Vision-based reinforcement learning using approximate policy iteration.
Marwan R. ShakerShigang YueTom DuckettPublished in: ICAR (2009)
Keyphrases
- approximate policy iteration
- reinforcement learning
- policy iteration
- policy search
- temporal difference
- markov games
- markov decision problems
- markov decision processes
- function approximation
- function approximators
- state space
- reinforcement learning algorithms
- model free
- control problems
- learning algorithm
- dynamic programming
- temporal difference learning
- partially observable
- machine learning
- infinite horizon
- optimal control
- optimal policy
- reinforcement learning methods
- action space
- transfer learning
- fixed point
- action selection
- markov decision process
- least squares
- learning problems
- evaluation function
- continuous state
- search space
- multi agent
- finite state