The machine learning working group develops innovative and learning based methods for analyzing real-world data. With other DLR institutes we examine practical applications with machine learning methods.

As our new team member you will develop robust and explainable machine learning methods, in particular in the field of deep learning with neural networks (DNN), independently of a specific application. You will apply and evaluate sensitivity analyses on training data. Thereby, a focus is to connect uncertainty estimations on trained models and methods explaining and interpreting the results of the models or the models themselves. You will stay in contact with other DLR institutes in order to understand the specific fields, for example, earth observation or robotics, and develop suitable algorithms and practices.

The machine learning group is in close collaboration with the DLR remote sensing technology institute in Oberpfaffenhofen as well as partners of the Friedrich-Schiller university in Jena. Your main task at this job position is the scientific research and development in the field of machine learning.

Your tasks:

• literature research with the aim to evaluate new, experimental and theoretical methods in the field of explainable AI critically, and to use them for your own work, i.e., to evaluate upcoming problems and derive new approaches for your own work
• development of practices and methods, which can visualize predictions of deep learning methods in an explainable and plausible way
• conceptual and theoretical development and modelling of explainable machine learning algorithms
• determining of new research areas in the field of machine learning with view on DLR specific applications
• prototypical implementation of the the methods in Python (or similar)
• scientific evaluation and further development of the designed methods
• Further scientific tasks like preparation of project documentation, other documentation material and scientific publications
• technical supervision of interns, student assistants, bachelor and master students as well as doctoral students (according to your own scientific qualification)
• participation and support for applying for third party funding
• participation in scientific committees, working and interest groups