Trace Equivalence Characterization Through Reinforcement Learning.
Josée DesharnaisFrançois LavioletteKrishna Priya Darsini MoturuSami ZhiouaPublished in: Canadian Conference on AI (2006)
Keyphrases
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- learning algorithm
- model free
- markov decision processes
- state space
- neural network
- transfer learning
- transition model
- temporal difference
- data sets
- dynamic programming
- machine learning
- active learning
- learning process
- search algorithm
- data mining
- markov decision process
- equivalence relation
- temporal difference learning
- stochastic approximation
- real world
- axiomatic characterization
- direct policy search