Maximal Cost-Bounded Reachability Probability on Continuous-Time Markov Decision Processes.
Hongfei FuPublished in: FoSSaCS (2014)
Keyphrases
- markov decision processes
- state space
- average cost
- finite state
- optimal policy
- markov chain
- stationary policies
- optimal control
- heuristic search
- reinforcement learning
- transition matrices
- dynamic programming
- factored mdps
- partially observable
- reinforcement learning algorithms
- risk sensitive
- decision theoretic planning
- finite horizon
- expected reward
- policy iteration
- probability distribution
- action space
- dynamical systems
- infinite horizon
- action sets
- planning problems
- reachability analysis
- state abstraction
- state and action spaces
- decision processes
- search space
- average reward
- reward function
- probabilistic planning
- planning under uncertainty
- expected cost
- belief state
- total cost
- learning algorithm