Galaxy Formation: Bayesian History Matching for the Observable Universe

TL;DR

This paper applies Bayesian history matching to galaxy formation modeling with Galform, enabling efficient uncertainty analysis and comparison with observations despite computational challenges.

Contribution

It introduces a Bayesian emulation approach for galaxy formation models, significantly advancing uncertainty quantification and model calibration techniques.

Findings

First detailed uncertainty analysis of galaxy formation simulation

Effective Bayesian emulation reduces computational costs

Improved model-data comparison methods

Abstract

Cosmologists at the Institute of Computational Cosmology, Durham University, have developed a state of the art model of galaxy formation known as Galform, intended to contribute to our understanding of the formation, growth and subsequent evolution of galaxies in the presence of dark matter. Galform requires the specification of many input parameters and takes a significant time to complete one simulation, making comparison between the model's output and real observations of the Universe extremely challenging. This paper concerns the analysis of this problem using Bayesian emulation within an iterative history matching strategy, and represents the most detailed uncertainty analysis of a galaxy formation simulation yet performed.