Deep learning dark matter map reconstructions from DES SV weak lensing data

Abstract

We present the first reconstruction of dark matter maps from weak lensing observational data using deep learning. We train a convolution neural network (CNN) with a Unet based architecture on over $3.6\times 10^5$ simulated data realizations with non-Gaussian shape noise and with cosmological parameters varying over a broad prior distribution. We interpret our newly created DES SV map as an approximation of the posterior mean $P(\kappa |\gamma )$ of the convergence given observed shear. Our DeepMass method is substantially more accurate than existing mass-mapping methods. With a validation set of 8000 simulated DES SV data realizations, compared to Wiener filtering with a fixed power spectrum, the DeepMass method improved the mean-square-error (MSE) by 11 per cent. With N-body simulated MICE mock data, we show that Wiener filtering with the optimal known power spectrum still gives a worse MSE than our generalized method with no input cosmological parameters; we show that the improvement is driven by the non-linear structures in the convergence. With higher galaxy density in future weak lensing data unveiling more non-linear scales, it is likely that deep learning will be a leading approach for mass mapping with Euclid and LSST.

arXiv commentAccepted MNRAS, 7 pages, 5 figures, added interpretation of DeepMass improvement

References in corpus (23)

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• The cosmological simulation code GADGET-2
• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• Weak Gravitational Lensing
• Dark Energy Survey Year 1 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• The Dark Energy Camera
• Power Spectra for Cold Dark Matter and its Variants
• Understanding Deep Convolutional Networks
• Galaxies in the Hubble Ultra Deep Field: I. Detection, Multiband Photometry, Photometric Redshifts, and Morphology
• Solving Large Scale Structure in Ten Easy Steps with COLA
• Wiener Reconstruction of The Large Scale Structure
• The MICE Grand Challenge Lightcone Simulation III: Galaxy lensing mocks from all-sky lensing maps
• Galaxy alignments: Observations and impact on cosmology
• Weak Gravitational Flexion
• High Resolution Weak Lensing Mass-Mapping Combining Shear and Flexion
• Improving Weak Lensing Mass Map Reconstructions using Gaussian and Sparsity Priors: Application to DES SV
• Wide-Field Lensing Mass Maps from DES Science Verification Data
• Denoising Weak Lensing Mass Maps with Deep Learning
• Bayesian weak lensing tomography: Reconstructing the 3D large-scale distribution of matter with a lognormal prior
• Dark Energy Survey Year 1 Results: Wide field mass maps via forward fitting in harmonic space
• Fast Sampling from Wiener Posteriors for Image Data with Dataflow Engines
• Dark energy constraints and correlations with systematics from CFHTLS weak lensing, SNLS supernovae Ia and WMAP5

Cited by in corpus (38)

• Likelihood-free inference with neural compression of DES SV weak lensing map statistics
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Dark Energy Survey Year 3 results: curved-sky weak lensing mass map reconstruction
• Dark Energy Survey Year 3 results: cosmology with moments of weak lensing mass maps -- validation on simulations
• Weak lensing scattering transform: dark energy and neutrino mass sensitivity
• Noise reduction for weak lensing mass mapping: An application of generative adversarial networks to Subaru Hyper Suprime-Cam first-year data
• Probabilistic Mass Mapping with Neural Score Estimation
• Deep learning approach for identification of HII regions during reionization in 21-cm observations
• Machine Learning for Observational Cosmology
• A roadmap to cosmological parameter analysis with third-order shear statistics I: Modelling and validation
• Weak lensing mass reconstruction using sparsity and a Gaussian random field
• Large-scale density and velocity field reconstructions with neural networks
• KaRMMa -- Kappa Reconstruction for Mass Mapping
• Revealing the Local Cosmic Web from Galaxies by Deep Learning
• Cosmic Velocity Field Reconstruction Using AI
• Single frequency CMB B-mode inference with realistic foregrounds from a single training image
• Avoiding baryonic feedback effects on neutrino mass measurements from CMB lensing
• Higher order statistics of shear field: a machine learning approach
• YOLO-CL: Galaxy cluster detection in the SDSS with deep machine learning
• Machine Learning and Cosmology
• Painting baryons onto N-body simulations of galaxy clusters with image-to-image deep learning
• Forklens: Accurate weak-lensing shear measurement with deep learning
• Deep learning approach for identification of HII regions during reionization in 21-cm observations -- III. image recovery
• Accurate Kappa Reconstruction Algorithm for masked shear catalog (AKRA)
• YOLO-CL cluster detection in the Rubin/LSST DC2 simulation
• AKRA 2.0: Accurate Kappa Reconstruction Algorithm for masked shear catalog
• Reconstructing the Gravitational Lensing Potential from the Lyman-$\alpha$ Forest
• Generative modelling for mass-mapping with fast uncertainty quantification
• Neural style transfer of weak lensing mass maps
• kSZ Pairwise Velocity Reconstruction with Machine Learning
• Galaxy Image Restoration with Shape Constraint
• Distribution-free uncertainty quantification for inverse problems: application to weak lensing mass mapping
• Impact of weak-lensing mass-mapping algorithms on cosmology inference
• Denoising weak lensing mass maps with diffusion model: systematic comparison with generative adversarial network
• Probabilistic Mapping of Dark Matter by Neural Score Matching
• Cosmological parameter estimation from large-scale structure deep learning
• A plug-and-play approach with fast uncertainty quantification for weak lensing mass mapping
• Advancing weak lensing mass mapping with a mask-aware HEALPix transformer