Basis refinement strategies for linear value function approximation in MDPs.
Gheorghe ComaniciDoina PrecupPrakash PanangadenPublished in: NIPS (2015)
Keyphrases
- linear value function approximation
- reinforcement learning
- markov decision processes
- reinforcement learning problems
- markov games
- optimal policy
- reinforcement learning algorithms
- state space
- linear programming
- markov decision process
- game theory
- reward function
- policy iteration
- average reward
- reinforcement learning methods