A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys

Abstract

We present a deep machine learning (ML)-based technique for accurately determining ${\sigma }_8$ and ${\Omega }_m$ from mock 3D galaxy surveys. The mock surveys are built from the AbacusCosmos suite of $N$-body simulations, which comprises 40 cosmological volume simulations spanning a range of cosmological models, and we account for uncertainties in galaxy formation scenarios through the use of generalized halo occupation distributions (HODs). We explore a trio of ML models: a 3D convolutional neural network (CNN), a power-spectrum-based fully connected network, and a hybrid approach that merges the two to combine physically motivated summary statistics with flexible CNNs. We describe best practices for training a deep model on a suite of matched-phase simulations and we test our model on a completely independent sample that uses previously unseen initial conditions, cosmological parameters, and HOD parameters. Despite the fact that the mock observations are quite small ($\sim 0.07h^{-3}{\mathrm{Gpc}}^3$) and the training data span a large parameter space (6 cosmological and 6 HOD parameters), the CNN and hybrid CNN can constrain ${\sigma }_8$ and ${\Omega }_m$ to $\sim 3\%$ and $\sim 4\%$, respectively.

arXiv commentSubmitted to The Astrophysical Journal

References in corpus (57)

• Planck 2015 results. XIII. Cosmological parameters
• The Sloan Digital Sky Survey: Technical Summary
• Planck 2013 results. XVI. Cosmological parameters
• Detection of the Baryon Acoustic Peak in the Large-Scale Correlation Function of SDSS Luminous Red Galaxies
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample
• The 2dF Galaxy Redshift Survey: Power-spectrum analysis of the final dataset and cosmological implications
• The 3D power spectrum of galaxies from the SDSS
• Observational Probes of Cosmic Acceleration
• Dark Energy Survey Year 1 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• Chandra Cluster Cosmology Project III: Cosmological Parameter Constraints
• Halo occupation numbers and galaxy bias
• The Halo Occupation Distribution: Towards an Empirical Determination of the Relation Between Galaxies and Mass
• Cosmology and fundamental physics with the Euclid satellite
• Theoretical Models of the Halo Occupation Distribution: Separating Central and Satellite Galaxies
• The 2dF Galaxy Redshift Survey: The power spectrum and the matter content of the universe
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• Extragalactic Science, Cosmology and Galactic Archaeology with the Subaru Prime Focus Spectrograph (PFS)
• The age dependence of halo clustering
• How Many Galaxies Fit in a Halo? Constraints on Galaxy Formation Efficiency from Spatial Clustering
• Detection of weak gravitational lensing distortions of distant galaxies by cosmic dark matter at large scales
• Detection of Weak Gravitational Lensing by Large-scale Structure
• The Dependence of Halo Clustering on Halo Formation History, Concentration, and Occupation
• Cosmology from cosmic shear power spectra with Subaru Hyper Suprime-Cam first-year data
• The Las Campanas Redshift Survey
• Planck 2013 Results. XXIV. Constraints on primordial non-Gaussianity
• KiDS+VIKING-450: Cosmic shear tomography with optical+infrared data
• Weighing the Giants IV: Cosmology and Neutrino Mass
• A catalog of visual-like morphologies in the 5 CANDELS fields using deep-learning
• nbodykit: an open-source, massively parallel toolkit for large-scale structure
• Cosmological Constraints from Galaxy Clusters in the 2500 square-degree SPT-SZ Survey
• Cosmological Constraints from a Combination of Galaxy Clustering and Lensing -- I. Theoretical Framework
• The power spectrum and bispectrum of SDSS DR11 BOSS galaxies I: bias and gravity
• Learning to Predict the Cosmological Structure Formation
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications
• Cosmological Constraints from a Combination of Galaxy Clustering and Lensing -- III. Application to SDSS Data
• Introducing Decorated HODs: modeling assembly bias in the galaxy-halo connection
• Cosmological constraints with deep learning from KiDS-450 weak lensing maps
• Non-Gaussian information from weak lensing data via deep learning
• Breaking the Degeneracies Between Cosmology and Galaxy Bias
• The Abacus Cosmos: A Suite of Cosmological N-body Simulations
• Cosmic Emulation: Fast Predictions for the Galaxy Power Spectrum
• A High-Fidelity Realization of the Euclid Code Comparison $N$-body Simulation with Abacus
• Weak lensing cosmology with convolutional neural networks on noisy data
• A Deep Learning Approach to Galaxy Cluster X-ray Masses
• Cosmological Constraints from a Combination of Galaxy Clustering & Lensing -- II. Fisher Matrix Analysis
• A Robust and Efficient Deep Learning Method for Dynamical Mass Measurements of Galaxy Clusters
• Identifying Exoplanets with Deep Learning II: Two New Super-Earths Uncovered by a Neural Network in K2 Data
• Distinguishing standard and modified gravity cosmologies with machine learning
• Classifying the Large Scale Structure of the Universe with Deep Neural Networks
• A volumetric deep Convolutional Neural Network for simulation of mock dark matter halo catalogues
• Exploring the squeezed three-point galaxy correlation function with generalized halo occupation distribution models
• Machine Learning Applied to the Reionization History of the Universe in the 21 cm Signal
• Testing the Accuracy of Halo Occupation Distribution Modelling using Hydrodynamic Simulations
• Beyond two-point statistics: using the minimum spanning tree as a tool for cosmology
• How to Optimally Constrain Galaxy Assembly Bias: Supplement Projected Correlation Functions with Count-in-cells Statistics
• Cosmological information content in redshift-space power spectrum of SDSS-like galaxies in the quasi-nonlinear regime up to $k=0.3h$Mpc${}^{-1}$

