Funding Opportunities

WGSSS Studentships

Aberystwyth University is a member of the ESRC Welsh Graduate School in Social Science (WGSSS) which supports a number of fully-funded studentships in the social sciences. Students can apply for studentships in four research areas or ‘pathways’: (1) Environmental Planning, (2) Health, Well-Being and Data Science, (3) Human Geography and (4) Politics, International Relations, and Area Studies. Information on each pathway is provided in the links below.  

Applicants should consider approaching potential supervisors before submitting their application to confirm that there is appropriate supervisory capacity and to discuss their draft application.

What will the studentship cover:

Studentship awards cover your tuition fees as well as a maintenance grant and include access to additional funding through Research Training Support Grants (RTSG).  There are other opportunities and benefits available to studentship holders, including an overseas fieldwork allowance (if applicable), internship opportunities, overseas institutional visits and other small grants.

Eligibility

WGSSS studentships are highly competitive. Applications should come from exceptional candidates with a first class or strong upper second-class honours degree, or appropriate Master’s degree. The University values diversity and equality at all levels and encourages applications from all sections of the community, irrespective of age, disability, sex, gender identity, marital or civil partnership status, pregnancy or maternity, race, religion or belief and sexual orientation.  In line with our commitment to supporting and promoting equality, diversity and inclusion, and to increase recruitment of currently underrepresented groups, applications from Black British, Asian British, minority ethnicity British and mixed-race British candidates are particularly encouraged and welcomed.  We welcome applications for both full and part-time study.

Pathways:

Detailed information on eligibility, topic areas and the application process can be found on the links below.

Environmental Planning

Health Wellbeing and Data Science

Human Geography

Politics, International Relations, and Area Studies

 

The Welsh Graduate School for the Social Sciences (WGSSS) recently hosted two webinars; ‘How to apply for a WGSSS studentship’ and ‘How to write a research proposal’. The webinars were designed to make the competition more accessible to those who are considering applying in the 2024 WGSSS studentship competition. The webinars covered topics such as; how to find a supervisor, how to prepare for an interview, and how to structure your proposal. Recordings of the webinars are available on the WGSSS studentships page.

 

WGSSS welcomes applications from students of all backgrounds. We value academic excellence and life skills, as well as the ability to meet challenges and student’s capacity to enrich the life of our community. Widening participation is a key goal for WGSSS and we are keen to receive applications from able and ambitious students. We are a collaboration between Cardiff University (the lead institution), Aberystwyth University, Bangor University, Cardiff Metropolitan University, the University of Gloucestershire, the University of South Wales and Swansea University.  

 

The closing date for applications in the General Competition is the 12th January 2024 (institutions may have earlier deadlines, these will be detailed in the individual General Competition adverts), the Collaborative Competition will be launching in March 2024.

 

Isabel Ann Robertson Scholarship

Isabel Ann Robertson, always known as Ann, was a tutor in the Computer Science Department at Aberystwyth University for 25 years from 1984 to 2009. But her links with the University spanned several generations. Ann Davies was born in London in 1932, the eldest of three children. Her mother, Enid Sayers, had graduated in English from the then University College of Wales in Aberystwyth in the 1920s and later (as Enid Davies) was Vice President of the Old Students’ Association. Ann’s father, C W Davies, was also an Aber graduate and was later a professor of Chemistry and Head of the Chemistry Department. Ann studied Physics when the department was still based in the Old College on the seafront, graduating with a BSc in 1954 and an MSc by research in 1957. Her research was on cavitation. She was also a College athlete and a member of the Sailing Club. In 1956 she married David Robertson, whom she had met through the Sailing Club. His work for the Forestry Commission took them to many different parts of the UK, including Glasgow, where Ann took an MSc in Computer Science. They returned to Aberystwyth to live in the 1980s. Their daughter, Sara Robertson, also studied at Aberystwyth from 1978 to 1981 and their granddaughter, Fiona Robertson, followed, from 2011 to 2015.

