Development of machine learning, mathematical optimisation and algorithmic techniques for materials chemistry problems

This opportunity will remain open until a suitable candidate has been found, therefore early applications are encouraged.

This is an opportunity to undertake one of our new and exciting cross-disciplinary projects lying at the interface between computer science / mathematics and materials chemistry. The candidate does not need to have any knowledge in chemistry, but will need strong mathematical knowledge through a degree in maths, computer science, physics or engineering, as well as good programming skills.

We have a number of different problems to be investigated and the projects intend to develop both new models & theories and also practical applications. The broader research areas to be employed include mathematical modelling & optimisation, machine learning & data analytics, as well as algorithms and statistical analysis methods. Examples of such techniques include combinatorial and constrained optimisation, scale-space approaches, relaxation methods, neural networks, deep and reinforcement learning, statistical, unsupervised and supervised machine learning, signal/image processing, large-scale data visualisation, object sequencing, graph-based methods, topological data analysis, computational geometry, and any other methods that are potentially useful to solve the problems at hand.

The student will work closely with our very strong teams of computer scientists, mathematicians, inorganic chemists, physicists and material scientists to develop ways of predicting and analysing new materials. The supervisory team has a strong track record in the defining ingredients of the underlying work and will closely contribute to the originality of the research. Supervision is provided from both Computer Science and Chemistry departments to appropriately support the discipline background of the student. Publications in top-tier theoretical and also application-oriented venues will be expected. These 42 month PhD projects will tackle multidisciplinary problems co-defined by our industrial partners working with the University of Liverpool. Core training in robotics, automation, data science, etc., will form part of a unifying curriculum, together with leadership and entrepreneurship training, to underpin the individual research projects.

Background: Most current societal problems are limited by materials; for example, new electronics, faster computers, higher efficiency solar cells, higher performance catalysts and batteries that store more energy. All of these problems require new materials, which have to be discovered. This discovery process is difficult because each problem occurs in the combinatorial search space of all possible compositions within the periodic table, and one doesn’t know beforehand whether something exists. Further, the underlying interactions are complex, and many industries have decades of data studying a problem for several classes of materials, but the integration of these large historical datasets into a single model is challenging. Mathematical modelling, optimisation and machine learning methods have been shown to be promising in many complex problems and recent work has demonstrated that such methods may also be viable to predict new functional materials with desirable properties. Chemical applications may involve the discovery of better materials for automobile catalytic converters, industrial catalysis, transparent computer displays, new batteries and superconductors.

The candidate should have at least a 2.1 BSc in Computer Science, Mathematics or related discipline, and also be competent in scientific programming (Matlab, R, Python, or C++).

Informal enquiries should be addressed Dr Vitaliy Kurlin - vkurlin@liverpool.ac.uk.

Tel. No. for Enquiries: +44 (0)151 794 8861

Please apply by completing the online postgraduate research application form:

https://www.liverpool.ac.uk/study/postgraduate-research/how-to-apply/

Please ensure you quote the following reference on your application: University of Liverpool Doctoral Training Centre in Next-Generation Materials Chemistry CDT03

Applications should be made as soon as possible.