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

Fluri, Janis; Kacprzak, T.; Lucchi, Aurélien; Réfrégier, Alexandre; Amara, Adam; Hofmann, Thomas; Schneider, Aurel

doi:10.1103/physrevd.100.063514

Cited by 151 publications

(121 citation statements)

References 81 publications

Supporting

Mentioning

118

Contrasting

Order By: Relevance

“…The data used in this work is the non-tomographic training and testing set introduced in (Fluri et al, 2019), without noise and intrinsic alignments. The simulation grid consists of 57 different cosmologies in the standard cosmological model: a flat Universe with cold dark matter (ΛCDM) (Lahav and Liddle, 2019).…”

Section: Sky Convergence Maps Datasetmentioning

confidence: 99%

“…This kernel is dependent on the relative distances between the observer and the lensed galaxies that are used to create the mass maps. The source galaxy redshift distribution n(z) used in this work is the non-tomographic distribution from (Fluri et al, 2019). The projected matter distribution is pixelized into images of size 128 px × 128 px, which corresponds to 5°× 5°of the sky.…”

Section: Sky Convergence Maps Datasetmentioning

confidence: 99%

“…This set of comparisons is the most exhaustive presentation of the capacity of generative models to learn the dark matter maps, to date. In this work we use the data generated by (Fluri et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Emulation of Cosmological Mass Maps with Conditional Generative Adversarial Networks

Perraudin

Marcon

Lucchi

et al. 2021

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

Weak gravitational lensing mass maps play a crucial role in understanding the evolution of structures in the Universe and our ability to constrain cosmological models. The prediction of these mass maps is based on expensive N-body simulations, which can create a computational bottleneck for cosmological analyses. Simulation-based emulators of map summary statistics, such as the matter power spectrum and its covariance, are starting to play increasingly important role, as the analytical predictions are expected to reach their precision limits for upcoming experiments. Creating an emulator of the cosmological mass maps themselves, rather than their summary statistics, is a more challenging task. Modern deep generative models, such as Generative Adversarial Networks (GAN), have demonstrated their potential to achieve this goal. Most existing GAN approaches produce simulations for a fixed value of the cosmological parameters, which limits their practical applicability. We propose a novel conditional GAN model that is able to generate mass maps for any pair of matter density Ωm and matter clustering strength σ8, parameters which have the largest impact on the evolution of structures in the Universe, for a given source galaxy redshift distribution n(z). Our results show that our conditional GAN can interpolate efficiently within the space of simulated cosmologies, and generate maps anywhere inside this space with good visual quality high statistical accuracy. We perform an extensive quantitative comparison of the N-body and GAN -generated maps using a range of metrics: the pixel histograms, peak counts, power spectra, bispectra, Minkowski functionals, correlation matrices of the power spectra, the Multi-Scale Structural Similarity Index (MS-SSIM) and our equivalent of the Fréchet Inception Distance. We find a very good agreement on these metrics, with typical differences are <5% at the center of the simulation grid, and slightly worse for cosmologies at the grid edges. The agreement for the bispectrum is slightly worse, on the <20% level. This contribution is a step toward building emulators of mass maps directly, capturing both the cosmological signal and its variability. We make the code1 and the data2 publicly available.

show abstract

Section: Sky Convergence Maps Datasetmentioning

confidence: 99%

Section: Sky Convergence Maps Datasetmentioning

confidence: 99%

See 1 more Smart Citation

Emulation of Cosmological Mass Maps with Conditional Generative Adversarial Networks

Perraudin

Marcon

Lucchi

et al. 2021

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

“…These characteristics are perfect to provide further results in cosmology. Some example of recent interesting applications of artificial neural networks in cosmology are the identification of galaxy mergers (Pearson et al 2019) and strongly gravitational lenses (Petrillo et al 2017) in astronomical images, a better estimation of cosmological constraints from weak lensing maps (Fluri et al 2019) and high fidelity generation of weak lensing convergence maps (Mustafa et al 2019), and cosmological structure formation simulations under different assumptions (Mathuriya et al 2018;He et al 2019;Perraudin et al 2019;Giusarma et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Point source detection with fully convolutional networks