Ann Robertson PhD Scholarship - Details of the Award & Available Projects

Open to applicants who qualify for Home (UK) fees status only, there are three full-time PhD scholarships available.  These will be allocated one per Faculty and on a competitive basis to three of the projects described in the Ann Robertson PhD Scholarship 2023 Project Details. Those awarded an Ann Robertson Scholarship will receive a grant for up to three years which will cover their tuition fees up to the UK rate of £4,712 per annum (2023/24 rate).  A maintenance allowance of approximately £18,622 per annum* and access to a travel and conference fund (max. £500 per annum*) will also be provided. Scholarships commence in September 2023 (although flexible starts up to February 2024 can be discussed).

How to Apply 

Closing date 27th September 2023

To be considered, candidates must complete the usual full online PhD application AND the specific  Ann Robertson PhD Scholarship Application Form 2023 

The completed Ann Robertson Scholarship Application Form should be submitted via our online Postgraduate Application Portal at the point of application.   

To make a full PhD application, firstly visit our course pages and find the details of the course for which you wish to apply.  Once you have found your chosen course page, select the “Apply Now” button to start your application.  

The Postgraduate Admissions Application Portal will ask you to provide us with your personal details, confirm your course selection(s) and upload documents in support of your application.  Please have you supporting documents saved in PDF format and ready to upload to your online application. 

At the same time, the completed Ann Robertson Scholarship Application Form should also be sent as an attachment by email to Prof Reyer Zwiggelaar (rrz@aber.ac.uk), Head of Graduate School, with the subject heading ANN ROBERTSON SCHOLARSHIP APPLICATION. 

Please ensure that you read the Ann Robertson PhD Scholarship Terms & Conditions thoroughly. 

Any Questions? 

If you have any specific queries regarding the projects listed, please contact the main supervisor associated with the project.  

If you have any queries about the postgraduate application process please contact pg-admissions@aber.ac.uk

Ann Robertson PhD Scholarship 2023 Project Details

Can Colloids Learn: A Path Towards Smart Matter

Dr Adil Mughal (Department of Mathematics) - aqm@aber.ac.uk 

This PhD project aims to develop an innovative method for controlling the self-assembly of patchy colloidal particles, which have potential applications in photonic crystals, targeted drug delivery, electronic and sensor technologies. The ultimate goal is to create smart colloids that can receive chemical or physical inputs and respond with a change in morphology and function, leading to ver-satile and adaptive materials.

Through a combination of numerical simulations and theoretical work, the project will explore a groundbreaking approach to self-assembly using transmutable nanoparticles, driven to crystallize along multiple thermodynamic trajectories. In this system, colloidal particles with complementary patches that enable them to stick together will "learn" to form desired structures through a process analogous to reinforcement learning. The strength of attractive interactions between patches will be adjusted based on the desirability of the arrangement, leading to more controlled self-assembly.

This innovative method has the potential for significant technological impact in various industries, as it addresses current limitations in self-assembly techniques and inverse statistical-mechanical methods. By overcoming these challenges, the project will pave the way for the development of next-generation materials with valuable optical, electrical, and mechanical properties, and enable new applications in areas such as cloaking, chemical sensing, imaging, nonlinear optics, and meta-fluids.

The candidate should have an undergraduate degree in Applied Mathematics, Physics, or a related discipline, with grade II(1) or above. Programming experience is essential.

Back to top

Investigating Area Selective Deposition (ASD) as an Enabler for Future Technologies

