Honours and Masters project

In this project, we aim at surveying relevant computational tools/models used for automatic question generation, and then comparing the effectiveness of these tools/models by using existing datasets.

Disruptive technologies such as artificial Intelligence (AI) systems can have unintended negative social and business consequences if not implemented with care. Specifically, faulty or biased AI applications may harm individuals, risk compliance and governance breaches, and damage to the corporate brand.

The aim for this project is to research, prototype and/or evaluate approaches to increase the explanatory effectiveness of the visualisations contained in analytics dashboards or similar support data-intensive tools. Explanatory visualisations are those whose main goal is the presentation and communication of insights. By contrast, exploratory visualisations are commonly targeted at experts in data analysis in search of insights from unfamiliar datasets.

The research challenge for this project is to research, prototype and evaluate approaches to automatically capture multimodal traces of team members’ activity using sensors (such as indoor positioning trackers, physiological wristbands and microphones), using learning analytics techniques to make sense of sensor data from healthcare contexts. Depending on the trajectory that you take, examples of the questions that such a project could investigate include:

Over the past decades, we have witnessed the emergence and rapid development of deep learning. DL has been successfully deployed in many real-life applications, including face recognition, automatic speech recognition, and autonomous driving, etc. However, due to the intrinsic vulnerability and the lack of rigorous verification, DL systems suffer from quality and security issues, such as the Alexa/Siri manipulation and the autonomous car accidents, which are introduced from both the development and deployment stages.

In multi-label classifications, a data sample is associated with more than one active label, which is a more challenging task than conventional single-label classifications. This project will focus on eXtreme Multi-Label (XML) classifications for text data (i.e., documents), where the label set can be extremely large, e.g., more than 10,000. For example, the input texts can be the item descriptions of an e-commerce website (e.g., Amazon) and one needs to classify them into a large set of item categories.

Despite the rapid progress made recently, deep learning (DL) approaches are data-hungry. To achieve their optimum performance, a significantly large amount of labeled data is required. Very often, unlabelled data is abundant but acquiring their labels is costly and difficult. Many domains require a specialist to annotate the data samples, for instance, the medical domain. Data dependency has become one of the limiting factors to applying deep learning in many real-world scenarios.

The Optimisation group is looking for multiple students to contribute to our world leading research. Our interests range from practical to theoretical. So whether you are interested in path finding for AI in games, solving a complex scheduling problem, designing new algorithms, or working on our specialised modelling language, we will have a project that is of interest to you!

To operate, interact and navigate safely in dynamic human environments, an autonomous agent, e.g. a mobile social robot, must be equipped with a reliable perception system, which is not only able to understand the static environment around it, but also perceive and predict intricate human behaviours in this environment while considering their physical and social decorum and interactions.

In this project, the goal is to develop a new method (using computer vision and machine learning techniques) for robotic navigation in which goals can be specified at a much higher level of abstraction than has previously been possible. This will be achieved using deep learning to make informed predictions about a scene layout and navigating as an active observer in which the predictions inform actions.