Fully funded PhD opportunity to investigate knowledge based approaches to support the autonomous formulation of materials through computational simulations. Application deadline October 22, 2021
This is an exciting opportunity to work on an interdisciplinary PhD at the intersection between Artificial Intelligence and Materials Science. The project is fully funded and undertaken in collaboration with Unilever; to be supervised by Dr Valentina Tamma, Dr Terry Payne, and Prof Boris Konev in the Department of Computer Science and Mr Andrew Mitchell in Unilever.
“Autonomous Formulation” refers to the use of autonomous software or robots to determine the composition and recipe of mass-market consumer products (e.g. shampoo, laundry detergents, hand sanitisers). Exploring the space of possible formulations is a complex problem that poses several challenges for material scientists; however, this complexity can be offset by using computational experiments that can, to some extent, predict behaviours and characteristics. These experiments are typically workflow-centric and pose new challenges to guarantee reproducibility. Some of the challenges are related to the vast amout of data, possibly heterogeneous, that is produced to represent specific tasks within experiments and specific measured quantities with respect to their performance
Knowledge graphs (KGs) have become popular thanks to Google, which uses them to power its knowledge panel describing people or events, are instrumental in helping scientists to automatically derive new information from their data. Recently, KGs models have been proposed to support the structuring, sharing, reusing and preservation of scientific assets and their annotations related to a computational experiment. A KG is a network graph of relationships between different assets, and their properties and additional information can be inferred by traversing such graph.
The objective of this PhD is to adapt and extend KG based approaches to effectively support computational simulations of formulations. This architecture will allow the use of machine learning techniques to optimise specific formulation profiles with the aim to:
1) Reduce the time needed to set up and execute formulation experiments;
2) Support the autonomous determination of the formulations, thus increasing the number of experimental runs;
3) Preserve experimental configurations and the associated meta-data, therefore providing explicit benchmarking for new formulations.
The successful applicant will work as part of a team of computer scientists, roboticists, and computational chemists to achieve the project vision. The project is open-ended and concept driven.
We are seeking creative and energetic individuals from a range of backgrounds. We require a 1st or 2:1 at first degree level and/or or a Master’s degree level or equivalent to apply. Typical degree subjects include (but are not limited to) Computer Science, Chemistry, or Materials Science, particularly those with skills directly related to this project. We also welcome those who have significant relevant work experience.
This project is funded by EPSRC iCase award studentship and Unilever for a duration of 4 years Full Time or part-time equivalent.
For any enquiries please contact Dr Valentina Tamma, Department of Computer Science, University of Liverpool on: V.Tamma@liverpool.ac.uk
To apply for this opportunity, please visit: https://www.liverpool.ac.uk/study/postgraduate-research/how-to-apply/ and click the 'Apply online' button.