Graph machine learning and algorithms for large-scale real-world graphs.

Graph ML and algorithms have gained popularity in the industry as they solve many business problems. However, the unique characteristics of real-world graphs pose many challenges to their effective implementation and deployment.

In this talk, I’ll cover a range of topics regarding challenges and opportunities with graphs, including:

1. Graph representations and adjacency compression.

2. Efficient implementation of graph algorithms.

3. Addressing the characteristics of real-world graphs (heterogeneity, imbalance etc.) in graph ML.

4. Running a large production graph system in the cloud.

I will also present a few MSc Master’s thesis and undergrad summer internship topics.

Neo4j is a company that tackles graph problems, spanning graph databases, algorithms, ML, query language, visualisation etc. Brian Shi is a senior software engineer at Neo4j and an Oxford alumnus.