Projects

Highlighted Research Directions

Short descriptions under blocks detail the research question(s) addressed by each block!

Multi-Agent Coordination

It is challenging to design decentralized multi-agent coordination algorithms that allow agents to perform collaboration in tasks such as search-and-rescue. We instead use multi-agent reinforcement learning (MARL) to allow decentralized robots to collectively share knowledge and develop collaboration plans.

Explainable AI

Recent advances in machine learning have led to growing interest in Explainable AI to enable humans to gain insight into the decision-making of machine learning models. However, 1) generating interpretable agents via reinforcement learning remains an open challenge, and 2) it is difficult to characterize the utility of interpretable models in human-machine teaming.

Learning from Demonstration

Having engineers program behavior for robots in each specific application across factories, hospitals, and households is intractable, can result in suboptimal robot policies, and may require access to domain information that is private. Thus, to truly achieve ubiquity of robotics, end-users must be able to interactively teach robots desired behaviors, thereby imparting their domain knowledge onto the robot.