Dr Anita Brady-Boyd (Department of Physics) - anb116@aber.ac.uk 

As feature sizes continue to decrease in integrated circuit chips, fabricating these devices becomes increasingly difficult. At present Apple’s most current chip boasts an incredibly small 28 nm width for their interconnects, which are the wires connecting the individual transistors. The transistors themselves are ~ 48 nm wide. Area selective deposition (ASD) allows for the nanoscale patterning of materials while limiting the use of traditional photolithographic and etch steps which can be time-consuming, expensive, and wasteful of materials. Although only a new field of research, ASD has been hailed as a driver for next generation technology. This PhD will take a fully interdiscipli-nary approach to answer fundamental questions about ASD with the core objective of the project to progress towards implementing ASD processes into nanoelectronics fabrication. ASD requires the use of small molecules called self-assembled monolayers (SAMs). These molecules create both a physical and chemical barrier to block deposition. The initial stages of the project will involve fundamental study into how these molecules interact and adhere to industry relevant substrates. Different methods of deposition for the SAMs will be investigated. Next, the ability of the SAM to block any subsequent deposited material will be investigated. This will utilise a lot of the world-class equipment already available to the Physics department. The final part of this project will in-volve collaboration with the industrial links of the PI. This stage involves the scale up of our ASD process to high volume manufacturing on 300 mm wafers used in industry.

Back to top

Soft matter approaches towards versatile and scalable photonics materials

Dr Chris Finlayson (Department of Physics) - cef2@aber.ac.uk

Polymer nano-spheres (with composite core-shell layers) can be synthesised and arranged into crystal structures, also known as Polymer Opals, to produce intense iridescent colours. In a real advance over other forms of synthetic opals, they are made by standard plastic manufacturing techniques, presenting a promising platform for next generation bulk-scale photonic structures, coatings, and sensors. They are flexible and durable, making them suited for mass production and incorporation into consumer items, and unlike with existing dyes/pigments, they are non-toxic, inexpensive and resistant to fading.

The recently developed bending induced oscillatory shearing (or BIOS) sample preparation meth-ods have had a transformative effect to the ordering and quality of such soft matter photonics. The next challenge is the general application of BIOS in generating a wider range of highly ordered opaline materials with advanced optical functionality. This studentship will offer significant pro-gress on multiple fronts; process development for new functional materials, furthering the re-search underpinning the scale-up to innovative applications, and the underlying science of order-ing in composite soft nanophotonics.

A key challenge is a deeper understanding of the rheological (fluid mechanics) properties of poly-meric viscoelastic media and the exact mechanisms and time evolution of crystallisation under shear flow. A combined experimental and theoretical approach will synergise detailed rheometry with simulation modelling and machine learning (in collaboration with the Maths Department). With applications in mind; the key scale-up of thin-film photonics to roll-to-roll processing, and the associated tolerances and quality control, will be examined using state-of-the-art in-line goniome-try and hyperspectral imaging techniques.

Back to top

Characterisation, Calibration and Testing of a Miniature Infrared Spectrometer for Planetary Explo-ration

Dr Dave Langstaff (Department of Physics) - cef2@aber.ac.uk 

Aberystwyth University is building a miniature spectrometer which is proposed to form part of the instrumentation suite for the Rosalind Franklin mars exploration rover due to launch on the re-scheduled European Space Agency ExoMars mission in 2028. The spectrometer, named Enfys after the Welsh word for rainbow, is to operate in the Near Infrared (NIR) and Short Wavelength Infra-red (SWIR) bands and will be used primarily for geological identification but also for studies of the Martian atmosphere. The instrument is required to survive the vibration and shock of launch and landing as well as the extreme cold of the Martian night down to -130C. It will be then required to operate during the Martian day over a temperature range from -50C to +40C.

As part of the development of this instrument, there is an opportunity for a PhD student to work on characterisation and testing of component parts and the completed instrument at low temper-atures; investigation of potential failure modes during thermal cycling, and procedures for pre-launch and in-situ calibration and testing of the completed instrument.

The successful candidate will ideally have experience of instrument design and calibration as well as practical laboratory and software skills. As well as their PhD thesis, they will contribute towards scientific papers written during the course of the project.

Back to top

How we See Motion in Depth

Dr David Hunter (Department of Computer Science) - dah56@aber.ac.uk 

This PhD studentship opportunity involves using machine learning techniques to understand how humans process moving objects. Using machine learning techniques on static images has proven a powerful tool in enabling researchers to understand human visual processing. However, extending these techniques to moving images is an on-going challenge. With human subjects this is further complicated by head motion and eye-tracking of moving objects.

