Interactive Value Iteration for Markov Decision Processes with Unknown Rewards.
Paul WengBruno ZanuttiniPublished in: IJCAI (2013)
Keyphrases
- markov decision processes
- policy iteration
- state space
- optimal policy
- reinforcement learning
- finite state
- transition matrices
- dynamic programming
- decision theoretic planning
- reachability analysis
- average reward
- sequential decision making under uncertainty
- partially observable
- reinforcement learning algorithms
- reward function
- model based reinforcement learning
- factored mdps
- markov decision process
- planning under uncertainty
- risk sensitive
- discounted reward
- infinite horizon
- finite horizon
- average cost
- partially observed
- decision processes
- objective function
- machine learning
- semi markov decision processes
- state and action spaces
- state abstraction
- action space