HCMDP: A hierarchical solution to Constrained Markov Decision Processes.
Seyedshams FeyzabadiStefano CarpinPublished in: ICRA (2015)
Keyphrases
- markov decision processes
- reinforcement learning
- finite state
- state space
- optimal policy
- dynamic programming
- transition matrices
- decision processes
- reachability analysis
- policy iteration
- factored mdps
- infinite horizon
- risk sensitive
- planning under uncertainty
- finite horizon
- reinforcement learning algorithms
- average cost
- partially observable
- decision theoretic planning
- action space
- policy evaluation
- reward function
- action sets
- average reward
- decision diagrams
- state and action spaces
- optimal solution
- model based reinforcement learning
- data mining