Therefore, this PhD studentship opportunity focuses on addressing the complexities associated with analysing motion and object tracking in dynamic environments. As a PhD student in this project, you will be responsible for gathering first-person perspective videos and eye-tracking data from individ-uals performing various tasks. You will use this data to develop AI models that mimic human visual processing. By creating robust models of early-stage visual processing, your work will contribute to advancing our understanding of the brain and its cognitive functions.

Back to top

To Develop a Machine Learning for Prediction of Childhood Asthma from Pre-school Information: a Super Learner Ensemble Approach.

Dr Faisal Rezwan (Department of Computer Science) - far8@aber.ac.uk 

Childhood asthma is a major cause of morbidity both in the UK and worldwide and also incurs con-siderable healthcare costs to the nation. A cure for childhood asthma is desperately needed but very little is understood about its early origins. However, at present we cannot predict which children with wheeze will develop asthma or which will outgrow their symptoms. To answer these questions, we need to focus on this early phase of asthma development during first few years of life. Different factors, including demographic, co-existing medical conditions, and environmental exposures, are likely to be important. Given the multiple contributions of epidemiological risk factors to the development of childhood asthma, a new approach is warranted to gain a great insight into disease pathogenesis and better estimation of disease risk. It has been clearly demonstrated that the use of machine learning models in asthma prediction can be beneficial in clinical decision making. Few studies have undertaken research for developing diagnostic or prognostic prediction of school age asthma development using machine learning. However, all of these studies suffer from generalisibility and lacks the approach of explainable AI. Therefore, the aim of this doctoral project is to develop a more robust and efficient prediction models using super learner ensemble approach to predict childhood asthma using larger cohort datasets. For this, we will use multiple cohort data (>14,0000 samples from five cohorts) from the Study Team for Early Life Asthma Research (STELAR) consortium. and further to develop an online prototype tool for predicting childhood asthma.

Back to top

Quantum Engineering Thermodynamics

Prof John Gough (Department of Physics) - jug@aber.ac.uk 

The Python package QuTiP will be used to simulate continuous observed quantum systems and study the entropy production. This contributes to the field of quantum thermodynamics which is currently gathering considerable interest: especially to the emerging subfield dealing with stochastic nonequilibrium problems. The novelty here is that we use insights from signal processing, feedback control, as well as the specific expertise of the supervisor in modelling interconnected open quantum systems and networks. There have been prior interpretations of the (Kalman) filter as a “Max-well’s demon” in both the classical and quantum case.

The quantum Kalman was studied in a recent paper but uses the Shannon entropy of the Gaussian Wigner function rather than the correct von Neumann entropy. We seek to remedy this and examine specific case studies. To study this we need to solve the algebraic Riccati equations associated with the filter and here numerical simulations will be necessary. There will be an opportunity to collaborate with international researchers. The project is suitable for a physics/mathematics graduate with experience of Python programming.

Back to top

Measurement Resources in Open Quantum Systems

Dr Jukka Kiukas (Department of Mathematics) - jek20@aber.ac.uk 

This project will build a theoretical framework for characterising the dynamical evolution of quantum properties in measurements induced by environmental interaction. This contributes to the re-search area of quantum information theory and correlations, and is motivated by practical applications to quantum technology.

An open quantum system interacts with its environment, which may consist of “noisy” surroundings or monitored ancillary systems. The resulting temporal evolution can be characterised by successive or continual transformations on the state of the system, or, alternatively, on the measurements available for extracting information from the system for further classical processing. This project will focus on the latter — while evolution of quantum properties in states (such as entanglement) have been studied extensively, much less is known about measurement resources.

