Extracting cosmological parameters from N-body simulations using machine learning techniques

Abstract

We make use of snapshots taken from the Quijote suite of simulations, consisting of 2000 simulations where five cosmological parameters have been varied (${\Omega }_m$, ${\Omega }_b$, $h$, $n_s$ and ${\sigma }_8$) in order to investigate the possibility of determining them using machine learning techniques. In particular, we show that convolutional neural networks can be employed to accurately extract ${\Omega }_m$ and ${\sigma }_8$ from the N-body simulations, and that these parameters can also be found from the non-linear matter power spectrum obtained from the same suite of simulations using both random forest regressors and deep neural networks. We show that the power spectrum provides competitive results in terms of accuracy compared to using the simulations and that we can also estimate the scalar spectral index $n_s$ from the power spectrum, at a lower precision.

arXiv comment11 pages, 5 figures

References in corpus (20)

• The cosmological simulation code GADGET-2
• Efficient Computation of CMB anisotropies in closed FRW models
• Planck 2018 results. I. Overview and the cosmological legacy of Planck
• Transients from Initial Conditions in Cosmological Simulations
• The Microwave Anisotropy Probe (MAP) Mission
• ANNz: estimating photometric redshifts using artificial neural networks
• Transients from Initial Conditions: A Perturbative Analysis
• The Quijote simulations
• Fast likelihood-free cosmology with neural density estimators and active learning
• Finding strong lenses in CFHTLS using convolutional neural networks
• Fast cosmic web simulations with generative adversarial networks
• Analysing the 21cm signal from the Epoch of Reionization with artificial neural networks
• ShapeFit: extracting the power spectrum shape information in galaxy surveys beyond BAO and RSD
• DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks
• Interpreting deep learning models for weak lensing
• The PAU Survey: Photometric redshifts using transfer learning from simulations
• Large-Scale Gravitational Lens Modeling with Bayesian Neural Networks for Accurate and Precise Inference of the Hubble Constant
• DeepSZ: Identification of Sunyaev-Zel'dovich Galaxy Clusters using Deep Learning
• Deep Forest: Neural Network reconstruction of the Lyman-alpha forest
• An Unbiased Estimator of the Full-sky CMB Angular Power Spectrum at Large Scales using Neural Networks

Cited by in corpus (17)

• The CosmoVerse White Paper: Addressing observational tensions in cosmology with systematics and fundamental physics
• Learning cosmology and clustering with cosmic graphs
• Machine Learning for Observational Cosmology
• Robust Neural Network-Enhanced Estimation of Local Primordial Non-Gaussianity
• Fast generation of mock galaxy catalogues with COLA
• Constraining cosmological parameters from N-body simulations with Variational Bayesian Neural Networks
• Cosmological constraints from low redshift 21 cm intensity mapping with machine learning
• Bayesian deep learning for cosmic volumes with modified gravity
• Inferring Cosmological Parameters on SDSS via Domain-Generalized Neural Networks and Lightcone Simulations
• Deep learning for cosmological parameter inference from a dark matter halo density field
• Non-Linearity-Free prediction of the growth-rate $f{\sigma }_8$ using Convolutional Neural Networks
• Clusternets: A deep learning approach to probe clustering dark energy
• kSZ Pairwise Velocity Reconstruction with Machine Learning
• Anisotropic cosmology using observational datasets: exploring via machine learning approaches
• Cosmology with Persistent Homology: Parameter Inference via Machine Learning
• Power Spectrum Emulators from Neural Networks and Tree-Based Methods
• Exploring Machine Learning Regression Models for Advancing Foreground Mitigation and Global 21cm Signal Parameter Extraction