Predictive uncertainty on improved astrophysics recovery from multifield cosmology

Abstract

We investigate how the constraints on cosmological and astrophysical parameters (${\Omega }_{\mathrm{m}}$, ${\sigma }_8$, $A_{SN1}$, $A_{SN2}$) vary when exploiting information from multiple fields in cosmology. We make use of a convolutional neural network to retrieve the salient features from different combinations of field maps from IllustrisTNG in the CAMELS project. The fields considered are neutral hydrogen (HI), gas density (Mgas), magnetic fields (B) and gas metallicity (Z). We estimate the predictive uncertainty on the predictions of our model by using Monte Carlo dropout, a Bayesian approximation. Results show that overall, the performance of the model improves on all parameters as the number of channels of its input is increased. As compared to previous works, our model is able to predict the astrophysical parameters with up to $5\%$ higher in accuracy. In the best setup which includes all fields (four channel input, Mgas-HI-B-Z) the model achieves $R^2>0.96$ on all parameters. Similarly, we find that the total uncertainty, which is dominated by the aleatoric uncertainty, decreases as more fields are used to train the model in general. The uncertainties obtained by dropout variational inference are overestimated on all parameters in our case, in that the predictive uncertainty is much larger than the actual squared error. After calibration, which consists of a simple $\sigma$ scaling method, the average deviation of the total uncertainty from the actual error goes down to $25\%$ at most (on $A_{SN1}$).

arXiv comment18 pages, 6 figures, 6 tables

References in corpus (39)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample
• The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Cosmological Implications from two Decades of Spectroscopic Surveys at the Apache Point observatory
• Simba: Cosmological Simulations with Black Hole Growth and Feedback
• Cosmological Parameters from the BOSS Galaxy Power Spectrum
• The clustering of galaxies in the SDSS-III Baryon Oscillation Spectroscopic Survey: RSD measurement from the power spectrum and bispectrum of the DR12 BOSS galaxies
• The CAMELS project: Cosmology and Astrophysics with MachinE Learning Simulations
• The shape of the SDSS DR5 galaxy power spectrum
• Constraining Early Dark Energy with Large-Scale Structure
• Combining Full-Shape and BAO Analyses of Galaxy Power Spectra: A 1.6% CMB-independent constraint on H0
• Fast Estimators for Redshift-Space Clustering
• The influence of baryons on the clustering of matter and weak lensing surveys
• Efficient Cosmological Analysis of the SDSS/BOSS data from the Effective Field Theory of Large-Scale Structure
• Fast Large Volume Simulations of the 21 cm Signal from the Reionization and pre-Reionization Epochs
• Cosmological constraints with deep learning from KiDS-450 weak lensing maps
• The Hubble Tension in Light of the Full-Shape Analysis of Large-Scale Structure Data
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: Cosmological implications of the Fourier space wedges of the final sample
• Non-Gaussian information from weak lensing data via deep learning
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: On the measurement of growth rate using galaxy correlation functions
• Deep learning from 21-cm tomography of the Cosmic Dawn and Reionization
• Cosmic Reionization and the 21-cm signal: Comparison between an analytical model and a simulation
• Full-shape cosmology analysis of SDSS-III BOSS galaxy power spectrum using emulator-based halo model: a $5\%$ determination of ${\sigma }_8$
• The DAWES review 10: The impact of deep learning for the analysis of galaxy surveys
• Galaxy Bias and Halo-Occupation Numbers from Large-Scale Clustering
• Cosmological constraints from the power spectrum of eBOSS emission line galaxies
• Constraining the astrophysics and cosmology from 21cm tomography using deep learning with the SKA
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• Simulation-Based Inference of Reionization Parameters From 3D Tomographic 21 cm Lightcone Images
• The CAMELS project: public data release
• Simulating the 21-cm signal from reionisation including non-linear ionisations and inhomogeneous recombinations
• Identifying Reionization Sources from 21cm Maps using Convolutional Neural Networks
• Machine learning astrophysics from 21 cm lightcones: impact of network architectures and signal contamination
• Removing Astrophysics in 21 cm maps with Neural Networks
• Constraining the reionization history using deep learning from 21cm tomography with the Square Kilometre Array
• Computing the Small-Scale Galaxy Power Spectrum and Bispectrum in Configuration-Space
• HInet: Generating neutral hydrogen from dark matter with neural networks
• A Faster Fourier Transform? Computing Small-Scale Power Spectra and Bispectra for Cosmological Simulations in $\mathcal{O}(N^2)$ Time
• AGN Outflows and the Matter Power Spectrum
• Multifield Cosmology with Artificial Intelligence

Cited by in corpus (1)

• Cosmological and lunar laser ranging constraints on evolving dark energy in a nonminimally coupled curvature-matter gravity model