Robust Control of Markov Decision Processes with Uncertain Transition Matrices.
Arnab NilimLaurent El GhaouiPublished in: Oper. Res. (2005)
Keyphrases
- transition matrices
- markov decision processes
- state space
- markov decision process
- optimal policy
- decision theoretic planning
- reinforcement learning
- infinite horizon
- finite state
- policy iteration
- average cost
- decision processes
- frequency counts
- reinforcement learning algorithms
- dynamic programming
- action space
- average reward
- markov chain
- reward function
- partially observable markov decision processes
- search algorithm
- decision making