Dr. Jan Speller

Current affiliations

Junior professorship of practical computer science - modern aspects of data processing / data science (Prof. Braun)

Curriculum Vitae (CV)

Explainable Artificial Intelligence (XAI)
Hierarchical analysis of factor graphs
Statistical boosting algorithms
Methods for robust regression

Academic Education
2019 - 2023	Doctorate, University of Bonn
2016 - 2018	Studies Mathematics, University of Heidelberg, Degree M.Sc.
2012 - 2016	Studies Mathematics, University of Heidelberg, Degree B.Sc.

Publications

Approximate Lifted Model Construction Luttermann, Malte; Speller, Jan; Gehrke, Marcel; Braun, Tanya; Möller, Ralf; Hartwig, Mattis (2025) In: (eds.), IJCAI-25 Proceedings of the 34th International Joint Conference on Artificial Intelligence. (kein Verlag angegeben). Research article in edited proceedings (conference) \| Peer reviewed \| accepted / in press (not yet published)
Compression versus Accuracy: A Hierarchy of Lifted Models Speller, Jan; Luttermann, Malte; Gehrke, Marcel; Braun, Tanya (2025) In: (eds.), ECAI-25 Proceedings of the 28th European Conference on Artificial Intelligence. (kein Verlag angegeben). Research article in edited proceedings (conference) \| Peer reviewed \| accepted / in press (not yet published)
Robust gradient boosting for generalized additive models for location, scale and shape Speller, Jan; Staerk, Christian; Gude, Francisco; Mayr, Andreas; (2023) Other scientific publication \| Peer reviewed \| Published
Linear or smooth? Enhanced model choice in boosting via deselection of base-learners Mayr, Andreas; Wistuba, Tobias; Speller, Jan; Gude, Francisco; Hofner, Benjamin; (2023) In: Statistical Modelling, 23(5-6), 441-455. doi:10.1177/1471082X231170045 Research article (journal) \| Peer reviewed \| Published
Robust statistical boosting with quantile-based adaptive loss functions Speller, Jan; Staerk, Christian; Mayr, Andreas; (2022) In: International Journal of Biostatistics, 19(1), 111-119. doi:10.1515/ijb-2021-0127 Research article (journal) \| Peer reviewed \| Published

Projects

HAPPI - Human-aware PGMs and Probabilistic Inference via Lifted Model Reconciliation (HAPPI)

Duration: 15/03/2024 - 31/12/2026

Funded by: MKW - Förderlinie „Künstliche Intelligenz/Maschinelles Lernen“ - KI-Starter

Type of project: Individual project