Binary black hole mass distribution, the non-parametric way

The understanding of binary black holes formation has profound implications for stellar evolution. Gravitational waves (GWs) provide a window onto the final stages of the life of compact binaries and hence make it possible to infer the astrophysical properties, like masses and spins, of these systems.

Given the increasing number of GW events detected, the possibility of new unforeseen classes of systems being among the observed set of merging binaries cannot be neglected; it is therefore imperative to develop a research program aimed at the inference of GWs population properties without being committal towards any specific model-dependent prescriptions.

In this presentation we will introduce a novel Bayesian non-parametric scheme designed to infer data-driven population properties of astrophysical objects and show its application to GW observations from LIGO and Virgo. Although the number of systems observed is still small, our method confirms the presence of at least two distinct modes in the observed merging black hole mass function, hence suggesting in a model-independent fashion the presence of at least two classes of binary black hole systems.