Non-Gaussian information from weak lensing data via deep learning

  title={Non-Gaussian information from weak lensing data via deep learning},
  author={Arushi Gupta and Jos{\'e} Manuel Zorrilla Matilla and Daniel J. Hsu and Zolt{\'a}n Haiman},
Weak lensing maps contain information beyond two-point statistics on small scales. Much recent work has tried to extract this information through a range of different observables or via nonlinear transformations of the lensing field. Here we train and apply a two-dimensional convolutional neural network to simulated noiseless lensing maps covering 96 different cosmological models over a range of ${{\mathrm{\ensuremath{\Omega}}}_{m},{\ensuremath{\sigma}}_{8}}$. Using the area of the confidence… 

Cosmological constraints from noisy convergence maps through deep learning

A new training strategy to train the neural network with noisy data is presented, as well as considerations for practical applications of the deep learning approach.

Interpreting deep learning models for weak lensing

This work analyze a DNN that has been found in previous work to accurately recover cosmological parameters in simulated maps of the WL convergence and applies a series of well established saliency methods to interpret the DNN, finding that the most relevant pixels are those with extreme $\kappa$ values.

Denoising Weak Lensing Mass Maps with Deep Learning

An image-to-image translation method with conditional adversarial networks (CANs), which learn efficient mapping from an input noisy weak lensing map to the underlying noise field, and improves the cosmological constraints by using observational data from ongoing and upcoming galaxy imaging surveys.

Learning from deep learning: better cosmological parameter inference from weak lensing maps

An improved CNN is presented that gives significantly better estimates of {\Omegam and {\sigma}8 cosmological parameters from simulated convergence maps than the state of art methods and also is free of systematic bias.

Decoding Cosmological Information in Weak-Lensing Mass Maps with Generative Adversarial Networks

Galaxy imaging surveys enable us to map the cosmic matter density field through weak gravitational lensing analysis. The density reconstruction is compromised by a variety of noise originating from

Cosmological parameter estimation from large-scale structure deep learning

CNN can be more promising than people expected in deriving tight cosmological constraints from the cosmic large scale structure and exhibits robustness against smoothing, masking, random noise, global variation, rotation, reflection, and simulation resolution.

Cosmological constraints with deep learning from KiDS-450 weak lensing maps

Convolutional neural networks (CNNs) have recently been demonstrated on synthetic data to improve upon the precision of cosmological inference. In particular, they have the potential to yield more

Weak lensing cosmology with convolutional neural networks on noisy data

Weak gravitational lensing is one of the most promising cosmological probes of the late universe. Several large ongoing (DES, KiDS, HSC) and planned (LSST, Euclid, WFIRST) astronomical surveys

Noise reduction for weak lensing mass mapping: an application of generative adversarial networks to Subaru Hyper Suprime-Cam first-year data

We propose a deep-learning approach based on generative adversarial networks (GANs) to reduce noise in weak lensing mass maps under realistic conditions. We apply image-to-image translation using

A new way to constrain the densities of intragroup medium in groups of galaxies with convolutional neural networks

Ram pressure (RP) can influence the evolution of cold gas content and star formation rates of galaxies. One of the key parameters for the strength of RP is the density of intra-group medium



Origin of weak lensing convergence peaks

Weak lensing convergence peaks are a promising tool to probe nonlinear structure evolution at late times, providing additional cosmological information beyond second-order statistics. Previous

Emulating the CFHTLenS weak lensing data: Cosmological constraints from moments and Minkowski functionals

Weak gravitational lensing is a powerful cosmological probe, with non-Gaussian features potentially containing the majority of the information. We examine constraints on the parameter triplet

Cosmology constraints from the weak lensing peak counts and the power spectrum in CFHTLenS data

Lensing peaks have been proposed as a useful statistic, containing cosmological information from non-Gaussianities that is inaccessible from traditional two-point statistics such as the power

Do dark matter halos explain lensing peaks

We have investigated a recently proposed halo-based model, Camelus, for predicting weak-lensing peak counts, and compared its results over a collection of 162 cosmologies with those from N-body

Probing cosmology with weak lensing peak counts

We propose counting peaks in weak lensing (WL) maps, as a function of their height, to probe models of dark energy and to constrain cosmological parameters. Because peaks can be identified in

Constraining dark energy by combining cluster counts and shear-shear correlations in a weak lensing survey

We study the potential of a large future weak lensing survey to constrain dark-energy properties by using both the number counts of detected galaxy clusters (sensitive primarily to density

The effective number density of galaxies for weak lensing measurements in the LSST project

Future weak lensing surveys potentially hold the highest statistical power for constraining cosmological parameters compared to other cosmological probes. The statistical power of a weak lensing

Probing Cosmology with Weak Lensing Minkowski Functionals

In this paper, we show that Minkowski Functionals (MFs) of weak gravitational lensing (WL) convergence maps contain significant non-Gaussian, cosmology-dependent information. To do this, we use a

Statistical connection of peak counts to power spectrum and moments in weak-lensing field

The number density of local maxima of weak lensing field, referred to as weak-lensing peak counts, can be used as a cosmological probe. However, its relevant cosmological information is still

Validity of the Born approximation for beyond Gaussian weak lensing observables

Accurate forward modeling of weak lensing (WL) observables from cosmological parameters is necessary for upcoming galaxy surveys. Because WL probes structures in the non-linear regime, analytical