Calendar - University of Houston
Skip to main content

[Seminar] Artificial Intelligence, Security & Privacy

Friday, February 18, 2022

11:00 am - 12:00 pm

Speaker

Lichao Sun
Assistant Professor, Computer Science & Engineering
Lehigh University

Location

Hybrid: PGH 232 and online via Zoom*
*Microsoft 365 @cougarnet.uh.edu authentication required to join via Zoom

Abstract

In the last ten years, benefiting from extensive data and powerful computing resources, more and more AI techniques achieved impressive breakthroughs for solving different problems in various areas, such as computer vision, natural language processing, robotics, healthcare, and social science. While society uses more AI technologies in real life, the security and privacy problems of AI technology become the top issues. First, some researchers use AI technology to solve different security problems, such as FaceID, touchID, malware detection systems, and so on. However, AI technology itself is very vulnerable, which becomes the first target of the attackers. For each application, we request a robust AI model and system against malicious intentions. Moreover, data privacy becomes a serious problem while using them to train a model in AI projects. In this talk, I will describe how to use AI to solve the security problem, how to protect the data with AI projects, and how to explore the weakness of current AI technology. I’ll also introduce future work in each direction and explain some possibilities for upcoming evolutions in AI security and privacy research.

About the Speaker

Lichao Sun is currently an Assistant Professor in the Department of Computer Science and Engineering at Lehigh University. His research interests include security and privacy in deep learning and data mining. He mainly focuses on AI security and privacy, social networks, and natural language processing applications. His current research is primarily focused on security and privacy issues in wireless networks, mobile devices, data sharing, and AI security, from both theoretical and practical perspectives. He is particularly interested in data-driven analytics and methods for applications, to reveal some new general and fundamental concepts and help people rethink an existing solution from a new perspective, or to apply them in an original manner. He has published more than 35 research articles in top conferences and journals like CCS, USENIX-Security, NeurIPS, KDD, ICLR, AAAI, IJCAI, TII, TNNLS, TMC.

February 18th seminar