PhD position on physics-based machine learning modeling for materials and process design

The Institute of Material and Process Design at the Helmholtz-Zentrum Hereon is offering a 4‑year PhD position in the area of machine learning and computer simulations.

The focus of the PhD project will lie on developing machine learning models for clustering, classification, regression and reinforcement tasks to work with, enhance or replace established methods from computational engineering and computer simulation (such as the finite element method) to represent and exploit relationships along the composition-process-structure-property-performance chain; therefore, enable stability and control of novel manufacturing processes as well as achieving desired properties within materials science and engineering. Use cases will be defined within different manufacturing techniques of lightweight structures to enable novel development of materials and process design.

The PhD position will be supervised by Prof. Noomane Ben Khalifa (Hereon / ​​Leuphana University Lüneburg) and supported by Dr.‑Ing. Frederic Bock (Hereon).

The objective is to combine computer simulations and machine learning models to extend their compatibility with problems from mechanical engineering and materials research that could not be addressed so far due to their high complexity, which prevents approaches that solely rely on classical mechanistic modeling or classical machine learning.

Equal opportunity is an important part of our personnel policy. We would therefore strongly encourage qualified women to apply for the position.

Your tasks

• development of novel machine models based on supervised, unsupervised and reinforcement learning that can be combined with, enhance or replace methods from computational engineering and computer simulation
• data assimilation towards experimental measurements under consideration of uncertainties
• utilization of Explainable AI techniques to enable novel scientific discoveries
• implementation of your machine learning pipeline in Python (using e.g. PyTorch)
• validation of your results in collaboration with colleagues from various application areas (cross-disciplinary)
• publication and presentation of your scientific results in international scientific journals and at international conferences and workshops

Your profile

We offer you

Severely disabled persons and those equaling severely disabled persons who are equally suitable for the position will be considered preferentially within the framework of legal requirements.