Effective cosmic density field reconstruction with convolutional neural network

Abstract

We present a cosmic density field reconstruction method that augments the traditional reconstruction algorithms with a convolutional neural network (CNN). Following Shallue $\&$ Eisenstein (2022), the key component of our method is to use the $\text{reconstructed}$ density field as the input to the neural network. We extend this previous work by exploring how the performance of these reconstruction ideas depends on the input reconstruction algorithm, the reconstruction parameters, and the shot noise of the density field, as well as the robustness of the method. We build an eight-layer CNN and train the network with reconstructed density fields computed from the Quijote suite of simulations. The reconstructed density fields are generated by both the standard algorithm and a new iterative algorithm. In real space at $z=0$, we find that the reconstructed field is $90\%$ correlated with the true initial density out to $k\sim 0.5h{\mathrm{Mpc}}^{-1}$, a significant improvement over $k\sim 0.2h{\mathrm{Mpc}}^{-1}$ achieved by the input reconstruction algorithms. We find similar improvements in redshift space, including an improved removal of redshift space distortions at small scales. We also find that the method is robust across changes in cosmology. Additionally, the CNN removes much of the variance from the choice of different reconstruction algorithms and reconstruction parameters. However, the effectiveness decreases with increasing shot noise, suggesting that such an approach is best suited to high density samples. This work highlights the additional information in the density field beyond linear scales as well as the power of complementing traditional analysis approaches with machine learning techniques.

arXiv comment10 pages, 8 figures. Matches version accepted for publication in MNRAS

References in corpus (42)

• A 2% Distance to z=0.35 by Reconstructing Baryon Acoustic Oscillations - I : Methods and Application to the Sloan Digital Sky Survey
• Improving Cosmological Distance Measurements by Reconstruction of the Baryon Acoustic Peak
• Nonlinear Evolution of Baryon Acoustic Oscillations
• On the Robustness of the Acoustic Scale in the Low-Redshift Clustering of Matter
• The Quijote simulations
• Bayesian physical reconstruction of initial conditions from large scale structure surveys
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: measurement of the BAO and growth rate of structure of the luminous red galaxy sample from the anisotropic power spectrum between redshifts 0.6 and 1.0
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Observational systematics and baryon acoustic oscillations in the correlation function
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Baryon Acoustic Oscillations in Fourier-space
• Baryonic Signatures in Large-Scale Structure
• The clustering of galaxies in the SDSS-III Baryon Oscillation Spectroscopic Survey: Measuring D_A and H at z=0.57 from the Baryon Acoustic Peak in the Data Release 9 Spectroscopic Galaxy Sample
• Measuring D_A and H at z=0.35 from the SDSS DR7 LRGs using baryon acoustic oscillations
• ELUCID - Exploring the Local Universe with reConstructed Initial Density field I: Hamiltonian Markov Chain Monte Carlo Method with Particle Mesh Dynamics
• High-precision predictions for the acoustic scale in the non-linear regime
• Non-Gaussian information from weak lensing data via deep learning
• Calibrating the Baryon Oscillation Ruler for Matter and Halos
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• The Initial Conditions of the Universe from Constrained Simulations
• Towards optimal extraction of cosmological information from nonlinear data
• Iterative initial condition reconstruction
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Weak lensing cosmology with convolutional neural networks on noisy data
• Reconstruction within the Zeldovich approximation
• Eulerian BAO Reconstructions and N-Point Statistics
• A Deep Learning Approach to Galaxy Cluster X-ray Masses
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: theoretical systematics and Baryon Acoustic Oscillations in the galaxy correlation function
• Modeling the reconstructed BAO in Fourier space
• Cosmological constraints from noisy convergence maps through deep learning
• Cosmological Reconstruction From Galaxy Light: Neural Network Based Light-Matter Connection
• A Robust and Efficient Deep Learning Method for Dynamical Mass Measurements of Galaxy Clusters
• Towards an Optimal Reconstruction of Baryon Oscillations
• An iterative reconstruction of cosmological initial density fields
• The foreground wedge and 21 cm BAO surveys
• A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys
• The Reconstructed Power Spectrum in the Zeldovich Approximation
• A gradient based method for modeling baryons and matter in halos of fast simulations
• A fast semi-discrete optimal transport algorithm for a unique reconstruction of the early Universe
• Baryon acoustic oscillations reconstruction using convolutional neural networks
• Reconstructing Cosmological Initial Conditions from Late-Time Structure with Convolutional Neural Networks
• Baryon Acoustic Oscillations reconstruction with pixels
• Optimal Transport Reconstruction of Baryon Acoustic Oscillations
• Accurate Baryon Acoustic Oscillations reconstruction via semi-discrete optimal transport

Cited by in corpus (12)

• Improving constraints on primordial non-Gaussianity using neural network based reconstruction
• Neural network reconstruction of density and velocity fields from the 2MASS Redshift Survey
• DarkAI: Reconstructing the density, velocity and tidal field of dark matter from a DESI-like bright galaxy sample
• Learning the Universe: Learning to Optimize Cosmic Initial Conditions with Non-Differentiable Structure Formation Models
• YOLO-CL cluster detection in the Rubin/LSST DC2 simulation
• Baryon Acoustic Oscillations analyses with Density-Split Statistics
• Neural Network Reconstruction of Non-Gaussian Initial Conditions from Dark Matter Halos
• Lagrangian space remapping and the angular momentum reconstruction from cosmic structures
• Searching optimal scales for reconstructing cosmological initial conditions using convolutional neural networks
• From Redshift to Real Space: Combining Linear Theory With Neural Networks
• On the Connection between Field-Level Inference and $n$-point Correlation Functions
• Fast Sampling of Cosmological Initial Conditions with Gaussian Neural Posterior Estimation