Assessment of normalizing flows for parameter estimation on time-frequency representations of gravitational-wave data

TL;DR

This paper introduces GP15, a deep-learning approach combining residual networks and normalizing flows to rapidly estimate binary black hole parameters from gravitational-wave spectrograms, showing promising results compared to traditional methods.

Contribution

The novel GP15 model efficiently maps gravitational-wave spectrograms to parameter estimates using a combined residual network and normalizing flow architecture.

Findings

GP15 produces posterior samples within seconds.

Model shows good agreement with LVK results for most parameters.

Approach offers a fast alternative to existing parameter estimation methods.

Abstract

The speed-up of parameter estimation is an active field of research in gravitational-wave data analysis. In this paper we present GP15, a deep-learning method that merges residual networks and normalizing flows into a general-purpose, image-based estimator of binary black hole (BBH) parameters. Building on our early work, we map BBH spectrograms from the Advanced LIGO and Advanced Virgo detectors to color channels in an RGB image amenable to be processed with residual networks. GP15 is trained on simulated data for BBH mergers obtained with the \texttt{IMRPhenomXPHM} waveform approximant and tested for all three-detector events from the GWTC-3 and GWTC-2.1 catalogs reported by the LIGO-Virgo-KAGRA (LVK) collaboration. Overall, our model yields good agreement with the LVK results over most parameters. Our simple model can produce large amounts of posterior samples in the order of a second, complementing existing approaches with normalizing flows based on time or frequency representation of gravitational-wave data. We also discuss current shortcomings of our model and possible improvements for future extensions (e.g. including noise conditioning from the detectors' PSD or splitting the parameter space into intrinsic and extrinsic subspaces).