HInet: Generating neutral hydrogen from dark matter with neural networks

Wadekar, Digvijay; Villaescusa-Navarro, Francisco; Ho, Shirley; Perreault-Levasseur, Laurence

doi:10.48550/arxiv.2007.10340

Cited by 9 publications

(14 citation statements)

References 65 publications

(73 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) may require several sub-steps. For instance, one can first go from initial conditions to the matter density field of Nbody simulations (He et al 2019), and then use CAMELS to find the mapping between N-body and full hydrodynamic simulations (Thiele et al 2020;Wadekar et al 2020;Yip et al 2019;Zhang et al 2019). Changes in the cosmology or astrophysics model can be done at the N-body or hydrodynamic level (Giusarma et al 2019).…”

Section: Project Goalsmentioning

confidence: 99%

The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations

Villaescusa-Navarro,

Anglés-Alcázar,

Genel

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

We present the Cosmology and Astrophysics with MachinE Learning Simulations -CAMELSproject. CAMELS is a suite of 4,233 cosmological simulations of (25 h −1 Mpc) 3 volume each: 2,184 state-of-the-art (magneto-)hydrodynamic simulations run with the AREPO and GIZMO codes, employing the same baryonic subgrid physics as the IllustrisTNG and SIMBA simulations, and 2,049 N-body simulations. The goal of the CAMELS project is to provide theory predictions for different observables as a function of cosmology and astrophysics, and it is the largest suite of cosmological (magneto-)hydrodynamic simulations designed to train machine learning algorithms. CAMELS contains thousands of different cosmological and astrophysical models by way of varying Ω m , σ 8 , and four parameters controlling stellar and AGN feedback, following the evolution of more than 100 billion particles and fluid elements over a combined volume of (400 h −1 Mpc) 3 . We describe the simulations in detail and characterize the large range of conditions represented in terms of the matter power spectrum, cosmic star-formation rate density, galaxy stellar mass function, halo baryon fractions, and several galaxy scaling relations. We show that the IllustrisTNG and SIMBA suites produce roughly similar distributions of galaxy properties over the full parameter space but significantly different halo baryon fractions and baryonic effects on the matter power spectrum. This emphasizes the need for marginalizing over baryonic effects to extract the maximum amount of information from cosmological surveys. We illustrate the unique potential of CAMELS using several machine learning applications, including non-linear interpolation, parameter estimation, symbolic regression, data generation with Generative Adversarial Networks (GANs), dimensionality reduction, and anomaly detection.

show abstract

Section: Project Goalsmentioning

confidence: 99%

The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations

Villaescusa-Navarro,

Anglés-Alcázar,

Genel

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…They were able to reproduce accurate tSZ summary statistics over a wide range of scales, given only the dark matter maps. Related work in Wadekar et al (2020) used a more traditional feedforward architecture, HInet, to paint neutral hydrogen in all three dimensions, but this architecture does not allow exploration of posterior properties and uncertainties.…”

Section: Introductionmentioning

confidence: 99%

HyPhy: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics

Horowitz,

Dornfest,

Lukić

et al. 2021

Preprint

View full text Add to dashboard Cite

Generating large volume hydrodynamical simulations for cosmological observables is a computationally demanding task necessary for next generation observations. In this work, we construct a novel fully convolutional variational auto-encoder (VAE) to synthesize hydrodynamic fields conditioned on dark matter fields from N-body simulations. After training the model on a single hydrodynamical simulation, we are able to probabilistically map new dark matter only simulations to corresponding full hydrodynamical outputs. By sampling over the latent space of our VAE, we can generate posterior samples and study the variance of the mapping. We find that our reconstructed field provides an accurate representation of the target hydrodynamical fields as well as a reasonable variance estimates. This approach has promise for the rapid generation of mocks as well as for implementation in a full Bayesian inverse model of observed data.

show abstract

“…For example, NNs have been used to predict the nonlinear structure formation from the linear cosmological initial condition (He et al 2019;Berger & Stein 2019;Bernardini et al 2020;Alves de Oliveira et al 2020) and from the density field (Ramanah et al 2019). Models have also been trained to predict various baryonic properties from dark matter only simulation, such as galaxy distribution (Modi et al 2018;Zhang et al 2019), thermal Sunyaev-Zeldovich (tSZ) effect (Tröster et al 2019), 21 cm emission from neutral hydrogen (Wadekar et al 2020), stellar maps and various gas properties (Dai & Seljak 2021), etc. Recently, Villaescusa-Navarro et al (2020a have started the Cosmology and Astrophysics with MachinE Learning Simulations (CAMELS), a set of over 4000 hydrodynamical simulations run with different hydrodynamic solvers and subgrid models for galaxy formation, providing a large training set to study baryonic effects with machine learning applications.…”

Section: Introductionmentioning

confidence: 99%

AI-assisted super-resolution cosmological simulations II: Halo substructures, velocities and higher order statistics

Ni,

Li,

Lachance

et al. 2021

Preprint

View full text Add to dashboard Cite

In this work, we expand and test the capabilities of our recently developed super-resolution (SR) model to generate high-resolution (HR) realizations of the full phase-space matter distribution, including both displacement and velocity, from computationally cheap low-resolution (LR) cosmological N-body simulations. The SR model enhances the simulation resolution by generating 512 times more tracer particles, extending into the deeply non-linear regime where complex structure formation processes take place. We validate the SR model by deploying the model in 10 test simulations of box size 100 ℎ −1 Mpc, and examine the matter power spectra, bispectra and 2D power spectra in redshift space. We find the generated SR field matches the true HR result at percent level down to scales of 𝑘 ∼ 10 ℎ Mpc −1 . We also identify and inspect dark matter halos and their substructures. Our SR model generate visually authentic small-scale structures, that cannot be resolved by the LR input, and are in good statistical agreement with the real HR results. The SR model performs satisfactorily on the halo occupation distribution, halo correlations in both real and redshift space, and the pairwise velocity distribution, matching the HR results with comparable scatter, thus demonstrating its potential in making mock halo catalogs. The SR technique can be a powerful and promising tool for modelling small-scale galaxy formation physics in large cosmological volumes.

show abstract

HInet: Generating neutral hydrogen from dark matter with neural networks

Cited by 9 publications

References 65 publications

The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations

The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations

HyPhy: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics

AI-assisted super-resolution cosmological simulations II: Halo substructures, velocities and higher order statistics

Contact Info

Product

Resources

About