Learning neutrino effects in Cosmology with Convolutional Neural Networks

Abstract

Measuring the sum of the three active neutrino masses, $M_{\nu }$, is one of the most important challenges in modern cosmology. Massive neutrinos imprint characteristic signatures on several cosmological observables in particular on the large-scale structure of the Universe. In order to maximize the information that can be retrieved from galaxy surveys, accurate theoretical predictions in the non-linear regime are needed. Currently, one way to achieve those predictions is by running cosmological numerical simulations. Unfortunately, producing those simulations requires high computational resources -- several hundred to thousand core-hours for each neutrino mass case. In this work, we propose a new method, based on a deep learning network, to quickly generate simulations with massive neutrinos from standard $\Lambda$CDM simulations without neutrinos. We computed multiple relevant statistical measures of deep-learning generated simulations, and conclude that our approach is an accurate alternative to the traditional N-body techniques. In particular the power spectrum is within $\simeq 6\%$ down to non-linear scales $k=0.7$~\rm h/Mpc. Finally, our method allows us to generate massive neutrino simulations 10,000 times faster than the traditional methods.

arXiv comment11 pages, 7 figures

References in corpus (43)

• Evidence for oscillation of atmospheric neutrinos
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Direct Evidence for Neutrino Flavor Transformation from Neutral-Current Interactions in the Sudbury Neutrino Observatory
• Observation of Reactor Electron Antineutrino Disappearance in the RENO Experiment
• Planck 2018 results. I. Overview and the cosmological legacy of Planck
• Planck 2018 results. V. CMB power spectra and likelihoods
• Measurement of Neutrino Oscillation with KamLAND: Evidence of Spectral Distortion
• Final Results from phase II of the Mainz Neutrino Mass Search in Tritium $\beta$ Decay
• Updated fit to three neutrino mixing: exploring the accelerator-reactor complementarity
• Updated fit to three neutrino mixing: status of leptonic CP violation
• Observation of Electron Neutrino Appearance in a Muon Neutrino Beam
• Neutrino oscillations refitted
• An improved upper limit on the neutrino mass from a direct kinematic method by KATRIN
• Unveiling $\nu$ secrets with cosmological data: neutrino masses and mass hierarchy
• Reactor electron antineutrino disappearance in the Double Chooz experiment
• The Quijote simulations
• Measurement of Neutrino Oscillations with the MINOS Detectors in the NuMI Beam
• Massive Neutrinos and the Non-linear Matter Power Spectrum
• Neutrino masses and cosmology with Lyman-alpha forest power spectrum
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• Learning to Predict the Cosmological Structure Formation
• Cosmology with massive neutrinos II: on the universality of the halo mass function and bias
• DEMNUni: The clustering of large-scale structures in the presence of massive neutrinos
• On the improvement of cosmological neutrino mass bounds
• Cosmology with massive neutrinos I: towards a realistic modeling of the relation between matter, haloes and galaxies
• Voids in massive neutrino cosmologies
• Constraining $M_{\nu }$ with the Bispectrum I: Breaking Parameter Degeneracies
• Massive Neutrinos Leave Fingerprints on Cosmic Voids
• Simulating nonlinear cosmological structure formation with massive neutrinos
• Scale-dependent galaxy bias, CMB lensing-galaxy cross-correlation, and neutrino masses
• The imprint of neutrinos on clustering in redshift-space
• Fast cosmic web simulations with generative adversarial networks
• Distinguishing standard and modified gravity cosmologies with machine learning
• DEMNUni: Massive neutrinos and the bispectrum of large scale structures
• How to add massive neutrinos to your $\Lambda$CDM simulation -- extending cosmology rescaling algorithms
• Cosmic voids: a novel probe to shed light on our Universe
• Accurate fitting functions for peculiar velocity spectra in standard and massive-neutrino cosmologies
• A Hybrid Deep Learning Approach to Cosmological Constraints From Galaxy Redshift Surveys
• Painting halos from cosmic density fields of dark matter with physically motivated neural networks
• Cosmological constraints from galaxy clustering in the presence of massive neutrinos
• Neutrino Signatures on the High Transmission Regions of the Lyman-alpha Forest
• A black box for dark sector physics: Predicting dark matter annihilation feedback with conditional GANs
• A Neural-Network based estimator to search for primordial non-Gaussianity in Planck CMB maps

Cited by in corpus (4)

• COmoving Computer Acceleration (COCA): $N$-body simulations in an emulated frame of reference
• High-precision measurements of the atomic mass and electron-capture decay $Q$ value of ${}^{95}$Tc
• nuGAN: Generative Adversarial Emulator for Cosmic Web with Neutrinos
• Predicting large scale cosmological structure evolution with generative adversarial network-based autoencoders