Stochastic Super-resolution of Cosmological Simulations with Denoising Diffusion Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Schanz, Andreas, List, Florian, Hahn, Oliver
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913576891973632
author Schanz, Andreas
List, Florian
Hahn, Oliver
author_facet Schanz, Andreas
List, Florian
Hahn, Oliver
contents In recent years, deep learning models have been successfully employed for augmenting low-resolution cosmological simulations with small-scale information, a task known as "super-resolution". So far, these cosmological super-resolution models have relied on generative adversarial networks (GANs), which can achieve highly realistic results, but suffer from various shortcomings (e.g. low sample diversity). We introduce denoising diffusion models as a powerful generative model for super-resolving cosmic large-scale structure predictions (as a first proof-of-concept in two dimensions). To obtain accurate results down to small scales, we develop a new "filter-boosted" training approach that redistributes the importance of different scales in the pixel-wise training objective. We demonstrate that our model not only produces convincing super-resolution images and power spectra consistent at the percent level, but is also able to reproduce the diversity of small-scale features consistent with a given low-resolution simulation. This enables uncertainty quantification for the generated small-scale features, which is critical for the usefulness of such super-resolution models as a viable surrogate model for cosmic structure formation.
format Preprint
id arxiv_https___arxiv_org_abs_2310_06929
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Stochastic Super-resolution of Cosmological Simulations with Denoising Diffusion Models
Schanz, Andreas
List, Florian
Hahn, Oliver
Cosmology and Nongalactic Astrophysics
Instrumentation and Methods for Astrophysics
Machine Learning
In recent years, deep learning models have been successfully employed for augmenting low-resolution cosmological simulations with small-scale information, a task known as "super-resolution". So far, these cosmological super-resolution models have relied on generative adversarial networks (GANs), which can achieve highly realistic results, but suffer from various shortcomings (e.g. low sample diversity). We introduce denoising diffusion models as a powerful generative model for super-resolving cosmic large-scale structure predictions (as a first proof-of-concept in two dimensions). To obtain accurate results down to small scales, we develop a new "filter-boosted" training approach that redistributes the importance of different scales in the pixel-wise training objective. We demonstrate that our model not only produces convincing super-resolution images and power spectra consistent at the percent level, but is also able to reproduce the diversity of small-scale features consistent with a given low-resolution simulation. This enables uncertainty quantification for the generated small-scale features, which is critical for the usefulness of such super-resolution models as a viable surrogate model for cosmic structure formation.
title Stochastic Super-resolution of Cosmological Simulations with Denoising Diffusion Models
topic Cosmology and Nongalactic Astrophysics
Instrumentation and Methods for Astrophysics
Machine Learning
url https://arxiv.org/abs/2310.06929