MillimeterDL: Deep Learning Simulations of the Microwave Sky

Abstract

We present 500 high-resolution, full-sky millimeter-wave Deep Learning (DL) simulations that include lensed CMB maps and correlated foreground components. We find that these MillimeterDL simulations can reproduce a wide range of non-Gaussian summary statistics matching the input training simulations, while only being optimized to match the power spectra. The procedure we develop in this work enables the capability to mass produce independent full-sky realizations from a single expensive full-sky simulation, when ordinarily the latter would not provide enough training data. We also circumvent a common limitation of high-resolution DL simulations that they be confined to small sky areas, often due to memory or GPU issues; we do this by developing a "stitching" procedure that can faithfully recover the high-order statistics of a full-sky map without discontinuities or repeated features. In addition, since our network takes as input a full-sky lensing convergence map, it can in principle take a full-sky lensing convergence map from any large-scale structure (LSS) simulation and generate the corresponding lensed CMB and correlated foreground components at millimeter wavelengths; this is especially useful in the current era of combining results from both CMB and LSS surveys, which require a common set of simulations.

Sims are public on NERSC at /global/cfs/cdirs/cmb/data/generic/mmDL; see also https://lambda.gsfc.nasa.gov/simulation/tb_sim_ov.cfm on LAMBDA. Added bispectra and covariance matrix plots

References in corpus (18)

• Five-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Likelihoods and Parameters from the WMAP data
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• From the warm magnetized atomic medium to molecular clouds
• A measurement of secondary cosmic microwave background anisotropies from the 2500-square-degree SPT-SZ survey
• Optical Design of the Atacama Cosmology Telescope and the Millimeter Bolometric Array Camera
• Full-sky Gravitational Lensing Simulation for Large-area Galaxy Surveys and Cosmic Microwave Background Experiments
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• Simulating polarized Galactic synchrotron emission at all frequencies, the Hammurabi code
• Nuisance hardened data compression for fast likelihood-free inference
• The Sunyaev-Zel'dovich effects from a cosmological hydrodynamical simulation: large-scale properties and correlation with the soft X-ray signal
• Full-sky maps for gravitational lensing of the CMB
• CMB Lensing Beyond the Power Spectrum: Cosmological Constraints from the One-Point PDF and Peak Counts
• Deep learning for Sunyaev-Zel'dovich detection in Planck
• DeepSZ: Identification of Sunyaev-Zel'dovich Galaxy Clusters using Deep Learning
• An Active Galactic Nucleus Recognition Model based on Deep Neural Network
• A Generative Model of Galactic Dust Emission Using Variational Inference
• Inpainting CMB maps using Partial Convolutional Neural Networks
• Fast Generation of Large-scale Structure Density Maps via Generative Adversarial Networks

Cited by in corpus (12)

• The CAMELS project: public data release
• Constraining Cluster Virialization Mechanism and Cosmology using Thermal-SZ-selected clusters from Future CMB Surveys
• Machine Learning for Observational Cosmology
• Baryon Pasting Algorithm: Halo-based and Particle-based Pasting Methods
• Simulated catalogs and maps of radio galaxies at millimeter wavelengths in Websky
• Mitigating Foreground Bias to the CMB Lensing Power Spectrum for a CMB-HD Survey
• Reconstructing Cosmic Polarization Rotation with ResUNet-CMB
• The HalfDome Multi-Survey Cosmological Simulations: N-body Simulations
• Deep learning approach for identification of HII regions during reionization in 21-cm observations -- III. image recovery
• Neural style transfer of weak lensing mass maps
• Fast Baryonic Field Painting for Sunyaev-Zel'dovich Analyses: Transfer Function vs. Hybrid Effective Field Theory
• Supervised Extraction of the Thermal Sunyaev$-$Zel'dovich Effect with a Three-Dimensional Convolutional Neural Network