Estimating Cosmological Parameters from the Dark Matter Distribution

By Siamak Ravanbakhsh, Junier B. Oliva, S. Fromenteau, Layne Price, Shirley Ho, Jeff Schneider, Barnab'as P'oczos. 2016

83 citations recorded by the source corpus.

A grand challenge of the 21st century cosmology is to accurately estimate the cosmological parameters of our Universe. A major approach in estimating the cosmological parameters is to use the large scale matter distribution of the Universe. Galaxy surveys provide the means to map out cosmic large-scale structure in three dimensions. Information about galaxy locations is typically summarized in a "single" function of scale, such as the galaxy correlation function or power-spectrum. We show that it is possible to estimate these cosmological parameters directly from the distribution of matter. This paper presents the application of deep 3D convolutional networks to volumetric representation of dark-matter simulations as well as the results obtained using a recently proposed distribution regression framework, showing that machine learning techniques are comparable to, and can sometimes outperform, maximum-likelihood point estimates using "cosmological models". This opens the way to estimating the parameters of our Universe with higher accuracy.

Subfields: Astronomy and Astrophysics, Computer Vision and Pattern Recognition, Ecology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Astronomy and Astrophysics

In Computer Vision and Pattern Recognition

In Ecology

R. Scoccimarro

Citations in the corpus

Learning to predict the cosmological structure formation (2018, background): “In particular, in astronomy and cosmology, a growing number of recent studies are using deep learning for a variety of tasks, ranging from analysis of cosmic microwave background (30–32), large-scale structure (33, 34), and gravitational lensing effects (35, 36) to classification of different light ”
Deep Sets (2017, methodology): “In these applications, point-cloud data is often converted to voxel or mesh representation at a preprocessing step ( e.g. Maturana & Scherer, 2015; Ravanbakhsh et al., 2016b; Lin et al., 2004).”
A volumetric deep Convolutional Neural Network for simulation of dark matter halo catalogues (2018, background): “In cosmology, machine learning techniques have shown promise for the purposes of cosmological parameter estimation (Ravanbakhsh et al. 2017; Gupta et al. 2018; Gillet et al. 2018), simulating twodimensional slices of the non-linear density field (Rodriguez et al. 2018), initial conditions…”
DeepSphere: Efficient spherical Convolutional Neural Network with HEALPix sampling for cosmological applications (2018, background): “Convolutional Neural Networks (CNNs) have been proposed as an alternative analysis tool in cosmology thanks to their ability to automatically design relevant statistics to maximise the precision 1 of the parameter estimation [9–16], while maintaining robustness to noise.”
CosmoFlow: Using Deep Learning to Learn the Universe at Scale (2018, methodology): “The aim of this paper is to predict three cosmological parameters by adapting the network topology developed by Ravanbakhsh et al. (2017) [3] and scale the problem to a dataset an order of magnitude larger in physical volume (i.e. 128 3 voxels), and an additional order of magnitude increase in…”
DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks (2018, background): “One interesting route would be to directly estimate cosmological parameters from the input CMB maps themselves, as introduced in e.g. Ravanbakhsh et al. (2017).”
The Quijote Simulations (2019, methodology): “Another way to approach the problem is to use advanced statistical techniques, such as machine/deep learning, to identify new and optimal statistics to extract cosmological information (Ravanbakhsh et al. 2017; 8 By information we mean the constraints on the value of the cosmological parameters.”