Related papers: Correcting for interloper contamination in the pow…

Deep neural networks have established as a powerful tool for large scale supervised classification tasks. The state-of-the-art performances of deep neural networks are conditioned to the availability of large number of accurately labeled…

We have developed a fast, accurate and generally applicable method for inferring the power spectrum and its uncertainties from maps of the cosmic microwave background (CMB) in the presence of inhomogeneous and correlated noise. For maps…

Coherent noise regularly plagues seismic recordings, causing artefacts and uncertainties in products derived from down-the-line processing and imaging tasks. The outstanding capabilities of deep learning in denoising of natural and medical…

The challenge of learning with noisy labels is significant in machine learning, as it can severely degrade the performance of prediction models if not addressed properly. This paper introduces a novel framework that conceptualizes noisy…

Noisy labels are ubiquitous in real-world datasets, especially in the large-scale ones derived from crowdsourcing and web searching. It is challenging to train deep neural networks with noisy datasets since the networks are prone to…

We describe a novel end-to-end approach using Machine Learning to reconstruct the power spectrum of cosmological density perturbations at high redshift from observed quasar spectra. State-of-the-art cosmological simulations of structure…

We derive and implement a full Bayesian large scale structure inference method aiming at precision recovery of the cosmological power spectrum from galaxy redshift surveys. Our approach improves over previous Bayesian methods by performing…

We exploit the synergy between low-resolution spectroscopy and photometric redshifts to study environmental effects on galaxy evolution in slitless spectroscopic surveys from space. As a test case, we consider the future Euclid Deep survey…

Recently deep neural networks have been successfully used for various classification tasks, especially for problems with massive perfectly labeled training data. However, it is often costly to have large-scale credible labels in real-world…

The uncertainties in photometric redshifts and stellar masses from imaging surveys affect galaxy sample selection, their abundance measurements, as well as the measured weak lensing signals. We develop a framework to assess the systematic…

Neural networks are central to modern artificial intelligence, yet their training remains highly sensitive to data contamination. Standard neural classifiers are trained by minimizing the categorical cross-entropy loss, corresponding to…

Line-intensity mapping (LIM) provides a promising way to probe cosmology, reionization and galaxy evolution. However, its sensitivity to cosmology and astrophysics at the same time is also a nuisance. Here we develop a comprehensive…

Methods are studied to compute the boosting effects produced by the observer motion that modifies and transfers to higher l the isotropic monopole frequency spectrum of the cosmic background. Explicit analytical solutions for spherical…

Estimating redshifts from broadband photometry is often limited by how accurately we can map the colors of galaxies to an underlying spectral template. Current techniques utilize spectrophotometric samples of galaxies or spectra derived…

Deep neural networks trained with standard cross-entropy loss memorize noisy labels, which degrades their performance. Most research to mitigate this memorization proposes new robust classification loss functions. Conversely, we propose a…

We present a method for suppressing contributions from higher-order terms in perturbation theory, greatly increasing the amount of information which may be extracted from the matter power spectrum. In an evolved cosmological density field…

A goal for pathfinder intensity mapping (IM) surveys will be detecting features in the neutral hydrogen (HI) power spectrum, which serve as conclusive evidence of cosmological signals. Observing such features at the expected scales in HI IM…

Training deep neural networks with noisy labels remains a significant challenge, often leading to degraded performance. Existing methods for handling label noise typically rely on either transition matrix, noise detection, or meta-learning…

Upcoming galaxy redshift surveys promise to significantly improve current limits on primordial non-Gaussianity (PNG) through measurements of 2- and 3-point correlation functions in Fourier space. However, realizing the full potential of…

Theoretical modeling of the redshift-space power spectrum of galaxies is crucially important to correctly extract cosmological information from redshift surveys. The task is complicated by the nonlinear biasing and redshift space distortion…