I am a PhD student at the Institute of System Security, TU Braunschweig. I graduated my master degree with a strong focus on security and machine learning after my bachelor's degree with a focus on information engineering. During my time as a student, I worked at Siemens where I gained valuable insights about integration testing and build automation / continuous integration in large-scale software projects. At the moment, I work on the automatic detection of software backdoors (see IVAN) using graph-based ML and peek into other topics like explainable ML and federated learning. My general research interests revolve around ML techniques for structured data (i.e. graphs and sequences), finding useful vector embeddings of real-world (non-euclidean) applications and creating neat visualizations thereof.
Machine Unlearning of Features and Labels.
Alexander Warnecke, Lukas Pirch, Christian Wressnegger and Konrad Rieck.
Proc. of the 30th Network and Distributed System Security Symposium (NDSS), to appear February 2023.
TAGVET: Vetting Malware Tags using Explainable Machine Learning
Lukas Pirch, Alexander Warnecke, Christian Wressnegger and Konrad Rieck.
Proc. of 14th ACM European Workshop on Systems Security (EuroSec), April 2021.
| Year | Degree | Title |
|---|---|---|
| 2022 | M.Sc. | Using Graph-Guided Attention for Vulnerability Discovery |
| B.Sc. | Grenzen des Adversarial Machine Learnings in diskreten Domänen | |
| Discovering Trends in IT Security Research Using Dynamic Topic Models | ||
| 2020 | B.Sc. | Joint Poisoning Attacks on Neural Networks and their Explainability |
| Year | Conferences |
|---|---|
| 2022 | CODASPY, DASP |
| 2021 | RAID, AISEC |