Location:

Innovation Center, Streatham Campus, Exeter

The University of Exeter’s Department of Computer Science is inviting applications for 4 PhD studentships fully-funded by the Faculty to commence on 23 September 2024 or as soon as possible thereafter.  For eligible students the studentship will cover Home or International tuition fees plus an annual tax-free stipend of at least £18,622 for 3.5 years full-time.  International applicants need to be aware that you will have to cover the cost of your student visa, healthcare surcharge and other costs of moving to the UK to do a PhD.
Applicants must choose and highlight one (or more, up to a maximum of 3) of the 8 listed projects in their application. Applicants are encouraged to discuss the topic with the project supervisor prior to application.

Project Descriptions:

Project 1: Game Theory for Cyber Security Economics

Supervisor: Dr Yunxiao Zhang, Y.Zhang12@exeter.ac.uk

This project will explore the economic aspects of cybersecurity to build secure, sustainable, hyper-connected digital societies through enhanced awareness and strong multi-stakeholder partnerships. Using game theory, it will examine the interactions of organisational roles and human factors in cybersecurity investment, risk management, and decision-making, alongside technical defensive measures and safeguards. The project's objectives include: 1. Analysing attack scenarios with potential strategic behaviours; 2. Developing and enhancing robust game-theoretical approaches in cybersecurity investment; 3. Creating user-friendly, adaptive decision support tools for firms and individuals to evaluate security levels and optimise cybersecurity investments.

Project 2:  Leveraging Human Feedback for Safer and More Transparent Reinforcement Learning

Supervisor: Dr Xiaoyang Wang, X.Wang7@exeter.ac.uk

Reinforcement learning (RL) has demonstrated significant potential in solving complex tasks, through interacting with the environment. Human feedback, encompassing explicit instructions, prior knowledge, and safety requirements, can accelerate the learning process, enhance algorithm interpretability and safety. This project will explore how human expertise can be effectively integrated into RL, including the use of Large Language Models (LLMs), multimodal data fusion, and human-AI communication. The outcome of this PhD project is applicable across multiple domains, including clinical decision support systems and robotics.

Project 3: Enabling Co-creation with Explainable Interactive Optimisation

Supervisor: Dr David Walker , D.J.Walker2@exeter.ac.uk

This project will explore the potential of using explainable AI approaches in combination with human-in-the-loop optimisation to provide better feedback to users designing solutions to complex optimisation problems. It will consider approaches for enabling teams of decision makers and stakeholders to co-create solutions, optimising and visualising the trade-off between solutions to a given problem as well as the trade-off between the preferences of different users. The work will evaluate whether such approaches lead to the development of more robust and fitter solutions to problems than traditional optimisation approaches, considering problem domains including the water sector, offshore renewable energy, and transport.

Project 4: Enhancing Public Safety through Explainable Multi-Modal Video Understanding

Supervisor: Dr Zeyu Fu, Z.Fu@exeter.ac.uk

As the volume of online video content rapidly grows, ensuring public safety and content compliance has become increasingly challenging. This PhD project will explore the use of large multi-modal models (LMMs) to understand and explain both short and long-form video content for public safety applications. The main goal is to develop novel adaptation methods for LMMs to detect and mitigate harmful content and abnormal events in social media videos. By leveraging multimodality (text, visual, and audio), the project aims to provide transparent and explainable insights, thereby enhancing online safety and public security. The candidate will contribute to advancements in AI-driven public safety, benefiting from interdisciplinary collaboration in machine learning, computer vision, natural language processing, and computational social science, as well as access to cutting-edge high-performance computing resources.

