Simulations of Quantum Chromodynamics (QCD), a microscopic theory of strong interactions in nuclei, are among the most challenging computational problems in physics, and typically use parallel computing, hardware acceleration, advanced statistical analysis of very large datasets (Big Data), and low-level code optimization. During their PhD studies, the student will acquire universally transferable skills in all these areas of scientific computing, in addition to the core knowledge of nuclear and hadronic physics.
One of the most important challenges for Monte-Carlo simulations is to equilibrate and de-correlate configurations of gluon fields, so that their stationary distribution reproduces the Boltzmann weight of QCD. This requires complex, non-local updates of gluon fields. In this project, the student will utilize neural networks such as Variational Autoencoders (VAEs) and/or Generative Adversarial Networks (GANs) to learn the most efficient non-local updates that lead to fastest equilibration and thermalization of ensembles of quantum fields.
The student will then apply these algorithmic developments to address some of the outstanding questions about possible exotic phases of hadronic matter, such as possible superfluidity and colour superconductivity of extremely dense nuclear matter in neutron star cores. For example, they will be able to study the dependence of transport coefficients such as electric conductivity and viscosity on fermion density within two-colour QCD at nonzero quark density. Transport coefficients directly feed into hydrodynamic simulations of quark-gluon plasma and play an important role in the interpretation of experimental results on hadron and lepton production in heavy ion collision experiments on LHC and RHIC colliders. Furthermore, as a knowledge transfer activity, they will also be able to apply Monte-Carlo methods used in lattice QCD to study condensed matter systems with strongly correlated electrons, and/or to apply machine learning methods to beam control in particle accelerators.
In collaboration with our industrial partner Art Recognition AG (Zurich), you will also explore machine learning methods for generating synthetic artwork, such as GANs and VAEs and their extensions. With art-generating Artificial Intelligence algorithms becoming more and more sophisticated, it becomes more and more important to distinguish authentic works of art created by real human artists from algorithmically generated ones. In the second part of the project, you will research the methods to reliably identify digitally generated images.
Throughout the project the student will have access to comprehensive postgraduate training in theoretical physics at the Department of Mathematical Sciences at the University of Liverpool, as well as to targeted training in data science provided by the University of Liverpool with the Centre for Doctoral Training LIV.INNO.
The student will also be given the opportunity to carry out a 6-months industry placement at the DiRAC supercomputing facility of the UK's Science & Technology Facilities Council to broaden your wider research and career skills. The placement would focus on Research Software Engineering projects in which you could bring your expertise to bear on research problems outside your field.
Student: Joseph Hadley
Back to: Centre for Doctoral Training for Innovation in Data Intensive Science