Related papers: deep21: a Deep Learning Method for 21cm Foreground…

Convolutional neural networks (CNN) have demonstrated remarkable performance when the training and testing data are from the same distribution. However, such trained CNN models often largely degrade on testing data which is unseen and…

Deep Neural Networks (DNN) and especially Convolutional Neural Networks (CNN) are a de-facto standard for the analysis of large volumes of signals and images. Yet, their development and underlying principles have been largely performed in…

The global 21 cm signal from the hyperfine transition of cosmic atomic hydrogen is theorised to track the state of the early Universe via the analysis of its absorption and emission with respect to the radio background. Detecting this…

We present a novel dehazing and low-light enhancement method based on an illumination map that is accurately estimated by a convolutional neural network (CNN). In this paper, the illumination map is used as a component for three different…

Modern neural networks are usually highly over-parameterized. Behind the wide usage of over-parameterized networks is the belief that, if the data are simple, then the trained network will be automatically equivalent to a simple predictor.…

Autonomous Raman instruments on Mars rovers, deep-sea landers, and field robots must interpret raw spectra distorted by fluorescence baselines, peak shifts, and limited ground-truth labels. Using curated subsets of the RRUFF database, we…

The redshifted 21-cm signal of neutral Hydrogen is a promising probe into the period of evolution of our Universe when the first stars were formed (Cosmic Dawn), to the period where the entire Universe changed its state from being…

We investigate the prospect of constraining scalar field dark energy models using HI 21-cm intensity mapping surveys. We consider a wide class of coupled scalar field dark energy models whose predictions about the background cosmological…

It is well-known that foreground subtraction in 21cm surveys removes large scale power. We investigate associated systematic biases. We show that removing line-of-sight fluctuations on large scales aliases into suppression of the 3D power…

Deep convolutional neural networks (CNNs) for image denoising have recently attracted increasing research interest. However, plain networks cannot recover fine details for a complex task, such as real noisy images. In this paper, we…

To compress deep convolutional neural networks (CNNs) with large memory footprint and long inference time, this paper proposes a novel pruning criterion using layer-wised Ln-norm of feature maps. Different from existing pruning criteria,…

Processing marine seismic data is computationally demanding and consists of multiple time-consuming steps. Neural network based processing can, in theory, significantly reduce processing time and has the potential to change the way seismic…

While deep convolutional neural networks (CNNs) have achieved impressive success in image denoising with additive white Gaussian noise (AWGN), their performance remains limited on real-world noisy photographs. The main reason is that their…

We present a new application of deep learning to reconstruct the cosmic microwave background (CMB) temperature maps from the images of microwave sky, and to use these reconstructed maps to estimate the masses of galaxy clusters. We use a…

Face recognition is one of the most active tasks in computer vision and has been widely used in the real world. With great advances made in convolutional neural networks (CNN), lots of face recognition algorithms have achieved high accuracy…

We propose a new sequential classification model for astronomical objects based on a recurrent convolutional neural network (RCNN) which uses sequences of images as inputs. This approach avoids the computation of light curves or difference…

We introduce TPCNet, a neural network predictor that combines Convolutional and Transformer architectures with Positional encodings, for neutral atomic hydrogen (HI) spectral analysis. Trained on synthetic datasets, our models predict cold…

The highly redshifted 21 cm line of neutral hydrogen has become recognized as a unique probe of cosmology from relatively low redshifts (z ~ 1) up through the Epoch of Reionization (z ~ 8) and even beyond. To date, most work has focused on…

Convolutional Neural Networks (CNNs) are generally prone to noise interruptions, i.e., small image noise can cause drastic changes in the output. To suppress the noise effect to the final predication, we enhance CNNs by replacing…

In this work, we investigate the value of employing deep learning for the task of wireless signal modulation recognition. Recently in [1], a framework has been introduced by generating a dataset using GNU radio that mimics the imperfections…