Transfer-Entropy-Regularized Markov Decision Processes.
Takashi TanakaHenrik SandbergMikael SkoglundPublished in: IEEE Trans. Autom. Control. (2022)
Keyphrases
- markov decision processes
- finite state
- optimal policy
- state space
- transition matrices
- dynamic programming
- reinforcement learning
- policy iteration
- reinforcement learning algorithms
- transfer learning
- reachability analysis
- factored mdps
- planning under uncertainty
- least squares
- infinite horizon
- model based reinforcement learning
- finite horizon
- average reward
- partially observable
- decision processes
- average cost
- data mining
- objective function
- learning algorithm
- action space
- risk sensitive
- decision theoretic planning
- reward function
- markov decision process
- action sets
- model checking