Baryon acoustic oscillations reconstruction using convolutional neural networks

2020/02/29 by Tian-Xiang Mao, Tianxiang Mao, Jie Wang +7 · 1 citation
Computer Science · Physics and Astronomy · #Algorithm #Artificial intelligence #Artificial neural network #Astrophysics #Baryon #Baryon acoustic oscillations #Computer science #Convolutional neural network #Cosmology #Dark energy #Data set #Galaxies: Formation, Evolution, Phenomena #Galaxy #Physics #Pulsars and Gravitational Waves Research #Radio Astronomy Observations and Technology #Set (abstract data type) #astro-ph.CO #cs.LG

paper · pdf · doi:10.1093/mnras/staa3741

published as MNRAS, 501, 1499-1510 (2021) · Accepted for publication in MNRAS

openalex created_date 2020/03/06 · arxiv created 2020/12/03 · openalex publication_date 2020/12/03 · arxiv updated 2021/02/09 · openalex updated_date 2026/08/05

Abstract

ABSTRACT We propose a new scheme to reconstruct the baryon acoustic oscillations (BAO) signal, which contains key cosmological information, based on deep convolutional neural networks (CNN). Trained with almost no fine tuning, the network can recover large-scale modes accurately in the test set: the correlation coefficient between the true and reconstructed initial conditions reaches 90 \rm per cent at k≤ 0.2 hMpc^-1, which can lead to significant improvements of the BAO signal-to-noise ratio down to k≃ 0.4 hMpc^-1. Since this new scheme is based on the configuration-space density field in sub-boxes, it is local and less affected by survey boundaries than the standard reconstruction method, as our tests confirm. We find that the network trained in one cosmology is able to reconstruct BAO peaks in the others, i.e. recovering information lost to non-linearity independent of cosmology. The accuracy of recovered BAO peak positions is far less than that caused by the difference in the cosmology models for training and testing, suggesting that different models can be distinguished efficiently in our scheme. It is very promising that our scheme provides a different new way to extract the cosmological information from the ongoing and future large galaxy surveys.

Citations

• Adam: A Method for Stochastic Optimization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• The 6dF Galaxy Survey: final redshift release (DR3) and southern large-scale structures
• Canadian Hydrogen Intensity Mapping Experiment (CHIME) pathfinder
• NINE-YEAR WILKINSON MICROWAVE ANISOTROPY PROBE ( WMAP ) OBSERVATIONS: FINAL MAPS AND RESULTS
• The WiggleZ Dark Energy Survey: improved distance measurements to z = 1 with reconstruction of the baryonic acoustic feature
• Observational Probes of Cosmic Acceleration
• Reconstructing the cosmic density field with the distribution of dark matter haloes
• Primeval Adiabatic Perturbation in an Expanding Universe
• Simulations of the formation, evolution and clustering of galaxies and quasars
• The Two Micron All Sky Survey (2MASS)
• Science at Very High Angular Resolution with the Square Kilometre Array
• The detectability of baryonic acoustic oscillations in future galaxy surveys
• FIVE-YEAR WILKINSON MICROWAVE ANISOTROPY PROBE OBSERVATIONS: DATA PROCESSING, SKY MAPS, AND BASIC RESULTS
• Primordial density and BAO reconstruction
• Update on the BINGO 21cm intensity mapping experiment
• Cosmological model discrimination with Deep Learning
• Don't Decay the Learning Rate, Increase the Batch Size
• nbodykit: an open-source, massively parallel toolkit for large-scale structure
• Deep learning
• CosmoFlow: Using Deep Learning to Learn the Universe at Scale
• Deep Residual Learning for Image Recognition
• Euclid Definition Study Report
• Planck 2015 results. XIII. Cosmological parameters

Cited by

• Interpreting Cosmological Information from Neural Networks in the Hydrodynamic Universe