Enhanced Temporal Difference Learning Using Compiled Eligibility Traces.
Peter VamplewRobert OllingtonMark HepburnPublished in: Australian Conference on Artificial Intelligence (2006)
Keyphrases
- temporal difference learning
- eligibility traces
- reinforcement learning algorithms
- reinforcement learning
- temporal difference
- markov decision processes
- model free
- state space
- function approximation
- reinforcement learning methods
- learning algorithm
- fixed point
- evaluation function
- machine learning
- game playing
- markov decision process
- state variables
- function approximators
- optimal control
- dynamic environments
- dynamic programming
- artificial neural networks
- training data