Project 5: Learning-Guided Evolutionary Optimisation for Noisy Combinatorial Problems

Supervisor: Dr. Aishwaryaprajna, Aishwaryaprajna@exeter.ac.uk

Current challenges of core AI research often involve combinatorial problems with uncertainty, where the best solution must be searched from a large space of possibilities, mingled with random noise. Combinatorial optimisation problems can be graph-based or of bin-packing style, have constraints and multiple conflicting objectives or require expensive function evaluations. This project will explore learning mechanisms that can steer evolutionary operators through noise towards discovering superior solutions of these combinatorial problems with faster convergence. The scope of this PhD will include benchmarking procedures to assess algorithmic performance, theoretical performance guarantees for algorithms and addressing real-world applications on feature selection with large-scale dataset for healthcare and multi-agent system for route finding problem.

Project 6: Integrating Multimodal Large Language Models and Knowledge Graphs for Disease Understanding

Supervisor: Dr Hang Dong, H.Dong2@exeter.ac.uk

Multimodal Large Language Models (MLLMs) are powerful but lack fine-grained, long-tail understanding of data, lack sufficient explainability, and can generate hallucinations (inaccurate, nonsensical, or irrelevant information), that hinder their real-world applications, e.g., in medicine and healthcare. To address the issues above, the methodology-focused project will explore novel approaches to integrate MLLMs with knowledge graphs (e.g., ontologies), for deep phenotyping of diseases (e.g., dementia) from patients’ unstructured clinical notes, structured data, medical imaging data, and scientific publications. The project candidate will be embedded in an excellent team of natural language processing, knowledge representation, computer vision, and medical experts in Exeter and the UK.

Project 7: Privacy Preserving Mechanisms for Multimodal Data

Supervisor: Prof. Anne Kayem, A.V.Kayem@exeter.ac.uk

This PhD project will use large multimodal datasets to study how privacy preserving mechanisms can be designed to efficiently detect personally identifiable information (PII). The project will draw on work in the data profiling field to highlight data inconsistencies and errors that could potentially result in privacy leaks. The successful candidate will be expected to develop a series of algorithms with the goal of studying both the adversarial and benign perspectives of the problem.

Project 8: Design & Development of Mitigation Mechanisms against Architectural & Microarchitectural Security Vulnerabilities

Supervisor: Prof. Khurram Bhatti, K.Bhatti@exeter.ac.uk

Microarchitectural attacks (e.g., Spectre, Meltdown, Flush+Reload, Prime+Probe etc.) are orchestrated by generating multiple direct and indirect events, both in software and hardware, and they cause multiple state changes for various microarchitectural parameters (e.g., memory access time, access pattern, cache miss and hit ratios etc.).This PhD position aims to investigate potential architectural and micro-architectural security vulnerabilities targeting memory sub-system, particularly cache memory, of a heterogeneous SoC architecture and to develop protection mechanisms against such attacks that can be integrated into system software (OS) and hardware.

Entry requirements

Applicants for this studentship must have obtained, or be about to obtain, a First or Upper Second Class UK Honours degree, or the equivalent qualifications gained outside the UK, in an appropriate area of science or technology. 
 

If English is not your first language you will need to meet the required level as per our guidance at https://www.exeter.ac.uk/pg-research/apply/english/

How to apply

Apply now

In the application process you will be asked to upload several documents. 

• CV
• Letter of application (outlining your academic interests, prior research experience and reasons for wishing to undertake the project,
• You will need to include in your letter of application which particular project(s) you wish to be considered for - please list title and supervisor. You must choose one (or more, up to a maximum of 3) of the 8 listed projects (above) in your application).
• Research proposal
• Transcript(s) giving full details of subjects studied and grades/marks obtained (this should be an interim transcript if you are still studying)
• Two references from referees familiar with your academic work. If your referees prefer, they can email the reference direct to PGRApplicants@exeter.ac.uk quoting the studentship reference number.
• If you are not a national of a majority English-speaking country you will need to submit evidence of your proficiency in English.

The closing date for applications is midnight on 19th July 2024.  Interviews will be held virtually on the 24th and 25th of July 2024.

If you have any general enquiries about the application process please email PGRApplicants@exeter.ac.uk or phone 0300 555 60 60 (UK callers) +44 (0) 1392 723044 (EU/International callers).  Project-specific queries should be directed to the main supervisor.

Summary