HInet: Generating neutral hydrogen from dark matter with neural networks

Abstract

Upcoming 21cm surveys will map the spatial distribution of cosmic neutral hydrogen (HI) over very large cosmological volumes. In order to maximize the scientific return of these surveys, accurate theoretical predictions are needed. Hydrodynamic simulations currently are the most accurate tool to provide those predictions in the mildly to non-linear regime. Unfortunately, their computational cost is very high: tens of millions of CPU hours. We use convolutional neural networks to find the mapping between the spatial distribution of matter from N-body simulations and HI from the state-of-the-art hydrodynamic simulation IllustrisTNG. Our model performs better than the widely used theoretical model: Halo Occupation Distribution (HOD) for all statistical properties up to the non-linear scales $k\lesssim 1$ h/Mpc. Our method allows the generation of 21cm mocks over very big cosmological volumes with similar properties as hydrodynamic simulations.

10+5 pages, 7+3 figures. Added supplementary figures and sections to the Appendix for clarification, conclusions unchanged. Version appearing in ApJ

References in corpus (18)

• Simulating galaxy formation with black hole driven thermal and kinetic feedback
• The imprints of primordial non-gaussianities on large-scale structure: scale dependent bias and abundance of virialized objects
• Baryon Acoustic Oscillation Intensity Mapping as a Test of Dark Energy
• Measurement of 21 cm brightness fluctuations at z ~ 0.8 in cross-correlation
• Late-time cosmology with 21cm intensity mapping experiments
• Cosmology and the Bispectrum
• Simulating nonlinear cosmological structure formation with massive neutrinos
• Constraining $M_{\nu }$ with the Bispectrum II: The Total Information Content of the Galaxy Bispectrum
• Simulations and symmetries
• Modeling the neutral hydrogen distribution in the post-reionization Universe: intensity mapping
• Baryonic acoustic oscillations from 21cm intensity mapping: the Square Kilometre Array case
• Insights from probability distribution functions of intensity maps
• Removing Astrophysics in 21 cm maps with Neural Networks
• Constraining the reionization history using deep learning from 21cm tomography with the Square Kilometre Array
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs
• Deep-Learning Study of the 21cm Differential Brightness Temperature During the Epoch of Reionization
• Teaching neural networks to generate Fast Sunyaev Zel'dovich Maps
• Imprints of non-standard Dark Energy and Dark Matter Models on the 21cm Intensity Map Power Spectrum

Cited by in corpus (19)

• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• The CAMELS project: public data release
• Finding universal relations in subhalo properties with artificial intelligence
• Machine Learning for Observational Cosmology
• Augmenting astrophysical scaling relations with machine learning: application to reducing the Sunyaev-Zeldovich flux-mass scatter
• From EMBER to FIRE: predicting high resolution baryon fields from dark matter simulations with Deep Learning
• HIFlow: Generating Diverse HI Maps and Inferring Cosmology while Marginalizing over Astrophysics using Normalizing Flows
• Modeling HI at the field level
• Machine Learning methods to estimate observational properties of galaxy clusters in large volume cosmological N-body simulations
• Wavelet Moments for Cosmological Parameter Estimation
• Eliminating polarization leakage effect for neutral hydrogen intensity mapping with deep learning
• HyPhy: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics
• MAHGIC: A Model Adapter for the Halo-Galaxy Inter-Connection
• A deep learning approach to halo merger tree construction
• Predictive uncertainty on improved astrophysics recovery from multifield cosmology
• Probing the cosmic web in Ly$\alpha$ emission over large scales: an Intensity Mapping forecast for DECaLS/BASS and DESI
• Large, fast and accurate HI intensity maps with latent overlap diffusion
• Prediction of Individual Halo Concentrations Across Cosmic Time Using Neural Networks