Markov Decision Processes with Ordinal Rewards: Reference Point-Based Preferences.
Paul WengPublished in: ICAPS (2011)
Keyphrases
- markov decision processes
- finite state
- reinforcement learning
- state space
- optimal policy
- dynamic programming
- policy iteration
- decision processes
- reinforcement learning algorithms
- finite horizon
- sequential decision making under uncertainty
- reachability analysis
- infinite horizon
- reward function
- transition matrices
- average cost
- factored mdps
- decision theoretic planning
- planning under uncertainty
- average reward
- model based reinforcement learning
- decision making
- probabilistic planning
- decision diagrams
- discounted reward
- partially observable
- semi markov decision processes
- action space
- markov decision process
- interval estimation