UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Simulation-Based Inference of Large Scale Structure

Lin, Kiyam; (2024) Simulation-Based Inference of Large Scale Structure. Doctoral thesis (Ph.D), UCL (University College London).

[thumbnail of Lin_10198087_Thesis2.pdf] Text
Lin_10198087_Thesis2.pdf
Access restricted to UCL open access staff until 1 May 2025.

Download (12MB)

Abstract

In the burgeoning era of high precision cosmology, cosmologists are rapidly gaining the high fidelity data they require to probe the cosmos, much like what The Large Hadron Collider gave to particle physicists. However, unlike particle colliders where thousands of independent collisions can be performed, there is only one universe to observe, and thus only one statistically independent sample. Combined with the problem that the Universe is so large that no single survey is currently capable of observing the entire Universe in one go, the statistical inference task of cosmology involves some unique and challenging problems. In recent years, cosmic shear, or weak lensing has proven itself to be a good tracer of the large-scale structure of the Universe. Cosmic shear is the effect of gravity deflecting light by a small amount, causing the observed shapes of galaxies to be weakly sheared. The correlations observed in the shapes of galaxies can then be used to infer no only the amount of matter, but also the distribution of matter in the large-scale structure of the Universe. Furthermore, as cosmological surveys become increasingly complex, there is a drive towards using better statistical inference techniques that are not limited to Gaussian assumptions. This has given rise to the class of methods known as simulation-based inference (SBI). In chapter 3 we present work that makes use of simplified simulations to test the SBI methodology in the context of a real cosmic shear survey, the Kilo-Degree Survey. Following the successful completion of the work presented in chapter 3, we then perform a full SBI analysis with a novel suite of forward simulations and testing methodologies of the 4th public ESO KiDS-1000 data as shown in chapter 4. Chapter 5 then presents work that makes use of wavelet representations to extract information from the field in the form of data summaries. We develop a methodology to extract the information related to higher order statistics and show both its robustness and performance in comparison to competing methods. Chapter 6 discusses the future of SBI alongside avenues for future work.

Type: Thesis (Doctoral)
Qualification: Ph.D
Title: Simulation-Based Inference of Large Scale Structure
Language: English
Additional information: Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/).Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Space and Climate Physics
URI: https://discovery.ucl.ac.uk/id/eprint/10198087
Downloads since deposit
Loading...
4Downloads
Download activity - last month
Loading...
Download activity - last 12 months
Loading...
Downloads by country - last 12 months
1.United Kingdom
4

Archive Staff Only

View Item View Item