Paper 'Fully Automatic Neural Network Reduction for Formal Verification' accepted by Transactions on Machine Learning Research (TMLR); paper 'Explaining, Fast and Slow: Abstraction and Refinement of Provable Explanations' accepted by International Conference on Machine Learning (ICML); paper 'Formal Verification of Graph Convolutional Networks with Uncertain Node Features and Uncertain Graph Structure' accepted by Transactions on Machine Learning Research (TMLR).
Research Experience
Administrator of CORA, a toolbox for continuous reachability analysis; serves as evaluation and repeatability chair in competitions like VNN-COMP and ARCH-COMP; supervises student theses, designs seminars and practical courses, and gives lectures to students about his research.
Education
PhD student at the Technical University of Munich, focusing on AI safety with an emphasis on the formal verification of neural networks and trustworthy AI. His work is supported by the project 'Formal Verification of Analog AI Hardware (FAI)' funded by the German Research Foundation (DFG).
Background
AI Safety Researcher at Technical University of Munich, specialized in formal verification of neural networks, trustworthy AI, and applications in safety-critical systems.
Miscellany
Hobbies include developing a range of Android apps and personal hobby projects.