Inductive Policy Selection for First-Order MDPs
Sung Wook YoonAlan FernRobert GivanPublished in: CoRR (2013)
Keyphrases
- optimal policy
- markov decision processes
- markov decision process
- finite horizon
- policy iteration
- reinforcement learning
- state space
- policy search
- state and action spaces
- higher order
- partially observable
- markov decision problems
- machine learning
- factored mdps
- decision diagrams
- policy evaluation
- inductive logic programming
- average cost
- decision problems
- action space
- decision processes
- average reward
- planning under uncertainty
- finite state
- dynamic programming
- concept learning
- inductive learning
- reward function
- action selection
- infinite horizon
- long run
- approximate dynamic programming