Join us from 12:45–1:45 p.m. on Monday, March 10, 2025, for a Computer Science Seminar at the Stuart Building, room 113. Speaker Nathaniel Hudson will give a talk.

Abstract

Conventional solutions for training and serving AI models rely on centralized systems (e.g., HPC clusters, data centers). While powerful, these systems are insufficient to train AI models on the unquantifiable amounts of data generated, collected, and sensed daily at the network edge. To address this limitation, the future of AI will require the utilization of the full computing continuum: from the cloud to the edge. However, resources at the edge are plagued with two critical challenges: (i) system heterogeneity and (ii) data/statistical heterogeneity. For the former challenge, the edge faces harsh resource constraints that must be considered for deploying, serving, and training AI models; for the latter, data at the edge are more likely to be skewed and non-iid, which complicates training accurate models. In this talk, I will present results from my research related to deploying, serving, and training AI at the network edge. Specifically, I will discuss optimal decision-making for serving and placing AI at the edge and balancing trade-offs associated with training AI at the edge with federated learning.

Biography

Nathaniel Hudson is a postdoctoral scholar at the University of Chicago, with a joint appointment at Argonne National Laboratory. He received his Ph.D. degree in computer science at the University of Kentucky in 2022. His research broadly focuses on decentralized learning methods, such as federated learning, with the aim to take advantage of the computing continuum by training, serving, and placing AI from the network edge to the cloud. He has developed the first federated learning framework with native support for hierarchical networks, AI placement and scheduling algorithms for edge computing systems, and new methods for interpreting large language models. His work has been applied to various domains, such as materials science, smart city use cases, and rural applications. His research has been recognized by numerous best paper awards and he has been recognized as a “Rising Star” in cyber-physical systems.