ECE Distinguished Lecture Series: Securing the Future of AI: Deep Learning Backdoor Attacks and Defenses

ABSTRACT

Deep learning has transformed our daily lives, revolutionizing efficiency and productivity. However, its widespread adoption has made it a prime target for cyberattacks. Today, the security of deep learning systems is a pressing concern. They can be poisoned, Trojanized, or infiltrated with backdoors, posing serious threats to their integrity and reliability. This talk will delve into the forefront of research on neural backdoor attacks, defense strategies, and their implications, aiming to stimulate novel ideas to advance this field and ensure the future security of deep learning technology.

BIOGRAPHY

Hongyi “Michael” Wu is a Kenneth Von Behren Professor and Department Head of Electrical and Computer Engineering at the University of Arizona. He received his B.S. degree in Scientific Instruments from Zhejiang University, Hangzhou, China, in 1996, and subsequently obtained his M.S. degree in Electrical Engineering and Ph.D. degree in Computer Science at the State University of New York (SUNY) at Buffalo in 2000 and 2002, respectively. His current research focuses on security and privacy in intelligent computing and communication systems. He has been a PI or Co-PI for more than 50 projects which have received over $24M in funding from various sources, including the NSF, NSA, DOD, DOE, NATO, State governments, and industry partners. He has published over 160 technical articles in journals and conference proceedings and acquired a US patent. He has chaired several conferences, including the prestigious IEEE Infocom 2020. He has also served on the editorial board of several journals, including as a Lead Topic Editor of IEEE Transactions on Computers (2023-now), an Area Editor for IEEE Internet of Things Journal (2022-now), and an Associate Editor for IEEE Transactions on Computers (2022-now), IEEE Transactions on Mobile Computing (2018-now), IEEE Internet of Things Journal (2014-2018), and IEEE Transactions on Parallel and Distributed Systems (2013-2016). He has received several awards and honors throughout his career, including an NSF CAREER Award in 2004, the UL Lafayette Distinguished Professor Award in 2011, the IEEE Percom Mark Weiser Best Paper Award in 2018, and the UB CSE Distinguished Alumni Award in 2022. He is a Fellow of IEEE.