PhD Studentship: Bayesian Modeling of High-dimensional Structural Data

Overview

The Department of Mathematics at The University of Manchester is offering a fully funded 3.5-year PhD studentship on Bayesian modelling of high-dimensional structural data. Fields such as biological sciences, social science and engineering produce high-dimensional observations, and a Bayesian approach offers a flexible way to model underlying structures such as covariance structures and conditional dependency graphs. When the data-generating mechanism changes, these structures may change over time, depending on latent factors or variables.

The projects will develop a comprehensive Bayesian learning framework for this broad class of problems, focused on specific applications. The goal is computationally efficient and scalable Bayesian learning methods with practical uses, together with their relevant theoretical properties.

Who can apply

Home students are eligible to apply. Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master's, or an international equivalent, in a relevant science or engineering discipline.

What you get

The successful candidate receives an annual tax-free stipend set at the UKRI rate, which is 21,805 pounds for 2026/27 and is expected to increase each year, and tuition fees are paid.

How to apply

Applicants contact the project supervisor, Dr Nilabja Guha, with their current level of study, academic background, relevant experience and a paragraph on their motivation. Early application is recommended because the advert may be removed before the deadline.

Key dates

Applications are accepted all year round, with a listed closing date of 28 October 2026, and the start date is October 2026.