network
- learning curve
- reinforcement learning
- continuous state
- dynamic programming
- version space
- artificial neural networks
- function approximators
- linear classifiers
- reinforcement learning algorithms
- radial basis function
- generalization error
- approximate solutions
- partially observable markov decision processes