Stopped Markov decision processes with multiple constraints.
Masayuki HoriguchiPublished in: Math. Methods Oper. Res. (2001)
Keyphrases
- multiple constraints
- markov decision processes
- optimal policy
- dynamic programming
- state space
- finite state
- transition matrices
- reinforcement learning
- planning under uncertainty
- finite horizon
- policy iteration
- decision processes
- decision theoretic planning
- reachability analysis
- average cost
- risk sensitive
- infinite horizon
- factored mdps
- partially observable
- average reward
- state and action spaces
- action space
- markov decision process
- state abstraction
- reinforcement learning algorithms
- discounted reward