A Hierarchical Reinforcement Learning Method for Persistent Time-Sensitive Tasks
This addresses a gap in reinforcement learning for persistent and time-sensitive tasks, but it appears incremental as it builds on existing STL and options methods.
The paper tackles the problem of learning policies for complex persistent and time-sensitive tasks by combining signal temporal logic (STL) with the options framework, showing in simulation that this approach can learn satisfactory policies with a small number of training cases.
Reinforcement learning has been applied to many interesting problems such as the famous TD-gammon and the inverted helicopter flight. However, little effort has been put into developing methods to learn policies for complex persistent tasks and tasks that are time-sensitive. In this paper, we take a step towards solving this problem by using signal temporal logic (STL) as task specification, and taking advantage of the temporal abstraction feature that the options framework provide. We show via simulation that a relatively easy to implement algorithm that combines STL and options can learn a satisfactory policy with a small number of training cases