Cited by in corpus (33)

• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• Large-scale dark matter simulations
• The $\text{Abacus}$ Cosmological $N$-body Code
• Deeply Uncertain: Comparing Methods of Uncertainty Quantification in Deep Learning Algorithms
• The CAMELS project: public data release
• Morphological classification of galaxies with deep learning: comparing 3-way and 4-way CNNs
• Learning cosmology and clustering with cosmic graphs
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• Robust Field-level Likelihood-free Inference with Galaxies
• Cosmology with one galaxy?
• Revisiting the cosmic distance duality relation with machine learning reconstruction methods: the combination of HII galaxies and ultra-compact radio quasars
• Machine Learning for Observational Cosmology
• ECoPANN: A Framework for Estimating Cosmological Parameters using Artificial Neural Networks
• Predicting dark matter halo formation in N-body simulations with deep regression networks
• Neural Network Reconstruction of $H^{\prime }(z)$ and its application in Teleparallel Gravity
• Recovering the CMB Signal with Machine Learning
• Effective cosmic density field reconstruction with convolutional neural network
• Likelihood-free Inference with Mixture Density Network
• Robust Neural Network-Enhanced Estimation of Local Primordial Non-Gaussianity
• Learning neutrino effects in Cosmology with Convolutional Neural Networks
• Deep learning insights into cosmological structure formation
• Fast generation of mock galaxy catalogues with COLA
• Cosmological constraints from low redshift 21 cm intensity mapping with machine learning
• Cosmology in the non-linear regime : the small scale miracle
• Inpainting hydrodynamical maps with deep learning
• Inferring Cosmological Parameters on SDSS via Domain-Generalized Neural Networks and Lightcone Simulations
• Cosmology from Point Clouds with Dark Matter Halos from the Quijote Simulations
• Cosmology with Topological Deep Learning
• Learning from Topology: Cosmological Parameter Estimation from the Large-scale Structure
• Predicting halo occupation and galaxy assembly bias with machine learning
• Towards an optimal extraction of cosmological parameters from galaxy cluster surveys using convolutional neural networks
• Constraining Galaxy-Halo Connection Using Machine Learning
• Cosmological constraints from the density gradient weighted correlation function