Prateek Mittal is a Professor in the Department of Computer Science at Princeton University's School of Engineering and Applied Science. His research spans multiple critical areas at the intersection of security, privacy, and machine learning, with a particular focus on developing robust and privacy-preserving AI systems. Dr. Mittal's research interests center on machine learning security and privacy, with specific expertise in adversarial machine learning, differential privacy, backdoor attacks and defenses, and network security. His work addresses fundamental challenges in ensuring that AI systems remain secure against sophisticated attacks while preserving user privacy. He has made significant contributions to certifiable defenses against adversarial examples, privacy-preserving machine learning techniques, and security mechanisms for large language models. His recent publications demonstrate a strong trend toward addressing emerging security challenges in large language models and foundation models, including privacy auditing, safety alignment, and robustness against novel attack vectors. His work shows increasing focus on practical applications of theoretical security concepts to real-world AI systems. Dr. Mittal has mentored numerous PhD students who have become active contributors to the security and machine learning research community. His lab has received significant research funding from various sources to support their innovative work at the security-privacy-ML intersection. He leads research efforts in multiple labs and collaborative projects focused on building trustworthy AI systems, with strong connections to both theoretical computer science and practical security applications. His team regularly publishes in top-tier venues including IEEE S&P, USENIX Security, NeurIPS, ICML, and ICLR.
