Related papers: Denoising weak lensing mass maps with diffusion mo…
Previous raw image-based low-light image enhancement methods predominantly relied on feed-forward neural networks to learn deterministic mappings from low-light to normally-exposed images. However, they failed to capture critical…
We analyse the first measurements from precision weak lensing (PWL): a new methodology for measuring individual galaxy-galaxy weak lensing through velocity information. Our goal is to understand the observed shear distribution from PWL,…
Gravitational waves (GWs) propagating through the universe can be microlensed by stellar and intermediate-mass objects. Lensing induces frequency-dependent amplification of GWs, which can be computed using \texttt{GLoW}, an accurate code…
We introduce the state-of-the-art deep learning Denoising Diffusion Probabilistic Model (DDPM) as a method to infer the volume or number density of giant molecular clouds (GMCs) from projected mass surface density maps. We adopt…
Weak gravitational lensing (WL) causes distortions of galaxy images and probes massive structures on large scales, allowing us to understand the late-time evolution of the Universe. One way to extract the cosmological information from WL is…
Image denoising (removal of additive white Gaussian noise from an image) is one of the oldest and most studied problems in image processing. An extensive work over several decades has led to thousands of papers on this subject, and to many…
The study of dark matter substructure through strong gravitational lensing has shown enormous promise in probing the properties of dark matter on sub-galactic scales. This approach has already been used to place strong constraints on a wide…
Intrinsic alignments of galaxies can mimic to an extent the effects of shear caused by weak gravitational lensing. Previous studies have shown that for shallow surveys with median redshifts z_m = 0.1, the intrinsic alignment dominates the…
Denoising diffusion probabilistic models (DDPMs) have achieved impressive performance on various image generation tasks, including image super-resolution. By learning to reverse the process of gradually diffusing the data distribution into…
Every signal propagating through the universe is at least weakly lensed by the intervening gravitational field. In some situations, wave-optics phenomena (diffraction, interference) can be observed as frequency-dependent modulations of the…
Recently, Generative Adversarial Networks (GANs) have emerged as a popular alternative for modeling complex high dimensional distributions. Most of the existing works implicitly assume that the clean samples from the target distribution are…
Galaxies, diffuse gas, and dark matter make up the cosmic web that defines the large-scale structure of the Universe. We constrained the joint distribution of these constituents by cross-correlating galaxy samples binned by stellar mass…
As a class of generative artificial intelligence frameworks inspired by statistical physics, diffusion models have shown extraordinary performance in synthesizing complicated data distributions through a denoising process gradually guided…
We consider the inverse problem of denoising an image where each point (pixel) is an element of a target set, which we refer to as a target-valued image. The target sets considered are either (i) a closed convex set of Euclidean space or…
Gravitational lensing is a universal phenomenon: it affects both gravitational waves (GWs) and electromagnetic signals travelling through the gravitational field of a massive object. In this work, we explore the prospects of observing…
We present a deep neural network to reduce coherent noise in three-dimensional quantitative phase imaging. Inspired by the cycle generative adversarial network, the denoising network was trained to learn a transform between two image…
Generative adversarial networks (GANs) are frequently utilized in astronomy to construct an emulator of numerical simulations. Nevertheless, training GANs can prove to be a precarious task, as they are prone to instability and often lead to…
Generative adversarial network (GAN) is a framework for generating fake data using a set of real examples. However, GAN is unstable in the training stage. In order to stabilize GANs, the noise injection has been used to enlarge the overlap…
Strong generative models can accurately learn channel distributions. This could save recurring costs for physical measurements of the channel. Moreover, the resulting differentiable channel model supports training neural encoders by…
Gravitational lensing is the relativistic effect generated by massive bodies, which bend the space-time surrounding them. It is a deeply investigated topic in astrophysics and allows validating theoretical relativistic results and studying…