Markov Decision Processes with Arbitrary Reward Processes.
Jia Yuan YuShie MannorNahum ShimkinPublished in: Math. Oper. Res. (2009)
Keyphrases
- markov decision processes
- average reward
- reinforcement learning
- reward function
- optimal policy
- expected reward
- total reward
- state space
- discounted reward
- policy iteration
- decision processes
- finite state
- reinforcement learning algorithms
- stationary policies
- dynamic programming
- transition matrices
- reachability analysis
- long run
- finite horizon
- action space
- planning under uncertainty
- semi markov decision processes
- factored mdps
- markov decision process
- infinite horizon
- decision theoretic planning
- average cost
- machine learning
- state and action spaces
- action sets
- function approximation
- partially observable markov decision processes
- stochastic processes
- partially observable
- model free
- markov chain
- model based reinforcement learning
- learning algorithm