From the secrets of the universe, to the cutting-edge of computer science, our Theoretical Physics programme is more than a collection of modules - it is a way of thinking that connects varied schools of thought and a broad range of scientific topics.
On your programme you will learn about the timeless, classical themes and universal laws of theoretical physics and combine these fundamentals with knowledge of modern, field-advancing developments, developed over the past fifty years.
Your research project will be an application of these universal laws and advanced developments and your topic will be drawn from an incredibly wide range of research areas - from quantum field theory, to machine learning, black holes and everything inbetween.
Explore a selection of our wide-ranging available projects:
This project is a computational project which involves analysis of data collected from large collections of text, using an algebraic approach to matrix theory.
The theoretical framework and some example applications are explained in https://arxiv.org/abs/1912.10839. A central role is played in the approach by permutation invariant polynomial functions of matrices and tensors, which are related to graph theory (see image below).
Pre-requisites: python programming and familiarity with finite groups.
In 2+1D, particles may exhibit so-called “anyonic" statistics that go beyond the usual Bose/Fermi paradigm. In this project, we will explore various quantum mechanical constructions involving anyons while looking at how they may be of use in building novel quantum computers.
Pre-requisites: a good grasp of quantum mechanics and symmetries; corequisite: quantum field theory
We will attempt to optimise the neural network metaparameters of depth and width for an ensemble of learned functions. This should then contribute to an understanding of how to pick neural network shape parameters. Pre-requisite: Python programming
This project will review and look to develop the recent approaches to machine learning based on quantum field theory with a specific focus on renormalisation as learning.
Pre-requisite: Quantum Field Theory module
This project will focus on the SIR model (Susceptible, Infectious, or Recovered) and its variants with their practical implementation in Python using different optimisation strategies. Pre-requisite: a good grasp of Python programming
This project will look at how we can construct the geometry of information spaces and how can this be used in machine learning. This project will focus on the construction of information metrics and their uses. Pre-requisites: Knowledge of general relativity, differential geometry