This PhD project is part of the CDT in Distributed Algorithms: The What, How and where of Next-Generation Data Science.
The University of Liverpool’s Centre for Doctoral Training in Distributed Algorithms (CDT) works in partnership with the STFC Hartree Centre and 20+ external partners from the manufacturing, defence and security sectors to provide a 4-year innovative PhD training programme that will equip up to 60 students with: the essential skills needed to become future leaders in distributed algorithms; the technical and professional networks needed to launch a career in next generation data science and future computing; and the confidence to make a positive difference in society, the economy and beyond.
The successful PhD student will be based at the University of Liverpool, co-supervised and work alongside the Materials Innovation Factory, Surface Science Research Centre, and our external partner Unilever.
Innovation in healthcare depends critically on creating novel materials that can control and treat infection. A similar approach is required to create new hygiene products in the personal care sector. Both sectors underpin a multi-billion pound UK economy.
There is an urgent need to accelerate the discovery of advanced materials and surfaces that can tackle emerging healthcare and personal care challenges. This project will utilise machine learning, bayesian optimisation, AI and high-performance computing to accelerate discovery, optimisation and development of a new generation of advanced materials and products that will protect public health and ensure societal wellbeing.
The molecular world provides the armoury to combat infection and safeguard public hygiene and health. Specifically, anti-infective agents (e.g. small molecules, natural compounds, antibiotics, etc.) are incorporated within surfaces and materials to create the operational core of many advanced technologies.
Currently experimental researchers at Liverpool are working with Unilever to investigate whether the individual components of this molecular arsenal can be combined in new and synergistic ways to create a step-change in performance and deliver a new generation of advanced products. This experimental research is creating a vast database, of the order of 103-106 discrete pieces of information, describing the relationship between specific agents and specific measured quantities related to their anti-infective performance (e.g. durability and stability of different agent combinations, the active display and release of each agent within specific combinations, response of the combination to environmental factors, and the adhesion, survival and growth of different types of bacteria exposed to each combination).
While this database contains the information necessary to interpolate and extrapolate, there is currently no technology that is applicable and can scale to the sizes of database that now exists. This motivates the need to develop machine learning algorithms that can extract trends and information. Given the desire to use this information to help researchers enhance their understanding and for the Artificial Intelligence to derive knowledge that can help inform and guide the discovery process, it will be imperative that any machine learning algorithm is transparent. We therefore plan to make use of Multi-Output Gaussian Processes to describe the understanding and to use emerging techniques and high-performance computational resources to ensure that the analysis can and does scale. We will then use Bayesian Optimisation to identify experiments that would be maximally informative about increasing this understanding.
This approach will enable faster exploration of new generations of technologies, accelerate innovation and provide an early assessment of performance that could streamline and shorten product development pipelines. Data correlations obtained between molecular combinations and performance will also generate a new knowledge base to inform future UK research and innovation in a number of other sectors.
The PhD will incorporate the following elements:
- Review of existing approaches to using Gaussian Processes to solve regression problems involving descriptions of molecules as the features and existing techniques for articulating similarities between different molecules and/or sets of molecules;
- Development of scalable Gaussian Process implementations (eg involving variational or distributed approaches to representing the uncertainty) and hyper-parameter estimation (eg using novel Sequential Monte Carlo methods) that exploit emerging many-cored compute resources to facilitate timely performance;
- Application of these approaches in the context of Bayesian Optimisation to help answer questions pertinent to the discovery of novel materials for healthcare.
UK and EU students are eligible to apply
Visit the CDT website for funding and eligibility information.
You must enter the following information:
- Admission Term: 2021-22
- Application Type: Research Degree (MPhil/PhD/MD) – Full time
- Programme of Study: Electrical Engineering and Electronics – Doctor in Philosophy (PhD)
The remainder of the guidance is found in the CDT application instructions on our website.