A Function Approximation Approach to Estimation of Policy Gradient for POMDP with Structured Policies
Huizhen YuPublished in: CoRR (2012)
Keyphrases
- policy gradient
- function approximation
- model free reinforcement learning
- reinforcement learning
- policy gradient methods
- partially observable markov decision processes
- policy search
- actor critic
- optimal policy
- function approximators
- finite state
- temporal difference
- reinforcement learning algorithms
- markov decision problems
- model free
- belief state
- learning tasks
- decision problems
- policy evaluation
- continuous state
- markov decision process
- temporal difference learning
- radial basis function
- dynamical systems
- partially observable
- markov decision processes
- average reward
- markov chain
- multi agent
- reward function
- reinforcement learning methods
- gradient method
- optimal control
- transfer learning
- state space
- dynamic programming
- learning algorithm
- machine learning
- planning problems
- natural actor critic