A comparative study of cosmological constraints from weak lensing using Convolutional Neural Networks

Abstract

Weak Lensing (WL) surveys are reaching unprecedented depths, enabling the investigation of very small angular scales. At these scales, nonlinear gravitational effects lead to higher-order correlations making the matter distribution highly non-Gaussian. Extracting this information using traditional statistics has proven difficult, and Machine Learning based summary statistics have emerged as a powerful alternative. We explore the capabilities of a discriminative, Convolutional Neural Networks (CNN) based approach, focusing on parameter constraints in the (${\Omega }_m$, ${\sigma }_8$) cosmological parameter space. Leveraging novel training loss functions and network representations on WL mock datasets without baryons, we show that our models achieve $\sim 5$ times stronger constraints than the power spectrum, $\sim 3$ stronger constraints than peak counts, and $\sim 2$ stronger constraints than previous CNN-learned summary statistics and scattering transforms, for noise levels relevant to Rubin or Euclid. For WL convergence maps with baryonic physics, our models achieve $\sim 2.3$ times stronger constraining power than the power spectrum at these noise levels, also outperforming previous summary statistics. To further explore the possibilities of CNNs for this task, we also discuss transfer learning where we adapt pre-trained models, trained on different tasks or datasets, for cosmological inference, finding that these do not improve the performance.

16 pages, 2 figures. Comments Welcome

References in corpus (50)

• The cosmological simulation code GADGET-2
• Weak Gravitational Lensing
• Cosmology from cosmic shear power spectra with Subaru Hyper Suprime-Cam first-year data
• Cosmology with cosmic shear observations: a review
• Cosmological constraints from cosmic shear two-point correlation functions with HSC survey first-year data
• Cosmology and the Bispectrum
• CFHTLenS revisited: assessing concordance with Planck including astrophysical systematics
• KiDS-450: The tomographic weak lensing power spectrum and constraints on cosmological parameters
• Cosmology with the shear-peak statistics
• CFHTLenS: Cosmological constraints from a combination of cosmic shear two-point and three-point correlations
• Improving lognormal models for cosmological fields
• SpinalNet: Deep Neural Network with Gradual Input
• Three-Point Correlations in Weak Lensing Surveys: Model Predictions and Applications
• Cosmology constraints from shear peak statistics in Dark Energy Survey Science Verification data
• Cosmological constraints with deep learning from KiDS-450 weak lensing maps
• A new approach to observational cosmology using the scattering transform
• Probing Cosmology with Weak Lensing Peak Counts
• Non-Gaussian information from weak lensing data via deep learning
• Likelihood-free inference with neural compression of DES SV weak lensing map statistics
• Probing Cosmology with Weak Lensing Minkowski Functionals
• Towards an Optimal Estimation of Cosmological Parameters with the Wavelet Scattering Transform
• Automatic physical inference with information maximising neural networks
• Cosmological Information in Weak Lensing Peaks
• Weak lensing cosmology with convolutional neural networks on noisy data
• Weak lensing from space: first cosmological constraints from three-point shear statistics
• Modelling the large scale structure of the Universe as a function of cosmology and baryonic physics
• New Interpretable Statistics for Large Scale Structure Analysis and Generation
• Lognormal Property of Weak Lensing Fields
• Cosmological constraints from noisy convergence maps through deep learning
• Testing the lognormality of the galaxy and weak lensing convergence distributions from Dark Energy Survey maps
• The three-point correlation function of cosmic shear. II: Relation to the bispectrum of the projected mass density and generalized third-order aperture measures
• The cosmology dependence of weak lensing cluster counts
• Weak lensing scattering transform: dark energy and neutrino mass sensitivity
• The Kurtosis of the Cosmic Shear Field
• Going Beyond the Galaxy Power Spectrum: an Analysis of BOSS Data with Wavelet Scattering Transforms
• Mocking the Weak Lensing universe: the LensTools python computing package
• Translation and Rotation Equivariant Normalizing Flow (TRENF) for Optimal Cosmological Analysis
• Interpreting deep learning models for weak lensing
• Going deep with Minkowski functionals of convergence maps
• Higher-order moments of the lensing shear and other spin two fields
• Implication of Omega_m through the Morphological Analysis of Weak Lensing Fields
• Minkowski Functionals of Convergence Maps and the Lensing Figure of Merit
• Cosmological constraints from HSC survey first-year data using deep learning
• Increasing the Lensing Figure of Merit through Higher Order Convergence Moments
• The Impact of Baryons on Cosmological Inference from Weak Lensing Statistics
• Neural networks as optimal estimators to marginalize over baryonic effects
• Towards a full $w$CDM map-based analysis for weak lensing surveys
• Higher order moments of lensing convergence - I. Estimate from simulations
• Cosmological model differentiation through weak gravitational lensing
• LightCurve MoE: A Dynamic Sparse Routing Mixture-of-Experts Architecture for Efficient Stellar Light Curve Classification

Cited by in corpus (12)

• Towards unveiling the large-scale nature of gravity with the wavelet scattering transform
• Diffusion-HMC: Parameter Inference with Diffusion-model-driven Hamiltonian Monte Carlo
• Full-waveform earthquake source inversion using simulation-based inference
• Massive $\nu$s through the CNN lens: interpreting the field-level neutrino mass information in weak lensing
• Hybrid summary statistics: neural weak lensing inference beyond the power spectrum
• Field-Level Comparison and Robustness Analysis of Cosmological N-body Simulations
• Detecting Modeling Bias with Continuous Time Flow Models on Weak Lensing Maps
• Interpreting Cosmological Information from Neural Networks in the Hydrodynamic Universe
• Towards an optimal extraction of cosmological parameters from galaxy cluster surveys using convolutional neural networks
• Denoising weak lensing mass maps with diffusion model: systematic comparison with generative adversarial network
• Cosmological Analysis with Calibrated Neural Quantile Estimation and Approximate Simulators
• Generative modeling of convergence maps based on predicted one-point statistics