3D ScatterNet: Inference from 21 cm Light-cones

TL;DR

This paper introduces the 3D ScatterNet, a novel deep learning method combining wavelet transforms and normalizing flows, to analyze 21 cm light-cones from the epoch of reionization, outperforming traditional approaches.

Contribution

The paper presents the 3D ScatterNet, a new approach that improves likelihood inference from 21 cm light-cones by integrating wavelet scattering and normalizing flows.

Findings

Outperforms existing likelihood inference methods.

Achieves better results than power spectrum-based inference.

Handles varied light-cone effects and contaminations effectively.

Abstract

The Square Kilometre Array (SKA) will have the sensitivity to take the 3D light-cones of the 21 cm signal from the epoch of reionization. This signal, however, is highly non-Gaussian and can not be fully interpreted by the traditional statistic using power spectrum. In this work, we introduce the 3D ScatterNet that combines the normalizing flows with solid harmonic wavelet scattering transform, a 3D CNN featurizer with inductive bias, to perform implicit likelihood inference (ILI) from 21 cm light-cones. We show that 3D ScatterNet outperforms the ILI with a fine-tuned 3D CNN in the literature. It also reaches better performance than ILI with the power spectrum for varied light-cone effects and varied signal contaminations.