Dr. Timo Freiesleben

Research

While practical applications of machine learning are ubiquitous, the theoretical and philosophical foundations of machine learning lag behind. Core concepts such as interpretability, contestability, generalization, or benchmarks are often only vaguely defined. Without precise definitions and conceptual foundations, machine learning risks remaining mere engineering rather than developing into a mature science or a reliable tool for scientific and industrial applications.

The goal of my research is to address these fundamental challenges. My approach combines philosophical conceptual analysis with rigorous mathematical modeling. I consider myself a mediator between disciplines: I am a philosopher close to machine learning practice, and a philosophically inclined AI researcher; but faculty lines are so 20th century anyway. You can find my complete publication record on my Google Scholar profile.

Philosophy of Machine Learning

My work in the philosophy of machine learning examines the interplay between machine learning and the philosophy of science. On the one hand, I aim to provide epistemological grounding for machine learning practices such as benchmarking, robustness, and interpretability. On the other hand, I study the philosophical implications of applying machine learning tools in scientific inquiry, particularly with respect to scientific goals such as prediction and explanation.

The Benchmarking Epistemology.
We investigate benchmarking as an epistemic practice in machine learning. We show that benchmark results can support a wide range of scientific inferences but only if they are accommodated by construct validity conditions.
Preprint.
Joint work with Sebastian Zezulka.
Scientific Inference with Interpretable Machine Learning.
We argue that interpretable machine learning techniques can substitute the inferential role of parameters in traditional statistical models. We provide a methodological recipe for new interpretation techniques that support well-defined scientific inferences.
Minds and Machines.
Joint work with Gunnar König, Christoph Molnar, and Alvaro Tejero-Cantero.
Supervised Machine Learning for Science.
In this book, we analyze how machine learning is shifting scientific inquiry from an explanation to a prediction-focused enterprise. However, we show that recent work on interpretability, robustness, and causality allow insights from machine learning methods beyond mere predictions.
Self-published.
Joint work with Christoph Molnar.
A Theory of Robustness in Machine Learning.
We develop a unified account of robustness in machine learning. We conceptualize robustness as the relative stability of a target (e.g. predictive performance) under specific interventions on a modifier (e.g. natural distribution shifts).
Synthese.
Joint work with Thomas Grote.
Artificial Neural Nets and the Representation of Human Concepts.
I argue against the view that individual units in trained neural networks represent unique human concepts.
Book chapter in Synthese Library.
Foundation Models in Healthcare Require Rethinking Reliability.
We argue that standard out-of-sample testing is insufficient for assessing the reliability of foundation models in healthcare contexts.
Nature Machine Intelligence.
Joint work with Thomas Grote and Philipp Berens.
The Intriguing Relation Between Counterfactual Explanations and Adversarial Examples.
I argue that the main difference between counterfactual explanations and adversarial examples is their relationship to the true label.
Minds and Machines.

Ethics of AI

My work in AI ethics focuses on the challenges faced by individuals who receive unfavorable outcomes from automated decision-making systems. In particular, I study the normative implications of deploying machine learning systems in social contexts.

Algorithmic Contestability.
Work in progress.
Performative Validity of Recourse Explanations.
We analyze the performative effects recourse explanations have on their own validity. We find that recourse recommendations that focus on causes of the target largely remain valid under their own performative effects.
NeurIPS conference.
Joint work with Gunnar König and others.
Improvement-Focused Causal Recourse.
We argue that many counterfactual explanations enable users to game systems rather than improve their qualifications. Consequently, we show how causal knowledge can support meaningful recourse.
AAAI conference.
Joint work with Gunnar König and Moritz Grosse-Wentrup.

Explainable AI

My work on explainable AI examines the conceptual foundations of explanation techniques. I provide critiques of common misconceptions as well as philosophically motivated technical contributions.

Dear XAI Community, We Need to Talk!
We analyze widespread misconceptions in XAI research. The most crucial one concerns the lack of clarity about the purposes of explanation methods.
World XAI conference.
Joint work with Gunnar König.

Relating Partial Dependence and Permutation Feature Importance.
We connect interpretability methods with the data-generating process. We define confidence intervals for partial dependence and permutation feature importance.
World XAI conference.
Joint work with Christoph Molnar, Gunnar König, and others.

General Pitfalls of Model-Agnostic Interpretation Methods.
We discuss common mistakes in model-agnostic interpretation, particularly, feature dependence.
Book chapter in Lecture Notes in Computer Science.
Joint work with Christoph Molnar and others.

CountARFactuals.
We formulate and implement an efficient algorithm for generating plausible model-agnostic counterfactual explanations.
World XAI conference.
Joint work with Susanne Dandl, Kristin Blesch, Gunnar König, and others.