{\sc CosmoNet}: fast cosmological parameter estimation in non-flat models using neural networks

Abstract

We present a further development of a method for accelerating the calculation of CMB power spectra, matter power spectra and likelihood functions for use in cosmological Bayesian inference. The algorithm, called {\sc CosmoNet}, is based on training a multilayer perceptron neural network. We compute CMB power spectra (up to $\ell =2000$) and matter transfer functions over a hypercube in parameter space encompassing the $4\sigma$ confidence region of a selection of CMB (WMAP + high resolution experiments) and large scale structure surveys (2dF and SDSS). We work in the framework of a generic 7 parameter non-flat cosmology. Additionally we use {\sc CosmoNet} to compute the WMAP 3-year, 2dF and SDSS likelihoods over the same region. We find that the average error in the power spectra is typically well below cosmic variance for spectra, and experimental likelihoods calculated to within a fraction of a log unit. We demonstrate that marginalised posteriors generated with {\sc CosmoNet} spectra agree to within a few percent of those generated by {\sc CAMB} parallelised over 4 CPUs, but are obtained 2-3 times faster on just a \emph{single} processor. Furthermore posteriors generated directly via {\sc CosmoNet} likelihoods can be obtained in less than 30 minutes on a single processor, corresponding to a speed up of a factor of $\sim 32$. We also demonstrate the capabilities of {\sc CosmoNet} by extending the CMB power spectra and matter transfer function training to a more generic 10 parameter cosmological model, including tensor modes, a varying equation of state of dark energy and massive neutrinos. {\sc CosmoNet} and interfaces to both {\sc CosmoMC} and {\sc Bayesys} are publically available at {\tt www.mrao.cam.ac.uk/software/cosmonet}.

8 pages, submitted to MNRAS

References in corpus (19)

• Efficient Computation of CMB anisotropies in closed FRW models
• Cosmological parameters from CMB and other data: a Monte-Carlo approach
• A Line of Sight Approach to Cosmic Microwave Background Anisotropies
• The 3D power spectrum of galaxies from the SDSS
• Three-Year Wilkinson Microwave Anisotropy Probe (WMAP) Observations: Temperature Analysis
• The 2dF Galaxy Redshift Survey: The power spectrum and the matter content of the universe
• High Resolution Observations of the CMB Power Spectrum with ACBAR
• Extended Mosaic Observations with the Cosmic Background Imager
• A Measurement of the CMB <EE> Spectrum from the 2003 Flight of BOOMERANG
• A Measurement of the Angular Power Spectrum of the CMB Temperature Anisotropy from the 2003 Flight of Boomerang
• High sensitivity measurements of the CMB power spectrum with the extended Very Small Array
• Efficient Cosmological Parameter Estimation from Microwave Background Anisotropies
• A measurement of the polarization-temperature angular cross power spectrum of the Cosmic Microwave Background from the 2003 flight of BOOMERANG
• Cosmic Calibration: Constraints from the Matter Power Spectrum and the Cosmic Microwave Background
• Fast cosmological parameter estimation using neural networks
• The entropic prior for distributions with positive and negative values
• Rapid Calculation of Theoretical CMB Angular Power Spectra
• Fast Cosmological Parameter Estimation from Microwave Background Temperature and Polarization Power Spectra
• CMBfit: Rapid WMAP likelihood calculations with normal parameters

Cited by in corpus (42)

• The Coyote Universe II: Cosmological Models and Precision Emulation of the Nonlinear Matter Power Spectrum
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• Cosmological constraints from the tomographic cross-correlation of DESI Luminous Red Galaxies and Planck CMB lensing
• SKYNET: an efficient and robust neural network training tool for machine learning in astronomy
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• BayesCLUMPY: Bayesian Inference with Clumpy Dusty Torus Models
• PkANN - II. A non-linear matter power spectrum interpolator developed using artificial neural networks
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Full-sky Cosmic Microwave Background Foreground Cleaning Using Machine Learning
• PkANN - I. Non-linear matter power spectrum interpolation through artificial neural networks
• Statistical methods for cosmological parameter selection and estimation
• Accelerating Large-Scale-Structure data analyses by emulating Boltzmann solvers and Lagrangian Perturbation Theory
• Decoding spectral energy distributions of dust-obscured starburst-AGN
• BAMBI: blind accelerated multimodal Bayesian inference
• CosmicNet II: Emulating extended cosmologies with efficient and accurate neural networks
• ECoPANN: A Framework for Estimating Cosmological Parameters using Artificial Neural Networks
• Machine Learning for Observational Cosmology
• Deciphering the Archaeological Record: Cosmological Imprints of Non-Minimal Dark Sectors
• Likelihood-free Cosmological Constraints with Artificial Neural Networks: An Application on Hubble Parameters and SNe Ia
• CosmicNet I: Physics-driven implementation of neural networks within Boltzmann-Einstein solvers
• Neural Network Reconstruction of $H^{\prime }(z)$ and its application in Teleparallel Gravity
• Intelligent Design: On the Emulation of Cosmological Simulations
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• Cleaning our own Dust: Simulating and Separating Galactic Dust Foregrounds with Neural Networks
• Inferring galaxy dark halo properties from visible matter with Machine Learning
• LINNA: Likelihood Inference Neural Network Accelerator
• Bayesian analysis of weak gravitational lensing and Sunyaev-Zel'dovich data for six galaxy clusters
• Accelerating cosmological inference with Gaussian processes and neural networks -- an application to LSST Y1 weak lensing and galaxy clustering
• Computing High Accuracy Power Spectra with Pico
• Accelerating Bayesian microseismic event location with deep learning
• Designing Decisive Detections
• Efficient cosmological parameter sampling using sparse grids
• Towards fast machine-learning-assisted Bayesian posterior inference of microseismic event location and source mechanism
• Exploring local fNL estimators based on the binned bispectrum
• MF-Box: Multi-fidelity and multi-scale emulation for the matter power spectrum
• Caching and Interpolated Likelihoods: Accelerating Cosmological Monte Carlo Markov Chains
• DeepSSM: an emulator of gravitational wave spectra from sound waves during cosmological first-order phase transitions
• Emulators for stellar profiles in binary population modeling
• Cosmic Kite: Auto-encoding the Cosmic Microwave Background
• Constraining the primordial power spectrum using a differentiable likelihood
• Design and optimization of neural networks for multifidelity cosmological emulation
• Neural Network Acceleration of Large-scale Structure Theory Calculations