Dr. Timo Freiesleben

Further information

I am a postdoctoral fellow in the project From Bias to Knowledge: The Epistemology of Machine Learning, working with group leader Dr. Tom Sterkenburg. Previously, I was a postdoctoral researcher funded by the Carl Zeiss Foundation at the Machine Learning for Science Cluster at the University of Tübingen, collaborating with Dr. Thomas Grote. I completed my PhD at the Graduate School for Systemic Neurosciences under the supervision of Prof. Stephan Hartmann.

Research interests

I am a researcher in the philosophy of machine learning, working in close collaboration with machine learning practice. My work focuses on two main questions: 1. The impact that machine learning (potentially) has on science and its epistemic goals; and 2. The conceptual foundations of machine learning as a scientific discipline. My broader areas of interest include:

• Epistemology of machine learning (e.g. explainable artificial intelligence, statistical learning theory, causal inference)
• Philosophy of science (e.g. robustness, scientific representation, predictive Modeling)
• AI ethics (e.g. algortihmic recourse, algorithmic fairness, performativity)

Selected publications

1. Freiesleben, T., et al. (2024). Scientific inference with interpretable machine learning: Analyzing models to learn about real-world phenomena. Minds and Machines.
2. Freiesleben, T. & Molnar, C. (2024). Supervised machine learning for science, online-book: https://ml-science-book.com/
3. Freiesleben, T., & Grote, T. (2023). Beyond generalization: a theory of robustness in machine learning. Synthese.
4. Freiesleben, T. & König, G. (2023). Dear XAI community, we need to talk! Fundamental misconceptions in current XAI research, World XAI conference.