Metrics for Finite Markov Decision Processes
Norman FernsPrakash PanangadenDoina PrecupPublished in: CoRR (2012)
Keyphrases
- markov decision processes
- state and action spaces
- state space
- optimal policy
- finite state
- reinforcement learning
- dynamic programming
- policy iteration
- action space
- planning under uncertainty
- decision theoretic planning
- average reward
- transition matrices
- reachability analysis
- finite horizon
- partially observable
- reinforcement learning algorithms
- stationary policies
- factored mdps
- model based reinforcement learning
- average cost
- decision processes
- infinite horizon
- markov decision process
- risk sensitive
- decision problems
- dynamical systems
- linear programming
- interval estimation
- stochastic shortest path