AI Seminar: Stella Yu – Unsupervised Data-Driven Learning of Visual Hierarchy

Zoom

https://umich.zoom.us/j/92216884113 (password: UMichAI)

Title

Unsupervised Data-Driven Learning of Visual Hierarchy

Abstract

The visual scene of an office could contain desks, chairs, bookshelves, and persons; each can be further decomposed into components such as legs, seats, surfaces, and human body parts. Could we discover this common organization of visual concepts in a data-driven manner, and in turn utilize hierarchical groupings during learning to facilitate the discovery of universal visual concepts? I will present a series of our recent work in this direction, including unsupervised learning of visual context, concurrent segmentation and recognition, unsupervised instance selection, and unsupervised learning of data prototypicality in hyperbolic space.

Bio

Stella Yu received her Ph.D. from Carnegie Mellon University, where she studied robotics at the Robotics Institute and vision science at the Center for the Neural Basis of Cognition.  Before she joined the University of Michigan faculty in Fall 2022, she has been the Director of Vision Group at the International Computer Science Institute, a Senior Fellow at the Berkeley Institute for Data Science, and on the faculty of Computer Science, Vision Science, Cognitive and Brain Sciences at UC Berkeley.  Dr. Yu is interested not only in understanding visual perception from multiple perspectives, but also in using computer vision and machine learning to automate and exceed human expertise in practical applications.  Her group currently focuses on complex-valued deep learning, sound-vision integration, and actionable mid-level representation learning from non-curated data with minimal human annotations.  Dr. Yu leads multiple interdisciplinary projects and has a strong track record of joint research and successful product deployment in the field.

This AI Seminar is sponsored by LG AI Research.