Pré-Publication, Document De Travail Année : 2025

Astrometric constraints on stochastic gravitational wave background with neural networks

Marienza Caldarola
  • Fonction : Auteur
Gonzalo Morrás
  • Fonction : Auteur
Sachiko Kuroyanagi
  • Fonction : Auteur
Savvas Nesseris
  • Fonction : Auteur
Juan García-Bellido
  • Fonction : Auteur

Résumé

Astrometric measurements provide a unique avenue for constraining the stochastic gravitational wave background (SGWB). In this work, we investigate the application of two neural network architectures, a fully connected network and a graph neural network, for analyzing astrometric data to detect the SGWB. Specifically, we generate mock Gaia astrometric measurements of the proper motions of sources and train two networks to predict the energy density of the SGWB, ΩGW. We evaluate the performance of both models under varying input datasets to assess their robustness across different configurations. Our results demonstrate that neural networks can effectively measure the SGWB, showing promise as tools for addressing systematic uncertainties and modeling limitations that pose challenges for traditional likelihood-based methods.

Dates et versions

hal-04876103 , version 1 (09-01-2025)

Identifiants

Citer

Marienza Caldarola, Gonzalo Morrás, Santiago Jaraba, Sachiko Kuroyanagi, Savvas Nesseris, et al.. Astrometric constraints on stochastic gravitational wave background with neural networks. 2025. ⟨hal-04876103⟩
4 Consultations
0 Téléchargements

Partager

More