This presentation is an extension of my review paper on stochastic block models (SBMs). There is this ever-present issue of selecting the number of blocks (or groups) when applying an SBM. This presentation focuses on the approaches that use a criterion to select this number. Various criteria that existed then are:

A taster lecture was given to high school students who want to do STEM degrees in university. It was done virtually (of course) at the height of the pandemic and first lockdown, and I thought, what better topic to talk about than epidemics? I didn’t know how the students took it though, but there was at least a differential equation to work on.

I still remember this presentationpresentation at CASA quite clearly, as it was my last talk before Covid lockdown.

Internal reading group on computational statistics and machine learning

Data science institute internal seminar

The title of this guest lecture (slides) was slightly misleading. It was more about data visualisations in the larger context of statistical analysis.