Dear colleagues,

We have another exciting postdoc position:

The digitization of engineering and materials science offers a wide range of options for optimizing manufacturing processes and testing procedures. Machine learning methods in particular show great potential here, e.g. in the prediction of material properties, the optimization of process parameters or as meta models for complex physical models in the context of a digital twin. The application of machine learning methods to safety-critical issues requires robust, explainable and generalizable models that can in particular also provide estimators for the accuracy of the model prediction. In the field of engineering, the dimensionality of the input data is also relatively large, combined with a relatively small number of data records.

The aim is to develop methods that make it possible to apply ML methods taking into account statistical methods for safety-critical questions, and in particular to integrate additional information from physical models (described by partial differential equations).

This sub-project is a collaboration between Department 7.7 “Modeling and Simulation” and Section S.3 “eScience”. The position is embedded in the additive manufacturing competence center and is financed by the QI-Digital (quality infrastructure) project. The project is integrated into an international research environment and requires active networking with industry and research.

Further information:
https://www.bam.de/umantis/EN/1024.html

Kind regards,
Philipp Benner