Goal Driven Learning of Quantum Chemical Energy Surfaces using Multi-Fidelity Bayesian Optimization

AI_CDT_DecisionMaking

Details

In recent years, Machine Learning Interatomic Potentials (MLIPs), have become increasingly important in computational materials modelling. The current state of the art MLIP algorithms are message-passing graph neural networks (GNN-IPs), which use non-linear activations, physically constrained by equivariance under translation and rotation, to accurately and efficiently match the predictions of high-fidelity, fully quantum mechanical (QM) calculations at a tiny fraction of the computational cost [1]. A particular advantage of GNN-IPs is the prospect of Foundation Model (FM) architectures, which exploit transfer learning of “chemical intuition”, when trained over large chemical databases, to render chemically accurate predictions with zero- or few-shot accuracy. Another important application of MLIPs is the use of inexpensive, “ephemeral” MLIPs [2] to efficiently identify new synthesisable compounds, as part of crystal structure prediction (CSP) algorithms.

In general, MLIP models do not generically benefit when trained on data from older models: different models often fail for different configurations, and in different ways, limiting generalisation to out-of-distribution molecules [3]. This often necessitates additional, expensive data generation steps (QM computations) when training GNN-IP FMs. Efficient sampling of the interatomic potential energy surface (PES), dictated by constituent atomic species and their relative positions – whether to parsimoniously generate transferable QM data or to rapidly identify minima for CSP – is therefore among the most important tasks for modern computational materials modelling. This is a particularly formidable task due to high-dimensional search space, mixed continuous and discrete variables and a highly non-convex energy landscape, problems which are further compounded by the high computational cost of the QM methods which, which provide “ground truth” values of relative energies used in model fitting.

Underlying many recent advances in MLIP methodology is the Atomic Cluster Expansion (ACE) [5], which provides a physically transparent and universal descriptor of local atomic environments. ACE allows physically interpretable cluster-based expansions of local atomic energies and forms the basis of a systematically improvable hierarchy of possible MLIP architectures [5]. This project will develop new methods for single (BO) and mutli-fidelity Bayesian optimization (MFBO) of complex chemical PE surfaces, allowing efficient PES sampling which will balance the utility and cost of additional QM calculations of proposed atomic configurations [6]. A key step in achieving this will be the implementation of a probabilistic measures within ACE, together with appropriate acquisition functions to guide PES sampling [7].

Advantages of the proposed approach will include greatly improved training time and model accuracy; MFBO will allow parameterisation of GNN-IP models using expensive, highest-fidelity “gold standard” QM data. This has a number of important practical applications, such as modelling of battery cathodes and nanoconfined water, and there is significant scope for theory-experiment collaboration and high-impact publications due to the highly collaborative and multi-disciplinary environment at the National Graphene Institute (NGI), where this PhD will be based. Additionally, while this project focuses on the applications of MLIP models, the cluster expansion provides a formally complete expansion of the statistical partition function of generic multicomponent systems [8], and is therefore highly likely to find applications beyond atomistic materials modelling.

Before you apply

We strongly recommend that you contact the supervisor(s) for this project before you apply. Please include details of your current level of study, academic background and any relevant experience and include a paragraph about your motivation to study this PhD project.

How to apply:

Please apply through the below link for the PhD Artificial Intelligence CDT:

https://pgapplication.manchester.ac.uk/psc/apply/EMPLOYEE/SA/s/WEBLIB_ONL_ADM.CIBAA_LOGIN_BT.FieldFormula.IScript_Direct_Login?Key=UMANC1251000021489F

When applying, you’ll need to specify the full name of this project, the name of your supervisor, if you already having funding or if you wish to be considered for available funding through the university, details of your previous study, and names and contact details of two referees.

Your application will not be processed without all of the required documents submitted at the time of application, and we cannot accept responsibility for late or missed deadlines. Incomplete applications will not be considered.

After you have applied you will be asked to upload the following supporting documents:

• Final Transcript and certificates of all awarded university level qualifications
• Interim Transcript of any university level qualifications in progress
• CV
• Supporting statement: A one or two page statement outlining your motivation to pursue postgraduate research and why you want to undertake postgraduate research at Manchester, any relevant research or work experience, the key findings of your previous research experience, and techniques and skills you’ve developed. (This is mandatory for all applicants and the application will be put on hold without it).
• Contact details for two referees (please make sure that the contact email you provide is an official university/work email address as we may need to verify the reference)
• English Language Certificate (if applicable)

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. We know that diversity strengthens our research community, leading to enhanced research creativity, productivity and quality, and societal and economic impact. (Equality, diversity and inclusion nearmejobs.eu The University of Manchester)

We actively encourage applicants from diverse career paths and backgrounds and from all sections of the community, regardless of age, disability, ethnicity, gender, gender expression, sexual orientation and transgender status.

We also support applications from those returning from a career break or other roles. We are dedicated to supporting work-life balance and offer flexible working arrangements to accommodate individual needs. Our selection process is free from bias, and we are committed to ensuring fair and equal opportunities for all applicants.

_{featuredproject22_nov24}