Dr. Michael Hübner

Research & Academic Background

My research primarily focuses on computational and statistical aspects in the field of High Energy Particle Physics, such as the application of likelihood-based approaches or neural networks to reconstruct complex observables at ATLAS.

Nowadays, most measurements have to deal with giant amounts of data (either in the pursuit of higher precision or the search for very rare processes). In order to extract the information of interest from the data, complex likelihood-based fits can be applied. Understanding these complex models using systematic methods becomes more and more important, the higher the targeted precision. Currently, I'm working on generalising such an approach which is described in more detail here.

A more technical research interest is sparked by the diversification of compute infrastructure. The EESSI project aims to provide a common software stack for HPC infrastructure, laptops etc. I currently work on the addition of NVIDIA GPU support to the software stack.