A unified approach to Markov decision problems and performance sensitivity analysis with discounted and average criteria: multichain cases.
Xi-Ren CaoXianping GuoPublished in: Autom. (2004)
Keyphrases
- sensitivity analysis
- markov decision problems
- optimal policy
- markov decision processes
- average cost
- average reward
- infinite horizon
- managerial insights
- state space
- partially observable
- policy iteration
- reinforcement learning
- influence diagrams
- long run
- dynamic programming
- finite state
- finite number
- markov decision process
- decision processes
- decision problems
- neural network
- reward function
- linear program
- multistage
- multi agent
- objective function