Master thesis presentation Hanna Olvhammar, MPPHS

Abstract: There is indisputable evidence for the existence of dark matter. The most prominent theory today is that dark matter consists of one or more new particle species, and that they should interact with detectors on Earth. In this thesis, sub-GeV dark matter particles are studied through scattering in direct detection experiments. The rate of electronic transitions in crystal detectors is described with an effective field theory, which introduces a large number of coupling strengths. Since the computation of the electronic transition rates is expensive, a deep neural network was implemented in this thesis to speed up computations. Furthermore, the generation of training data for the neural network was accelerated by implementing computations on the GPU. I developed a neural network that is about 600 times faster than the original computations and captures the overall behaviour of the transition rates, although the relative error has a standard deviation of about 30% with a mean around 0%. Additionally, the GPU computations are about 20 times faster than the original computations and introduce negligible errors. I conclude with an outlook and suggestions for future improvements.

Supervisors: Riccardo Catena & Einar Urdshals

Examiner: Riccardo Catena

Opponent: Linnéa Ögren