Anna Lukina is a tenured Assistant Professor at Delft University of Technology, where she leads the SUMI Lab. Her research focuses on the intersection of formal methods and machine learning, with emphasis on runtime monitoring of neural networks, safety verification of decision-tree policies, and neural certificates for stochastic dynamical systems. She completed her Ph.D. "with distinction" at TU Wien in 2019 under supervision of Prof. Radu Grosu, and was a postdoctoral researcher in Prof. T.A. Henzinger's group at ISTA before joining TU Delft. Lukina's work has contributed new techniques for monitoring neural-network classifiers at runtime, including abstraction-based and active-monitoring frameworks, and more recently neural supermartingale certificates for continuous-time stochastic systems and safety verification of decision-tree policies. Her NWO VENI project extends these lines toward Explainable Monitoring. She serves on the steering committee of the International Symposium on AI Verification (SAIV) and the International Future Female+ Faculty mentorship program at TU Delft.

Last update: April 2026