School of Electronic Engineering and Computer Science

Bayesian Machine Learning for Causal Discovery

Supervisor: Dr Anthony Constantinou

Research group(s): Risk & Information Management

Are you interested in the theory of causality? Do you want to improve the algorithms we use to discover causal, or other, relationships from data and possibly knowledge? This project will focus on the problem of Bayesian Network structure learning when data incorporate missing data points and latent confounders (important missing variables). Research may also extend to other data issues, such as when dealing with ordinal or nominal distributions, discrete or continuous data, and the various statistical distributions that can be used to represent each of the variables. The project will be adjusted to the skills and interests of the successful candidate. For example, theoretical advancements may be evaluated by applying them to an area (or areas) of your interest, such as in economics, finance, medicine, gaming, and sports.