Inference of the optical depth to reionization $\tau$ from $\text{Planck}$ CMB maps with convolutional neural networks

Abstract

The optical depth to reionization, $\tau$, is the least constrained parameter of the cosmological $\Lambda$CDM model. To date, its most precise value is inferred from large-scale polarized CMB power spectra from the $Planck$ High-Frequency Instrument (HFI). These maps are known to contain significant contamination by residual non-Gaussian systematic effects, which are hard to model analytically. Therefore, robust constraints on $\tau$ are currently obtained through an empirical cross-spectrum likelihood built from simulations. In this paper, we present a likelihood-free inference of $\tau$ from polarized $Planck$ HFI maps which, for the first time, is fully based on neural networks (NNs). NNs have the advantage of not requiring an analytical description of the data and can be trained on state-of-the-art simulations, combining information from multiple channels. By using Gaussian sky simulations and $Planck$ $SRoll2$ simulations, including CMB, noise, and residual instrumental systematic effects, we train, test and validate NN models considering different setups. We infer the value of $\tau$ directly from $Q$ and $U$ maps at $\sim 4^{\circ }$ pixel resolution, without computing power spectra. On $Planck$ data, we obtain ${\tau }_{NN}=0.058\pm 0.008$, compatible with current EE cross-spectrum results but with a $\sim 30\%$ larger uncertainty, which can be assigned to the inherent non-optimality of our estimator and to the retraining procedure applied to avoid biases. While this paper does not improve on current cosmological constraints, our analysis represents a first robust application of NN-based inference on real data and highlights its potential as a promising tool for complementary analysis of near-future CMB experiments, also in view of the ongoing challenge to achieve a detection of primordial gravitational waves.

arXiv comment13 pages, 10 figures, 5 tables

References in corpus (36)

• Array Programming with NumPy
• Planck 2015 results. XIII. Cosmological parameters
• Overcoming catastrophic forgetting in neural networks
• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• Nine-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Cosmological Parameter Results
• Efficient Computation of CMB anisotropies in closed FRW models
• Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methods
• The Simons Observatory: Science goals and forecasts
• Planck 2018 results. V. CMB power spectra and likelihoods
• A Survey of z>5.8 Quasars in the Sloan Digital Sky Survey I: Discovery of Three New Quasars and the Spatial Density of Luminous Quasars at z~6
• Evidence for Reionization at z ~ 6: Detection of a Gunn-Peterson Trough in a z=6.28 Quasar
• Planck 2015 results. XI. CMB power spectra, likelihoods, and robustness of parameters
• Planck intermediate results. XLVI. Reduction of large-scale systematic effects in HFI polarization maps and estimation of the reionization optical depth
• Planck intermediate results. LVII. Joint Planck LFI and HFI data processing
• Hydrogen reionisation ends by $z=5.3$: Lyman-$\alpha$ optical depth measured by the XQR-30 sample
• The Discovery of a Luminous z=5.80 Quasar from the Sloan Digital Sky Survey
• Likelihood Analysis of CMB Temperature and Polarization Power Spectra
• How to measure CMB polarization power spectra without losing information
• Cosmological constraints with deep learning from KiDS-450 weak lensing maps
• Planck 2013 results. VI. High Frequency Instrument data processing
• Reionization and galaxy inference from the high-redshift Lyα forest
• Reionization optical depth determination from Planck HFI data with ten percent accuracy
• DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks
• The CAMELS Multifield Dataset: Learning the Universe's Fundamental Parameters with Artificial Intelligence
• SRoll2: an improved mapmaking approach to reduce large-scale systematic effects in the Planck High Frequency Instrument legacy maps
• Inference of the optical depth to reionization from low multipole temperature and polarisation Planck data
• Convolutional Neural Networks on the HEALPix sphere: a pixel-based algorithm and its application to CMB data analysis
• TEASING: a fast and accurate approximation for the low multipole likelihood of the Cosmic Microwave Background temperature
• Inferring the Thermal History of the Intergalactic Medium from the Properties of the Hydrogen and Helium Lyman-alpha Forest
• Likelihood methods for CMB experiments
• ForSE: a GAN based algorithm for extending CMB foreground models to sub-degree angular scales
• Recovering the CMB Signal with Machine Learning
• Single frequency CMB B-mode inference with realistic foregrounds from a single training image
• Convolutional Neural Network-reconstructed velocity for kinetic SZ detection
• CENN: A fully convolutional neural network for CMB recovery in realistic microwave sky simulations
• A novel CMB polarization likelihood package for large angular scales built from combined WMAP and Planck LFI legacy maps

Cited by in corpus (3)

• The state of the dark energy equation of state circa 2023
• Measuring the reionization optical depth without large-scale CMB polarization
• From few to many maps: A fast map-level emulator for extreme augmentation of CMB systematics datasets