Bonavera

Gómez

González‐Nuevo

et al. 2021

A&A

View full text Add to dashboard Cite

Context. Point sources are one of the main contaminants to the recovery of the cosmic microwave background signal at small scales, and their careful detection will be important for the next generation of cosmic microwave background experiments like LiteBird. Aims. We want to develop a method based on fully convolutional networks to detect sources in realistic simulations, and to compare its performance against one of the most used point source detection method in this context, the Mexican hat wavelet 2 (MHW2). The frequencies for our analysis are the 143, 217, and 353 GHz Planck channels. Methods. We produce realistic simulations of point sources at each frequency taking into account potential contaminating signals as the cosmic microwave background, the cosmic infrared background, the Galactic thermal emission, the thermal Sunyaev-Zel'dovich effect, and the instrumental and point source shot noises. We first produce a set of training simulations at 217 GHz to train the neural network that we named PoSeIDoN. Then we apply both PoSeIDoN and the MHW2 to recover the point sources in the validating simulations at all the frequencies, comparing the results by estimating the reliability, completeness, and flux density estimation accuracy. Moreover, the receiver operating characteristic (ROC) curves are computed in order to asses the methods'performance. Results. In the extra-galactic region with a 30 • galactic cut, the neural network successfully recovers point sources at 90% completeness corresponding to 253, 126, and 250 mJy for 143, 217, and 353 GHz respectively. In the same validation simulations the wavelet with a 3σ flux density detection limit recovers point sources up to 181, 102, and 153 mJy at 90% completeness. To reduce the number of spurious sources, we also apply a safer 4σ flux density detection limit, the same as in the Planck catalogues, increasing the 90% completeness levels: 235, 137, and 192 mJy. In all cases PoSeIDoN produces a much lower number of spurious sources with respect to MHW2. As expected, the results on spurious sources for both techniques worsen when reducing the galactic cut to 10 •. Conclusions. Our results suggest that using neural networks is a very promising approach for detecting point sources using data from cosmic microwave background experiments, providing overall better results in dealing with spurious sources with respect to the more usual filtering approaches. Moreover, PoSeIDoN gives competitive results even at the 217 GHz nearby channels where the network was not trained.

show abstract

“…Because their growth has spanned the entire age of the Universe, and has depended upon the density of building material and its collapse under gravity, versus its disruption by supernovae, active galactic nuclei, and dark energy, measurements of the precise number and properties of clusters are a highly sensitive test of the standard cosmological model (e.g. Bahcall & Cen 1993;Bahcall & Bode 2003;Ho, Bahcall & Bode 2006;Rozo et al 2010;Weinberg et al 2015;Jauzac et al 2016;Schwinn et al 2017;Mao et al 2018;Fluri et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Mapping dark matter and finding filaments: calibration of lensing analysis techniques on simulated data

Tam

Massey

Jauzac

et al. 2020

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

ABSTRACT We quantify the performance of mass mapping techniques on mock imaging and gravitational lensing data of galaxy clusters. The optimum method depends upon the scientific goal. We assess measurements of clusters’ radial density profiles, departures from sphericity, and their filamentary attachment to the cosmic web. We find that mass maps produced by direct (KS93) inversion of shear measurements are unbiased, and that their noise can be suppressed via filtering with mrlens. Forward-fitting techniques, such as lenstool, suppress noise further, but at a cost of biased ellipticity in the cluster core and overestimation of mass at large radii. Interestingly, current searches for filaments are noise-limited by the intrinsic shapes of weakly lensed galaxies, rather than by the projection of line-of-sight structures. Therefore, space-based or balloon-based imaging surveys that resolve a high density of lensed galaxies could soon detect one or two filaments around most clusters.

show abstract

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

Cited by 151 publications

References 81 publications

Emulation of Cosmological Mass Maps with Conditional Generative Adversarial Networks

Emulation of Cosmological Mass Maps with Conditional Generative Adversarial Networks

Point source detection with fully convolutional networks

Mapping dark matter and finding filaments: calibration of lensing analysis techniques on simulated data

Contact Info

Product

Resources

About