Limits of Multi-Discounted Markov Decision Processes.
Hugo GimbertWieslaw ZielonkaPublished in: LICS (2007)
Keyphrases
- markov decision processes
- optimal policy
- finite state
- state space
- infinite horizon
- reinforcement learning
- finite horizon
- average reward
- average cost
- dynamic programming
- transition matrices
- risk sensitive
- policy iteration
- decision theoretic planning
- planning under uncertainty
- reinforcement learning algorithms
- model based reinforcement learning
- partially observable
- factored mdps
- markov decision process
- action space
- discounted reward
- decision processes
- long run
- real time dynamic programming
- partially observable markov decision processes
- reward function
- reachability analysis
- discount factor
- semi markov decision processes
- real valued
- learning algorithm