This repository contains the work of my masters thesis on dynamic network embeddings. I worked on this during my masters course at Cambridge from 2019-2020. It was a very interesting research project and inspired me to learn more about how I can apply my skills using machine learning to bridge my interest for mathematics with the real world.
The thesis highlights the challenges presented by embedding large dynamic graphs, like social networks, effectively. It highlights the importance of embeddings not being orthogonally invariant and the increased sensitivity of dynamic embeddings to dimensionality. If you find yourself interested in embedding dynamic graphs, please give it a read and I hope it helps you on your journey to understanding this increasingly useful area of machine learning.