Robust Markov Decision Processes.
Wolfram WiesemannDaniel KuhnBerç RustemPublished in: Math. Oper. Res. (2013)
Keyphrases
- markov decision processes
- optimal policy
- reinforcement learning
- finite state
- reinforcement learning algorithms
- state space
- policy iteration
- transition matrices
- decision theoretic planning
- model based reinforcement learning
- reachability analysis
- action space
- finite horizon
- factored mdps
- dynamic programming
- average cost
- risk sensitive
- planning under uncertainty
- partially observable
- infinite horizon
- decision processes
- action sets
- machine learning
- average reward
- state and action spaces
- markov decision process