Massive $\nu$s through the CNN lens: interpreting the field-level neutrino mass information in weak lensing

Abstract

Modern cosmological surveys probe the Universe deep into the nonlinear regime, where massive neutrinos suppress cosmic structure. Traditional cosmological analyses, which use the 2-point correlation function to extract information, are no longer optimal in the nonlinear regime, and there is thus much interest in extracting beyond-2-point information to improve constraints on neutrino mass. Quantifying and interpreting the beyond-2-point information is thus a pressing task. We study the field-level information in weak lensing convergence maps using convolution neural networks. We find that the network performance increases as higher source redshifts and smaller scales are considered -- investigating up to a source redshift of 2.5 and ${\ell }_{\max }\simeq 10^4$ -- verifying that massive neutrinos leave a distinct effect on weak lensing. However, the performance of the network significantly drops after scaling out the 2-point information from the maps, implying that most of the field-level information can be found in the 2-point correlation function alone. We quantify these findings in terms of the likelihood ratio and also use Integrated Gradient saliency maps to interpret which parts of the map the network is learning the most from. We find that, in the absence of noise, the network extracts a similar amount of information from the most overdense and underdense regions. However, upon adding noise, the information in underdense regions is distorted as noise disproportionately washes out void-like structures.

10+5 pages, 5 figures, minor changes to reflect final version accepted by JCAP

References in corpus (41)

• Observation of electron-antineutrino disappearance at Daya Bay
• The DESI Experiment Part I: Science,Targeting, and Survey Design
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• Measurement of Neutrino Oscillation by the K2K Experiment
• Captum: A unified and generic model interpretability library for PyTorch
• Prime Focus Spectrograph (PFS) for the Subaru Telescope: Overview, recent progress, and future perspectives
• 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
• 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 GIGANTES dataset: precision cosmology from voids in the machine learning era
• Nuw CDM cosmology from the weak lensing convergence PDF
• Constraining neutrino mass with weak lensing Minkowski Functionals
• Weak lensing scattering transform: dark energy and neutrino mass sensitivity
• An $n$-th order Lagrangian Forward Model for Large-Scale Structure
• Translation and Rotation Equivariant Normalizing Flow (TRENF) for Optimal Cosmological Analysis
• FlowPM: Distributed TensorFlow Implementation of the FastPM Cosmological N-body Solver
• Bayesian evidence for the tensor-to-scalar ratio $r$ and neutrino masses $m_{\nu }$: Effects of uniform vs logarithmic priors
• Aemulus $\nu$: Precise Predictions for Matter and Biased Tracer Power Spectra in the Presence of Neutrinos
• Bayesian forward modelling of cosmic shear data
• The MillenniumTNG Project: The impact of baryons and massive neutrinos on high-resolution weak gravitational lensing convergence maps
• Beware of Fake $\nu$s: The Effect of Massive Neutrinos on the Nonlinear Evolution of Cosmic Structure
• Cosmological constraints from HSC survey first-year data using deep learning
• A Parameter-Masked Mock Data Challenge for Beyond-Two-Point Galaxy Clustering Statistics
• The look-elsewhere effect from a unified Bayesian and frequentist perspective
• Explaining deep learning of galaxy morphology with saliency mapping
• HIFlow: Generating Diverse HI Maps and Inferring Cosmology while Marginalizing over Astrophysics using Normalizing Flows
• Starlet l1-norm for weak lensing cosmology
• Joint velocity and density reconstruction of the Universe with nonlinear differentiable forward modeling
• Neural networks as optimal estimators to marginalize over baryonic effects
• A comparative study of cosmological constraints from weak lensing using Convolutional Neural Networks
• Explaining dark matter halo density profiles with neural networks
• pmwd: A Differentiable Cosmological Particle-Mesh $N$-body Library
• Field-level inference of cosmic shear with intrinsic alignments and baryons
• Evidence Networks: simple losses for fast, amortized, neural Bayesian model comparison
• Deep learning insights into cosmological structure formation
• Field-Level Inference with Microcanonical Langevin Monte Carlo
• Improving initialization and evolution accuracy of cosmological neutrino simulations
• The HalfDome Multi-Survey Cosmological Simulations: N-body Simulations
• Significance of void shape: Neutrino mass from Voronoi void halos?
• Optimal Neural Summarisation for Full-Field Weak Lensing Cosmological Implicit Inference

Cited by in corpus (3)

• The Power of the Cosmic Web
• Field-Level Comparison and Robustness Analysis of Cosmological N-body Simulations
• Interpreting Cosmological Information from Neural Networks in the Hydrodynamic Universe