The main aim is to understand how quantum measurement resources degrade with time in open systems — this is motivated by the development of noisy quantum devices. As measurements are mathematically constructed from the semidefinite cone, one starting point is to study the temporal contraction of the cone and the resulting reduction in specific resources. Preliminary results exist in the case of incompatibility of a given set of measurements, that is, nonexistence of a common refinement which could simulate all of them. This naturally leads to further related topics, including quantum contextuality, steering, and uncertainty relations, among plenty of other possibilities.

The candidate should have an undergraduate degree in Mathematics or Physics, with grade II(1) or above. Interest in the mathematical structure of quantum theory is essential. Master’s degree and/or knowledge on quantum information theory / open systems is desirable.

Back to top

Better Multimodal Benchmarks for Theory of Evolutionary Computation

Dr Maxim Buzdalov (Department of Computer Science) - mab168@aber.ac.uk 

Evolutionary computation is a computational intelligence discipline studying randomised algorithms acting as black-box optimisers. Over years it developed sound theoretical grounds with proven results about the performance of algorithms on certain problem classes. Here, properly designed benchmark functions play an important role: they can be used to understand the working principles of evolutionary algorithms by considering in isolation different features that make optimisation problems hard.

One of such features is multimodality, that is, presence of local optima, and the most famous bench-mark function dedicated to multimodality in pseudo-Boolean optimisation is called Jump. However, there is evidence that the Jump function has a number of properties that makes it too easy for some algorithms. This may lead to overly optimistic conclusions about their performance, which do not hold in practice. Recent papers studied few variations of it, but the research field calls for more.

The aim of the project is to develop variations of the Jump function, possibly having multiple parameters that control its shape, that are more realistic in terms of carrying the results over to real-world problems, but still allows rigorous runtime analysis. The project will also require to perform such analysis for a number of evolutionary algorithms, and to augment this analysis with computational experiments that clarify the behaviour for the cases where the theory is not precise enough.

Back to top

Why is Reasoning about Function Widely used by Engineers yet so Difficult for Computers? (Computer based Functional Reasoning: Analysis and Automation)

Dr Neal Snooke (Department of Computer Science) - nns@aber.ac.uk 

Engineers use notions of function and purpose as an abstraction to efficiently understand, reason, and explain how and why products and systems work. In contrast, behaviour and structure explain what a product does and is well supported and understood by industry. Despite many decades of academic research there remains little consensus and no explicit standard functional reasoning (FR) approach incorporated into software or tool design used by industry software or design automation tools.

Building on experience from previous work at Aberystwyth on model-based (deterministic, explain-able, deductive) AI tools for a variety of engineering design tasks that rely on functional interpretation, this project will consider the fundamental similarities and differences of the competing functional representations.

The research will determine if an encompassing framework incorporating concepts including multi-view flexible structural hierarchy, context and deployment-mode and separation of interpretative purpose and physical effects is possible. If the different approaches and views of function cannot be reconciled, then a clear schema and understanding of these fundamental differences with respect to a broad range of target tasks such as design, redesign, planning, explanation, diagnosis, failure mode analysis, prognosis, reverse engineering, design verification, etc., will be developed.

Overall the work will support, and guide engineers tasked with approach and tool selection and enhance development of (AI) software and tools that support the product lifecycle and exchange of functional information.

Back to top

Anytime Analysis for Dynamic Optimisation Problems

Dr Thomas Jansen (Department of Computer Science) - thj10@aber.ac.uk 

Many optimisation problems are too difficult to be solved efficiently by standard algorithms. Heuristic optimisation methods like evolutionary algorithms are frequently applied in these situations. Theory still puts an emphasis on runtime analysis and is at odds with the way these heuristics are actually applied. This project addresses this gap by concentrating on anytime analysis targeting dynamic problems that change over time.

Building on existing anytime analysis results (sometimes also called fixed budget results) as well as recent results from fixed target analysis, the project performs a systematic study of dynamic optimisation. The starting point are simple static unimodal and multimodal benchmarks and simple different tools to construct dynamic optimisation problems from static ones. Starting from simple heuristics like random sampling and local search the tools and methods are developed to com-pare these baseline methods with more advanced methods, employing populations, crossover, and different approaches to deal with dynamic optimisation problems like hall of fame approaches or diploidy.

