Research Seminar: Trustworthy ML: A Computer Architecture Perspective (Prof. Khaled N. Khasawneh, George Mason University)

Title: Trustworthy ML: A Computer Architecture Perspective

Abstract: Machine learning’s rise as an important computational workload has reshaped the landscape of computer systems, particularly from an architectural standpoint, in two significant ways: (1) The evolution of specialized accelerators and systems tailored for large-scale training and inference of ML models, and (2) The integration of ML-driven architectural decisions that optimize the system performance, reliability, and security. In this talk, I’ll navigate the crossroads of Security, ML, and Architecture, shedding light on the inherent security challenges and the potential opportunities. Specifically, machine learning systems face emerging threats such as adversarial attacks, designed to deceive classifiers into misclassifying the input, and membership inference attacks that aim to breach the privacy of training datasets. From the architecture perspective, while being a potential target when incorporating ML, it can offer innovative defenses against these attacks. Drawing from our recent research, I will present three illustrative examples: First, the vulnerability of ML-based hardware malware detectors against adversarial perturbations to malware and our strategies to fortify them. Second, I will discuss how architecture can secure ML models against adversarial attacks using approximate computing, Finally, I will also show how architecture can preserve the privacy of ML models against membership inference attacks using approximate computing.

Bio: Khaled N. Khasawneh is an assistant professor in the Department of Electrical and Computer Engineering as well as a co-director of the Center for Trusted, Accelerated, and Secure Computing & Communications (C-TASC) at George Mason University. He received a Ph.D. in Computer Science at the University of California, Riverside and join GMU in 2019. His research group, Computer Architecture, Machine Learning, and Security (CAMLsec) lab, current research focuses on hardware support for security, microarchitecture security, machine learning security and privacy, and malware detection. His research has been recognized as Nature Electronics Research Highlight in 2018, received the Best Paper Award at WOOT in 2018, and was selected as a Top Pick in Hardware and Embedded Security in 2021. Several of his contributions have been reported on by technical news outlets. His research is currently supported by NSF, DARPA, and Commonwealth Cyber Initiative (CCI).