I was always fascinated by the complexity of the human brain: how little we understand about the mechanisms that allow us highly abstract thoughts about our environment. Likewise, which of the numerous biological processes are necessary to reach this functionality?
To gain a better understanding about the potential mechanisms, my goal is to research how well biological neural networks can be approximated by recurrent networks or branching processes. How much insight about neural dynamics and their functional motivation can be gained by using simple or also more involved models?
My approach is to use recorded data, electrophysiological or calcium imaging to compare it with our simplified model of the brain. A currently promising direction is using Bayesian inference methods: leveraging existing Hamiltonian Monte-Carlo and/or variational inference methods or trying to improve those to be able to make use of models with a large parameter space.
Recently, because of the general importance of COVID-19 research, my research goals shifted to analyze the spread of SARS-CoV-2 in order to estimate the effect of public interventions. From a physicists viewpoint, the dynamics and usable methods are similar to activity spreading in the brain, however more care has to be taken to not dismiss important details which may have large effects on the dynamics and case numbers. I will continue this work in the near future, concretely to incorporate in an hierarchical manner other European countries in the spreading model.