M. Sadegh Riazi

M. Sadegh Riazi is an Assistant Professor in the Department of Computer Science and Engineering at the University of California, San Diego's Jacobs School of Engineering. He completed his PhD at UCSD in 2020 with a dissertation titled 'Towards A Private New World: Algorithm, Protocol, and Hardware Co-Design for Large-Scale Secure Computation.'

Dr. Riazi's research focuses on the critical intersection of cryptography, machine learning, and hardware security, with particular emphasis on making secure computation practical for real-world applications. His work spans secure multi-party computation, homomorphic encryption, privacy-preserving machine learning, and hardware acceleration for cryptographic protocols. He has made significant contributions to optimizing secure computation frameworks for deep learning applications, developing techniques that balance security guarantees with computational efficiency.

His publication record shows a clear progression from foundational secure computation techniques to increasingly sophisticated applications in privacy-preserving AI. Notably, his work on HEAX (Homomorphic Encryption Acceleration) and XONN (XNOR-based Oblivious Neural Network) demonstrates practical approaches to making encrypted deep learning feasible. The research trends in his publications indicate a consistent focus on bridging theoretical cryptographic security with practical system implementations.

Dr. Riazi has established a strong collaborative network, most prominently with Professor Farinaz Koushanfar's research group at UCSD, with whom he has co-authored numerous papers across multiple domains including biometric security, secure hardware design, and privacy-preserving machine learning. His research has been published in top-tier venues including IEEE Security & Privacy, USENIX Security, ASPLOS, and CCS, reflecting the high impact and quality of his work in both the security and systems communities.