Delayed Nondeterminism in Continuous-Time Markov Decision Processes.
Martin R. NeuhäußerMariëlle StoelingaJoost-Pieter KatoenPublished in: FoSSaCS (2009)
Keyphrases
- markov decision processes
- state space
- stationary policies
- markov chain
- reinforcement learning
- optimal policy
- dynamic programming
- policy iteration
- finite state
- transition matrices
- optimal control
- risk sensitive
- planning under uncertainty
- situation calculus
- dynamical systems
- average cost
- heuristic search
- infinite horizon
- decision processes
- reachability analysis
- finite horizon
- reinforcement learning algorithms
- factored mdps
- average reward
- model based reinforcement learning
- markov decision process
- partially observable
- decision theoretic planning
- action sets
- monte carlo
- state and action spaces
- search space
- action space
- planning domains
- fixed point
- multistage
- sufficient conditions