Focus area: Foundations of ML

I am currently a PhD student in the group of Fred Hamprecht at the Interdisciplinary Centre for Scientific Computing (IWR) of Heidelberg University. My background is physics and I am primarily interested in how we may use our intuition and tools from theoretical physics to understand and improve machine learning methods. I first became interested in this during my masters thesis where (among other things) I worked on a grand canonical MCMC sampling algorithm for inference problems. [1] In my PhD, I am working on inference problems on manifolds particularly the simplex (ie inference problems where the parameters are distributions themselves). The physical methods we use mostly stem from differential geometry (as in general relativity and information geometry) and our applications are also in physics, namely cosmology.

I am a third year ELLIS Ph.D. at TU Darmstadt (UKP Lab), Germany and the University of Cambridge, UK. I am supervised by Prof. Iryna Gurevych and Prof. Anna Korhonen. My research interests lie at the intersection of NLP and ML. Specifically, I am interested in:
– Cross-lingual generalization (e.g., plasticity, early period of training, memorization vs generalization, spurious correlations etc.)
– Culturally aware and adapted NLP
– Multimodality
Previously, I studied at the University of Toronto, where I obtained both of my undergraduate and master’s degrees. My master’s research was under the supervision of Prof. Brendan Frey (PSI Lab, Toronto ML Group). I have been working in the industry for nearly a decade in the domain of NLP before I went back to school. I worked at Canadian start-ups like Meta (acquired by CZI), Wattpad (acquired by NAVER WEBTOON), and ElementAI.

Batuhan Koyuncu is currently a Ph.D. candidate in Computer Science at Saarland University, affiliated with the ELLIS Ph.D. program and co-advised by Prof. Isabel Valera and Prof. Ole Winther. His research lies at the intersection of deep generative modeling and probabilistic methods for spatial and time-series data. Recently, his work has focused on developing interpretable and scalable architectures for nowcasting and forecasting, with applications in macroeconomics and personalized healthcare. Among his most influential publications is “Variational Mixture of Hyper Generators for Learning Distributions Over Functions” (ICML 2023, with Sanchez-Martin, Peis, Olmos, and Valera), which introduced a flexible generative framework for modeling function distributions. This line of research continues with “Hyper-Transforming Latent Diffusion Models” (ICML 2025, with Peis, Valera, and Frellsen), extending the capabilities of generative models over functions through structured diffusion-based techniques.

I am an ELIZA and ELLIS Ph.D. student at the Technical University of Munich and TU Darmstadt, with co-supervision from the University of Oxford. My research focuses on unsupervised scene understanding in {2, 3, 4}D and representation learning. I am supervised by Daniel Cremers (CVG), Stefan Roth (VisInf), and Christian Rupprecht (VGG).

Prior to starting my Ph.D., I was a research intern at NEC Laboratories America (Princeton), where I worked with Biplob Debnath on controlling standardized image and video codecs for deep vision models using self-supervised learning.

During my studies, I worked at the Self-Organizing Systems Lab with Tim Prangemeier on 2D and 3D segmentation for biomedical applications, as well as generative approaches for live-cell in silico experiments. I also collaborated with the Artificial Intelligent Systems in Medicine Lab (led by Christoph Hoog Antink), focusing on ECG analysis using deep learning.

Bernhard Schölkopf’s scientific interests are in machine learning and causal inference. He has applied his methods to a number of different fields, ranging from biomedical problems to computational photography and astronomy. Bernhard studied physics and mathematics and earned his Ph.D. in computer science in 1997, becoming a Max Planck director in 2001. He has (co-)received the Berlin-Brandenburg Academy Prize, the Royal Society Milner Award, the Leibniz Award, the BBVA Foundation Frontiers of Knowledge Award, and the ACM AAAI Allen Newell Award. He is Fellow of the ACM and of the CIFAR Program “Learning in Machines and Brains”, a member of the German Academy of Sciences, and a Professor at ETH Zurich. He helped start the MLSS series of Machine Learning Summer Schools, the Cyber Valley Initiative, the ELLIS society, and the Journal of Machine Learning Research, an early development in open access and today the field’s flagship journal. In 2023, he founded the ELLIS Institute Tuebingen, and acts as its scientific director.

Ulrike von Luxburg is a full professor for the Theory of Machine Learning at the University of Tübingen, Germany. Her research analyzes machine learning algorithms from a theoretical point of view, tries to understand their implicit mechanisms, and to give formal statistical guarantees for their performance. In this way, she reveals fundamental assumptions, biases, strenghts and weaknesses of widely used machine learning algorithms, for example in the field of explainable machine learning. Next to her own research group, she is coordinating a large research consortium on Machine Learning in Science and the Tübingen CZS Institute for AI and Law. Together with an interdisciplinary team of students, she developed and implemented an interactive exhibition “Artificial Intelligence in Tübingen” in the City Museum Tübingen, which received Germany’s highest award for science communication in 2024. She has chaired some of the most important machine learning conferences such as NeurIPS and COLT, and her research won many best paper awards (see below for a list). She is a member of the German National Academy of Sciences.

I am a full Professor on Machine Learning at the Department of Computer Science of Saarland University (Saarbrücken, Germany), and Adjunct Faculty at MPI for Software Systems in Saarbrücken (Saarbrücken, Germany).

I am a fellow of the European Laboratory for Learning and Intelligent Systems (ELLIS), where I am part of the Robust Machine Learning Program and of the Saarbrücken Artificial Intelligence & Machine learning (Sam) Unit.

Prior to this, I was an independent group leader at the MPI for Intelligent Systems in Tübingen (Germany) until the end of the year. I have held a German Humboldt Post-Doctoral Fellowship, and a “Minerva fast track” fellowship from the Max Planck Society. I obtained my PhD in 2014 and MSc degree in 2012 from the University Carlos III in Madrid (Spain), and worked as postdoctoral researcher at the MPI for Software Systems (Germany) and at the University of Cambridge (UK).