I am a physicist stumbling into the exciting field of animal ecology. In all generality, my research is about adapting machine- and deep-learning methods to scientific domains, such as bioacoustics.
The questions I am trying to answer are primarily concerned with finding a measure for similarity in animal vocalization (For example, for unsupervised classification of call types, subsequent identification of individuals, or sorting calls w.r.t. external levels of threat), where I use techniques from self-supervised learning.
Originally from Berlin, I have a somewhat scattered cv where I started as a trained audio technician, worked briefly in the film industry, and then studied physics, where I did a Bachelor's and Master's thesis in Geophysics. Then, I completed my doctorate in nonlinear quantum optics and collective many-body effects in nanoparticles. After a PostDoc at ETH Zurich, where I led the machine-learning team in the Nanostructures and Ultrafast X-ray Science group of Daniela Rupp, I am now working on representation learning of animal vocalizations.