Discounted Markov decision processes with fuzzy costs.
Abdellatif SemmouriMostafa JourhmaneZineb BelhallajPublished in: Ann. Oper. Res. (2020)
Keyphrases
- markov decision processes
- average cost
- optimal policy
- state space
- reinforcement learning
- finite state
- infinite horizon
- dynamic programming
- policy iteration
- decision theoretic planning
- transition matrices
- factored mdps
- average reward
- finite horizon
- planning under uncertainty
- long run
- partially observable
- markov decision process
- reinforcement learning algorithms
- reachability analysis
- real time dynamic programming
- total reward
- discounted reward
- decision processes
- control system
- expected cost
- total cost
- decision problems
- linear program
- semi markov decision processes
- model based reinforcement learning
- optimal solution
- objective function