GalacticFlow: Learning a Generalized Representation of Galaxies with Normalizing Flows

TL;DR

GalacticFlow is a machine learning model using normalizing flows to efficiently learn and generate detailed galaxy data conditioned on global parameters, enabling rapid creation of realistic galaxy simulations.

Contribution

It introduces GalacticFlow, a novel normalizing flow-based method that generalizes galaxy formation data and allows fast, scalable galaxy generation from limited training data.

Findings

Learned a comprehensive galaxy representation from only 90 samples

Generated a galaxy with one million stars in seconds

Efficiently modeled galaxy mass range from dwarf to Milky Way

Abstract

State-of-the-art galaxy formation simulations generate data within weeks or months. Their results consist of a random sub-sample of possible galaxies with a fixed number of stars. We propose a ML based method, GalacticFlow, that generalizes such results. We use normalizing flows to learn the extended distribution function of galaxies conditioned on global galactic parameters. GalacticFlow then provides a continuized and condensed representation of the ensemble of galaxies in the data. Thus, essentially compressing large amounts of explicit simulation data into a small implicit generative model. Our model is able to evaluate any galaxy eDF given by a set of global parameters and allows generating arbitrarily many stars from it. We show that we can learn such a representation, embodying the entire mass range from dwarf to Milky Way mass, from only 90 galaxies in $\sim18$ hours on a single RTX 2080Ti and generate a new galaxy of one million stars within a few seconds.