LyNNA: A Deep Learning Field-level Inference Machine for the Lyman- Forest
arXiv:2311.02167 · doi:10.1051/0004-6361/202348485
Abstract
The inference of astrophysical and cosmological properties from the Lyman- forest conventionally relies on summary statistics of the transmission field that carry useful but limited information. We present a deep learning framework for inference from the Lyman- forest at field-level. This framework consists of a 1D residual convolutional neural network (ResNet) that extracts spectral features and performs regression on thermal parameters of the IGM that characterize the power-law temperature-density relation. We train this supervised machinery using a large set of mock absorption spectra from Nyx hydrodynamic simulations at with a range of thermal parameter combinations (labels). We employ Bayesian optimization to find an optimal set of hyperparameters for our network, and then employ a committee of 20 neural networks for increased statistical robustness of the network inference. In addition to the parameter point predictions, our machine also provides a self-consistent estimate of their covariance matrix with which we construct a pipeline for inferring the posterior distribution of the parameters. We compare the results of our framework with the traditional summary (PDF and power spectrum of transmission) based approach in terms of the area of the 68% credibility regions as our figure of merit (FoM). In our study of the information content of perfect (noise- and systematics-free) Ly forest spectral data-sets, we find a significant tightening of the posterior constraints -- factors of 10.92 and 3.30 in FoM over power spectrum only and jointly with PDF, respectively -- that is the consequence of recovering the relevant parts of information that are not carried by the classical summary statistics.
arXiv comment15 pages, 16 figures, to be published in Astronomy and Astrophysics. This update contains the peer-reviewd, accepted version of the paper
References in corpus (65)
- The Baryon Oscillation Spectroscopic Survey of SDSS-III
- Sloan Digital Sky Survey IV: Mapping the Milky Way, Nearby Galaxies, and the Distant Universe
- Constraining Warm Dark Matter candidates including sterile neutrinos and light gravitinos with WMAP and the Lyman-alpha forest
- Warm Dark Matter as a solution to the small scale crisis: new constraints from high redshift Lyman-alpha forest data
- Equation of State of the Photoionized Intergalactic Medium
- Cosmological parameters from combining the Lyman-alpha forest with CMB, galaxy clustering and SN constraints
- Baryon Acoustic Oscillations in the Ly-α forest of BOSS quasars
- Model-independent evidence in favor of an end to reionization by z~6
- Reionization of the Inhomogeneous Universe
- First constraints on fuzzy dark matter from Lyman- forest data and hydrodynamical simulations
- Overview of the Instrumentation for the Dark Energy Spectroscopic Instrument
- Evidence of patchy hydrogen reionization from an extreme Ly trough below redshift six
- Recovery of the Power Spectrum of Mass Fluctuations from Observations of the Lyman-alpha Forest
- The thermal history of the intergalactic medium
- The Completed SDSS-IV extended Baryon Oscillation Spectroscopic Survey: Baryon acoustic oscillations with Lyman- forests
- The Observed Probability Distribution Function, Power Spectrum, and Correlation Function of the Transmitted Flux in the Lyman-alpha Forest
- Detection of Extended He II Reionization in the Temperature Evolution of the Intergalactic Medium
- Strong bound on canonical ultra-light axion dark matter from the Lyman-alpha forest
- Neutrino masses and cosmology with Lyman-alpha forest power spectrum
- Constraining the mass of light bosonic dark matter using SDSS Lyman- forest
- Nyx: A Massively Parallel AMR Code for Computational Cosmology
- Measurement of Baryon Acoustic Oscillations in the Lyman-alpha Forest Fluctuations in BOSS Data Release 9
- Baryons in the CosmicWeb of IllustrisTNG - I: Gas in Knots, Filaments, Sheets and Voids
- The Lyman- forest in optically-thin hydrodynamical simulations
- A Refined Measurement of the Mean Transmitted Flux in the Ly-alpha Forest over 2 < z < 5 Using Composite Quasar Spectra
- Constraints on neutrino masses from Lyman-alpha forest power spectrum with BOSS and XQ-100
- New Constraints on IGM Thermal Evolution from the Lyα Forest Power Spectrum
- The one-dimensional power spectrum from the SDSS DR14 Ly forests
- A Measurement of Small Scale Structure in the 2.2 < z < 4.2 Lyman-alpha Forest
- Possible evidence for an inverted temperature-density relation in the intergalactic medium from the flux distribution of the Lyman-alpha forest
- Cosmological constraints with deep learning from KiDS-450 weak lensing maps
- Revealing reionization with the thermal history of the intergalactic medium: new constraints from the Lyman- flux power spectrum
- Probing the thermal state of the intergalactic medium at with the transmission spikes in high-resolution Ly forest spectra
- Non-Gaussian information from weak lensing data via deep learning
- The thermal history of the intergalactic medium down to redshift z=1.5: a new curvature measurement
- A consistent and robust measurement of the thermal state of the IGM at from a large sample of Ly forest spectra: Evidence for late and rapid HeII reionization
- Weak lensing cosmology with convolutional neural networks on noisy data
- A consistent determination of the temperature of the intergalactic medium at redshift z=2.4
- On the intergalactic temperature-density relation
- Cosmological and astrophysical constraints from the Lyman-alpha forest flux probability distribution function
- IGM Constraints from the SDSS-III/BOSS DR9 Ly-alpha Forest Flux Probability Distribution Function
- The intergalactic medium thermal history at redshift z=1.7--3.2 from the Lyman alpha forest: a comparison of measurements using wavelets and the flux distribution
- Cosmological constraints from noisy convergence maps through deep learning
- Deep Learning of Quasar Spectra to Discover and Characterize Damped Lya Systems
- Reionization history constraints from neural network based predictions of high-redshift quasar continua
- Inhomogeneous Reionization Models in Cosmological Hydrodynamical Simulations
- The Dark Energy Spectroscopic Instrument: One-dimensional power spectrum from first Lyman- forest samples with Fast Fourier Transform
- Constraining Reionization with the Lyman- Forest Power Spectrum: the Outlook after Planck
- A wavelet analysis of QSO spectra
- Searching for fluctuations in the IGM temperature using the Lyman alpha forest
- Inhomogeneous HeII reionization in Hydrodynamic Simulations
- Constraints on the cosmic expansion rate at redshift 2.3 from the Lyman- forest
- Simulating intergalactic gas for DESI-like small scale Lymanα forest observations
- Machine Learning for Observational Cosmology
- Spectral Analysis of the Lyman-Alpha Forest Using Wavelets
- Autoencoding Galaxy Spectra I: Architecture
- Autoencoding Galaxy Spectra II: Redshift Invariance and Outlier Detection
- Thermal state of the intergalactic medium at
- Modeling the Lyman- forest with Eulerian and SPH hydrodynamical methods
- A Novel Statistical Method for Measuring the Temperature-Density Relation in the IGM Using the - Distribution of Absorbers in the Ly Forest
- Fast, high-fidelity Lyman forests with convolutional neural networks
- Deep Forest: Neural Network reconstruction of the Lyman-alpha forest
- Improving IGM temperature constraints using wavelet analysis on high-redshift quasars
- Measuring the thermal and ionization state of the low- IGM using likelihood free inference
- Deep Forest: Neural Network reconstruction of intergalactic medium temperature