Research projects in Information Technology

Agent-based computational simulations are now widely employed to study the evolution of behaviour, e.g., predator-prey simulations, the evolution of cooperation and altruism, the evolution of niches and food chains. These methods implement evolutionary processes in virtual populations of software agents and explore the evolution of their behaviour in diverse environments. Many behaviours that are not well understood in biological systems, that are difficult or impossible to measure in real environments, can be accurately captured and understood using these techniques.

"A picture is worth a thousands words"... or so the saying goes. How much information can we extract from an image of an insect on a flower? What species is the insect? What species is the flower? Where was the photograph taken? And at what time of the year? What time of the day? What was the weather like on the day the photograph was taken? This project aims to extract useful ecological and/or horticultural data from digital images by analysing their content.

In this project we are interested in exploring new paradigms for solving Combinatorial Optimisation problems, and generally NP-hard ones. One direction of research could consist in using approximation algorithms for deriving dual bound within a branch-and-bound algorithms. Other directions could use Machine Learning or new decompositions. This subject is generally quite open so it is important to be highly creative.

The brain is a complex machine and brain function remains yet to be fully understood. This project works at the intersection of dynamical modelling, statistical signal processing, statistical inference and machine learning to develop subject specific mathematical models of the brain that can be used to infer brain states and monitor and image the brain.

Life is exciting and challenging at the same time. Some people have it easy and some have it hard but we all need advice that can help us to live rich, fulfilling lives. By drawing on psychological, sociological, economic and web data this project will apply machine learning to build a recommendation engine that can provide life advice and predictions to individuals based on their current life situation and past experiences. The goal is to help people find ways to improve the quality of their lives through incremental adjustments suggested by the recommendation engine.

In multicultural societies diversity is a valuable asset. However, diversity faces critical challenges from extremism: the adoption of radical beliefs that are rejected by the mainstream.

This project aims to develop a model to explain the way in which conflicting beliefs divide societies into extremes, ultimately leading to the formation of ideology-based clusters, such as terrorist cells. These can occur in real world and online settings.

This project will seek to further the research into and development of machine learning techniques that may be used to triage, classify, and otherwise process material of a distressing nature (such as child exploitation material). It will involve the use of deep neural networks for image, video, audio, social network, and/or text classification.

This project aims to identify novel methods for inferring where and when photographs and videos were recorded from features of the material itself. A key requirement of image processing in a Law Enforcement (LE) context is to augment classification of material by identifying its spatio-temporal context.

Adversarial Machine Learning (AML) is a technique to fool a machine learning model through malicious input. Due to its significance in many scenarios, including security, privacy, and health application, AML has attracted a large amount of attention in recent years. However, the underlying theoretical foundation for AML still remains unclear and how to design effective and efficient attack and defence algorithms are remain a challenge in the research community. Furthermore, most existing  AML algorithms can only apply to Euclidean space.

The brain is a complex system and monitoring and imaging methods to observe critical neurophysiological variables underlying brain function are limited. This project works at the intersection of statistical signal processing, inference, machine learning and dynamical systems theory to develop new semi-analtyical filtering approaches for state and parameter estimation to infer neurophysiological variables such as network connection strengths between neural population networks underlying brain activity.