Policy Iteration within Logico-Numerical Abstract Domains.
Pascal SotinBertrand JeannetFranck VédrineEric GoubaultPublished in: ATVA (2011)
Keyphrases
- policy iteration
- markov decision processes
- model free
- fixed point
- reinforcement learning
- least squares
- finite state
- sample path
- markov decision process
- optimal policy
- infinite horizon
- policy evaluation
- average reward
- linear programming
- random walk
- dynamic programming
- search space
- artificial neural networks
- neural network