Estimation and Approximation Bounds for Gradient-Based Reinforcement Learning.
Peter L. BartlettJonathan BaxterPublished in: J. Comput. Syst. Sci. (2002)
Keyphrases
- reinforcement learning
- error bounds
- model free
- approximation methods
- stage stochastic programs
- state space
- upper bound
- upper and lower bounds
- learning algorithm
- importance sampling
- accurate estimation
- lower bound
- function approximation
- error tolerance
- maximum likelihood estimator
- approximation error
- markovian decision
- parameter estimation
- reinforcement learning algorithms
- temporal difference
- temporal difference learning
- learning problems
- supervised learning
- neural network