Cleaning our own Dust: Simulating and Separating Galactic Dust Foregrounds with Neural Networks

Abstract

Separating galactic foreground emission from maps of the cosmic microwave background (CMB), and quantifying the uncertainty in the CMB maps due to errors in foreground separation are important for avoiding biases in scientific conclusions. Our ability to quantify such uncertainty is limited by our lack of a model for the statistical distribution of the foreground emission. Here we use a Deep Convolutional Generative Adversarial Network (DCGAN) to create an effective non-Gaussian statistical model for intensity of emission by interstellar dust. For training data we use a set of dust maps inferred from observations by the Planck satellite. A DCGAN is uniquely suited for such unsupervised learning tasks as it can learn to model a complex non-Gaussian distribution directly from examples. We then use these simulations to train a second neural network to estimate the underlying CMB signal from dust-contaminated maps. We discuss other potential uses for the trained DCGAN, and the generalization to polarized emission from both dust and synchrotron.

References in corpus (18)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• HEALPix -- a Framework for High Resolution Discretization, and Fast Analysis of Data Distributed on the Sphere
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
• The Simons Observatory: Science goals and forecasts
• Improved Training of Wasserstein GANs
• SPT-3G: A Next-Generation Cosmic Microwave Background Polarization Experiment on the South Pole Telescope
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• CosmoGAN: creating high-fidelity weak lensing convergence maps using Generative Adversarial Networks
• BICEP2 / Keck Array V: Measurements of B-mode Polarization at Degree Angular Scales and 150 GHz by the Keck Array
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks
• Planck intermediate results. XLVIII. Disentangling Galactic dust emission and cosmic infrared background anisotropies
• {\sc CosmoNet}: fast cosmological parameter estimation in non-flat models using neural networks
• Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect
• Microwave multiplexing on the Keck Array
• Fast Wiener filtering of CMB maps with Neural Networks

Cited by in corpus (16)

• New Interpretable Statistics for Large Scale Structure Analysis and Generation
• Full-sky Cosmic Microwave Background Foreground Cleaning Using Machine Learning
• Impact of polarised extragalactic sources on the measurement of CMB B-mode anisotropies
• ForSE: a GAN based algorithm for extending CMB foreground models to sub-degree angular scales
• A new approach for the statistical denoising of Planck interstellar dust polarization data
• Recovering the CMB Signal with Machine Learning
• Non-Gaussian modelling and statistical denoising of Planck dust polarization full-sky maps using scattering transforms
• MillimeterDL: Deep Learning Simulations of the Microwave Sky
• Inpainting Galactic Foreground Intensity and Polarization maps using Convolutional Neural Network
• Single frequency CMB B-mode inference with realistic foregrounds from a single training image
• A Generative Model of Galactic Dust Emission Using Variational Inference
• Machine Learning and Cosmology
• Generative Models of Multi-channel Data from a Single Example -- Application to Dust Emission
• Self-supervised component separation for the extragalactic submillimeter sky
• ForSE+: Simulating non-Gaussian CMB foregrounds at 3 arcminutes in a stochastic way based on a generative adversarial network
• Simulating a full-sky high resolution Galactic synchrotron spectral index map using neural networks