Planning in Discrete and Continuous Markov Decision Processes by Probabilistic Programming.
Davide NittiVaishak BelleLuc De RaedtPublished in: ECML/PKDD (2) (2015)
Keyphrases
- markov decision processes
- continuous state spaces
- probabilistic planning
- action space
- planning under uncertainty
- macro actions
- decision theoretic planning
- state space
- partially observable
- finite state
- transition matrices
- optimal policy
- planning problems
- continuous state
- policy iteration
- reinforcement learning
- dynamic programming
- decision processes
- reachability analysis
- state and action spaces
- markov decision process
- markov decision problems
- partially observable markov decision processes
- average reward
- infinite horizon
- factored mdps
- heuristic search
- bayesian networks
- model based reinforcement learning
- action sets
- finite horizon
- probabilistic model
- average cost
- sufficient conditions
- initial state
- ai planning
- domain independent
- real valued
- multistage
- decision theory
- planning domains
- reward function
- action selection