EPSRC DTP PhD Studentship: Building interpretable generative models of images (2022)

What you get

Standard UKRI stipend of £16,062 stipend, research training grant of £1,650 per annum and full tuition fees up to the overseas rate for 3.5 years.

Type of award

Postgraduate Research

PhD project

In recent years computer vision has made fantastic progress in employing deep learning to create descriptive representations of image data. These methods have enabled accurate predictions (e.g. classification, segmentation etc.) to be made from images, where the learned representations are optimised specifically for these tasks. While effective in many situations, such approaches may struggle to generalise to new examples, particularly where data is drawn from different distributions to the training set. Moreover, learned representations lack both direct interpretability and uncertainty, which limits our ability to understand or resolve reasons for failure.

The aim of this project is to create a machine learning framework that learns human interpretable representations of images, where variations are explicitly decomposed into separate semantically meaningful components such as shape, identify, pose, texture, lighting etc. Important considerations include:devising appropriate model formulations, effectively integrate prior knowledge into the learning process;and exploring how to capture the uncertainty, or ambiguity, of describing an image using these explicit factors.

This research project will employ state of the art deep learning methods to advance our understanding of how interpretable image representations can be learned and used for both discriminative (e.g. classification) and creative (e.g. image editing) tasks. The specific application area for this project is flexible and will depend on the student’s interest. Potential topics include: analysing photographs of human/animal facesor bodies, landscapes and cityscapes,or building models of neurodegenerative disease from brain MRI.

This is a computational project, which would suit a student with good mathematical and programming skills and a keen interest in probabilistic machine learning and computer vision. Students will be expected to present their work at top-tier computer vision, medical image analysis or machine learning venues such as CVPR, ICCV/ECCV, MICCAI, NeurIPS etc.

Eligibility

The stipend is available to: UK / EU / Overseas.

The fee waiver is available to: UK / EU / Overseas.

Eligible candidates will have a 2:1 degree or equivalent in a related field.

Deadline

How to apply

Apply online for a full time PhD in Informatics using our step-by-step guide (http://www.sussex.ac.uk/study/phd/apply). Here you will also find details of our entry requirements.

Please clearly state on your application form that you are applying for the EPSRC DTP 2022 under the supervision of Dr Ivor Simpson (I.Simpson@sussex.ac.uk) 

Contact us

For queries related to the admissions process please contact PhD.Informatics@sussex.ac.uk. For queries related to the research project please contact I.Simpson@sussex.ac.uk.

Timetable

Deadline: June 10th (Extended deadline)

Interview Date: Late June 

Decision: Early July 

Start Date: 19th September