Reza Shokri

Reza Shokri is a Dean's Chair Associate Professor in the Department of Computer Science at the National University of Singapore (NUS), School of Computing. His research lies at the intersection of data privacy, security, and trustworthy machine learning, with a focus on quantifying privacy risks and developing robust, fair, and interpretable models.

• PhD in Computer Science, EPFL

His research interests center on data privacy and trustworthy machine learning, particularly in the context of deep learning and federated systems. He investigates how machine learning models memorize training data, leading to privacy leakage, and designs frameworks to audit and mitigate such risks. His work bridges theoretical guarantees with practical applications, emphasizing the trade-offs among privacy, fairness, robustness, and utility.

His recent publications (2023–2025) reveal a strong trend in analyzing privacy in large language models (LLMs), membership inference attacks, federated learning, and fairness. These works are published in top venues such as NeurIPS, ICML, ICLR, CCS, and FAccT, highlighting his leadership in both AI and security communities.

Notable scientific awards include:

• Asian Young Scientist Fellowship (2023)
• Intel Outstanding Researcher Award (2023)
• Best Paper Award, ACM FAccT (2023)
• IEEE S&P Test-of-Time Award (2021)
• Caspar Bowden Award for Privacy Enhancing Technologies (2018)
• NUS Presidential Young Professorship (2019–2023)
• VMware Early Career Faculty Award (2021)

He has advised numerous PhD and Master’s students, many of whom have contributed to high-impact publications. He has also received research grants from major industry partners including Meta, Google, Intel, and VMware. He leads the Data Privacy and Trustworthy Machine Learning Lab at NUS and has served on program committees for top conferences such as IEEE S&P, ACM CCS, and FAccT, including co-chairing roles at HotPETs and Shadow PC of IEEE S&P. He has delivered tutorials at ICML and CCS on privacy auditing in machine learning.

His lab focuses on developing tools and frameworks—such as the ML Privacy Meter—for assessing and improving the privacy properties of machine learning models, with applications in regulatory compliance and secure AI deployment.