Tracking value function dynamics to improve reinforcement learning with piecewise linear function approximation.
Chee Wee PhuaRobert FitchPublished in: ICML (2007)
Keyphrases
- function approximation
- piecewise linear
- reinforcement learning
- function approximators
- dynamic programming
- temporal difference
- model free
- temporal difference learning algorithms
- temporal difference learning
- radial basis function
- reinforcement learning algorithms
- learning tasks
- particle filter
- state space
- temporal difference methods
- data sets
- machine learning
- reinforcement learning problems
- dynamical systems
- learning experience
- linear combination
- policy gradient
- data mining