Extremely expensive likelihoods: A variational-Bayes solution for precision cosmology

TL;DR

This paper introduces a variational-Bayes method to efficiently estimate non-Gaussian posteriors in cosmology when likelihood evaluations are extremely costly, enabling analysis that was previously computationally infeasible.

Contribution

The authors develop a variational-Bayes approach that can handle expensive likelihoods and limited model evaluations, providing a practical alternative to MCMC in complex cosmological analyses.

Findings

Successfully reconstructed KiDS-450 posterior with only 0.6% of original likelihood evaluations.

Reduced computational cost allows inclusion of systematic effects previously too expensive to consider.

Demonstrated applicability to real cosmological data with significant efficiency gains.

Abstract

We present a variational-Bayes solution to compute non-Gaussian posteriors from extremely expensive likelihoods. Our approach is an alternative for parameter inference when MCMC sampling is numerically prohibitive or conceptually unfeasible. For example, when either the likelihood or the theoretical model cannot be evaluated at arbitrary parameter values, but only previously selected values, then traditional MCMC sampling is impossible, whereas our variational-Bayes solution still succeeds in estimating the full posterior. In cosmology, this occurs e.g. when the parametric model is based on costly simulations that were run for previously selected input parameters. We demonstrate the applicability of our posterior construction on the KiDS-450 weak lensing analysis, where we reconstruct the original KiDS MCMC posterior at 0.6% of its former numerical posterior evaluations. The reduction in numerical cost implies that systematic effects which formerly exhausted the numerical budget could now be included.