Neural style transfer of weak lensing mass maps

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shirasaki, Masato, Ikeda, Shiro
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914805775859712
author Shirasaki, Masato
Ikeda, Shiro
author_facet Shirasaki, Masato
Ikeda, Shiro
contents We propose a new generative model of projected cosmic mass density maps inferred from weak gravitational lensing observations of distant galaxies (weak lensing mass maps). We construct the model based on a neural style transfer so that it can transform Gaussian weak lensing mass maps into deeply non-Gaussian counterparts as predicted in ray-tracing lensing simulations. We develop an unpaired image-to-image translation method with Cycle-Consistent Generative Adversarial Networks (Cycle GAN), which learn efficient mapping from an input domain to a target domain. Our model is designed to enjoy important advantages; it is trainable with no need for paired simulation data, flexible to make the input domain visually meaningful, and expandable to rapidly-produce a map with a larger sky coverage than training data without additional learning. Using 10,000 lensing simulations, we find that appropriate labeling of training data based on field variance allows the model to reproduce a correct scatter in summary statistics for weak lensing mass maps. Compared with a popular log-normal model, our model improves in predicting the statistical natures of three-point correlations and local properties of rare high-density regions. We also demonstrate that our model enables us to produce a continuous map with a sky coverage of $\sim166\, \mathrm{deg}^2$ but similar non-Gaussian features to training data covering $\sim12\, \mathrm{deg}^2$ in a GPU minute. Hence, our model can be beneficial to massive productions of synthetic weak lensing mass maps, which is of great importance in future precise real-world analyses.
format Preprint
id arxiv_https___arxiv_org_abs_2310_17141
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Neural style transfer of weak lensing mass maps
Shirasaki, Masato
Ikeda, Shiro
Cosmology and Nongalactic Astrophysics
Instrumentation and Methods for Astrophysics
We propose a new generative model of projected cosmic mass density maps inferred from weak gravitational lensing observations of distant galaxies (weak lensing mass maps). We construct the model based on a neural style transfer so that it can transform Gaussian weak lensing mass maps into deeply non-Gaussian counterparts as predicted in ray-tracing lensing simulations. We develop an unpaired image-to-image translation method with Cycle-Consistent Generative Adversarial Networks (Cycle GAN), which learn efficient mapping from an input domain to a target domain. Our model is designed to enjoy important advantages; it is trainable with no need for paired simulation data, flexible to make the input domain visually meaningful, and expandable to rapidly-produce a map with a larger sky coverage than training data without additional learning. Using 10,000 lensing simulations, we find that appropriate labeling of training data based on field variance allows the model to reproduce a correct scatter in summary statistics for weak lensing mass maps. Compared with a popular log-normal model, our model improves in predicting the statistical natures of three-point correlations and local properties of rare high-density regions. We also demonstrate that our model enables us to produce a continuous map with a sky coverage of $\sim166\, \mathrm{deg}^2$ but similar non-Gaussian features to training data covering $\sim12\, \mathrm{deg}^2$ in a GPU minute. Hence, our model can be beneficial to massive productions of synthetic weak lensing mass maps, which is of great importance in future precise real-world analyses.
title Neural style transfer of weak lensing mass maps
topic Cosmology and Nongalactic Astrophysics
Instrumentation and Methods for Astrophysics
url https://arxiv.org/abs/2310.17141