Adaptive Discretization Using Voronoi Trees for Continuous-Action POMDPs.
Marcus HörgerHanna KurniawatiDirk P. KroeseNan YePublished in: WAFR (2022)
Keyphrases
- continuous action
- continuous state
- partially observable markov decision processes
- policy search
- reinforcement learning
- decision problems
- robot navigation
- partially observable
- finite state
- reinforcement learning algorithms
- planning problems
- dynamical systems
- action space
- state dependent
- optimal policy
- search space
- multi agent systems