Cosmology as an optimisation problem
File(s)
Author(s)
Mäkinen, Timothy Lucas
Type
Thesis
Abstract
This thesis presents several studies under a common objective: accelerating, automating, and enhancing cosmological inference and Bayesian statistical data analysis using information-theoretic objectives and neural networks. I present introductions to modern cosmological theory, Bayesian probability, and statistics and unite them under implicit, simulation-based inference. The research chapters cover advances made in both cosmological and statistical methods, culminating in an application of “hybrid statistics” to measure the Dark Energy equation of state parameter from weak gravitational lensing fields captured by the Dark Energy Survey.
Version
Open Access
Date Issued
2025-05-12
Date Awarded
01/07/2025
License URL
Advisor
Heavens, Alan
Wandelt, Benjamin
Porqueres, Natalia
Sponsor
Imperial College President's Scholarship
Publisher Department
Department of Physics
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)