Policy evaluation with temporal differences: a survey and comparison.
Christoph DannGerhard NeumannJan PetersPublished in: J. Mach. Learn. Res. (2014)
Keyphrases
- temporal difference
- policy evaluation
- td learning
- reinforcement learning
- function approximation
- least squares
- evaluation function
- monte carlo
- model free
- markov decision processes
- step size
- action selection
- policy iteration
- reinforcement learning algorithms
- supervised learning
- semi parametric
- radial basis function
- variance reduction
- feature selection
- training data