The project aims to develop Artificial Intelligent (AI) algorithms, which can map multi-functional material properties with multi-objective search algorithms to accelerate the discovery of novel materials exhibiting desirable functional properties. Some of these desirable characteristics would be piezoelectric or thermoelectric properties, allowing the novel materials to be used in the development of highly robust and miniaturised printable sensors.
Furthermore, such an AI model will be integrated with a suitable time integrator such as molecular dynamics to validate the predicted properties to yield a complete property map which can be validated by experimental flexible printing using a ink printing method.
Additionally, the project aims at the autonomous discovery of chemical rules from large-scale data via semi-supervised learning and local descriptor search. Infusion of domain knowledge will be tried out to make the ML framework more data efficient.
To utilise data from multiple experimental and computational sources, databases such as Materials Genome Initiative with varying levels of fidelity, will be use with inclusion of first principal DFT/ab-initio calculations.
Works with similar approaches are growing in the literature and some of these works will readily guide the methodology required in this project. In this regard, the following references will serve as suitable reference papers:
Requirements: Applicants must be of outstanding academic merit and should have (or be expected to gain) either a first class or an upper second-class Honours degree (or the international equivalent), or an MSc/MRes with distinction. Enthusiastic and self-motivated candidates from all countries with a background in either Engineering, Materials Science, Physics or Mathematics or a related discipline are encouraged to apply. Candidates should be able to demonstrate that they are highly motivated, have excellent communication skills and undertake challenging tasks using their own initiative.
Continue with Facebook
