Nonatomic total rewards Markov decision processes with multiple criteria.
Eugene A. FeinbergAlexei B. PiunovskiyPublished in: CDC (2000)
Keyphrases
- markov decision processes
- multiple criteria
- decision problems
- optimal policy
- reinforcement learning
- multi criteria
- decision makers
- finite state
- state space
- dynamic programming
- multi objective
- transition matrices
- finite horizon
- mathematical programming
- average cost
- multi attribute
- decision theoretic planning
- infinite horizon
- reinforcement learning algorithms
- policy iteration
- state and action spaces
- reward function
- markov decision process
- action space
- average reward
- total reward
- linear programming
- partially observable
- decision making
- lower bound
- machine learning
- fuzzy logic
- sufficient conditions
- neural network
- influence diagrams