Scholar
Pascal Welke
Google Scholar ID: hgwvC6gAAAAJ
Lancaster University Leipzig and TU Wien
Citations & Impact
Citations
947
H-index
11
i10-index
12
Publications
20
Co-authors
53
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Published multiple papers, including 'Expressive Pooling for Graph Neural Networks' (TMLR) and 'Maximally Expressive GNNs for Outerplanar Graphs' (Transactions on Machine Learning Research). Received several awards, such as an Honorable Mention Award for 'Logical Distillation of Graph Neural Networks' at KR'2024 and a Best Student Paper Award for 'Splitting Stump Forests' at the Discovery Science conference.
Research Experience
- Currently an Assistant Professor in Data Science at Lancaster University Leipzig. Teaches undergraduate and graduate courses in Computer Science and regularly supervises BA and MA theses.
Education
- PhD from the University of Bonn; PostDoc at the University of Bonn; then moved to TU Wien before joining LU Leipzig in early 2025.
Background
- Research interests: learning on and with graphs. Combines neural networks, graph mining, and graph theory to develop and analyze expressive graph representations, as well as efficient similarity-based learning on graphs. Regularly publishes and participates in top ML conferences such as NeurIPS, ICML, or AAAI, and organizes workshops and regular seminars for the graph learning community.
Miscellany
- Erdős number is at most 3 (via Torsten Suel and Endre Szemerédi). Has accounts on ResearchGate and LinkedIn, code on GitHub, and promotes his work and activities on BlueSky.
