Honours and Masters project

Multiple master thesis/honour projects on MLLM/ human understanding are available.

Benefits: we aim for publication at top conferences and journals. You will have chance for full PhD scholarship. For those working hard, paid RA opportunities will also be provided.

requirement: WAM>80 and high self-motivation 

Inclusive learning environments are essential for ensuring that all students can fully participate and succeed. Neurodivergent learners—including those with ADHD, autism, dyslexia, and other cognitive differences—often navigate university settings that were not designed with their learning profiles in mind. Challenges related to feedback interpretation, cognitive load, communication, and assessment design can create barriers that impact learning, confidence, and wellbeing.

Autoregressive moving average (ARMA) models remain a competitive tool for forecasting low signal-to-noise ratio time series, due to their flexibility, low complexity and physical plausibility. They predict the next observation in a time series as a linear combination of a number of previous observations as well as a number of hidden (latent) random innovations. The AutoARIMA package remains a staple benchmarking tool against which forecasting techniques must be compared.

Traditional marketing analytics rely on predictive models that estimate the probability of customer behaviours such as churn or purchase. However, these models identify customers who are likely to act, not those whose behaviour can be influenced by an intervention. Uplift modelling addresses this limitation by estimating the causal effect of a marketing intervention on individual customers, enabling firms to target those whose behaviour is expected to change as a result of treatment rather than those who would act regardless.

This project aims to design and evaluate a hybrid optimisation framework using Qiskit and complementary classical solvers to address complex urban transport optimisation challenges.

The research will benchmark quantum-assisted and classical optimisation methods in terms of accuracy, scalability, and computational efficiency, and explore how hybrid algorithms can improve routing, scheduling, and energy management in next-generation urban mobility systems.

Studies using computed tomography (CT), particularly high-resolution CT and quantitative CT have become a crucial non-invasive method for examining airway thickness and the structural changes known as airway remodeling in chronic respiratory diseases such as asthma. Bronchial thermoplasty is a treatment option for patients with severe asthma and works by applying heat energy to reduce the amount of excess airway smooth muscle, which is often abnormally thickened in patients with asthma.

Modern transport and logistics rely on efficient routing to ensure that goods are delivered on time and at minimal cost. Determining the optimal order in which vehicles visit a set of customers is a fundamental challenge in mathematics and computer science. Despite decades of research, this problem remains computationally difficult. To achieve reasonable cost-effectiveness in industry-scale applications, the best-performing methods rely on heuristic and suboptimal solvers. This project will develop new suboptimal algorithms for tackling large instances relevant to practice.

Shortest path algorithms are used to advise travelers on the best route from an origin to destination across various modes including car and public transport. One important variant of the standard shortest path problem is the inclusion of resource constraints, where a user has a limited budget in cost, time, carrying capacity, etc. and needs to find a shortest path within this budget. Unlike the regular shortest path problem, the resource-constrained shortest path problem is theoretically proven to be NP-hard, making its computation difficult.

Providing timely, individualised feedback is a persistent challenge in large-scale computing units. This project investigates how Generative AI models can automatically produce pedagogically aligned, rubric-based feedback on student submissions. A prototype system will interface with an LLM API (e.g., OpenAI GPT) and generate structured feedback, which will be evaluated for accuracy, usefulness, and tone against educator benchmarks.

Quantum algorithms such as Grover’s Search promise quadratic speed-ups over classical search but are sensitive to hardware noise. This project will use Qiskit Aer to model realistic noise channels (decoherence, gate and readout errors) and evaluate their impact on algorithmic performance. By varying circuit depth, qubit count, and noise parameters, the student will identify conditions under which quantum advantage remains achievable and investigate possible error-mitigation strategies.