Optimizing Quantiles in Preference-Based Markov Decision Processes.
Hugo GilbertPaul WengYan XuPublished in: AAAI (2017)
Keyphrases
- markov decision processes
- optimal policy
- transition matrices
- finite state
- dynamic programming
- reinforcement learning
- state space
- reachability analysis
- policy iteration
- average cost
- risk sensitive
- finite horizon
- partially observable
- model based reinforcement learning
- decision theoretic planning
- planning under uncertainty
- sliding window
- action space
- infinite horizon
- average reward
- markov decision process
- reinforcement learning algorithms
- factored mdps
- real time dynamic programming
- learning algorithm
- discounted reward
- continuous state spaces
- decision processes
- decision making