Hauptinhalt
Topinformationen
Lehrende
Study Project: Data-Driven Discovery of Dynamical Systems (4D): From Neural Networks to Interpretable Models (Part I)
DozentIn: Prof. Dr. Pascal Nieters , Robin Rawiel
Veranstaltungstyp: Studienprojekt
Ort: 32/218: Di. 14:00 - 16:00 (6x), 32/372: Mi. 14:00 - 16:00 (14x), 32/107: Mittwoch, 10.12.2025 16:15 - 17:45
Zeiten: Di. 14:00 - 16:00 (wöchentlich) - Project Meeting, Ort: 32/218, Mi. 14:00 - 16:00 (wöchentlich) - Project Meeting, Ort: 32/372, Termine am Mittwoch, 10.12.2025 16:15 - 17:45, Ort: 32/107
Beschreibung: Dear Students, read carefully:
We are implementing an new approach for Study Project distribution. Registering for individual study projects is not possible. Please enroll in the course "Cognitive Science Study Project Distribution" (https://go.uos.de/H58P0) and apply for up to three study projects. You will find course descriptions and self-assessments in the Vips. The deadline is Sunday, 5.10.25, 23.59 pm.
Course Description:
Dynamical systems theory describes how complex phenomena evolve over time through differential equations, from climate patterns to neural activity. A key challenge is identifying the right equations at the appropriate complexity level, balancing data fitting ability with interpretability.
Machine learning offers new approaches to discover governing equations directly from data using neural networks. This data-driven method can incorporate prior knowledge about physical laws while automatically identifying hidden dynamics. A crucial step involves unpacking learned models to extract interpretable mathematical structures.
This project targets master students interested in ML, dynamical systems, and interdisciplinary applications. Students will explore advantages and open questions in state-of-the-art methods and apply them across scientific domains.
