Towards Versatile AI Task Assistants

Please join us for an Expeditions in Experiential AI seminar with Ehsan Elhamifar, associate professor in the Khoury College of Computer Sciences and the director of the Mathematical Data Science (MCADS) Lab at Northeastern University, on Wednesday, Oct. 16, 2024 from 1-2 p.m. ET at Northeastern University in Boston and online.

Abstract

What does it take to build an AI system that can understand complex physical and skilled human activities? If successful, such an AI system can serve as an intelligent task assistant to guide people in their activities at work and home, and it can also be integrated into robots to perform complex tasks or collaborate with humans.

In this seminar, Dr. Elhamifar will discuss his research on several fundamental problems that need to be solved in order to build such an AI system, including understanding complex actions, progress, and human mistakes in long videos and multimodal data, with a focus on the real-time aspect of such understanding. He will demonstrate examples of an AI task assistant that his team has developed over the past few years.

Bio

Ehsan Elhamifar is an associate professor in the Khoury College of Computer Sciences and the director of the Mathematical Data Science (MCADS) Lab at Northeastern University.

His research interests span computer vision and machine learning, with the overarching goal of developing systems that learn from and make inferences about data in a manner analogous to humans.

He is a recipient of the DARPA Young Faculty Award. Prior to joining Northeastern, he was a postdoctoral scholar in the EECS department at UC Berkeley. He earned his PhD in Electrical and Computer Engineering (ECE) at Johns Hopkins University (JHU).

‍