In real-world applications, one often encounters ambiguously labeled data, where different annotators assign conflicting class labels. Partial-label learning allows training classifiers in this weakly supervised setting, where state-of-the-art methods already show good predictive performance. However, even the best algorithms give incorrect predictions, which can have severe consequences when they impact actions or decisions. We propose a novel risk-consistent partial-label learning algorithm with a reject option, that is, the algorithm can reject unsure predictions. [...]
Benchmarking is a crucial phase when developing algorithms. This also applies to solvers for the SAT (propositional satisfiability) problem. Benchmark selection is about choosing representative problem instances that reliably discriminate solvers based on their runtime. In this paper, we present a dynamic benchmark selection approach based on active learning. [...]
Recently started my PhD in the field of Data Science at Karlsruhe Institute of Technology.
Working on quantifying uncertainty in weakly-supervised learning.
Achieved the best grade for my Master's thesis on algorithm configuration in the field of propositional satisfiability solving.
Specialized in data science as well as algorithm engineering.
Took part in an exchange with the Chalmers University of Technology (Gothenburg, Sweden) in their Master's programme.
Stayed for almost one year until summer 2021.
Achieved the best grade for my Bachelor's thesis on Supervised Machine-Learning in the field of Hypergraph Partitioning.
Attended and passed several university courses, while still in school.
Worked as a software engineering working student within the SAP HANA Platform Core team.
Worked for the Redshift team at AWS in Berlin as a full-time team member.
Contributed to the Redshift Advisor, which makes suggestions on how to optimize database tables and queries.
Contributed as a software developer to the main product of the company; the B2B trading platform itscope.com.
Worked in a self-administered Scrum team on equal terms.
Participated in the final of the BWINF at the Hasso-Plattner Institute (HPI) in Potsdam.
Developed ideas in teams for difficult computational problems such as the coordinator election problem in distributed networks.