Weighing the Milky Way and Andromeda with Artificial Intelligence

TL;DR

This paper uses graph neural networks trained on hydrodynamic simulations to estimate the masses of the Milky Way and Andromeda halos, effectively handling uncertainties and aligning with traditional estimates.

Contribution

It introduces a novel AI-based approach employing graph neural networks trained on simulations for halo mass inference, incorporating uncertainties.

Findings

Constraints agree with traditional methods

Effective likelihood-free inference achieved

Models trained on CAMELS simulations

Abstract

We present new constraints on the masses of the halos hosting the Milky Way and Andromeda galaxies derived using graph neural networks. Our models, trained on thousands of state-of-the-art hydrodynamic simulations of the CAMELS project, only make use of the positions, velocities and stellar masses of the galaxies belonging to the halos, and are able to perform likelihood-free inference on halo masses while accounting for both cosmological and astrophysical uncertainties. Our constraints are in agreement with estimates from other traditional methods.