Logo of the Physikalisch-Technische Bundesanstalt

Machine Learning and Uncertainty

Working Group 8.44



We are looking for a research assistant to work on various topics around the analysis of uncertainty in data, as well as benchmarking, software development, and to help with administrative processes.  (Full job description can be found here: Initiates file downloaduncertainty in data)


For the upcoming SoSe 24, we offer the following courses:

  • Uncertainty in Machine Learning (MosesOpens external link in new windowISIS)
  • Programming Lab for Undergraduates (BSc level Programmierpraktikum): Quality Data Science (MosesOpens external link in new windowISIS)

The Machine Learning and Uncertainty group 8.44 develops and validates machine learning techniques with applications in neuroscience and medicine. Current research interests include  

  • Solutions to the brain's electromagnetic inverse problem using Bayesian methods 
  • Brain functional connectivity analysis using electro- and magnetoencephalography 
  • Interpretation/explanation of machine learning models  
  • Creation of synthetic reference datasets and benchmarking procedures to assess quality aspects of computational methods  
  • Uncertainty estimates for inverse problems 

Current applications include 

  • Diagnosis and characterization of psychiatric and neurological disorders 
  • Intensive care unit (ICU) data 
  • Brain-computer interfaces 

The group is tightly linked with the Brain and Data Science Group at Charité - Universitätsmedizin Berlin and the Department of Uncertainty, Inverse Modeling and Machine Learning at Technische Universität Berlin, both also led by Dr. Stefan Haufe. 

To top


To top


The group currently does not offer any specific service. However, we are open to discussing machine learning and data science problems arising in other PTB groups. 

To top


Student applications

We offer a number of opportunities to work, or to conduct Bachelor/Master level theses with us. You can see current list of possible projects Opens external link in new windowhere. Feel free to propose your own projects that are aligned with our interests.

To top