Related papers: Evaluating Physically Motivated Loss Functions for…

Subjective image quality measures based on deep neural networks are very related to models of visual neuroscience. This connection benefits engineering but, more interestingly, the freedom to optimize deep networks in different ways, make…

I propose an analysis method, based on spin-spherical harmonics and spherical Bessel functions, for large-scale weak lensing surveys which have source distance information through photometric redshifts. I show that the distance information…

In this paper, orthogonal to the existing data and model studies, we instead resort our efforts to investigate the potential of loss function in a new perspective and present our belief ``Random Weights Networks can Be Acted as Loss Prior…

We analyze gradient descent with randomly weighted data points in a linear regression model, under a generic weighting distribution. This includes various forms of stochastic gradient descent, importance sampling, but also extends to…

Supervised artificial neural networks are used to predict useful properties of galaxies in the Sloan Digital Sky Survey, in this instance morphological classifications, spectral types and redshifts. By giving the trained networks unseen…

We present new tests to identify stationary position-dependent additive shear biases in weak gravitational lensing data sets. These tests are important diagnostics for currently ongoing and planned cosmic shear surveys, as such biases…

This paper analyses the effects of random noise in determining errors and confidence levels for galaxy redshifts obtained by cross-correlation techniques. The main finding is that confidence levels have previously been overestimated, and…

Current methods based on Neural Radiance Fields fail in the low data limit, particularly when training on incomplete scene data. Prior works augment training data only in next-best-view applications, which lead to hallucinations and model…

Photometric Redshift is critical for analyzing astronomical objects, but existing ML methods often overlook the aleatoric uncertainties inherent in observed data. We introduce Starkindler, a novel training objective that explicitly…

This study investigates the potential accuracy boundaries of physics-informed neural networks, contrasting their approach with previous similar works and traditional numerical methods. We find that selecting improved optimization algorithms…

Forecasting weather accurately and efficiently is a critical capability in our ability to adapt to climate change. Data driven approaches to this problem have enjoyed much success recently providing forecasts with accuracy comparable to…

Pruning - that is, setting a significant subset of the parameters of a neural network to zero - is one of the most popular methods of model compression. Yet, several recent works have raised the issue that pruning may induce or exacerbate…

Photometric redshift estimation algorithms are often based on representative data from observational campaigns. Data-driven methods of this type are subject to a number of potential deficiencies, such as sample bias and incompleteness.…

Most image segmentation algorithms are trained on binary masks formulated as a classification task per pixel. However, in applications such as medical imaging, this "black-and-white" approach is too constraining because the contrast between…

Why rely on dense neural networks and then blindly sparsify them when prior knowledge about the problem structure is already available? Many inverse problems admit algorithm-unrolled networks that naturally encode physics and sparsity. In…

Recent work has focused on generating synthetic imagery to increase the size and variability of training data for learning visual tasks in urban scenes. This includes increasing the occurrence of occlusions or varying environmental and…

A recent article by Wojtak {\it et al} (arXiv:1504.00718) pointed out that the local gravitational redshift, despite its smallness ($\sim 10^{-5}$), can have a noticeable ($\sim 1\%$) systematic effect on our cosmological parameter…

Deep Neural Networks have achieved remarkable success relying on the developing availability of GPUs and large-scale datasets with increasing network depth and width. However, due to the expensive computation and intensive memory,…

Cosmology constraints serve as a crucial criterion in discriminating cosmological models. The traditional combined method to constrain the cosmological parameters designates the corresponding theoretical value and observational data as…

We present a cosmic shear analysis of the 100 square degree weak lensing survey, combining data from the CFHTLS-Wide, RCS, VIRMOS-DESCART and GaBoDS surveys. Spanning ~100 square degrees, with a median source redshift z~0.78, this combined…