Time: 1:00 PM, October 25th (Friday), 2024
Location: POTR 234
Coffee and bagels will be provided.
Building Trustworthy AI Models for Autonomous Cyber Physical Systems
Qiben Yan, Associate Professor, Department of Computer Science and Engineering, Michigan State University
As Internet-of-Things (IoT) and Cyber-Physical Systems (CPS) become integral to daily life, advanced sensing technologies powered by AI models play a pivotal role. However, their deep integration with the physical world has made them susceptible to physical adversarial and backdoor attacks. In this talk, I will present our latest research on securing autonomous driving systems. I will highlight the vulnerabilities of camera systems in drones and autonomous vehicles against adversarial attacks, designed to inject false data, such as imaginary obstacles. The talk will conclude with insights into building trustworthy, resilient AI-enabled cyber-physical systems.
Qiben Yan is an Associate Professor of Computer Science and Engineering at Michigan State University. His research focuses on cyber-physical systems security, trustworthy AI systems, wireless communication, and blockchain security. His recent research identifies security vulnerabilities in autonomous driving and voice AI systems. He received the NSF CRII award (2016), best paper awards at IEEE SECON and ACM SenSys (2021), and a Best Paper Honorable Mention at ACM CCS (2022). His work has been featured in BBC Radio, Scientific American, and Forbes. His recent research has been supported by NSF, DoE, and FORD. He received his Ph.D. from Virginia Tech.