講題：Learning Representations For Robust Human-Robot Interaction

講者：美國維吉尼亞大學資訊科學系 郭彥伶助理教授

時間：2023.12.19 (二) 15:00-16:00

地點：人工智慧研究中心（管理大樓11樓）

線上直播：https://reurl.cc/q0MmZE

講者介紹：

Yen-Ling Kuo is an Anita Jones Faculty Fellow and Assistant Professor in Computer Science at the University of Virginia. 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 commonsense reasoning. Yen-Ling received her Ph.D. in Computer Science from MIT and BS/MS degrees in Computer Science and Information Engineering from National Taiwan University. She is a recipient of the CBMM-Siemens Graduate Fellowship.

演講大綱：

For robots to robustly and flexibly interact with humans, they need to acquire skills to use across scenarios. One way to enable the generalization of skills is to learn representations that are useful for downstream tasks. Learning a representation for interactions requires an understanding of what (e.g., objects) as well as how (e.g., actions, controls, and manners) to interact with. Failure to generalize in different scenarios 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 the compositional nature of language and reward functions to learn representations that generalize to novel scenarios. I will show that together with the information from multiple modalities, the learned representation can reason about task progress, future behaviors, and the goals/beliefs of an agent. I will demonstrate how these ideas can be used for language understanding and social interactions. I will conclude with research directions on endowing robots with generalizable reasoning skills and long-term human-AI interactions.

主辦單位：人工智慧研究中心、智慧運算學院

※本活動無需報名。