A value iteration method for undiscounted multichain Markov decision processes.
Katsuhisa OhnoPublished in: ZOR Methods Model. Oper. Res. (1988)
Keyphrases
- markov decision processes
- dynamic programming
- state space
- average reward
- reinforcement learning
- transition matrices
- policy iteration
- optimal policy
- finite state
- decision theoretic planning
- reinforcement learning algorithms
- markov decision process
- planning under uncertainty
- finite horizon
- action sets
- stochastic games
- optimal solution
- decision processes
- data mining
- partially observable
- infinite horizon
- learning algorithm
- model based reinforcement learning
- state and action spaces
- factored mdps
- continuous state spaces
- policy evaluation
- least squares