Approximate Equivalence of Markov Decision Processes.
Eyal Even-DarYishay MansourPublished in: COLT (2003)
Keyphrases
- markov decision processes
- factored mdps
- state space
- optimal policy
- policy evaluation
- finite state
- reinforcement learning
- decision diagrams
- dynamic programming
- transition matrices
- policy iteration
- finite horizon
- planning under uncertainty
- decision theoretic planning
- risk sensitive
- action space
- reinforcement learning algorithms
- partially observable
- decision processes
- markov decision process
- infinite horizon
- reachability analysis
- action sets
- average cost
- semi markov decision processes
- least squares
- state and action spaces
- state abstraction
- average reward
- data mining