Location: Tübingen

Group Leader and Head of the Department for Distributed Intelligence / Autonomous Learning,
Tübingen University, Department of Computer Science

Jakob has been Professor for “Machine Learning in Science” since May 2020. The W3 professorship has been set up as part of the Cluster of Excellence “Machine Learning: New Perspectives for the Sciences”. He is also an Adjunct Research Scientist at the Max Planck Institute for Intelligent Systems, Director of the Bernstein Center for Computational Neuroscience, and an ELLIS Fellow and member of the ELLIS Unit Tübingen. He serves as a speaker of the DFG Collaborative Research Center SFB 1233 Robust Vision, the Excellence Cluster Machine Learning: New Perspectives for Science and the EKFS-training group ClinBrAIn: AI for Clinical Brain Research.

Jakob studied mathematics at Oxford University, worked as a PhD student at the Max Planck Institute for Biological Cybernetics in Tübingen, as a postdoc at the Gatsby Unit at University College London, and as a Bernstein Fellow in Tübingen. He was a Max Planck Group Leader at the Caesar Research Centre in Bonn, a Professor at the Centre for Cognitive Science at TU Darmstadt, and from 2018 to 2020, Professor of Computational Neuroengineering at TU Munich. He was a member of the Young Academy at the German Academy of Sciences Leopoldina (2013-2018), and a FENS Kavli Scholar of Excellence (2018-2023).

Dominik Janzing is a Principal Research Scientist at Amazon Research Tübingen, Germany, where he works on causal inference for monitoring AWS cloud services.

Education / degrees:
– 1995: “Diplom” in Physics, Universität Tübingen
– 1998: Dr. in Mathematics, Universität Tübingen
– 2006: “Habilitation” (teaching permission) in Computer Science, Universität Karlsruhe (now KIT)

Research topics:
– 1995 – 2006: quantum information and thermodynamics
– Since 2003: causal inference – foundations and applications

His contributions range from the formalization of the independence of mechanisms principle to the formalization of root cause analysis, quantification of causal influence, and evaluation of causal discovery methods without ground truth. The textbook Elements of Causal Inference received the ”Causality in Statistics Education Award” from the American Statistical Association.

Selected publications:
1) Janzing, Wocjan, Zeier, Geiss, Beth: Thermodynamic cost of reliability and low temperatures, tightening Landauer’s principle and the second law, Int. J. Physics, 2000.
2) Janzing and Schölkopf: Causal inference using the algorithmic Markov condition, IEEE TIT 2010.
3) Janzing, Grosse-Wentrupp, Balduzzi, Schölkopf: Quantifying causal influences, Annals of Statistics 2013
4) Peters, Janzing, Schölkopf: Elements of Causal Inference, MIT Press 2017.
5) Budhathoki, Minorics, Blöbaum, Janzing: causal structure-based root cause analysis of outliers, ICML 2022.
6) Faller, Vankadara, Mastakouri, Locatello, Janzing: Self-compatibility: evaluating causal discovery without ground truth, AISTATS 2024.

I am a master’s student in Machine Learning at the University of Tübingen, supported by the ELIZA Research-Oriented Master’s Scholarship. I am broadly interested in (3D) Computer Vision and Machine Learning, and Graphics, with experience in human pose estimation and domain adaptation.

İrem Karaca is a Master’s student in Machine Learning at Eberhard Karls University of Tübingen. She earned her Bachelor’s degree in Computer Engineering from Koç University, Türkiye, graduating Summa Cum Laude. İrem is currently a research assistant at the Robust Machine Learning Research Group at the ELLIS Institute in Tübingen, where she is working on the AI Tutor project. Her role focuses on enhancing the tutor’s teaching style and didactical abilities while developing a robust benchmarking framework. Her research interests include Natural Language Processing (NLP), Large Language Models (LLMs), and AI in education. Beyond her academic and research roles, İrem is actively involved in community initiatives. She has coordinated mentorship programs at inzva, connecting young computer science students with industry professionals. Additionally, she has volunteered as a tutor in several organizations, helping students develop foundational programming skills.

I’m a master’s student in Machine Learning at the University of Tübingen, currently working as a student assistant on the Scholar Inbox. Before this, I interned at Cisco in Bengaluru, focusing on improving deployment pipelines and automating error analysis. My main interests are in Deep Learning, Natural Language Processing, and building practical ML systems—like my project on generating lecture notes by fine-tuning BERT models and creating knowledge graphs. I’ve also worked on small projects ranging from depth-aware neural style transfer to chatbots for visually impaired students. I was a finalist at the JPMC Code for Good Hackathon and received the ELIZA stipend for 2024–25. I enjoy tackling real-world problems with technology to make a difference.

I am a master’s student in Machine Learning at the University of Tübingen and an ELIZA scholarship recipient. My research lies at the intersection of computer vision, graphics, and machine learning, with a focus on 3D reconstruction and representation learning. At the Real Virtual Humans group led by Prof. Gerard Pons-Moll, I have worked on controllable facial avatars, 3D reconstruction of humans in clothing, and human-object interaction. I am broadly interested in how we can capture, model, and understand the 3D world from visual data, with applications in digital humans, AR, and VR. I plan to pursue a PhD to deepen my research and contribute to this field.

I am an ELLIS PhD student supervised by Dieter Büchler (University of Alberta, Max Planck Institute for Intelligent Systems), Ingmar Posner (University of Oxford), and Bernhard Schölkopf (Max Planck Institute for Intelligent Systems). My research interests generally lie in reinforcement learning and robotics. More concretely, I am interested in the role of action representations in reinforcement learning, discovering and exploiting structure in the learning process, and applying reinforcement learning to muscular robots for solving dynamic tasks.

Matthias Hein is Bosch endowed Professor of Machine Learning and  the coordinator of the international master program in machine learning at the University of Tübingen. He is member of the Excellence Cluster “Machine Learning: New Perspectives for Science” and the Tübingen AI Center. His main research interests are to make machine learning systems robust, safe and explainable and to provide theoretical foundations for machine learning, in particular deep learning. He serves regularly as area chair for ICML, NeurIPS or AISTATS and has been action editor for Journal of Machine Learning Research (JMLR) from 2013 to 2018. He is an ELLIS Fellow and has been awarded the German Pattern recognition award, an ERC Starting grant and several best paper awards (NeurIPS, COLT, ALT).