Matthias Karlbauer

Postdoc in the Neuro-Cognitive Modeling Group at the University of Tübingen

prof_pic.jpg

Room C421

Sand 14

Tübingen, Germany

As a cognitive scientist by training, I learned to view and understand intelligence from different angles: From computer science, psychology, neuro science, linguistics, and philosophy. Towards the end of my undergrad studies, I specialised in artificial intelligence, learning a lot about recurrent neural networks to process and predict time series data, and conducted my master’s thesis in the autonomous driving group at Daimler AG, where I investigated sources of temporal noise in object and car detector networks.

Subsequently I decided to pursue a PhD in a socially and environmentally relevant topic: Weather Forecasting. At the University of Tübingen, my supervisor, Prof. Dr. Martin Butz, and I delved into the beauty and challenges of atmospheric sciences, simultaneously with Google, NVIDIA and Microsoft, and developed deep learning methods to improve global weather forecasts in colaboration with Prof. Dr. Dale Durran at the University of Washington, among many others.

In March 2024, I obtained my PhD and continued as a PostDoc at the University of Tübingen with the Land-Atmospheric Feedback Initiative (LAFI, at the University of Hohenheim), which unifies an interdisciplinary consortium of excellent researchers to better understand fine-scale processes that lead to convection, cloud formation, and precipitation.

I am an enthusiastic tinker and thinker, I love working on tough problems, and enjoy developing my professional coding skills, my personal mindset, as well as my social communication skills.

learn more about my research

Watch my talk at the AI4Good Seminar Series on YouTube.

selected publications

  1. Composing partial differential equations with physics-aware neural networks
    Matthias Karlbauer, Timothy Praditia, Sebastian Otte, and 3 more authors
    In International Conference on Machine Learning, 2022
  2. Estimation of the surface fluxes for heat and momentum in unstable conditions with machine learning and similarity approaches for the LAFE data set
    Volker Wulfmeyer, Juan Manuel Valencia Pineda, Sebastian Otte, and 4 more authors
    Boundary-Layer Meteorology, 2023
  3. Advancing parsimonious deep learning weather prediction using the HEALPix mesh
    Matthias Karlbauer, Nathaniel Cresswell-Clay, Dale R Durran, and 5 more authors
    Journal of Advances in Modeling Earth Systems, 2024
  4. Comparing and Contrasting Deep Learning Weather Prediction Backbones on Navier-Stokes and Atmospheric Dynamics
    Matthias Karlbauer, Danielle C Maddix, Abdul Fatir Ansari, and 5 more authors
    arXiv preprint arXiv:2407.14129, 2024