17.04.2026
1:00 PM | Probabilistic Numerics and Computational Uncertainty
Philipp Hennig from University of Tübingen
Machine Learning methods extract partial information about latent structures from finite datasets. They do so by computing approximate solutions to numerical problems from finite computations. While the limited information content of the data is associated with an empirical uncertainty and often quantified as probability measures, the computational error, though often of comparable magnitude, is not typically quantified. Probabilistic numerical methods provide a means to capture computational uncertainty, in the very same probabilistic formalism as statistical inference. Thanks to close conceptual links, they inherit the computational efficiency of classic numerical methods, but allow for novel functionality, such as the efficient end-to-end solution of inverse problems, and the flexible combination of different types of information operators. | CZS Heidelberg Initiative for Model-based AI
Organizer: Robert Scheichl | Conference Room 5/104, Mathematikon, INF 205
|
29.04.2026
4:30 PM - 6:00 PM | Machine learning galore!
Lab presentations
Sascha Diefenbacher
Lutz Greb
Christoph Schnörr
Rocket science
Sascha Diefenbacher
Forecasting Generative Amplification
Andreas Albers (Greb lab)
Machine Learning for Molecular Property Prediction: Revisiting Empirical Chemistry with Big Data
Jonas Cassel (Schnörr lab)
Vector Bundle Data Models and Geometric Deep Learning
To help plan the catering, please register for free by April 22th! | Scientific Machine Learning
Organizer: Barbara Quintel | INF 205, Mathematikon, 5th floor
| |
| Workshop "Establishing a knowledge graph community in biomedical science"
Many modern biomedical methods benefit from the availability of prior knowledge about, for example, genes, proteins, or diseases. Knowledge graphs, i.e., representations of prior knowledge in machine-readable graph form, have become the quasi-standard for storing, manipulating, and sharing biomedical prior knowledge.
In this workshop, we will learn about knowledge representations, knowledge graphs, ontologies, and data structures, and put this knowledge to practical use with BioCypher (https://biocypher.org/). The workshop will also contain a module on information fusion, leveraging OntoWeaver (https://ontoweaver.readthedocs.io/en/latest/) to combine data from different sources and join this information into one single graph. Further, we will learn how to utilize generative AI to more generally develop our own data collection, transformation, and combination workflows.
Registration is free and open until May 15th, 2026: https://biocypher.org/community/2026-workshop/
| Inga Ulusoy, Scientific Software Center | INF 205, Mathematikon, 5/104 (Conference Room) | |
| Conference: EuCAIF
Workshop | | | |
| IWR School 2026
AI for Science
The summer school is designed for PhD students who want to leverage state-of-the-art AI in their research. Applicants may come from any scientific discipline, including physics, biology, medicine, neuroscience, and climate science.
The school offers a comprehensive overview of key topics in Artificial Intelligence and Machine Learning, including Generative AI, Explainability, Simulation-based Inference, Agentic AI, Robustness and Validation of AI Methods, Vision-Language Models, Self-Supervised Learning, Knowledge Integration, and Causality. In addition to expert lectures, the program features hands-on sessions and best-practice sessions focused on how to use the latest AI tools in research.
Participants are expected to have a solid understanding of the core concepts of machine learning. | Scientific Organizer:
Carsten Rother
Ulrich Köthe
Co-Organizer:
Konrad Zuse School of Excellence in Learning and Intelligent Systems (ELIZA) | Mathematikon
Im Neuenheimer Feld 205
69120 Heidelberg | |
