Planning using hierarchical constrained Markov decision processes.
Seyedshams FeyzabadiStefano CarpinPublished in: Auton. Robots (2017)
Keyphrases
- optimal solution
- markov decision processes
- macro actions
- decision theoretic planning
- planning under uncertainty
- partially observable
- finite state
- state space
- linear programming
- optimal policy
- reinforcement learning
- transition matrices
- reachability analysis
- objective function
- partially observable markov decision processes
- infinite horizon
- policy iteration
- factored mdps
- markov decision problems
- planning problems
- dynamic programming
- markov decision process
- action space
- reinforcement learning algorithms
- ai planning
- average reward
- average cost
- finite horizon
- decision processes
- domain independent
- model based reinforcement learning
- semi markov decision processes
- heuristic search
- action sets
- state abstraction
- probabilistic planning
- finite number
- data mining
- discounted reward
- state and action spaces
- hierarchical reinforcement learning
- planning domains
- decision theoretic