Nonstationary denumerable state Markov decision processes - with average variance criterion.
Xianping GuoPublished in: Math. Methods Oper. Res. (1999)
Keyphrases
- markov decision processes
- non stationary
- state space
- finite horizon
- average cost
- discounted reward
- finite state
- action space
- optimal policy
- markov decision process
- real time dynamic programming
- policy iteration
- partially observable
- transition matrices
- planning under uncertainty
- average reward
- reinforcement learning
- decision theoretic planning
- dynamic programming
- reinforcement learning algorithms
- state abstraction
- risk sensitive
- random fields
- initial state
- optimality criterion
- infinite horizon
- state and action spaces
- reachability analysis
- model based reinforcement learning
- total reward
- planning problems
- action sets
- state variables
- stochastic shortest path