Thesis Defense: On the statistics of the Universe: An effective field theory approach to large scale structures

Ph.D. Candidate: Henry Zhang
Research Advisor: Eva Silverstein

Zoom Link: https://stanford.zoom.us/j/96346039371
Password: Email physicsstudentservices@stanford.edu for password.

The cosmic microwave background (CMB) has been exhaustively measured and analyzed, solidifying its role in shaping our understanding of the early universe. As a result, large-scale structure (LSS) now stands as the leading frontier for extracting new, high-precision cosmological information. The Effective Field Theory of Large-Scale Structure (EFTofLSS), after many stages of theoretical development, is now poised for application to observational data. In this work, we compute next-to-leading order (NLO) corrections to both the power spectrum and bispectrum of biased tracers within the EFTofLSS framework. We further introduce a fast and accurate algorithm for evaluating these corrections to sub-percent precision in a cosmology-independent manner, significantly reducing computational cost during inference. We demonstrate how the one-loop bispectrum predictions tighten constraints on early universe physics and enhance the sensitivity of upcoming galaxy surveys. In addition, we highlight the role of NLO corrections in enhancing the robustness of neutrino mass measurements under various beyond the standard model $\Lambda$CDM extensions. Lastly, we introduce a new optimization algorithm, $\mathsf{ECD}q=1$, adapted from a similar algorithm used in Bayesian inference, the MicroCanonical Hamiltonian Monte Carlo (MCHMC) algorithm.