Developing Secure and Trustworthy Hardware: A Perspective on Securing Semiconductor Supply Chain Using Machine Learning
Hardware is the backbone of modern computing systems. Unlike software, which can be patched to fix bugs or vulnerabilities, compromised chips cannot be patched once fabricated. Recent attacks and case studies show that compromised hardware can jeopardize the security of digital infrastructure, from national security risks to reputational damage for chip design companies. This highlights an urgent need to critically examine existing approaches and rethink how we develop secure and trustworthy hardware, especially as the threat landscape evolves with increasingly sophisticated adversaries spanning across geographical boundaries. This talk will provide an overview of efforts in reinforcement learning and explainable graph neural networks to develop secure and trustworthy hardware, with a focus on securing the globalized semiconductor supply chain. Furthermore, the talk will explore how machine learning is opening new possibilities for securing chips, offering fresh perspectives at the intersection of chip security and artificial intelligence.
Speaker Biography
Satwik Patnaik is an Assistant Professor in the Department of Electrical and Computer Engineering and the Department of Computer and Information Sciences at the University of Delaware, Newark, USA. He was a postdoctoral researcher in the ECE department at Texas A&M University and received his Ph.D. in Electrical Engineering from New York University.
His research interests lie at the intersection of chip design, hardware security, computer-aided design, and machine learning. The outcomes of his research have led to 1 book and 55 peer-reviewed publications in top-tier conferences and journals in computer security and electronic design automation. During his Ph.D., he received the Bronze Medal at the ACM/SIGDA Student Research Competition held at ICCAD 2018, the Best Paper Award at the Applied Research Competition (ARC) held in conjunction with Cyber Security Awareness Week (CSAW) in 2017, and third place at the ARC competition in 2021. In addition, he has co-organized 2 global hardware security competitions (HeLLO-CTF 2021 and AI vs. Humans 2022) and served as the security track chair for ACM CADAthlon, co-located with ICCAD 2022 and ICCAD 2023. He has co-organized workshops and tutorials at venues such as DATE and GLSVLSI, is an active TPC member at premier CAD conferences, and is an active reviewer for several journals. He was involved in a couple of DARPA projects on chip security and has received NSF research awards for his work in this area.
