Entropic Regularization of Markov Decision Processes.
Boris BelousovJan PetersPublished in: CoRR (2019)
Keyphrases
- markov decision processes
- optimal policy
- finite state
- reinforcement learning
- transition matrices
- policy iteration
- state space
- dynamic programming
- reachability analysis
- decision theoretic planning
- finite horizon
- average cost
- markov decision process
- infinite horizon
- factored mdps
- reinforcement learning algorithms
- planning under uncertainty
- partially observable
- decision processes
- state and action spaces
- sufficient conditions
- action sets
- machine learning
- average reward
- regularization parameter
- dynamical systems