Reinforcement Learning with Reward Shaping and Mixed Resolution Function Approximation.
Marek GrzesDaniel KudenkoPublished in: Int. J. Agent Technol. Syst. (2009)
Keyphrases
- function approximation
- reward shaping
- reinforcement learning
- reinforcement learning algorithms
- temporal difference
- temporal difference learning
- model free
- state space
- learning algorithm
- optimal policy
- dynamic programming
- function approximators
- markov decision problems
- complex domains
- machine learning
- action selection
- learning process
- transfer learning
- markov decision processes
- transition model
- reinforcement learning methods
- radial basis function
- learning tasks
- multi agent
- policy search
- temporal difference methods
- learning agent
- markov decision process
- optimal control
- utility function
- supervised learning