Add video for Roozbeh's talk

Author: angelxuanchang

contents/seminars/seminar_info.yaml

@@ -3,8 +3,8 @@ organizers_by_year:
     terms:
       - term: Spring
         main_student_coordinator: Xingguang Yan
-        student_volunteers: Xingguang Yan, Xiaoliang Huo, Jiayi Liu
-        faculty_sponsors: Angel Chang, Manolis Savva, Linyi Li
+        student_volunteers: Xingguang Yan, Xiaoliang Huo, Jiayi Liu, Yalda Foroutan
+        faculty_sponsors: Angel Chang, Manolis Savva, Linyi Li, Yasu Furukawa
   - year: 2024
     terms:
       - term: Fall

contents/seminars/seminars.yaml

@@ -18,6 +18,7 @@
   title: Embodied AI at Scale
   abstract: The remarkable progress in AI across domains like language and computer vision has been largely fueled by the use of large-scale datasets for training models. However, achieving a similar transformative impact in the field of Embodied AI remains a challenge, primarily due to scaling difficulties. In this talk, I will delve into approaches that facilitate scaling in Embodied AI, including the utilization of simulation environments and the use of internet video resources. Additionally, I will discuss methods for scaling not only generic training data but also the creation and scaling of datasets, with a focus on developing a large-scale benchmark for embodied planning and reasoning in dynamic environments. I will present a comprehensive study on how state-of-the-art planning models struggle with these tasks and discuss methods for distilling knowledge from models trained on large-scale data into smaller models that are suitable for deployment on robots.
   bio: Roozbeh Mottaghi is a Senior Research Scientist Manager at FAIR and an Affiliate Associate Professor in the Paul G. Allen School of Computer Science and Engineering at the University of Washington. Before joining FAIR, he was the Research Manager of the Perceptual Reasoning and Interaction Research (PRIOR) group at the Allen Institute for AI (AI2). He completed his master’s degrees at Simon Fraser University and Georgia Tech and earned his PhD in Computer Science from the University of California, Los Angeles, in 2013. After his PhD, he joined the Computer Science Department at Stanford University as a post-doctoral researcher. His research primarily focuses on embodied AI, reasoning through perception, and learning via interaction. His work on large-scale Embodied AI received the Outstanding Paper Award at NeurIPS 2022.
+  video: https://stream.sfu.ca/Media/Play/34b3bbc58e314b04b7caa31fc543c5bd1d
 
 - date: "2025-02-21"
   location: TASC1 9204 11:00am