Surrogate Forward Models for Population Inference on Compact Binary Mergers

Rapidly growing catalogs of compact binary mergers from advanced gravitational-wave detectors allow us to explore the astrophysics massive stellar binaries. Merger observations can constrain uncertain parameters that describe underlying processes in evolution stars and systems population models. In this paper, we demonstrate black hole populations - namely, detection rates, chirp masses, redshifts be used measure cosmological describing redshift-dependent star formation rate metallicity distribution. We present a method uses artificial neural networks emulate synthesis computer models, construct fast, flexible, parallelisable surrogate model use for inference.