Fast and effortless computation of profile likelihoods using CONNECT

Abstract

The frequentist method of profile likelihoods has recently received renewed attention in the field of cosmology. This is because the results of inferences based on the latter may differ from those of Bayesian inferences, either because of prior choices or because of non-Gaussianity in the likelihood function. Consequently, both methods are required for a fully nuanced analysis. However, in the last decades, cosmological parameter estimation has largely been dominated by Bayesian statistics due to the numerical complexity of constructing profile likelihoods, arising mainly from the need for a large number of gradient-free optimisations of the likelihood function. In this paper, we show how to accommodate the computational requirements of profile likelihoods using the publicly available neural network framework CONNECT together with a novel modification of the gradient-based $basin$-$hopping$ optimisation algorithm. Apart from the reduced evaluation time of the likelihood due to the neural network, we also achieve an additional speed-up of 1$-$2 orders of magnitude compared to profile likelihoods computed with the gradient-free method of $simulated$ $annealing$, with excellent agreement between the two. This allows for the production of typical triangle plots normally associated with Bayesian marginalisation within cosmology (and previously unachievable using likelihood maximisation because of the prohibitive computational cost). We have tested the setup on three cosmological models: the $\Lambda$CDM model, an extension with varying neutrino mass, and finally a decaying cold dark matter model. Given the default precision settings in CONNECT, we achieve a high precision in ${\chi }^2$ with a difference to the results obtained by CLASS of $\Delta {\chi }^2\approx 0.2$ (and, importantly, without any bias in inferred parameter values) $-$ easily good enough for profile likelihood analyses.

arXiv comment23 pages, 9 figures

References in corpus (28)

• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Efficient Computation of CMB anisotropies in closed FRW models
• Cosmological parameters from CMB and other data: a Monte-Carlo approach
• A Unified Approach to the Classical Statistical Analysis of Small Signals
• Global Optimization by Basin-Hopping and the Lowest Energy Structures of Lennard-Jones Clusters Containing up to 110 Atoms
• The Cosmic Linear Anisotropy Solving System (CLASS) II: Approximation schemes
• Planck 2018 results. V. CMB power spectra and likelihoods
• Conservative Constraints on Early Cosmology: an illustration of the Monte Python cosmological parameter inference code
• Neutrino masses and the number of neutrino species from WMAP and 2dFGRS
• COSMOPOWER: emulating cosmological power spectra for accelerated Bayesian inference from next-generation surveys
• II: Bayesian Methods for Cosmological Parameter Estimation from Cosmic Microwave Background Measurements
• New constraint on Early Dark Energy from Planck and BOSS data using the profile likelihood
• Current cosmological constraints from a 10 parameter CMB analysis
• Restoring cosmological concordance with early dark energy and massive neutrinos?
• A fast test to assess the impact of marginalization in Monte Carlo analyses, and its application to cosmology
• Resolving the Hubble tension with Early Dark Energy
• New constraints on neutrino physics from Boomerang data
• Evidence for extra radiation? Profile likelihood versus Bayesian posterior
• New constraint on the tensor-to-scalar ratio from the $Planck$ and BICEP/Keck Array data using the profile likelihood
• CONNECT: A neural network based framework for emulating cosmological observables and cosmological parameter inference
• Data compression in cosmology: A compressed likelihood for Planck data
• Cosmological parameter estimation via iterative emulation of likelihoods
• CosmicNet II: Emulating extended cosmologies with efficient and accurate neural networks
• Capse.jl: efficient and auto-differentiable CMB power spectra emulation
• Profiling Cold New Early Dark Energy
• Fast and robust Bayesian Inference using Gaussian Processes with GPry
• LINNA: Likelihood Inference Neural Network Accelerator
• Uncertainty-aware and Data-efficient Cosmological Emulation using Gaussian Processes and PCA

Cited by in corpus (5)

• The CosmoVerse White Paper: Addressing observational tensions in cosmology with systematics and fundamental physics
• Robust constraints on tensor perturbations from cosmological data: a comparative analysis from Bayesian and frequentist perspectives
• Addressing the $H_0$ tension through matter with pressure and no early dark energy
• Calculating Bayesian evidence for inflationary models using CONNECT
• A frequentist view on the two-body decaying dark matter model