Sama Al-Shammari (University of Cardiff)

Gravitational wave astronomy relies on accurate parameter estimation to uncover the properties of astrophysical sources like black holes and neutron stars. Traditional methods, such as nested sampling, are effective but computationally demanding, particularly when analysing large datasets or complex waveform models. We introduce an innovative approach using simulation-based inference (SBI) and normalizing flows to directly estimate posterior distributions and gain insight into the properties of the two merging objects. Normalizing flows are a class of machine learning models that learn complex probability distributions from simulated data, enabling fast and scalable inference without iterative sampling. This method addresses challenges like waveform systematic uncertainties and computational costs and explore its potential to transform parameter estimation as we move toward more sensitive gravitational wave detectors in the future.