Compositional reasoning for weighted Markov decision processes.
Yuxin DengMatthew HennessyPublished in: Sci. Comput. Program. (2013)
Keyphrases
- markov decision processes
- state space
- finite state
- policy iteration
- optimal policy
- transition matrices
- reinforcement learning
- dynamic programming
- decision theoretic planning
- reachability analysis
- factored mdps
- planning under uncertainty
- average cost
- markov decision process
- decision processes
- finite horizon
- action space
- reinforcement learning algorithms
- average reward
- reasoning process
- partially observable
- state and action spaces
- action sets
- infinite horizon
- risk sensitive
- reward function
- model based reinforcement learning
- decision making
- semi markov decision processes
- real time dynamic programming
- stochastic shortest path