My research is on skill discovery for hierarchical reinforcement learning.
My two main research questions are:
- How can AI develop high-level skills that assist in completing complex tasks?
- What are the different ways of representing how an agent interacts with an environment? And how can these different representations be unified into a single theory?