•   When: Friday, March 04, 2022 from 02:00 PM to 03:00 PM
•   Speakers: Yen-Ling Kuo
•   Location: ZOOM only
•   Export to iCal

Abstract:

To effectively understand environments and safely interact with humans, robots must generalize their learned models to scenarios they have never been trained on before, such as new commands and new agents. Failure to generalize in those situations could cause robots to develop confusing or harmful behaviors in a variety of human-centric applications ranging from elderly care to driving.

In this talk, I will present my work on leveraging compositionality to enable generalization in language understanding and social interactions. I will first show how compositional linguistic structure can be incorporated into robotic models to enable robots to follow novel commands and act rationally in new scenarios. Then I will show how recursive reward estimation can enable robots to reason about sequences of actions and about novel social interactions. I will conclude with future research directions on endowing robots with generalizable reasoning skills and adapting robotic behavior for long-term Human-AI interactions.

Bio:

Yen-Ling Kuo is a final-year Ph.D. candidate at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). Her research interests lie at the intersection of artificial intelligence and cognitive science. She develops machine learning models that provide robots with generalizable reasoning skills including language understanding, social interactions, and common sense reasoning. Yen-Ling received her BS and MS degrees in Computer Science and Information Engineering from National Taiwan University. She is a recipient of the CBMM-Siemens Graduate Fellowship and the MIT Greater China Computer Science Fellowship.