Supervisor: Dr Omer Bobrowski and Dr Primoz Skraba
Project description:
This project will focus on a series of conjectures, recently discovered experimentally, arguing that topological features have strong universality properties (i.e., the distribution of topological descriptors is independent of the model generating the data). The main goal of this project is to develop both the theory related to these conjectures, and their potential applications in statistics and machine learning. The project will roughly be equally divided between the two parts. The theoretical part will combine probability theory, with algebraic topology and geometry.
The applications part will mainly address how universality can contribute to assessing the statistical significance of structures detected in data using topological tools.
No funding is currently assigned to this project.
Further information:
How to apply
Entry requirements
Fees and funding