Learning Policies for Markov Decision Processes in Continuous Spaces.
Santiago PaternainJuan Andrés BazerqueAustin SmallAlejandro RibeiroPublished in: CDC (2018)
Keyphrases
- markov decision processes
- optimal policy
- reinforcement learning
- state space
- macro actions
- action space
- partially observable
- real time dynamic programming
- model based reinforcement learning
- decision theoretic planning
- stochastic games
- finite state
- infinite horizon
- markov decision process
- policy iteration
- decision processes
- continuous state
- supervised learning
- hierarchical reinforcement learning
- learning algorithm
- reinforcement learning algorithms
- transition matrices
- search space
- semi markov decision processes
- dynamic programming
- planning under uncertainty
- multistage
- average cost