Date: Tuesday, March. 25
Time: 10:30am - 11:30am
Location: NWC 14th FLR conference room
Seminar: Empowering the Next Billion Devices with AI
Host: Xiaofan (Fred) Jiang
Abstract:
The proliferation of edge devices and the gigantic amount of data theygenerate make it no longer feasible to transmit all the data to the cloud forprocessing. Such constraints fuel the need to move the intelligence from thecloud to the edge where data resides. In this talk, I will present our works onhow we bring the power of AI, in particular, deep learning, to edge devices torealize the vision of Artificial Intelligence of Things (AIoT).
This talk consists of two parts. The first part focuses on how weaddress some of the most fundamental problems that act as the key barriers ofachieving the vision of AIoT. First, I will present our work on designingadaptive frameworks that empower AI-embedded edge devices to adapt to theinherently dynamic runtime system resources in real-world deployments. Second,I will talk about our work on developing automated machine learning (AutoML)frameworks that provide an automated and scalable solution to the device delugechallenge in AIoT. In the second part of this talk, I will present how we useAI as the core component to design AIoT systems for a broad range of problemdomains. I will focus on one killer application of edge computing, and presentan AI-empowered distributed edge system for low-latency, high-throughput, andscalable live video analytics. Finally, I will talk about our work on spatialcomputing, which pushes the frontier and opens up new opportunities of AIoTresearch.
Bio:
Mi Zhang is an Associate Professor and the Director of AIoT andMachine Learning Systems Lab at The Ohio State University (OSU). He receivedhis Ph.D. in Computer Engineering from University of Southern California (USC)and B.S. from Peking University, and spent one year as a Postdoctoral Associateat Cornell University. The key mission of his lab is to Empower Billions ofEveryday Devices with AI to realize the vision of Artificial Intelligence ofThings. To achieve this mission, he and his students focus on its corechallenges related to sensing, intelligence, connectivity, efficiency as wellas its real-world applications. Dr. Zhang’s work has been recognized by sevenbest paper awards and nominations, NSF CRII Award, Facebook/Meta FacultyResearch Award, and Amazon Research Award. He is the 4th Place Winner of GoogleMicroNet Challenge, the Third Place Winner of NSF Hearables Challenge, and thechampion of NIH Pill Image Recognition Challenge. He is also the recipient ofthe inaugural USC ECE SIPI Distinguished Alumni Award in the Junior/Academiacategory for his contributions to mobile computing, edge AI, and AIoT in hisearly career.