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.

14 pages, 6 figures, Submitted to APJ

References in corpus (20)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Simulating galaxy formation with black hole driven thermal and kinetic feedback
• The DESI Experiment Part I: Science,Targeting, and Survey Design
• Simba: Cosmological Simulations with Black Hole Growth and Feedback
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• EdgeConnect: Generative Image Inpainting with Adversarial Edge Learning
• eROSITA Science Book: Mapping the Structure of the Energetic Universe
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Large-scale Structure Catalogs for Cosmological Analysis
• 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
• Towards an Optimal Estimation of Cosmological Parameters with the Wavelet Scattering Transform
• Constraining $M_{\nu }$ with the Bispectrum II: The Total Information Content of the Galaxy Bispectrum
• Detecting Neutrino Mass by Combining Matter Clustering, Halos, and Voids
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• Joint analysis of anisotropic power spectrum, bispectrum and trispectrum: application to N-body simulations
• 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
• Cosmological model discrimination with Deep Learning
• 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