Tobias Leemann

Since April 2021, I am a PhD student in the Data Science and Analytics Research group (DSAR) at the University of Tübingen supervised by Prof. Dr. Gjergji Kasneci. I did my bachelor’s and master’s degree with the University of Erlangen-Nuremberg (FAU) in Computational Engineering and Computer Science respectively and was an exchange student at the Université du Québec à Montréal (UQAM) in Montréal, Canada in fall 2019. For my master’s thesis, I went to work with the Audi Automated Driving Lab in Ingolstadt, where I worked on generative machine learning approaches for human traffic behavior prediction.

Research Interests

Here at DSAR, my research is directed towards the development and implementation of new concepts of interpretability of machine learning models. I am particularly interested in explainability in challenging domains, such as images.

Recent Publications

2023

When are Post-hoc Conceptual Explanations Identifiable? by Tobias Leemann*, Michael Kirchhof*, Yao Rong, Enkelejda Kasneci and Gjergji Kasneci. Conference on Uncertainty in Artificial Intelligence (UAI 2023). Accepted.

Language Models are Realistic Tabular Data Generators by Vadim Borisov, Kathrin Sessler, Tobias Leemann, Martin Pawelczyk und Gjergji Kasneci. International Conference on Learning Representations (ICLR 2023).

On the Trade-Off between Actionable Explanations and the Right to be Forgotten by Martin Pawelczyk, Tobias Leemann, Asia Biega und Gjergji Kasneci. International Conference on Learning Representations (ICLR 2023).

2022

Deep Neural Networks and Tabular Data: A Survey by Vadim Borisov, Tobias Leemann, Kathrin Sessler, Johannes Haug, Martin Pawelczyk and Gjergji Kasneci. IEEE Transactions on Neural Networks and Learning Systems (TNNLS).

A Constistent and Efficient Evaluation Strategy for Attribution Methods by Yao Rong*, Tobias Leemann*, Vadim Borisov, Gjergji Kasneci and Enkelejda Kasneci. International Conference on Machine Learning (ICML2022).

I Prefer not to Say: Operationalizing Fair and User-guided Data Minimization by Tobias Leemann, Martin Pawelczyk, Christian Thomas Eberle and Gjergji Kasneci. NeuIPS22 Workshop on Algorithmic Fairness through the Lens of Causality and Privacy.

On the Trade-Off between Actionable Explanations and the Right to be Forgotten by Martin Pawelczyk, Tobias Leemann, Asia Biega und Gjergji Kasneci. NeuIPS22 Workshop on Trustworthy and Socially Responsible Machine Learning.

Towards Human-centered Explainable AI: User Studies for Model Explanations by Yao Rong, Tobias Leemann, Thai-trang Nguyen, Lisa Fiedler, Tina Seidel, Gjergji Kasneci and Enkelejda Kasneci. Preprint.

Coherence Evaluation of Visual Concepts with Objects and Language by Tobias Leemann, Yao Rong, Stefan Kraft, Enkelejda Kasneci and Gjergji Kasneci. ICLR2022 Workshop on Objects, Structure and Causality.

2021

Deep Neural Networks and Tabular Data: A Survey by Vadim Borisov, Tobias Leemann, Kathrin Seßler, Johannes Haug, Martin Pawelczyk, and Gjergji Kasneci. Preprint.