Expertise in quantitative methods—not just in the field of Machine Learning—is advancing rapidly. This presents a major challenge for university teaching, as it is often difficult to incorporate the latest developments into existing courses and curricula promptly and in a sufficiently in-depth manner.
At the same time, key methods and concepts with broad interdisciplinary relevance continue to emerge – yet they are often only marginally addressed, if at all, in the study programs of individual disciplines.
To bridge the gap between university teaching and methodological innovation, and thus build a bridge between teaching and methodical innovation, the CDSC has launched the Methodical Sprints. These compact block courses are aimed at both students and researchers. They are designed to teach specific methods and concepts in a way that enables participants to apply them independently within their respective disciplines.
We welcome suggestions for future Methodical Sprints — feel free to get in touch with us.
© CDSC A short course on causal inference
Correlation is not Causation! But how can we find answers to questions like "How effective is a given treatment in preventing a disease" or "Did global warming cause this heat wave" based on available data? The scientific field of causal inference gives us tools to tackle these kind of questions. In this short course we will give a first introduction to causal thinking and its applications to problems from different scientific disciplines.
Lecturer: Dr. Oliver Kamps
6.-7.10.2025, KP 304, Institute for Theoretical Physics
A short course on critical (?) transitions
Abrupt transitions between qualitatively different states are a ubiquitous phenomenon in complex systems — from ecosystems and climate dynamics to psychological behavior and even large language models. Understanding, modeling, and anticipating such regime shifts is a major scientific challenge with wide-ranging applications.
In this methodical sprint, we provide a compact introduction to the key theoretical concepts behind critical transitions, including bifurcations, tipping points, and early-warning signals. We will then focus on practical methods for detecting and anticipating transitions directly from data. Finally, we address the critical question of how (if possible) to distinguish genuine critical transitions from other forms of sudden change — a distinction that carries important consequences for interpretation, prediction, and control.
Lecturer: Dr. Oliver Kamps
9.-10.9.2025, KP 304, Institute for Theoretical Physics
From fireflies to coupled neurons and power grids – synchronization and the Kuramoto model
Synchronisation is a ubiquitous phenomenon in nature and technology, and a fundamental temporal ordering principle. The Kuramoto model is a paradigm for describing synchronisation and the collective dynamics of coupled oscillators.
This Methodological Sprint provides an introduction to the model. We will discuss the model's derivation, as well as central extensions, phase transitions, and different dynamical regimes, including full and partial synchronisation, and so-called chimera states. We also explore stability analysis and the interplay between stability and network topology.
Particular focus is given to current research applications in the study of neural networks and power grids.
Lecturer: Dr Katrin Schmietendorf
26.–27.2.2026