Time-Bounded Reachability Probabilities in Continuous-Time Markov Decision Processes.
Martin R. NeuhäußerLijun ZhangPublished in: QEST (2010)
Keyphrases
- markov decision processes
- state space
- markov chain
- optimal policy
- reinforcement learning
- dynamic programming
- finite state
- dynamical systems
- policy iteration
- transition probabilities
- partially observable
- decision theoretic planning
- infinite horizon
- factored mdps
- average reward
- markov decision process
- reinforcement learning algorithms
- finite horizon
- decision processes
- state and action spaces
- transition matrices
- reachability analysis
- belief state
- model based reinforcement learning
- planning under uncertainty
- stationary policies
- action sets
- initial state
- belief networks
- planning problems
- probability distribution
- risk sensitive
- stochastic processes
- real time dynamic programming