Back to top

On-site Breed Classification from Low Coverage DNA Sequencing

Dr Wayne Aubrey (Department of Computer Science) - waa2@aber.ac.uk 

This project aims to use Machine Learning (ML) methods and state-of-the-art DNA sequencing to assign cattle by breed origin. There are more than 1000 recognized breeds of cattle globally and ~12 commercially important breeds in the UK. The MinION sequencer by Oxford Nanopore has made the real-time genetic testing of animals on farms affordable. Genetic testing has the potential to increase the value of animals to breeders by providing empirical evidence of breeding lineage, but current methods rely on central genotyping facilities that are time-consuming, with breeders often waiting 4-8 weeks to obtain the result of testing for a given animal. Research by Dr Wayne Aubrey and Dr Matt Hegarty has demonstrated that ML methods can be applied to cattle genomes to clas-sify a set of single nucleotide variants (SNVs) into individual breeds. Identifying SNVs relies on align-ing DNA sequence reads to a reference genome to identify single base differences, which typically results in ~3 million SNVs.

The first step is to establish a database of SNV allele frequencies derived from the 1000 Bulls dataset (121 breeds, over 2000 animals) to train a ML model to identify what combinations of SNVs and their location in the genome contribute the most to breed classification. Reducing the number of SNVs under consideration avoids the computationally slow process of aligning reads to the entire genome. The project also aims to improve the affordability of testing by determining what is the minimum sequence coverage necessary for accurate classification.

Back to top

Trust Management in Vehicular Networks using Artificial Intelligence

Dr Yasir Saleem (Department of Computer Science) - yss1@aber.ac.uk 

A vehicular network contains vehicles that could be autonomous, semi-autonomous or with human drivers. Vehicles collect several types of data from the environment that include accident data, emergency data, advertisements, and backup data to name a few. Such information is exchanged with other vehicles using vehicle-to-vehicle (V2V) communication and with the infra-structure, called road-side unit (RSU), using vehicle-to-infrastructure (V2I) communication. In addition, vehicles also exchange information related to their location, speed and heading direction with other vehicles and RSUs. However, in such a scenario, some vehicles may misbehave for their own benefit (e.g., to get more resources) and in this regard, can provide false information or services. Such misbehaving vehicles can either act alone or in a group. Hence, trust management systems have the important role in providing trustable communications, preventing data manipulation by unauthorized vehicles, and can be developed to guarantee the detection of malicious behaviours. Trust management in vehicular networks is challenging because firstly, vehicles meet other vehicles for short duration and secondly, it is less likely that vehicles meet the same vehicles again. It is in-teresting to investigate how Artificial Intelligence (AI) (such as reinforcement learning) can be used for trust management in vehicular networks.

The aim of this project is to investigate trust issues in vehicular networks and develop a trust man-agement system using AI that considers high mobility, high dynamicity, and heterogeneity. The per-formance should be evaluated using a network simulator (OMNeT++ or NS-3 with SUMO) by con-sidering real-world vehicular mobility datasets.

AberDoc

AberDoc Scholarships are part of a prestigious fund for Research Postgraduates.

These awards are tailored to enable students to develop the necessary skills required to meet their career choices and offer a breadth of development opportunities to enhance their research, teaching and transferable skills.

For more information, check out the dedicated page for AberDoc.

AHRC Scholarship

Aberystwyth University is one of a number of institutions in the South, West & Wales Doctoral Training Partnership (DTP) and has successfully secured funding for PhD scholarships in the arts and humanities. Successful students may benefit from potential supervision and training opportunities available at more than one university within the DTP.

Please note that these awards are only available to UK students and are for new PhD students rather than current PhD students.

Please visit the South, West & Wales Doctoral Training Partnership website for further information.

Other Funding

Other funding opportunities are available. For more information check the dedicated Other Funding page.