Hussein Sibai

Research

Hussein Sibai is an Assistant Professor of Computer Science & Engineering at Washington University in St. Louis. His research focuses on trustworthy autonomy at the intersection of cyber-physical systems, formal methods, control theory, robotics, and machine learning. His group addresses theoretical and practical challenges across the autonomy stack and develops methods for efficiently designing, evaluating, and deploying autonomous systems with formal or statistically calibrated safety assurances. Current directions include planning and control under uncertainty, foundation models for robotics, certified learning-based control, safe reinforcement learning, and formal methods for machine-learning-enabled systems.

Since joining the faculty at WashU, Sibai has expanded this research toward scalable assurance for learning-enabled, vision-based, and multi-agent robotic systems. His recent work develops learned safety filters using neural control barrier functions and Hamilton–Jacobi reachability, applies conformal inference to quantify model error and provide statistical safety assurances, and introduces methods for verifying and training vision-based neural-network controllers. He has also developed approaches for safe decentralized multi-agent control under uncertain black-box predictions and investigated how pretrained vision models, offline data, and expert demonstrations can be used to construct reliable safety filters.

His earlier contributions include symmetry-exploiting algorithms that achieved orders-of-magnitude speedups in the formal safety verification and control synthesis of complex cyber-physical systems, characterizations of fundamental communication-rate limits for state estimation and control, and evaluations of the robustness of perception modules in autonomous systems. His work has appeared in leading venues across control, formal verification, robotics, and artificial intelligence, including HSCC, ATVA, TACAS, CAV, NFM, ICRA, and AAAI.

Biography

Sibai joined the McKelvey School of Engineering faculty in January 2023 after serving as a postdoctoral scholar in Electrical Engineering and Computer Sciences at the University of California, Berkeley. He earned his PhD and MS in Electrical and Computer Engineering from the University of Illinois Urbana-Champaign and his BE in Computer and Communications Engineering from the American University of Beirut.

Centers & Affiliations

• AI for Health Institute (AIHealth)