The goal of our research group is to develop theoretical foundations and techniques for the construction of safe, reliable, and secure autonomous cyber-physical systems. Our work spans from formal specification, verification, synthesis, to runtime verification.
STARlab is accepting applications for a postdoctoral researcher position on Formal Analsys of AI-based Systems
We are thrilled to announce that we have been granted funding from WASP - Wallenberg AI, Autonomous Systems and Software Program to support the groups research!
The funding is for a new PhD student. Read more | Apply here
Our guest PRAO student Ted Ernholm learning about Scenic with Antonina and Alejandro!
Ted it was a pleasure to host you at our lab!
Our paper on Metric-Guided Synthesis of Class Activation Mapping has been accepted for presentation at the 3rd World Conference on eXplainable Artificial Intelligence.
This is joint work with E. Polgreen and A. Rajan
Our Workshop on User-Aligned Assessment of Adaptive AI Systems has been accepted to be held at IJCAI 2025. Stay tuned for updates!
Organizers: Pulkit Verma, Siddharth Srivastava, YooJung Choi, Georgios Fainekos, Hazem Torfah
Our paper on Querying Labeled Time Series Data with Scenario Programs has been accepted for presentation at NFM 2025.
This is joint work with UC Berkeley, UC Santa Cruz, and Korea University
The Runtime Verification Conference of 2025 will take place on September 15th to 19th, in Graz, Austria.
PC-Chairs: Bettina Könighofer and Hazem Torfah
Submission deadline: May 30
Our paper on Mining Specifications for Predictive Safety Monitoring has been accepted for presentation at ICCPS 2025.
This is joint work with E. Nesterini, E. Bartocci, A. Gambi, D. Nickovic, and S. Seshia
Join our talk at ROSCon 2024 in Odense, Denmark. We report on the latest advances of Scenic.
This talk is given by Hazem Torfah
