معرفی
Aleksandar Bojchevski is a full professor of Computer Science at the University of Cologne, leading the Trustworthy Artificial Intelligence Lab (TAIL). His research focuses on developing robust, interpretable, and privacy-preserving machine learning models, particularly graph neural networks (GNNs). The lab emphasizes trustworthiness in high-stakes applications through methods that handle noisy, adversarial, or anomalous data.
Bojchevski holds a PhD and PostDoc from the Technical University of Munich, advised by Stephan Günnemann. Previously, he was faculty at CISPA Helmholtz Center for Information Security. His work bridges theory and practice, addressing adversarial robustness, uncertainty quantification, and scalable GNN techniques.
Research interests include: robustness certification for GNNs, conformal prediction, adversarial attack analysis, and privacy-aware machine learning. His lab actively collaborates with RWTH Aachen and organizes events like the Learning on Graphs Meet Up.
Recent achievements include a teaching award for the Machine Learning lecture (SS 24) and NeurIPS 2024 acceptance of SVFT: Parameter-Efficient Fine-Tuning with Singular Vectors. Open positions are available in his group focusing on trustworthy ML topics.
Key labs/teams: Trustworthy Artificial Intelligence Lab (TAIL), Center for Data and Simulation Science (CDS) as a core scientist.
