Extracting Axion String Network Parameters from Simulated CMB Birefringence Maps using Convolutional Neural Networks

TL;DR

This paper demonstrates that spherical convolutional neural networks can effectively extract parameters of axion string networks from simulated CMB birefringence maps, surpassing traditional statistical methods, and assesses the impact of realistic noise levels.

Contribution

It introduces a novel pipeline using SCNNs to estimate axion string network parameters from birefringence maps, highlighting the potential of neural networks in this cosmological context.

Findings

Neural networks can extract information inaccessible to two-point statistics.

Noise significantly biases parameter estimation, requiring noise modeling in training.

SCNNs show promise for analyzing future CMB birefringence data.

Abstract

Axion-like particles may form a network of cosmic strings in the Universe today that can rotate the plane of polarization of cosmic microwave background (CMB) photons. Future CMB observations with improved sensitivity might detect this axion-string-induced birefringence effect, thereby revealing an as-yet unseen constituent of the Universe and offering a new probe of particles and forces that are beyond the Standard Model of Elementary Particle Physics. In this work, we explore how spherical convolutional neural networks (SCNNs) may be used to extract information about the axion string network from simulated birefringence maps. We construct a pipeline to simulate the anisotropic birefringence that would arise from an axion string network, and we train SCNNs to estimate three parameters related to the cosmic string length, the cosmic string abundance, and the axion-photon coupling. Our results demonstrate that neural networks are able to extract information from a birefringence map that is inaccessible with two-point statistics alone (i.e., the angular power spectrum). We also assess the impact of noise on the accuracy of our SCNN estimators, demonstrating that noise at the level anticipated for Stage IV (CMB-S4) measurements would significantly bias parameter estimation for SCNNs trained on noiseless simulated data, and necessitate modeling the noise in the training data.