Non-Gaussian information from weak lensing data via deep learning

TL;DR

This paper demonstrates that a deep learning approach applied to weak lensing maps can extract significantly more cosmological information than traditional power spectrum and peak statistics, especially on small scales.

Contribution

The authors develop and apply a CNN to simulated noiseless lensing maps, achieving substantially tighter constraints on cosmological parameters than existing statistical methods.

Findings

CNN yields ~5x tighter constraints than power spectrum.

CNN outperforms lensing peaks by ~4x.

Deep learning captures non-Gaussian information beyond traditional statistics.

Abstract

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 2D convolutional neural network to simulated noiseless lensing maps covering 96 different cosmological models over a range of {$\Omega_m,\sigma_8$}. Using the area of the confidence contour in the {$\Omega_m,\sigma_8$} plane as a figure-of-merit, derived from simulated convergence maps smoothed on a scale of 1.0 arcmin, we show that the neural network yields $\approx 5 \times$ tighter constraints than the power spectrum, and $\approx 4 \times$ tighter than the lensing peaks. Such gains illustrate the extent to which weak lensing data encode cosmological information not accessible to the power spectrum or even other, non-Gaussian statistics such as lensing peaks.