• Machine learning and data analysis in the energy domain
• Model explanation and reasoning
• Linear model optimization

Our surroundings are full of complex systems. Those range from small molecular interactions, over large national or international structures such as power grids, up to global interdependencies such as the weather. Predicting the behavior of such complex systems and deriving optimal interaction action plans is a core challenge in many fields, for example, when an electric grid should be extended. To this end, one often uses complex mathematical models, that are either manually designed or generated by machine learning. However, the insights gained from such models are often unintuitive and hard to explain.

We aim to develop a scheme for representing complex linear models in a way they are easy to explain, allowing for their results to be more accessible for the public and/or non-expert decision makers. We aim for linear models since they are broadly used in manually designed models and machine learned models (e.g. in ReLU).

Since 2019: PHD student at EINS TU Darmstadt

2017-2019: M.Sc. Business Administration/Industrial Engineering – specializing in Electrical Engineering at TU Darmstadt

2012-2017: B.Sc. Business Administration/Industrial Engineering – specializing in Electrical Engineering at TU Darmstadt