Weak lensing peak statistics in the era of large scale cosmological surveys

Fluri, Janis; Kacprzak, T.; Sgier, Raphaël; Réfrégier, Alexandre; Amara, Adam

doi:10.1088/1475-7516/2018/10/051

Cited by 34 publications

(23 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This can be difficult. Certain approximated and fast simulation methods have been proposed (Fluri et al 2018). The full analyses of the covariance calculations are beyond the scope of the current paper, and will be studied in our future investigations.…”

Section: The Dependence Of Cosmological Gains On the Number Of Redshi...mentioning

confidence: 99%

Cosmological Studies from Tomographic Weak Lensing Peak Abundances and Impacts of Photo-z Errors

Yuan

Pan

Liu³

et al. 2019

ApJ

View full text Add to dashboard Cite

Weak lensing peak abundance analyses have been applied in different surveys and demonstrated to be a powerful statistics in extracting cosmological information complementary to cosmic shear two-point correlation studies. Future large surveys with high number densities of galaxies enable tomographic peak analyses. Focusing on high peaks, we investigate quantitatively how the tomographic redshift binning can enhance the cosmological gains. We also perform detailed studies about the degradation of cosmological information due to photometric redshift (photo-z) errors. We show that for surveys with the number density of galaxies ∼ 40 arcmin −2 , the median redshift ∼ 1, and the survey area of ∼ 15000 deg 2 , the 4-bin tomographic peak analyses can reduce the error contours of (Ω m , σ 8 ) by a factor of 5 comparing to 2-D peak analyses in the ideal case of photo-z error being absent. More redshift bins can hardly lead to significantly better constraints. The photo-z error model here is parametrized by z bias and σ ph and the fiducial values of z bias = 0.003 and σ ph = 0.02 is taken. We find that using tomographic peak analyses can constrain the photo-z errors simultaneously with cosmological parameters. For 4-bin analyses, we can obtain σ(z bias )/z bias ∼ 10% and σ(σ ph )/σ ph ∼ 5% without assuming priors on them. Accordingly, the cosmological constraints on Ω m and σ 8 degrade by a factor of ∼ 2.2 and ∼ 1.8, respectively, with respect to zero uncertainties on photo-z parameters. We find that the uncertainty of z bias plays more significant roles in degrading the cosmological constraints than that of σ ph .

show abstract

Section: The Dependence Of Cosmological Gains On the Number Of Redshi...mentioning

confidence: 99%

Cosmological Studies from Tomographic Weak Lensing Peak Abundances and Impacts of Photo-z Errors

Yuan

Pan

Liu³

et al. 2019

ApJ

View full text Add to dashboard Cite

show abstract

“…Such approaches include weak lensing peak statistics (e.g. [10,[16][17][18][19][20][21][22]), the three-point correlations function (e.g. [23,24]) or machine learning based methods (e.g.…”

Section: Introductionmentioning

confidence: 99%

A Full $w$CDM Analysis of KiDS-1000 Weak Lensing Maps using Deep Learning

Fluri,

Kacprzak,

Lucchi

et al. 2022

Preprint

View full text Add to dashboard Cite

We present a full forward-modeled wCDM analysis of the KiDS-1000 weak lensing maps using graphconvolutional neural networks (GCNN). Utilizing the CosmoGrid, a novel massive simulation suite spanning six different cosmological parameters, we generate almost one million tomographic mock surveys on the sphere. Due to the large data set size and survey area, we perform a spherical analysis while limiting our map resolution to HEALPix n side = 512. We marginalize over systematics such as photometric redshift errors, multiplicative calibration and additive shear bias. Furthermore, we use a map-level implementation of the non-linear intrinsic alignment model along with a novel treatment of baryonic feedback to incorporate additional astrophysical nuisance parameters. We also perform a spherical power spectrum analysis for comparison. The constraints of the cosmological parameters are generated using a likelihood free inference method called Gaussian Process Approximate Bayesian Computation (GPABC). Finally, we check that our pipeline is robust against choices of the simulation parameters. We find constraints on the degeneracy parameter of S8 ≡ σ8 ΩM /0.3 = 0.78 +0.06 −0.06 for our power spectrum analysis and S8 = 0.79 +0.05 −0.05 for our GCNN analysis, improving the former by 16%. This is consistent with earlier analyses of the 2-point function, albeit slightly higher. Baryonic corrections generally broaden the constraints on the degeneracy parameter by about 10%. These results offer great prospects for full machine learning based analyses of on-going and future weak lensing surveys.

show abstract

“…Approaches based on weak lensing peak statistics (e.g. [13][14][15][16][17][18]) or the three-point correlations function (e.g. [19,20]) have been * janis.fluri@phys.ethz.ch applied to obtain cosmological constraints.…”

Section: Introductionmentioning

confidence: 99%

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

et al. 2019

View full text Add to dashboard Cite

Convolutional Neural Networks (CNN) have recently been demonstrated on synthetic data to improve upon the precision of cosmological inference. In particular they have the potential to yield more precise cosmological constraints from weak lensing mass maps than the two-point functions. We present the cosmological results with a CNN from the KiDS-450 tomographic weak lensing dataset, constraining the total matter density Ωm, the fluctuation amplitude σ8, and the intrinsic alignment amplitude AIA. We use a grid of N-body simulations to generate a training set of tomographic weak lensing maps. We test the robustness of the expected constraints to various effects, such as baryonic feedback, simulation accuracy, different value of H0, or the lightcone projection technique. We train a set of ResNet-based CNNs with varying depths to analyze sets of tomographic KiDS mass maps divided into 20 flat regions, with applied Gaussian smoothing of σ = 2.34 arcmin. The uncertainties on shear calibration and n(z) error are marginalized in the likelihood pipeline. Following a blinding scheme, we derive constraints of S8 = σ8(Ωm/0.3) 0.5 = 0.777 +0.038 −0.036 with our CNN analysis, with AIA = 1.398 +0.779 −0.724 . We compare this result to the power spectrum analysis on the same maps and likelihood pipeline and find an improvement of about 30% for the CNN. We discuss how our results offer excellent prospects for the use of deep learning in future cosmological data analysis.

show abstract

Weak lensing peak statistics in the era of large scale cosmological surveys

Cited by 34 publications

References 56 publications

Cosmological Studies from Tomographic Weak Lensing Peak Abundances and Impacts of Photo-z Errors

Cosmological Studies from Tomographic Weak Lensing Peak Abundances and Impacts of Photo-z Errors

A Full $w$CDM Analysis of KiDS-1000 Weak Lensing Maps using Deep Learning

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

Contact Info

Product

Resources

About