-discount and finite-horizon optimality for continuous-time Markov decision processes.
Quanxin ZhuXianping GuoPublished in: J. Syst. Sci. Complex. (2014)
Keyphrases
- finite horizon
- markov decision processes
- average cost
- stationary policies
- state space
- optimal policy
- average reward
- optimal stopping
- infinite horizon
- finite state
- markov decision process
- reinforcement learning
- policy iteration
- optimal control
- dynamic programming
- markov chain
- transition matrices
- decision problems
- initial state
- decision theoretic planning
- single item
- partially observable
- factored mdps
- dynamical systems
- control policies
- multistage
- state dependent
- reward function
- integer programming
- linear program
- stochastic shortest path