Rollout Sampling Approximate Policy Iteration.
Christos DimitrakakisMichail G. LagoudakisPublished in: ECML/PKDD (1) (2008)
Keyphrases
- approximate policy iteration
- reinforcement learning
- policy iteration
- markov decision problems
- policy search
- markov decision processes
- markov games
- temporal difference
- fixed point
- linear programming
- state space
- model free
- function approximation
- multiagent reinforcement learning
- reinforcement learning algorithms
- markov decision process
- optimal policy
- least squares
- partially observable
- function approximators
- decision processes
- continuous state
- learning algorithm
- finite state
- dynamic programming
- control problems
- infinite horizon
- evaluation function
- convergence rate
- machine learning
- cooperative