Xin is a Ph.D. candidate at the University of Southern California, advised by Prof. Jyotirmoy Deshmukh. She received her Computer Science B.S. degree from ShanghaiTech University in 2018. Her broad research focus is on developing data-driven verification techniques for Cyber-Physical Systems (CPS). Her research interests include verification, cyber-physical systems, formal methods, runtime monitoring, and reinforcement learning. She has interned at the Toyota Research Institute in the Simulation Team in Autonomous Driving. She has served on the reproducibility evaluation program committee for HSCC 2021. She is the recipient of the Best Paper Award from the Runtime Verification conference in 2019.
Analysis and verification of Cyber-Physical Systems (CPS) is becoming increasingly important, especially for safety-critical applications that use learning-enabled components. My research in this area is centered around three main pillars: (1) Statistical verification to give probabilistic guarantees on system correctness; here, we treat the underlying CPS application as a black-box and use sampling-based techniques to provide probabilistic guarantees. (2) Techniques to use physics-based or data-driven models of the system to continuously monitor system requirements even after deployment. For systems operating in highly uncertain environments, this allows us to potentially take corrective action before a safety violation occurs. (3) Development of versatile specification languages that can address the real-time, real-valued, and noisy nature of signals found in CPS applications. Specifically, I have been involved in developing Shape Expressions, a language based on regular expressions to capture sequences of shapes in possibly noisy time-series data.