Related papers: A Generative Modeling Approach to Reconstructing 2…
The redshifted 21 cm line is an emerging tool in cosmology, in principle permitting three-dimensional surveys of our Universe that reach unprecedentedly large volumes, previously inaccessible length scales, and hitherto unexplored epochs of…
This paper proposes a new framework to regularize the highly ill-posed and non-linear phase retrieval problem through deep generative priors using simple gradient descent algorithm. We experimentally show effectiveness of proposed algorithm…
The accurate and precise removal of 21-cm foregrounds from Epoch of Reionization redshifted 21-cm emission data is essential if we are to gain insight into an unexplored cosmological era. We apply a non-parametric technique, Generalized…
21 cm cosmology is a promising new probe of the evolution of visible matter in our universe, especially during the poorly-constrained Cosmic Dawn and Epoch of Reionization. However, in order to separate the 21 cm signal from bright…
The cosmic dawn 21-cm signal is a highly sensitive probe of any process which injects energy into the intergalactic medium, enabling novel searches for anomalous energy injection by through dark matter interactions. In addition to modifying…
There remains an important need for the development of image reconstruction methods that can produce diagnostically useful images from undersampled measurements. In magnetic resonance imaging (MRI), for example, such methods can facilitate…
A proposed method for dealing with foreground emission in upcoming 21-cm observations from the epoch of reionization is to limit observations to an uncontaminated window in Fourier space. Foreground emission can be avoided in this way,…
Whereas cryptography easily arouses attacks by means of encrypting a secret message into a suspicious form, steganography is advantageous for its resilience to attacks by concealing the message in an innocent-looking cover signal. Minimal…
Recently deep generative models have achieved impressive progress in modeling the distribution of training data. In this work, we present for the first time a generative model for 4D light field patches using variational autoencoders to…
Given the widespread availability of grids of models for stellar atmospheres, it is necessary to recover intermediate atmospheric models by means of accurate techniques that go beyond simple linear interpolation and capture the intricacies…
This paper shows how data-driven deep generative models can be utilized to solve challenging phase retrieval problems, in which one wants to reconstruct a signal from only few intensity measurements. Classical iterative algorithms are known…
Phase retrieval is the problem of reconstructing images from magnitude-only measurements. In many real-world applications the problem is underdetermined. When training data is available, generative models allow optimization in a…
The 21-cm forest is a promising probe of the Epoch of Reionization. The local state of the intergalactic medium (IGM) is encoded in the spectrum of a background source (radio-loud quasars or gamma ray burst afterglow) by absorption at the…
We consider the evolution of the sky-averaged 21 cm background during the early phases of structure formation. Using simple analytic models, we calculate the thermal and ionization histories, assuming that stellar photons dominate the…
End-to-end generative methods are considered a more promising solution for image restoration in physics-based vision compared with the traditional deconstructive methods based on handcrafted composition models. However, existing generative…
The cosmological global (sky-averaged) 21-cm signal is a powerful tool to probe the evolution of the intergalactic medium (IGM) in high-redshift Universe ($z \leq 6$). One of the biggest observational challenges is to remove the foreground…
Detection of redshifted \ion{H}{i} 21-cm emission is a potential probe for investigating the Universe's first billion years. However, given the significantly brighter foreground, detecting 21-cm is observationally difficult. The Earth's…
This paper addresses the classical problem of one-bit compressed sensing using a deep learning-based reconstruction algorithm that leverages a trained generative model to enhance the signal reconstruction performance. The generator, a…
This paper proposes a novel deep generative model, called BSDE-Gen, which combines the flexibility of backward stochastic differential equations (BSDEs) with the power of deep neural networks for generating high-dimensional complex target…
Cryogenic electron tomography is a technique for imaging biological samples in 3D. A microscope collects a series of 2D projections of the sample, and the goal is to reconstruct the 3D density of the sample called the tomogram.…