Neural Networks for cosmological model selection and feature importance using Cosmic Microwave Background data

Abstract

The measurements of the temperature and polarisation anisotropies of the Cosmic Microwave Background (CMB) by the ESA Planck mission have strongly supported the current concordance model of cosmology. However, the latest cosmological data release from ESA Planck mission still has a powerful potential to test new data science algorithms and inference techniques. In this paper, we use advanced Machine Learning (ML) algorithms, such as Neural Networks (NNs), to discern among different underlying cosmological models at the angular power spectra level, using both temperature and polarisation Planck 18 data. We test two different models beyond $\Lambda$CDM: a modified gravity model: the Hu-Sawicki model, and an alternative inflationary model: a feature-template in the primordial power spectrum. Furthermore, we also implemented an interpretability method based on SHAP values to evaluate the learning process and identify the most relevant elements that drive our architecture to certain outcomes. We find that our NN is able to distinguish between different angular power spectra successfully for both alternative models and $\Lambda$CDM. We conclude by explaining how archival scientific data has still a strong potential to test novel data science algorithms that are interesting for the next generation of cosmological experiments.

24 pages, 9 figures, 2 tables, comments welcome

References in corpus (63)

• Deep Learning in Neural Networks: An Overview
• First Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Determination of Cosmological Parameters
• Nine-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Parameter Results
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample
• Deep Learning using Rectified Linear Units (ReLU)
• Planck 2018 results. X. Constraints on inflation
• A Comprehensive Measurement of the Local Value of the Hubble Constant with 1 km/s/Mpc Uncertainty from the Hubble Space Telescope and the SH0ES Team
• Models of f(R) Cosmic Acceleration that Evade Solar-System Tests
• Dark Energy Survey Year 3 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing
• Planck 2013 results. XXII. Constraints on inflation
• The Dark Energy Survey: more than dark energy - an overview
• The Atacama Cosmology Telescope: DR4 Maps and Cosmological Parameters
• Euclid. I. Overview of the Euclid mission
• The Large Scale Structure of f(R) Gravity
• The Cosmic Linear Anisotropy Solving System (CLASS) I: Overview
• Planck 2018 results. VII. Isotropy and Statistics of the CMB
• Overview of the Instrumentation for the Dark Energy Spectroscopic Instrument
• The Primordial Lithium Problem
• The Kilo-Degree Survey
• Matter density perturbations and effective gravitational constant in modified gravity models of dark energy
• Planck Early Results XVIII: The power spectrum of cosmic infrared background anisotropies
• Features and New Physical Scales in Primordial Observables: Theory and Observation
• Foreground component separation with generalised ILC
• Observational constraints on viable f(R) parametrizations with geometrical and dynamical probes
• New observational constraints on $f(R)$ gravity from cosmic chronometers
• Hubble tension or a transition of the Cepheid SnIa calibrator parameters?
• Probing primordial features with future galaxy surveys
• Distinguishing standard and modified gravity cosmologies with machine learning
• A comparison of Einstein-Boltzmann solvers for testing General Relativity
• Constraints on $f(R)$ theories of gravity from GW170817
• f(R) Gravity and its Cosmological Implications
• Cosmological Constraints on sub-horizon scales modified gravity theories with MGCLASS II
• Probing primordial features with next-generation photometric and radio surveys
• Machine Learning for Observational Cosmology
• Dipole Cosmology: The Copernican Paradigm Beyond FLRW
• Observational constraints on Starobinsky $f(R)$ cosmology from cosmic expansion and structure growth data
• CMB Anomalies and the Hubble Tension
• SHAPing the Gas: Understanding Gas Shapes in Dark Matter Haloes with Interpretable Machine Learning
• Euclid: The search for primordial features
• Recovering the CMB Signal with Machine Learning
• Lecture Notes on CMB Theory: From Nucleosynthesis to Recombination
• New late-time constraints on $f(R)$ gravity
• The information content of cosmic microwave background anisotropies
• Investigating the accelerated expansion of the Universe through updated constraints on viable $f(R)$ models within the metric formalism
• Screenings in Modified Gravity: a perturbative approach
• Machine learning cosmic inflation
• Machine Learning and Cosmology
• Pantheon+ constraints on dark energy and modified gravity: An evidence of dynamical dark energy
• Constraining f(R) gravity with cross-correlation of galaxies and cosmic microwave background lensing
• Recovering Cosmic Microwave Background Polarization Signals with Machine Learning
• Searching for local features in primordial power spectrum using genetic algorithms
• Bayesian deep learning for cosmic volumes with modified gravity
• Enhancing Cosmological Model Selection with Interpretable Machine Learning
• Joint survey processing: combined resampling and convolution for galaxy modelling and deblending
• Unraveling the CMB lack-of-correlation anomaly with the cosmological gravitational wave background
• Lensing reconstruction from the cosmic microwave background polarization with machine learning
• Deep learning for cosmological parameter inference from a dark matter halo density field
• Classifying CMB time-ordered data through deep neural networks
• Non-Linearity-Free prediction of the growth-rate $f{\sigma }_8$ using Convolutional Neural Networks
• Cosmic Kite: Auto-encoding the Cosmic Microwave Background
• Introducing the DREAMS Project: DaRk mattEr and Astrophysics with Machine learning and Simulations
• Signal-preserving CMB component separation with machine learning

Cited by in corpus (1)

• Conditional variational autoencoders for cosmological model discrimination and anomaly detection in cosmic microwave background power spectra