CIIR Talk Series: Ismini Lourentzou

Speaker: Ismini Lourentzou, Virginia Tech

Title: Robust Visual Understanding with Limited Supervision

Date: Friday, March 24, 2023 - 1:30 - 2:30 PM EST (North American Daylight Saving Time) via Zoom. On campus attendees will gather in CS 151 to view the presentation.

Abstract: Over the past few years, a wide range of computer vision and natural language understanding models have been proposed for various perception and reasoning tasks such as classifying images, recognizing objects, understanding natural language, or even navigating to a goal location. Despite recent progress, there still exist several critical deficiencies that need to be addressed before deployment, including data scarcity and privacy, robustness, generalization, and human-AI collaboration. This talk will cover some of our research on learning with limited supervision for various applied computer vision tasks. Finally, I will outline open research directions in human-agent collaboration and embodied intelligence.

Bio: Ismini Lourentzou is an Assistant Professor at the Computer Science Department of Virginia Tech, where she leads the Perception and Language (PLAN) Lab. She is also a faculty member of the Sanghani Center for Artificial Intelligence and Discovery Analytics, and an affiliate faculty of the National Security Institute and the Center for Advanced Innovation in Agriculture. Ismini's research focus is multimodal machine learning, primarily the intersection of vision and language in settings with limited supervision, and its applications in healthcare, embodied AI, and other fields. She served on the organizing committee of NeurIPS'22 and is currently serving on the NeurIPS'23 organizing committee, as well as holding editorial and area chair roles for top-tier AI journals and conferences. Her research has received support from NSF, DARPA, CCI, and Amazon.