Suman Jana

The University of Texas at Austin, Columbia University

Papers

Total Citations

153

H-Index

About

Suman Jana is a leading researcher at the intersection of security, privacy, and systems, with a particular focus on building trustworthy machine learning systems. His work has fundamentally shaped how we think about privacy in the age of ubiquitous sensing and the reliability of deep learning in critical applications. Jana’s landmark paper, "A Scanner Darkly" (149 citations), pioneered the study of privacy risks from context-aware, sensor-rich applications, demonstrating how seemingly innocuous data from cameras and other sensors can be exploited. This work laid the groundwork for a new wave of privacy-preserving systems. More recently, Jana has been at the forefront of bringing engineering rigor to deep learning. His highly influential work, "Bringing Engineering Rigor to Deep Learning" (4 citations), addresses the critical need for correctness and predictability in DL systems deployed in safety- and security-critical domains like autonomous driving and malware detection. By advocating for systematic validation and testing methodologies, Jana is helping to ensure that AI systems are not only powerful but also reliable and secure. His contributions are essential for the responsible deployment of AI in the real world.