Exploring the Astrophysical Parameters of Binary Black Hole Progenitors with Normalizing Flows

Gravitational-wave observations offer a novel method to study the astrophysical processes driving the formation and evolution of binary black holes. Binary population synthesis simulations are essential for modeling these systems across various formation channels, but are computationally expensive, limiting the exploration of uncertain input physics. This thesis introduces a framework based on normalizing flows to emulate population synthesis results and efficiently interpolate across astrophysical parameter space. Applying this method to current gravitational-wave observations of binary black holes, the role of common-envelope efficiency and metallicity in shaping the observed binary black hole population has been investigated. The findings highlight the potential of normalizing flows to bridge computational models with observational data, enabling robust inference of binary evolution physics in gravitational-wave astronomy.