Inpainting hydrodynamical maps with deep learning

Abstract

From 1,000 hydrodynamic simulations of the CAMELS project, each with a different value of the cosmological and astrophysical parameters, we generate 15,000 gas temperature maps. We use a state-of-the-art deep convolutional neural network to recover missing data from those maps. We mimic the missing data by applying regular and irregular binary masks that cover either $15\%$ or $30\%$ of the area of each map. We quantify the reliability of our results using two summary statistics: 1) the distance between the probability density functions (pdf), estimated using the Kolmogorov-Smirnov (KS) test, and 2) the 2D power spectrum. We find an excellent agreement between the model prediction and the unmasked maps when using the power spectrum: better than $1\%$ for $k<20h/$Mpc for any irregular mask. For regular masks, we observe a systematic offset of $\sim 5\%$ when covering $15\%$ of the maps while the results become unreliable when $30\%$ of the data is missing. The observed KS-test p-values favor the null hypothesis that the reconstructed and the ground-truth maps are drawn from the same underlying distribution when irregular masks are used. For regular-shaped masks on the other hand, we find a strong evidence that the two distributions do not match each other. Finally, we use the model, trained on gas temperature maps, to perform inpainting on maps from completely different fields such as gas mass, gas pressure, and electron density and also for gas temperature maps from simulations run with other codes. We find that visually, our model is able to reconstruct the missing pixels from the maps of those fields with great accuracy, although its performance using summary statistics depends strongly on the considered field.

arXiv comment14 pages, 6 figures, Submitted to APJ

References in corpus (56)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• The Baryon Oscillation Spectroscopic Survey of SDSS-III
• Simulating Galaxy Formation with the IllustrisTNG Model
• The Simons Observatory: Science goals and forecasts
• Simulating galaxy formation with black hole driven thermal and kinetic feedback
• The DESI Experiment Part I: Science,Targeting, and Survey Design
• GIZMO: A New Class of Accurate, Mesh-Free Hydrodynamic Simulation Methods
• Simba: Cosmological Simulations with Black Hole Growth and Feedback
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• Euclid Definition Study Report
• EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
• The Arepo public code release
• eROSITA Science Book: Mapping the Structure of the Energetic Universe
• The Quijote simulations
• The VIMOS Public Extragalactic Redshift Survey (VIPERS). Galaxy clustering and redshift-space distortions at z=0.8 in the first data release
• Cosmology with Phase 1 of the Square Kilometre Array; Red Book 2018: Technical specifications and performance forecasts
• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• The clustering of galaxies in the SDSS-III Baryon Oscillation Spectroscopic Survey: Analysis of potential systematics
• Dark Energy Survey Year 3 Results: Photometric Data Set for Cosmology
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale Structure Catalogs for Cosmological Analysis
• Cosmological constraints with deep learning from KiDS-450 weak lensing maps
• Prime Focus Spectrograph (PFS) for the Subaru Telescope: Overview, recent progress, and future perspectives
• Image Inpainting by End-to-End Cascaded Refinement with Mask Awareness
• Constraining $M_{\nu }$ with the Bispectrum I: Breaking Parameter Degeneracies
• Non-Gaussian information from weak lensing data via deep learning
• Likelihood-free inference with neural compression of DES SV weak lensing map statistics
• Towards an Optimal Estimation of Cosmological Parameters with the Wavelet Scattering Transform
• Fisher for complements: Extracting cosmology and neutrino mass from the counts-in-cells PDF
• Constraining $M_{\nu }$ with the Bispectrum II: The Total Information Content of the Galaxy Bispectrum
• Nearest Neighbor distributions: new statistical measures for cosmological clustering
• Using the Marked Power Spectrum to Detect the Signature of Neutrinos in Large-Scale Structure
• Weak lensing cosmology with convolutional neural networks on noisy data
• New Interpretable Statistics for Large Scale Structure Analysis and Generation
• Patch-Based Image Inpainting with Generative Adversarial Networks
• Detecting Neutrino Mass by Combining Matter Clustering, Halos, and Voids
• Constraining the astrophysics and cosmology from 21cm tomography using deep learning with the SKA
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• Primordial non-Gaussianity without tails -- how to measure fNL with the bulk of the density PDF
• Joint analysis of anisotropic power spectrum, bispectrum and trispectrum: application to N-body simulations
• Matter trispectrum: theoretical modelling and comparison to N-body simulations
• Cosmological cross-correlations and nearest neighbor distributions
• Interpreting deep learning models for weak lensing
• A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Pairwise-Inverse-Probability and Angular Correction for Fibre Collisions in Clustering Measurements
• Information Content of Higher-Order Galaxy Correlation Functions
• Exploring KSZ velocity reconstruction with $N$-body simulations and the halo model
• What Can We Learn by Combining the Skew Spectrum and the Power Spectrum?
• High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis
• Weighing neutrinos with the halo environment
• Cosmological model discrimination with Deep Learning
• Inpainting Galactic Foreground Intensity and Polarization maps using Convolutional Neural Network
• Image Inpainting for Irregular Holes Using Partial Convolutions
• An Inpainting Approach to Tackle the Kinematic and Thermal SZ Induced Biases in CMB-Cluster Lensing Estimators
• The unequal-time matter power spectrum: impact on weak lensing observables
• Confronting missing observations with probability weights: Fourier space and generalised formalism
• Inpainting CMB maps using Partial Convolutional Neural Networks

Cited by in corpus (4)

• The CAMELS project: public data release
• Precise Cosmological Constraints from BOSS Galaxy Clustering with a Simulation-Based Emulator of the Wavelet Scattering Transform
• How the Galaxy-Halo Connection Depends on Large-Scale Environment
• Probing the cosmic web in Ly$\alpha$ emission over large scales: an Intensity Mapping forecast for DECaLS/BASS and DESI