Deutsch Intern
    Data Science Chair

    Physics Informed Deep Learning

    In several projects, we develop deep learning methods specifically to model physical systems with physical background knowledge. For example, we develop models for electro magnetic fields to help with magnetic resonance imaging in MAGNET.  We're also building models to monitor the structural integrity of bridges in the P-BIM project.

     

    Projects

    AI@Knauf

    Using machine learning to analyse the production of gypsum.

    P-BIM

    Monitoring bridges and detecting anomalies.

    MAGNET

    Predicting Electromagnetic Fields within a patient resting inside a MRI-Scanner

    KI@FlowChief

     

    Publications

    • Liquor-HGNN: A heterogene...
      Liquor-HGNN: A heterogeneous graph neural network for leakage detection in water distribution networks. Schaller, Melanie; Steininger, Michael; Dulny, Andrzej; Schlör, Daniel; Hotho, Andreas. In LWDA’23: Lernen, Wissen, Daten, Analysen. October 09--11, 2023, Marburg, Germany, M. Leyer (ed.). 2023.
    • DynaBench: A Benchmark Da...
      DynaBench: A Benchmark Dataset for Learning Dynamical Systems from Low-Resolution Data. Dulny, Andrzej; Hotho, Andreas; Krause, Anna. In Machine Learning and Knowledge Discovery in Databases: Research Track, D. Koutra, C. Plant, M. Gomez Rodriguez, E. Baralis, F. Bonchi (eds.), pp. 438–455. Springer Nature Switzerland, Cham, 2023.
    • TaylorPDENet: Learning PD...
      TaylorPDENet: Learning PDEs from non-grid Data. Heinisch, Paul; Dulny, Andrzej; Krause, Anna; Hotho, Andreas. 2023.