Finite State Markov Decision Processes with Transfer Entropy Costs.
Takashi TanakaHenrik SandbergMikael SkoglundPublished in: CoRR (2017)
Keyphrases
- finite state
- markov decision processes
- average cost
- state space
- optimal policy
- dynamic programming
- policy iteration
- action sets
- decision theoretic planning
- transition matrices
- markov chain
- reinforcement learning
- markov decision process
- average reward
- infinite horizon
- reinforcement learning algorithms
- finite horizon
- model checking
- planning under uncertainty
- action space
- decision processes
- continuous state
- model based reinforcement learning
- partially observable
- approximate dynamic programming
- risk sensitive
- expected cost
- policy iteration algorithm
- stationary policies
- vector quantizer