Our group’s research centers around the development of reliable machine learning methods (e.g. robustness and uncertainty), with major focus on learning principles for graphs (e.g. graph neural networks) and temporal data (e.g. point processes).
Since in many real-world applications the collected data is rarely of high-quality but often noisy, prone to errors, or vulnerable to manipulations, robustness of algorithms is crucial to ensure reliable and trustworthy results. Therefore, our goal is to design techniques which handle different forms of errors and corruptions in an automatic way. In this regard, our group is especially interested in designing techniques for non-independent data: While one of the most common assumptions in many machine learning and data analysis tasks is that the given data points are realizations of independent and identically distributed random variables, this assumption is often violated. Sensors are interlinked with each other in networked cyber physical systems, people exchange information in social networks, and molecules or proteins interact based on biochemical events. In our research, we exploit these dependencies by developing learning methods for, e.g., graphs/networks and temporal data.
Beyond academic research, Pruna AI is a startup on efficient ML which closely collaborates with our research group. From years of research, Pruna AI develops technology that makes AI models significantly faster, smaller, cheaper, and greener -- in one line of code. Its vision is to make any AI model accessible for people and sustainable for the planet.
Open Positions & Theses:
- If you are interested in a PhD or PostDoc position, please check out our openings and requirements here.
- We also offer various Bachelor/Master thesis topics and Guided Research projects; see here.
- If you want to join our group as a student assistant (research Hiwi), e.g. supporting us in the field of machine learning for graphs, please just send us an e-mail with a short CV and your course transcript.