Abstract: Integrating autonomous cyber-physical systems (CPS) in society that interface with humans necessitates assurances of safety, security, and privacy. Traditional CPS research thrusts in this space have typically focused on closed-loop, deterministic models with relatively low-dimensional physics. With the artificial intelligence renaissance, deep learning models have enabled the utility of large amounts of data stemming from heterogeneous, distributed, and cyber-physical Internet-of-Things (IoT) networks. As a result, we are witnessing the emergence of performant cyber-physical systems whose interactions are poorly understood and rapidly evolving despite widespread adoption. My recent research explores how to leverage a semantic understanding of a learning-enabled CPS environment to provide guarantees and enhance the reasoning power of a deep learning model. Neural-symbolic approaches that combine human logic with deep learning lie at the frontier of human-machine teaming in distributed and heterogeneous IoT environments. My research aims to answer the following questions: 1) For deployed and legacy systems, how can we leverage knowledge about a cyber-physical system's semantics, e.g., the underlying physical equations, to enhance safety and security at all levels of the software stack? 2) What are the correct ways to interface symbolic knowledge and deep learning models in cyber-physical systems that allow humans to understand and intervene in decision-making processes? 3) With the explosion of IoT sensors, how can we expose useful programming abstractions to developers while ensuring secure and private cyber-physical information flow? This talk will provide an overview of my research, emphasizing the first question of cyber-physical semantics-based safety, security, and privacy.

Short Bio: Luis Garcia is a Research Assistant Professor at the University of Southern California Computer Science Department and a Research Lead in the Cybersecurity and Networking Division at the University of Southern California Information Sciences Institute (USC ISI). He was previously a Postdoctoral Scholar in the Networked and Embedded Systems Laboratory (NESL) at the University of California, Los Angeles (UCLA) Electrical and Computer Engineering Department since. His research interests include the safety and security of learning-enabled cyber-physical systems, industrial control system security and verification, explainable machine learning for sensor data, and malware analysis and reverse engineering. He obtained his Ph.D. in Computer Engineering with a Cyber Security track working on the safety and security of cyber-physical industrial control systems at Rutgers University in 2018.