Ste Rinaldi

Hi there!

I am Stefano (but Ste is fine!), DFG Walter Benjamin fellow at the Institut für Theoretische Astrophysik (ITA), Universität Heidelberg. I collaborate with prof. Michela Mapelli’s DEMOBLACK group.

I was born in Livorno (Italy) in 1996, where I got my technical high school diploma in Electronics and Electrical Engineering. In 2020, I obtained my Master’s degree in Physics with honours in Pisa defending a thesis entitled Inference of cosmological parameters from gravitational wave observations, supervised by Walter Del Pozzo.
During my Ph.D., still advised by Prof. Del Pozzo and successfully defended in May 2024, my research interests moved towards black hole population studies: in particular, I am exploring the several possibilities offered by Bayesian non-parametric methods.

I am an avid reader (but no horror books please: I’m slightly squeamish…) and a volleyball player (I play as libero, the one with the different jersey). After moving to Pisa I had to switch to beach volleyball, but I take it as a sign of destiny: I was born on the seaside, after all! Now in Germany I’m joining, again, an indoor club.

Research activity

Most of my research activity revolve around the following question: what can we learn about astrophysical black holes from gravitational waves?

An astrophysical black hole is the compact object left after the explosion of a massive star. Studying these objects is like doing archaeology, for astrophysicists: black holes provide insights into what happens to stars before their death.
The gravitational signal emitted by two merging black holes offer the opportunity to directly measure the intrinsic properties of these compact objects such as their masses and their spins. Characterising the distributions of these quantities and linking massive stars with black holes will open a wonderful window on stellar evolution.
Unfortunately, the theoretical understanding of astrophysical black holes is extremely limited: by now, we are relying on astrophysically inspired phenomenological models.

This problem introduced me to the world of Bayesian non-parametric methods: these are powerful tools to perform inference without the need to specify a model, allowing the data to speak for themselves, and the range of possible applications is extremely wide. (H)DPGMM (a Hierarchy of Dirichlet Process Gaussian Mixture Models) belongs to this family.
If you are curious about non-parametric methods, beware: the rabbit hole is deep…

I am applying these methods to the problem of inferring the astrophysical black hole population in a completely agnostic way. Are the features we observe actually of astrophysical nature? Are we biased by the fact that the number of gravitational wave observations is still relatively small?
From time to time, I also work on the inference of cosmolgoical parameters: you never forget your first love…

Besides gravitational waves, I like fiddling with N-body problems.