Markov Decision Processes Based Optimal Control Policies for Probabilistic Boolean Network.
Osman AbulReda AlhajjFaruk PolatPublished in: BIBE (2004)
Keyphrases
- control policies
- markov decision processes
- optimal policy
- finite horizon
- action space
- reinforcement learning
- average cost
- dynamic programming
- state space
- control policy
- policy iteration
- average reward
- finite state
- continuous state
- infinite horizon
- reward function
- real valued
- long run
- decision problems
- markov decision process
- action sets
- decision diagrams
- decision theoretic planning
- transition matrices
- probabilistic planning
- stationary policies
- multistage
- probabilistic model
- bayesian networks
- machine learning
- initial state
- optimal control
- latent variables
- sufficient conditions
- control system
- discounted reward